wb_bm

ANSYS TurboSystem User Guide

Release 14.5ANSYS, Inc.October 2012Southpointe

275 Technology DriveCanonsburg, PA 15317 ANSYS, Inc. is

certified to ISO9001:2008.

[email protected]://www.ansys.com(T) 724-746-3304(F) 724-514-9494

Copyright and Trademark Information

© 2012 SAS IP, Inc. All rights reserved. Unauthorized use, distribution or duplication is prohibited.

ANSYS, ANSYS Workbench, Ansoft, AUTODYN, EKM, Engineering Knowledge Manager, CFX, FLUENT, HFSS and anyand all ANSYS, Inc. brand, product, service and feature names, logos and slogans are registered trademarks ortrademarks of ANSYS, Inc. or its subsidiaries in the United States or other countries. ICEM CFD is a trademark usedby ANSYS, Inc. under license. CFX is a trademark of Sony Corporation in Japan. All other brand, product, serviceand feature names or trademarks are the property of their respective owners.

Disclaimer Notice

THIS ANSYS SOFTWARE PRODUCT AND PROGRAM DOCUMENTATION INCLUDE TRADE SECRETS AND ARE CONFID-ENTIAL AND PROPRIETARY PRODUCTS OF ANSYS, INC., ITS SUBSIDIARIES, OR LICENSORS. The software productsand documentation are furnished by ANSYS, Inc., its subsidiaries, or affiliates under a software license agreementthat contains provisions concerning non-disclosure, copying, length and nature of use, compliance with exportinglaws, warranties, disclaimers, limitations of liability, and remedies, and other provisions. The software productsand documentation may be used, disclosed, transferred, or copied only in accordance with the terms and conditionsof that software license agreement.

ANSYS, Inc. is certified to ISO 9001:2008.

U.S. Government Rights

For U.S. Government users, except as specifically granted by the ANSYS, Inc. software license agreement, the use,duplication, or disclosure by the United States Government is subject to restrictions stated in the ANSYS, Inc.software license agreement and FAR 12.212 (for non-DOD licenses).

Third-Party Software

See the legal information in the product help files for the complete Legal Notice for ANSYS proprietary softwareand third-party software. If you are unable to access the Legal Notice, please contact ANSYS, Inc.

Published in the U.S.A.

Table of Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12. Workflows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2.1. Geometry Sources .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.2. Mesh Sources .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.3. Solution Sources .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.4. Examples of TurboSystem Workflows .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42.5. Usage Notes for Specific Workflows .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.5.1. Tips on using ANSYS Workbench .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.5.2. Using BladeGen .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2.5.2.1. Changing the Active Document in BladeGen .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.5.3. Using BladeEditor ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2.5.3.1. Loading versus Importing a BladeGen Geometry .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.5.3.2. Restarting a BladeEditor Session .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.5.3.3. Modifying Spline Curves .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.5.3.4. Adding a Hub Fillet to an Imported BladeGen Geometry .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.5.3.5. Creating a Full 360-Degree Fluid Zone for an Impeller ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2.5.4. Using ANSYS TurboGrid .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82.5.5. Using ANSYS CFX .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2.5.5.1. Connecting from a Turbo Mesh Cell .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92.5.5.2. Changing the Geometry .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.6. Using ANSYS Workbench Journaling and Scripting with TurboSystem ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92.6.1. Acquiring a Journal File with a TurboSystem Component in ANSYS Workbench .... . . . . . . . . . . . . . . . . . . . 10

2.6.1.1. Journal of an Operation that uses Vista TF .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.6.1.2. Journal of an Operation that uses ANSYS TurboGrid .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.6.2. Scripting .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122.6.2.1. Example: Using a Script to Change the Mesh Density in ANSYS TurboGrid .... . . . . . . . . . . . . . . . . . . . 12

2.7. Quick Pump Tutorial ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132.7.1. Designing the Blade and Creating the Mesh .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132.7.2. Setting up the Turbomachinery Simulation .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.7.3. Viewing the Turbo Report ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

3. BladeGen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.1. Tips on Using the Create New Blade CFD Mesh Command .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

4. BladeEditor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.1. Configuring the ANSYS BladeModeler License .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.2.The BladeEditor User Interface .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

4.2.1. Tree View and Details View .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284.2.2. Contour Sketch Management .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294.2.3. BladeEditor Toolbars ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

4.2.3.1. Feature Creation Toolbar .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304.2.3.2. Active Selection Toolbar .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324.2.3.3. Display Control Toolbar .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

4.2.4. Auxiliary Views .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324.2.4.1. Blade-to-Blade View .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334.2.4.2. Blade Lean Graph .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344.2.4.3. Curvature View .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354.2.4.4. Meridional Curvature View .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

4.2.5. Angle and Thickness Views .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.2.5.1. Angle View .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.2.5.2. Thickness View .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

4.2.6. Section Definition and Stacking Views .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

iiiRelease 14.5 - © SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information

of ANSYS, Inc. and its subsidiaries and affiliates.

4.2.6.1. Section Definition View .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404.2.6.2. Section Stacking View .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

4.2.7. User Preferences and Properties ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454.3. Blade Editing Features .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

4.3.1. Flow Path Contour Creation .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474.3.2. FlowPath Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484.3.3. Blade Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

4.3.3.1. Blades made using Camberline/Thickness sub-features .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514.3.3.1.1. Adding/Removing CamThkDef sub-features .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

4.3.3.2. Blades made using Blade Section (Airfoil Design Mode) Sub-features .... . . . . . . . . . . . . . . . . . . . . . . . . . . 534.3.4. Camberline/Thickness Definition Sub-feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

4.3.4.1. Interpolated and Non-interpolated Angle/Thickness Data .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584.3.4.2. Importing and Exporting Angle Definition Data .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584.3.4.3. Importing and Exporting Thickness Definition Data .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 594.3.4.4. Converting Curves to Bezier or Spline .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 594.3.4.5. Converting Angle Definition Data .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

4.3.5. Blade Section (Airfoil Design Mode) Sub-feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 604.3.6. Splitter Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

4.3.6.1. Cloned Splitter ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 654.3.6.2. Independent Splitter ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

4.3.6.2.1. Camberline/Thickness Definition Sub-features of Independent Splitters ... . . . . . . . . . . . . . . 674.3.7. Stage Fluid Zone Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674.3.8. Throat Area Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

4.4. Blade Comparison .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 704.5. Importing Blades from ANSYS BladeGen .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

4.5.1. Limitations of the ImportBGD Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 774.6. Loading and Modifying Blades from ANSYS BladeGen .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 774.7. Using and Exporting Blades .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

4.7.1. Export to Vista TF (.geo) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 784.7.2. Export as Meanline Data (.rtzt file) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 794.7.3. Export to ANSYS TurboGrid .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

4.8. Blade Parameterization .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844.8.1. Meridional Contours .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844.8.2. FlowPath Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844.8.3. Blade Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844.8.4. Camberline/Thickness Definition Sub-feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 854.8.5. Splitter Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 854.8.6. StageFluidZone Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 864.8.7. ExportPoints Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 864.8.8.VistaTFExport Feature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

4.9. Tutorial 1: Blade Editing With Emphasis On Sketches, Layers, and Blade Comparison .... . . . . . . . . . . . . . . . . . . . . . 864.9.1. Creating the Blade .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 874.9.2. Editing the Main Blades and Splitter Blades in BladeEditor ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

4.9.2.1. Modifying the Shroud .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 884.9.2.2. Using Blade Comparison Mode While Updating a Control Point on the Shroud .... . . . . . . . . . . . . 914.9.2.3. Changing the Number of Blades .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 944.9.2.4. Changing the Leading Edge .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 944.9.2.5. Adding Camberline/Thickness Definitions on New Layers .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 944.9.2.6. Calculating the Throat Area .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

4.9.3. Summary .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974.10. Tutorial 2: Blade Editing With Emphasis On Camberline and Thickness Distributions .... . . . . . . . . . . . . . . . . . . . 97

4.10.1. Creating the Blade .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Release 14.5 - © SAS IP, Inc. All rights reserved. - Contains proprietary and confidential informationof ANSYS, Inc. and its subsidiaries and affiliates.iv

TurboSystem

4.10.2. Editing the Main Blades and Splitter Blades in BladeEditor ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 994.10.2.1. Adding Blade Clearance .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 994.10.2.2. Modifying Camberline/Thickness Definitions at the Hub and Shroud .... . . . . . . . . . . . . . . . . . . . . . . . . 994.10.2.3. Modifying the Thickness Distribution of the Splitter Blade .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

4.10.3. Looking at the Auxiliary Views .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1024.10.4. Exporting Geometry to ANSYS TurboGrid .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1044.10.5. Using ANSYS Meshing .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1054.10.6. Summary .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

5. Vista RTD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1075.1. Vista RTD Workflows .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1075.2. Data Review and Edit ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

5.2.1. Aerodynamics Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1105.2.2. Geometry Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

5.3. Viewing the Results ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1125.3.1. Results Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1125.3.2. Velocity Triangles Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

5.4. Context Menu Commands of the Blade Design Cell .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1145.5. Launching a New BladeGen Model ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1145.6. Creating a New Throughflow System ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1155.7. Launching a New BladeEditor Model ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1155.8. Linking to a New Vista TF Cell .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1165.9. Using Vista RTD to Model an Existing Turbine .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1165.10. Appendix 1: Definition of Parameters on Results Tab (Ideal and Semi-perfect Gas) ... . . . . . . . . . . . . . . . . . . . . . 116

6. Vista CCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1216.1. Vista CCD Workflows .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1216.2. Data Review and Edit ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

6.2.1. Duty and Aerodynamic Data Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1276.2.2. Gas Properties Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1296.2.3. Geometry Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

6.3. Viewing the Results ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1316.4. Common Options .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1346.5. Context Menu Commands of the Blade Design Cell .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1356.6. Launching a New BladeGen Model ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1366.7. Creating a New Throughflow System ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1366.8. Launching a New BladeEditor Model ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1376.9. Using Vista CCD to Model an Existing Compressor .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1386.10. Predicting a Performance Map .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1386.11. Appendix 1: Definition of Parameters on Results Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1406.12. Notation .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1436.13. References .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

7. Vista CPD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1457.1. Vista CPD Workflows .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1457.2. Vista CPD Interface Details ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

7.2.1. Global Controls ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1477.2.2. Graphics Display .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

7.2.2.1. Sketches .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1487.2.2.2. Efficiency Chart ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

7.2.3. Component Controls ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1507.2.3.1. Operating conditions Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

7.2.3.1.1. Units ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1507.2.3.1.2. Duty .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1507.2.3.1.3. Efficiencies .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

vRelease 14.5 - © SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information


TurboSystem

7.2.3.2. Geometry Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1537.2.3.2.1. Impeller Geometry .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

7.2.3.2.1.1. Hub Diameter .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1537.2.3.2.1.2. Leading Edge Blade Angles .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1547.2.3.2.1.3. Tip Diameter .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1557.2.3.2.1.4. Trailing Edge Blade Angles .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1567.2.3.2.1.5. Miscellaneous .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158

7.2.3.2.2. Volute Geometry .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1607.2.3.2.2.1. Casing rotation angle .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1627.2.3.2.2.2. Section Type .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1627.2.3.2.2.3. Diffuser ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

7.2.3.3. Results Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1637.2.3.3.1. Impeller Results ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

7.2.3.3.1.1. Overall Performance .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1637.2.3.3.1.2. Impeller Inlet ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1677.2.3.3.1.3. Impeller Exit ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

7.2.3.3.2. Volute Results ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1697.2.3.3.2.1. Key Dimensions .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1697.2.3.3.2.2. Sections, cutwater to throat .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1707.2.3.3.2.3. Diffuser ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

7.3. Context Menu Commands of the Blade Design Cell .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1717.4. Launching a New BladeGen Model ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1727.5. Creating a New Throughflow System ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1737.6. Launching a New BladeEditor Model ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1747.7. Creating a New Volute .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

8. Vista AFD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1778.1. Vista AFD Workflow .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

8.1.1. Meanline Calculation .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1798.1.2. Design (Throughflow) Calculation .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1808.1.3. Analysis (Throughflow) Calculation .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

8.2. Data Review and Edit ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1838.2.1. Aerodynamics Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1908.2.2. Geometry Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1918.2.3. Controls Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192

8.3. Results Tab .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1948.4. Context Menu Commands of the Cells in the Vista AFD System ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1958.5. Creating a Blade Design .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196

8.5.1. Launching a New BladeGen Model ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1978.5.2. Launching a New BladeEditor Model ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

8.6. Troubleshooting and Error Messages .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1978.7. Notation .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199

9. TurboGrid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20110. CFX-Pre . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20511. CFD-Post . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20712. Vista TF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

12.1. Vista TF User's Guide .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20912.1.1. Vista TF Setup Cell Properties ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21112.1.2. Customizing the Vista TF Template Files ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21712.1.3. Vista TF Context Menu Commands .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

12.2. Vista TF Reference Guide .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21912.2.1. Running Vista TF from the Command Line .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22012.2.2. Input and Output Data Files for Vista TF .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220

Release 14.5 - © SAS IP, Inc. All rights reserved. - Contains proprietary and confidential informationof ANSYS, Inc. and its subsidiaries and affiliates.vi

TurboSystem

12.2.2.1.The Auxiliary File with the Default Name: vista_tf.fil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22012.2.2.1.1. Backwards Compatibility and Cases without Real Gas Data .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221

12.2.2.2. Overview of Input Files ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22212.2.2.3. Overview of Output Files ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22412.2.2.4. Specification of the Control Data File (*.con) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22612.2.2.5. Specification of the Geometry Data File (*.geo) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23712.2.2.6. Specification of Aerodynamic Data File (*.aer) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24512.2.2.7. Specification of Correlations Data File (*.cor) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25512.2.2.8. Specification of the Real Gas Properties Data File (*.rgp) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26612.2.2.9. Specification of the Output Data File (*.out) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26712.2.2.10. Specification of the Text Data Files (*.txt) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27012.2.2.11. Specification of the CFD-Post Output Files (*.csv) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27312.2.2.12. Specification of Convergence History Data File (*.hst) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273

12.2.3. Software Limitations .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27412.2.4. Streamline Curvature Throughflow Theory .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275

12.2.4.1. The Equations .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27612.2.4.2.The Mean Stream Surface .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27612.2.4.3.The Grid .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27712.2.4.4. Ductflow and Throughflow .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27812.2.4.5. Iterative Solution Procedure .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27812.2.4.6. Initial Estimate .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27912.2.4.7. Radial Equilibrium Equation .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27912.2.4.8. Combination of Velocity Gradient and Continuity Equations .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27912.2.4.9. Relaxation Factors ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28012.2.4.10. Streamline curvature .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28012.2.4.11. Equations for Enthalpy and Angular Momentum ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28012.2.4.12. Boundary Conditions .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28112.2.4.13. Empirical Data .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28112.2.4.14. Blade-to-blade Solution .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28112.2.4.15. Spanwise Mixing .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28212.2.4.16. Streamline Curvature Throughflow Theory: Bibliography .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

12.2.5. Appendices .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28412.2.5.1. Appendix A: A Note on Sign Convention for Angles and Velocities in Vista TF .... . . . . . . . . . . . 284

12.2.5.1.1. Definition of Blade Lean Angles .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28912.2.5.1.2. Definition of Meridional Streamline Inclination Angle or Pitch Angle .... . . . . . . . . . . . . . . 29112.2.5.1.3. Definition of Blade Angle .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291

12.2.5.2. Appendix B: Example of a Control Data File (*.con) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29112.2.5.3. Appendix C: Example of a Geometry Data File (*.geo) for a Radial Impeller ... . . . . . . . . . . . . . . . . 29212.2.5.4. Appendix D: Example of an Aerodynamic Data File (*.aer) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29412.2.5.5. Appendix E: Examples of Correlations Data Files (*.cor) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29512.2.5.6. Appendix F: Troubleshooting .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300

12.2.5.6.1. Input-output Errors ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30112.2.5.6.2. Convergence .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30212.2.5.6.3. Reverse Flow .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30412.2.5.6.4. Choking .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30412.2.5.6.5. Computational Grid .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30612.2.5.6.6. Other Numerical Issues .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30712.2.5.6.7. Using real gas files ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

12.2.5.7. Appendix G: The RTZTtoGEO Program ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30912.2.5.8. Appendix H: Example of a Real Gas Property Data File (*.rgp) ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313

viiRelease 14.5 - © SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information


TurboSystem

Release 14.5 - © SAS IP, Inc. All rights reserved. - Contains proprietary and confidential informationof ANSYS, Inc. and its subsidiaries and affiliates.viii

Chapter 1: TurboSystem Introduction

TurboSystem is a set of software applications and software features for designing turbomachinery inthe ANSYS Workbench environment. It consists of the following components:

• ANSYS BladeGen: a geometry creation tool that is specialized for turbomachinery blades. It is available onWindows only. For details, see TurboSystem: ANSYS BladeGen (p. 19).

• Vista CCD, Vista CPD, Vista RTD, and Vista AFD are 1-D blade design tools. They are available on Windowsonly. For details, see TurboSystem: ANSYS BladeGen (p. 19).

• ANSYS DesignModeler: a general purpose geometry preparation tool that is integrated in ANSYS Workbench.This CAD-like program is primarily used to prepare CAD geometry models for analysis by other ANSYSWorkbench based tools. For details, see DesignModeler Help.

• ANSYS BladeEditor: a plugin for DesignModeler for creating blade geometry. ANSYS BladeEditor providesthe geometry link between BladeGen and DesignModeler, and therefore links BladeGen with other ANSYSWorkbench based applications. For details, see TurboSystem: ANSYS BladeEditor (p. 27).

• ANSYS TurboGrid: a meshing tool that is specialized for CFD analyses of turbomachinery blade rows. Fordetails, see TurboSystem: ANSYS TurboGrid (p. 201).

• ANSYS CFX-Pre: a general-purpose CFD preprocessor that has a turbomachinery setup wizard for facilitatingthe setup of turbomachinery CFD simulations. For details, see TurboSystem: ANSYS CFX-Pre (p. 205).

• ANSYS CFD-Post: a general-purpose CFD postprocessor that has features for facilitating the postprocessingof turbomachinery CFD simulations. For details, see TurboSystem: ANSYS CFD-Post (p. 207).

• Vista TF: a streamline curvature throughflow program for the analysis of turbomachinery. This programenables you to rapidly evaluate radial blade rows (pumps, compressors and turbines) at the early stagesof the design. For details, see TurboSystem: ANSYS Vista TF (p. 209).

For information about using TurboSystem in various workflows, see TurboSystem Workflows (p. 3).

1Release 14.5 - © SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information


Release 14.5 - © SAS IP, Inc. All rights reserved. - Contains proprietary and confidential informationof ANSYS, Inc. and its subsidiaries and affiliates.2

Chapter 2: TurboSystem Workflows

TurboSystem is a set of software applications and software features that help you to perform turboma-chinery analyses in ANSYS Workbench. The software applications and features are listed in TurboSystemIntroduction (p. 1). This chapter describes the primary ways of using TurboSystem.

The following topics are discussed:2.1. Geometry Sources2.2. Mesh Sources2.3. Solution Sources2.4. Examples of TurboSystem Workflows2.5. Usage Notes for Specific Workflows2.6. Using ANSYS Workbench Journaling and Scripting with TurboSystem2.7. Quick Pump Tutorial

2.1. Geometry Sources

The turbomachinery geometry is typically provided by one of the following cells:

• Blade Design cell of a BladeGen system

• Geometry cell of a Geometry or Mesh system

This can be based on an upstream Blade Design cell.

If you want to pass the geometry to the Turbo Mesh cell of a TurboGrid system, you must useBladeEditor to export to ANSYS TurboGrid.

2.2. Mesh Sources

The mesh is typically provided by one of the following cells:

• Turbo Mesh cell of a TurboGrid system

• Mesh cell of a Mesh system

2.3. Solution Sources

You can use one of the following systems to compute and report performance data:

• CFX component system

Use a CFX component system to perform a CFD analysis.

• Vista TF component system

Use a Vista TF component system to perform a throughflow analysis.



• Static Structural analysis system

Use a Static Structural analysis system to compute stresses and strains due to fluid forces and centri-fugal forces.

2.4. Examples of TurboSystem Workflows

This section shows examples of TurboSystem workflows, represented as connected systems in theProject Schematic view. Many other configurations are possible.

CFD analysis of a centrifugal pump:

CFD analysis of a turbine stage:

Throughflow analysis and CFD analysis of a turbine stage:


Workflows

2.5. Usage Notes for Specific Workflows

This section describes TurboSystem-related workflow issues and recommended practices:2.5.1.Tips on using ANSYS Workbench2.5.2. Using BladeGen2.5.3. Using BladeEditor2.5.4. Using ANSYS TurboGrid2.5.5. Using ANSYS CFX

2.5.1. Tips on using ANSYS Workbench

The following is a list of tips that you may find useful when working in ANSYS Workbench:

• Try right-clicking on different parts of the interface to see shortcut menus.

• You may find Compact Mode to be useful. Select View > Compact Mode from the ANSYS Workbench

menu or click to turn the Project Schematic view into a small, non-intrusive “title bar”that is always visible. To return to Full Mode, hover the mouse over the title bar, then, after the window

has expanded, click Restore Full Mode in the upper-right corner of the application window.

• Use the Files view to determine which files were created for each cell/system. It is easiest to find files as-sociated with a specific cell by sorting the view by Cell ID. This will sort the list by system and then bycell.



Usage Notes for Specific Workflows

• When selecting a system in the toolbox, ANSYS Workbench will highlight the cells in any systems alreadyin the Project Schematic view to which a valid connection can be made.

• Give unique meaningful names to all of your systems, especially if there are multiple systems of the sametype.

2.5.2. Using BladeGen

The following topic(s) are discussed:2.5.2.1. Changing the Active Document in BladeGen

2.5.2.1. Changing the Active Document in BladeGen

If you want to replace the active document in BladeGen, then:

1. Reset the corresponding Blade Design cell.

2. Edit the Blade Design cell or right-click it and select Import Existing Case.

Note that opening a subsequent .bgd file in the same instance of BladeGen will not replace the modelassociated with the Blade Design cell.

2.5.3. Using BladeEditor

The following topic(s) are discussed:2.5.3.1. Loading versus Importing a BladeGen Geometry2.5.3.2. Restarting a BladeEditor Session2.5.3.3. Modifying Spline Curves2.5.3.4. Adding a Hub Fillet to an Imported BladeGen Geometry2.5.3.5. Creating a Full 360-Degree Fluid Zone for an Impeller

2.5.3.1. Loading versus Importing a BladeGen Geometry

There are two basic ways of transferring a BladeGen geometry into BladeEditor

• Importing

If you import the geometry, the connection to the BladeGen geometry is maintained, and bladegeometry changes must be made by editing the upstream Blade Design cell.

• Loading

If you load the geometry, native BladeEditor features are created to represent the complete geometry,the connection to the BladeGen geometry is lost, and all geometry changes must be made by editingthe Geometry cell.

Linking from a Blade Design cell to a Geometry cell causes the BladeGen geometry to be imported intoBladeEditor. The desired import options should be set in the Blade Design cell properties. (See


Workflows

Table 3.1: BladeGen Blade Design Cell Properties (p. 20) for more information.) After you make the link,the Geometry cell should be updated to process the imported geometry.

Note

If you edit the Geometry cell before updating it, then the ImportBGD feature details that areshown in BladeEditor may not accurately reflect the Blade Design cell properties. To refresh

the ImportBGD feature properties, click in BladeEditor. It is not recommended thatyou edit the ImportBGD properties inside BladeEditor because they will be overwritten bythe properties from the Blade Design cell the next time you update the Geometry cell.

For details on importing BladeGen geometries, see Importing Blades from ANSYS BladeGen (p. 71).

For details on loading BladeGen geometries, see Loading and Modifying Blades from ANSYS Blade-Gen (p. 77).

2.5.3.2. Restarting a BladeEditor Session

If you want to clear a BladeEditor session while maintaining the link to the upstream cell, do not usethe Start Over command from the File menu in BladeEditor. Doing so would erase the incoming datafrom any upstream connections — notably when there is an upstream link to a BladeGen system. Instead,return to the Project Schematic view, right-click the Geometry cell and select Reset from the shortcutmenu. When you subsequently edit the Geometry cell, the upstream data is imported correctly.

2.5.3.3. Modifying Spline Curves

When a spline is initially created, it is uniquely defined by its fit points.

Spline control points (if exposed via the option when the spline is created or through the spline edit)can be modified using the normal drag operation. It is recommended that you use control points ratherthan fit points when manipulating splines. If you move any of the fit points, the spline may or may notbe uniquely-defined, depending on the way in which you move the points.

2.5.3.4. Adding a Hub Fillet to an Imported BladeGen Geometry

Assuming that you have already created the hub when you initially imported the BladeGen model, youcan add a hub fillet using the following procedure:

1. Edit the Geometry cell.

2. In BladeEditor/DesignModeler, in the feature tree, right-click the first feature that is listed below boththe hub and blade features, then select Insert > Fixed Radius from the shortcut menu.

This inserts a new Blend feature immediately above the feature that you right-click in the tree.

3. Select an edge along the intersection of a blade and the hub, for each edge that is to have a fillet of aspecified size.

You may need to use viewer toolbar icons to manipulate the view beforehand. If you cannot select

an edge, confirm that the selection filter is set for selecting edges (the toolbar icon).

4. Select Extend to Limits from the toolbar.



Usage Notes for Specific Workflows

This causes the selected edge to reach as far as possible around the blade.

5. Click Apply beside the Geometry property in the details view.

6. Set an appropriate value for the Radius property in the details view.

7. Click .

2.5.3.5. Creating a Full 360-Degree Fluid Zone for an Impeller

Assuming that you have not created the fluid zone when you initially imported the BladeGen model,you can create a fluid zone for a full, 360° impeller model using the following procedure:

1. View the Blade Design cell properties.

To do this, right-click the Blade Design cell and select Properties from the shortcut menu.

2. In the Properties view, ensure that Create All Blades is selected.

3. Update the Geometry cell if required.

4. Edit the Geometry cell.

5. Create a Revolve feature using the master profile, and revolve it around the axis for the full 360° togenerate the annulus volume.

6. Use an Enclosure feature to subtract the solid impeller from the annulus to generate the desired full360° fluid zone.

2.5.4. Using ANSYS TurboGrid

The following is a list of tips that you may find useful when working with ANSYS TurboGrid:

• Session files are primarily for use in standalone mode and batch mode. When using ANSYS TurboGrid inANSYS Workbench, session file playback is limited and unsupported. In particular, you must be cautiouswhen playing session files that involve file input/output. To play a session file when running in ANSYSWorkbench, open the Command Editor dialog box in ANSYS TurboGrid and process a command of theform: > readsession filename=[name] .

• ANSYS Workbench units and appearance options are not passed to ANSYS TurboGrid.

2.5.5. Using ANSYS CFX

The following topic(s) are discussed:2.5.5.1. Connecting from a Turbo Mesh Cell


Workflows

2.5.5.2. Changing the Geometry

Note

For more tips on using ANSYS CFX in ANSYS Workbench, refer to the ANSYS CFX in ANSYS

Workbench chapter of the ANSYS CFX online help, which is accessible from the Help menuin CFX-Pre.

2.5.5.1. Connecting from a Turbo Mesh Cell

It is recommended that you connect the Turbo Mesh cell of a TurboGrid system to the Setup cell of aCFX component system rather than to the Setup cell of a Fluid Flow (CFX) system.

2.5.5.2. Changing the Geometry

After setting up a turbomachinery simulation in a CFX component system, if you change topology ornumber of blades in the mesh, then refreshing or updating the CFX Setup cell (directly or indirectly)will fail to propagate the new information correctly.

If you generate a turbo report in CFD-Post (the latter accessed by editing the Results cell), then sub-sequently make changes that affect the upstream solution, and then update the Results cell, the resultingturbo report will use the updated values for any variables (such as pressure and velocity) and expressionsthat it uses, but the report will continue to use the old turbomachinery data (such as the number ofblades and the machine axis). After updating the solution, you will have to take one of the followingactions to update the turbo report:

• If you have no work that needs to be preserved with respect to the Results cell, then:

1. Reset the Results cell.

2. Edit the Results cell.

3. Load a turbo report.

• If you have done work in CFD-Post but have no work that needs to be preserved with respect to the turboreport, then:

1. Update the Results cell.

2. Reload the turbo report.

This is slower than the previously-mentioned method because the update process includes an un-wanted reprocessing of the old turbo report (with new results, but old turbomachinery data). However,because this method does not involve resetting the Results cell, any work you have done in CFD-Postis preserved (except for any modifications you have made to the original report).

2.6. Using ANSYS Workbench Journaling and Scripting with TurboSystem

Journaling is the capturing of ANSYS Workbench actions (creating a project, opening a system, and soon) to a file. For ANSYS CFX applications, CCL and command actions are embedded within ANSYSWorkbench actions. Scripting refers to the processes of editing and running a journal file in ANSYSWorkbench. With scripting, you could, for example, implement a prescribed workflow.



Using ANSYS Workbench Journaling and Scripting with TurboSystem

This section describes how to acquire, edit, and run script files that have commands that affectTurboSystem components. For more general information on journal files as well as scripting, refer toWorkbench Scripting Guide.

Note

• Journal actions such as a CFD-Post Export or the loading of a static .res file record the pathof the file. You may need to manually adjust this file path before attempting to rerun thejournal, particularly if you have created the journal using an unsaved project. More generally,when you create a project, you should save the project immediately to set file paths that ANSYSWorkbench uses (rather than require ANSYS Workbench to use file paths that have temporarydirectories, as happens before the project is saved).

• The handling of file paths described in File Path Handling in ANSYS Workbench in the Workbench

Scripting Guide applies to file references that are made outside of CCL and command actions.

• Workbench journal files for TurboGrid and TurboGrid session files from 12.0/12.1 may not workwith 13.0 and later releases. If the recorded file does not explicitly set the topology type, thefollowing CCL block needs to be added:

TOPOLOGY SET:ATM Topology Optimizer = offEND

For Workbench journal files, this CCL block should be added immediately before the fol-lowing line occurs:

> um mode=normal, object=/TOPOLOGY SET

The line above appears only in journal files that were recorded while the Topology Setobject was processed (unsuspended).

For session files, this CCL block should be added to the beginning of the session file.

2.6.1. Acquiring a Journal File with a TurboSystem Component in ANSYS

Workbench

Journaling is not available for any actions/operations made in either BladeGen or BladeEditor.

The basic workflow for acquiring a journal file with a TurboSystem-related system (for example, a Tur-boGrid system or a Vista TF system) is as follows:

1. Start ANSYS Workbench.

2. Start journaling: File > Scripting > Record Journal.

3. From Toolbox panel, open a TurboGrid or other TurboSystem-related system.

4. Create a TurboSystem-related system and then work with it (for example, create a TurboSystem-relatedmesh or set up a throughflow analysis). The actions you perform are captured by the journaling processand written to a .wbjn file).

5. Stop journaling: File > Scripting > Stop Recording Session.


Workflows

6. Optionally, edit the journal file (this is the process of scripting).

7. Run File > Scripting > Run Script File and select the .wbjn file.

2.6.1.1. Journal of an Operation that uses Vista TF

When you record a journal file of operations that involve Vista TF, the contents will be similar to thefollowing code snippets. In this example, a Vista TF system is added to an existing Geometry cell, thenthe Vista TF system is renamed, then a Vista TF setting is changed in the Setup cell properties:

Create a Vista TF system downstream of a Geometry cell

template2 = GetTemplate(TemplateName="Vista TF")component1 = system1.GetComponent(Name="Geometry")componentTemplate1 = GetComponentTemplate(Name="VistaTFSetupTemplate")system2 = template2.CreateSystem( DataTransferFrom=["FromComponent": component1, "TransferName": None, "ToComponentTemplate": componentTemplate1], Position="Right", RelativeTo=system1)

Rename the Vista TF system

system2.DisplayText = "Centrifugal Compressor Throughflow Analysis"

Change the Handedness of the Machine in the Setup cell of the Vista TF system

component2 = system2.GetComponent(Name="Setup")component2.Refresh()setup1 = system2.GetContainer(ComponentName="Setup")setupEntity1 = setup1.GetSetupEntity()setupEntity1.RotationalDirection = "LeftHanded"Update()

2.6.1.2. Journal of an Operation that uses ANSYS TurboGrid

When you record a journal file of an operation that uses ANSYS TurboGrid, the contents will be similarto the following code snippets. In this example, a mesh is created and the project is then saved.

Create the TurboGrid system

template1 = GetTemplate(TemplateName="TurboGrid")system1 = template1.CreateSystem()

Edit the Turbo Mesh cell and read a BladeGen file

turboMesh1 = system1.GetContainer(ComponentName="Turbo Mesh")turboMesh1.Edit()turboMesh1.SendCommand(Command=r"""VIEW:View 1 Camera Mode = User Specified CAMERA: Option = Pivot Point and Quaternion Pivot Point = 0, 0, 0 Scale = 1 Pan = 0, 0 Rotation Quaternion = 0.279848, -0.364705, -0.115917, 0.880476 Send To Viewer = False END

END

>readinf filename=C:\Program Files\ANSYS Inc\v121\TurboGrid\examples\rotor37\BladeGen.inf, guess=Off""")



Using ANSYS Workbench Journaling and Scripting with TurboSystem

Create a topology by unsuspending the Topology Set object

turboMesh1.SendCommand(Command="""VIEW:View 1 Camera Mode = User Specified CAMERA: Option = Pivot Point and Quaternion Pivot Point = 21.2526, -0.338768, 3.24 Scale = 0.0965047 Pan = 0, 0 Rotation Quaternion = 0.279848, -0.364705, -0.115917, 0.880476 Send To Viewer = False END

END

> um mode=normal, object=/TOPOLOGY SET""")

Create a mesh

turboMesh1.SendCommand(Command="""VIEW:View 1 Camera Mode = User Specified CAMERA: Option = Pivot Point and Quaternion Pivot Point = 21.2526, -0.338768, 3.24 Scale = 0.0965047 Pan = 0, 0 Rotation Quaternion = 0.279848, -0.364705, -0.115917, 0.880476 Send To Viewer = False END

END

> addlayer type=auto> mesh""")

Quit ANSYS TurboGrid

turboMesh1.Exit()

Save the Project file

Save( FilePath="C:/demo.wbpj", Overwrite=True)

In the above snippets, note how CCL and command actions for ANSYS TurboGrid are encapsulated asarguments of CFX.SendCommand instructions.

2.6.2. Scripting

Scripting refers to the processes of editing and running a journal file in ANSYS Workbench. You cancreate your own scripts and include the power of the Python programming language to implementhigh-level programming constructs for input, output, variables, and logic.

Full support for scripting is available for ANSYS TurboGrid. Scripting is not available for BladeGen.

2.6.2.1. Example: Using a Script to Change the Mesh Density in ANSYS TurboGrid

This example illustrates how a script can request and use user input. In this case, the mesh density iscontrolled based on user input:

system1 = GetSystem(Name="TS")

x = int(raw_input("Enter: 1=Medium-density mesh, 2=High-density mesh: "))

if x == 1:


Workflows

print 'Medium-density mesh' turboMesh1 = system1.GetContainer(ComponentName="Turbo Mesh") turboMesh1.Edit() turboMesh1.SendCommand(Command=r"""MESH DATA: Target Mesh Granularity = Medium END""") turboMesh1.SendCommand(Command=r"""> mesh""") turboMesh1.Exit()elif x == 2: print 'High-density mesh' turboMesh1 = system1.GetContainer(ComponentName="Turbo Mesh") turboMesh1.Edit() turboMesh1.SendCommand(Command=r"""MESH DATA: Target Mesh Granularity = Fine END""") turboMesh1.SendCommand(Command=r"""> mesh""") turboMesh1.Exit()

The script includes the CCL in the appropriate CFX.SendCommand argument to set the TargetMesh Granularity option in the MESH DATA object for either a medium-density mesh or a finemesh.

Before running this script, you would have to first open the Command Window dialog box (by selectingFile > Scripting > Open Command Window from the ANSYS Workbench main menu), and you wouldhave to have a TurboGrid system present, with ANSYS TurboGrid having a geometry already loadedand an unsuspended Topology Set object with suitable settings. To run the script, you would selectFile > Scripting > Run Script File from the ANSYS Workbench main menu and then use the browserto open the file containing the script. When the script is running, you input the value of x at the promptin the Command Window dialog box.

2.7. Quick Pump Tutorial

In this tutorial, you will quickly run through the steps required to simulate a water pump while usingTurboSystem.

2.7.1. Designing the Blade and Creating the Mesh

1. Open ANSYS Workbench and click File > Save As.

2. Save the project as QuickPump in a suitable directory.

3. From the Toolbox view, under Component Systems, drag a Vista CPD system onto the Project

Schematic view.

A new Vista CPD system appears in the Project Schematic view.

4. Double-click the Blade Design cell in the Vista CPD system.

Vista CPD appears.

5. Use the default settings for the pump specification and click Calculate.

A meridional sketch of the pump is shown alongside the calculated 1D performance data.

6. Click Close to exit Vista CPD and return to the Project Schematic view.

7. Right click the Blade Design cell and select Create New > BladeGen.

After a short time a new BladeGen system is created on the Project Schematic view.



Quick Pump Tutorial

8. Right click the Blade Design cell in the BladeGen system and select Transfer Data to New > TurboGrid.

A new TurboGrid system is created on the Project Schematic view, linked to the Blade Design

cell of the BladeGen system.

9. Double-click the Turbo Mesh cell in the TurboGrid system.

TurboGrid appears.

10. Edit Blade Set .

11. In the details view, select Trailing Edge Definition > Line of rotation on hub and shroud.

This makes it possible to split the hub at the trailing edge of the impeller so that the downstreampart of the hub can be stationary.

12. Click Apply.

13. Edit Mesh Data.

14. In the details view, set Size Factor to 0.7 .

This will generate a coarser mesh, which in turn will produce a faster CFD solution.

15. Click Apply.

16. Select File > Close TurboGrid to return to the Project Schematic view.

17. Right-click the Turbo Mesh cell and select Update to generate the mesh.

After the mesh has been generated, a green check mark appears on the Turbo Mesh cell.

18. Right-click the Turbo Mesh cell and select Transfer Data to New > CFX.

A new CFX system is created on the Project Schematic view.


Workflows

2.7.2. Setting up the Turbomachinery Simulation

You will use CFX-Pre in Turbo mode to set up physics and boundary conditions. In this case, you willspecify the same rotational rate and mass flow as specified in Vista CPD.

1. Double-click the Setup cell in the CFX system to open CFX-Pre.

2. Select Tools > Turbo Mode.

Basic Settings is displayed.

3. Accept the default settings in the Basic Settings panel by clicking Next.

4. In the Component Definition panel, select R1.

5. Set Component Type > Value to 1450 [rev min^-1] .

6. Click Next.

7. In the Physics Definition panel, configure the following options:

ValueSetting

WaterFluid

Shear Stress TransportModel Data > Turbulence

P-Total Inlet Mass Flow OutletInflow/Outflow BoundaryTemplates

0 [atm]Inflow > P-Total

Cylindrical ComponentsInflow > Flow Direction

Per MachineOutflow > Mass Flow

77.8 [kg s^-1]Outflow > Mass Flow Rate

(selected)Solver Parameters

Physical TimescaleSolver Parameters > Conver-gence Control

1e-2 [s]Solver Parameters > PhysicalTimescale

8. Click Next.

9. In the Interface Definition panel, note that periodic interfaces have been set up, then click Next.

10. In the Boundary Definition panel, modify the R1 Hub boundary:

1. Select R1 Hub

2. Click the multi-select from extended list icon.

This opens the Selection Dialog dialog box.

3. In the Selection Dialog dialog box, hold the Ctrl key and click each of Hub DOWNSTREAM andOUTBlock HUB to remove them from the selection.



Quick Pump Tutorial

4. Click OK.

This closes the Selection Dialog dialog box.

11. Add a new boundary:

1. Right-click Boundaries in the tree view and select Add Boundary....

2. Set the boundary name to R1 Hub Outlet and click OK.

3. Select the R1 Hub Outlet boundary and configure the following settings:

ValueSetting

WallR1 Hub Outlet > BoundaryType

HUB DOWNSTREAM,OUTBlock

HUB a

R1 Hub Outlet > Location

Counter Rotating WallWall Influence On Flow >Option

aUse the multi-select from extended list icon to select multiple locations.

12. Click Next.

Final Operations is displayed.

13. Ensure that Operation is set to Enter General Mode .

14. Click Finish.

15. In the Outline tree view, edit R1 Outlet .

16. In the details view, configure the following settings:

ValueSettingTab

(Selected)Mass And Momentum > MassFlow Update

Boundary De-tails

Shift PressureMass And Momentum > MassFlow Update > Option

17. Click OK.

18. Select File > Close CFX-Pre to return to the Project Schematic view.

19. Double-click the Solution cell in the CFX system to launch CFX-Solver Manager.

20. In the Define Run dialog box, click Start Run to begin the CFD calculation.

The residual monitors show the progress of the calculation, which ends after about 44 iterationloops.


Workflows

21. After the solution has been generated, click OK to dismiss the completion message, then select File >Close CFX-Solver Manager to return to the Project Schematic view.

2.7.3. Viewing the Turbo Report

To see a report for the pump performance:

1. Double-click the Results cell in the CFX system to open CFD-Post.

2. Select File > Report > Load 'Pump Impeller Report' Template.

3. After the report has been generated, select the Report Viewer tab at the bottom of the graphics windowto display the report.

The report includes both numerical and graphical results.

Note that if you have visited the Report Viewer tab before loading the template, or have otherwise

made any changes to the report definition after first viewing the report, you need to click inthe Report Viewer tab to update the report as displayed.



Quick Pump Tutorial


Chapter 3: TurboSystem: ANSYS BladeGen

ANSYS BladeGen is a geometry creation tool that is specialized for turbomachinery blades. BladeGenhas its own documentation that can be accessed through the user interface, or by browsing the install-ation directory. The main documentation, “ANSYS BladeGen User's Guide”, is available from the Helpmenu in BladeGen. There, you will find general help and tutorials.

Other Vista programs are available from Workbench, under the Component Systems and are documentedin the Workbench help:

• See Vista AFD for details on using Vista AFD (for axial fans).

• See TurboSystem: Vista CCD for details on using Vista CCD (for centrifugal compressors).

• See TurboSystem: Vista RTD for details on using Vista RTD (for radial inflow turbines).

Note

ANSYS BladeGen and the Vista programs mentioned above are available only on Windows.

Note

Another Vista program, Vista TF, is for throughflow analyses in turbomachinery. It has a ref-erence guide named "Vista TF Reference Guide" that is in this set of documentation. You canaccess it by clicking the following link: Vista TF Reference Guide (p. 219).

To launch BladeGen from ANSYS Workbench, add the BladeGen component system to your projectschematic, then edit the Blade Design cell of that system.

The Blade Design cell has properties that need to be configured in order to transfer the blade geometryfrom BladeGen to BladeEditor. This transfer is represented by a link that connects a Blade Design cellto a Geometry cell.

A sample of the cell properties is shown in Figure 3.1: Properties of the BladeGen Blade Design Cell (p. 20).



Figure 3.1: Properties of the BladeGen Blade Design Cell

Table 3.1: BladeGen Blade Design Cell Properties (p. 20) describes each of the cell properties. These arerelated to the properties of the ImportBGD feature in BladeEditor (described at Table 4.3: Properties forthe ImportBGD Feature (p. 72)).

Table 3.1: BladeGen Blade Design Cell Properties

DescriptionNameGroup

If this property is selected, thenBladeEditor will create a HubPro-

Create Hub*Import Op-tions

file sketch for the non-flow pathhub geometry, and will create arevolved body feature called Hub-Body .

If this property is selected, thenBladeEditor will create all the

Create AllBlades

blades using the number of bladesspecified in the BladeGen model.

If this property is not selected, thenonly the first blade will be created.

If this property is not selected, thenBladeEditor will create the blade

Merge BladeTopology

with four faces corresponding tothe leading edge, pressure side,trailing edge and suction side. Thiscan make it easier to create astructural mesh for the blades inthe Mechanical application.

If this property is selected, then theblade faces will be merged wherethey are tangent to one another.


BladeGen


If this property is set to Stream-wise , then BladeEditor will loft the

Blade LoftDirection

blade surfaces in the streamwisedirection through curves that runfrom hub to shroud. This is the de-fault because the surface is morewell defined, especially for flank-milled blades.

If this property is set to Spanwise ,then BladeEditor will loft the bladesurfaces in the spanwise directionthrough the blade profile curves.

For an illustration of these loftingmethods, see the figure after thistable.

This property specifies whether ashroud clearance is created. If

Shroud Clear-ance

None is selected, then no shroudclearance is created. To create ashroud clearance, choose eitherRelative Layer or AbsoluteLayer . The blade(s) will betrimmed off at the selected Blade-Gen output layer, and the layercontour will be created in theLayerProfile sketch.

If Relative Layer is selected,then the selected Layer Number

is relative to the shroud layer, e.g.,1 implies the first layer closest tothe shroud layer, 2 implies thesecond closest layer to the shroud,etc.

If Absolute Layer is selected,then the selected layer indexcounts up from the hub layer,which is zero.

If this property is selected, thenBladeEditor will create a Stage-

Create FluidZone*

FluidZone body for the flowpassage, and an Enclosure featureto subtract the blade body. Theresulting Enclosure can be used fora CFD analysis of the blade pas-sage.




If this property is selected, thenBladeEditor will create NamedSe-

CreateNamed Selec-tions* lections (regions) for the typical

faces of the blade passage, i.e.,Blade , Hub, Shroud , Inflow ,Outflow , PeriodicA andPeriodicB . These NamedSelec-tions can be used as selectiongroups in other ANSYS Workbenchapplications.

Note that this property is availableonly if Create Fluid Zone is selec-ted.

This property defines the surfaceextension length (as a percentage

Blade Exten-sion (%)

of the average hub to shroud dis-tance) for the blade surfaces. Thesesurfaces are extended and thentrimmed to the MasterProfilesketch to ensure that the bladesolid correctly matches the hub andshroud contours.

This property defines the surfaceextension length (as a percentage

Periodic Sur-face Exten-sion (%) of the average hub to shroud dis-

tance) for the periodic surfaces.These surfaces are extended to en-sure that the StageFluidZoneis properly cut.

This property specifies the style ofthe periodic interface surfaces.

Periodic Sur-face Style

If Three Pieces is selected, thenthe periodic surface is created inthree connected pieces: one up-stream of the blade, one within thepassage, and one downstream ofthe blade. This style can better ac-commodate highly curved or twis-ted blades, and is similar to theANSYS TurboGrid style of periodicsurface.

If One Piece is selected, then theperiodic surface is created as asingle surface.


BladeGen


Note that this property is availableonly if Create Fluid Zone is set toYes .

* These properties will only apply when you initially import the model. Any subsequent changes tothese properties will not be propagated to downstream cells upon updating the latter.

Figure 3.2: Spanwise Lofting versus Streamwise Lofting (p. 23) shows how spanwise lofting andstreamwise lofting differ.

Figure 3.2: Spanwise Lofting versus Streamwise Lofting

Once the ImportBGD feature has been created (for example, by updating the Geometry cell), changingthe following Blade Design cell properties will have no effect:

• Create Hub



• Create Fluid Zone

• Create Named Selections

Note

In order to facilitate the use of a BladeGen model in another application (such as BladeEditoror ANSYS TurboGrid), you should always ensure that the model units are set. When you createa new BladeGen model (File > New > BladeGen Model), the default units are always “Un-known ” (regardless of ANSYS Workbench preferences). In this case, select Model > Properties

from the BladeGen main menu and set Model Units in the Model Property Dialog.

You can access a context menu for the Blade Design cell in the BladeGen component system by right-clicking the cell. Most of the commands that are available are standard, and are described in Systemsand Cells. The context menu commands that are specific to the Blade Design cell are described inTable 3.2: Context Menu Commands Specific to the BladeGen Blade Design Cell (p. 24).

Table 3.2: Context Menu Commands Specific to the BladeGen Blade Design Cell

DescriptionCommand

This command creates a linkeda Mesh systemand generates a mesh for the associated

Create New Blade CFDMesh

BladeGen model. This command is availableonly if the cell is up to date and the geometryhas been saved (either by saving it from withinBladeGen or by closing the BladeGen window).

Also see Tips on Using the Create New BladeCFD Mesh Command (p. 25).

This command opens BladeGen. If the cell isup to date, then BladeGen will load the associ-ated BladeGen model.

Edit

This command opens a file browser for select-ing a .bgd file. This command is only available

Import Existing Case

if the cell is in the Edit Required state. If thiscommand is not available due to havingalready selected a .bgd file, and you want tomake the command available so that you canselect a different .bgd file, then choose theReset command in the context menu.

This command creates a Fluid Flow (CFX) sys-tem and links it to the BladeGen system.

Transfer Data to New >Fluid Flow (CFX)

This command creates a Geometry system andlinks it to the BladeGen system.

Transfer Data to New >Geometry

This command creates a Mesh system and linksit to the BladeGen system.

Transfer Data to New >Mesh

This command is not recommended becausethe blade data is not suitable for export toVista TF.

Transfer Data to New >Throughflow


BladeGen

DescriptionCommand

This command creates a TurboGrid system andlinks it to the BladeGen system.

Transfer Data to New >TurboGrid

aNote that the link between systems causes the geometry to be imported, rather than loaded, into BladeEditor. The blade will bedependent on the BladeGen BGD file, but you can still modify the HubProfile sketch that appears under the associated MerPlanefeature. For more information on importing and loading geometry into BladeEditor, see Importing Blades from ANSYS BladeGen inthe TurboSystem User Guide and Loading and Modifying Blades from ANSYS BladeGen in the TurboSystem User Guide.

3.1. Tips on Using the Create New Blade CFD Mesh Command

• Visual Expansion is on by default. This displays a full 360° mesh in ANSYS Meshing for visualization pur-poses. To display only the actual single passage mesh, clear Visual Expansion in the View menu of ANSYSMeshing.

• If the mesh fails to be generated, try the following procedure to adjust the fineness of the mesh:

1. Edit the Mesh cell of the Mesh system.

The ANSYS Meshing window appears.

2. In the tree view, select Mesh.

3. In the details view, under the Defaults group, change the Relevance option.

This option affects the values set for some of the controls under the Sizing group. You couldalternatively adjust those controls directly, for finer control.

For details on the Relevance option, see Relevance in the Meshing User's Guide.

• If the geometry has a small radius on the leading edge or trailing edge, the resulting mesh might be finerthan wanted. In this case, you can try increasing the minimum element size manually by, say, a factor of4. For details on the Min Size control, see Min Size in the Meshing User's Guide.

• If proximity and curvature are not an issue (for example, if the hub and shroud are not in close proximity,nor are adjacent blades), you can set the Use Advanced Size Function option to Off. This will cause themesh to be generated faster; however, mesh generation will be less robust (more likely to fail).

For details on the Use Advanced Size Function option, see Use Advanced Size Function in theMeshing User's Guide.



Tips on Using the Create New Blade CFD Mesh Command


Chapter 4: TurboSystem: ANSYS BladeEditor

ANSYS BladeEditor is a plugin for ANSYS DesignModeler for creating, importing, and editing bladegeometry. Using BladeEditor, you can create a blade from scratch. You can also import a blade fromANSYS BladeGen. Once you have a blade in DesignModeler, you can use it as you would any geometrycreated in DesignModeler. In addition, you can use BladeEditor to export the geometry for use in ANSYSTurboGrid (a meshing tool) and Vista TF (a throughflow analysis tool).

The following topics are discussed:4.1. Configuring the ANSYS BladeModeler License4.2.The BladeEditor User Interface4.3. Blade Editing Features4.4. Blade Comparison4.5. Importing Blades from ANSYS BladeGen4.6. Loading and Modifying Blades from ANSYS BladeGen4.7. Using and Exporting Blades4.8. Blade Parameterization4.9.Tutorial 1: Blade Editing With Emphasis On Sketches, Layers, and Blade Comparison4.10.Tutorial 2: Blade Editing With Emphasis On Camberline and Thickness Distributions

4.1. Configuring the ANSYS BladeModeler License

BladeModeler requires an ANSYS license to use, although BladeGen will run in "demonstration" modewithout the license (no saving or exporting is possible in this mode). Without the license the BladeGensystem does not appear in the Toolbox, and the only way to access BladeGen is to open a previouslysaved project that contains a BladeGen system. Please contact your ANSYS representative to obtain alicense for BladeModeler if you do not already have one.

One BladeModeler license will permit a single user to have BladeGen, DesignModeler, and BladeEditorrunning together in the same ANSYS Workbench session. This license sharing ability means that noadditional DesignModeler license will be required to use the full functionality of BladeModeler.

In order to use BladeEditor, you must set the Geometry license preference to ANSYS BladeModeler asfollows:

1. In the ANSYS Workbench menu, select Tools > License Preferences.

2. In the License Preferences dialog box, click the Geometry tab.

3. If ANSYS BladeModeler is not the first license listed, then select it and click Move up as required tomove it to the top of the list. Furthermore, you should select ANSYS DesignModeler in the list and setits value to 0 (which means “Don't Use”). This prevents DesignModeler from using an ANSYS Design-

Modeler license when an ANSYS BladeModeler license is not available.

If ANSYS BladeModeler is not in the list then you need to obtain an ANSYS BladeModeler license.

4. Click OK to close the dialog box.



Click the Help button in the License Preferences dialog box for more information.

Note

A BladeEditor model cannot be processed by DesignModeler under the ANSYS DesignModel-

er license. If you do edit a Geometry cell that contains BladeEditor feature parameters, andyou are using DesignModeler under the ANSYS DesignModeler license, then do the followingto prevent your model from becoming corrupted:

1. Close your project (by selecting File > New from the main menu in Workbench) WITHOUTsaving it.

2. Configure your license preferences as specified above.

3. Open your project.

4. Edit the Geometry cell and confirm that DesignModeler runs as BladeEditor (with BladeEditor-related icons appearing in the tool bar).

4.2. The BladeEditor User Interface

The BladeEditor user interface extends the DesignModeler user interface in the following ways:

• There are new feature types aimed at creating a blade and flow passage.

• There is a set of toolbar icons, most of which are used to create the new feature types.

• There are new views associated with some of the feature types.

• There are new context menu commands associated with the new views.

The following topics will be discussed:4.2.1.Tree View and Details View4.2.2. Contour Sketch Management4.2.3. BladeEditor Toolbars4.2.4. Auxiliary Views4.2.5. Angle and Thickness Views4.2.6. Section Definition and Stacking Views4.2.7. User Preferences and Properties

4.2.1. Tree View and Details View

As with DesignModeler features, the primary user control of the new blade geometry features is throughthe tree view and details view. The tree view shows the features in the current model, and providesmechanisms for inserting, editing, and deleting features. The tree view also shows the order of operationsin which DesignModeler will generate the model (top to bottom), and the feature dependencies. Fig-ure 4.1: Tree View (p. 29) shows an example of the tree view diagram with the sketches (HubContour1 ,ShroudContour1 , and so on), FlowPath (FlowPath1 ), and Blade (Blade1 ) features for two bladerows.


BladeEditor

Figure 4.1: Tree View

You might want to temporarily suppress various features to speed up the regeneration time. This canbe managed in DesignModeler by right-clicking the feature in the tree view and selecting “Suppress”.Note that suppressing a feature also suppresses all dependent features below it. Sketches cannot besuppressed; they can only be hidden or deleted.

When you click a contour sketch in the tree view, the sketch will be highlighted in the 3D viewer. Whenyou click a camberline definition, the corresponding FlowPath layer contour will be highlighted in the3D viewer, and the angle and thickness views will be displayed for this definition. Any changes madeto the sketches or camberline definitions will be immediate. However, updates to the dependentFlowPath or Blade features will happen only when you subsequently click the “Generate” button.

4.2.2. Contour Sketch Management

The contour sketches are managed just as normal sketches are managed in DesignModeler. To modifya meridional contour, you should first select the sketch in the tree view and then click the “Sketching”tab to open the sketch toolbox. If you already have the sketch toolbox open, you can select a differentactive sketch from the toolbar drop-down list. For models with many different sketches, and forswitching between editing one contour or another, this could be tedious. For this reason, it is possibleto modify or delete existing edges in a sketch that is not the active sketch as long as it belongs to thesame plane as the active sketch. However, newly created edges are added only to the active sketch, soit is imperative that you have the active sketch selected appropriately when creating new edges.

4.2.3. BladeEditor Toolbars

BladeEditor utilizes three separate toolbars: one for feature creation, one for selection, and one for displaycontrol.

The following topics are discussed4.2.3.1. Feature Creation Toolbar4.2.3.2. Active Selection Toolbar



The BladeEditor User Interface

4.2.3.3. Display Control Toolbar

4.2.3.1. Feature Creation Toolbar

The feature creation toolbar has icons for the BladeEditor specific features that can be added to themodel (and that appear in the tree view).

The following icons exist in this toolbar:

DescriptionIcon

Creates an ImportBGD fea-ture, which is used to import

Import BGD

a BladeGen file and constructa solid model of the bladeand optionally the hub andfluid zone. The flow path andblade shape cannot be editedin BladeEditor, but the importcan be refreshed if there areany changes to the BladeGenfile. For details, see ImportingBlades from ANSYS Blade-Gen (p. 71).

Loads a BladeGen file andconstructs the FlowPath,

Load BGD

Blade, and Splitter featuresso that the flow path andblade shape may be editedin BladeEditor. For details, seeLoading and ModifyingBlades from ANSYS Blade-Gen (p. 77).

Creates a flow path(FlowPath) feature, which is

FlowPath

used to define the flow pathfor blade geometry (that is,the meridional shape of thepassage). For details, seeFlowPath Feature (p. 48).

Creates a blade (Blade) fea-ture, which is used to create

Blade

the blade bodies. For details,see Blade Feature (p. 51).

Creates a splitter (Splitter)feature, which is used to cre-

Splitter

ate the splitter bodies. Fordetails, see Splitter Fea-ture (p. 64).

Creates a Vista TF export(VistaTFExport) feature for

VistaTFExport


BladeEditor

DescriptionIcon

exporting the flow path andblade geometry to the VistaTF through flow analysis tool.For details, see Export toVista TF (.geo) (p. 78).

Creates an ExportPoints fea-ture for exporting blade point

ExportPoints

data to TurboGrid or to ameanline file. For details, seeExport to ANSYS Tur-boGrid (p. 81) and Export asMeanline Data (.rtztfile) (p. 79).

Creates a stage fluid zone(StageFluidZone) feature for

StageFluid-Zone

generating the blade passagebodies. The stage fluid zoneis a 3D fluid region to sup-port CFD analyses. For details,see Stage Fluid Zone Fea-ture (p. 67).

Creates a throat area(ThroatArea) feature for calcu-

ThroatArea

lating the blade throat area.For details, see Throat AreaFeature (p. 68).

Opens the BladeEditor prefer-ences in the details view. For

Preferences

details, see User Preferencesand Properties (p. 45).

Note

Note that, because of limitations in the FlowPath, and Blade features, the ‘Load BGD’ featuresupports loading BladeGen files with the following restrictions, as applied to BladeGen settings:

• Model

Only the ‘Angle/Thickness’ mode is supported.

• If loading multiple BladeGen files, the Beta Definition, ‘from Axial’ or ‘from Tangential’, mustbe consistent for all files.

• Layers

Only specified span fraction layers and data layers are supported.

The Spanwise Calculation mode must be ‘Geometric’.

• Angle Definition




Only Theta and/or Beta definitions are supported, not End Angle.

Only 'General' and 'Ruled Element' spanwise distributions are supported.

At least one angle definition must exist on either the hub or shroud layer.

The Angle View Data Location must be 'Meanline'.

After loading a .bgd file, you cannot further split or join curve segments.

• Thickness Definition

Only the 'Normal to Meanline on Layer Surface' thickness data type is supported.

The 'vs. Cam' and '% Cam vs. % Cam' thickness specifications are not supported.

Only the General spanwise distribution is supported

At least one thickness definition must exist on either the hub or shroud layer.

• Meridional Profile

Only the 'Design Profile' is supported.

After loading a .bgd file, you cannot further split or join curve segments.

Note

When you load a .bgd file that specifies Angle/Thickness data from trailing edge to leadingedge, the geometry will not be loaded correctly (will not appear exactly as in BladeGen). Ifyou load such a .bgd file, a warning message will appear. You can change the data directionby going into BladeGen and selecting Model > Ang/Thk Data Direction > Data from LE

to TE.

4.2.3.2. Active Selection Toolbar

You can select the Blade and layer used in the angle and thickness views from a pair of drop-down listboxes on the toolbar, beside the label Select Layer:. The first box indicates the active Blade feature;the second box indicates the CamThkDef sub-feature.

4.2.3.3. Display Control Toolbar

The display control toolbar enables you to control display properties such as the angle/thickness graphvisibility and graph view actions such as “zoom fit”, etc.

4.2.4. Auxiliary Views

The Auxiliary View provides a variety of formats for reviewing the blade design. It is used for data displayonly.

The following topics are discussed:4.2.4.1. Blade-to-Blade View


BladeEditor

4.2.4.2. Blade Lean Graph4.2.4.3. Curvature View4.2.4.4. Meridional Curvature View

4.2.4.1. Blade-to-Blade View

When using Camberline/Thickness design mode (enabled via the Blade Design Mode setting of theblade object), the blade-to-blade view is available. This view shows the blade profile for the selectedCamThkDef sub-feature in either of the following coordinate systems:

• M-Prime vs. Theta

• M vs. R*Theta

Note

Profiles are shown for all blades in the selected blade row.

To show the blade-to-blade view:

1. Select the CamThkDef sub-feature for which you want the blade-to-blade view. (The CamThkDef sub-feature you select must be a sub-feature of a Blade.)

2. If you do not see the Auxiliary view, click Show/Hide Auxiliary Pane .

3. If the Auxiliary view shows a plot other than blade-to-blade, right-click in the Auxiliary view and selectBlade-to-blade.

Within the blade-to-blade view, you can use the context menu to:

• Change the type of plot.




• Toggle the coordinate system between “M-Prime vs. Theta” and “M vs. R*Theta”.

• Zoom to fit the view

4.2.4.2. Blade Lean Graph

When using Camberline/Thickness design mode (enabled via the Blade Design Mode setting of theblade object), the blade lean graph is available. This graph shows a plot of the blade lean angle versusmeridional coordinate (%M). The blade lean angle is essentially the angle between a constant-theta lineand a straight line connecting the hub and shroud camberlines, as viewed in a plane of constant meri-dional coordinate.

To show the blade lean graph:

1. Select the blade (or any CamThkDef sub-feature of it) for which you want the blade lean graph.


3. If the Auxiliary view shows a plot other than blade lean angle, right-click in the Auxiliary view and selectBlade Lean Angle.

Within the blade lean graph, you can use the context menu to:


• Zoom to fit the view


BladeEditor

4.2.4.3. Curvature View

When using Airfoil design mode (enabled via the Blade Design Mode setting of the blade object), theCurvature view is available. This view shows a plot of either the curvature or the radius of curvature ofthe outline of the airfoil section versus percent arc length. Curvature is evaluated in the applicable layer(the layer containing the blade section). Arc length is measured along the outline of the airfoil section,starting at the major axis of the leading edge ellipse and ending at the major axis of the trailing edgeellipse.

To show the Curvature view:

1. Select the any Blade Section sub-feature for which you want the Curvature view.

2. If you do not see the Curvature view at this point, click Show/Hide Auxiliary Pane .

Within the Curvature view, you can use the context menu to:

• Toggle display of the curvature/radius of curvature data for the high-theta side only.

• Toggle display of the curvature/radius of curvature data for the low-theta side only.




• Toggle between viewing curvature or radius of curvature.

• Zoom to fit the view.

4.2.4.4. Meridional Curvature View

This view shows a plot of meridional curvature of the hub/shroud versus either meridional, or normalizedmeridional, coordinate. Meridional curvature is computed as the inverse of the radius of curvature inthe axial-radial plane. For a plot that uses normalized meridional coordinate, normalization is carriedout such that the coordinate ranges from 0 to 100 over the plot. The portions of the hub and shroudcurves that are plotted depend on the selected feature: a FlowPath feature or a Blade feature.

To show the Meridional Curvature view:

1. Select either a FlowPath feature or a Blade feature.


3. If the Auxiliary view shows a plot other than meridional curvature, right-click in the Auxiliary view andselect Meridional Curvature.

Within the Meridional Curvature view, you can use the context menu to:


• Toggle display of the hub curvature data.

• Toggle display of the shroud curvature data.

• Change the X-Axis Scale between meridional coordinate and normalized meridional coordinate.

• Zoom to fit the view.


BladeEditor

When you click a line in a plot, the details view shows the following coordinates of the point you clicked:

• Z (axial coordinate)

• R (radial coordinate)

• M (meridional coordinate)

• % M (normalized meridional coordinate)

• Meridional curvature

4.2.5. Angle and Thickness Views

When using Camberline/Thickness design mode (enabled via the Blade Design Mode setting of theblade object) the Angle and Thickness views are available to enable you to see and modify the Camber-line/Thickness definitions. Both views present the data for the active CamThkDef sub-feature. See ActiveSelection Toolbar (p. 32) for more details on how the active CamThkDef sub-feature is selected.

If the Blade feature or any of its parent features have been modified, the curves shown in theAngle/Thickness views, when displayed under the Blade feature, are updated when the model is regen-erated.

The following topics are discussed:4.2.5.1. Angle View4.2.5.2.Thickness View

4.2.5.1. Angle View

The Angle view displays a graph of the Theta and/or Beta angle definition for the selected CamThkDefsub-feature.

To show the Angle view:

1. Select the blade for which you want the Angle view.

2. If you do not see the Angle view, click Show/Hide Angle Pane .

The Angle view will show a Beta curve (colored cyan) or a Theta curve (colored dark blue) — whicheveris specified in the properties of the selected CamThkDef sub-feature. You can add the other curve tothe Angle view using the context menu in the Angle view.

Within the Angle view, you can:

• Access a shortcut menu by right-clicking the mouse.

• Drag (using the mouse) the control points defining the Theta/Beta curve.

• Change the curve type: Bezier, cubic spline, piecewise-linear (context menu)

• Insert or delete points (context menu).

• Convert the selected curve segment to a Bezier or spline curve of the specified order (context menu).

• Set control point coordinates (double-click a control point and edit the coordinate values in a dialog box).




• View the coordinates of any location on a curve (hold the Ctrl key and left click any control point or linesegment of a curve).

• Pan the view (right-click and drag).

• Zoom to fit the view (context menu).

• Zoom in and out (middle-click and drag).

• Zoom in via a zoom box (hold the Alt key and left-click and drag from the top-left corner to the bottom-right corner of a rectangular region; drag in the opposite direction to re-fit the view).

• Change the x-axis display type: m, m’, %m, or %m’ (context menu).

• Read angle data points from a file or save points to a file (context menu). In a crude way, this enablescopying and pasting data from one CamThkDef sub-feature to another, or to and from an external applic-ation, for example a spreadsheet.

• Use as Input Parameter (context menu)

After right-clicking in the view and selecting Use as Input Parameter from the context menu, click apoint on the curve. The settings for that point will appear in the properties of the relevant CamThkDeffeature; each property has a check box that can be selected to cause the property to become andinput parameter.

• Show all defining Theta/Beta curves in the current view (context menu).

• Show/hide the second angle curve (the Beta or Theta curve — whichever is not specified for the view inthe properties of the selected CamThkDef sub-feature) (context menu).

Note

When the meridional contours are changed, the angle data points are scaled relative to thepercent m’-coordinate (% m-prime).

Note

X-axis value behavior: preference used as default. Change of x-axis value will affect all layersviewed for a given Blade feature.

Note

Some context menu commands are not available whenever at least one control point is eligibleto be selected as an input parameter due to having been processed by the Use as Input

Parameter command.

4.2.5.2. Thickness View

The Thickness view displays a graph of the thickness definition for the selected CamThkDef sub-feature.

To show the Thickness view:


BladeEditor

1. Select the blade for which you want the Thickness view.

2. If you do not see the Thickness view, click Show/Hide Thickness Pane .

Within the Thickness view, you can:


• Drag the points defining the thickness curve (mouse).

• Change the curve type: Bezier, cubic spline, piecewise-linear (context menu).

• Insert or delete points (context menu).



• View the coordinates of any location on a curve (hold the Ctrl key and left click any control point or linesegment of a curve).


• Zoom to fit the view (context menu).



• Change the x-axis display type: m, m’, %m or %m’ (context menu).

• Read thickness data points from a file or save points to a file (context menu). In a crude way, this enablescopying and pasting data from one CamThkDef sub-feature to another, or to and from an external applic-ation, for example a spreadsheet.

• Use as Input Parameter (context menu)

After right-clicking in the view and selecting Use as Input Parameter from the context menu, click apoint on the curve. The settings for that point will appear in the properties of the relevant CamThkDeffeature; each property has a check box that can be selected to cause the property to become andinput parameter.




• Show all defining thickness curves in the current view (context menu).

Note

When the meridional contours are changed, the thickness data points are scaled relative tothe percent m-coordinate (% m).

Note

X-axis value behavior: preference used as default. Change of x-axis value will affect all layersviewed for a given Blade feature.

Note


Parameter command.

4.2.6. Section Definition and Stacking Views

When using Airfoil design mode (enabled via the Blade Design Mode setting of the blade object) theSection Definition and Section Stacking views are available to enable you to see and modify the airfoilshapes and their theta distribution across (either spanwise or user-defined) layers. These views havebehavior similar to the Angle and Thickness views (see Angle and Thickness Views (p. 37)). Both viewspresent the data for the active Blade Section sub-feature.

If the Blade feature or any of its parent features have been modified, the curves shown in the SectionDefinition/Stacking views, when displayed under the Blade feature, are updated when the model is re-generated.

The following topics are discussed:4.2.6.1. Section Definition View4.2.6.2. Section Stacking View

4.2.6.1. Section Definition View

The Section Definition view displays, and enables you to adjust, the airfoil shape for each Blade Sectionsub-feature.

To show the Section Definition view:

1. Select the Blade Section sub-feature (under the Blade feature) for which you want the Section Definitionview.

2. If you do not see the Section Definition view, click Show/Hide section shape .

The blade section shape is displayed in the Section View (Figure 4.2: Blade Section View (p. 41)). Theblade section shape can be modified by changing the section dimensions in the details view, or bydragging the Bezier curve control points in the Section View. The Section View also displays the:

• pitch line


BladeEditor

• adjacent blade position

• specified throat line

• calculated throat (throat with the minimum throat length) and its midpoint

• airfoil centroid

• both ellipses, their axes and centers

• stagger line

Any of the defining dimensions can be parameterized by clicking in the check box next to the dimension.

Figure 4.2: Blade Section View

Within the Section Definition view, you can:


• Drag the Bezier curve control points defining the pressure and suction side edges of the airfoil shape (left-click a control point and drag it with the mouse).




• Insert a point on a Bezier curve (right-click in view to insert the new point).

• Delete a point from a Bezier curve (right-click the point to be deleted; only interior points may be deleted).


• Reset the airfoil input parameters and control points to their original values (context menu)

• Toggle the display of the blade pitch spacing line (context menu)

• Toggle the display of the specified throat width (context menu)

• Toggle the display of the calculated throat width and its midpoint (context menu)

• Toggle the display of the airfoil centroid location (context menu)

• Toggle the display of the LE ellipse (context menu)

• Toggle the display of the TE ellipse (context menu)

• Toggle the display of the stagger line (context menu)

• Toggle the display of a point on the airfoil surface that corresponds to the point currently selected in thecurvature view (context menu)



• Zoom to fit the view (context menu)


Note

Toggling the display options will make them take effect for the remainder of the session,but not from session to session.

You can see the curvature of an airfoil section via the Curvature view. For details, see CurvatureView (p. 35).

4.2.6.2. Section Stacking View

The Section Stacking view displays, and enables you to adjust, the theta-versus-span distribution ofblade section shapes (airfoils).

To show the Section Stacking view:

1. Select the blade for which you want the Section Stacking view.

2. If you do not see the Section Stacking view, click Show/Hide Stacking Pane .


BladeEditor

The individual airfoil sections are stacked in 3D space by the meridional position of the section and bythe Theta position of the section. Each section has a reference location that defines the stacking point.The Section Reference Location may be one of the following:

• Section Origin

• Airfoil Centroid

• LE Camberline Point

• TE Camberline Point

• Chord Midpoint

• Throat Midpoint

The meridional position is defined by the Meridional Stacking Method, for which there is currently oneavailable option:

• LE/TE Contours – the meridional extent of each airfoil section is determined by the intersections of theLE and TE contours with the section-defining layer contour. In this way, you can control the meridionalextent of the blade by adjusting the LE/TE contours. In this mode, the section reference location is onlyused for Theta positioning of the blade section.

The absolute Theta position of each section (at the Section Reference Location) is defined by a stackingcurve. The stacking curve specifies how the Theta coordinate of the Section Reference Location varieswith non-dimensional span location. You can control the shape of the stacking curve in the SectionStacking Graph by dragging the curve points, insert/deleting points, or by changing the curve type.




Figure 4.3: Section Stacking Graph

For example, if you wanted to stack the sections in a straight radial line based on the airfoil centroids,you would set Section Reference Location to Airfoil Centroid and set the section stackinggraph to have the same angle for all span values.

Within the Section Stacking view, you can:


• Drag (using the mouse) the control points defining the stacking curve.


• Change the curve type: Bezier, cubic spline, piecewise-linear (context menu).

• Insert a point on the stacking curve (right-click in view to insert the new point).

• Delete a point from the stacking curve (right-click the point to be deleted).



BladeEditor



• Zoom to fit the view (context menu)


• Read stacking data points from a file or save points to a file (context menu).

• Assign parameters to the stacking curve control points - X, Y values (context menu)

The technique for setting input parameters for control points in the Stacking view is the same as thatfor the control points in the Angle and Thickness views. This technique is described in Camberline/Thick-ness Definition Sub-feature (p. 85).

Note

The lean angle is defined as atan(R * dTheta / dSpan) for any point on the stacking curve. Apurely radial stacking curve would have zero lean angle.

Note

Toggling the display options will make them take effect for the remainder of the session,but not from session to session.

Note


Parameter command.

4.2.7. User Preferences and Properties

The model properties are listed in the details view for the root node (along with the DesignModelerfile properties). Model properties are saved with the .agdb file. These properties include:

• Length Tolerance – Sketch edge tessellation tolerance in current length unit

• Angle Tolerance – Angle graph tolerance in degrees

• Beta definition – axial or tangential

• Angle View – x-axis type

• Thickness View – x-axis type

• Show Advanced Properties – this specifies whether the advanced parameters or expert parameters areshown in the details view. This only applies to specific features, such as the Blade and Splitter.




4.3. Blade Editing Features

The BladeEditor toolbar icons related to creating/editing a blade are:

DescriptionIcon

Creates a flow path(FlowPath) feature, which is

FlowPath

used to define the flow pathfor blade geometry (that is,the meridional shape of thepassage). For details, seeFlowPath Feature (p. 48).

Creates a blade (Blade) fea-ture, which is used to create

Blade

the blade bodies. For details,see Blade Feature (p. 51).

Creates a splitter (Splitter)feature, which is used to cre-

Splitter

ate the splitter bodies. Fordetails, see Splitter Fea-ture (p. 64).

Creates a Vista TF export(VistaTFExport) feature for

VistaTFExport

exporting the flow path andblade geometry to the VistaTF through flow analysis tool.For details, see Export toVista TF (.geo) (p. 78).

Creates an ExportPoints fea-ture for exporting blade point

ExportPoints

data to TurboGrid or to ameanline file. For details, seeExport to ANSYS Tur-boGrid (p. 81) and Export asMeanline Data (.rtztfile) (p. 79).

Creates a stage fluid zone(StageFluidZone) feature for

StageFluid-Zone

generating the blade passagebodies. The stage fluid zoneis a 3D fluid region to sup-port CFD analyses. For details,see Stage Fluid Zone Fea-ture (p. 67).

Creates a throat area(ThroatArea) feature for calcu-

ThroatArea

lating the blade throat area.For details, see Throat AreaFeature (p. 68).


BladeEditor

The following topics will be discussed:4.3.1. Flow Path Contour Creation4.3.2. FlowPath Feature4.3.3. Blade Feature4.3.4. Camberline/Thickness Definition Sub-feature4.3.5. Blade Section (Airfoil Design Mode) Sub-feature4.3.6. Splitter Feature4.3.7. Stage Fluid Zone Feature4.3.8.Throat Area Feature

4.3.1. Flow Path Contour Creation

The main elements of a multi-blade row machine are the flow path and the individual blade rows, asshown in Figure 4.4: Flow Path and Blade Row Concepts (p. 47).

Figure 4.4: Flow Path and Blade Row Concepts

The first step in creating new blade row geometry is creating the flow path contours that define thehub, shroud, inlet and outlet. The flow path contours are defined by sketch edges, which can be createdusing the existing DesignModeler sketch tools. Each contour (hub, shroud, etc.) should be defined in aseparate sketch. This implicitly identifies all the edges belonging to a given contour. All contour sketchesare expected to lie on the same Plane feature. Not only will this guarantee that the contours are coplanar,it will enable you to apply constraints and dimensions between sketches.

Note

• Contour sketches must be created on the (global) ZX-plane. The local X and Y axes on thesketch plane correspond to the global Z and X axes, respectively. The local X axis correspondsto the machine axis and the local Y axis corresponds to the radial coordinate axis. Consequently,all flow contours in the sketch must have positive Y coordinates.

• It is advisable that you turn off the sketch global autoconstraint setting. Click the Sketching

tab, click the Constraints toolbox, click Auto Constraints and clear the Global check box.Having this global constraint turned on may cause undesirable constraints to be added to theLayerContour edges.

• The hub, shroud, inlet, and outlet contour end points must be coincident.



Blade Editing Features

4.3.2. FlowPath Feature

The FlowPath feature specifies the flow path. The contour sketches described in Flow Path ContourCreation (p. 47) are used in the definition of this feature. The FlowPath feature also creates the layercontours used to define the blade profiles, and automatically creates the LayerContour sketch, whichshows the locations of the layers. The feature properties are listed in the table below. Note that all itemsunder each bold heading in the table are grouped under that heading.

Details of [Feature Name]

[Name]FlowPath

[Pump | Centrifugal Com-pressor | Axial Compressor |

Machine Type

Fan | Radial Turbine | AxialTurbine | Hydraulic Turbine |Other | Undefined]

[Right-handed | Left-handed]Theta Direc-tion

[Sketch selection]Hub Contour

[Sketch selection]Shroud Con-tour

[Sketch selection]Inlet Contour

[Sketch selection]Outlet Con-tour

[Yes | No]Hub Cut?

[Sketch selection](If “Hub Cut?”= “Yes”)

Hub Cut

[Yes | No]Shroud Cut?

[Sketch selection](If “ShroudCut?” = “Yes”)

Shroud Cut

Implicitly defined number oflayers

Number ofLayers

[Sketch selection]Sketches forDefined Layer

Sketches of User Defined Layers:

[Sketch name]Sketch Name

[Another sketch name]Sketch Name

Layer Details: 1

[Fixed Span | User DefinedLayer]

Layer Type

0Span Fraction1

Layer Details: 2


BladeEditor


Layer Type

[0 < Value < 1]Span Fraction2

Layer Details: 3


Layer Type

[0 < Value < 1]Span Fraction3

Layer Details: 4


Layer Type

1Span Fraction4

The Machine Type is used by downstream applications, for example in CFD-Post, for determining theappropriate default postprocessing report. The Theta Direction specifies the interpretation of the thetadirection for the CamDef angle definitions. The theta direction can be specified as either right-handedor left-handed relative to the direction of the global Z axis (the machine axis).

By default, the flow path starts with two layers: one for the hub and one for the shroud. You can insertor delete layers. The first and last layers must be at span fractions of 0 and 1, respectively; these willalways exist and cannot be deleted or changed. A given blade does not need to have defined profileson all layers. The layers are defined in the context of the FlowPath feature for consistency betweenblade rows.

A layer can be defined by a fixed value of span fraction, or it can be defined by a sketch:

• To insert a new fixed-span-fraction layer, right-click an existing fixed-span-fraction layer in the details viewand select Insert Layer Above or Insert Layer Below. To delete a fixed-span-fraction layer, right-click itand select Delete.

• To add one or more layers using sketches, specify the applicable sketches (which must be created before-hand) in the Sketches for Defined Layer property. Each sketch should contain one or more (separate)curves. Each curve in each specified sketch is used to create one layer. For example, if you select twosketches, one having one curve and the other having four curves, then five sketch-based layers will begenerated.

After you apply your sketch selection, a new category, Sketches of User Defined Layers, appears andlists the sketches that were selected. You can remove a sketch listed under Sketches of User DefinedLayers by right-clicking the sketch and selecting Remove from selection.




After you select sketches, or if you make any changes to the sketch selection in general, you mustthen regenerate the FlowPath feature. After regeneration is complete, the user-defined layers willappear in the “Layer Details” portion of the FlowPath details view.

Note

After the FlowPath feature has been generated, you will not be able to edit the Sketches forDefined Layer property until you right-click the FlowPath feature in the tree view and selectEdit Selections from the shortcut menu.

All layers are presented in the details view in the order of average fractional span value. If you changea layer’s sketch curve or span fraction value such that the order of layers changes, then:

• The layers are renumbered (in accordance with their average fractional span values), and

• Any objects that refer to layers will continue to refer to the same layers under their new numbers.

For example, if the Blade sub-feature Blade1_Camberline6 refers to layer 6 and you subsequently inserta layer before layer 6, then the original layer 6 will be renumbered to become layer 7 andBlade1_Camberline6 will then automatically change so that it refers to layer 7 (and it will be renamedto Blade1_Camberline7).

Note

The layers used by a blade are required to not touch or intersect each other anywherewithin the blade.

Note

When you change sketches that are used to define layers, the software will automaticallyattempt to maintain the association between the layers and the objects that refer to them.For example, if you add a curve near an existing curve, then delete the latter, the objectsthat originally referenced the deleted curve may be changed automatically to refer to thenew curve. In cases where the software encounters difficulty maintaining assignments, awarning message may be issued.

If automatic reassignment is not applicable, manual reassignment can be carried out. Forexample, if you delete a curve that was being referenced by an object, and there is no othercurve available for that object, then the object will be left in an invalid state. In that case,you could correct the problem by creating a new curve and manually reassigning that curveto the object. Note that for section definition objects, the manual reassignment must bedone before regenerating the object.

You can create the FlowPath feature automatically via the Load BGD feature, or you can manually createthe FlowPath feature. In either case, you can change the sketch selections once the feature has beencreated.

The FlowPath feature has optional hub cut and shroud cut properties to support trimming the flowpath for a 'flow cut'. An example is if you have a base centrifugal compressor blade design that is usedfor a range of compressors. In this case, the FlowPath and Blade features would define the base bladegeometry. You then want to use the same blade, but in a compressor of different size, so the blade


BladeEditor

height must be trimmed. The flow cut would be used to define the final flow path boundaries, andwould trim the height of all blades in the flow path.

The flow cut affects all blades that reference the flow path. The flow cut properties take as input thesketch contours that define the flow cut. The additional feature properties are in a group below theOutlet Contour property.

4.3.3. Blade Feature

The Blade feature is the geometry feature that is responsible for creating a blade or blade set. Thisfeature defines the key properties of the blade, such as the blade type and the leading and trailingedge location and shape. A given Blade feature (and its child Splitter features) must not overlap withother Blade features (and their child Splitter features) within a given FlowPath.

Blade profiles can be defined in two ways:

• By referring to angle and thickness definitions (CamThkDef sub-features: see Blades made using Camber-line/Thickness sub-features (p. 51)) to define the blade profiles, or

• By referring to blade sections and their theta positions (Blade Section sub-features: see Blades made usingBlade Section (Airfoil Design Mode) Sub-features (p. 53)).

4.3.3.1. Blades made using Camberline/Thickness sub-features

When using the Camberline/Thickness blade design mode, the blade shape is defined on flow pathlayers according to angle and thickness definitions.

In the tree view, when using the Camberline/Thickness blade design mode, CamThkDef sub-featuresappear as sub-nodes of the Blade feature (or Splitter feature, if the Splitter is of Type “Independent”).For details on CamThkDef sub-features, see Camberline/Thickness Definition Sub-feature (p. 55).

When using the Camberline/Thickness blade design mode, the feature properties of the Blade/Splitterfeature are as follows:


[Feature Name]Blade

[ Camberline/Thickness |Airfoil ]

Blade DesignMode

[FlowPath selection]Flow Path

[Rotor | Stator]Type

[Integer Value]Number ofBlade Sets

[General | Ruled Element]Surface Con-struction

[Value]Blade Exten-sion (%)

[None | Layer]Shroud TipClearance

This property also affects anyand all splitter blades.




Layer number at which tomake a blade tip (at theshroud end of the blade).

Shroud TipLayer

(Only for“Shroud Tip This property also affects any

and all splitter blades.Clearance”set to “Lay-er”)

Leading Edge Details

[Sketch selection]LEContour

[Ellipse | Cut Off | Square]a

Type

[Value]LE Ratio atHub (for El-lipse only)

[Value]LE Ratio atShroud (forEllipse only)

Trailing Edge Details

[Sketch selection]TEContour

[Ellipse | Cut-off | Square]b

Type

[Value]Ellipse Ratioat Hub

[Value]Ellipse Ratioat Shroud

Advanced Properties (Hidden by default)

[Int Value]Number ofPoints Alongthe Blade

[Int Value]Number ofPoints for LE(for Ellipseonly)

[Int Value]Number ofPoints for TE(for Ellipseonly)

Camberline/Thickness Definitions: [number]

YesLayer 1

[Yes | No]Layer 2

[Yes | No]...

YesLayer naNote that you cannot specify a perfectly sharp leading edge . As a workaround, you can specify a square trailing edge with a small,but finite, thickness (for example, 0.05 mm).


BladeEditor

bNote that you cannot specify a perfectly sharp trailing edge . As a workaround, you can specify a square trailing edge with a small,but finite, thickness (for example, 0.05 mm).

The first Type property specifies whether the blade is a rotating or stationary component.

If the Surface Construction property is set to “Ruled Element”, then exactly two CamThkDef sub-featuresmust be used; one at the hub and one at the shroud.

When a curve in the angle or thickness view is modified, the properties displayed for the correspondingCamThkDef sub-feature are updated simultaneously, but the graphics view will not be updated untilyou regenerate the Blade.

4.3.3.1.1. Adding/Removing CamThkDef sub-features

When a new blade or splitter is created, CamThkDef sub-features are created automatically for the Huband Shroud layers. When a new layer is added to the flow path, the layer will appear in the CamberlineDefinitions group, but will be marked as 'No' to indicate that it is not a defining layer.

If you specify a curved leading or trailing edge contour, you must have at least three angle and/orthickness defining camberlines to create the desired blade shape. Furthermore, if you specify a cut-offleading or trailing edge and you do not have sufficient camberlines to follow the edge curvature, thenthe cut-off operation might fail. Adding additional camberline definitions (especially where the lead-ing/trailing edge curve deviates the most from the blade shape) should fix this problem.

Whenever the Camberline Thickness Definition Layer property is set to ‘No’, no CamThkDef sub-featureexists under the Blade/Splitter for this layer. Also in this case, no information is displayed in the Angle,Thickness, or Auxiliary Views for this layer.

When a Camberline Thickness Definition Layer property is switched from 'No' to 'Yes', a CamThkDefsub-feature will be created for the Blade or Splitter, and the data will initially be interpolated from theexisting blade data. The Angle Definition and Thickness Definition properties of the CamThkDef sub-feature will both be set to ‘User Specified’ so that the angle and thickness data is editable (otherwiseit would remain interpolated from the existing blade data, but would be uneditable).

When a layer that was used for a CamThkDef sub-feature is removed from the flow path, the Blade orSplitter feature will generate an error. The error can be corrected by inserting a new layer in the flowpath at the same layer index, or by switching the Camberline Thickness Definition Layer property forthat layer to 'No' (therefore deleting the associated CamThkDef sub-feature). In the latter case, the layerwill be removed from the list. If you insert more than one layer into the flow path after deleting a layer,the CamThkDef should be reconnected automatically to the layer with the closest average span value.

4.3.3.2. Blades made using Blade Section (Airfoil Design Mode) Sub-features

When using the Airfoil blade design mode, the blade shape is defined by a stacked sequence of airfoilsections. The airfoil sections are defined on the flow path layers, as with the Camberline/Thicknessmode. The basic topology of each blade section (the set of edges that describe the shape of the airfoil)is defined by the section template, which is shown in Figure 4.5: Blade Section Sketch Template (p. 62).

In the tree view, when using the Airfoil blade design mode, Blade Section sub-features appear as sub-nodes of the Blade feature . For details on Blade Section sub-features, see Blade Section (Airfoil DesignMode) Sub-feature (p. 60).

When using the Airfoil blade design mode, the feature properties of the Blade feature are as follows:





[Feature Name]Blade

[ Camberline/Thickness | Airfoil ]Blade DesignMode

[ Positive | Negative ]Section Ori-entation

[FlowPath selection]Flow Path

[Rotor | Stator]Type

[Int Value]Number ofBlade Sets

[General | Ruled Element]Surface Con-struction

[Value]Blade Exten-sion (%)

[None | Layer]Shroud TipClearance

This property also affects anyand all splitter blades.

Layer number at which tomake a blade tip (at theshroud end of the blade).

Shroud TipLayer

(Only for“Shroud Tip This property also affects any

and all splitter blades.Clearance”set to “Lay-er”)

Meridional Stacking Definitions

[LE/TE Contours]StackingMethod

[Section Origin | AirfoilCentroid | LE Camberline

Section Refer-ence Location

Point | TE Camberline Point |Chord Midpoint | ThroatMidPoint ]

[Sketch selection]LE Contour

[Sketch selection]TE Contour

[Inside Contours]Location

Section Definitions [number]

YesSection1 onLayer1

[Yes | No]Section2 onLayer2

[Yes | No]...

YesSection n on Lay-ern


BladeEditor

The Section Orientation property specifies how the blade section is oriented with respect to the Thetadirection. The choices are as follows:

• Positive - makes the suction side of the blade face the low-Theta direction.

• Negative - makes the suction side of the blade face the high-Theta direction.

When you make a change to the Section Orientation property, the section input data is preserved sothat the section shape is effectively mirrored.

The FlowPath selection is the reference to the FlowPath feature, which is mandatory.

The Type property specifies whether the blade is a rotating or stationary component.

You can turn defining sections on or off from an existing airfoil design mode blade feature. This givesyou flexibility in controlling the blade shape without having to recreate the blade feature.

You control which sections are defining sections in the Properties view for the blade feature. The bladeproperty group Section Definitions lists all available sections. The property value for each section hasa Yes or No value depending on whether it is a defining section (Yes) or not. When the blade featureis first created, all sections are marked as defining sections. (The hub (layer1) and shroud (layer n) arealways defined.)

If the flow path layers are modified, updating the blade feature shows any added sections in the bladeSection Definitions. Newly added sections are marked as not defining sections unless you change theproperty value. Existing defining sections remain unchanged.

You cannot delete a flow path layer if a section is defined on it. If a section definition is switched fromYes to No, the section data will be preserved (including save/resume) until the flow path layer is deleted.Therefore, you can switch it back to Yes, keeping the previous data, as long as the layer still exists. Hublayer and shroud layer are always defined.

The section names are numbered in the blade Properties view as the layer number on which they aredefined (for example, Section1, Section2, Section3, and so on). The section sub-features listed underthe blade feature in the tree are only the defining sections, and the names correspond with the namesin the Properties view.

Note that Airfoil Design mode is not compatible with the following features:

• Throat Area

• Export Points (meanline)

• Stage Fluid Zone

• Vista TF Export

• Splitter Blades

4.3.4. Camberline/Thickness Definition Sub-feature

The Camberline/Thickness Definition (CamThkDef ) sub-feature, found under the Blade (or Splitter) feature,defines a blade profile on a particular layer (of the FlowPath feature) according to:

• Angle data that defines the shape of the camberline, and




• Thickness data that defines the thickness of the blade along the camberline.

The Angle and Thickness views can be used to view/edit the angle data and thickness data, respectively.For details, see Angle and Thickness Views (p. 37). As a side note, the blade lean angle is plotted in theblade lean graph. For details, see Blade Lean Graph (p. 34).

The CamThkDef sub-feature properties are as follows:

Details of [Sub-feature Name]

[CamThkDef Sub-FeatureName] is derived from [par-

Camber-line/Thickness

ent Blade name]_Layer[Layernumber]

Details of Camberline

[Interpolated | User Specified]Angle Defini-tion

For the hub and shroud lay-ers, this property is always setto User Specified.

[Theta | Beta | Theta withLE/TE Beta]

Angle Defini-tion Type

[Leading Edge | Trailing Edge](For Beta defini-tion only)

Theta Reference

[Value](For Beta defini-tion only)

Theta at Refer-ence

The Beta angle at the leadingedge.

(For Theta withLE/TE Betadefinition only)

Beta at LE

The Beta angle at the trailingedge.

(For Theta withLE/TE Betadefinition only)

Beta at TE

[Camberline | Side 1 | Side 2]Angle Data Loc-ation

[% M’ | M’ | % M | M]X-Axis Defini-tion Type

(Controls the type of dataused for Angle Point inputparameters.)

Details of Thickness

[Interpolated | User Specified]ThicknessDefinition


BladeEditor

For the hub and shroud lay-ers, this property is always setto User Specified.

[Normal to Camberline onLayer surface]

ThicknessDefinition Type

[% M’ | M’ | % M | M]X-Axis Defini-tion Type

(Controls the type of dataused for Thickness Point in-put parameters.)

Details of Angle Point 1

[Value]X of AnglePoint 1

[Value]Y of AnglePoint 1

Details of Thickness Point 1

[Value]X of ThicknessPoint 1

[Value]Y of ThicknessPoint 1

CamThkDef sub-features are named according to the associated Blade name and Layer number.

When a CamThkDef sub-feature is selected in the feature tree, its angle data is displayed on the anglegraph and its thickness data is displayed on the thickness graph. The angle and thickness distributionsare plotted from leading edge to trailing edge.

If the Angle Definition Type is set to Theta, then the specified angle curve is used as the theta definitionfor the camberline.

If the Angle Definition Type is set to Beta, then the specified angle curve is used as the Beta definitionfor the camberline. In this case, the theta definition is derived by integrating the Beta curve using thespecified Theta Reference value. The Theta Reference value can be specified at either the leading edgeor the trailing edge of the blade.

If the Angle Definition Type is set to Theta with LE/TE Beta, then you can control the theta distributionas well as specify the leading and trailing edge Beta values. This type is not offered until you havegenerated the camberline with:

• Angle Definition Type set to Theta, and

• A Theta curve that consists of a Bezier curve having at least four control points.

When the Angle Definition Type is set to Theta with LE/TE Beta, the Beta at LE and Beta at TE values inthe details view show/control the leading and trailing edge blade metal angles. These same angle valuescan be seen in the Angle view as the angle values of the first and last points on the Beta curve. Youcan also control these angle values in the Angle view by moving the second and second-last controlpoints on the Beta curve. Note that you can parameterize the Beta at LE and Beta at TE values in thedetails view, which will restrict the motion of the control points in the Angle view. You can also set thesecond and second-last control points to be used as input parameters, causing new properties (for ex-ample, X of Angle Point 1, under the Details of Angle Point 1 property group) to be added in the details




view; if you parameterize those properties, the motion of the control points in the Angle view is con-strained.

When the Angle Definition Type is set to Theta with LE/TE Beta, control points can be inserted or deletedfor the Bezier curve with the exception that the first, second, second-last, and last control points cannotbe removed. In this mode, the curve type cannot be changed.

For parameterization purposes, the Angle and Thickness points have an X-Axis Definition Type propertythat defines what scale the data uses: % M', M’, % M, or M. This property cannot be changed when anyof the graph values are parameterized.

The following topics are discussed:4.3.4.1. Interpolated and Non-interpolated Angle/Thickness Data4.3.4.2. Importing and Exporting Angle Definition Data4.3.4.3. Importing and Exporting Thickness Definition Data4.3.4.4. Converting Curves to Bezier or Spline4.3.4.5. Converting Angle Definition Data

4.3.4.1. Interpolated and Non-interpolated Angle/Thickness Data

If the Angle Definition property of the CamThkDef feature is set to ‘Interpolated’, then the angle defin-ition is interpolated from the blade/splitter and cannot be edited. In addition, any custom changes youmay have made to this data will be discarded.

If the Thickness Definition property of the CamThkDef feature is set to ‘Interpolated’, then the thicknessdefinition is interpolated from the blade/splitter and cannot be edited. In addition, any custom changesyou may have made to this data will be discarded.

Setting both the Angle Definition and Thickness Definition to ‘Interpolated’ can be useful when theleading and/or trailing edges are curved on a blade/splitter with insufficient CamThkDef features torepresent the curvature.

If you switch the Angle Definition property from ‘Interpolated’ to ‘User Specified', the angle definitionwill be initialized by interpolation from the blade definition, and will then become editable.

If you switch the Thickness Definition property from ‘Interpolated’ to ‘User Specified', the thicknessdefinition will be initialized by interpolation from the blade definition, and will then become editable.

4.3.4.2. Importing and Exporting Angle Definition Data

There is an import/export points mechanism for the Angle Definition to support the manipulation ofthe Angle Definition points outside of BladeEditor, and to provide a means to read data from a fileproduced externally. In the Angle view, the context menu operation 'Read Points' lets you replace theactive curve data points from the data in a text file called a '.ha' file. The context menu operation 'SavePoints' lets you save the active data points to a '.ha' text file. These options are available only if theAngle Definition is active, that is, set to 'User Specified'.

The first line in the '.ha' file specifies the number of points. Subsequent lines in the file, one for eachdata point, contain the point coordinate values 'h, a', where 'h' is the horizontal or x-axis value and 'a'is the angle value. The point coordinates correspond to the defining points for the angle definitioncurve, which may be either Theta or Beta, in radians, versus m-prime, % m-prime, m or % m.

When reading the '.ha' file, the data point coordinates will be interpreted based on the Angle DefinitionType and the selected x-axis type. If the Angle Definition Type is set to Theta, then the angle values


BladeEditor

will be interpreted as Theta values. Otherwise, the angle values will be interpreted as Beta values. If thex-axis is set to m-prime or % m-prime, then the 'h' coordinates for the data points will be treated as %m-prime values, and the values will be normalized based on the first and last data points so that thevalues start at zero and end at 100%. If the x-axis is set to m or % m, then the 'h' coordinates for thedata points will be converted to m values by normalizing based on the first and last data points andthen multiplying by the maximum m on the layer. These values will then be converted to % m-prime.The existing angle curve data for all segments will be replaced by a single curve segment with the datafrom the '.ha' file, and the curve type will be set to 'cubic spline'.

When you choose 'Save Points', the defining points for all curve segments will be written sequentiallyto the file starting with the leading edge point and ending at the trailing edge point. The 'h' coordinateswill be written in the value corresponding to the selected x-axis type. If the Angle Definition Type isset to Theta, then the Theta values will be written as the angle values. Otherwise, the Beta values willbe written as the angle values.

4.3.4.3. Importing and Exporting Thickness Definition Data

There is an import/export points mechanism for the Thickness Definition to support the manipulationof the Thickness Definition points outside of BladeEditor, and to provide a means to read data from afile produced externally. In the Angle view, the context menu operation 'Read Points' lets you replacethe active curve data points from the data in a text file called a '.ht' file. The context menu operation'Save Points' lets you save the active data points to a '.ha' text file. These options are available only ifthe Thickness Definition is active, that is, set to 'User Specified'.

The first line in the '.ht' file specifies the number of points. Subsequent lines in the file, one for eachdata point, contain the point coordinate values 'h, t', where 'h' is the horizontal or x-axis value and 't'is the thickness value. The point coordinates correspond to the defining points for the thickness defin-ition curve in the length unit being used in DesignModeler, versus m-prime, % m-prime, m or % m.

When reading the '.ht' file, the data point coordinates will be interpreted based on the selected ThicknessDefinition Type and the x-axis type. If the x-axis is set to m or % m, then the 'h' coordinates for the datapoints will be treated as % m values, and the values will be normalized based on the first and last datapoints so that the values start at zero and end at 100%. If the x-axis is set to m-prime or % m-prime,then the 'h' coordinates for the data points will be converted to m-prime values by normalizing basedon the first and last data points and then multiplying by the maximum m-prime on the layer. Thesevalues will then be converted to % m. The existing thickness curve data for all segments will be replacedby a single curve segment with the data from the '.ht' file, and the curve type will be set to 'cubic spline'.

When you choose 'Save Points', the defining points for all curve segments will be written sequentiallyto the file starting with the leading edge point and ending at the trailing edge point. The 'h' coordinateswill be written in the value corresponding to the selected x-axis type, while the thickness coordinateswill be written in the value corresponding to the selected Thickness Definition Type.

4.3.4.4. Converting Curves to Bezier or Spline

This feature lets you convert a curve segment in either the Angle or Thickness views to an equivalentspline or Bezier curve with a specified number of points. The new curve is an approximation of theoriginal curve, and the error in the approximation depends on the complexity of the original curve andthe specified number of points for the new curve. This operation is very helpful for simplifying a curvewith many points, either because it was imported or was interpolated.

The context menu items are: 'Convert to Bezier' and 'Convert to Spline'. They are available in both theAngle and Thickness views. When choosing either of these options, you are asked to specify the number




of points for the new curve in a dialog box. After clicking OK, you are required to select the curvesegment to convert in the graph view. The curve will be converted immediately following your selection.

4.3.4.5. Converting Angle Definition Data

If you select the CamThkDef sub-feature under a Blade feature and change the Angle Definition Type,then the angle data is converted to the new type. This conversion will try to retain the shape of thecamberline by interpolation. Thirty points are used in the interpolated curve.

4.3.5. Blade Section (Airfoil Design Mode) Sub-feature

The blade section definitions enable you to view/edit the airfoil input data on each defining layer. TheBlade Section sub-features appear under the Blade feature in the tree view. By selecting a sectiondefinition, it becomes visible for editing in the Section View (for details on showing this view, see SectionDefinition View (p. 40)), and the section properties become visible in the details view. When a newBlade airfoil design mode feature is first created, the section details are not available until valid FlowPathand LE/TE contours are selected. Once these are selected, blade sections are automatically defined forall layers in the FlowPath.

The Blade Section sub-feature properties are as follows (Note that angles are in degrees and lengthsare in the length unit used in the active session of Blade Editor.):

Details of [Sub-feature Name]

[Name of Section] (for information only)Section

[Index Value] (for information only)Layer

Details of Dimensions

The angle between the M axis and the stagger line.StaggerAngle

Half the length of the LE ellipse's major axis, which isthe axis that is at an angle of LEBeta with respect to theM axis.

MajorLERadi-us

Half the length of the LE ellipse's minor axis.MinorLERadi-us

Angle between the major axis of the LE ellipse and theline of tangency on the LE ellipse at the point where the

LESS-WedgeAngle

LE suction side curve contacts the LE ellipse. Note thatthis line of tangency on the LE ellipse is also a line oftangency to the LE suction side curve at the LE end ofthat curve. Note also that the value of LESSWedgeAngleaffects the location of the point where the LE suctionside curve contacts the LE ellipse.

Angle between the major axis of the LE ellipse and theline of tangency on the LE ellipse at the point where the

LEPSWedgeAngle

pressure side curve contacts the LE ellipse. Note thatthis line of tangency on the LE ellipse is also a line oftangency to the pressure side curve at the LE end of thatcurve. Note also that the value of LEPSWedgeAngle af-fects the location of the point where the pressure sidecurve contacts the LE ellipse.


BladeEditor

Angle between the major axis of the LE ellipse and theM axis.

LEBeta

Angle between the major axis of the TE ellipse and theM axis.

TEBeta

Length of the throat line (the specified throat, not thecalculated throat).

ThroatWidth

Angle between the throat line and the R*Theta axis.SSThroat-Angle

Half the length of the TE ellipse's major axis, which isthe axis that is at an angle of TEBeta with respect to theM axis.

MajorTERadi-us

Half the length of the TE ellipse's minor axis.MinorTERadi-us

Half the TE wedge angle.TE-HalfWedgeAngle

These Blade Section Sub-feature properties, in conjunction with the sketch template, define the airfoilshape. The blade section template, as applied to an example blade section, is shown in the followingfigures: Figure 4.5: Blade Section Sketch Template (p. 62), Figure 4.6: Blade Section Sketch Template:Details in Leading Edge Region (p. 63), Figure 4.7: Blade Section Sketch Template: Details in TrailingEdge Region (p. 64).

In these figures, the red “asterisk” symbols represent control points for the Bezier curves. Note that eachend of the stagger line (that is, the chord line) is at an intersection of the major axis of an ellipse andthe ellipse itself. Note that the R*Theta axis is tangent to the blade. The M axis is at the intersection ofthe R*Theta axis and the extended major axis of the leading edge ellipse.




Figure 4.5: Blade Section Sketch Template


BladeEditor

Figure 4.6: Blade Section Sketch Template: Details in Leading Edge Region




Figure 4.7: Blade Section Sketch Template: Details in Trailing Edge Region

The Blade Section Sub-feature properties can be parameterized as for other properties in DesignModeler.For details, see Creating Parameters in the DesignModeler User Guide.

When the sections are first created, the dimensions are scaled to the meridional chord defined by theLE/TE contours.

To edit the airfoil shapes and stacking arrangement, use the Section and Stacking views, respectively.For details, see Section Definition and Stacking Views (p. 40). You can see the curvature of an airfoilsection via the Curvature view. For details, see Curvature View (p. 35).

After modifying one of the blade sections, but before regenerating the blade, the graphics windowdisplays the blade sections and the curve that passes through the Section Reference Location of eachblade section. Note that the Theta-versus-span distribution of this curve follows the relation given bythe stacking curve (see Section Stacking View (p. 42)).

4.3.6. Splitter Feature

This feature creates a new blade relative to an existing (main) blade such that the new blade (or splitter)shares the same flow path, tip clearance, and blade count, and is generally assumed to be part of thesame blade row. The Reference Blade of this feature is the selected Blade feature; only one Blade featurecan be selected per Splitter feature.

The Splitter feature can have one of two forms: a cloned splitter or an independent splitter. The formis specified in the feature properties. Once the feature has been created, you cannot change the type.

The following topics will be discussed:


BladeEditor

4.3.6.1. Cloned Splitter4.3.6.2. Independent Splitter

4.3.6.1. Cloned Splitter

A cloned splitter shares the same blade shape as the main blade, but is offset in theta from the mainblade.

The feature properties are as follows:


[Feature Name]Splitter

[Blade feature selection]ReferenceBlade

CloneType

[Angle | Pitch Fraction]Offset

[Value: 0-1]Pitch Fraction

or

[Value]Angle

The Offset specifies how the splitter will be offset relative to the main blade. If “Angle” is selected, thenthe angular offset of the cloned splitter blade relative to the main blade is specified directly. If “PitchFraction” is selected, then the pitch fraction as a value between 0 and 1 is specified. The pitch fractionrepresents the fraction of distance between two adjacent main blades. The actual angular offset is:

offset = pitch fraction * 2 * π / n

where n is the number of main blades.

4.3.6.2. Independent Splitter

An independent splitter can have a completely different shape than the main blade (except that theblade clearance properties are inherited from the main blade). An independent splitter shares the sameblade count as, and has a fixed offset from, the main blade. The feature properties are nearly identicalto the Blade feature except for the additional main blade reference and offset value.

The properties are as follows:


[Feature Name]Splitter

[Blade feature selection]ReferenceBlade

IndependentType

[Angle | Pitch Fraction]Offset

[Value: 0-1]Pitch Fraction

or




[Value]Angle

[Hub at Leading Edge | Hubat Trailing Edge]

Offset Refer-ence

(Only for “Off-set” set to“Pitch Frac-tion”)

...

(The remaining details are the same as for theBlade feature selected as the Reference Blade.)

The interpretation of the Offset is different for an independent splitter than for the cloned splitter. If“Angle” is selected, then the angular position of the splitter will be determined by the camber surfacetheta, shifted by the specified angle. If “Pitch Fraction” is selected, then you enter the pitch fraction asa value between 0 and 1, for the cloned splitter. However, because an independent splitter can have adifferent shape than the main blade, an Offset Reference must be specified. This determines how thesplitter will be positioned relative to the main blade. The angular offset and the offset reference determinethe reference theta for the splitter blade.

When “Hub at Leading Edge” is selected for the Offset Reference, the reference theta and the camberlinelocations are calculated as follows:

1. An offset reference meridional location is calculated at the intersection of the hub contour and thesplitter leading edge contour.

2. For the offset meridional location, the main blade theta value is calculated from the main blade cam-bersurface.

3. The reference theta is then calculated by adding the angular offset (from the pitch fraction or specifiedangle) to the main blade theta value.

4. All splitter camberline theta values are offset internally so that the splitter hub camberline starts at thereference theta.

Alternatively, when “Hub at Trailing Edge” is selected for the Offset Reference, the reference theta andthe camberline locations are calculated as follows:

1. An offset reference meridional location is calculated at the intersection of the hub contour and thesplitter trailing edge contour.

2. For the offset meridional location, the main blade theta value is calculated from the main blade cam-bersurface.

3. The reference theta is then calculated by adding the angular offset (from the pitch fraction or specifiedangle) to the main blade theta value.

4. All splitter camberline theta values are offset internally so that the splitter hub camberline ends at thereference theta.

When a BladeGen model with splitters is imported, the theta values shown for the splitter camberlinesmay appear to differ in BladeEditor. The reason is that BladeGen always calculates the splitter referencetheta using a meridional location at the main blade leading edge rather than at the splitter blade


BladeEditor

leading edge. When the splitter camberline definitions are imported they are converted to preserve theblade position and shape, and therefore the point data for the definitions may differ.

4.3.6.2.1. Camberline/Thickness Definition Sub-features of Independent Splitters

An independent splitter (as opposed to a cloned splitter) has its own CamThkDef sub-features if itsReference Blade is a (main) blade that is defined using CamThkDef sub-features (as opposed to bladesection (airfoil design mode) sub-features). The properties of a CamThkDef sub-feature of a splitter arethe same as those of a CamThkDef sub-feature of a main blade (described in Camberline/ThicknessDefinition Sub-feature (p. 55)), with the following exceptions:

• The Angle Definition property has an extra option, From Reference Blade, that causes the angle data tobecome the same as that of the main blade for the length of the splitter camberline. You can select FromReference Blade then User Specified if you want to make the angle definition editable but initialized frommain blade data.

• The Thickness Definition property has a corresponding extra option with similar behavior.

4.3.7. Stage Fluid Zone Feature

The “fluid zone” represents the 3D fluid region within the flow passage, and is used to define the extentof the CFD domain. In release 11.0, a fluid zone was automatically created for each ImportBGD feature,which included the fluid region around a single blade row. For multiple blade rows, the fluid zone mustinclude separate regions for, and have distinct interfaces between, each blade row.

In release 12.0 and higher, a StageFluidZone feature takes as a reference the selected FlowPath feature,and creates solid bodies for the fluid region in every blade row of the flow path. The first stage fluidzone in the flow path will start at the inlet contour and extend to the first interface (or to the outletcontour if there is only one blade row). The last stage fluid zone starts at the last interface and extendsto the outlet contour. All intermediate stage fluid zones start and end at the interfaces adjacent to theblade row. The individual bodies are “frozen” so that they are not merged to adjacent solid bodies, in-cluding adjacent stage fluid zone bodies. (Individual bodies can be suppressed or hidden using existingDM functionality.) All active bodies (blades, hub, etc.) are subtracted from the fluid zone.

The stage fluid zone upstream and downstream interfaces are located by default halfway between theclosest extents of the adjacent blade rows. For this to work, the StageFluidZone feature assumes thatthere is no overlap between adjacent blade rows; the StageFluidZone feature reports an error if thiscondition is violated. (Overlapping blade designs should be created using a combination of the Bladeand Splitter features.) You may adjust the relative location of the interface by using one value in therange 0-1.


Table 4.1: Details View for StageFluidZone Feature


[Feature Name]StageFluid-Zone

[FlowPath feature selection]Flow Path

[Yes | No]CreateNamed Selec-tions




Interface Detail 1:

[Value: 0-1, default 0.5]aInterface Loc-ation

Interface Detail 2:

[Value: 0-1, default 0.5]aInterface Loc-ation

a0 corresponds to the aftmost trailing edge of the set of blades in the upstream blade row. 1 corresponds to the foremost leadingedge of the set of blades in the downstream blade row.

The Create Named Selections property determines whether the feature will automatically create namedselections for each stage fluid zone. The Interface Details groups are created dynamically dependingon the number of interfaces involved (number of blade rows less one.)

Named selections are created automatically for each blade row of the StageFluidZone. The namingconvention is Hub_[Main Blade Name], Shroud_[Main Blade Name], Blade_[Blade Name], etc. whereMain Blade Name is the name of the main blade in the given blade row. For the Blade_ regions, eachblade in the blade row will have a separate region defined.

Named selections will not appear in DesignModeler, but they will appear if you load the .agdb file intothe Mechanical application or the Meshing application, provided that you have set the properties ofthe Geometry cell (in the Tools > Options > Geometry Import in ANSYS Workbench) as follows:

1. CAD Attributes is selected.

2. Filtering Prefixes are cleared.

(This setting appears when CAD Attributes check box is selected.)

3. Named Selections is selected.

You can make the Geometry cell settings this way by default by selecting Tools > Options from theANSYS Workbench main menu, browsing to the Geometry Import branch, and setting CAD Attributes(same as Attributes), CAD Attributes > Filtering Prefixes (same as Attribute Key), and Named Selections.

Note

If the StageFluidZone periodic interfaces do not appear to be trimmed appropriately on thehub or shroud boundaries, you can try increasing the Blade Extension property for the cor-responding blade feature to remedy the problem. The StageFluidZone feature uses thisproperty to control how the periodic interfaces are created.

4.3.8. Throat Area Feature

The throat area (ThroatArea) feature is used to find and display the minimum-area throat surface(s) forone or more selected blade rows. The calculation is normally updated each time you issue the Generatecommand. The feature may be suppressed to avoid the calculation time when it is not needed, or thefeature may be deleted if it is no longer needed.


BladeEditor


Table 4.2: Details View for ThroatArea Feature


[Feature Name]ThroatArea

[Blade feature selection]Blade

[Leading Edge | Trailing Edge| Minimum Area]

Location

[Value]Throat Sur-face Exten-sion

(AdvancedProperty)

Throat Details:

[Value]Area

[Value]Layer 1Length

[Value]Layer 2Length

...

...

...

Blade itemsare listed forall selectedblades.

The input property, Blade, is a Blade feature selection used to indicate the blade row.

The Location property specifies the region for which to begin searching for the minimum throat area.The Location property is only an initial guess for finding the location of the minimum throat area ofthe passage. The choices are as follows:

• Leading Edge

Finds the minimum passage area in the region of the main blade starting with leading edge. TheLeading Edge option can report the throat area for a throat location that is up to halfway towardthe trailing edge.

• Trailing Edge

Finds the minimum passage area in the region of the main blade starting with the trailing edge. TheTrailing Edge option can report the throat area for a throat location that is up to halfway towardthe leading edge.

• Minimum Area




Finds the minimum passage area in region of the main blade starting with both the leading andtrailing edge. The Minimum Area option searches the entire blade for the minimum throat area.

The minimum area calculation uses a minimization algorithm based on the blade profiles on the designlayers of the blade. If there are only one or two defining layers, then additional layers will be insertedfor the purpose of the calculation.

The output property group contains the throat area and the throat lengths for each layer defined inthe flow path.

The throat surface is displayed in the viewer as a frozen sheet body. The Throat Surface Extensionproperty enables you to control the surface extension to ensure that the throat surface is cut by theblade bodies properly. This is an advanced property and it is shown only if the Show Advanced Propertiespreference is selected.

If you use the VistaTFExport feature, and the ThroatArea feature is used for a selected Blade, the throatinformation for that blade will be written to the .geo file. This information may improve the calculationof the choke mass flow rate in the Vista TF solver. Without this information, Vista TF will make its ownestimate of the throat area.

Note

If the throat surface does not appear to fully capture the shape of the blade passage, it maybe because the selected Blade feature has too few defining camberlines. Specifically, if theselected Blade feature only has two defining camberlines, you may need to insert additionalcamberlines to sufficiently define the shape of the throat surface.

Note

The minimization calculation of the throat surface area uses the raw blade profile data fromthe Blade feature and not the final solid model. The actual throat surface area is calculatedfrom the solid model.

4.4. Blade Comparison

You can take a snapshot of a blade design so that, as you modify the current design, you can easilycompare it against the snapshot to visualize the changes you have made. When the FlowPath, Blade,

and Splitter features have been successfully generated, you can click Create snapshot to take asnapshot, or to update the snapshot, of the current blade design.

The snapshot also appears in the following views:

• Model View

The snapshot appears in the 3D Viewer as a transparent (by default) blade set.

Until you change the blade geometry, the snapshot will coincide with the blade in the Model View.To see the snapshot by itself, suppress the display of the blade geometry (for example, by clicking

Display Model to hide the geometry).

• Angle view


BladeEditor

• Thickness view

• Blade to Blade view

Click Comparison On/Off to toggle the visibility of the snapshot in all views that can display asnapshot. While comparison mode is on, you cannot update or clear the snapshot.

To clear the snapshot, click Clean snapshot . Deleting the snapshot has a (usually minor) benefit; itreduces the disk space required to save your case.

You can control display options for the snapshot by clicking Comparison display option and makingchanges in the details view.

Blade comparison is demonstrated in Tutorial 1: Blade Editing With Emphasis On Sketches, Layers, andBlade Comparison (p. 86).

4.5. Importing Blades from ANSYS BladeGen

BladeEditor provides a geometry connection between BladeGen and DesignModeler. Reasons for im-porting BladeGen blades into DesignModeler include:

• ANSYS BladeGen can output geometry in many different point data formats, but its surface output in IGESformat is cumbersome to use.

• BladeGen does not produce a solid model in a standardized format such as Parasolid.

• You can combine an imported blade with other CAD geometry imported via one of the many Design-Modeler-supported CAD file formats.

Through BladeEditor, one or more BladeGen models can be linked into a DesignModeler session, sothat any changes to the BladeGen models will be reflected in DesignModeler the next time you updatethe Geometry cell.

When you import a BladeGen model, BladeEditor does the following:

• constructs blade surfaces

• creates a solid model for the blades and hub

• creates 2D sketches for the meridional contours and non-flow-path hub geometry

• creates periodic fluid zones

The preferred method of importing a BladeGen file is to create a link from the Blade Design cell of aBladeGen system to the Geometry cell. This link maintains the data transfer relationship between thetwo cells. The desired import options should be set in the Blade Design cell properties. (SeeTable 3.1: BladeGen Blade Design Cell Properties (p. 20) for more information.) After you make the link,the Geometry cell should be updated to process the imported geometry.



Importing Blades from ANSYS BladeGen

Note

If you edit the Geometry cell before updating it, then the ImportBGD feature details that areshown in BladeEditor may not accurately reflect the Blade Design cell properties. To refresh

the ImportBGD feature properties, click in BladeEditor. It is not recommended thatyou edit the ImportBGD properties inside BladeEditor because they will be overwritten bythe properties from the Blade Design cell the next time you update the Geometry cell.

Alternatively, you can import a BladeGen file from outside the project. To do this, click inthe BladeEditor toolbar. When you click this icon, you will be prompted for the location and name ofthe BladeGen (.bgd ) file. Once the filename is selected, the details view enables you to select theproperties for the import. These properties are listed in Table 4.3: Properties for the ImportBGD Fea-ture (p. 72). As with other DesignModeler feature properties, you can double-click in a property valuebox to change the selection to the next choice, or single-click the property and select the value fromthe drop-down list.

Table 4.3: Properties for the ImportBGD Feature

FunctionDefault

Value

Property

This property defines name of the import feature.Import-BGD#

Import-BGD

This property defines the name and path of the impor-ted .bgd file. You can change the source to a new.bgd file if Refresh is set to Yes .

(selec-tedBGDFile)

Source

You may change the value of this property to the in-tended BladeGen model length unit if the latter does

(defaultis the

UnitPrefer-ence not match the DesignModeler length unit. If theDesign-

BladeGen model length unit is specified as “Unknown”Model-er (in the BladeGen model properties), then BladeEditorlengthunit)

will interpret the model as having the length unitspecified here, and will process the model by convert-ing from this unit into the DesignModeler length unit.Otherwise, the unit specified here will be ignored, andthe model will be converted from the unit specifiedin the .bgd file into the DesignModeler length unit.It is recommended that you specify a length unit inBladeGen so that this information is stored in the .bgdfile.

Make sure that the length unit specified here is appro-priate for the model. If the BladeGen dimensions are


BladeEditor

FunctionDefault

Value

Property

too small, DesignModeler may fail to import theBladeGen model.

If this property is set to Yes , then BladeEditor willcreate a HubProfile sketch for the non-flow path

YesCreateHub

hub geometry, and will create a revolved body featurecalled HubBody .

This property defines the default line offset (in thepreferred length unit) for creating the initial HubPro-file sketch.

1 (Inch)HubOffset

Note that this property is available only if Create Hub

is set to Yes .

If this property is set to All , then BladeEditor willcreate all the blades using the number of blades spe-cified in the BladeGen model.

AllCreateBlades

If this property is set to 1, then only the first blade willbe created.

If this property is set to No, then BladeEditor will createthe blade with four faces corresponding to the leading

YesMergeBladeTopo-logy

edge, pressure side, trailing edge and suction side. Thiscan make it easier to create a structural mesh for theblades in the Mechanical application.

If this property is set to Yes , then the blade faces willbe merged where they are tangent to one another.

If this property is set to Streamwise , then BladeEditorwill loft the blade surfaces in the streamwise direction

Stream-wise

BladeLoft Dir-ection through curves that run from hub to shroud. This is

the default because the surface is more well defined,especially for flank-milled blades.

If this property is set to Spanwise , then BladeEditorwill loft the blade surfaces in the spanwise directionthrough the blade profile curves.

For an illustration of these lofting methods, see thefigure after this table.

This property specifies whether a shroud clearance iscreated. If No is selected, then no shroud clearance is

NoCreateShroudClear-ance

created. To create a shroud clearance, choose eitherRelative Layer or Absolute Layer . Theblade(s) will be trimmed off at the selected BladeGenoutput layer, and the layer contour will be created inthe LayerProfile sketch.

If Relative Layer is selected, then the selectedLayer Number is relative to the shroud layer, e.g., 1




FunctionDefault

Value

Property

implies the first layer closest to the shroud layer, 2implies the second closest layer to the shroud, etc.

If Absolute Layer is selected, then the selectedlayer index counts up from the hub layer, which iszero.

This property defines the selected layer index for theshroud clearance.

1LayerNum-ber

Note that this property is available only if Create

Shroud Clearance is selected.

If this property is set to Yes , then BladeEditor willcreate a StageFluidZone body for the flow passage,

YesCreateFluidZone and an Enclosure feature to subtract the blade body.

The resulting Enclosure can be used for a CFD analysisof the blade passage.

If this property is set to Yes , then BladeEditor willcreate NamedSelections (regions) for the typical

YesCreateNamedSelec-tions

faces of the blade passage, that is, Blade , Hub,Shroud , Inflow , Outflow , PeriodicA andPeriodicB . These NamedSelections can be usedas selection groups in other ANSYS Workbench applic-ations.

Note that this property is available only if Create Fluid

Zone is set to Yes .

This property defines the surface extension length (asa percentage of the average hub to shroud distance)

2BladeExten-

for the blade surfaces. These surfaces are extendedsion(%) and then trimmed to the MasterProfile sketch to

ensure that the blade solid correctly matches the huband shroud contours.

This property defines the surface extension length (asa percentage of the average hub to shroud distance)

5Period-ic Surf

for the periodic surfaces. These surfaces are extendedto ensure that the StageFluidZone is properly cut.

Exten-sion(%)

This property specifies the style of the periodic inter-face surfaces.

ThreePieces

Period-ic SurfStyle

If Three Pieces is selected, then the periodic sur-face is created in three connected pieces: one upstreamof the blade, one within the passage, and one down-stream of the blade. This style can better accommodatehighly curved or twisted blades, and is similar to theANSYS TurboGrid style of periodic surface.


BladeEditor

FunctionDefault

Value

Property

If One Piece is selected, then the periodic surfaceis created as a single surface.

Note that this property is available only if Create Fluid

Zone is set to Yes .

This property specifies whether the imported BladeGenmodel should remain linked to the DesignModeler

YesRefresh

session. If this property is set to Yes , then when theDesignModeler Generate button is clicked, BladeEditorwill check to see if the BladeGen file has changed. Ifthe BladeGen file has been modified, BladeEditor willreload the file. If the BladeGen file has moved or hasbeen deleted, BladeEditor will switch this property toNo and will leave the blade geometry unchanged.

If this property is set to No, then BladeEditor will notreload the BladeGen file when the DesignModelermodel is regenerated.

Figure 4.8: Spanwise Lofting versus Streamwise Lofting (p. 76) shows how spanwise lofting andstreamwise lofting differ.




Figure 4.8: Spanwise Lofting versus Streamwise Lofting

Once the properties for the ImportBGD feature have been set, click the Generate button, and theBladeGen model will be imported.

The following features will then be created in the tree view:

• MerPlane : this plane is a copy of the Z-X plane; it is the plane on which the blade design sketches arecreated.

– MasterProfile : a sketch defining the hub, shroud, leading edge, trailing edge, inflow and outflowboundaries of the blade passage (imported from the .bgd file) will be created. This sketch is usedduring the creation of the blade bodies, and can be used to create the fluid zone. You should not

modify this sketch.1

1This constraint is to prevent the MasterProfile and the blades from becoming inconsistent, because the blade surface datacomes from BladeGen.


BladeEditor

– BladeProfile : a sketch defining the locations of the leading and trailing edges for the main bladewill be created.

– HubProfile : a sketch defining the hub body will be (optionally) created. This sketch can be modified;however, you should take care to ensure that the sketch loop remains intact or the hub body will failto regenerate.

HubBody - (Optionally) the HubProfile sketch is revolved to create the HubBody feature inthe tree view.

• BladeBody - The blade surface data is imported and lofted in DesignModeler to create the BladeBodyfeature in the tree view.

• StageFluidZone - (Optionally) the MasterProfile sketch is revolved and cut into a sector by theperiodic surface to form the StageFluidZone body. This feature forms a sector of the fluid volumearound a single blade, but the blade has not been removed from the volume.

• Enclosure - When the StageFluidZone is created, the blade (and any other connected geometry)is removed from the StageFluidZone body by the Enclosure feature.

• Named Selections - These are the labeled regions on the final Enclosure body: Blade , Hub, Shroud ,Inflow , Outflow , PeriodicA and PeriodicB .

4.5.1. Limitations of the ImportBGD Feature

There is a known limitation with the ImportBGD feature when importing multiple BladeGen files. If youhave imported two or more BladeGen files using separate ImportBGD features, and have turned onshroud clearance for one of these features, then the import process may fail. The workaround is to importthe case(s) with shroud clearance first, then import the others.

Furthermore, changing the Blade Design cell Shroud Clearance property from "Relative Layer" or "AbsoluteLayer" to "None" will have no effect on the ImportBGD feature. In this case, you must change the ShroudClearance property directly in the ImportBGD feature.

4.6. Loading and Modifying Blades from ANSYS BladeGen

If you create a blade from scratch in BladeEditor, or if you convert a BladeGen BGD file into BladeEditor

features (that is, load a BGD file) by clicking , then you can modify the blade by editing anyof the features in the Tree Outline. By contrast, if you instead import a blade from BladeEditor (for ex-

ample, by clicking (see Importing Blades from ANSYS BladeGen (p. 71))), the blade will bedependent on the BladeGen BGD file, but you can still modify the HubProfile sketch that appearsunder the associated MerPlane feature.

Note

If you edit the hub sketch, make sure to maintain a closed edge loop when modifying thissketch or the hub body will fail to be generated. You can check the loop while editing thesketch by right-clicking in the viewer, choosing Select Loop/Chain and then selecting anedge of the loop. This will highlight the loop and let you inspect it in order to make surethat it is uniquely closed.



Loading and Modifying Blades from ANSYS BladeGen

You can modify the blade geometry using any of the standard DesignModeler geometry features. Someexamples of the latter are:

• Adding edge blends to the blade geometry

• Trimming the blade and/or hub bodies

• Adding extrusions

4.7. Using and Exporting Blades

Once the model is ready in DesignModeler, you can quickly analyze it in the Mechanical application.BladeEditor can also construct a solid model of the periodic flow passage that can be used to performa CFD analysis of the flow path geometry.

When needed, you can use DesignModeler to export the blade model to the Parasolid, IGES, or STEPfile format to bring it into your native CAD system.

You can also use BladeEditor to export the blade model for use in ANSYS TurboGrid or Vista TF.

The following topics are discussed:4.7.1. Export to Vista TF (.geo)4.7.2. Export as Meanline Data (.rtzt file)4.7.3. Export to ANSYS TurboGrid

4.7.1. Export to Vista TF (.geo)

The VistaTFExport feature defines the parameters necessary to write the flow path geometry data (.geofile) for Vista TF. The .geo file contains a 2D mesh of the flow path geometry, including the locationsof the blade leading and trailing edges. The .geo data also includes the blade type, blade camber surfacedata, blade thickness data and blade count for each blade row.

If the ThroatArea feature (see Throat Area Feature (p. 68)) is used for a selected Blade, the throat inform-ation for that blade will be written to the .geo file. This information may improve the calculation of thechoke mass flow rate in the Vista TF solver. Without this information, Vista TF will make its own estimateof the throat area.

The feature properties are listed below.

Table 4.4: Details View for Meanline Export


[Feature Name]Export Points

[Blade feature selection]Blade(s)

[Yes | No]Export to file

[.geo filename](If “Export tofile” = “Yes”)

Geo FileName

[Integer Value > 2]StreamwiseMesh Count


BladeEditor

[Integer Value > 1]SpanwiseMesh Count

Selected Blades: [number]a

[Blade reference]Blade 1

[Blade reference]Blade 2

[Blade reference]Blade 3aBlade items are listed for all selected blades.

Note

The Vista TF solver expects the flow path (from inlet to outlet) to be oriented in the directionof the machine axis, which is the Z-axis.

The VistaTFExport feature requires you to select the Blade features that will be exported to the .geofile. The selected blades must all belong to the same flow path. Therefore, the VistaTFExport featurecan only export the data for a single flow path. Only the selected blades in the flow path will be con-sidered for export; all other blades in the flow path will be ignored.

The ‘Export to file’ option specifies whether to explicitly save the data to a .geo file. When using theANSYS Workbench project schematic, exporting to a file is unnecessary because the data is transferredautomatically. If the .geo file is explicitly written, then you need to specify the .geo file name.

The .geo file contains the geometry coordinates for the flow path in terms of quasi-orthogonal (q-o)‘lines’ running from hub to shroud. The q-o lines are spaced approximately uniformly from inlet tooutlet, based on the mid-span meridional length. The number of q-o lines is specified by the ‘StreamwiseMesh Count’ property, where the minimum count is determined based on the number of selectedblades in the flow path. If the specified Streamwise Mesh Count is less than the calculated minimum,then the calculated minimum will be used.

The number of q-o lines has a direct impact on the throughflow calculation accuracy and computationtime. Increasing the number can improve accuracy, but it will also increase computation time. Increasingthe number beyond a certain maximum may cause instabilities in the convergence process. Typical ra-dial impellers with an axial inducer should be calculated with around 15 q-o lines in the bladed region,which equates to approximately a q-o line every 5° of curvature.

The Spanwise Mesh Count property specifies the number of points used to define each quasi-orthogonalline; the points are uniformly spaced from hub to shroud. The minimum number is two. Increasing thisparameter will improve the exported geometry representation for highly curved blades, but it shouldnot be necessary to go beyond about 30. This parameter has no effect on the number of streamlinesthat are actually used in the throughflow calculation.

When the .geo file is generated, a warning will be given for the VistaTFExport feature if the streamwiseto spanwise aspect ratio of any of the mesh elements is greater than 15 or less than 1/15. Dependingon the aspect ratio, you will be suggested to increase or decrease the appropriate streamwise orspanwise mesh count.

4.7.2. Export as Meanline Data (.rtzt file)

To export a file that contains the camberline/thickness definitions for a given blade row, create an Ex-portPoints feature as follows:



Using and Exporting Blades

1. Click the icon.

2. Set Export Type to Meanline .

3. Set File Name to indicate the name of the file that is to be exported.

4. Select a blade to indicate the blade row for which data is to be exported.

5. Specify which layers are to be used in generating the meanline data.

6. Click Generate .

Upon generation, the meanline data for the applicable blade row is written to a file that containsr, theta, z, and normal thickness data. Note that this file will not appear in the Files view of Work-bench.

Meanline data is exported based on the camber surface and thickness surface data represented internally.Therefore, this data may not be representative of the blade if solid model operations have been madeto the blade geometry.

An ExportPoints (Meanline) feature defines the point output for exactly one blade row. A multi-bladerow model would require multiple ExportPoints features to output data for more than one blade row.All blades (the main blade and all existing splitter blades) in a single blade row are exported by a singleExportPoints feature. Regardless of whether you select a main blade or a splitter blade when definingthe Blade parameter of an ExportPoints feature, the Blade parameter will always display the main blade.If you suppress any blade in a blade row, the ExportPoints feature for that blade row will also be sup-pressed.

The feature properties are listed below:

Table 4.5: Details View for Meanline Export



[Meanline | TurboGrid]Export Type

[File name]File Name

[Blade feature selection]Blade

[Uniform in M | Uniform in camber-line length]

Point Distribu-tion

[Value>1]Number ofPoints

Output Layers:a

[Yes | No]Output Layer1?



aLayer items are listed for all layers defined in the Flow Path.


BladeEditor

For the Meanline export type, all blades in the selected blade row are exported. The ‘Number of Points’property specifies how many points are exported from the leading edge to the trailing edge for eachblade. The number of points upstream and downstream of the blade(s) is determined by the lengthtolerance and the shape of the flow path contours.

The Blade selection lets you select the Blade feature of interest. However, all blades in a blade row areexported.

4.7.3. Export to ANSYS TurboGrid

To export blade profile data to ANSYS TurboGrid (so that you can use TurboGrid to generate a meshsuitable for CFD analysis), do the following:

• In BladeEditor, for each blade to be exported, create an ExportPoints feature as follows:

1. Click the icon.

2. Set Export Type to TurboGrid .

3. Optionally set the 'Export to file' property to 'Yes' and fill in a prefix to be used in naming the outputfiles.

It is not necessary to export the profile data to files unless you want to use this data in an externalapplication.

4. Select a Blade feature or FlowPath feature.

If you select a Blade feature then BladeEditor will attempt to select the pressure side, suctionside, leading edge, and trailing edge surfaces of the blade. If it does not do this automatically,or if you select a FlowPath feature, then you must manually select the blade surfaces from whichto extract the blade profile data. To do this, click the cell next to Blade Surfaces and then selectthe surfaces of the blade; these surfaces must collectively cover the entire blade surface exceptfor the hub-facing and shroud-facing surfaces.

5. Specify which layers are to be used in generating the blade profile data.

6. Click Generate .

• On the Project page, link the Geometry cell to the Turbo Mesh cell of a TurboGrid system.

The blade profile (point) data is automatically extracted from the blade solid model at the eligible specifiedoutput layers by slicing the blade surfaces to give profile (intersection) curves. The profile curves arethen tessellated into piecewise linear segments using the Point Tolerance value. This value defines themaximum deviation (in model units) between the piecewise linear curve and the profile curve. Thepiecewise linear curve points are output to TurboGrid or to the specified output file.

Regarding layer eligibility (for use in creating profiles for output) and the use of blade tips and hub/shroudcuts:

• If you select a Blade feature and it has a shroud tip, the eligible layers exclude any layers past the shroudtip; layers lying on or above the shroud tip will not be used to generate profiles for export.




The export of the tip profile itself can be controlled with the “Export Shroud Tip” option of the Ex-portPoints feature.

Tip clearance settings have no effect on the exported hub and shroud curves.

• If the FlowPath (that is selected directly or that is associated with the selected Blade) has a hub and/orshroud cut applied, the eligible layers exclude any layers past the cut(s); layers lying on or below the hubcut, or on or above the shroud cut, will not be used to generate profiles for export.

If the FlowPath has a hub and/or shroud cut applied, then the hub and shroud curves that are exportedwill be those used to define the cut.

Because profiles are not exported for the Hub Cut or Shroud Cut layers, if the Hub Cut or Shroud Cutis an existing FlowPath layer, this layer will not be exported. The workaround is to ensure the neigh-boring FlowPath layer to the Hub Cut or Shroud Cut layer is sufficiently close so that TurboGrid canrecreate the blade surface.

Selecting ineligible layers will result in a warning message associated with the ExportPoints feature.

By default, the profiles are extracted from all eligible layers defined in the applicable FlowPath: thespecified FlowPath feature or, if a Blade feature is specified, its parent FlowPath feature.

If output is specified on the hub and/or shroud layers (these layers are in the FlowPath feature by default),then a small offset, 'Hub/Shroud Offset %', is required in order to guarantee a complete intersection ofthe layer with the blade; such an offset is applied by default. The offset applies inwards in the spanwisedirection. The offset only affects the location where the point data is extracted from the blade; it doesnot cause the blade profile at the hub or shroud to be applied at the offset location.

Care must be taken to ensure that sufficient output layers are used when exporting so that the TurboGridblade surface is a reasonable approximation of the BladeEditor solid model. Note that using manyoutput layers with a small point tolerance can cause Bspline surface lofting problems in TurboGrid. Inthese cases, you may need to override the Blade Geometric Representation in TurboGrid and changethe Curve and Surface Types to 'Piece-wise linear' and 'Ruled', respectively.

You will need to use one ExportPoints feature for each blade that you want to export, even if they arein the same blade row. For example, if you have a main blade and a splitter, you will need to createtwo ExportPoints features and select the appropriate blade surfaces for each feature.

You specify which exported blade(s) to load into TurboGrid by editing the Turbo Mesh cell properties.See Table 9.1: TurboGrid Turbo Mesh Cell Properties (p. 201) for more information.

Because the blade profile data is extracted from the blade surfaces, this feature can also be used toextract blade profile data from an imported CAD model. The only requirement is that a FlowPath featuremust be defined to specify the output layers. If you specify a FlowPath as the Blade or Flow Pathproperty, you will need to specify the number of blades and the blade row number. The blade rownumber lets TurboGrid know how the exported blades are ordered, and whether any exported bladesare part of the same blade row. The blade rows are sorted in ascending order from inlet to outlet.

The feature properties are listed below:

Table 4.6: Details View for Export to ANSYS TurboGrid




BladeEditor

[Meanline | TurboGrid]Export Type

[Yes | No]Export to file

[Folder name](If “Export tofile” = “Yes”)

File Folder

[Filename prefix](If “Export tofile” = “Yes”)

File Prefix

[FlowPath or Blade featureselection]

Blade or FlowPath

[Blade | User Specified]Blade InfoFrom

This will be set to “User Spe-cified” if “Blade or Flow Path”is set to the name of aFlowPath feature (as opposedto a Blade feature).

[Value>1](If “Blade InfoFrom” =“User Spe-cified”)

Number ofBlades

[Int Value](If “Blade InfoFrom” =“User Spe-cified”)

Blade RowNumber

[Named Selection for bladesurfaces]

Blade Sur-faces

[0<Value<100]Hub/ShroudOffset %

[Value>0]Point Toler-ance

[Yes | No]ExportShroud Tip

Output Layers:a







aLayer items are listed for all layers defined in the FlowPath.

4.8. Blade Parameterization

ANSYS DesignModeler supports the ability to specify input parameters for sketch dimensions and featureproperties that accept real values. Through the Parameter Manager, these parameters can be used tomake parametric changes to the model, for example, in a design exploration study or for optimization.For information about the use of parameters in DesignModeler, see DesignModeler Parameters in theDesignModeler User Guide.

Many BladeEditor features support parameterization so that the shape of each blade can be controlledthrough input parameters, including the meridional flow path, the angle/thickness definition for eachdefining camberline, and the stacking arrangement of blade section sub-features.

The following sections describe the blade definition features that can be parameterized:4.8.1. Meridional Contours4.8.2. FlowPath Feature4.8.3. Blade Feature4.8.4. Camberline/Thickness Definition Sub-feature4.8.5. Splitter Feature4.8.6. StageFluidZone Feature4.8.7. ExportPoints Feature4.8.8.VistaTFExport Feature

4.8.1. Meridional Contours

The meridional contour sketches can be parameterized just like any sketch in DesignModeler. Splineshave an extra property that specifies how they should be updated when the sketch is re-evaluated.

You can apply dimensional constraints to the spline control points, so these may be parameterized.

4.8.2. FlowPath Feature

The FlowPath properties that can be parameterized are the Layer Type: Fixed Span: Span Fraction values.Only intermediate layers may be parameterized. Once the span fraction value is parameterized, thelayer cannot be deleted, unless it is de-parameterized.

4.8.3. Blade Feature

The following Blade feature properties can be parameterized:

• Number of Blade Sets (Note that this must be an integer value.)

• LE Details: Type: Ellipse: Ratios at Hub and Shroud

• TE Details: Type: Ellipse: Ratios at Hub and Shroud

• Advanced Properties: Number of Points Along the Blade (Note that this must be an integer value.)

• Advanced Properties: Number of Points for LE (Note that this must be an integer value.)


BladeEditor

• Advanced Properties: Number of Points for TE (Note that this must be an integer value.)

• If the ellipse ratio is parameterized, the LE/TE type cannot be changed.

4.8.4. Camberline/Thickness Definition Sub-feature

The CamThkDef sub-feature controls the shape of the blade profile on a given layer based on userdefined curves for the angle and thickness distributions. Because the curves can use any number ofdefining points, you can choose which points to define through parameters; not all points need to beparameterized.

To parameterize a control point, select the CamThkDef sub-feature (under a Blade or Splitter feature),right-click in the angle/thickness graph and select Use as Input Parameter, then select the controlpoint. It will appear in the details view for you to assign a parameter to the X and/or Y values, just asyou would for any other DesignModeler feature.

Basic functional behavior:

• Details view for parameters for graph points are grouped as angle or thickness parameter points. Thesepoints are displayed only on the root node, not under the Blade or Splitter feature.

• BladeEditor highlights a point on the graph when that point's parameters are selected in the details view.

• Changes to a free graph point are updated in the details view.

• Selecting a check box to set a point's X or Y value as a parameter will constrain the point's movement inthe graph view. The point will be displayed in gray if there is only one constraint, or black if both X andY are constrained.

• If the check box is cleared, the parameter will be removed from the parameter set.

• Point highlighting on the graph is cleared if the active feature is changed.

• In the details view that shows the input points, you can right-click for a context menu that contains theDelete command, which deletes an input point. This menu item is available only if the check boxes forboth the X and Y values are cleared.

• If a Beta Definition is specified, you can parameterize the theta reference value.

• If a point in a graph is used as an input parameter, and you right-click it and delete it using the contextmenu, the corresponding point parameters are removed from the details view.

Note

The technique for setting input parameters for control points in the Stacking view is thesame as that for the control points in the Angle and Thickness views. The stacking view isdescribed at Section Stacking View (p. 42).

4.8.5. Splitter Feature

In addition to the Blade feature properties (for an Independent Splitter), the following Splitter featureproperties can be parameterized:

• Pitch Fraction



Blade Parameterization

• Angle (Offset)

If the pitch fraction or offset angle is parameterized, then the offset type cannot be changed.

4.8.6. StageFluidZone Feature

For the StageFluidZone feature, the Interface Location property for each interface can be parameterized.A unique parameter name will be created for each interface.

4.8.7. ExportPoints Feature

For the TurboGrid export, the following properties can be parameterized:

• (User Specified) Number of Blades – the most important parameter

• (User Specified) Blade Row Number

• Hub/Shroud Offset

• Point Tolerance

4.8.8. VistaTFExport Feature

For the VistaTFExport feature, the following properties can be parameterized:

• Streamwise Mesh Count (Note that this must be an integer value.)

• Spanwise Mesh Count (Note that this must be an integer value.)

4.9. Tutorial 1: Blade Editing With Emphasis On Sketches, Layers, and

Blade Comparison

This tutorial demonstrates many of the different ways to manually edit a blade in ANSYS BladeEditor,as well as how to use Blade Comparison mode to take a snapshot of the blade that you are editing.You will launch a BladeEditor model using data from a Vista CCD system. You will then use BladeEditorto manually edit the blade.

Note

At Release 14.5, Vista CCD is not supported on Linux platforms.

You are first going to set up the license correctly:

1. Open ANSYS Workbench.






BladeEditor


4.9.1. Creating the Blade

1. In ANSYS Workbench, click File > Save As.

2. Save the project as BE_tutorial1 in a suitable directory.

3. In the Toolbox view, under Component Systems, double-click Vista CCD.

A new Vista CCD system appears in the Project Schematic view.

4. Right-click the Blade Design cell in the Vista CCD system and select Update.

After a short time, the Blade Design cell should have a check-mark as shown in Figure 4.9: ProjectSchematic View (p. 87).

Figure 4.9: Project Schematic View

5. If the Properties view is not already displayed, show the Blade Design cell properties by right-clickingthe cell and selecting Properties.

6. In the Properties view, set BladeModeler units to cm.

When using this method, DesignModeler does not prompt for units; it uses whatever units you sethere.

7. In the Properties view, ensure that Geometry export style is set to Parametric , then update theBlade Design cell if it requires an update.

The parametric geometry style is useful for passing sketch dimensions as parameters to an optimizer.

Leave the other settings at default values because you are using Vista CCD only to create a geometry.The focus will be on editing this geometry in BladeEditor.

8. Right-click the Blade Design cell and select Create New > Geometry.

Over the next several seconds, a Geometry system is added in the Project Schematic view andthe geometry is created in DesignModeler.

9. Right-click the Geometry cell, and select Edit Geometry... to launch DesignModeler.

You have created the geometry using Vista CCD and are now ready to edit the geometry usingBladeEditor (which is accessed via DesignModeler).

4.9.2. Editing the Main Blades and Splitter Blades in BladeEditor

Nine blades and nine splitter blades are displayed.



Tutorial 1: Blade Editing With Emphasis On Sketches, Layers, and Blade Comparison

Now that the geometry has been loaded in BladeEditor, you can modify it. Instructions for doing thisare provided in the following sections:

4.9.2.1. Modifying the Shroud4.9.2.2. Using Blade Comparison Mode While Updating a Control Point on the Shroud4.9.2.3. Changing the Number of Blades4.9.2.4. Changing the Leading Edge4.9.2.5. Adding Camberline/Thickness Definitions on New Layers4.9.2.6. Calculating the Throat Area

4.9.2.1. Modifying the Shroud

In this section, you will change the shroud in various places. Some of these changes will affect thegeometries of the main blade and splitter blade.

Modify the meridional sketch at the shroud as follows:

1. In the Tree Outline view, right-click Blade1 and select Suppress & All Below.

This is done so that sketches can be seen clearly without the geometry obstructing your view.

Alternatively to suppressing the blade, you can click Display Model to toggle the visibility ofthe blade.

2. Hide the Angle and Thickness views by clicking the Show/Hide Angle Pane and Show/Hide Thickness

Pane toolbar icons respectively.

3. Select the Sketching tab at the bottom of the Tree Outline view to switch to the Sketching Toolboxes.

4. Click Look At Face/Plane/Sketch and zoom in as appropriate.

You should see the following sketch in the graphics view:


BladeEditor

5. Select SY1 from the graphics view.

6. In the details view for SY1, change Reference Only? to Yes .

SY1 no longer governs the control point at the inlet end of the shroud.

7. In the Modify toolbox, select Drag.

8. Using the mouse, drag the control point at the inlet end of the shroud upward (in the local Y direction)a short distance.




The upstream portion of the shroud curve, and curves on the other layers, are modified as a result.

9. Select SY1.

10. In the details view for SY1, change Reference Only? to No.

SY1 now governs the control point at the inlet end of the shroud.

Note that you could have caused the same modification by leaving SY1 as a dimension (Reference

Only? set to No) and changing its value in the details view.

Changing a key dimension, such as SY1, usually causes multiple control points to move. Thistechnique enables you to make large changes to the geometry quickly.

11. Select SY2.

12. In the details view for SY2, change Value to 3cm.

This change to the shroud affects both the main blade and the splitter blade.

13. Select the Modeling tab.

14. In the Tree Outline view, right-click Blade1 and select Unsuppress & All Below.


BladeEditor

4.9.2.2. Using Blade Comparison Mode While Updating a Control Point on the Shroud

Blade Comparison mode enables you to create a snapshot of the blade. After you create the snapshot,you can make changes to the blade and then compare the current version of the blade to the snapshot.

In this section, you will:

• Take a snapshot of the blade

• Modify the blade angle

• Modify a control point on the shroud

• View the changes you have made

Start Blade Comparison mode.

1. Click Create snapshot .

A snapshot of the blade in its current state is now stored in BladeEditor.

Note that BladeEditor can only create a snapshot when the FlowPath, Blade, and Splitter featureshave been successfully generated.

2. Click Display Model to hide the geometry.

This is done so that the modifications made to the blade can be seen clearly.

3. Click Comparison On/Off to toggle the visibility of the snapshot.

Modify the blade angle.

1. In the Tree Outline, select Blade1 > Blade1_Camberline1

2. In the Angle view, drag a control point to change the shape of the curve.




Note that Blade Comparison mode shows you the shape of the curve from the snapshot as wellas the present shape of the curve.

Modify a shroud control point.

1. On the Modeling tab, select MerPlane1 > ShroudControl1 .

This is a sketch that controls the location of control points by enforcing horizontal and verticalcontrol point coordinates via two sets of fixed angles.

2. Click the Sketching tab.

3. Select the angular dimension S1XA3.

4. In the details view for S1XA3, change Reference Only? to Yes .

S1XA3 is no longer a dimension on the horizontal position of the third control point from the inletend of the shroud (counting from the inlet end). That control point can now move horizontally.

Alternatively to changing the dimensions to Reference Only, you could have deleted the entirecontrol sketch. Note that there is no “undo” function available for this action.

5. In the Modify toolbox, select Drag.

6. Using the mouse, drag the indicated point (circled in the figure below) to the left (opposite the localX and global Z directions) a short distance.


BladeEditor

Note that Blade Comparison mode shows you the shape of the shroud from the snapshot (shownin pink) as well as the present shape of the shroud.

Instead of dragging the point, you can cause the same modification by leaving S1XA3 as a dimension(Reference Only? set to No) and changing its value in the details view.

Note

Changes to the flow path can affect the main blade and the splitter blade, depending onwhich portions of the flow path are changed.

Click Generate to compare the snapshot of the geometry to the present geometry in the 3Dviewer.

After the geometry has been generated, you are able to view the differences between the snapshot(shown in pink) and the present geometry. Note that you may have to rotate the geometry and zoomin to see the differences in the geometry.




Click Comparison On/Off to turn off Blade Comparison mode.

Note

To control the snapshot display settings in Blade Comparison mode, click Comparison display

option .

4.9.2.3. Changing the Number of Blades

Generate and view the blade row, then reduce the number of blade sets to seven.

1. On the Modeling tab, select Blade1 .

2. In the Details view for Blade1 , change FD1, Number of Blade Sets to 7.

3. Click Generate .

There are now seven main blades and seven splitter blades.

4.9.2.4. Changing the Leading Edge

Adjust the shape of the leading edge of the main blade.

1. In the Details view for Blade1 , ensure that Leading Edge Details > Type is set to Ellipse .

2. Set FD2, Ellipse Ratio at Hub to 1.5 .

3. Set FD3, Ellipse Ratio at Shroud to 1.0 .

4. Click Generate .

Note

You can adjust the leading edge of the splitter blade in the same way.

4.9.2.5. Adding Camberline/Thickness Definitions on New Layers

There is currently a layer at the hub and a layer at the shroud. You can confirm this by clicking Blade1in the Tree Outline view and looking at the Camberline/Thickness Definitions property group in thedetails view.

In this section you will:

• Add a new layer at a constant span fraction value, and use it to add a new camberline/thickness definitionsubfeature to the main blade and also the splitter blade

• Add a user-defined layer (which involves using a sketch to define the layer) and use it to add a newcamberline/thickness definition subfeature to the main blade

• Specify that the thickness distribution on the user-defined layer should be interpolated from the surroundinglayers.


BladeEditor

Insert a new layer and camberline/thickness definition and interpolate it from the existing layers:

1. In the Tree Outline view select FlowPath1 .

The Details view for FlowPath1 is displayed.

2. Right-click Layer Details: 2 and select Insert Layer Above.

3. Set Span Fraction 2 to 0.3 .

4. Click Generate .

The new layer is now ready to be used by any of the blades.

5. In the Tree Outline view select Blade1 .

Note that the tree shows two camberline subfeatures under Blade1 . Although you have addeda layer to the flowpath used by the blade, the layer is not initially being used to define a camber-line/thickness definition.

6. In the details view for Blade1 , under Camberline/Thickness Definitions, set Layer 2 to Yes .

This causes a camberline/thickness definition to be created on that layer, as can be confirmed inthe tree view, which now shows three camberline subfeatures under Blade1 .

7. In the details view for Splitter1 , under Camberline/Thickness Definitions, set Layer 2 to Yes .

This causes a camberline/thickness definition to be created on that layer, as can be confirmed inthe tree view, which now shows three camberline subfeatures under Splitter1 .

8. Click Generate .

Create a sketch to be used in the definition of a user-defined layer:

1. In the Tree Outline view, right-click Blade1 and select Suppress & All Below.

This is done so that sketches can be seen clearly without the geometry obstructing your view.

2. Select MerPlane1 in the Tree Outline view.

3. Click New Sketch .

A new sketch, Sketch14 , appears in the Tree Outline view under MerPlane1 .

4. Ensure that Sketch14 is selected in the Tree Outline view.

5. Select the Sketching tab at the bottom of the Tree Outline view to switch to the Sketching Toolboxes.


7. Draw an open-ended curve (for example, a line, polyline, spline, or combination thereof ) that passesthrough the blade passage from inlet to outlet, between the previously-added layer and the shroud,without crossing any existing layers.




If you use more than one curve to produce a chain of curves, test that the chain is unbroken.

Next, use the sketch to create a user-defined layer:

1. On the Modeling tab, right-click FlowPath1 and select Edit Selections.

2. Click Sketch14 in the tree view, then, in the details view, click the field beside Sketches for Defined

Layer, then click Apply.

3. Click Generate .

A new layer appears in the Layer Details portion of the details view.

The span fraction value is an average of the span value over the sketch used to define the layer.

Next, make a camberline/thickness definition for Blade1 using the user-defined layer:


Note that the tree shows three camberline subfeatures under Blade1 . Although you have addeda layer to the flowpath used by the blade, the layer is not initially being used to define a camber-line/thickness definition.

2. In the details view for Blade1 , under Camberline/Thickness Definitions, set Layer 3 to Yes .

This causes a camberline/thickness definition to be created on that layer, as can be confirmed inthe tree view, which now shows four camberline subfeatures under Blade1 .

Next, make the angle and thickness data on the user-defined layer non-editable, so that it is always in-terpolated:

1. In the Tree Outline view select Blade1 > Blade1_Camberline3 .

2. In the details view, set Angle Definition and Thickness Definition to Interpolated .

This makes the Angle view and Thickness view read-only. Angle data and thickness data are nowinterpolated from the adjacent camberline-thickness definitions. The control points disappear fromthe thickness graph.

3. On the Modeling tab, right-click Blade1 in the Tree Outline view and select Unsuppress & All Below.

The blade row should be displayed.

Note

If you wanted to add more user-defined layers, you could add more sketch curves by doingeither or both of the following:

• Add more curves to an existing sketch that is already used by the FlowPath feature to defineat least one layer.

• Create new sketches, and then add them to the FlowPath feature as sketches to be used forlayers (under the Sketches for Defined Layer property).


BladeEditor

Remember that, before you can add sketches to the FlowPath feature, you must right-clickthe FlowPath feature in the Tree Outline view and select Edit Selections.

4.9.2.6. Calculating the Throat Area

The ThroatArea feature enables you to calculate the minimum passage area between the blades. Youare going to calculate the throat area at the leading edge of the main blade. The throat area is usefulfor choke calculations.

1. Click ThroatArea .

2. Click the cell next to Blade, and select Blade1 in the Tree Outline view.

3. Click Apply.

4. Click Generate .

A surface should be displayed in the graphics view, indicating the recently calculated ThroatArea. Thecalculated area along with other information is displayed in the Details view. For more information onthis feature, see Throat Area Feature (p. 68).

4.9.3. Summary

This tutorial has demonstrated how to manually edit a blade in ANSYS BladeEditor, as well as how touse Blade Comparison mode to take a snapshot of the blade that you are editing. For information onhow to export the geometry to various meshing programs, see Tutorial 2: Blade Editing With EmphasisOn Camberline and Thickness Distributions (p. 97).

4.10. Tutorial 2: Blade Editing With Emphasis On Camberline and Thick-

ness Distributions

This tutorial demonstrates some of the features in ANSYS BladeEditor, with emphasis on editing thecamberline and thickness distributions. You will launch a BladeEditor model using data from a VistaCCD system, make a few modifications to the blade in BladeEditor, and export the geometry for use inANSYS TurboGrid. Additionally, the tutorial provides information on how to bring the geometry intoANSYS Meshing.

Note

At Release 14.5, Vista CCD is not supported on Linux platforms.

You are first going to set up the license correctly:

1. Open ANSYS Workbench.





Tutorial 2: Blade Editing With Emphasis On Camberline and Thickness Distributions




4.10.1. Creating the Blade

1. In ANSYS Workbench, click File > Save As.

2. Save the project as BE_tutorial2 in a suitable directory.

3. In the Toolbox view, under Component Systems, double-click Vista CCD.

A new Vista CCD system appears in the Project Schematic view.

4. Right-click the Blade Design cell in the Vista CCD system and select Update.

After a short time, the Blade Design cell should have a check-mark as shown in Figure 4.10: ProjectSchematic View (p. 98).

Figure 4.10: Project Schematic View

5. If the Properties view is not already displayed, show the Blade Design cell properties by right-clickingthe cell and selecting Properties.

6. In the Properties view, set BladeModeler units to cm.

When using this method, DesignModeler does not prompt for units; it uses whatever units you sethere.

7. In the Properties view, ensure that Geometry export style is set to Interactive , then update theBlade Design cell if it requires an update.

The Interactive geometry export style uses a few key dimensions to set basic machine data. Thisexport style simplifies the manual blade design process by enabling you to drag the non-key di-mensions to modify the blade.

Leave the other settings at default values because you are using Vista CCD only to create a geometry.The focus will be on editing this geometry in BladeEditor.

8. Right-click the Blade Design cell and select Create New > Geometry.

Over the next several seconds, a Geometry system is added in the Project Schematic view andthe geometry is created in DesignModeler.

9. Right-click the Geometry cell, and select Edit Geometry... to launch DesignModeler.


BladeEditor

You have created the geometry using Vista CCD and are now ready to edit the geometry usingBladeEditor (which is accessed via DesignModeler).

4.10.2. Editing the Main Blades and Splitter Blades in BladeEditor

Nine blades and nine splitter blades are displayed.

Now that the geometry has been loaded in BladeEditor, you can modify it. Instructions for doing thisare provided in the following sections:

4.10.2.1. Adding Blade Clearance4.10.2.2. Modifying Camberline/Thickness Definitions at the Hub and Shroud4.10.2.3. Modifying the Thickness Distribution of the Splitter Blade

4.10.2.1. Adding Blade Clearance

Look at the sketch:

1. In the Tree Outline view, click MerPlane1 .

This will display the blade sketch if it is not already displayed.


BladeEditor shows a sketch with a small number of key dimensions that are used to set the basic machinedata.

You can manually modify the sketch by clicking the sketching tab and using tools in the various tool-boxes. Tutorial 1: Blade Editing With Emphasis On Sketches, Layers, and Blade Comparison (p. 86) de-scribes some of the methods available for modifying geometry in BladeEditor.

Add blade clearance to the main blade and the splitter blade:

1. In the Tree Outline view select FlowPath1 .

The Details view for FlowPath1 is displayed.

2. Set Span Fraction 4 to 0.95.


4. In the details view, set Shroud Tip Clearance to Layer .

5. Set Shroud Tip Layer to 4.

6. Click Generate .

The main blade and splitter blade now extend to layer 4.

4.10.2.2. Modifying Camberline/Thickness Definitions at the Hub and Shroud

In this section, you will make changes to the angle and thickness distributions of the main blade asfollows:

• Adjust the camberline at the hub by changing the leading edge Beta angle




• Set the thickness distribution at the hub to a constant value from leading edge to trailing edge

• Adjust the camberline at the shroud, and use a Bezier spline to control the Beta distribution

Note

These changes will also affect the splitter blade because the splitter blade currently inheritsthe angle and thickness distributions from the main blade. Later in this tutorial, you willchange the thickness distribution of the splitter blade without affecting the main blade.

Start by adjusting the camberline at the hub:

1. In the Tree Outline view, select Blade1 > Blade1_Camberline1 .

The Angle and Thickness views for Blade1_Camberline1 are displayed.

The Angle view displays one or two curves, one of which has control points along it. The curvewith control points shows whatever is specified by the Angle Definition Type and X-Axis Definition

Type properties of the blade. In this case, the curve with control points shows the distribution of

Beta versus M′.

2. In the Angle view, double-click the leftmost point on the graph to manually type in its coordinates.

3. Set the coordinates to 0, 40 and press Enter.

Note

If you drag any of the points using the mouse, the curve will be adjusted accordingly. If anyother changes are made, for example, by using the shortcut menu, you must then click thecurve to update it.

Next, set the thickness at the hub to a constant value from leading edge to trailing edge, as follows:

1. Right-click anywhere in the Thickness view, and select Delete Point.

2. In the Thickness view select the point near the middle of the graph to delete it.

A straight line should now form, because there are two points remaining.

3. Double-click the leftmost point to manually set its coordinates.

4. Set the coordinates to 0, 0.18 and press Enter.

The straight line should now be completely horizontal, as in Figure 4.11: Angle and Thickness Viewsat Hub After Modifications (p. 101) below.


BladeEditor

Figure 4.11: Angle and Thickness Views at Hub After Modifications

5. Click Generate .

Finally, adjust the camberline at the shroud, as follows:

1. In the Tree Outline view select Blade1 > Blade1_Camberline5 .

The Angle and Thickness views for Blade1_Camberline5 are displayed.

2. Right-click anywhere in the Angle view, and select Convert to Bezier.

3. In the ANSYS BladeEditor dialog box that appears, set the number of points to 6 and click OK.




4. Click the angle curve that has control points.

That angle curve becomes a Bezier spline.

5. Double-click the fourth point (counting from the left) to manually set its coordinates.

6. Set the coordinates to 0.5, 10 and press Enter.

7. Click Generate .

You have changed the angle and thickness distributions of the main blade; these changes have beeninherited by the splitter blade.

4.10.2.3. Modifying the Thickness Distribution of the Splitter Blade

After the main blade has been designed, the splitter blade will already have the same angle and thicknessdistributions by default. In this section, you will modify the thickness distribution of the splitter bladealone without affecting the main blade.

1. In the Tree Outline view select Splitter1 > Splitter1_Camberline1 .

2. Set Thickness Definition to User Specified .

3. Right-click anywhere in the Thickness view, and select Convert to Bezier.

4. In the ANSYS BladeEditor dialog box, set Number of Points to 4.

5. Click the curve (not a point on the curve) in the Thickness view.

The curve is replaced by a similar curve with fewer control points.

6. Drag a control point to change the shape of the curve.

7. Click Generate .

You have changed the thickness distribution of the splitter blade without affecting the main blade.

Note

Setting Thickness Definition to From Reference Blade and then to User Specifiedhas the effect of resetting the thickness distribution to that of the main blade.

4.10.3. Looking at the Auxiliary Views

The Auxiliary view can show either the blade to blade geometry or the blade lean angle at a giventime.

Make the Auxiliary view visible as follows:

1. Click Show/Hide Auxiliary Pane as required to show the Auxiliary view.

2. Select Blade1 from the Tree Outline view.


BladeEditor

The Blade to blade view is displayed for Blade1 .

3. Right-click inside the Blade to blade view and select Blade Lean Angle.

The Lean Angle view is displayed for Blade1 .




The Blade to blade and Lean Angle views are updated if changes are made in the Angle or Thickness

views, after you click Generate .

Note

You can view the blade to blade and blade lean angle auxiliary views for the splitter bladein the same way.

4.10.4. Exporting Geometry to ANSYS TurboGrid

To proceed to the meshing stage of your design, the geometry may be exported to TurboGrid.

1. Click ExportPoints .

A new feature ExportPoints1 is created, and its Details view is displayed.

2. Click the cell next to Blade or Flow Path and select Blade1 in the Tree Outline view.

3. Click Apply.

BladeEditor automatically selects the pressure side, suction side, leading edge, and trailing edgesurfaces of the blade.

4. Note that Export Shroud Tip is set to Yes .

This will cause information about the blade tip to be exported.

5. Click Apply.

6. Click Generate .

7. Create a second ExportPoints feature, ExportPoints2 , for the splitter blade, Splitter1 .

BladeEditor automatically selects the pressure side, suction side, leading edge, and trailing edgesurfaces of the blade.

Two ExportPoints features have been specified, and are now ready to be used by TurboGrid.

• Return to ANSYS Workbench. In the Toolbox view under Component Systems, drag TurboGrid ontothe Geometry cell.

A TurboGrid system appears with the Geometry cell linked to the Turbo Mesh cell.

It is possible to proceed with meshing in TurboGrid from this point by double-clicking the Turbo Mesh

cell. Your Project Schematic view should look like the one in Figure 4.12: Project Schematic View withLink to TurboGrid (p. 105).


BladeEditor

Figure 4.12: Project Schematic View with Link to TurboGrid

4.10.5. Using ANSYS Meshing

An alternate method to proceed with the meshing process is to use ANSYS Meshing.

1. In BladeEditor, click StageFluidZone .

The Details view for this appears.

2. Click the cell next to Flow Path and select FlowPath1 in the Tree Outline view.

3. Click Apply.

4. Click Generate .

Note

If the generation fails, follow these steps:


The Details view for Blade1 is displayed.

2. Change Blade Extension (%) to 5.

3. Click Generate .

The newly created StageFluidZone1 feature will be useful for bringing the geometry into ANSYSMeshing. To do this:

• Return to ANSYS Workbench. In the Toolbox view under Component Systems, drag Mesh onto theGeometry cell.

A new Mesh system appears with the Geometry cells being linked to one another.




It is possible to proceed with meshing in ANSYS Meshing from this point by double-clicking the Mesh

cell. Your Project Schematic view may look like the one in Figure 4.13: Project Schematic View withLinks to TurboGrid and ANSYS Meshing (p. 106) below.

Figure 4.13: Project Schematic View with Links to TurboGrid and ANSYS Meshing

4.10.6. Summary

This tutorial has demonstrated how to launch a BladeEditor model using data from a Vista CCD system,make a few modifications to the blade, and export the geometry for use in ANSYS TurboGrid or ANSYSMeshing. For more information on manually modifying the blade in BladeEditor, see Tutorial 1: BladeEditing With Emphasis On Sketches, Layers, and Blade Comparison (p. 86).


BladeEditor

Chapter 5: TurboSystem: Vista RTD

The Vista range of turbomachinery software for Windows includes 1D design and off-design performanceprograms for axial turbines, radial turbines, axial compressors, axial fans, centrifugal compressors, andcentrifugal pumps.

The subject of this user guide is Vista RTD. Vista RTD is a program for the preliminary design of radialinflow turbines. It can be used in an iterative fashion to create a 1D design. The resulting geometry canbe passed to BladeGen, BladeEditor, and Vista TF. Vista RTD can also be used to model an existing turbine.An accurate 1D model can provide insight into the performance of the machine that goes beyond thetest measurements.

Vista RTD is integrated into ANSYS Workbench so that it may be used to generate an optimized 1Dturbine design before moving rapidly to a full 3D geometry model and CFD analysis.

Vista RTD is provided by PCA Engineers Limited, Lincoln, England.

The following topics are discussed:5.1.Vista RTD Workflows5.2. Data Review and Edit5.3.Viewing the Results5.4. Context Menu Commands of the Blade Design Cell5.5. Launching a New BladeGen Model5.6. Creating a New Throughflow System5.7. Launching a New BladeEditor Model5.8. Linking to a New Vista TF Cell5.9. Using Vista RTD to Model an Existing Turbine5.10. Appendix 1: Definition of Parameters on Results Tab (Ideal and Semi-perfect Gas)

5.1. Vista RTD Workflows

There are two main ways to use Vista RTD:

• Vista RTD can be used in an iterative fashion to create a 1D design as follows:

1. With ANSYS Workbench running, create a new Vista RTD system (available in the Component Systemstoolbox).

(Double-click the Vista RTD system or drag it onto the Project Schematic view.)

2. Either Edit the Blade Design cell (of the new Vista RTD system) or show that cell's properties.

(Right-click the Blade Design cell and select either Edit or Properties from the shortcut menu.)

Editing the cell causes the Vista RTD dialog box to appear; the first two tabs of this dialog boxcontain the input data and the last two tabs display the results.



The cell properties can be used as an alternative to the aforementioned two tabs that acceptinput. One advantage of using cell properties is that you can make use of ANSYS Workbenchinput parameters to specify the values of input settings.

3. Specify the required input settings:

– Information about the operating conditions such as mass flow rate, expansion ratio, and rotationalspeed

– The fluid properties

– Geometric constraints such as the mean vane thickness at the throat

– Some aerodynamic and geometric settings (for example, the number of vanes) that are open tochange during this 1D design procedure.

See Data Review and Edit (p. 108) for details on the input settings.

4. Compute a turbine design by clicking Calculate.

5. Assess the results.

See Viewing the Results (p. 112) for descriptions of the results.

6. Continue to revise the input data and refresh (regenerate) the results until you obtain a satisfactory1D design.

• Vista RTD can be used to model an existing turbine. This use of Vista RTD is described in Using Vista RTDto Model an Existing Turbine (p. 116).

5.2. Data Review and Edit

The input data can be specified on the Aerodynamics and Geometry tabs of the Vista RTD dialog box,and also (or alternatively) in the Properties view for the Blade Design cell. The Properties view is shownin Figure 5.1: Properties View for a Blade Design cell of a Vista RTD system (p. 109).


Vista RTD

Figure 5.1: Properties View for a Blade Design cell of a Vista RTD system

Note

All input and result parameters are tabulated in Appendix 1: Definition of Parameters onResults Tab (Ideal and Semi-perfect Gas) (p. 116).

Input data may be declared as parameters via the Properties view. Once an input is designated as aparameter, its value can be modified only via the Parameter Manager.

The data defining the case are also displayed in the first two tabs of the user interface as follows:5.2.1. Aerodynamics Tab



Data Review and Edit

5.2.2. Geometry Tab

5.2.1. Aerodynamics Tab

Figure 5.2: Aerodynamics Tab

Frame 1: Operating conditions

This frame contains the basic flow parameters at the design point:

• Inlet stagnation temperature

• Inlet stagnation pressure

• Mass flow rate

• Expansion ratio t-t

• Rotational speed

• Blade speed ratio (U/C t-t)

Frame 2: Stage efficiency

This panel contains the stage isentropic efficiency options.

The default option is User specify; if this remains selected, enter the value of total-total stage efficiency

in the text box. Alternatively, if the other option, Activate correlation is chosen, then the program will,using internal empirical expressions for losses, calculate an efficiency value and use it in the design.

Frame 3: Fluid properties

This panel contains the gas property options.

To define the fluid properties, you can either input fixed values of gas constant (R), specific heat atconstant pressure (Cp), or request that the program find its own values. The latter facility is selectedby clicking either the Air or the Air fuel ratio option. The program will then compute Cp at each stationthrough the turbine from a polynomial expression in terms of local temperature (and, if selected, air-


Vista RTD

fuel ratio) for pure air or for a combustion products mixture defined by the air-fuel ratio. The programalso then provides a value for gas constant.

Frame 4: Flow Angles

This frame contains the Inlet Angle and the Exit Angle frames.

• The Inlet Angle frame contains the data items Zero relative inlet angle and Calculate from nozzle area.

– Zero relative inlet angle indicates zero incidence at the leading edge. When this option is selectedthe absolute inlet angle must be specified, but the exit angle is calculated by the code.

– Calculate from nozzle area indicates that the inlet angle is calculated from the nozzle area (specifiedon the Geometry Tab (p. 111)). With this option enabled there are two possible solutions, a high (oftensupersonic) relative inlet velocity and a low (normally subsonic) relative inlet velocity. The low speedoption is normally the most desirable, but sometimes a high speed solution is the only viable solution(often when using the zero relative inlet angle option above).

The inlet angle can either be specified as a relative or absolute value, except when the Zero relative

inlet angle option is selected.

• The Exit Angle frame enables you to specify the exit angle in either relative or absolute terms when youuse the Calculate from nozzle area inlet angle.

5.2.2. Geometry Tab

Figure 5.3: Geometry Tab

This tab contains all the required geometric input data.

Frame 1: Nozzle

This frame contains the following data items: Mean vane thickness at throat, Number of vanes, andThroat Area.

Frame 2: Impeller

This frame contains the following data items:




• Mean vane thickness at exit.

• Number of vanes.

• Shroud exit/inlet radius ratio. This refers to the ratio of the exducer shroud radius to the impeller inletradius

• Hub exit/inlet radius ratio. This is the ratio of the exducer hub radius to the impeller inlet radius

• Axial length. This is the length of the impeller in the axial direction as a percentage of the tip of diameter(measured at the impeller inlet).

5.3. Viewing the Results

The results are shown in the Vista RTD dialog box, which you can invoke by editing the Blade Designcell of the Vista RTD system.

To compute a turbine design, click Calculate. The Results tab is then automatically made visible (Fig-ure 5.4: Results Tab (p. 112)) and the numerical results appear in its text box.

5.3.1. Results Tab

Figure 5.4: Results Tab

This tab contains the output results data and is automatically displayed on execution of the calculation.

The results are arranged in the following main blocks:

• Input Data Summary

• Performance

• Inlet Velocities

• Outlet Velocities (at shroud)

• Impeller Geometry

• Nozzle Geometry


Vista RTD

Appendix 1: Definition of Parameters on Results Tab (Ideal and Semi-perfect Gas) (p. 116) is a quick-ref-erence for the meaning and units of all the parameters.

The velocity triangles for the computed design are shown on the Velocity triangles tab (Figure 5.5: Ve-locity Triangles Tab (p. 113))

5.3.2. Velocity Triangles Tab

Figure 5.5: Velocity Triangles Tab

This tab contains the velocity triangles, and the appropriate velocity values, for the inlet and shroudexit locations.

The data is displayed as follows:

Units

These radio buttons control whether the data (both input and results) are presented in either SI or im-perial units. The data switches as soon as the units are changed, no recalculation is required afterswitching and any subsequent recalculation will yield identical results.

Impeller sketch

This displays a simple meridional sketch of the impeller, hub, and shroud curves.

The Calculate button updates the data model with the input data in the user interface and subsequentlyperforms the calculation.

Note

When the user interface loses focus, the data model is automatically updated with the inputdata in order to keep the user interface and data model synchronized. Conversely, on receivingthe focus the user interface input data will be updated from the data model.



Viewing the Results

5.4. Context Menu Commands of the Blade Design Cell

You can access a context menu for the Blade Design cell in the Vista RTD component system by right-clicking the cell. Most of the commands that are available are standard, and are described in Systemsand Cells. The context menu commands that are specific to the Blade Design cell are described inTable 5.1: Context Menu Commands Specific to the Vista RTD Blade Design Cell (p. 114).

Table 5.1: Context Menu Commands Specific to the Vista RTD Blade Design Cell

DescriptionCommand

This command opens the Vista RTD dialog box.Edit

This command first prompts you to select aBladeGen file, then imports the Vista RTD set-

Import BladeGen File

tings from the selected file into the cell prop-erties.

The BladeGen file is required to have beenwritten by an older version of BladeGen thatincluded Vista RTD; such an older version ofBladeGen wrote Vista RTD settings into theBladeGen file. By contrast, a BladeGen file pro-duced by the Create New > BladeGen com-mand does not contain the Vista RTD data re-quired for the Import BladeGen File com-mand.

This command creates a BladeGen system thatcontains a 3D model based on the 1D model

Create New > BladeGen

in the Vista RTD system. For details, seeLaunching a New BladeGen Model (p. 114).

This command creates a Throughflow systemthat can be used to perform a throughflow

Create New > Through-flow

study using Vista TF. For details, see Creatinga New Throughflow System (p. 115).

This command creates a Geometry system thatcontains a 3D model based on the 1D model

Create New > Geo-metry

in the Vista RTD system. For details, seeLaunching a New BladeEditor Model (p. 115).

This command creates a Vista TF system andlinks it to the Vista RTD system. For details, seeLinking to a New Vista TF Cell (p. 116).

Transfer Data to New >Vista TF

5.5. Launching a New BladeGen Model

A new BladeGen model can be generated from an up-to-date Vista RTD system using the Create New

> BladeGen command in the right-click context menu of the Blade Design cell.


Vista RTD

The 1D turbine design is converted to a 3D turbine geometry model in BladeGen. To achieve the 3Dvane shape, Vista RTD creates initial guesses of vane camber and thickness distributions and these arethen combined with its computed meridional design.

Note

• The BladeGen model is detached from the Vista RTD system (that is, no link is generated).Therefore, any changes made to the Vista RTD system will not be reflected in the BladeGenmodel following an update of the Vista RTD system.

• The BladeGen file that you create in this way cannot be imported into another Vista RTD system,because it does not contain Vista RTD settings. A BladeGen file created by an older version ofBladeGen (that included Vista RTD) contains Vista RTD settings and can be imported into aVista RTD system.

5.6. Creating a New Throughflow System

You can create a Throughflow system, which is essentially a Vista TF system with an added Geometrycell, using the Create New > Throughflow command in the right-click context menu of the BladeDesign cell. The Geometry cell will be populated with the BladeEditor geometry based on the current,up-to-date parameters from Vista RTD. The Setup cell will be the same as for a Vista TF system, exceptthat the machine type and number of blade rows cannot be specified in the properties because thatinformation is taken from the Geometry cell. Upon creating a Throughflow system using data from aVista RTD system, each of the cells in the Throughflow system is updated automatically:

For details, see Vista TF User's Guide (p. 209).

5.7. Launching a New BladeEditor Model

A new BladeEditor model can be generated from an up-to-date Vista RTD system using the Create New

> Geometry command in the right-click context menu of the Blade Design cell. This command launchesANSYS DesignModeler (with BladeEditor) and generates the new model.

Note

The DesignModeler model is detached from the Vista RTD system (that is, no link is generated).Therefore, any changes made to the Vista RTD system will not be reflected in the Design-Modeler model following an update of the Vista RTD system.



Launching a New BladeEditor Model

5.8. Linking to a New Vista TF Cell

There are three ways to link the Blade Design cell of a Vista RTD system with the Setup cell of a VistaTF system:

• Drag and drop a Vista TF system from the toolbox onto the Blade Design cell of a Vista RTD system

• Drop a new Vista TF system on the project schematic and then create a link between the Blade Designcell of a Vista RTD system and the Setup cell of the Vista TF system

• Right-click the Blade Design cell of the Vista RTD system, then select Transfer Data to New > Vista TF

to transfer data to a new Vista TF system.

Once the link is in place, updating the Setup cell of the Vista TF system will populate it with the geometryand operating conditions from the Blade Design cell of a Vista RTD system, making it ready for analysiswith Vista TF.

5.9. Using Vista RTD to Model an Existing Turbine

Vista RTD can be of use in the analysis of turbine test data – or operating point data from some othersource. A 1D model of a selected operating point can be created by synthesizing the known geometryand test data.

In the Geometry section of the data, all items will be known and fixed. From the test data, the massflow, expansion ratio, and rotational speed (corrected for measured inlet temperature and pressure) areall known. By running the code in an iterative mode, the ‘unknowns’ (for example blade speed ratio)can be gradually adjusted until geometric parameters (such as impeller tip diameter, tip width, and soon) all attain their actual known values.

5.10. Appendix 1: Definition of Parameters on Results Tab (Ideal and

Semi-perfect Gas)

Imperial

units

SI

units

ParameterAbbreviated

name

INPUT DATA SUMMARY

Line 1

degFKInlet stagnation temperatureT0

lbf/in^2PaInlet stagnation pressureP0

lb/skg/smass flow rateMass

——Total- total expansion ratioExp tt

rev/minrev/minRotational speedN

——Blade speed ratioU/C

%%Stage total-static isentropic effi-ciency estimate

Eff

Line 2

inmmNozzle vane thicknessNozz Thk

——Number of nozzle vanesNum nozz


Vista RTD

Imperial

units

SI

units


name

in2mm2Nozzle throat areaNozz Area

inmmImpeller vane thickness at exitImp Thk

inmmNumber of impeller vanesNum Imp

——Exducer shroud diameter/inlet dia-meter

Shr O/I

——Exducer hub diameter/inlet diamet-er

Hub O/I

PERFORMANCE

Line 1

——Ratio of exit rms relative velocityto inlet relative velocity

W5/W4

——Ratio of exit rms axial velocity toinlet relative velocity

Vx5/U4

——Total-static expansion ratioExp ts

——Flow function/flow function chok-ing value

Q/Qchk

—kg-

T0.5/s-kPa

⋅ MrtT/P(e-5)

Btu/hwattsPowerPwr(w)

——ReactionRctn

——Blade speed ratio with exit staticisentropic conditions as reference

U/C ts

for C. C is the spouting velocity;that is, the velocity that would beobtained if the gas were expandedisentropically from the inlet stagna-tion enthalpy to the exit condition,

so is given by: = ⋅

——dh/U^2

——Specific speedNs

Line 2

LOSSES:

——Loss coefficient due to blade load-ing

Load

——Loss coefficient due to curvatureCrv

——Loss coefficient due to frictionFric

——Loss coefficient due to clearanceClear

——Total of above lossesTot

——EFF:



Appendix 1: Definition of Parameters on Results Tab (Ideal and Semi-perfect Gas)

Imperial

units

SI

units


name

——Total-static stage efficiencyStg ts

——Total-total stage efficiencyStg tt

——Total-static impeller efficiencyImp ts

——Total-total impeller efficiencyImp tt

INLET VELOCITIES

——Absolute Mach numberMabs

——Relative Mach numberMrel

ft/sm/sBlade speedU4

ft/sm/sAbsolute velocityV4

ft/sm/sRelative velocityW4

ft/sm/sAbsolute whirl velocityVw4

ft/sm/sRadial velocityVr4

degdegAbsolute flow angle, relative to ra-dial

alpha4

degdegRelative flow angle, relative to radi-al

beta4

EXIT VELOCITIES (at shroud)

Line 1

——Absolute Mach numberMabs


ft/sm/sBlade speedU5

ft/sm/sAbsolute velocityV5

ft/sm/sRelative velocityW5

ft/sm/sAbsolute whirl velocityVw5

ft/sm/sAxial velocityVx5

degdegAbsolute flow angle, relative toaxial

alpha5

degdegRelative flow angle, relative to axialbeta5

Line 2

degdegRelative flow angle at rmsbeta5rms

degdegRelative flow angle at shroudbeta5shr

degdegRelative flow angle at hubbeta5hub

IMPELLER GEOMETRY

inmmInlet diameterd4

inmmTip widthtip width

inmmExducer diameter at hubd5hub

inmmExducer diameter at shroudd5shr

——Ratio of inlet diam/exit diam at rmsd4/d5rms


Vista RTD

Imperial

units

SI

units


name

——Vane chord / pitchSolidity

NOZZLE GEOMETRY

——Flow function/flow function chok-ing value

Q/Qchk

inmmDiameter at trailing-edgeInner dia

inmmA/R ratio for a vaneless nozzleVless A/R



Appendix 1: Definition of Parameters on Results Tab (Ideal and Semi-perfect Gas)


Chapter 6: TurboSystem: Vista CCD


The subject of this user guide is Vista CCD. Vista CCD is a program for the preliminary design of centri-fugal compressors. It can be used in an iterative fashion to create a 1D design. The resulting geometrycan be passed to BladeGen or BladeEditor. Vista CCD can be used to model an existing compressor and,if known, its measured performance at single operating points. An accurate 1D model can provide insightinto the performance of the machine that goes beyond the test measurements.

Vista CCD is integrated into ANSYS Workbench so that it may be used to generate an optimized 1Dcompressor design before moving rapidly to a full 3D geometry model and CFD analysis.

Vista CCD is provided by PCA Engineers Limited, Lincoln, England.

The following topics are discussed:6.1.Vista CCD Workflows6.2. Data Review and Edit6.3.Viewing the Results6.4. Common Options6.5. Context Menu Commands of the Blade Design Cell6.6. Launching a New BladeGen Model6.7. Creating a New Throughflow System6.8. Launching a New BladeEditor Model6.9. Using Vista CCD to Model an Existing Compressor6.10. Predicting a Performance Map6.11. Appendix 1: Definition of Parameters on Results Tab6.12. Notation6.13. References

6.1. Vista CCD Workflows

There are two main ways to use Vista CCD:

• Vista CCD can be used in an iterative fashion to create a 1D design as follows:

1. With ANSYS Workbench running, create a new Vista CCD system (available in the Component Systemstoolbox).

(Double-click the Vista CCD system or drag it onto the Project Schematic view.)

2. Either edit the Blade Design cell (of the new Vista CCD system) or show that cell's properties.

(Right-click the Blade Design cell and select either Edit or Properties from the shortcut menu.)



Editing the cell causes the Vista CCD dialog box to appear; the first three tabs of this dialog boxcontain the input data and the last tab displays the results.

The cell properties can be used as an alternative to the aforementioned two tabs that acceptinput. One advantage of using cell properties is that you can make use of ANSYS Workbenchinput parameters to specify the values of input settings.


– The compressor duty (pressure ratio, mass flow, rotational speed)

– The gas properties

– Fixed geometric constraints (e.g. inducer hub diameter, vane thickness)

– Some geometric and aerodynamic variables which are open to choice and may be changed duringthe iterative 1D design procedure (e.g. vane number, relative velocity ratio, incidence, etc)

See Data Review and Edit (p. 122) for details on the input settings.

4. Compute a compressor design by clicking Calculate.

The 1D geometric and aerodynamic dependent parameters are computed and displayed numer-ically; a meridional view of the impeller is also displayed. If efficiency correlations have been se-lected, a chart of the correlation with the current design point superimposed is also shown.


See Viewing the Results (p. 131) for descriptions of the results.

6. Continue to revise the input data and refresh (regenerate) the results until you obtain a satisfactory1D design.

• Vista CCD can be used to model an existing compressor and, if known, its measured performance at singleoperating points. This use of Vista CCD is described in Using Vista CCD to Model an Existing Com-pressor (p. 138).


The input data can be specified on the Duty and Aerodynamic Data, Gas properties and Geometry

tabs of the Vista CCD dialog box, and also (or alternatively) in the Properties view for the Blade Designcell. The Properties view is shown in Figure 6.1: Properties view for a Blade Design cell of a Vista CCDsystem (Part 1 of 4) (p. 123) and the following related figures.


Vista CCD

Figure 6.1: Properties view for a Blade Design cell of a Vista CCD system (Part 1 of 4)






Vista CCD






Descriptions of the properties follow:

• Geometry export style

This property controls the way that sketches appear in BladeEditor. Sketches appear in one of twostyles:

– Interactive

The interactive style is useful for manual blade design. It uses a small number of blade dimensionsto set the basic machine size. Other points and edges in the sketch can be easily moved by hand.The interactive style is used in BladeEditor sketches that are generated from all Vista tools.

– Parametric

The parametric style is useful for passing sketch dimensions to an optimizer as parameters.

• All other input parameters and all result parameters are tabulated in Appendix 1: Definition of Parameterson Results Tab (p. 140). The appendix is best used as a quick reference.

The data defining the case are also displayed in the first three tabs of the user interface as follows:6.2.1. Duty and Aerodynamic Data Tab6.2.2. Gas Properties Tab


Vista CCD

6.2.3. Geometry Tab

6.2.1. Duty and Aerodynamic Data Tab

Figure 6.5: Duty and Aerodynamic Data Tab

Frame 1: Duty

This frame contains: Overall pressure ratio, Mass flow, and Rotational speed. The pressure ratio canbe total-total or total-static but then the Stage efficiency must be on the same basis.

Frame 2: Inlet Stagnation Conditions

This frame contains: The stagnation Temperature and Pressure at the inlet to the machine.

Frame 3: Inlet Gas Angle

This frame contains the options for specifying the upstream flow angle. The RMS angle is specifiedalong with a Radial distribution from hub to shroud. The distribution options are:

• Constant angle

• Free vortex

• Solid body rotation

• Linear variation of Vw

If the linear variation of the whirl velocity is selected, the Vw ratio must also be specified.

Frame 4: Incidence at Shroud

This frame contains the option to either specify the Incidence at shroud directly, or indirectly by spe-cifying the choke margin for the machine and letting the program calculate the appropriate incidenceto achieve this value.

Frame 5: Stage Efficiency




You may specify Stage efficiency (Select User specify, then choose Isentropic or Polytropic). Altern-atively, the program will determine an efficiency appropriate to the compressor duty by means of inbuiltcorrelations (if you select Correlations). There is a choice of three correlations:

• Casey-Robinson [1] (p. 144)

• Casey-Marty [2] (p. 144)

• Rodgers [3] (p. 144)

Casey-Robinson is a combination of the other two older correlations, combining data from high Machnumber compressors with data from lower-speed industrial machines. However, you may want to useone of the two classic methods, so both of these have been retained as options.

The correlations provide “base efficiency” for the following class of compressor stage:

• large size (at least 300 mm impeller diameter)

• low axial tip clearance (2% of axial tip width or less)

• unshrouded impeller

• vaned diffuser

• machined vane surface

For smaller machines or machines with higher clearances, you may select Reynolds no. correction

and/or the Tip clearance and shroud correction.

If Reynolds no. correction is selected, then you should review the Vane roughness options on theGeometry tab.

If Tip clearance and shroud correction is selected, then you should review Tip clearance or Tip

Clearance/vane height on the Geometry tab.

Corrections for vaneless diffusers and shrouded impellers may be applied by selecting the appropriateoption on the Geometry tab.

Frame 6: Impeller Isentropic Efficiency

Impeller isentropic efficiency may be specified by you or set as linked to stage, where the latter hasbeen selected to come from the correlations.

Frame 7: Power Input Factor

Power Input Factor (PIF) may be optionally specified by you or calculated automatically from a correl-ation by Vista CCD.

Frame 8: Other aerodynamic data

This frame contains: Meridional velocity gradient, and Relative velocity ratio.

Meridional velocity gradient is the ratio of the shroud station meridional velocity to the rms stationand meridional velocity at the impeller inlet plane.


Vista CCD

Relative velocity ratio is the ratio of the relative exit velocity to the inlet shroud relative velocity andso is a measure of impeller diffusion.

6.2.2. Gas Properties Tab

Figure 6.6: Gas Properties Tab

This tab contains the required gas property input data.

Frame 1: Gas Properties Model

This frame gives the following options:

• Ideal gas

• Real gas

For the Real gas option, material properties are derived from the Aungier-Redlich-Kwong equation,which is mentioned in Overview and Limitations in the FLUENT User's Guide.

You may also select whether to choose a standard material from the gas properties database, or chooseto specify the detailed material properties as shown in Figure 6.6: Gas Properties Tab (p. 129).

The list of available gases can be seen in the drop-down menu in Vista CCD. At the time of writing, thislist was:

• air

• carbon-dioxide

• hydrogen

• methane

• nitrogen

• oxygen




• parahydrogen

• propylene

• R123

• R125

• R134a

• R141b

• R142b

• R245fa

• water (steam)

Frame 2: User Specified Properties

This frame is enabled should you choose to specify the material properties. Only fields relevant to theselected gas model are enabled. This frame contains the Kinematic Viscosity and Cp polynomial

coefficients frames. For the Ideal Gas model, only the Gas constant, Gamma and Kinematic viscosity

are required. For the real gas model Gamma is no longer required; but the Critical pressure, Critical

temperature, Critical specific volume and Acentric factor along with the Cp polynomial definitionsas a function of temperature need to be specified. If the Kinematic viscosity is specified as Air, thenit is calculated using Sutherland's law as a function of temperature.

6.2.3. Geometry Tab

Figure 6.7: Geometry Tab

Frame 1: Inducer

This frame contains the parameters relating to the impeller inlet section. This is mainly separated intotwo frames:


Vista CCD

• The Hub frame enables you to specify the Diameter, Vane inlet angle and Vane normal thickness. TheVane inlet angle is optional. The radial distribution of beta angles may be calculated using either a tangentor sine formula.

• The Shroud frame contains geometry data for the inducer at the shroud. You can specify the Diameter

and the Vane normal thickness. The Diameter has the following three settings:

– You can specify the shroud diameter value directly.

– You can specify the Vane inlet angle and have the program calculate the appropriate diameter accord-ingly.

– You can select Optimise diameter where the program calculates the appropriate diameter to achievea minimum inlet relative Mach number.

In addition to the values in the Hub and Shroud frames, you may also specify a non-zero value of LE

inclination to radial in the inducer frame.

Frame 2: Diffuser

Vaned or Vaneless may be selected. Vaneless activates an empirical correction to the baseline stageefficiency obtained from correlations.

Frame 3: Impeller Shroud and Clearance Geometry

Unshrouded or Shrouded impeller may be selected. The Axial Tip Clearance can either be given asa fraction of the vane height, or directly as a specific value.

Frame 4: Other Impeller Geometry

This frame contains other miscellaneous geometry data required for the calculation, specifically thefollowing:

• Main Vanes

• Intervanes

• Backsweep Angle

• Rake Angle

• Axial Length

• Vane Roughness (choices of Machined Finish or Cast Finish)

6.3. Viewing the Results

The results are shown in the Vista CCD dialog box, which you can apply by editing the Blade Designcell of the Vista CCD system.

To compute an impeller design, click Calculate. The Results tab is then automatically made visible(Figure 6.8: Results Tab (p. 132)) and the numerical results appear in its text box.



Viewing the Results


The results are arranged in four sections:

• Input Data and Derived Parameters

• Impeller Inlet

• Impeller Exit

• Overall Performance

At the three radial locations of hub, RMS and shroud, Impeller inlet shows the principal geometric andaerodynamic parameters. Impeller exit shows aerodynamic and geometric data at the impeller tip.Overall performance includes specific speed, flow coefficient, stage loading, and other significantperformance quantities.

Appendix 1: Definition of Parameters on Results Tab (p. 140) is a quick-reference for the meaning andunits of all the parameters.

The efficiency correlation selected on the Geometry tab determines which chart is displayed on theEfficiency Chart tab at the bottom-right of the dialog box. One of the three charts shown in the followingfigures will be displayed. The current design is indicated by the circle symbol in each case.


Vista CCD

Figure 6.9: Casey-Robinson stage efficiency correlation

Figure 6.10: Casey-Marty stage efficiency correlation



Viewing the Results

Figure 6.11: Rodgers stage efficiency correlation

In Figure 6.9: Casey-Robinson stage efficiency correlation (p. 133), the legend shows values of tip speed

Mach number, . In Figure 6.11: Rodgers stage efficiency correlation (p. 134) the legend shows values

of stage pressure ratio, R. A sketch of the meridional view of the designed impeller appears in the Im-

peller Sketch tab at the bottom-right of the dialog box.

Figure 6.12: Impeller sketch

6.4. Common Options

The following options and frames are always visible in the Vista CCD dialog box:

Units: These radio buttons control whether the data (both input and results) are presented in either SIor imperial units. The data switches as soon as the units are changed; no recalculation is required afterswitching and any subsequent recalculation will yield identical results.


Vista CCD

Impeller sketch: This displays a simple meridional sketch of the impeller, hub, and shroud curves.

The Calculate button updates the data model with the input data in the user interface and subsequentlyperforms the calculation.

The Close button updates the data model with the input data in the user interface before closing it.

Note

When the user interface loses focus, Workbench is automatically updated with the inputdata. Conversely, on receiving the focus the user interface input data will be updated fromWorkbench.


You can access a context menu for the Blade Design cell in the Vista CCD component system by right-clicking the cell. Most of the commands that are available are standard, and are described in Systemsand Cells. The context menu commands that are specific to the Blade Design cell are described inTable 6.1: Context Menu Commands Specific to the Vista CCD Blade Design Cell (p. 135).

Table 6.1: Context Menu Commands Specific to the Vista CCD Blade Design Cell

DescriptionCommand

This command opens the Vista CCD dialog box.Edit

This command first prompts you to select aBladeGen file, then imports the Vista CCD set-



The BladeGen file is required to have beenwritten by an older version of BladeGen thatincluded Vista CCD; such an older version ofBladeGen wrote Vista CCD settings into theBladeGen file. By contrast, a BladeGen file pro-duced by the Create New > BladeGen com-mand does not contain the Vista CCD data re-quired for the Import BladeGen File com-mand.



in the Vista CCD system. For details, seeLaunching a New BladeGen Model (p. 136).






in the Vista CCD system. For details, seeLaunching a New BladeEditor Model (p. 137).



Context Menu Commands of the Blade Design Cell


A new BladeGen model can be generated from an up-to-date Vista CCD system using the Create New


The 1D impeller design is converted to a 3D impeller geometry model in BladeGen. To achieve the 3Dvane shape, Vista CCD creates initial guesses of vane camber and thickness distributions and these arethen combined with its computed meridional design.

Note

• The BladeGen model is detached from the Vista CCD system (that is, no link is generated).Therefore, any changes made to the Vista CCD system will not be reflected in the BladeGenmodel following an update of the Vista CCD system.

• The BladeGen file that you create in this way cannot be imported into another Vista CCD system,because it does not contain Vista CCD settings. A BladeGen file created by an older version ofBladeGen (that included Vista CCD) contains Vista CCD settings and can be imported into aVista CCD system.


You can create a Throughflow system, which is essentially a Vista TF system with an added Geometrycell, using the Create New > Throughflow command in the right-click context menu of the BladeDesign cell. The Geometry cell will be populated with the BladeEditor geometry based on the current,up-to-date parameters from Vista CCD. The Setup cell will be the same as for a Vista TF system, exceptthat the machine type and number of blade rows cannot be specified in the properties because thatinformation is taken from the Geometry cell. Upon creating a Throughflow system using data from aVista CCD system, each of the cells in the Throughflow system is updated automatically:



Vista CCD


A new BladeEditor model can be generated from an up-to-date Vista CCD system using the Create New


Note

The DesignModeler model is detached from the Vista CCD system (that is, no link is generated).Therefore, any changes made to the Vista CCD system will not be reflected in the Design-Modeler model following an update of the Vista CCD system.

In BladeEditor you can change the key dimensions of the model (that is, the size of the inlet, outlet,and locations of the leading and trailing edge end points) and you can modify the shape of the contoursby modifying the B-spline control points. Both types of changes typically begin in the same way:

1. In the BladeEditor Tree Outline, expand the flow path-defining plane (for example, MerPlane1).

2. Click the Hide Model icon then click the Look At Face icon to center the geometry in the viewer.

3. Expand MerPlane1 so that you can see the sketches that define the flow path.

Tip

You can right-click the sketches under MerPlane1 and hide visible sketches or show hiddensketches.

To change a key dimension:

1. Click the Sketching tab, then click the desired dimension in the Model View. The properties for thatdimension appear in the Details View.

2. Change the dimension:

• To set a particular value, type in the Value field and press Enter.

• To drag a key point:

1. In the Details View set Reference Only to Yes. The dimension name displays in parentheses;that is SY1 becomes (SY1) .

2. In the Sketching tab's Modify toolbox, select Drag and move the key point as required.

Note

After either of these types of changes, the contours automatically update proportionallyto the changes in key dimensions.

To modify the shape of a contour by moving B-spline control points:



Launching a New BladeEditor Model

1. In the Sketching tab, select a dimension that you want to modify and in the Details View set Reference

Only? to Yes. Repeat for any other dimensions you want to change. The dimension names display inparentheses.

2. Select Drag and move the B-spline control points.

Note

If key points are shown in dark blue, they are locked; control points, shown in light blue,can be dragged.

6.9. Using Vista CCD to Model an Existing Compressor

Vista CCD can be of use in the analysis of compressor test data – or operating point data from someother source. A 1D model of a selected operating point may be created by synthesizing the knowngeometry and test data.

In the Geometry section of the data, all items will be known and fixed. From the test data, the massflow, pressure ratio, work factor, rotational speed (corrected for measured inlet temperature and pressure)are all known. By running the program in an iterative mode, the ‘unknowns’ (relative velocity ratio, in-cidence, and possibly lambda and choke margin) may be gradually adjusted until parameters such asimpeller tip diameter, tip width, inlet vane angles, and so on, all attain their actual known values. Effectivebacksweep may also have to be adjusted to obtain the known work factor or pressure ratio.

6.10. Predicting a Performance Map

You can generate performance maps for Vista CCD designs via a performance prediction module. Toaccess this feature, add a Vista CCD (with CCM) system to your project.

Having created a satisfactory design in Vista CCD, you may then predict the overall performance of thedesigned compressor stage using the Performance Map cell.

Note

Performance maps are not available with real gases.


Vista CCD

Figure 6.13: Performance Prediction Module

Figure 6.13: Performance Prediction Module (p. 139) shows the data on which the performance predictionis based and the results in the form of a performance characteristic. The data on which the performanceprediction is based is also accessible from the Properties view for the Performance Map cell.

The data shown is partially enabled and partially disabled. The performance prediction method is basedon the Stage Parameters such as Flow Coefficient, Work Factor, and so on. The parameters for inletstagnation temperature and inlet stagnation pressure, and the rotational speed values, can be modifiedafter clearing the Use Design Point Data check box. Note that the performance then predicted wouldcorrespond to the stage designed in Vista CCD under the modified operating conditions. Clearing theUse Design Point Data check box may be useful at times to see the effect on performance of smallvariations in the enabled parameters.

You cannot change either the geometric or stage parameters. The values shown are based on the VistaCCD geometry carried over so that you may view them; these parameters can be changed only by ex-ecuting another design run. If the Performance Map cell becomes “out-of-date” with respect to theBlade Design cell (and the latter is up-to-date), then you can use the Refresh button to update thegeometric and stage parameters that are displayed. Executing the performance calculation will alsorefresh the geometric and stage parameters; whenever these parameters are refreshed, the optionaloperating conditions (such as inlet stagnation temperature and flow coefficient) will also be refreshedif you select Use Design Point Data.

The default setting for the performance chart is Pressure ratio vs Mass Flow. By right-clicking thechart, you can select additional plots from a context menu (Figure 6.14: Additional PerformanceCharts (p. 140)).



Predicting a Performance Map

Figure 6.14: Additional Performance Charts

The performance chart itself features a menu bar that contains the following icons:

• Save

This icon enables the current chart data to be exported as a comma separated variable (*.csv ) filefor import into 3rd-party tools such as Microsoft Excel.

• Maximize

This icon enables the chart to be viewed in a full-screen view. The full-screen view contains a Restore

down icon to restore the chart to its original size.

6.11. Appendix 1: Definition of Parameters on Results Tab

Imperial

unitsSI unitsParameter

Abbrevi-

ated

name

DATA ECHO AND DERIVED PARAMETERS

Line 1

lb/skg/sMass flow rateMass

rpmrpmMachine rotational

speedN

——Total pressure ratioP0out/P0in

——Polytropic stage effi-

ciencyEta Poly

——Isentropic stage effi-

ciencyEta Isen

——Isentropic impeller effi-

ciencyEta Imp

°FKInlet stagnation temper-

atureT0in


Vista CCD

Imperial


Abbrevi-

ated

name

psikPaInlet stagnation pres-

sureP0in

degdegBlade backsweep angleBeta'5

——Relative velocity ratioW ratio

Line 2

btu/lb-°FJ/kg-KGas constantRgas

——Ratio of specific heatsGamma

ft2/sm2/sKinematic viscosityNu

——Number of main vanesMain

——Number of splitter

vanesInter

inmmImpeller hub thicknessThub

inmmImpeller shroud thick-

nessTshr

degdegInlet flow angleAlpha3

——Power input factorPIF

——Tip clearance to bladeheight ratio at outlet

k/h

RESULTS

Impeller inlet

——Meridional velocity ratio

at impeller inletVm ratio

——Whirl velocity ratio at

impeller inletVw ratio

——RMS Mach number at

impeller inletM_rms

ft/sm/sRMS absolute velocity

at impeller inletV_rms

in2mm2Throat area at impellerinlet

A_throat

btu/lbkJ/kgTotal enthalpy at im-

peller inletH0

btu/lb-°FkJ/kg-KSpecific entropy at im-

peller inlets

Values at the impeller inlet hub, rms, and shroud positions for

inmmDiameterDia

ft/sm/sWhirl velocityVw

ft/sm/sMeridional velocityVm




Appendix 1: Definition of Parameters on Results Tab

Imperial


Abbrevi-

ated

name

degdegRelative flow angleBeta

degdegIncidence angleInc

degdegBlade angleBeta'

Impeller exit

inmmDiameter at impeller

exitDia

inmmDistance between huband shroud at the im-

peller exitTip width

psikPaStagnation pressure at

impeller exitP0

psikPaStatic pressure at im-

peller exitP

°FKStagnation temperature

at impeller exitT0

btu/lbkJ/kgStagnation enthalpy at

impeller exitH0

btu/lb-°FkJ/kg-KSpecific entropy at im-

peller exits

ft/sm/sBlade speed at impeller

exitU

——Blade Mach number at

impeller exitM_U

——Mach number at im-

peller exitM_rms

ft/sm/sRMS relative velocity at

impeller exitW_rms

ft/sm/sRMS absolute velocity

at impeller exitV_rms

degdegRMS absolute flow

angle at impeller exitAl-

pha_rms

degdegRMS relative flow angle

at impeller exitBeta_rms

Overall performance

——Specific speedNs

——Flow coefficientphi

——Loading parameterDelH/U2

——Choke ratiom/mch

——Annulus choke ratiom/mch_a

hpkWPowerpower


Vista CCD

Imperial


Abbrevi-

ated

name

——Reynolds number usingimpeller exit tip widthas the defining length

Re tipwidth

——Reynolds number usingimpeller exit diameteras the defining length

Re tip dia

6.12. Notation

Imperi-

al Units

SI

unitsMeaning

Sym-

bol

ft/sm/sAbsolute velocityC

ftmDiameterD

Btu/lbmJ/kgStagnation enthalpyH0

lbm/skg/sMass flowm

lbm/skg/sMass flow at inducer chokemch

lbm/skg/sMass flow at inlet annulus chokemch

a

——Mach numberM

rev/srev/sRotational speedN

lbf/in^2PaTotal pressureP0

——Qualityqu

Btu/lbm-RJ/kg-K

Entropys

FKTotal temperatureT0

ft/sm/sBlade speedU

ft3/sm3/sVolume flowV

ft/sm/sRelative velocityW

ft/sm/sAbsolute gas angleα

degdegRelative gas angleß

degdegBlade angleß'

Btu/lbmJ/kgIncrease in stagnation enthalpy∆H

s^-1s^-1

Angular velocityΩ

——Work factor∆H/U25

——Specific speed: = ⋅

NsND



Notation

——Flow coefficient: =⋅

——

Tip speed Mach number:

=

MU5

Suf-

fixes

flowfield inlet1

IGV exit2

impeller inlet3

impeller exit5

compressor delivery7

at inducer hubhub

isentropicis

meridional direction or componentm

at inducer shroudshr

tangential direction or componentw

root-mean-square inlet radiusrms

NB — the units listed above are those that are consistent with the dimensionless group definitions(NsND, etc). The definitions used in the Vista CCD GUI are listed in Appendix 1: Definition of Parameterson Results Tab (p. 140).

6.13. References

Bibliography

[1] M V Casey and C J Robinson. A guide to turbocharger compressor characteristics. Dieselmotorentechnik”,

10th Symposium, 30-31 March, 2006, Ostfilder. Ed. M. Bargende, , TAE Esslingen, ISBN 3-924813-65-5.

[2] P Dalbert, B Ribi, and M V Casey. Radial compressor design for industrial compressors. Proc Inst Mech

Engrs Part C, Journal of Mechanical Engineering Sciences, 1999, 213 (C1), 71-83.

[3] C Rodgers. Centrifugal Compressor Design . Cranfield University Short Course on Centrifugal Compressors.1992.

[4] NIST Reference Fluid Thermodynamic and Transport Properties - REFPROP Version 7.0 Users' Guide. Appendix

A.


Vista CCD

Chapter 7: TurboSystem: Vista CPD


The subject of this user guide is Vista CPD. Vista CPD is a program that employs a 1D approach for thepreliminary design of pumps. Vista CPD is able to produce designs for a wide range of pumps from themixed flow (Ns ~ 5500), through Francis type, to high head radial machines (Ns ~ 500). It is integratedinto ANSYS Workbench so that it may be used to generate an optimized initial pump impeller designbefore moving rapidly to a full 3D geometry model, throughflow, and CFD analysis.

The following topics are discussed:7.1.Vista CPD Workflows7.2.Vista CPD Interface Details7.3. Context Menu Commands of the Blade Design Cell7.4. Launching a New BladeGen Model7.5. Creating a New Throughflow System7.6. Launching a New BladeEditor Model7.7. Creating a New Volute

Vista CPD is provided by PCA Engineers Limited, Lincoln, England (www.pcaeng.co.uk).

7.1. Vista CPD Workflows

1. With ANSYS Workbench running, create a new Vista CPD system (available in the Component Systemstoolbox).

(Double-click the Vista CPD system or drag it onto the Project Schematic view.)

2. Either edit the Blade Design cell of the new Vista CPD system or show that cell's properties.

(Right-click the Blade Design cell and select Edit, or click View, and select Properties).

Editing the cell causes the Vista CPD dialog box to appear. The first two tabs of this dialog boxcontain the input data and the last tab displays the results.

The cell properties can be used as an alternative to the aforementioned two tabs that accept input.One advantage of using cell properties is that you can make use of ANSYS Workbench inputparameters to specify the values of input settings.


• Information about the duty of the pump such as rotational speed, volume flow rate, and density

• The pump efficiencies

• Geometric constraints such as the shaft minimum diameter factor



www.pcaeng.co.uk

• Some aerodynamic and geometric settings (for example, the number of vanes) that are open tochange during this 1D design procedure.

See Operating conditions Tab (p. 150) and Geometry Tab (p. 153) for details on the input settings.

4. Compute a pump design by clicking Calculate.


See Results Tab (p. 163) for descriptions of the results.

6. Continue to revise the input data and refresh (regenerate) the results until you obtain a satisfactory 1Ddesign.

7.2. Vista CPD Interface Details

You can enter the input data and view the results in either the Vista CPD interface, or in the Properties

view of the Blade Design cell. The information provided in this section, which discusses the Vista CPDinterface, also applies to the Properties view. For more information on the different ways to enter inputdata, see Vista CPD Workflows (p. 145).

The Vista CPD interface enables the easy 1D design of a pump. The main areas of the interface are de-scribed below and shown in Figure 7.1: Vista CPD Interface (p. 147).

• Global Controls

The controls in this area affect what is displayed in the text and graphics areas. The three globalcontrols are the component selection control, the Calculate button, and the Close button. For moreinformation, see Global Controls (p. 147).

• Graphics Display

Graphical results of the design for each component are presented here. This area contains two tabs:the Sketches tab, and the Efficiency Chart tab. For more information, see Graphics Dis-play (p. 147).

• Component Controls

Input data is specified and numerical results values are presented for each component of the pumphere. This area contains three tabs: the Operating conditions tab, the Geometry tab, and the Resultstab. For more information, see Component Controls (p. 150).

• Error Messages

By default, the Error Messages area is inactive. If an error occurs, the Error Messages area will displayan appropriate error message.


Vista CPD

Figure 7.1: Vista CPD Interface

7.2.1. Global Controls

• Component selection control

This drop-down list enables you to select between the impeller and volute components. You shouldspecify details for both the impeller and volute components before performing a calculation. This listis set to Impeller by default.

• Calculate

Click Calculate to perform the calculation. If you modify the input controls after you click Calculate,the results will be out-of-date. To update the results, click Calculate again.

• Close

Click Close to close Vista CPD.

7.2.2. Graphics Display

The graphics display contains two tabs that display the following after a design has been calculated:7.2.2.1. Sketches



Vista CPD Interface Details

7.2.2.2. Efficiency Chart

7.2.2.1. Sketches

• Impeller

After you click Calculate , you can view the meridional sketch of the impeller in the graphics displayas shown in Figure 7.2: Example Meridional Sketch of the Impeller (p. 148). To view this sketch, ensurethat Impeller is selected in the component selection control, and click the Sketch tab.

Figure 7.2: Example Meridional Sketch of the Impeller

• Volute

After you click Calculate, you can view the central section sketch of the volute in the graphics displayas shown in Figure 7.3: Example Central Section Sketch of the Volute Geometry (p. 149). To view thissketch, ensure that Volute is selected in the component selection control, and click the Sketch tab.


Vista CPD

Figure 7.3: Example Central Section Sketch of the Volute Geometry

7.2.2.2. Efficiency Chart

After you click Calculate, the pump efficiency chart is shown in the Graphics Display under the Effi-ciency Chart tab. The plot shows the overall pump efficiency against the non-dimensional specificspeed for flow rate to speed ratios (Q/N) ranging between 0.0001 and 1.0. The current design is indicatedon the plot by a black square.

Note that if imperial units are used then , the specific speed measure commonly used in the United

States, is used in place of . An example efficiency chart is shown in Figure 7.4: Overall Pump Efficiency

Chart (p. 150).




Figure 7.4: Overall Pump Efficiency Chart

7.2.3. Component Controls

The component controls allow you to specify input data and view the results values for the pump impellerand volute.

7.2.3.1. Operating conditions Tab

7.2.3.1.1. Units

You can use either SI or Imperial units for the input data and results. Note that the standard unit systemsare sometimes modified to reflect commonly used units in pump design. For example, rotational speedis specified in rpm and not in rad/s.

7.2.3.1.2. Duty

The Duty frame contains all the data needed to define the duty for which a pump is to be designed:

• Rotational speed

This setting controls the design point rotational speed of the machine (rpm).

• Volume flow rate

This setting controls the delivery volumetric flow rate of the pump. The volume of flow that passesthrough the impeller is normally higher than this because some of the flow leaks past the impellerback into the inlet eye. The specified volumetric efficiency is used to account for this leakage.

• Density

This setting is used to determine the operating fluid. The default value of 1000 kg/m^3 is for water.

• Head rise


Vista CPD

This setting controls the total dynamic head rise required of the pump at the design point. The headrise is the sum of the static head rise and the velocity head rise. For a fixed volume flow rate, thevelocity head is determined by the area of the aperture. In the case of a centrifugal pump, a smallerinlet gives a larger inlet velocity head and vice versa. If the pump has the same inlet and outlet areas,the total dynamic head rise will be equal to the static head rise.

• Inlet flow angle

This setting controls the angle of the flow, denoted by , at the impeller leading edge, measuredwith respect to the tangential direction. The default value of 90 degrees is for an approach flowwithout pre-rotation. When the flow approaches from a plane pipe, you can generally leave this valueat the default of 90°. However, when an upstream inducer is employed in order to reduce cavitation,you may need to adjust this value.

For an inducer that rotates in the same direction as the impeller, the inlet flow angle will be less than90°. This may happen when the inducer and the impeller are both mounted on the same shaft.Conversely, for a counter-rotating inducer/impeller pair, the inlet flow angle will be greater than 90°.

The inlet swirl angle is treated as a constant value from hub to shroud.

Figure 7.5: Velocity Triangles at the Impeller Leading Edge Indicating the Inlet Flow Angle

• Meridional velocity ratio

This setting is used to describe a linear velocity profile from the hub to the shroud at the leadingedge. It sets the gradient of the profile by specifying the ratio of the meridional velocity at the shroudleading edge radius to the meridional velocity at the average leading edge radius.

The default value of 1.1 indicates a larger meridional velocity at the shroud than at the hub. A valueof less than 1 indicates a larger meridional velocity at the hub than at the shroud. A value of 1 indicatesa uniform meridional velocity distribution.




Figure 7.6: Typical Linear Leading Edge Velocity Profile

7.2.3.1.3. Efficiencies

You may specify the pump efficiencies individually or have Vista CPD calculate them automatically usingcorrelations based on historical data for a range of machine specific speeds.

When you specify individual efficiencies, only three of the four efficiencies may be set, since the efficien-cies are related by the following equation:

(7.1)= × ×

• Hydraulic

The hydraulic efficiency (

) results from the reduction in head due to the pressure loss resulting

from the pump hydrodynamic design (for example, friction losses, turning losses, and so on). This isnormally the most significant of the efficiency components that can be influenced by the designer.Hydraulic efficiency is calculated from the following equation:

= −

where is the ideal head rise and is the head loss due to the hydrodynamic design.

There is often a trade-off between peak hydraulic efficiency and a flatter efficiency profile over awider operating range. Therefore, a pump with a high design point hydraulic efficiency may performmore poorly over the rest of the operating range compared with a pump with a lower design pointhydraulic efficiency.

• Volumetric

The volumetric efficiency ( ) results primarily from the leakage of flow past the impeller back into

the inlet eye. This normally occurs between the shroud ring and the outer casing of the pump.Therefore, in order to deliver the specified volume of flow at the outlet, the volume of flow thatpasses through the impeller must be increased by this leakage volume. Volumetric efficiency is calcu-lated from the following equation:


Vista CPD

=

+

where is the volume of flow delivered at the outlet and

is the leakage flow.

Pumps designed to have tight performance will have less leakage and therefore a higher volumetricefficiency. However, these designs may be more susceptible to wear, especially when the pumpedfluid has a significant suspended solids content. In this case, a “loose” performance pump with alower volumetric efficiency may be favorable because it will provide a more consistent performanceover a longer operating period.

• Mechanical

The mechanical efficiency (

) results from drag on the rotating component of the pump due to

mechanical friction and viscous friction on the outside surface of the impeller shroud (disk friction).Disk friction is the dominant component in the mechanical loss. Mechanical efficiency is calculatedfrom the following equation:

=−

where is the shaft input power of the pump and !"#$ is the power lost due to disk friction.

Pumps designed to have tight performance with smaller clearances will generally suffer more fromdisk friction effects compared to “loose” performance pumps.

• Pump

The overall pump efficiency. As indicated in Equation 7.1 (p. 152), this is the product of the hydraulic,volumetric, and mechanical efficiencies.

7.2.3.2. Geometry Tab

7.2.3.2.1. Impeller Geometry

To enter the impeller geometry data, select Impeller in the component selection control, and clickthe Geometry tab.

7.2.3.2.1.1. Hub Diameter

• Shaft minimum diameter factor

The shaft minimum diameter is calculated based on the maximum allowable shear stress of the shaft.The shaft minimum diameter factor is then applied to the resulting value as a factor of safety. Thedefault value of 1.1 represents a 10% increase in the shaft diameter.

• Dhub/Dshaft

This is the ratio of the impeller hub diameter to the shaft diameter. The hub and shaft diameters areshown in Figure 7.7: Hub and Shaft Diameter Locations (p. 154).




Figure 7.7: Hub and Shaft Diameter Locations

7.2.3.2.1.2. Leading Edge Blade Angles

• Hub and Meanline

This drop-down menu controls how the leading edge blade angles are calculated at the hub andmeanline locations. You can select from three methods using the Hub and Meanline drop-downmenu:

– Cotangent (default)

In this method, the angles are calculated relative to the shroud leading edge blade angle. Theangle is calculated as follows:

′ = ′−

for the hub, and similarly

′ = ′−

for the meanline.

– Cosine

This method uses a similar approach calculating the angles as follows:

′ = ′−

for the hub, and similarly


Vista CPD

′ = ′−

for the meanline.

– User defined

This method allows you to specify the angles directly.

• Shroud

The leading edge blade angle at the shroud is defined either indirectly by specifying the incidenceangle at the shroud, (default, 0 degrees incidence), or directly by specifying the value of the angle.

7.2.3.2.1.3. Tip Diameter

This sets the diameter of the impeller at the meanline trailing edge location. The tip diameter is probablythe most important early decision in the hydraulic design of a centrifugal pump, since the impeller

diameter and the tip speed,, influence all other dimensions of the pump hydraulic design and per-

formance characteristics.

There are three methods for specifying the tip diameter. You can select from the three methods byusing the Tip diameter drop-down menu:

• Automatic (using stability factor) (default)

A prerequisite for a new pump design is that the head-flow characteristic is stable, that is, con-tinuously rising to zero flow. Unstable head curves may be due to excessive diffusion of the im-peller relative velocity at low flows and may also be due to excessive blade shape effect comparedwith centrifugal effect in head generation.

Impeller diffusion can be defined as:

−

where and are the meanline relative velocities at the inlet and outlet respectively. Analysis

of pump tests show that when = at pump best efficiency flow, instability in a head-flow

curve at lower flows is very unlikely. For this condition it can be shown that the tip diametershould be such that:

− =

where and are the meanline blade speeds at the leading and trailing edges, respectively,

and ! "# is the meanline tangential flow velocity at the trailing edge.

This leads to the definition of the stability factor,$%:

= −&

' '

()*

+ +

+

, -

,




For a stable head-flow characteristic, should be > 0.9. Vista CPD calculates a value for based on the speed of the machine. With already established, the tip diameter is specified.

• Specify head coefficient

In this method, Vista CPD calculates the tip diameter based on a given head coefficient,. This

is a non-dimensional parameter that is useful when the new pump is based on an existing designof known head coefficient.

The head coefficient is defined by:

=

• User defined

This method allows you to directly specify the tip diameter. This method is useful when the newdesign is a replacement for an existing machine and the tip diameter is already a constraint. Inthis case, you should first use the Automatic method to obtain a design close to what isneeded, and then you should switch to the User defined method to establish the exact tipdiameter needed.

7.2.3.2.1.4. Trailing Edge Blade Angles

• Blade angle

The trailing edge blade angle, ′

, is the angle the blade makes with respect to the tangential direction

at the trailing edge as shown in Figure 7.8: Trailing Edge Blade Angle (p. 157).


Vista CPD

Figure 7.8: Trailing Edge Blade Angle

The trailing edge blade angle is a key factor in determining the impeller width at the trailing edge,

also called the tip width, shown in Figure 7.9: Tip Width (p. 158). This is a logical relationship since

the exit flow rate is determined by the meridional velocity and the cross sectional area at that point,as defined by the usual continuity equation:

=

At a given rotational speed, a reduction in the blade angle results in a smaller meridional velocity,

. In order to maintain the flow rate,

, the cross sectional area, , must be increased. For

a fixed impeller diameter, can only be raised by increasing the tip width, . Conversely, increas-

ing the blade angle at a fixed rotational speed and impeller diameter reduces the tip width.




Figure 7.9: Tip Width

The default value of 22.5 degrees is considered to be a standard design. Studies have shown thatincreasing the blade angle can lead to an enhanced head rise, but with an associated reduction inhydraulic efficiency.

• Rake angle

This is the angle the trailing edge makes with a line perpendicular to the hub surface, also referredto as the blade lean at the trailing edge. Since only the hub and shroud sections are considered inVista CPD, only straight lean, as opposed to compound lean, is possible.

The default value of 0 degrees is very common in pumps, likely for manufacturing reasons, althougha positive rake angle can be used to reduce secondary flows by influencing the distribution of flowin the spanwise (hub to shroud) direction. This approach is more common in centrifugal compressors,but the same mechanism applies to centrifugal pumps.

7.2.3.2.1.5. Miscellaneous

• Number of vanes

The number of blades used in the impeller. A larger number of blades gives greater control over theflow direction in the impeller, but with an increased blockage to flow due to the larger solid to fluidratio. This will also impact the blade angle and tip width described in Trailing Edge Blade Angles (p. 156).

As the number of vanes is increased both the tip width, , and the relative flow angle at the trailing

edge,

, are influenced. The plot shown below in Figure 7.10: Influence of the Number of Vanes on

Impeller Tip Width and Relative Flow Angle at the Trailing Edge (p. 159) shows a typical example of

the behavior of and

with the variation of the number of vanes.


Vista CPD

Figure 7.10: Influence of the Number of Vanes on Impeller Tip Width and Relative Flow Angle

at the Trailing Edge

The plot shows how as the number of vanes is increased

also increases, gradually becoming closer

to the trailing edge blade angle, ′

, because the impeller imposes greater control on the flow direc-

tion. As a consequence of the increase in

the meridional velocity at the trailing edge,, is also

increased. This in turn acts to reduce the flow area at the trailing edge and, as a result, the tip width,

also decreases for a fixed tip diameter. However, another effect of increasing the number of vanes

is increasing the blockage to the flow. This increased blockage acts to increase in order to maintain

the flow area at the trailing edge. Consequently, an increase in the number of vanes has two competinginfluences on the tip width of the impeller. Figure 7.10: Influence of the Number of Vanes on ImpellerTip Width and Relative Flow Angle at the Trailing Edge (p. 159) shows that the influence of the variation

in

dominates for a low number of vanes, with the tip width decreasing as the number of vanes

is increased. As the number of vanes increases further, the effects of the blockage to the flow dom-inate and the tip width increases.

• Thickness/tip diameter

The thickness to tip diameter ratio is a non-dimensional parameter used to define the impeller vanethickness. Increasing the thickness to tip diameter ratio also increases the blockage to flow. This resultsin a larger tip width although, unlike when increasing the number of vanes, this has no impact ofthe trailing edge flow angle.

• Hub inlet draft angle

The hub inlet draft angle,, is the angle between the hub and the horizontal line at the hub inlet,

as shown in




Figure 7.11: Hub Inlet Draft Angle

Reducing the hub inlet draft angle moves the hub inlet point forward, which results in a smaller hubradius. Conversely, a larger value moves the hub inlet point backward, which results in a bigger hub

radius. For low specific speed machines, it is common to use a larger , whereas higher specific

speed machines may benefit from a lower .

7.2.3.2.2. Volute Geometry

The volute design in Vista CPD is calculated to maintain a constant angular momentum in the scroll,with a small adjustment to account for friction losses at the walls. Starting at the tongue and endingat the throat, the required cross sectional areas are established at 8 equally spaced locations aroundthe scroll and are used to determine the appropriate dimensions of the selected cross section shape.The throat is co-located at the diffuser inlet. The length and exit area of the diffuser section may optionallybe specified by the user. The central section of the volute is shown in Figure 7.12: Central Section ofthe Volute (p. 161).


Vista CPD

Figure 7.12: Central Section of the Volute

The inlet width of the volute is calculated from the sum of the impeller tip width, the hub and shroudsolid thicknesses, and the clearances between the rotating impeller and the stationary casing at thehub and at the shroud. This arrangement is show in Figure 7.13: Impeller and Volute Interface Arrange-ment (p. 161).

Figure 7.13: Impeller and Volute Interface Arrangement

Note that the volute inlet width can often measure twice as much as the impeller tip width.




7.2.3.2.2.1. Casing rotation angle

The casing rotation,, is the angle between the vertical line and the tongue location when viewing

the central section through the volute, as shown in Figure 7.12: Central Section of the Volute (p. 161).The default value of 14 degrees is suitable in most cases, but small adjustments may be made to ensurea smooth transition to the diffuser section.

7.2.3.2.2.2. Section Type

There are two options for the volute section shape:

• Elliptical/circular

This section type begins as a straight line at the tongue and transitions to an elliptical sectionthereafter. In the smaller sections, the major axis is fixed by the volute width and the minor axis isadjusted to achieve the desired area. Where the required value of the minor axis would be greaterthan the value of the major axis, both axes are made equal and the section is then circular.

• Rectangular

This section type begins as a straight line at the tongue and transitions to a rectangular section. Anadditional parameter to specify for rectangular sections is the aspect ratio (width/height) at the throat.

Given the volute width, a critical height can be calculated:

=

where is the specified aspect ratio. This translates to a critical area:

= !"#$ %&'$( !"#$

or


Vista CPD

=

Below this critical area, the width of the rectangle is fixed by the volute width and the height is ad-justed to achieve the desired cross sectional area with an aspect ratio less than that specified. Abovethis critical area the aspect ratio of the rectangle is fixed by the value specified and the height andwidth are adjusted to match the required cross sectional area accordingly.

7.2.3.2.2.3. Diffuser

The exit diameter and length of the diffuser section are calculated to achieve a reasonable rate of diffusionwhile avoiding stall. To define these settings manually, select the check box next to the setting youwould like to define, and enter the desired value in the appropriate box. If the check boxes are cleared,these values will be calculated automatically by Vista CPD.

7.2.3.3. Results Tab

7.2.3.3.1. Impeller Results

To view these results, click the Results tab and set the component selection control to impeller .

7.2.3.3.1.1. Overall Performance

This section presents the overall performance parameters that characterize the pump impeller.

• Specific speeds: (non-dimensional), (US units), (European units)




Specific speed is a number that loosely defines the geometric shape of a pump. For example, a low

, high head pump would have a narrow radial impeller and a small section volute throat. A higher

, lower head, mixed flow design would involve a wider, more conical impeller and the volute throat

area would be comparatively larger.

Pump efficiency reaches a peak value when is close to 1.0. It may be possible to arrange the pump

design duty or pump speed to take advantage of this fact.

=

Equivalent forms of which are commonly used in the US and in Europe are also included here for

convenience. They can be related to simply as:

=

and

=

A value of !" between 0.3 and 0.8 indicates a radial flow impeller, a value of #$ between 1.0 and 2.5

indicates a mixed flow impeller, and a value of %& above 3.5 indicates an axial flow impeller.

The specific speed has a significant influence on pump shape. A low '( pump has a narrow radial

flow impeller and the outlet diameter is significantly larger than the inlet diameter. A high )* pump

has a mixed flow impeller and the outlet diameter is only slightly larger than the inlet diameter.

Figure 7.14: Typical Pump Shapes at Low and High Specific Speeds.

The specific speed also has a significant influence on the shape of the pump performance curve. A

low +,, radial flow impeller pump has a head/flow curve with a low head rise to zero flow. The pump


Vista CPD

power/flow curve usually rises continuously from about 50% at zero flow. A medium , mixed flow

impeller pump has a more steeply falling head/flow curve and the power may be at its maximum atthe pump design flow.

• Suction specific speed,

Suction specific speed is a non-dimensional parameter that can be useful in the evaluation of pumpcavitation performance.

=

where and are taken at the highest efficiency, or design point, of the impeller.

Note that Vista CPD uses the non-dimensional form of . Similar to the specific speed, alternative

forms are also in use for the US and European unit systems. Since the units for ! and head arethe same, the same conversion factors shown above also apply.

For overhung impeller volute pumps, with "#$%& for 3% head loss, '(( indicates the following

performances:

– 1.5 indicates generally poor cavitation performance

– 2.5 indicates reasonable cavitation performance

– 4.0 indicates good cavitation performance

– Above 4.0 is exceptional (possible enlarged impeller inlet area)

• Power

The shaft power of the impeller. This is defined as a combination of the hydraulic power and theoverall pump efficiency:

= =⋅

)*+,-)*+,-

.

/01

.

234

5 5

where ⋅6 is the impeller mass flow rate, 7 is the head rise and 8 9 is the overall pump efficiency.

• Head Coefficient,:(head coeff)

Where the characteristics of an established pump are known, it is common to scale this design toproduce a family of geometrically similar pumps which operate at different speeds. A key parameterthat remains constant through such a scaling is the head coefficient. This is a measure of the energytransfer to the fluid (sometimes called the energy transfer coefficient) and is defined as:

=;<=

>??




where is the head rise, and is the blade speed at the meanline trailing edge location.

• Flow coefficient, (flow coeff)

Similar to the head coefficient, the flow coefficient remains the same for geometrically similar pumps.As the name suggests, this is a measure of the flow rate through the pump and is defined as

=

where is the impeller volume flow rate, is the rotational speed, and is the meanline tip dia-

meter.

• Stability factor,

The stability factor is a measure of how stable the pump's performance characteristic curve is likelyto be. A value of less than 0.9 at the design point indicates that the head curve may fall as the flowrate approaches zero, a so-called unstable characteristic. The stability factor is defined as

= −

where and are the meanline blade speeds at the leading and trailing edges respectively,

and ! is the tangential velocity at the meanline trailing edge location.

• Net positive suction head required, "#$%&

If the pressure at a point in the flow field drops below the vapor pressure of the liquid, the liquidwill vaporize, a process known as cavitation. As the vapor bubble moves back into a region of pressurehigher than the vapor pressure, the bubble will collapse as it reverts back to liquid form. This is a vi-olent process due to the large density change involved which causes noise and, above all, damageto pumps.

In centrifugal pumps, the liquid accelerates into the eye of the pump causing the pressure to drop.If there is insufficient head at the eye to accommodate this local pressure drop then the pump will

cavitate. The '()*+ is the level of head required at the impeller eye in order to avoid significant

noise and damage due to cavitation.

It is possible to reduce the ,-./0 by increasing the eye diameter to reduce the acceleration effect.

However, this increases the risk of recirculation at the shroud inlet, which itself can result in severeflow oscillations and cavitation. A common approach to mitigate against cavitation, where insufficient

1234 is available, is the addition of an inducer ahead of the impeller inlet to provide the extra headrequired.

• Diffusion Ratio

The diffusion ratio is defined as

−5 5

5

6 7

6


Vista CPD

where and are the meanline relative velocities at the inlet and outlet respectively. Analysis of

pump tests show that when = at highest efficiency flow, instability in a head-flow curve at

lower flows is very unlikely. Consequently, a value of diffusion ratio close to zero is desirable, a valuegreater than 0.25 is considered high.

7.2.3.3.1.2. Impeller Inlet

This section describes the calculated dimensions, angles and velocities at the impeller leading edge.

• Basic Dimensions

The hub diameter, Dh, and the eye diameter, De, are the inlet diameters at hub and shroud as indicatedin

Figure 7.15: Hub Diameter and Eye Diameter

The vane thickness, Thk, is the normal thickness of the vane calculated from the specified thicknessto diameter ratio.

• Detailed Parameters

The following parameters are calculated at the hub, meanline, and shroud sections, and are listed intable format:

– Diameter,

– Tangential velocity,




– Meridional velocity,

– Blade speed,

– Flow relative velocity,

– Blade angle, ′

– Relative flow angle,

– Incidence,

Here both the blade angles, ′ and the relative flow angles, are measured relative to the tangential

direction, similar to the specification of the inlet flow angle (see Figure 7.8: Trailing Edge Blade

Angle (p. 157)). The incidence is simply calculated as ′ − and is presented for convenience.

7.2.3.3.1.3. Impeller Exit

This section describes the calculated parameters at the impeller trailing edge.

• Tip diameter,

• Tip width, (see Figure 7.9: Tip Width (p. 158))

• Lean angle (rake),

• Relative flow angle,

• Flow relative velocity,

• Absolute flow angle,

• Flow absolute velocity,

• Slip factor,

• Blade speed (tip speed),

• Flow tangential velocity,!"#

The slip factor is a non-dimensional parameter which indicates the degree to which the flow is expectedto deviate from the blade. Figure 7.16: Exit velocity triangles with slip (red) and without slip (black) (p. 169)


Vista CPD

shows the exit velocity triangles for both the hypothetical case where the flow angle is the same as the

blade exit angle, ′

, and the true case with the flow angle of

.

Figure 7.16: Exit velocity triangles with slip (red) and without slip (black)

The slip velocity,, is defined as the difference between the no-slip tangential velocity, ′ , and

the true tangential velocity,. The slip factor, , is defined as the ratio of the slip velocity to the

trailing edge tip speed,:

=

7.2.3.3.2. Volute Results

To view these results, click the Results tab and set the component selection control to Volute .

7.2.3.3.2.1. Key Dimensions

This section describes some key dimensions defining the volute geometry.

• Inlet width

For details on how the volute inlet width relates to the impeller tip width, see Figure 7.13: Impellerand Volute Interface Arrangement (p. 161).

• Base circle radius

The base circle radius is the radius of the circle that touches the tongue (or cutwater). It is shownin Figure 7.12: Central Section of the Volute (p. 161).

• Cutwater clearance

The cutwater clearance is the distance between the impeller tip and the volute tongue, calculatedas the difference between the base circle radius and the meanline impeller tip radius.




• Cutwater thickness

The cutwater thickness is the thickness of the tongue at the point where it meets the base circle.

7.2.3.3.2.2. Sections, cutwater to throat

This section presents the geometrical properties of the calculated volute cross sections in tabular form.The data displayed varies slightly depending on whether the volute sections are elliptical/circular orrectangular. Figure 7.17: Sample Elliptical/circular Cross-section Data (p. 170) shows an example of ellipt-ical/circular cross section data, and Figure 7.18: Sample Rectangular Cross-section Data (p. 170) showsan example of rectangular cross section data.

Figure 7.17: Sample Elliptical/circular Cross-section Data

Figure 7.18: Sample Rectangular Cross-section Data

The areas, centroid radii, and outer radii of the cross sections are shown for both elliptical/circular andrectangular section types.

For the elliptical/circular section type the major and minor axes of the ellipses are shown. When themajor and minor axes are equal the section is circular.

For the rectangular section type the height and width is listed. In the sections where the width is equalto the volute inlet width, the aspect ratio (height/width) of the section is usually less than the valuespecified in the geometry panel, (except where the height equals the critical height). When the widthexceeds the volute inlet width, the required aspect ratio is met.


Vista CPD

7.2.3.3.2.3. Diffuser

This section describes the basic dimensions of the diffuser section of the volute.

The exit hydraulic diameter is calculated as the diameter of the equivalent circular section with an areaequal to the calculated exit area. The cone angle is calculated as the angle between the sloping sidesof the equivalent circular based conic frustum, as shown in Figure 7.19: Equivalent Conic Frustum Dif-fuser (p. 171).

Figure 7.19: Equivalent Conic Frustum Diffuser


You can access a context menu for the Blade Design cell in the Vista CPD component system by right-clicking the cell. Most of the commands that are available are standard, and are described in Systemsand Cells. The context menu commands that are specific to the Blade Design cell are described inTable 7.1: Context Menu Commands Specific to the Vista CPD Blade Design Cell (p. 171).

Table 7.1: Context Menu Commands Specific to the Vista CPD Blade Design Cell

DescriptionCommand

This command opens the Vista CPD dialog box.Edit

This command first prompts you to select aBladeGen file, then imports the Vista CPD set-



The BladeGen file is required to have beenwritten by an older version of BladeGen thatincluded Vista CPD. Older versions of BladeGenwrote Vista CPD settings into the BladeGen file.By contrast, a BladeGen file produced by the



Context Menu Commands of the Blade Design Cell

DescriptionCommand

Create New > BladeGen command does notcontain the Vista CPD data required for theImport BladeGen File command.



in the Vista CPD system. For details, seeLaunching a New BladeGen Model (p. 114).

This command creates a system with a geo-metry cell and a mesh cell. The volute mesh

Create New > Volute

from this system can be combined with thematching impeller mesh for CFD analysis. Fordetails, see Creating a New Volute (p. 174).






in the Vista CPD system. For details, seeLaunching a New BladeEditor Model (p. 115).


A new BladeGen model can be generated from an up-to-date Vista CPD system using the Create New


The 1D impeller design may be converted to a 3D impeller geometry model in ANSYS BladeGen. Toachieve the 3D vane shape, Vista CPD invokes initial guesses of vane camber and thickness distributionsand these are then combined with its computed meridional design. You can modify the camber andthe thickness in the subsequent BladeGen model if desired.

Figure 7.20: BladeGen Model (p. 173) shows the BladeGen model corresponding to the default data usedin the earlier examples.


Vista CPD

Figure 7.20: BladeGen Model

Note

• The BladeGen model is detached from the Vista CPD system (that is, no link is generated).Therefore, any changes made to the Vista CPD system will not be reflected in the BladeGenmodel following an update of the Vista CPD system.

• The BladeGen file that you create in this way cannot be imported into another Vista CPD system,because it does not contain Vista CPD settings. A BladeGen file created by an older version ofBladeGen (that included Vista CPD) contains Vista CPD settings and can be imported into aVista CPD system.


You can create a Throughflow system, which is essentially a Vista TF system with an added Geometrycell, using the Create New > Throughflow command in the right-click context menu of the BladeDesign cell. The Geometry cell will be populated with the BladeEditor geometry based on the current,up-to-date parameters from Vista CPD. The Setup cell will be the same as for a Vista TF system, exceptthat the machine type and number of blade rows cannot be specified in the properties because thatinformation is taken from the Geometry cell. Upon creating a Throughflow system using data from aVista CPD system, each of the cells in the Throughflow system is updated automatically:



Creating a New Throughflow System



A new BladeEditor model can be generated from an up-to-date Vista CPD system using the Create New


You can either generate only the impeller in BladeEditor, or you can generate the impeller and the volute.To generate only the impeller, set the Impeller export type property in the Properties view of theBlade Design cell to Isolated impeller . To generate the impeller and the volute, set the Impeller

export type property to Coupled to Volute . This property is set to Isolated impeller bydefault.

Note

The DesignModeler model is detached from the Vista CPD system (that is, no link is generated).Therefore, any changes made to the Vista CPD system will not be reflected in the Design-Modeler model following an update of the Vista CPD system.

7.7. Creating a New Volute

You can create a system that contains the volute geometry and volute mesh from the pump you havedesigned in Vista CPD using the Create New > Volute command in the right-click context menu of anup-to-date Blade Design cell. The geometry cell contains a parametric volute geometry model, theparameters of which may be modified after creation, similar to a regular DesignModeler parametricmodel. The mesh cell contains a tetrahedral mesh which includes inflation layers on the volute walls.This mesh is ready for combining with the matching impeller mesh for CFD analysis.

The volute parameters are as follows:

• r3, volute inlet radius

• b3, volute inlet width

• b2, impeller tip width

• clear, the clearance between the impeller and the volute tongue

• thk, the volute tongue thickness

• theta2, the angle of inclination of volute inlet to the horizontal.


Vista CPD

An angle of zero gives a regular cylindrical inlet, whereas a non zero angle results in a conicalinlet section. This is the same as the impeller trailing edge angle of inclination.

• thetaCR, the volute casing rotation angle

• Elliptical/circular volute only: minor1 to minor8, the ellipse cross section minor axis

• Rectangular volute only: height1 to height8, the rectangular cross sectional height

• Rectangular volute only: aspectRatio, the aspect ratio of the throat cross section

• diffLength, the axial length of the volute diffuser section

• diffDiam, the exit diameter of the volute diffuser section



Creating a New Volute


Chapter 8: Vista AFD


The subject of this user guide is Vista AFD. Vista AFD is a program for the preliminary design of axialfans. It creates axial fan geometry data for use in BladeGen or BladeEditor. It also provides estimates ofthe performance of the axial fan.

Vista AFD is integrated into ANSYS Workbench so that it may be used to generate a preliminary fandesign before moving rapidly to a full 3D geometry model and CFD analysis.

Vista AFD is provided by PCA Engineers Limited, Lincoln, England.

The following topics are discussed:8.1.Vista AFD Workflow8.2. Data Review and Edit8.3. Results Tab8.4. Context Menu Commands of the Cells in the Vista AFD System8.5. Creating a Blade Design8.6.Troubleshooting and Error Messages8.7. Notation

8.1. Vista AFD Workflow

The fan design calculation involves two mandatory steps (a meanline calculation and a design calculation)and one optional step (an analysis calculation).

The workflow in Vista AFD is illustrated by Figure 8.1: Principle of Operation Flowchart (p. 178).



Figure 8.1: Principle of Operation Flowchart

Begin by creating a Vista AFD system.

Figure 8.2: Vista AFD System

Details of the calculations are provided in the following sections:8.1.1. Meanline Calculation8.1.2. Design (Throughflow) Calculation8.1.3. Analysis (Throughflow) Calculation


Vista AFD

8.1.1. Meanline Calculation

To set up and run a meanline case in Vista AFD:

1. With ANSYS Workbench running, create a new Vista AFD system (available in the Component Systemstoolbox).

(Double-click the Vista AFD system or drag it onto the Project Schematic view.)

2. Either edit the Meanline cell (of the new Vista AFD system) or show that cell's properties.

(Right-click the Meanline cell and select either Edit or Properties from the shortcut menu.)

Editing the cell causes the Vista AFD dialog box to appear; the first two tabs of this dialog boxcontain the input data and the last tab displays the results.


3. Specify the basic aerodynamic and geometric requirements, namely:

• the fan duty (e.g. head rise, mass flow, rotational speed)

• the machine configuration (with/without IGV or OGV)

• fixed geometric constraints (e.g. annulus diameters, aspect ratio)

Refer to Data Review and Edit (p. 183) for details about each input setting.

With the appropriate inputs made, the meanline calculation can be run.

4. Run a meanline calculation by one of these methods:

• Updating the Meanline cell after configuring its properties.

• Editing the Meanline cell, configuring the settings in the Aerodynamics and Geometry tabs, thenclicking the Calculate button, which is shown in Figure 8.3: Meanline Results Summary (p. 180).



Vista AFD Workflow

Figure 8.3: Meanline Results Summary

The Results tab is then automatically made visible. The 1D aerodynamic dependent parametersare computed and displayed numerically; values that are outside of the recommended rangesare highlighted.

A sketch of the meridional view of the designed fan is displayed. This outlines the annulus inblack, the IGV in red (if enabled), the rotor in blue and the OGV in green (if enabled).

Refer to Results Tab (p. 194) for details.

8.1.2. Design (Throughflow) Calculation

If the results of the meanline calculation are satisfactory the design (throughflow) calculation may thenbe performed, if not the input data must be revised accordingly and the meanline calculation rerun.

To set up and run a design (throughflow) case in Vista AFD:

1. Set up and run a meanline case.

2. Either edit the Design cell (of the new Vista AFD system) or show that cell's properties.

(Right-click the Design cell and select either Edit or Properties from the shortcut menu.)

Editing the cell causes the Vista AFD dialog box to appear; the first tab of this dialog box containsthe input data and the last tab displays the results.


3. Specify the required input settings. Refer to Data Review and Edit (p. 183) for details about each inputsetting.


Vista AFD

4. Run a design (throughflow) calculation (which is more detailed than the meanline calculation) by oneof these methods:

• Updating the Design cell after configuring its properties.

• Editing the Design cell, configuring the settings in the Controls tab, then clicking the Calculate

button, which is shown in Figure 8.4: Design Results Summary (p. 181).

Figure 8.4: Design Results Summary

The Results tab is then automatically made visible. Compared to the meanline calculation, moredetailed aerodynamic dependent parameters are computed; these are displayed numericallyusing the design calculation; again values that are outside of the recommended ranges arehighlighted.

A sketch of the meridional view of the designed fan appears in the sketch window. This outlinesthe annulus in black, the IGV in red (if enabled), the rotor in blue and the OGV in green (if en-abled).


If the results of the design calculation are satisfactory then a BladeGen or BladeEditor model may becreated. If they are not satisfactory then the input data must be revised accordingly and the processbegins again with the meanline calculation (as shown in Figure 8.1: Principle of Operation Flow-chart (p. 178)).

At this point, the geometry data is available for a Blade model to be created (see Creating a BladeDesign (p. 196)). However, it may be useful to verify the performance of the design by running thethroughflow calculation in analysis mode. See Analysis (Throughflow) Calculation (p. 181) for details.

8.1.3. Analysis (Throughflow) Calculation

Optionally, before creating a BladeGen or BladeEditor model, an analysis calculation may be performed.This uses a similar throughflow method to the design calculation but simply analyses the design created



Vista AFD Workflow

in the previous step, rather than adjusting the geometry. A significant difference between the designand analysis results indicates a potentially flawed design.

To set up an analysis (throughflow) case in Vista AFD:

1. Set up and run a design (throughflow) case.

2. Either edit the Analysis cell (of the new Vista AFD system) or show that cell's properties.

(Right-click the Analysis cell and select either Edit or Properties from the shortcut menu.)

Editing the cell causes the Vista AFD dialog box to appear; the first tab of this dialog box containsthe input data and the last tab displays the results.


3. Specify the required input settings. Refer to Data Review and Edit (p. 183) for details about each inputsetting.

4. Run an analysis (throughflow) calculation (which is more detailed than the design (throughflow) calcu-lation) by one of these methods:

• Updating the Analysis cell after configuring its properties.

• Editing the Analysis cell, configuring the settings in the Controls tab, then clicking the Calculate

button, which is shown in Figure 8.5: Analysis Results Summary (p. 182).

Figure 8.5: Analysis Results Summary

The Results tab is then automatically made visible. The parameters shown on the Results tabfor an analysis calculation should be very similar to those for a design calculation, with the ex-ception of the deviation, which is limited to 20 degrees in the design calculation only. Shouldthe results be significantly different between the design and analysis calculations, this indicatesa potentially flawed design and the input data should be reviewed carefully.


Vista AFD

A sketch of the meridional view of the designed fan appears in the sketch window. This outlinesthe annulus in black, the IGV in red (if enabled), the rotor in blue and the OGV in green (if en-abled).



The input data can be specified on the Aerodynamics and Geometry tabs of the Vista AFD dialog boxthat is invoked from the Meanline cell. The input data can also (or alternatively) be specified in theProperties view of the Meanline cell in the Vista AFD system.

The Properties view for the Meanline cell is shown in the following figures:

• Figure 8.6: Properties view for a Meanline cell of a Vista AFD system (Part 1 of 3) (p. 183)



Figure 8.6: Properties view for a Meanline cell of a Vista AFD system (Part 1 of 3)






Vista AFD


The control data can be specified on the Controls tab of the Vista AFD dialog box that is invoked fromthe Design cell. The control data can also (or alternatively) be specified in the Properties view of theDesign cell.

The Properties view for the Design cell is shown in the following figures:

• Figure 8.9: Properties view for a Design cell of a Vista AFD system (Part 1 of 2) (p. 186)

• Figure 8.10: Properties view for a Design cell of a Vista AFD system (Part 2 of 2) (p. 187)




Figure 8.9: Properties view for a Design cell of a Vista AFD system (Part 1 of 2)


Vista AFD

Figure 8.10: Properties view for a Design cell of a Vista AFD system (Part 2 of 2)

The control data can be specified on the Controls tab of the Vista AFD dialog box that is invoked fromthe Analysis cell. The control data can also (or alternatively) be specified in the Properties view of theAnalysis cell.

The Properties view for the Analysis cell is shown in the following figures:

• Figure 8.11: Properties view for an Analysis cell of a Vista AFD system (Part 1 of 2) (p. 188)

• Figure 8.12: Properties view for an Analysis cell of a Vista AFD system (Part 2 of 2) (p. 189)




Figure 8.11: Properties view for an Analysis cell of a Vista AFD system (Part 1 of 2)


Vista AFD

Figure 8.12: Properties view for an Analysis cell of a Vista AFD system (Part 2 of 2)

Input data may be declared as parameters via the Properties views. Once an input is designated as aparameter, its value can be modified only via the Parameter Manager.

Descriptions of the aforementioned tabs are given in the following sections:8.2.1. Aerodynamics Tab8.2.2. Geometry Tab8.2.3. Controls Tab




8.2.1. Aerodynamics Tab

Figure 8.13: Aerodynamics Input Tab

Operating Conditions:

Rotational speed, Mass flow rate, Inlet total pressure, Inlet total temperature, Total head rise andEfficiency estimate.

The total head rise used here is the net head rise after any downstream pressure loss (see Downstream

mixing losses parameter below). This also neglects any downstream swirl component of dynamicpressure.

While the efficiency estimate will be used as the specified value in the meanline calculation, this willdiffer from the efficiency reported by the design calculation as the throughflow method uses empiricalcorrelations to derive the efficiency.

Stator exit angles:

IGV exit angle and OGV exit angle.

Should the IGV and/or OGV be omitted in the machine configuration, then these options will be disabledaccordingly.

Additional parameters:

Downstream mixing losses, Hub velocity deficit factor and Hub loading parameter.

The downstream mixing losses parameter specifies the proportion of axial dynamic pressure at rotorexit that is assumed to be lost due to the aerodynamic mixing process. The default value of 0.25 isreasonable for most industrial fan designs and should not require significant adjustment.

The hub velocity deficit factor is used to adjust the meanline calculation to take into account the influenceof the boundary layer at the hub. Reducing this value can be useful to make the meanline calculationfor the hub gas exit angle correlate more closely with that of the design calculation.


Vista AFD

Typically an initial fan design will be more highly loaded at the hub of the machine and more lightlyloaded at the tip. Reducing the hub loading parameter redistributes the workload to reduce the loadingat the hub of the machine whilst maintaining the overall pressure rise.

Machine configuration:

Check boxes indicating whether an IGV and/or OGV will be included in the calculation.

Units:

Radio buttons to select either SI or Imperial units. This impacts both input data and results and may beswitched at any time with immediate effect.

8.2.2. Geometry Tab

Figure 8.14: Geometry Input Tab

Annulus dimensions:

Outer diameter, Hub/tip rotor in and Hub/tip rotor out.

These values specify overall annulus geometry. An outer diameter that is too small for the duty willgive loadings that are too high and DeHaller numbers that are too low. Equally an outer diameter thatis too large for the duty will produce too much swirl and associated pressure loss.

The hub/tip ratios at inlet and outlet are often the same, although it can be an advantage to have ahigher outlet hub/tip ratio than that at the inlet in order to accelerate the axial flow and prevent stall.Machines with a low hub/tip ratio commonly have a high hub loading and may require adjustmentsusing the hub loading parameter.

IGV:

Aspect ratio, Number of vanes and Profile trim.




The aspect ratio is defined as the blade height/hub chord. Acceptable values for aspect ratio are between0.5 and 10, although values between 3 and 5 are more common.

The number of vanes is simply the number of IGV vanes in the bladerow. Acceptable values for thenumber of vanes are between 3 and 100, although values between 10 and 30 are more common.

The profile trim is defined as the chord at the tip chord/hub chord. Acceptable values for profile trimare between 0.2 and 1.8. Often the profile trim for a rotor will be < 1, reducing the chord with increasingspan, and the corresponding IGVs and OGVs will have a profile trim > 1 in order to maintain a constantgap between bladerows.

Entries for Rotor and OGV are similar to those for IGV. The entries for IGV and OGV will be disabledaccordingly depending on the machine configuration.

Machine configuration:

Check boxes indicating whether an IGV and/or OGV will be included in the calculation.

Units:

Radio buttons to select either SI or Imperial units. This impacts both input data and results and may beswitched at any time with immediate effect.

8.2.3. Controls Tab

Figure 8.15: Control Input Tab

Throughflow controls:

Max iterations, Error tolerance and Relaxation factor.

The design and analysis calculations are made using an iterative throughflow solver. Max iterations,Error tolerance and Relaxation factor control the maximum number of attempted iterations, the rel-ative error in mass balance and the damping between iterations respectively. The default solver settings


Vista AFD

are robust and accurate and should only be adjusted where difficulties are encountered and modificationof the input data does not help.

Design Iterations (available only from the Design cell):

Inner loop iterations and Outer loop iterations.

In the design calculation, the throughflow code is executed multiple times in order to achieve the re-quired duty. The inner loop makes adjustments to the blade angles and requested pressure rise in orderto achieve the duty, while the outer loop updates the efficiency estimate and reruns both the meanlineand throughflow calculations in order to achieve a consistent result.

The process is illustrated by Figure 8.16: Design Calculation Flowchart (p. 193).

Figure 8.16: Design Calculation Flowchart

Export options

Blade to export (BladeGen) is used to select the blade row to export when exporting to BladeGen(BladeGen can only model a single blade row at a time.). By contrast, when exporting to BladeEditor,all blade rows are always exported.




Layers to export settings control which spanwise layers are exported. Although the throughflow calcu-lation uses geometry on 5 spanwise layers, the number of layers used to create the blade model canbe reduced by clearing the check boxes for any of the 3 internal layers. The hub and shroud layers aremandatory.

8.3. Results Tab

The results are arranged in four main panels, one for the overall machine performance and one for eachof the three blade rows. Results are presented for the blade rows specified by the machine configurationas appropriate:


Performance:

The rotor power and torque are listed here, together with the dynamic pressure at the outlet, whichneglects the swirl component, and the dynamic pressure downstream of the machine.

The downstream dynamic pressure is taken to be the dynamic pressure in a simple pipe of diameterequal to the outer diameter of the machine. This is assumed to be “well mixed” at 2.5 diametersdownstream of the rotor and again neglects any swirl component.

The aerodynamic efficiency is the adiabatic efficiency of the rotor (taken as the efficiency estimate inthe meanline calculation).

The system efficiency t-t is the efficiency of the assembly based on the specified total head rise.

The system efficiency t-s is the efficiency of the assembly based on the static head rise (that is, totalhead rise - outlet dynamic pressure).

The downstream efficiency t-s is the efficiency of the assembly based on the static head rise 2.5 diametersdownstream of the machine, i.e. (total head rise - downstream dynamic pressure).

Clearly all of the performance results must be positive for the design to be viable.


Vista AFD

Rotor:

This panel displays results for the following parameters at the meanline and hub sections:

Flow coefficient:

Loading:

DeHaller:

Deviation: −

Diffusion factor: − + −

Gas exit angle

It is advised that the flow coefficient is < 0.8, deHaller number > 0.7, deviation < 15°, diffusion factor <0.6 and the gas exit angle < 90°. The upper recommendation for loading is case dependent and is cal-culated based on a deHaller number of 0.7.

All of the above limits are shown in tooltip boxes that appear when the mouse rests over the particularparameter for a few seconds. Should the limits be exceeded in the calculation, the parameter in questionis highlighted in red for easy identification of the potential problem.

IGV and OGV:

A similar, but more limited, set of parameters, to those for the rotor, is presented in the panels for bothIGV and OGV as appropriate for the machine configuration.

8.4. Context Menu Commands of the Cells in the Vista AFD System

You can access a context menu for any cell in the Vista AFD system by right-clicking that cell. Most ofthe commands that are available are standard, and are described in Systems and Cells.

The context menu commands that are specific to the Meanline cell are described in Table 8.1: ContextMenu Commands Specific to the Vista AFD Meanline Cell (p. 195).

Table 8.1: Context Menu Commands Specific to the Vista AFD Meanline Cell

DescriptionCommand

This command opens the Vista AFD dialog boxwith the Aerodynamics, Geometry, and Res-

Edit

ults tabs available for performing and viewingthe results of a meanline simulation.

This command first prompts you to select aBladeGen file, then imports the Vista AFD set-



The BladeGen file is required to have beenwritten by an older version of BladeGen thatincluded Vista AFD; such an older version of



Context Menu Commands of the Cells in the Vista AFD System

DescriptionCommand

BladeGen wrote Vista AFD settings into theBladeGen file. It must not have been producedby the Create New > BladeGen command.

The context menu commands that are specific to the Design cell are described in Table 8.2: ContextMenu Commands Specific to the Vista AFD Design Cell (p. 196).

Table 8.2: Context Menu Commands Specific to the Vista AFD Design Cell

DescriptionCommand

This command opens the Vista AFD dialog boxwith the Controls and Results tabs available

Edit

for performing and viewing the results of aniterative series of meanline and throughflowsimulations.



in the Vista AFD system. For details, seeLaunching a New BladeGen Model (p. 197).



in the Vista AFD system. For details, seeLaunching a New BladeEditor Model (p. 197).

The context menu commands that are specific to the Design cell are described in Table 8.3: ContextMenu Commands Specific to the Vista AFD Analysis Cell (p. 196).

Table 8.3: Context Menu Commands Specific to the Vista AFD Analysis Cell

DescriptionCommand

This command opens the Vista AFD dialog boxwith the Controls and Results tabs available

Edit

for performing and viewing the results of athroughflow simulation.



in the Vista AFD system. For details, seeLaunching a New BladeGen Model (p. 197).



in the Vista AFD system. For details, seeLaunching a New BladeEditor Model (p. 197).

8.5. Creating a Blade Design

The preliminary fan design may be converted to a 3D geometry model in ANSYS BladeGen or ANSYSBladeEditor. To achieve the 3D vane shape, Vista AFD uses empirical estimates of vane camber andthickness distributions and these are then combined with the computed throughflow design. You canmodify the camber and the thickness in the subsequent BladeGen or BladeEditor model if required.


Vista AFD

To access the export options, edit the Design or Analysis cell. A dialog box will appear with the exportoptions on the Controls tab. For details, see Controls Tab (p. 192).

Details of creating a new BladeGen or BladeEditor model are described in the following sections:8.5.1. Launching a New BladeGen Model8.5.2. Launching a New BladeEditor Model

8.5.1. Launching a New BladeGen Model

A new BladeGen model can be generated from an up-to-date Vista AFD system using the Create New

> BladeGen command in the right-click context menu of the Design and Analysis cells.

Note

• The BladeGen model is detached from the Vista AFD system (that is, no link is generated).Therefore, any changes made to the Vista AFD system will not be reflected in the BladeGenmodel following an update of the Vista AFD system.

• The BladeGen file that you create in this way cannot be imported into another Vista AFD system,because it does not contain Vista AFD settings. A BladeGen file created by an older version ofBladeGen (that included Vista AFD) contains Vista AFD settings and can be imported into aVista AFD system.

8.5.2. Launching a New BladeEditor Model

A new BladeEditor model can be generated from an up-to-date Vista AFD system using the Create New

> Geometry command in the right-click context menu of the Design and Analysis cells. This commandlaunches ANSYS DesignModeler (with BladeEditor) and generates the new model.

Note

The DesignModeler model is detached from the Vista AFD system (that is, no link is generated).Therefore, any changes made to the Vista AFD system will not be reflected in the Design-Modeler model following an update of the Vista AFD system.

8.6. Troubleshooting and Error Messages

Data Input Errors:

Where a numerical value is required as an input but a text or blank entry is made, an error messagesimilar to the following will appear at the bottom of dialog box:

Invalid input for mass flow rate. Please check the value and try again.

This indicates exactly which input is causing the problem and suggests revision. Often this will be wherethe letter “O” has been typed rather than the number “0”. To rectify the problem, click OK to enablethe user interface, then edit the value that contains the error.

A different error message, similar to the following, will appear where an input value lies outside of theadvised range in the tooltip:

mass flow rate must be > 0. Please check the value and try again.



Troubleshooting and Error Messages

This is similarly specific and similarly resolved.

Calculation Errors:

During the meanline calculation, where the geometry has no OGV, the following error may appear:

Figure 8.18: Meanline Calculation Error Message

The usual cause of this error is that the specified mass flow rate is too large for the size of the machine.In this situation, either gradually increasing the outer diameter or reducing the mass flow rate shouldgive a valid meanline calculation.

This error may also occur during the design calculation. This is because the meanline calculation is up-dated using a new efficiency estimate during the design calculation outer loop. This suggests that theinitial efficiency estimate is too high and lowering this in the meanline calculation may give the sameerror. Again increasing the outer diameter or reducing the mass flow rate should resolve the problem.

During the design or analysis calculation, the following error message may occasionally appear:

Figure 8.19: Design Calculation Error Message

Even though the results of the meanline calculation will seem to be fine, with all parameters within therecommended guideline, the throughflow code may still fail. It is important to note that the bladegeometry (number of vanes, aspect ratio, and profile trim) does not influence the meanline calculationbut has a significant impact on the throughflow calculation.

Less aggressive settings for the geometry parameters (a larger number of vanes, lower aspect ratio, andprofile trim closer to 1) should lead to a successful design/analysis calculation. Exploring the designspace around the desired geometry in this manner will give a good indication of why the calculationfails under such conditions and what can be done to remedy the situation.

During the design and analysis calculations, a number of files associated with the throughflow are createdin subdirectories below the case name directory. Should the design and/or analysis calculation fail andprove difficult to remedy using the above techniques, these files will prove helpful to technical supportin diagnosing the issue.


Vista AFD

8.7. Notation

MeaningAbbrevi-

ation

inlet guide vaneIGV

outlet guide vaneOGV

adiabatic efficiency

total head riseP

relative angle

air absolute velocityC

air relative velocityV

blade tangential velocityU

change in enthalpyH

Superscripts

at the inlet to the blade rowin

at the outlet to the blade rowout

Subscripts

with respect to the bladegeometry

blade

with respect to the air flowgas

axial componenta

estimated valueest



Notation


Chapter 9: TurboSystem: ANSYS TurboGrid

ANSYS TurboGrid is a meshing tool that is specialized for CFD analyses of turbomachinery bladerows.The ANSYS TurboGrid documentation is available from the Help menu in ANSYS TurboGrid.

The main documentation is available from the Help menu in ANSYS TurboGrid, and consists of the fol-lowing parts:

• TurboGrid Introduction

• TurboGrid Tutorials

• TurboGrid User's Guide

• TurboGrid Reference Guide

To launch ANSYS TurboGrid from ANSYS Workbench, add the TurboGrid component system to yourproject schematic, then edit the Turbo Mesh cell of that system.

The geometry can be loaded from the File menu in ANSYS TurboGrid, or it can be specified by linkinga Geometry or Blade Design cell upstream of the Turbo Mesh cell.

In the case when data is transferred to a Turbo Mesh cell from a Geometry cell, the Turbo Mesh cellhas properties that control this transfer. These are described in Table 9.1: TurboGrid Turbo Mesh CellProperties (p. 201). Before attempting to modify these properties, be sure to refresh the Turbo Mesh cellif it is in a Refresh Required state. Refreshing this cell causes the properties to be updated.

Table 9.1: TurboGrid Turbo Mesh Cell Properties


This property displays a list of theavailable flow paths and bladerows.

Flowpath Op-tions

Geometry Se-lection

Use this information as a guidewhen specifying the Flowpath andBladerow properties (describedbelow). Use the Refresh commandin the context menu to update thelist after linking.

This property specifies whichFlowpath feature in BladeEditor

Flowpath

contains the bladerow that is to beloaded in ANSYS TurboGrid.

This property specifies whichbladerow (within the specified

BladerowNumber

Flowpath feature) is to be loadedin ANSYS TurboGrid.




This property specifies how the in-let points are positioned in Tur-

Inlet PositionMethod

boGrid. The Manual option meansthe user will specify these in Tur-boGrid. The Adjacent Blade optionmeans the inlet points will be cal-culated using the upstreambladerow, specified below.

This property is available only whenmultiple bladerows have beenspecified in BladeEditor.

This property specifies thebladerow number for the bladerow

UpstreamBladerowNumber that is immediately upstream of the

current bladerow.

This property is available only whenInlet Position Method is set to Ad-jacent Blade.

This property specifies how theoutlet points are positioned in Tur-

Outlet Posi-tion Method

boGrid. The Manual option meansthe user will specify these in Tur-boGrid. The Adjacent Blade optionmeans the outlet points will becalculated using the downstreambladerow, specified below.

This property is available only whenmultiple bladerows have beenspecified in BladeEditor.

This property specifies thebladerow number for the bladerow

DownstreamBladerowNumber that is immediately downstream of

the current bladerow.

This property is available only whenthe Outlet Position Method is setto Adjacent Blade.

If this property is selected, Tur-boGrid will generate an inlet do-main as part of the mesh.

Inlet DomainMeshing

If this property is selected, Tur-boGrid will generate an outlet do-main as part of the mesh.

Outlet Do-main

The minimum face angle of themesh is displayed here.

MinimumFace Angle


TurboGrid


The maximum face angle of themesh is displayed here.

MaximumFace Angle

This property specifies a string ofcharacters that is prefixed to all

Region NamePrefix

Mesh OutputOptions

mesh region names when the meshis written to file.

This property is blank by default.Note that when importing meshesinto ANSYS CFX-Pre, any duplicatemesh region names will be appen-ded with numbers to make themunique.

Each ExportPoints feature in BladeEditor defines a single blade, is associated with a FlowPath, and hasa Bladerow Number. To model a series of consecutive bladerows in a turbomachine, you should definea series of ExportPoints features associated with the same FlowPath feature, with Bladerow Numbersin numerical order (lowest number at the inlet end of the machine). By using the same FlowPath number:

• You have access to the Turbo Mesh cell properties that collectively control the position of the inlet andoutlet ends of each bladerow: Inlet Position Method, Upstream Bladerow Number, Outlet Position Method,Downstream Bladerow Number.

• The machine is eligible to be analyzed by Vista TF. For help on Vista TF, see Vista TF User's Guide (p. 209).

If you want a given bladerow to contain more than one blade geometry (for example, main blades withsplitter blades), create one ExportPoints feature for each unique blade in the bladerow, with each Ex-portPoints feature based on the same FlowPath and given the same Bladerow Number. When morethan one ExportPoints feature matches the FlowPath and Bladerow Number criteria set in the TurboMesh cell properties, ANSYS TurboGrid will create a bladerow with splitter blades.

You can access a context menu for the Turbo Mesh cell in the TurboGrid component system by right-clicking the cell. Most of the commands that are available are standard, and are described in Systemsand Cells. The context menu commands that are specific to the Turbo Mesh cell are described inTable 9.2: Context Menu Commands Specific to the Turbo Mesh Cell (p. 203).

Table 9.2: Context Menu Commands Specific to the Turbo Mesh Cell

DescriptionCommand

This command opens TurboGrid and loads the geometryand state, regenerating the mesh if appropriate.

Edit

This command opens TurboGrid and loads the geometryand state, but suspends the Topology from regenerating.

Edit with To-pology Sus-pended Select this option if you want to open TurboGrid more

quickly to make changes to the mesh settings withoutfirst regenerating the mesh.




Chapter 10: TurboSystem: ANSYS CFX-Pre

ANSYS CFX-Pre is a general-purpose CFD preprocessor that has a turbomachinery setup wizard for facil-itating the setup of turbomachinery CFD simulations.

See CFX-Pre User's Guide.




Chapter 11: TurboSystem: ANSYS CFD-Post

ANSYS CFD-Post is a general-purpose CFD postprocessor that has features for facilitating the postpro-cessing of turbomachinery CFD simulations.

See CFD-Post User's Guide.




Chapter 12: TurboSystem: ANSYS Vista TF

The Vista TF program is a streamline curvature throughflow program for the analysis of any type ofturbomachine, but has been developed in the first instance primarily as a tool for radial turbomachineryanalysis. The program enables you to rapidly evaluate radial blade rows (pumps, compressors and tur-bines) at the early stages of the design.

Vista TF is provided by PCA Engineers Limited, Lincoln, England.

The documentation for Vista TF is provided in the following sections:12.1.Vista TF User's Guide12.2.Vista TF Reference Guide

12.1. Vista TF User's Guide


Vista TF is operated from ANSYS Workbench by working with either the Vista TF component system orthe Throughflow analysis system. The Vista TF component system is comprised of three cells: a Setupcell, a Solution cell, and a Results cell:

The Throughflow analysis system is essentially a Vista TF system with an added Geometry cell:



The name of the system is changeable upon first adding the system, or by right-clicking the blue celland selecting Rename, then typing in a new name. Note that the name of the system appears belowthe system.

To create a new Throughflow analysis system in Workbench, you can drag the Throughflow analysissystem from the Toolbox to the Project Schematic, or double-click the system in the Toolbox. You canalso create a new Throughflow analysis system by right-clicking a Blade Design cell in a Vista CCD, VistaCPD, or Vista RTD component system, and selecting Create New > Throughflow. Upon creating aThroughflow system using data from any one of the Vista CCD, Vista CPD, or Vista RTD systems, eachof the cells in the Throughflow system is updated automatically.

Vista TF uses the following input to define a run:

• A geometry (*.geo) file

• Setup cell properties

• Three Vista TF template files

– a control data file (*.cont)

– an aerodynamic data file (*.aert)

– a correlations data file (*.cort)

The general procedure for running a simulation in Vista TF is:

1. Drag the Vista TF component system from the Toolbox to the Project Schematic, or double-click thesystem in the Toolbox.

2. Specify a geometry file using either one of the following methods:

• Connect an upstream Geometry cell that contains a VistaTFExport feature to the Setup cell.

If there is more than one VistaTFExport feature, then the first valid and unsuppressed one isused.

• Right-click the Setup cell, select Import Geometry, and browse to select a geometry (*.geo) file.

3. Double-click the Setup cell, then configure the Setup cell properties.

For details, see Vista TF Setup Cell Properties (p. 211).

4. Optionally customize one or more of the three template files (*.cont, *.aert, *.cort).

For details on customizing the template files, see Customizing the Vista TF Template Files (p. 217).

5. Update the Solution cell, or update the Project, to generate a solution.

6. Double-click the Solution cell to view the solver output.

7. Double-click the Results cell to view the results in CFD-Post.

The Results cell has one property that you can edit to control report generation.


Vista TF

8. When the analysis is complete and the project is finished, you save the project (and therefore the asso-ciated files). Once a project has been saved, it can be re-opened at a later date for review or modificationof any aspect of the simulation.

Important

Saving a project enables you to re-open the project on the machine that originally createdit. To make the project available on another machine, you need to use File > Archive tocreate a project archive. To open the project on a different machine, run File > Restore

Archive on that machine.

The following topics are discussed:12.1.1.Vista TF Setup Cell Properties12.1.2. Customizing the Vista TF Template Files12.1.3.Vista TF Context Menu Commands

12.1.1. Vista TF Setup Cell Properties

The main properties that control Vista TF are associated with the Setup cell. To see the properties, doany one of the following:

• Right-click the Setup cell and select Edit.

• Double-click the Setup cell.

• Right-click the Setup cell and select Properties.

A sample of the cell properties is shown in Figure 12.1: Properties of the Vista TF Setup Cell (p. 212).



Vista TF User's Guide

Figure 12.1: Properties of the Vista TF Setup Cell

Table 12.1: Vista TF Setup Cell Properties (p. 212) describes each of the cell properties.

Table 12.1: Vista TF Setup Cell Properties


The name of the cell with whichthe present set of properties is as-sociated.

ComponentID

General

This is the system directory namethat appears within the Projectfiles.

DirectoryName

This property appears when thereis no cell upstream of the Setupcell.

Machine Type


Vista TF


This property defines the machinetype.

The choices are:

• Pump

• Axial Compressor

• Centrifugal Compressor

• Fan

• Axial Turbine

• Radial Turbine

• Hydraulic Turbine

• Other

• Unknown

This property specifies which tem-plate files are used and which re-port is used for the results. You cancustomize these templates. For de-tails, see Customizing the Vista TFTemplate Files (p. 217).

Note that this property does notappear in Throughflow systemsbecause Throughflow systems re-ceive this data from the Geometrycell, which is upstream of the Setupcell.

This property appears when thereis no cell upstream of the Setupcell.

Number ofBlade Rows

This property defines the numberof blade rows in the geometry(.geo) file.

Note that this property does notappear in Throughflow systemsbecause Throughflow systems re-ceive this data from the Geometrycell, which is upstream of the Setupcell.

This property defines the numberof meridional streamlines to use in

Number ofStreamlines

Solver Set-tings





the calculation. For details, seeSpecification of the Control DataFile (*.con) (p. 226).

This property defines the maximumnumber of solver iterations. For

Maximum Iter-ations

details, see Specification of theControl Data File (*.con) (p. 226).

This property defines the directionof rotation of the blades about theZ axis.

Machine Rota-tional Direc-tion

OperatingConditions

The choices are:

• Right-handed

• Left-handed

This property defines the rotationalspeed of the machine.

Machine Rota-tional Speed

This property specifies the types ofboundary conditions.

Flow Option

The choices are:

• Mass Flow

• Pressure Ratio

• Pressure Difference

For details, see Specification of theControl Data File (*.con) (p. 226) andSpecification of Aerodynamic DataFile (*.aer) (p. 245).

This property defines the outlet-to-inlet ratio of absolute total pres-

Pressure Ra-tio

sures (each pressure measured inthe stationary frame).(For Flow Op-

tion = Pres-sure Ratio)

This property defines the outlet-to-inlet difference of total pressures

Pressure Dif-ference

(each pressure measured in thestationary frame).(For Flow Op-

tion = Pres-sure Differ-ence)

This property defines the inlet massflow rate.

Mass FlowRate


Vista TF


This property defines the inlet ab-solute total pressure (measured inthe stationary frame).

Inlet TotalPressure

This property defines the inlet ab-solute total temperature (measuredin the stationary frame).

Inlet TotalTemperature

This property defines the angle ofthe inlet velocity measured with

Inlet SwirlAngle

respect to the meridional plane. Apositive angle implies that the flowswirls in the Machine RotationalDirection.

This property defines the referencediameter for all blade rows. For

ReferenceDiameter

ReferenceValues

details, see Specification of Aerody-namic Data File (*.aer) (p. 245).

This property defines the small-scale polytropic efficiency for themachine.

Polytropic Effi-ciency

This property defines the type offluid that flows through the ma-chine.

Fluid OptionFluid Proper-ties

The choices for this property are:

• Ideal Gas

• Real Gas

• Liquid

This property defines the specificheat capacity (at constant pressure)of the ideal gas.

Gas SpecificHeat Cp

(For Fluid Op-tion = IdealGas)

This property defines the specificheat ratio of the ideal gas.

Specific HeatRatio

(For Fluid Op-tion = IdealGas)

This property is used to specify areal gas from a list of available real

Material

(For Fluid Op-tion = RealGas)

gases. You can select Custom tospecify a custom real gas propertyfile. The equivalent real gas files forthe standard materials appear in





the Ansys installation under AnsysInc/v145/Addins/VistaTF/Real-Gas and can be used as a startingpoint for you to generate your owncustom real gas files. The format ofthe real gas files is described inSpecification of the Real Gas Prop-erties Data File (*.rgp) in theTurboSystem User Guide.

This property defines the densityof the liquid.

Fluid Density

(For Fluid Op-tion = Liquid)

This property defines the specificheat capacity of the liquid.

Fluid SpecificHeat

(For Fluid Op-tion = Liquid)

This property indicates the dynamicviscosity of the fluid as follows:

Dynamic Vis-cosity

• A value less than 1 [N s m^-2] (orequivalent value in other units) is in-terpreted as the dynamic viscosity.Note that the value must be greaterthan 0.0000001 [N s m^-2].

• A value of 0 causes Vista TF to calcu-late the dynamic viscosity from aninbuilt equation for dynamic viscositybased on Sutherland’s law and theInlet Total Temperature. This worksonly for an ideal gas.

• A value greater than 1 [N s m^-2] (orequivalent value in other units) is in-terpreted as the Reynolds number, inwhich case Vista TF calculates thedynamic viscosity using this Reynoldsnumber, the Reference Diameter, theMachine Rotational Speed, and thefluid density.

This property serves as an initialguess for the meridional velocity

InitialCm/U_ref

Initial Condi-tions

divided by a characteristic velocity,where the latter is half the Refer-ence Diameter multiplied by theMachine Rotational Speed. Formore information, see the descrip-


Vista TF


tion for cm_start in Specificationof the Control Data File(*.con) (p. 226).

12.1.2. Customizing the Vista TF Template Files

When you run Vista TF, one of each of the *.cont, *.aert, and *.cort data files are copied from the VistaTF template directory into your working directory as required (that is, if they are not already present inthe working directory). The exact *.cont, *.aert, and *.cort files that are copied (and then used during arun) depend on the Machine Type setting in the Setup cell properties. You can use custom versions ofany of the *.cont, *.aert, and *.cort files. To customize one of these files:

1. Import the template you want to customize by right-clicking the Setup cell and selecting one of theImport Template commands.

The selected template file is copied to the working directory. If you import the same template filemore than once, the name of the copied file is changed automatically to produce a unique filename; the last one imported will take effect when you start a run.

2. From the Workbench main menu, select View > Files to see the template files you have imported.

3. Right-click the template file that you want to customize, and select Open Containing Folder.

4. Open the template file in a text editor and change it.

In these files, do not change setting values that are between a pair of braces (“” and “”). You canchange setting values which are not wrapped in braces.

The settings of the template files correspond with the settings of the *.con, *.aer, and *.cor fileswhich are described in Specification of the Control Data File (*.con) (p. 226), Specification of Aero-dynamic Data File (*.aer) (p. 245), and Specification of Correlations Data File (*.cor) (p. 255).

12.1.3. Vista TF Context Menu Commands

You can access a context menu for each cell in the Vista TF component system by right-clicking a cellin the system. Most of the commands that are available are standard, and are described in Systems andCells. The context menu commands that are specific to the Vista TF system cells are described inTable 12.2: Context Menu Commands Specific to the Vista TF System Cells (p. 217).

Table 12.2: Context Menu Commands Specific to the Vista TF System Cells

DescriptionCommandCell

This command opens theVista TF Properties view.

EditSetup

This command enablesyou to specify the geo-

Import Geometry

metry file, provided thatthere is no upstreamGeometry cell linked tothe Setup cell.





Note that, when a geo-metry file is imported,Vista TF saves a referenceto the imported file, butthe file is not copied intothe project. This is so thatthe geometry file can beupdated outside of theproject. If the geometryfile is updated, the pro-ject will not recognizethe update automatically.To force the system toreread the updated file,you need to right-clickthe Setup cell and selectClear Generated Data

before updating the sys-tem.

If you archive the Project,you will be presentedwith an option to archiveImported files external

to project directory. Ifyou choose this option,the imported geometryfile will be added to theimport_files directory inthe archived version ofthe project (but not inthe original project). Evenwhen the imported geo-metry file is in the im-port_files directory, up-dates to the geometryfile are not recognizedautomatically, and it isnecessary to use theClear Generated Data

command, as describedabove.

This command enablesyou to import a real gasproperty (.rgp ) data file.

Import Real Gas...

This command is onlyavailable when Fluid Op-tion = Real Gas and Ma-terial = Custom in theProperties view.


Vista TF


This command enablesyou to import a template

Import Template

file. For details, see Cus-tomizing the Vista TFTemplate Files (p. 217).

This command opens theVista TF screen output

View Solver OutputSolution

(.scn file) for viewing. Fordetails, see Screen Out-put Files(screen.scn) (p. 226).

This command restartsthe solver. You can use

Continue Calculation

this command to contin-ue a run that did notconverge; in this case, theUpdate command maynot work because the cellis already up-to-date.

This command opens theresults in CFD-Post.

EditResults

12.2. Vista TF Reference Guide


The key aspect of this document is the input and output data specification for the program and howto run it.

The input files include a control file, an aerodynamics file, a geometry file, a correlations file, and canalso include a restart file providing data from a previous converged simulation, and a file with real gasdata. The input files include comment lines to help the reader to identify the parameters.

The output files include a results file with text output for analysis of the simulation giving:

• Data on streamlines and quasi-orthogonals (short for “quasi-orthogonal calculating stations”)

• Various files with the same information that can be used for plotting the results

• A file which monitors the history of the simulation

• A file to act as interface to other software systems

• A restart file which can be used to initialise a further simulation.

For turbomachinery calculations in subsonic flow the program is very robust. Note, however, that ro-bustness decreases with increasing Mach number in transonic cases. Some tips on dealing with possibleproblems with running the program are provided in Appendix F: Troubleshooting (p. 300).



Vista TF Reference Guide

The topics in this guide are:12.2.1. Running Vista TF from the Command Line12.2.2. Input and Output Data Files for Vista TF12.2.3. Software Limitations12.2.4. Streamline Curvature Throughflow Theory12.2.5. Appendices

12.2.1. Running Vista TF from the Command Line

You can run Vista TF from the command line of your operating system. By default, the executable willlook in the current directory for the required input files for your case. One of the input files, which hasa default name of vista_tf.fil , contains the names of the other input files that are to be used forthe case. If this input file has a different name, specify that name as the first command line argument.For example:

VistaTF.exe myfile.fil

When the executable runs, it writes text output messages to the console. To redirect these messagesto a file, append -silent to the command line. The default file name for storing the redirected outputmessages is screen.scn . To store the output messages in a file of a different name, specify that namefollowing -silent .

After running the program, several files are produced in .csv format as the basis for producing plotswith CFD-Post or in .txt format as the basis for producing plots with Tecplot.

The input and output files for VistaTF.exe are described in detail in Input and Output Data Files forVista TF (p. 220).

12.2.2. Input and Output Data Files for Vista TF

The following sections describe the input and output files for VistaTF.exe .12.2.2.1.The Auxiliary File with the Default Name: vista_tf.fil12.2.2.2. Overview of Input Files12.2.2.3. Overview of Output Files12.2.2.4. Specification of the Control Data File (*.con)12.2.2.5. Specification of the Geometry Data File (*.geo)12.2.2.6. Specification of Aerodynamic Data File (*.aer)12.2.2.7. Specification of Correlations Data File (*.cor)12.2.2.8. Specification of the Real Gas Properties Data File (*.rgp)12.2.2.9. Specification of the Output Data File (*.out)12.2.2.10. Specification of the Text Data Files (*.txt)12.2.2.11. Specification of the CFD-Post Output Files (*.csv)12.2.2.12. Specification of Convergence History Data File (*.hst)

12.2.2.1. The Auxiliary File with the Default Name: vista_tf.fil

You specify the input data and output data file names in an auxiliary data file that has the default file-name vista_tf.fil . This file can have another name if this name is passed to the program througha command-line argument to specify the auxiliary filename, as described in Running Vista TF from theCommand Line (p. 220). If no command-line argument is specified in this way, then the program assumesthat the file has the name vista_tf.fil . This auxiliary file in turn must contain the necessary filenamesfor the input and output files in the following order and form:


Vista TF

Version number Version 3.0control datafile name prefix.congeometry datafile name prefix.geoaerodynamic datafile name prefix.aercorrelation datafile name prefix.corresults output file name prefix.outconvergence history filename prefix.hsttext data output file name prefix.txtCfx-post output file name prefix.csvrestart datafile name prefix.rstInterface output file prefix.intReal gas properties data file prefix.rgp

Note that this file has changed its structure from version 12.1 to version 13, but provided that no realgas calculations are needed, files used in version 12.1 will still work without modification.

Note that the prefixes need not be identical for a given run. In fact this is not usually the case. An exampleof a vista_tf.fil file is:

Version 3.0standard_control.conimpeller_XYZa.geodesign_point.aerradial_impeller.corresults.outhistory.hstimpeller.txtcfx_post.csvrestart.rststream.intreal_gas_CO2.rgp

Note that the program also produces and uses other files in special situations as outlined below; theirnames do not need to be specified separately because they are determined by the program. The datafiles, results files, and the file vista_tf.fil are usually in the same directory. If the history (.hst )file already exists in the working directory before the program is run, it will be overwritten. If the restart(.rst ) file already exists in the working directory before the program is run, it will be overwritten onlyif the solution has converged or reached the maximum number of iterations that you have specified.If the output (.out ) file and the plot files (.txt , and .csv ) already exist in the working directory beforethe program is run, they will be overwritten. The program will not prompt you for permission to overwritethese files.

12.2.2.1.1. Backwards Compatibility and Cases without Real Gas Data

The capability to run with real gas data (starting in version 13.0) has called for an additional input filethat provides the coefficients needed in the real gas equations. This has required a change in the inputfile description. If calculations are made that do not make use of real gas properties then the real gasfile is not needed, and the format of the vista_tf.fil file can remain the same as in versions priorto 13.0 (that is without the first line giving the version number and without the last line giving the realgas properties file name), as in the example that follows:

standard_control.conimpeller_XYZa.geodesign_point.aerradial_impeller.corresults.outhistory.hstimpeller.txtcfx_post.csvrestart.rststream.int




12.2.2.2. Overview of Input Files

Four input data files are always needed:

Control data file (.con)Geometrical data file (.geo)Aerodynamic data file (.aer)Correlations data file (.cor)

A fifth input file will be used if it is available and if you specify that it should be used:

Restart data file (.rst)

A sixth input file is needed if the calculation requires real gas properties:

Real Gas Property data file (.rgp)

The division of the input data into separate files provides a simple and clear way to vary or retain theannulus geometry, the aerodynamic conditions, the blade element data, or the correlations being used,without changing all the files in use. Typically during the design process, you will change the .geo fileto examine a new geometry, and the .aer file to examine new operating points or boundary conditions,and you will leave the .con and .cor files untouched once you have configured their settings to meetthe requirements.

The data specified in the individual files is structured to be as logical as possible, but some small overlapbetween the different files is inevitably necessary. The structure may appear more complicated thannecessary, but this arises from the requirement that ultimately the program should calculate all typesof turbomachinery in single stage and multistage configurations, both as ductflow and as throughflowcalculations. During the development an attempt has been made to include a built-in “expert system”in the program. For example, the program itself is able to identify whether a particular blade row is aradial compressor impeller or a radial turbine inlet guide vane (from the geometry) and ultimately willbe able to select automatically the most appropriate correlations to be used. In general many parametersmay be set to zero and the program selects the value it deems appropriate. “Expert parameters” allowyou to override the selections that the program would automatically make.

The functions of the six input data files are summarized next:

Control Data File (.con)

This is a short file giving values of identifiers of the file (title and headers) and control information andconstants defining such choices as the number of streamline calculating planes, the convergence toler-ance, the relaxation factors, and so on. Various “expert” parameters are also specified in this file. Alsospecified in this file are the planes and stations for which output information is required, and the levelof detail requested on these planes.

To make the program easier to use, you can specify many of these parameters as 0.0, 1.0, or 0, and theprogram will then make a sensible choice of the value for the parameter concerned, so that typicallyyou are only concerned with two or three parameters in this file. The control parameters that determinethe selection of particular numerical models are also defined in this file, for example the type of span-wise mixing or the model for blade row choking. In general, this input file does not need to be changedfrom run to run.

Geometry Data File (.geo)

The geometry data file contains the dimensions of the annulus in terms of the axial and radial coordinatesof the quasi-orthogonal calculating planes at hub and casing, and information to identify the type of


Vista TF

calculating station (such as duct, stator, or rotor). Calculating planes can represent regions of a duct orblades (leading edges, trailing edges, and internal stations) and can be curved or linear. For linear cal-culating planes (which, by definition, are straight in the meridional plane), details of the geometry ofeach calculating plane are specified at only two points and intermediate geometric values are interpolatedlinearly from these, whereas curved planes require more points to be specified across the span. Forcurved duct calculating planes, this geometry data specifies the co-ordinate points along the curvedcalculating plane. In blade regions, the coordinates of the calculating plane, together with informationabout the blade geometry at this location, must be specified (including the number of blades, bladelean angles, and blade thickness). For each blade row, additional geometrical parameters can be specifiedthat might be relevant for the correlations (such as the throat area, the location of the throat, the locationof maximum camber, the maximum thickness, the trailing edge thickness, and the tip clearance).

In general, the geometry data file will be generated automatically using a blade geometry definitionprogram (such as ANSYS BladeEditor, or Vista GEO of PCA). A geometry conversion program is availableto convert data from the BladeGen meanline RTZT output format into the .geo file format for thethroughflow program and this has been tested for radial impeller rotors and stators, and axial statorand rotor blade rows (compressors and turbines); see Appendix G: The RTZTtoGEO Program (p. 309).Other custom tools are available for conversion of geometrical data from specific formats into the VistaTF .geo file format, and others can be prepared as required.

Aerodynamic Data File (.aer)

The aerodynamic data file contains the definition of the fluid, boundary conditions, and operating datasuch as inlet conditions and rotational speed. It includes parameters related to aerodynamic modelsfor the mean stream surface description, and the spanwise mixing coefficient.

Correlation Data File (.cor)

The correlation data file provides details of control parameters and empirical constants and data forthe particular choice of empiricism that has been chosen. The method allows a general specification oflosses, flow angle, and blockage for all calculating planes and across the span through the definitionof the spanwise variation of these parameters at particular quasi-orthogonal locations and for particularblade rows. Ultimately, in many cases, if default values of zero are chosen for these parameters thenthe program should automatically select appropriate correlations and make its own choice of correlationparameters.

Restart Data File (.rst)

This restart file contains some key information from a previous calculation in a non-dimensional form.Note that the restart file can be for different flow conditions and for a different geometry but it musthave the same number of quasi-orthogonal calculating stations and streamlines as the current calculation.If the restart data file has been generated from a calculation with similar geometry and flow conditionsas the current calculation, it provides a much better initial estimate of the flow and the streamline pos-itions than the first estimate generated internally within the program, promoting more rapid convergence.A restart with unchanged conditions and geometry will generally have a meridional velocity error ofless than 2% and will converge almost immediately, except for choked flows where more iterations areneeded. Convergence with the restart file is never immediate, even with unchanged geometry and flowconditions, because not all of the solution is saved to the restart file, and so some data needs to beregenerated over a few iterations of the solution. For small changes in flow conditions or geometry,the number of iterations when using the restart file is generally less than 50% of that required whenstarting from the program's own first estimate.




An existing restart file cannot be used if the number of streamlines or quasi-orthogonals is changed.The program recognizes if the number of streamlines or quasi-orthogonals has been changed and makesa new cold start in this case.

You do not have to be concerned with the content and format of the restart file because it is generatedautomatically at the end of a run of the program, and is automatically used if it is available. No furtherinformation is provided here with regards to the content of the restart file. In some situations where itis difficult to obtain convergence, the restart file can be used to store results for a converged operatingpoint (at lower speed, for example) and then the required operating condition can be obtained bystarting from the restart file with new flow conditions. In other cases where an un-converged solutionhas been stored in the restart file, it is possible that using the restart file can be disadvantageous as astarting point for a new simulation, and a cold-start may be better.

The restart file can also be used for reducing the number of computations when the program is coupledto an optimizer. In this case, an additional restart file with the name best_restart.rst is used andgenerated.

Real gas property data file (.rgp)

This file contains coefficients and data required to define the gas properties of a real gas. Note thatstandard files have been prepared for the most usual gases. Note this file does not have the same formatas the typical ANSYS .rgp real gas property data files.

12.2.2.3. Overview of Output Files

The program always creates the following three output files:

• Results output file (.out )

• Convergence history file (.hst )

• Restart data file (.rst )

In addition, the program can create the following output files of tabular data for plot and display pur-poses, depending on the value of the parameter i_display in the control file:

• Several comma separated variable output files for CFD-Post (.csv )

• Several text output files in a format suitable for Tecplot (.txt )

There is one CFD-Post output file for a calculation with no blade rows and four additional files for eachblade row. There are two Tecplot output files for a calculation with no blade rows, and an additionalfile for each blade row.

In addition, the program can create a data file containing data in a specific format for use with otherprograms:

• Interface output file (.int )

and a file which can be produced as an alternative to the screen output:

• Screen output file (screen.txt )


Vista TF

Results Output File (.out)

This contains rudimentary details of the data used for the calculation and the results of the calculationat every plane and radial station for which output has been requested. An overview of the content ofthis file is given in Specification of the Output Data File (*.out) (p. 267).

Convergence History File (.hst)

This contains a recording of the input data, followed by details of the convergence of the main iterativeprocedures, and extensive details of the terms in the radial equilibrium equation for each stream tubeand calculating plane. It is rare for this to be examined in any depth, but this can be useful to identifyproblems if the solution fails to converge.

Restart Data File (.rst)

This restart file stores information from a converged calculation in a non-dimensional form. It providesa much better initial estimate of the flow and the streamline positions than the first estimate generatedinternally within the program. It reduces the calculation time for a calculation with slightly modifiedgeometry or changed aerodynamic data by more than 25%. If an existing restart file is available, it willbe overwritten by the program.

Comma Separated Variable Output Files for CFD-Post (.csv)

Depending on the value of i_display in the control file, the following files are produced:

• prefix.csv

• global_prefix.csv

together with four additional files produced for each blade row from 1 to n:

• row_0n_hub_prefix.csv

• row_0n_mean_prefix.csv

• row_0n_tip_prefix.csv

• row_0n_loading_prefix.csv

The first file (prefix.csv ) contains key results of the calculation at every calculating plane andstreamline in a form that can be used for setting up a meridional contour plot of the results. The secondcontains a summary of the global performance and reference parameters for the calculation. The addi-tional four .csv files are produced for each blade row in the calculation. These contain the same in-formation as in the row data from the .txt files, but separated into hub, mean, and tip streamlinedata, which is information along the blade calculating station from leading to trailing edge on the hub,mean, and tip streamlines. The additional file contains spanwise variation of data. This can be used todefine typical blade loading diagrams and incidence plots for each blade row.

Even if no .csv file is required, the prefix.csv file still needs to be specified in the vista_tf.filfile.

Text Data Output Files for Tecplot (.txt)

Depending on the value of I_display in the control file, the following files are produced:




• prefix.txt

• test_prefix.txt

Together with one additional file produced for each blade row from 1 to n:

• row_0n_prefix.txt

The first file (prefix.txt ) contains key results of the calculation at every calculating plane andstreamline in a form that can be used for setting up a meridional contour plot of the results, usingTecplot software. This is an ASCII file which is formally correct for presenting the results in graphicalform with the plot processing software Tecplot, but can be used by other plot systems (such as Excel)with appropriate conversion or macros. A standard layout file for Tecplot (flowfield_2d.lay ) hasbeen prepared for typical meridional plots from Vista TF calculations. There are no macros included inthis so this may need some adjustment for a typical case (scale of axes, level of contour values, and soon). The second text file (test_prefix.txt ) contains the grid of the initial estimate of the streamlinesand quasi-orthogonals. This can be useful for debugging purposes and can be used to plot the initialgrid of an un-converged calculation to identify any specific problems with this. The additional .txtfiles are produced for each blade row in the calculation (row_0n_prefix.txt where n is the numberof the blade row from the inlet). These can be used to define typical blade loading diagrams and incid-ence plots for each blade row. The Layout files for Tecplot that have been prepared in advance assumethat the .txt file has the prefix “impeller ”.

Even if no .txt file is required, the prefix.txt file still must be specified in the vista_tf.filfile.

Interface Output Files (.int)

If you request the generation of an interface file for another analysis program then the appropriate filesare also generated. The first use of this has been established to allow a summary of the results to beobtained as input to an optimizing software system. A second option will be to generate the .streamfile for MISES blade-to-blade calculations providing the streamtube thickness and the radius along astream section. Even if no interface file is required, the .int file still must be specified in thevista_tf.fil file.

Screen Output Files (screen.scn)

In normal operation, the progress of the program can be seen on the screen. If the command line includesa parameter -silent , the screen results will be written to a separate file called screen.scn and notto the screen. If this parameter is not present, this output goes to the screen. If a name follows thisparameter, the screen results are printed to a file with this name.

12.2.2.4. Specification of the Control Data File (*.con)

The control data file includes lines of text that have no function other than to help you to identify theparameters defined here. Note that if values are set to 0.0 or 0 then standard values are used, so typicallyyou do not have to worry about this input. If zero values are specified for some parameters then thevalues actually selected by the program are written to the output file. Standard forms of this file areavailable for editing to meet the specific requirements, whereby in most cases no modification of thefile is necessary. An example of a control data file is given in Appendix B: Example of a Control DataFile (*.con) (p. 291).

Section 1: Character strings identifying the control data (max 72 characters/line)

The syntax is:


Vista TF

Character string - title(1)Character string - title(2)Character string - title(3)

Section 2: Integer control parameters

The syntax is:

n_sl max_it_main max_it_mass

DescriptionParameter

Number of meridional streamlinesn_sl

Notes:

• Must be an odd number so that there is always a midstreamline.

• Typically n = 9 or 17. If n_sl = 0, then 9 will be used.

• Maximum n_sl = max_n_sl = 29.

• If the mixing model is being used, (i_mix > 0 in section4) then there has to be a minimum of 9 streamlines.

Maximum number of iterations of the main streamlinecurvature loop in the iterative method.

max_it_main

Notes:

• Typically specified as 500 but fewer are generally neededfor simple radial compressor calculations to attain conver-gence.

• If max_it_main < 4 then max_it_main = 4, so that atleast 4 iterations are always done as a minimum.

• If max_it_main = 0 then max_it_main = 500.

• If the flow reaches the convergence limit before the max-imum number of iterations is reached then the calculationis automatically stopped earlier.

Maximum number of iterations for internal mass flowloop at each quasi-orthogonal calculating station.

max_it_mass

Notes:

• Typically 10 and if max_it_mass = 0, then max_it_mass= 10.

• If max_it_mass < 5 then max_it_mass = 5; experienceshows that this is a sensible value.

• If max_it_mass > 20 then max_it_mass = 20.

• If the mass flow convergence tolerance at a particular quasi-orthogonal is reached before the maximum number of it-





erations are completed then mass flow iteration is stoppedearly.

Section 3: Integer control parameters that control input and output data

The syntax is:

i_print_plane i_print_level i_progress i_display i_restart i_interface

Note that setting all of these parameters to 0 gives a standard form of output.


Determines the quasi-orthogonal calculating planes atwhich data is output into the results file.

i_print_plane

= 0, as i_print_plane = 4

= 1, output at no planes

= 2, data at inlet and outlet planes only

= 3, data at leading edges and trailing edges and inletand outlet planes only

= 4, data at all planes

Note:

The extent of the data printed at each plane is determ-ined by i_print_level .

Determines the level of output data printed into theresults file at each output plane.

i_print_level

= 0, standard output (as iprint_level = 3)

= 1, very limited data at each plane

= 2, generous level of data at each plane

= 3, extensive data at each plane

Note:

The planes at which output is available are defined bythe parameter i_print_plane .

Determines the extent of intermediate data that is prin-ted to the various files.

i_pro-gress

• If i_progress = 0 then no intermediate information isprinted.


Vista TF


• If i_progress = 1 then intermediate progress of the iter-ations are printed to the history file.

• If i_progress = 2 then data is printed to the history fileand to the screen.

Determines extent of tabular data output which is pre-pared for displaying the results with other plot and post-processing tools:

i_display

Notes:

• If i_display = 1, then no plot files are produced.

• If i_display = 0 then the output files of type .txt areproduced for display of the results with Tecplot.

• If i_display = 2, then a comma separated variable filewith extension .csv is produced for display of the resultswith CFD-Post.

• If i_display = 3, then both i_display = 0 and 2 aboveare activated.

• Other output formats can be incorporated as requested.

Determines whether the restart file should be used andwhether the results will overwrite the restart file con-tents.

i_restart

Notes:

• If i_restart = 0, the restart file (prefix.rst ) will beused automatically if it is present (a warm start) and itscontent will be overwritten automatically at the end of anormal calculation. Note that this also overwrites the restartfile even if the iterations are not converged, so that asecond start with the same number of iterations starts witha better approximation. This is the normal way to use theprogram. The restart file includes the number of quasi-or-thogonals and streamlines. If i_restart = 0 and thisnumber has changed then the program makes a cold startwith its own estimate of initial conditions (as in i_re-start = 1).

• If i_restart = 1 the restart file (prefix.rst ) will notbe used even if it is available and the program will set upits own initial conditions (a cold start). The content of therestart file will be overwritten as under 1 above. This is notgenerally recommended but can be useful in debuggingdifficult cases. This is equivalent to deleting the existingrestart file and using option 1 above.





• i_restart = 2 and 3 are special options for running theprogram when coupled to an automatic optimizer. In thesecases the use of a good restart file reduces the number ofiterations needed and brings a reduction in calculatingtime. Unfortunately some of the geometries being ex-amined may be poor and so it is inadvisable to overwritethe restart file with poor results. If i_restart = 2 thenthe file runs with a restart file called best_restart.rstand writes the results on to prefix.rst . If i_restart= 3 then the file runs with a restart file called best_re-start.rst and also writes the results on to the same filebest_restart.rst . In both cases if the restart filebest_restart.rst is not present then the internal ini-tial estimate is used (a cold start).

Determines the type of output interface file that is gen-erated.

i_inter-face

Notes:

• If i_interface = 0 then no output interface file is gen-erated.

• If i_interface = 1 then the prefix.int file containsa summary of the results for use in radial compressor op-timization.

• Other options are in preparation, such as a link to theblade-to-blade program MISES.

Section 4: Integer control parameters for various models and reference parameters

The syntax is:

i_expert i_flow i_fluid i _inbc i_mass i_mix i_ree


Allows special calculations to be carried out making useof development features of the program. Normally you

i_expert

would set the value of this parameter to 0 or 1, butother expert features of the program may be modifiedwith this control parameter. Each digit of the parameterhas an influence on its effect.

Notes:

• The last digit controls the choke calculation mode. Usinga value of zero for the last digit causes the choke massflow limitation to be eliminated which may be more robustin difficult cases. Calculations of new cases should start inthis mode.


Vista TF


• Using a value of 1 for the last digit, enables the choke massflow limitation calculation. This requires exact data for thethroat areas to be specified and should only be used if thisis available. It is necessary to set this value to 1 when usingiteration to pressure ratio in choked stages.

• The second-last digit controls the blending function calcu-lation for the deviation between the blade angle and theflow angle as follows:

0 - Turbines use departure angle at the leading andtrailing edge ends, Compressors use swirl at the lead-ing edge and departure angle at the trailing edge.

1 - Turbines and compressors use the departure anglefor the leading and trailing edges.

2 - Turbines and compressors use departure angle atthe trailing edge and relative swirl at the leading edge.

3 - Turbines and compressors use departure angle atthe trailing edge and the absolute swirl at the inlet.

4 - Compressors use swirl at the leading edge anddeparture angle at the trailing edge; turbines use swirlat the outlet and departure angle at the leading edge.

5 - Compressors use swirl at the leading edge, depar-ture angle at the trailing edge; turbines use swirl atthe trailing edge and departure angle at the leadingedge.

Determines the input definition for the reference flowparameters (which are input in .aer file).

i_flow

Notes:

• i_flow = 0 then ref_mach , ref_phi , and ref_d , arespecified, but if ref_mach > 3 then it is interpreted asref_u , so this is equivalent to ref_u , ref_phi , andref_d .

• i_flow = 1 then ref_n , ref_mass , and ref_d arespecified.

Note: If there is more than one spool in the calculationwith different rotational speeds, then this is taken intoaccount as follows:

- If i_spool = 2 and i_flow = 1 then ref_n1 ,ref_n2 , ref_mass , and ref_d are specified.





- If i_spool = 3 and i_flow = 1 then ref_n1 ,ref_n2 , ref_n3 , ref_mass , and ref_d are spe-cified.

• i_flow = 2 then ref_n , ref_volume , and ref_d arespecified.

• i_flow = 3 then ref_u , ref_mass , and ref_d arespecified.

• i_flow = 4 then ref_u , ref_volume , and ref_d arespecified

• i_flow = 5 then ref_n , ref_mass , and ref_d arespecified together with the ref_pr (total to static pressureratio between inlet plane and the last trailing edge on themid-streamline). The value of ref_mass is a start valuefor mass flow in the iteration to pressure ratio and has noeffect on the final solution.

• i_flow = 6 then ref_n , ref_mass , and ref_d arespecified together with ref_pr (total to static pressureratio) together with n_p_te (the total number of trailingedges at which a guessed value of the static pressure ratiois specified, followed by the guessed pressure ratios ateach trailing edge, including the last, which is also definedby ref_pr .

• i_flow = 7 then ref_n , ref_mass , and ref_d arespecified together with the ref_p (static pressure at thelast trailing edge on the mid-streamline). The value ofref_mass is a start value for mass flow in the iterationto outlet pressure and has no effect on the final solution.This option may be useful for low speed devices wherepressure ratio becomes indeterminate.

• i_flow = 8 then ref_n , ref_mass , and ref_d arespecified together with the ref_p (static pressure attrailing edge pane) together with n_p_te (the totalnumber of trailing edges at which a guessed value of thestatic pressure is specified, followed by the guessed pres-sures at each trailing edge, including the last, which is alsodefined by ref_p .

• i_flow = 9 then ref_n , ref_mass , and ref_d arespecified together with the ref_cu (absolute swirl velocityat the last trailing edge on the mid-streamline). The valueof ref_mass is a start value for mass flow in the iterationto outlet swirl and has no effect on the final solution. Thisoption may be useful for turbine calculations where thelast blade row is a turbine rotor.


Vista TF


• Other options are available for debugging purposes butare not described here.

• The geometry definition of Vista TF assumes clockwise ro-tation. This leads to a certain convention for the sign ofthe blade angles (see Appendix A: A Note on Sign Conven-tion for Angles and Velocities in Vista TF (p. 284)). In somecases you may have a counterclockwise machine with bladeangles of the opposite sign. To avoid the need to changeall the angles specified in the .geo file, an option isprovided whereby the value of i_flow is given a negativesign. As a more sophisticated alternative the blade speedmay be defined as negative.

Determines the model for the equation of state of thefluid:

i_fluid

• If i_fluid = 0 then ideal gas with constant specific heats.

• If i_fluid = 1 then liquid.

• If i_fluid = 2 then a real gas calculated with theAungier-Redlich-Kwong equations.

• If i_fluid = 3 then real gas calculated with the Redlich-Kwong equations.

• If i_fluid = 4 then an ideal gas calculation with variablespecific heats is carried out. This is done by using the realgas equations but setting the coefficients of the Aungier-Redlich-Kwong equations to the appropriate values to re-produce an ideal gas equation. Steam properties can beapproximated with the Aungier-Redlich-Kwong equationsusing the values of the coefficients to model steam.

Determines the type of inlet boundary conditions:i_inbc

• i_inbc = 0 then input values are total pressure, totaltemperature and swirl (r x cu , that is radius times circum-ferential component of the absolute velocity) on the inputplane.

• i_in_bc = 1 then input values are total pressure, totaltemperature, and absolute flow angle.

See also the section on n_inbc in the aerodynamicsfile (described in Specification of Aerodynamic Data File(*.aer) (p. 245)) where the inlet profile across the spancan be specified.

Determines whether the mass flow is uniformly distrib-uted across streamlines or not.

i_mass





• i_mass = 0 then mass flow between each streamline isthe same.

Determines which mixing model is used:i_mix

• i_mix = 0 then no mixing model.

• i_mix = 1 then a spanwise mixing model based on eddydiffusion across the streamlines will be used. Note that thisrequires a minimum of 9 streamlines (n_sl => 9).

Determines the form of the radial equilibrium equationthat is used to determine the velocity gradient alongthe quasi-orthogonal.

i_ree

• i_ree = 0 then the equations as given in the paper ofCasey and Roth (1984) are used, except that the dissipationterm is set to zero and the blade force term is set to zeroat a trailing edge and at a leading edge.

• i_ree = 1 then the solution is as for i_ree = 0 but thedissipation term is not set to zero but the equations asgiven in the theory documentation are used.

• i_ree = 2 then the dissipation term is not set to zero butthe equations given in the paper of Casey and Roth (1984)are used.

• i_ree = 3 then the velocity gradient in the radial equilib-rium equation is reduced by the factor grad_ree givenin section 5. This is useful for debugging difficult cases andthe simulation becomes similar to a mean-line calculationwith no gradient of meridional velocity across the span. Ifgrad_ree = 1.0 then selecting i_ree = 3 has no effect.

Section 5: Convergence and damping factors

The syntax is:

damp_sc damp_vl cm_start tolerance_cm tolerance_mass grad_ree

The damping factor model automatically chooses the most appropriate values of these parametersbased on the type of turbomachine and the grid. You then typically need to specify the followingvalues for this section:

0.00 0.00 0.00 0.00 0.00 1.00

In some rare cases it may still be necessary to select these values differently.


Damping factor for streamline curvature terms.damp_sc

Note:


Vista TF


• The Wilkinson stability analysis for streamline curvatureprograms indicates that the streamline curvature dampingterm has to be reduced for long closely spaced quasi-or-thogonals (high aspect ratio). For details, see the sectionon computational grid (Computational Grid (p. 306)).

• If damp_sc = 0.0 then the program determines the valueof the damping factor from the theory of Wilkinson, or usesan internally determined value that differs for each typeof turbomachine — whichever is smaller.

• Typically values of damp_sc between 0.05 and 0.25 areused, but lower values may be necessary for high aspectratio quasi-orthogonals (as in end stages of steam turbines).

• If the specified value of damp_sc is larger than the valuepredicted by the Wilkinson stability theory then the pro-gram automatically reduces the damping factor to a stablevalue.

• If the program has convergence problems with an increas-ing error then the value is automatically reduced internallywithin the program during the convergence process.

• Several different schemes for the damping are applied. Theoriginal scheme is obtained with a value of damp_scbetween 1.0 and 1.25. New schemes which are more stableand robust in most cases can be obtained with the valueof damp_sc between 0.0 and 0.25. A value between or2.0 and 2.25 uses the original scheme with changed con-stants. The first digit then defines which scheme is used(0 - original, 1 - modified original, 2 - new scheme) and thedigits after the decimal point are the damping factor itself.

Damping factor on velocity level. Note this dampingfactor is also used internally in the program for all para-meters which are under-relaxed.

damp_vl

Notes:

• There is no stability theory to define this, and a typicalvalue used is 0.50, indicating that 50% of the new paramet-er together with 50% of its original value is used.

• If damp_vl = 0.0 then 0.50 is used.

Value of meridional velocity on the mean streamline asa fraction of u_ref , as used in the initial conditions.

cm_start

This is then a sort of flow coefficient (cx /u) of the deviceconcerned and is used as a guide to the velocity levelsthat can be expected.

Notes:





• Recommended that this should be less than the actualvalue when converged because this avoids choking duringthe early streamline curvature iterations.

• If simulations from a cold-start (with no restart file) fail toconverge, it may be useful to modify this parameter be-cause it strongly influences the start velocities.

• Not used if the restart initial condition is used.

• If cm_start = 0.0 then 0.25 is used for all calculationsexcept radial turbines where 0.1 is used. The value of 0.25is probably adequate for radial compressors only.

Tolerance level on change in meridional velocity duringstreamline curvature iterations. The value specified is

toler-ance_cm

the maximum percentage change in meridional velocityfor convergence. Iterations stop when all streamlinesand all quasi-orthogonals have a lower value than this.

Notes:

• Typical value 0.01 (that is, 0.01%, which is 1 part in 10,000).

• If tolerance_cm = 0.0 then 0.01% is used.

• Note that if the meridional velocity is low at a certain pointin the flow field, it may be necessary to use a higher valuethan this.

• Note that the extremely low value of 0.01% does not implythat the solution is as accurate as this, but just providesconfidence that convergence has really been achieved.

Tolerance level on mass flow for internal mass flow iter-ation. Note that because this controls the convergence

toler-ance_mass

of the innermost loop, it should be a factor of 2 to 10lower than the tolerance value for the meridional velocity(above).

Notes:

• Typical value 0.001 (that is 0.001%, which is 1 part in100000).

• If error_max = 0.0 then 0.001% is used.

Factor to reduce the spanwise velocity gradient fromthe radial equilibrium equation. Normally equal to 1.0

grad_ree

indicating that the gradient from the radial equilibriumequation is used without change. For calculations withi_ree = 3, if grad_ree is set to 0.0, the program takesa meridional velocity gradient of 0.0 (that is constantmeridional velocity across the span) and a value between1.0 and 0.0 reduces the spanwise gradient of meridional


Vista TF


velocity determined by the radial equilibrium equationby this amount.

12.2.2.5. Specification of the Geometry Data File (*.geo)

The geometry data file includes sections of text lines that help you to identify the parameters definedhere. You should read the section on geometry in Appendix A: A Note on Sign Convention for Anglesand Velocities in Vista TF (p. 284) to become familiar with sign conventions and angle definitions usedin this file. An example of a geometry data file is given in Appendix C: Example of a Geometry Data File(*.geo) for a Radial Impeller (p. 292).

Section 1: Character strings identifying the geometry data (max 72 characters/line)

The syntax is:


Section 2: Number of quasi-orthogonal lines and scale factor (one line)

The syntax is:

n_qo scale


n_qo = number of quasi-orthogonal lines from inlet tooutlet of the domain. From version V1.31 onwards themaximum value of this parameter is unlimited.

n_qo

The scaling factor for all geometry data that is input.scale

Notes:

• Usually the geometry data is input in SI units (that is, allvalues are expected in m and not mm) and then this valueis 1.0.

• If the input geometry data comes from a CAD system thenit may be in mm. In this case, the value of scale must be0.001. Similarly the value can be adjusted to allow thegeometry data to be input in other systems of units; forexample, inches.

• This scale factor only scales the data in the geometry inputfile and has no effect on other dimensions elsewhere; forexample, it does not scale the reference diameter, whichmust be input with units of metres, in the aerodynamicfile.

• If you specify scale = 0.0, unity is used.

For each quasi-orthogonal, the following data is required to define the flow channel for the meridi-onal through-flow calculation and the meridional spacing of the quasi-orthogonals. Note that someof this data is also repeated in the section on the blade geometry. This duplication allows calculations




to be made in a channel that is not the same as the hub and casing line of the actual blade definition(blade cropping).

Section 3: Definition of quasi-orthogonal types and end points (n_qo lines: i = 1 to n_qo)

The syntax (of a single line) is:

i r_hub(i) r_shr(i) z_hub(i) z_shr(i) n_blade(i) n_curve(i) i_type(i) i_row(i) i_spool(i)


Number of a particular quasi-orthogonal line.i

Notes:

• The actual value is not used by the program internally be-cause it recounts the quasi-orthogonals as they are input.The number specified is only used as a guide to the loca-tion in the input file.

• This allows you to merge two different geometry files of,say, a rotor and stator, to a single stage geometry filewithout the need to renumber the quasi-orthogonals. Inthis case, the same number may appear more than once.In a similar way, a single quasi-orthogonal may be removedwithout the need for renumbering the lines.

Radial coordinate at hub end of quasi-orthogonal [m].r_hub(i)

Radial coordinate at casing end of quasi-orthogonal [m].r_shr(i)

Axial co-ordinate at hub end of quasi-orthogonal [m].z_hub(i)

Axial co-ordinate at casing end of quasi-orthogonal [m].z_shr(i)

Notes on co-ordinates:

• r_hub may not be close to zero; adapt the grid if necessaryto avoid small values of r_hub .

• The aspect ratio of the quasi-orthogonal lines determinesthe stability of the solver. The lines should not be tooclosely spaced.

• The end coordinates of the leading and trailing edges ofblade rows should be included in the list of coordinates.

• The end coordinates of the leading edge of a splitter vaneshould be included in the list of coordinates.

Number of blades in blade row.n_blade(i)

Notes:

• = 0 in duct regions.

• The number of blades changes at a splitter blade leadingedge in a compressor or at a splitter vane trailing edge ina turbine, and changes again for multiple splitters. This isthe only information that the program has about the


Vista TF


splitters, so the location of the splitter vane leading ortrailing edge needs to be a quasi-orthogonal line in theinput data.

Number of defining points along the i th quasi-orthogon-al line.

n_curve(i)

Notes:

• = 1, a special case for duct stations. This indicates thatthere are 2 defining points (as for n_curve(i) = 2) butthat no further information for this calculating station isprovided in section 4 below, because it is already fullydefined by the hub and casing points given in section 3.

• = 2 for a linear calculating plane in which only the endpoints of the quasi-orthogonal are defined. In this case,similar information can be found in section 4.

• > 2 for a non-linear or curved calculating plane.

• The number of defining points can vary from station tostation but n_curve(i) should typically be the same forall stations because streamline section data is usuallyavailable on a fixed number of spanwise sections.

• Note that the hub and casing geometry information doesnot necessarily have to be the same as that defined insection 4. In this case the data in section 3 will be used tocrop the blade row or to make a section through the bladeinformation in section 4.

Parameter to identify type of calculating station:i_type(i)

1 - for duct region

2 - for non-rotating blade row (stator)

3 - for rotating blade row (rotor)

Notes:

• The program internally identifies which lines are leadingand trailing edges from the changes of type of blade row,and which line is the leading edge of a splitter vane (bythe change in blade number).

• There must be at least two duct calculating stations up-stream of the first blade row, and downstream of the lastblade row.

• A blade row must consist of at least two calculating stations(leading and trailing edge). Typically a radial impeller willhave around 15 calculating stations, because this gives a





6° turn between each station and improves the calculationof the curvature terms.

• There must be at least two blade calculating stations up-stream and downstream of a splitter vane leading edge.

• Other types of blade row may be defined at a later stage.

This parameter is used to identify type of blade row andstage of the quasi-orthogonal calculating station. In fact

i_row(i)

the program can usually identify the type of blade rowitself from the geometry and the context, so it is notnecessary to specify these values at all and, in the firstinstance, this parameter may be set to zero. They areincluded here for special cases where the program mayhave difficulty with the rules that are coded to identifyblade row types.

= n11 - radial compressor inlet guide vane1

= n12 - radial compressor inlet guide vane2

= n13 - radial compressor impeller blade

= n14 - radial compressor diffuser vane

= n15 - radial compressor return channel vane

= n16 - radial compressor axial de-swirl vane

= n21 - axial compressor inlet guide vane1

= n22 - axial compressor inlet guide vane2

= n23 - axial compressor rotor blade

= n24 - axial compressor stator vane

= n25 - axial compressor outlet guide vane

= n31 - radial turbine inlet guide vane1

= n32 - radial turbine inlet guide vane2

= n33 - radial turbine impeller blade

= n34 - radial turbine stator vane

= n35 - radial turbine outlet guide vane

= n41 - axial turbine inlet guide vane1

= n42 - axial turbine inlet guide vane2

= n43 - axial turbine rotor blade


Vista TF


= n44 - axial turbine stator vane,

= n45 - axial turbine outlet guide vane

Notes:

The value of n determines which stage is being con-sidered, such that a multistage axial compressor with anIGV would begin with values of 121 for the inlet guidevane, continue with values of 123 for the rotor, and 124for the downstream stator, so that the next rotor wouldbe 223, and so on. A double row of stators would bedenoted as n24 and n25 for the successive blade rows.

Parameter to identify rotational speed of spool or shaftswhere different blade rows have different speeds.

i_spool

0 single shaft with one speed (ref_n )

1 (or 0) first shaft with speed (ref_n1 )

2 second spool with second speed (ref_n2 )

3 third spool with third speed (ref_n3 )

Note that counter-rotating blade rows can be dealt withby specifying negative speeds for the second spool. Theprogram determines the number of different spools(n_spool ) from the number of different values ofi_spool(i) found in the geometry input file. (Themaximum is set to be 3.)

Notes:

• The data is supplied on n_curve lines spaced fairly evenly from hub to casing. For example, ifn_curve = 5, there may be points at 0%, 25%, 50%, 75%, and 100% of span. If n_curve = 2, therewill be just two points, at 0% and 100% span.

• The blade data and quasi-orthogonal data in section 4 may extend outside of the flow channel definedby the meridional coordinates given in section 3. In a normal calculation, the data overlaps partly withthat given in section 3, because the end points of the quasi-orthogonal lines are defined twice (exceptfor duct calculating stations with n_curve(i) = 1; see above). The flow channel defined in section3 is generally congruent with the blade hub and shroud defined in section 4. The blade, as definedin section 4, may extend outside of the region of the flow channel because this allows a calculationto be made in a cropped or trimmed flow channel only by changing the data in section 3; section 4does not need to be changed.

• The end points of the quasi-orthogonal lines, as defined by coordinates r_hub(i) , r_shr(i) ,z_hub(i) , and z_shr(i) , should lie along the quasi-orthogonal lines as defined by r_qo (j,i)and z_qo(j,i) below. In many cases, the end points will be coincident with the blade data, but ifthis is not the case, it is not acceptable to define end points that do not lie on the blade data point.




• All the angles are specified in degrees because this is more convenient in those cases where it maybe necessary to define the geometry by hand, and also allows a simple consistency check, but internallythe program converts them into radians. Further description of the angles is provided in Appendix A:A Note on Sign Convention for Angles and Velocities in Vista TF (p. 284).

The sections above define only the hub and casing walls and provide information on the type ofcalculating station. The detailed orientation and position of the curved quasi-orthogonal line andthe details of the blade surface geometry are provided in the next section.

Section 4: Geometry of quasi-orthogonal line and blade

The syntax (of the single line) is:

i j r-qo(j,i) theta_qo(j,i) z_qo(j,i) thu_qo(j,i) gamma_r_qo(j,i) gamma_z_qo(j,i)


Number of a particular quasi-orthogonal for data input(increasing from inlet to outlet). Note that this value is

i

not read in as input data but is simply used as orienta-tion in the data file when examining the geometry inputdata.

Number of streamline for data input (increasing fromhub to shroud). Note that this value is not read in as in-

j

put data but is simply used as orientation in the datafile when examining the geometry input data. Note thatfor each q-o (i) , the spanwise data is entered (j)before continuing with the next q-o .

Radial coordinate of point j along q-o (i) [m].r_qo(j,i)

Circumferential coordinate of blade camber line at point(r_qo , z_qo ) [degrees].

theta_qo(j,i)

Note:

• The angular coordinate (theta) is taken as positive in thedirection of rotation and negative in the other direction.

• In a duct region this angle may be zero.

• This angle is not used by the program but helps to visualizethe blade shape and may be useful for plots of the bladeshape.

• In a region where there is a splitter vane this angle is theblade camber angle of the main blade and not of thesplitter.

Axial coordinate of point along quasi-orthogonal [m].z_qo(j,i)

Circumferential thickness of blade at point (r_qo , z_qo )[m].

thu_qo(j,i)

Note:

• In a duct region this thickness should be specified as zero.


Vista TF


• In a region where there is a splitter vane this thickness isthe mean thickness of the main blade and the splitter.

• At leading and trailing edges the value supplied is not usedby the program, but the calculating station is taken to beat the limit of the chord with zero thickness.

• The thickness is not the thickness normal to the camberline.

Lean angle of the blade with a radial line [degrees] asdefined in Appendix A: A Note on Sign Convention forAngles and Velocities in Vista TF (p. 284).

gamma_r_qo(j,i)

Lean angle of the blade with an axial line [degrees] asdefined in Appendix A: A Note on Sign Convention forAngles and Velocities in Vista TF (p. 284).

gamma_z_qo(j,i)

Notes:

• This geometry includes the coordinates of the blade camber surface (r-qo(j,i) , theta_qo(j,i) ,z_qo(j,i) ) so that it would theoretically be possible for the program to differentiate this informationto determine the slope angles of the surface (gamma_r_qo(j,i) , gamma_z_qo(j,i) ). This is notdone for two reasons. Firstly, experience shows that with the crude grids typically used for throughflowcalculations, this differentiation would be an unwanted source of error leading to poor estimates ofthe blade angles, so a system was chosen in which the blade angles are supplied. In fact the value oftheta_qo(j,i) is not used by the program and can be specified as zero. Secondly, in many casesthe slope angles of a blade row are known (inlet and outlet angles) whereas the circumferential co-ordinate is unknown.

• This system of geometry with the definition of two angles is designed for radial turbomachinery ap-plications because it allows the complex shape of three-dimensional blades to be defined by the useof the two lean angles, gamma_r and gamma_z.

• In conventional axial turbomachinery throughflow programs, it is not usual to define the blade inmuch detail because often simulations are carried out with only inlet and outlet blade angles. This isalso possible with Vista TF. In a ductflow calculation with only leading and trailing edges, the valueof gamma_r can be set to 0° . Because the leading and trailing edges are not considered to have anyblade force, this has no effect on the simulation. The value of gamma_z defines the inlet and outletblade angles of sections through the blade at constant radius.

• Vista TF assumes that the geometry is always specified for a clockwise rotation (see Appendix A: ANote on Sign Convention for Angles and Velocities in Vista TF (p. 284)). If the geometry is correctlyspecified and a negative rotational speed is used, and Vista TF performs calculations assuming thatthe shaft is rotating in the wrong direction with all the wrong incidences, loading, and so forth. Forthe case where the geometry is specified for counterclockwise rotation, the value of i_flow in the.con file should be specified as a negative number, for example, -1 or -2. This causes the programto internally switch the angles from positive to negative, and vice-versa, before it proceeds to performcalculations as usual, assuming that the shaft rotates clockwise. A machine that has a geometry suitedfor anti-clockwise rotation should have a negative value for the rotational speed in the .aer file.

• The next sections provide additional blade-row geometry data for use in the correlations in the program.Although this data is not always needed, the format of this section includes 8 real parameters.




Section 5: Additional geometry data (n_curve(i) lines for each blade row)

The syntax is:

j throat throat_pos clearance max_thickness te_thickness dummy dummy dummy


Number of a particular streamline for data input (increas-ing from hub to shroud).

j

Throat width of section j [m].throat

Note:

• If the value of zero is input then the program estimatesthe throat width from the blade angles and the bladethickness, taking into account that the throat is close tothe leading edge for compressors and close to the trailingedge for turbines. For radial compressors this estimate isnot particularly accurate and for cases close to choke youshould provide more precise data here.

• For axial compressors it is assumed that the blade has acircular arc camber line and the program includes an estim-ate of the throat area based on the geometrical relation-ships for circular arc blades.

• For other blade types, you should specify the value.

Position of throat on this blade section.throat_pos(i)

Note:

• If the value of zero is input then the program estimatesthe throat position from the blade angles and the bladethickness, taking into account that it is close to the leadingedge for compressors and close to the trailing edge forturbines.

• For radial compressors this estimate is not particularly ac-curate and for cases close to choke you should providemore precise data here.

• For axial compressors it is assumed that the blade has acircular arc camber line and the program includes an estim-ate of the throat position based on the geometrical rela-tionships for circular arc blades.

• For other blade types, you should specify the value.

Tip clearance of hub section [m].tip clear-ance

Tip clearance of shroud section [m].hub/shroud

Note:

• If the first value is non-zero, this is interpreted as the hubclearance.


Vista TF


• If the last value (n_curve(i) ) is non-zero, this is inter-preted as the casing clearance.

• In a variable stator vane with a hub clearance and a clear-ance gap for the shaft, it is possible to have non-zero valueson both the hub and the tip and then both will be takeninto account in the loss correlations.

Maximum normal thickness of blade section [m].max_thick-ness

Note:

• Although the circumferential thickness of the blade row isalready specified in section 3, the location of the quasi-or-thogonal lines may not coincide with the location of normalmaximum thickness, so this value must be specified separ-ately.

• If the value of zero is specified then the program searchesfor the normal maximum thickness of the blade, as givenin section 3, and uses it.

Trailing edge normal thickness of blade section [m].te_thick-ness

Note:

• Although the circumferential thickness of the blade row isalready specified in section 3, this value must be specifiedseparately.

• If the value of zero is specified then the program estimatesthe trailing-edge thickness of the blade from the circumfer-ential thickness given in section 3, and uses it.

In order to allow for parameters that may be needed byother correlations in the future, the program includes

dummy

three dummy parameters which should each be set to0.0.

12.2.2.6. Specification of Aerodynamic Data File (*.aer)

The aerodynamic input data file includes lines of text that help you to identify the parameters definedhere. Although in some areas this file may appear to be complex, a typical simulation uses only one ofthe allowed options and so, once an input file with the correct format has been established, further useof the file is less complex.

An example of an aerodynamic data file is given in Appendix D: Example of an Aerodynamic Data File(*.aer) (p. 294).

Section 1: Character strings identifying the aerodynamic data (max 72 characters/line)

The syntax is:

Character string – title(1)Character string – title(2)Character string – title(3)




Many options are supplied for specifying the flow data, but typically the options with i_flow = 1(specified mass flow) and i_flow = 5 (specified pressure ratio) are used. Note that i_flow isspecified in the control file.

Section 2: Reference aerodynamic parameters (depends on value of i_flow in .con file)

The syntax is:

ref_mach ref_phi ref_d

if i_flow = 0 and ref_mach < 3, or

ref_u ref_phi ref_d

if i_flow = 0 and ref_u > 3, or

ref_n1 ref_mass ref_d

if i_flow = 1 and n_spool = 0 or 1, or

ref_n1 ref_n2 ref_mass ref_d

if i_flow = 1 and n_spool = 2, or

ref_n1 ref_n2 ref_n3 ref_mass ref_d

if i_flow = 1 and n_spool = 3, or

ref_n1 ref_volume ref_d

if i_flow = 2, or

ref_u ref_mass ref_d

if i_flow = 3, or

ref_u ref_volume ref_d

if i_flow = 4, or

ref_n1 ref_mass ref_d ref_pr

if i_flow = 5, or

ref_n1 ref_mass ref_d ref_pr n_p_te guess_pr_1 guess_pr_2...

if i_flow = 6, or

ref_n1 ref_mass ref_d ref_pd

if i_flow = 7, or

ref_n1 ref_mass ref_d ref_pd n_p_te guess_pd_1 ...

if i_flow = 8, or


Vista TF

ref_n1 ref_mass ref_d ref_cu

if i_flow = 9.

Note

The very large number of possible ways of specifying the flow and speed appears, atfirst, to be slightly overwhelming. Generally i_flow = 1 is used. Note that the referencediameter is given here and not in the geometry file. This is because you may prefer touse the hub diameter, the tip diameter, the inlet diameter, or the outlet diameter, as areference value without changing the geometry file.

Note

The geometry definition of Vista TF assumes clockwise rotation. This leads to a certainconvention for the sign of the blade angles (see Appendix A: A Note on Sign Conventionfor Angles and Velocities in Vista TF (p. 284)). Vista TF expects that the geometry is definedfor a clockwise-rotating machine (viewed from the inlet along the positive axis). In somecases you may have a counterclockwise-rotating machine with blade angles of the op-posite sign to that expected by Vista TF. To avoid the need to change all the anglesspecified in the .geo file, an option is provided whereby the value of i_flow is givena negative sign. As an alternative, the rotational speed may be specified as a negativevalue, which means that the blade is rotating in the counterclockwise direction.


“Iteration to mass flow”i_flow =0

ref_mach = Machine Mach number (based on inlettotal conditions) [-] or reference blade speed [m/sec]. (Inthis document, “[-]” means dimensionless.)

Notes:

• If ref_mach < 3 then ref_mach is interpreted asref_mach .

• If ref_mach > 3 then ref_mach is interpreted as ref_u .

ref_phi = Inlet flow coefficient (based on total inletconditions) [-]. (In this document, “[-]” means dimension-less.)





ref_d = Reference blade diameter for the definition of

flow coefficient. [m]


ref_n1 = Machine rotational speed [rpm]

ref_mass = Mass flow [kg/sec]



If the machine has separate spools of different speeds(i_spool(i) > 2) then the speed of each spool canbe provided up to a maximum of three different spools.

ref_n1 = Rotational speed of shaft 1 [rpm]



Counter-rotating blade rows require the second bladerow to be provided with a negative speed.



ref_volume = Volume flow at inlet total conditions[m3/sec]




ref_u = Reference blade speed [m/s]

ref_mass = Mass flow [kg/sec]

ref_d = Reference blade diameter for the definition ofthe blade speed [m]


ref_u = Reference blade speed [rpm]

ref_volume = Volume flow at inlet total conditions[m3/sec]


Vista TF


ref_d = Reference blade diameter for the definition ofthe blade speed [m]

“Iteration to pressure ratio”i_flow =5


ref_mass = Mass flow [kg/sec] (estimate of actual massflow but final converged mass flow is determined by thepressure ratio and this only serves as an initial guess)



ref_pr = Ratio of static pressure at the trailing edgeto total pressure at the inlet, with both of these pressuresevaluated on the mean streamline.

“Iteration to pressure ratio”i_flow =6


ref_mass = Mass flow [kg/sec] (estimate of actual massflow)



ref_pr = Ratio of static pressure at the trailing edgeto total pressure at the inlet, with both of these pressuresevaluated on the mean streamline.

n_p_te = Number of trailing edges at which a guessedvalue of the pressure ratio is specified

guess_pr_1 = Guessed value of the pressure ratio atthe first trailing edge

guess_pr_2 = Guessed value of the pressure ratio atthe second trailing edge

...

guess_pr_n_p_te = This continues up to and includ-ing the last trailing edge.

Note that if the pressure ratio of the last trailing edgediffers to that of ref_pr , than all values at all trailingedges are scaled with the value of ref_pr .

“Iteration to pressure difference”i_flow =7






ref_mass = Mass flow [kg/sec] (estimate of actual massflow but final converged mass flow is determined by thepressure ratio and this only serves as an initial guess)



ref_pd = Static pressure at the trailing edge minustotal pressure at the inlet, with both of these pressuresevaluated on the mean streamline.

“Iteration to pressure difference”i_flow =8



ref_d = reference blade diameter for the definition of

flow coefficient [m]

ref_pd = Static pressure at the trailing edge minustotal pressure at the inlet, with both of these pressuresevaluated on the mean streamline.

n_p_te = Number of trailing edges at which a guessedvalue of the pressure ratio is specified

guess_pd_1 = Guessed value of the pressure differenceat the first trailing edge

guess_pd _2 = Guessed value of the pressure differ-ence at the second trailing edge

...

guess_pd_n_p_te = This continues up to and includ-ing the last trailing edge

“Iteration to outlet swirl”i_flow =9



ref_d = reference blade diameter for the definition of


ref_cu = Swirl velocity on mean streamline at rotoroutlet


Vista TF

Note

• The reference blade speed, tip speed Mach number and/or the machine rotational speedare also needed for the calculation of a stator blade row. This is because these parametersare used to define various non-dimensional flow and work coefficients and the referenceblade speed is also used to determine the flow velocities for the initial estimate of the flowfield (together with parameter cm_start . See section 5 of the .con file specificationgiven in Specification of the Control Data File (*.con) (p. 226)).

• Iteration to a defined pressure ratio makes use of the so-called target pressure ratio methodof Denton. This requires the program to make a fist guess of the pressure at each trailingedge of the machine. The algorithm currently incorporated makes a crude estimate of these,but it has been found that this may not be sufficient to secure convergence. For this reason,you can define the first guess of the pressure at each trailing edge by setting i_flow =6 (instead of 5).

• A line prepared with data for i_flow = 6 can formally also be used with i_flow = 5 ori_flow =1 with no change, so that a calculation can switch from “iteration to pressureratio” to “iteration to mass flow” with no formal change to the aerodynamic input data file.

Section 3: Reynolds number or viscosity

The syntax is:

ref_re

or

ref_mue


Reynolds number based on ref_u ( ), ref_D

( ), and inlet total conditions [-]. (In this document,“[-]” means dimensionless.)

ref_re

Dynamic viscosity at the inlet plane and mean inlettotal conditions.

ref_mue

Note

The program identifies which of these parameters has been provided from the absolutevalue of the numerical input. If the value is greater than 1.0 [N s m^-2] (or equivalentvalue in other units), it is interpreted as a Reynolds number; if it is less than 1.0 [N s m^-




2] but greater than 0.0000001 [N s m^-2], it is interpreted as the dynamic viscosity. Avalue of 0 causes the program to determine the dynamic viscosity from an inbuilt equationfor the dynamic viscosity based on Sutherland’s law and the inlet total temperature.

Section 4: Fluid data (depends on value of i_fluid in .con file)

The syntax is:

cp_gas gamma_gas

if i_fluid = 0, which indicates an ideal gas with constant specific heats, or

cw_fluid rho_fluid

if i_fluid = 1, which indicates a liquid, or

R_gas gamma_gas

if i_fluid = 2, 3 or 4, which indicate a real gas.


cp_gas = Specific heat at constant pressure [J/kgK]i_fluid = 0

(ideal gaswith constantspecificheats)

gamma_gas = ratio of specific heats [-] (In this docu-ment, “[-]” means dimensionless.)

cw_fluid = Specific heat of fluid [J/kgK]i_fluid = 1

(liquid)

rho_fluid = density of liquid [kg/m3]

Real gas option in which full details of the gas propertiesare provided in the .rgp file. In order to avoid changing

i_fluid = 2

i_fluid = 3 the format of this file, these values also need to be spe-

i_fluid = 4cified here. The values given here are specified as addi-tional data but are then overwritten internally in the

(real gas)program by the values in the real gas data file (see Spe-cification of the Real Gas Properties Data File(*.rgp) (p. 266)).

R_gas = Gas constant [J/kgK]

gamma_gas = mean value of ratio of specific heats [-]

=


Vista TF

Section 5: Number of points on the inlet boundary where flow conditions are specified

The syntax is:

n_inbc


Number of points at which the inlet flow conditions arespecified across the inlet plane.

n_inbc

Note:

• n_inbc < n_sl

• if n_inbc = 1 then only a single value is input and thevalues are constant across the span. Note that a singlevalue implies that the inlet conditions have a constant totalpressure, constant total temperature, and constant swirl(rcu) across the span at the inlet.

Section 6: Fraction of mass flow where inlet conditions are specified (n_inbc values)

The syntax is:

f_mass_inbc


n_inbc values of the fraction of mass along the inletboundary at which inlet boundary conditions are spe-

f_mass_in-bc

cified. The first value should be 0.0 and the last valueshould be 1.0. Note that if n_inbc = 1 then this has nofunction and a dummy value can be specified, but thisshould not be omitted.

Section 7: Pressure on the inlet boundary (n_inbc values which depend on i_inbc in .con file)

The syntax is:

pt_inbc

if i_inbc = 0.


Total pressure on the inlet boundary [Pa].pt_inbc

Note that if an incompressible calculation is carried outit is still necessary to specify the absolute value of thetotal pressure on the inlet boundary.

(i_inbc =0)

Section 8: Temperature on the inlet boundary (n_inbc values, depend on i_inbc in .con file)

The syntax is:




tt_inbc

if i_inbc = 0.


Total temperature on the inlet boundary [K].tt_inbc

(i_inbc = 0)

Section 9: Swirl or angle on the inlet boundary (n_inbc values, depend on i_inbc in .con file)

The syntax is:

rcu_inbc

if i_inbc = 0, or

alpha_inbc

if i_inbc = 1.


Swirl on the inlet boundary [m2/sec]. Note that this isthe product of the local radius of the streamline and the

rcu_inbc

(i_inbc = 0) local circumferential velocity component and is positiveif the swirl is in the clockwise direction of rotation.

Flow angle on the inlet boundary [°]. Note that this isfrom the axial direction and positive in the clockwisedirection of rotation.

alpha_in-bc

(i_inbc = 1) Note also that, if a single value is specified, it is used to cal-culate the swirl on the mean streamline of the inlet boundary,and the swirl is then kept constant across the span. If a con-stant flow angle across the span is required then 2 valuesneed to be specified across the span (n_inbc = 2). Experi-ence with radial turbines with high swirl at the inlet showthat the specification of a single value of swirl across the spanis more robust than specifying a variation of flow angle acrossthe span.

Section 10 : Aerodynamic model parameters

The syntax is:

eddy f_bl_le f_bl_te


Spanwise mixing parameter (eddy diffusivity).eddy

Notes:

• Typically 0.0001 to 0.001. See Streamline CurvatureThroughflow Theory (p. 275).


Vista TF


• This has no function if i_mix = 0.

• If i_mix = 1 and eddy = 0.0 then a value of eddy = 0.0005will be used.

f_bl_te is the meridional fraction of blade length atthe leading edge where the blade is partly transparent

f_bl_le

to the flow (accounts artificially for the increased loadingdue to incidence). It is recommended that you specifya value of 0.0 for both f_bl_te and f_bl_le . If bothf_bl_te and f_bl_le are equal to zero then theprogram estimates the values of these from the bladespacing using the equations given in StreamlineCurvature Throughflow Theory (p. 275). If you want, youcan specify these values. If you specify 0.0 then the flowis congruent with the mean blade camber line at theleading edge. A typically value is roughly equal to theblade spacing.

Fraction of blade length after which the blade is partlytransparent to the flow at the trailing edge (accounts

f_bl_te

artificially for the decreased loading due to deviation asthe trailing edge is approached). It is recommended thatyou specify a value of 0.0 for both f_bl_te andf_bl_le . If you specify 1.0 then the flow is congruentwith the mean blade camber line right up to the trailingedge. If both f_bl_te and f_bl_le are equal to zerothen the program estimates these from the blade spacingusing the equations given in Streamline CurvatureThroughflow Theory (p. 275).

12.2.2.7. Specification of Correlations Data File (*.cor)

The input data file includes sections of text lines that help you to identify the parameters defined here;see the examples in Appendix E: Examples of Correlations Data Files (*.cor) (p. 295). The first part of thefile contains control parameters that define the layout of the remainder of the input data that is needed.The file is arranged so that it always has a standard format at the beginning (first 8 lines) but may requiredifferent data towards the end depending on the type of calculation under consideration (axial com-pressor, radial turbine, single stage, multi-stage, and so on). Similar types of simulations always usesimilar formats for this file, but different types of simulations may require different formats for the latterpart of this file.

You specify the empirical data using one of two approaches:

• You specify the losses, the blockage, and deviation associated with individual, specifically-defined, quasi-orthogonals.

This approach is oriented around the quasi-orthogonals, enabling you to specify different local effi-ciency and blockage values for each quasi-orthogonal. Although deviation is only meaningful at thetrailing edge, a similar approach is used for deviation to be consistent with losses and blockage.

This approach is recommended for beginners.




• You provide information that enables the program to calculate the needed empirical data for blade rowsfrom built-in correlations.

This approach is oriented around blade rows and requires data to be specified at each blade rowtrailing edge (or upstream of this – see later).

Note that the correlations used for individual blade rows can be changed from one blade row to another.You can also use the correlations for deviation from one source and the correlations for losses fromanother. In general, and for simplicity, you should apply the correlations in such a way that all bladerows use the same correlations and the losses, blockage, and deviation, are also from the same source.

The correlations currently programmed and in preparation can be obtained with the use of the followingparameters for i_loss_type , i_dev_type and i_ewb_type :

i_loss_type / i_dev_type / i_ewb_typeTurbine cor-

relations

1 / 1 / 1Kacker-Okapuu

2 / 2 / 2Dunham-Came

i_loss_type / i_dev_type / i_ewb_typeCompressor

correlations

11 / 11 / 11Miller-Wright

12 / 12 / 12Miller-Wright(as modifiedby Calvert)

If required, you can specify single values of the losses, deviation, and blockage parameters, for the firstblade row or first quasi-orthogonal. These are then used throughout the domain provided that theyare not changed again. In addition, it is possible to apply multiplicative and additive correction factorsto the values predicted by the correlations, by the application of “fudge factors” (user-defined corrections).Generally this file does not need to be changed once it has been set up, so you might not need tounderstand all the intricacies of the many possibilities that it allows, and typically a standard type ofcorrelations file can be established which can then be used for all subsequent simulations with thesame set of correlations.

Examples of some typical correlations data files are given in Appendix E: Examples of Correlations DataFiles (*.cor) (p. 295), showing several specific examples for specific types of calculation.

Section 1: Character strings identifying the correlation data (max 72 characters/line)

The syntax is:

Character string – title(1)Character string – title(2)Character string – title(3)

Section 2: Integer control parameters for loss, deviation and blockage models (one line)

The syntax is:

i_loss i_dev i_ewb


Vista TF


Determines the loss specification to be used.i_loss

Notes:

• i_loss = 0

No losses are specified and the calculation uses con-stant entropy. Note, however, that in order to retainsimilar structures for the correlations file, the linesdescribed in sections 3 and 4 are still needed, but haveno effect.

• If i_loss = 1 then a user-defined variation of efficiency,loss coefficient, or dissipation coefficient across the spanis specified at the spanwise positions given in section 4.The number of definition positions across the span andthrough the domain are given in section 3. Different valuesof the efficiency, loss coefficient, or dissipation coefficientcan be specified for different quasi-orthogonal calculatingstations. A value on the first quasi-orthogonal is alwaysneeded. This value remains constant until changed by thenext quasi-orthogonal at which the loss is defined. If onlya single value is specified across the span then this is ap-plied on the mean streamline and each streamline has thesame entropy rise as on the mean streamline. If a calcula-tion with a constant efficiency for all streamlines is requiredthen the efficiency needs to be specified constant on atleast two points across the span.

• If i_loss = 2 then built-in loss correlations are used ac-cording to the values given in sections 3 and 4.

Determines the method for calculation of a blade outletflow angle (deviation or slip).

i_dev

Notes:

• If i_dev = 0 , the flow follows the mean blade camberline given in the .geo file with no deviation. This optionallows a mean S2 though-flow calculation to be carried outusing the mean stream surface obtained from a series ofS1 blade-to-blade calculations, provided that these arethen given as the geometry of the camber surface in the.geo file. Note that, in order to retain similar structuresfor the correlations file, the lines described in sections 5and 6 are still needed but have no effect.

• If i_dev = 1 then data is specified in sections 5 and 6 topredict the variation of the flow angle at the blade outletfrom the mean blade camber line across the span. The in-formation can be either in the form of a deviation angle,a specified flow outlet angle, a specified slip factor, or a





modification to the cosine rule, according to the type ofblade row.

• If i_dev = 2 then inbuilt correlations are used to determ-ine the flow angle according to the type of blade row asspecified in sections 5 and 6.

Determines the blockage correlation for boundary layersthat is to be used (ewb means “end-wall blockage”)

i_ewb

Notes:

• If i_ewb = 0 then no end-wall blockage is applied. Notethat, in order to retain similar structures for the correlationsfile, the lines described in sections 7 and 8 are still needed,but have no effect.

• If i_ewb = 1 then data is specified in sections 7 and 8 topredict the variation of the flow blockage for each quasi-orthogonal and for each streamline. A single value impliesthat a constant blockage value is applied in the wholecalculation domain.

• If i_ewb = 2 then builtin correlations are used to determ-ine the blockage according to the type of blade row asspecified in sections 7 and 8.

Section 3: Loss input data locations

The syntax is:

n_loss_sl n_loss_qo

or

n_loss_bladerow n_dummy

If i_loss equals 0 or 1 then the losses for each quasi-orthogonal can be specified, and the followingvalues are required:


Number of the spanwise positions for which the lossdata is supplied.

n_loss_sl

Number of quasi-orthogonal calculating stations throughthe domain on which the loss information is specified.

n_loss_qo

If i_loss equals 2 then the losses for each blade row can be specified, and the following valuesare required:


Number of separate blade rows for which the loss datais supplied.

n_loss_bladerow


Vista TF


A dummy integer value to ensure that section 3 alwayshas two integer values. This enables you to switch

n_dummy

between various alternatives without changing the datain the file.

If i_loss equals 1 or 2 then the following version of section 4 is required:

Section 4: User-defined loss data specification (n_loss_qo * n_loss_sl lines)

This version of section 4 is applicable when i_loss equals 0 or 1.

The syntax is:

i_qo_loss k_loss f_loss loss


Number of the quasi-orthogonal calculating station onwhich the loss, deviation, and blockage information is

i_qo_loss

specified. The specified values will be applied for allblade rows and calculating stations downstream of thisquasi-orthogonal until the value is changed by a sub-sequent section 4 with a new value of j_qo_loss . Thefirst value of i_qo_loss must be 1. Subsequent linesmay change the way in which the losses are defined.

Parameter to define how the losses are specified. Thevalue of the polytropic efficiency, the loss coefficient, or

k_loss

the dissipation coefficient, determines the entropy in-crease depending on the values of k_loss .

Notes:

• If k_loss = 1 then the small-scale static to static polytropicefficiency (etapoly ) is specified. This is used in such away that the entropy always increases so that, for an accel-erating flow with a decrease in static enthalpy, it definesa turbine efficiency, and for a decelerating flow, it definesa compressor efficiency. This small-scale efficiency is appliedin blade rows and in ducts.

• If k_loss = 2 then the entropy loss coefficient (xsi ) isspecified. Note that the loss coefficient is the entropy losscoefficient with respect to outlet plane dynamic head foraccelerating blade rows (turbine rotors and stators) andwith respect to the inlet plane dynamic head for decelerat-ing blade rows (compressor rotors and stators). An inletguide vane for a compressor is therefore calculated as aturbine blade row, and an outlet guide vane for a turbineas a compressor blade row. Like the deviation, this valueis only used at the trailing edge of a blade row.

• If k_loss = 3 then the dissipation loss coefficient (cd ) isspecified. This is applied in blade rows and in ducts.





• If k_loss = 4 to 9 then various forms of efficiency areused to specify the rotor blade row losses, as follows:

k_loss = 4 total-total polytropic efficiency

k_loss = 5 total-static polytropic efficiency

k_loss = 6 static-static polytropic efficiency

k_loss = 7 total-total isentropic efficiency

k_loss = 8 total-static isentropic efficiency

k_loss = 9 static-static isentropic efficiency

These are applied in blade rows only. The full rangeis allowed, but options 6 and 9 are probably not veryrelevant.

• If k_loss = 10 then the loss in a stator vane can be spe-cified as a total pressure loss coefficient.

• If k_loss = 11 then the performance of a stator van canbe specified as a static pressure rise coefficient.

Note

These options have been programmed in sucha way that different blade rows may havedifferent values of k_loss .

The fraction of the span at which the value of loss ap-plies. If only a single value is given, it is applied at themean streamline independent of the value of f_loss .

f_loss

Depending on the value of k_loss , this is interpretedas a small-scale polytropic efficiency (etapol ), a losscoefficient (xsi ) or a dissipation coefficient (cd ).

loss

If i_loss equals 2 then the following version of section 4 is required:

Section 4: Correlation-based loss data specification (n_loss_bladerow lines)

This version of section 4 is applicable when i_loss equals 2.

The syntax (of a single line) is:

i_loss_bladerow i_loss_type factor1 factor2 factor3 factor4 factor5 factor6


Number of the blade row on which the loss informationis specified. The specified values will be applied for all

i_loss_bladerow

blade rows downstream of this blade row until the value


Vista TF


is changed by a subsequent section 4 with a new valueof i_loss_bladerow , and then this will again be ap-plied until changed. The first value must be 1.

Parameter to define which correlations are used for theblade row losses. The current loss correlations incorpor-ated into the method are as follows:

i_loss_type

• i_loss_type = 1

Turbine loss correlations of Kacker and Okapuu

• i_loss_type = 2

Turbine loss correlations of Dunham and Came

• i_loss_type = 11

Compressor loss correlations of Miller and Wright

User-defined multiplication or addition factors to thelosses determined by the correlations. This allows the

factor1 tofactor6

loss correlations to be modified to improve matchingwith experimental or CFD data. These have differentfunctions for the different correlation systems.

factor1 : Multiplication factor on profile losses

factor2 : Multiplication factor on secondary losses

factor3 : Multiplication factor on tip clearance losses

factor4 : Multiplication of penetration of secondarylosses

factor5 : Multiplication of penetration of tip clearancelosses

factor6 : Not in use

Section 5: Deviation input data locations

The syntax is:

n_dev_sl n_dev_qo

or

n_dev_bladerow n_dummy

If i_dev equals 0 or 1 then the deviation for each quasi-orthogonal can be specified, and the fol-lowing values are required:





Number of spanwise positions for which the flow angledata is supplied.

n_dev_sl

Number of quasi-orthogonal calculating stations onwhich the flow angle information is specified.

n_dev_qo

If i_dev equals 2 then the deviation flow angle data can be specified for each blade row, and thefollowing values are required:


Number of separate blade rows for which the flow angledata is supplied.

n_dev_bladerow

A dummy integer value to ensure that section 5 alwayshas two integer values.

n_dummy

If i_dev equals 0 or 1 then the following version of section 6 is required:

Section 6: Flow angle data specification (n_dev_qo * n_dev_sl lines)

This version of section 6 is applicable when i_dev equals 0 or 1.

The syntax is:

i_qo_dev k_dev f_dev dev


Number of the quasi-orthogonal calculating station onwhich the deviation information is specified. The spe-

i_qo_dev

cified values will be applied for all blade row trailingedges downstream of this quasi-orthogonal until thevalue is changed by a subsequent section 6 with a newvalue of i_qo_dev . The first value must be 1.

Determines the type of outlet flow angle calculation, asfollows:

k_dev

• If k_dev = 1 then the deviation angle is specified (in de-grees).

• If k_dev = 2 then the relative outlet flow angle is specified(in degrees).

• If k_dev = 3 then a slip factor is specified and applied onthe mean line (dimensionless). Note that if no value isspecified then the Wiesner slip factor is used. Only recom-mended for radial impellers and not for mixed flow stages,which should use deviation.

• If k_dev = 4 then a slip factor is applied at each radius ofthe trailing edge (dimensionless). Note that if no value isspecified then the Wiesner slip factor is used. Only recom-mended for radial impellers and not for mixed flow stages,which should use deviation.


Vista TF


• If k_dev = 5 then a correction to the flow angle calculatedby the cosine rule is specified. If zero is specified for thiscorrection, then the cosine rule is used to calculate therelative flow angle at the outlet for all radii. If a non-zerovalue is specified then this modifies the flow angle by thisconstant amount (in degrees). A positive value increasesthe deviation.

The fraction of the span of at which the value of thedeviation (and similar things) applies. If only a single

f_dev

value is given, it is applied at the mean streamline inde-pendent of its value.

Depending on the value of k_dev , this is interpreted asa deviation angle, flow angle, slip factor, or correctionto the cosine rule.

dev

If i_dev equals 2 then the deviation for each blade row must be specified using the followingversion of section 6:

Section 6: Flow angle data specification (n_dev_bladerow lines)

This version of section 6 is applicable when i_dev equals 2.

The syntax is:

i_bladerow_dev i_dev_type factor1 factor2 factor3 factor4 factor5 factor6


Number of the blade row on which the deviation inform-ation is specified. The specified values will be applied

i_bladerow_dev

for all blade row trailing edges downstream of this quasi-orthogonal until the value is changed by a subsequentsection 6 with a new value of i_bladerow_dev . Thefirst value must be 1.

Parameter to define which correlations are used for theblade row deviations. The current deviation correlationsincorporated into the method are as follows:

i_dev_type

• i_dev_type = 1 (Kacker Okapuu)

Turbine deviation using the cosine rule.

• i_dev_type = 2 (Dunham and Came)

Turbine deviation using the cosine rule.

• i_dev_type = 11 (Miller Wright)

Compressor deviation correlations of Miller and Wright.

User-defined multiplication or addition factors to thedeviations determined by the correlations. This allows

factor1 tofactor6

the deviation correlations to be modified to improve





matching with experimental or CFD data. These havedifferent functions for the different correlation systems.

factor1 : Additive factor in degrees (°) on hub deviation

factor2 : Additive factor in degrees (°) on mean devi-ation

factor3 : Additive factor in degrees (°) on tip deviation




Section 7: Blockage input data locations

The syntax is:

n_ewb_sl n_ewb_qo

or

n_ewb_bladerow n_dummy

If i_ewb equals 0 or 1 then the blockage for each quasi-orthogonal can be specified, and the fol-lowing values are required:


Number of spanwise positions for which the blockagedata is supplied.

n_ewb_sl

Number of quasi-orthogonal calculating stations onwhich the blockage information is specified.

n_ewb_qo

If i_ewb equals 2 then the blockage correlation can be specified for each blade row, and the fol-lowing values are required:


Number of separate blade rows for which the blockageis supplied.

n_ewb_bladerow

A dummy integer value to ensure that section 7 alwayshas two integer values.

n_dummy

If i_ewb equals 0 or 1 then the blockage for each quasi-orthogonal must be specified using thefollowing version of section 8:

Section 8: End wall blockage data specification (n_dev_qo * n_dev_sl lines)

This version of section 8 is applicable when i_ewb equals 0 or 1.

The syntax is:


Vista TF

i_qo_ewb k_ewb f_ewb ewb


Number of the quasi-orthogonal calculating station onwhich the blockage information is specified. The specified

i_qo_ewb

values will be applied for all blade rows and calculatingstations downstream of this quasi-orthogonal until thevalue is changed by a subsequent section 4 with a newvalue of j_qo_ewb . The first value must be 1.

Determines the type of blockage calculation. Has no ef-fect because the blockage is input in only one form asbelow.

k_ewb

The spanwise location of the value of the blockage. Ifonly a single value is given, it is applied at all streamlines.

f_ewb

End-wall boundary layer blockage is specified. A valueof 0.05 represents 5% blockage of the flow channel by

ewb

the end-wall boundary layers. If zero is specified thenthere is no end-wall boundary layer blockage. Typicallya constant value is specified across the whole span, butthis parameter can be varied across the span to allowfor the higher blockage in the end-walls related to bladeends with tip clearance.

If i_ewb equals 2 then the following version of section 8 is required:

Section 8: End wall blockage data specification (n_dev_qo * n_dev_sl lines)

This version of section 8 is applicable when i_ewb equals 2.

The syntax is:

i_bladerow_ewb i_ewb_type factor1 factor2 factor3 factor4 factor5 factor6


Number of the blade row on which the blockage inform-ation is specified. The specified values will be applied

i_bladerow_ewb

for all blade row trailing edges downstream of this quasi-orthogonal until the value is changed by a subsequentsection 6 with a new value of i_bladerow_ewb . Thefirst value must be 1.

Parameter to define which correlations are used for theblade-row blockage. The current blockage correlationsincorporated into the method are as follows:

i_ewb_type

• i_ewb_type = 1

No blockage consistent with Kacker Okapuu.

• i_ewb_type = 2

No blockage consistent with Dunham and Came.

• i_ewb_type = 11





Compressor blockage correlations of Miller and Wright.

User-defined multiplication or addition factors to theblockage determined by the correlations. This allows the

factor1 tofactor6

end-wall blockage correlations to be modified to improvematching with experimental or CFD data. These havedifferent functions for the different correlation systems.







12.2.2.8. Specification of the Real Gas Properties Data File (*.rgp)

The real gas property input data file includes lines of text that help you to identify the parametersdefined there. The name .rgp is used to describe this file, but the content is different compared tothe usual (ANSYS CFX) real gas property file, which has a similar name.

An example of a real gas property data file is given in Appendix H: Example of a Real Gas Property DataFile (*.rgp) (p. 313).

Section 1: Character strings identifying the aerodynamic data (max 72 characters/line)

The syntax is:


Section 2: Name of gas (72 characters)

The syntax is:

gas_name


Text characters defining the gas namegas_name

Section 3: Molecular mass and/or Gas constant

The syntax is:

MW Gas_R


Molecular mass of gas (kg/kmol)MW

Gas constant = Universal gas constant/MW J(kg/K)Gas_R


Vista TF

Note that if either of these values is specified with a numerical value of less than 0.1 then it is cal-culated from the universal gas constant using the other value, which needs to be specified correctly.

Section 4: Critical point parameters and acentric factor

The syntax is:

Pc (Pa) Tc (K) Vc (m3/kg) gas_omega (-)


Critical pressure (Pa)Pc

Critical temperature (K)Tc

Specific volume at critical point (m3/kg)Vc

Acentric factor (-)gas_omega

Section 5: Temperature limits of specific heat curve polynomial

The syntax is:

T_min (K) T_max(K) order_T_poly (max 8)


Lowest temperature in range (K)T_min

Highest temperature in range (K)T_max

number of coefficients = Order of polynomial plus 1n_T_coeff

Section 6: Coefficients of cp polynomial (T_min < T < T_max)

The syntax is:

A1 A2 A3 A4 A5 A6 A7 A8


First coefficient in polynomialA1

Second coefficient in polynomialA2

Subsequent coefficients…

Eighth coefficient in polynomialA8

Note that this representation only allows a single interval dependant polynomial to be defined. If athird order polynomial is specified then only four coefficients are needed.

Note that if the value of the first coefficient specified is less than the value of the gas constant then allof the coefficients will be multiplied by the gas constant as in some databases this form of the coefficientsis used.

12.2.2.9. Specification of the Output Data File (*.out)

The output data file includes lines of text in ASCII format that show the results of the simulation. Thefile consists of several sections:




Section 1: Input data

At the start of this file, a list of the input data files that were used is recorded in this file so that a recordof the file names is available.

INPUT DATA FILES: Control data file: agard.con Geometric data file: agard.geo Aerodynamic data file: agard.aer Correlation data file: agard.cor Restart data file: agard.rstOUTPUT DATA FILES: Results file: agard.out Tecplot data file: impeller.txt CFD-POST data file: agard.csv Convergence history file: agard.hst Interface output file: agard.int

Section 2: Reference data

The program uses the input data to set up the values of various other parameters; for example, wherethe reference mass flow is specified, the reference volume flow is calculated. The values of all otherparameters not included in the input specification are given in this section. In addition, the internalcalculation of the damping factors is provided. Other data listed here relates to the program's own es-timates of the throat area and the throat position. An example is given below.

Example radial compressor calculation:

Reference flow parameters-------------------------

Mass flow distribution across streamlines: 0.000 0.062 0.125 0.188 0.250 0.312 0.375 0.438 0.500 0.562 0.625 0.688 0.750 0.812 0.875 0.938 1.000

Inlet distribution across streamlines: Fraction of flow: 0.00 Total pressure: 100000.00 Total temperature: 293.00 Inlet swirl: 0.00

Ideal gas calculation

gamma_gas gas_m r_gas cp_gas 1.400 3.500 287.200 1005.200

ref_pt(bar) ref_tt ref_rhot ref_soundt 1.0000 293.00 1.1884 343.23

Reference rotational direction: clockwise

ref_d ref_u ref_omega ref_n0.2700 398.10 2948.91 28160.0

ref_mach ref_phi ref_mass ref_volume1.1599 0.0667 2.3000 1.94

ref_re ref_mue 6963614.0 0.00001834

ref_n1 ref_n2 ref_n3 28160.0 0.0 0.0

Simulation with no spanwise mixing

Iterate to mass flow, ref_mass = 2.30000

First approximation of streamline positionstaken from an earlier calculation

Estimated max. thicknesses for blade row 1span 0.00000 0.17343 0.34429 0.51305 0.67879 0.84126 0.97669 1.00000thickne 0.00351 0.00319 0.00291 0.00266 0.00241 0.00215 0.00195 0.00191


Vista TF

User-input throat widths for blade row 1span 0.00000 0.17343 0.34429 0.51305 0.67879 0.84126 0.97669 1.00000throat 0.01760 0.02137 0.02460 0.02732 0.02959 0.03141 0.03257 0.03276

Estimated throat positions for blade row 1span 0.00000 0.17343 0.34429 0.51305 0.67879 0.84126 0.97669 1.00000throat_pos 0.00969 0.02084 0.03424 0.04835 0.06224 0.07573 0.08688 0.08882

Empirical correlation data

User specified losses User specified deviation No blockage specified

Relaxation factor calculation of Wilkinson------------------------------------------

Maximum aspect ratio: 6.0667at calculating station: 5damp_sc (user input): 0.2500damp_sc (Wilkinson): 0.0577

Warning: damp_sc reduced by the code to the value suggested by Wilkinson

Section 3: Short history of the convergence

The .out file contains a statement about convergence of the results. A converged calculation includesa header such as the following:

************************************************ Vista TF converged -it_main: 6 ** ** cm_error(%) p_error(%) mass_error(%) ** 0.042 0.000 0.000 ************************************************* Global performance ** ** mass (kg/s): 0.093200 ** est. choke: 0.107995 ** t-t t-s s-s ** eta_p: 0.8453 0.7415 ** eta_s: 0.8553 0.7583 ** pr: 0.5800 0.5374 0.6100 ** er (1/pr): 1.7243 1.8609 1.6392 ** power (kW): -5.4316 ************************************************

A more detailed convergence history is output to the history file, which is described in Specificationof Convergence History Data File (*.hst) (p. 273).

Section 4 : Simulation results on quasi-orthogonal planes and streamlines

Results are provided at every quasi-orthogonal requested (see control file parameter i_print_plane )in the level of detail requested (parameter i_print_level ). The data is provided across the span foreach streamline. If the number of streamlines is greater than 9 then data for every second streamline isgiven. The structure of this data and the information provided varies depending on the type of calculationstation. The example below is for a radial impeller leading edge:

Quasi-orthogonal - i = 3 n_blade n_curve i_type i_row i_spool 13 6 3 0 0Rotor blade - leading edgeAt throat (or just upstream of throat)CompressorRadial impellerThroat area = 0.00274713Annulus choke parameters: dm_dcm = 0.79601 Mach_eff = 0.45166choke mass at this q_o = 7.72100




choke_mass of machine = 7.72100choke_mass of this blade row = 7.72100current_mass at this q_o = 6.70000inlet mass flow = 6.70000Factor on slope of ree: grad_re = 1.00000 cm_guess_save = 152.57564Flow parameters - print_level=1streamline 1 3 5 7 9 11 13 15 17r [m] 0.07612 0.08753 0.09759 0.10647 0.11445 0.12176 0.12855 0.13493 0.14100z [m] -0.13500-0.13500-0.13500-0.13500-0.13500-0.13500-0.13500-0.13500-0.135001/rc[1/m] -5.996 1.353 2.758 2.818 2.475 1.984 1.438 0.864 0.244throat mm 23.00 24.41 25.66 26.77 27.40 27.84 28.17 28.30 29.00f_sl [-] 0.17503 0.17777 0.18110 0.18431 0.18734 0.19021 0.19295 0.19556 0.19806f_qo [-] 0.00000 0.17593 0.33098 0.46775 0.59076 0.70340 0.80807 0.90652 1.00000f_bl [-] 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000f_thr [-] 0.02584 0.03400 0.04157 0.04823 0.05408 0.05923 0.06379 0.06785 0.07150gamma_in 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000gamma_out 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000cm [m/s] 146.87 142.25 145.30 149.18 152.57 155.22 157.11 158.32 158.87- error % 0.0001 -0.0002 0.0001 0.0001 0.0001 0.0001 0.0000 -0.0002 -0.0001- max 169.57 166.48 168.30 172.19 174.87 177.33 179.61 180.96 186.11cu [m/s] 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00M_cm [-] 0.437 0.423 0.432 0.444 0.455 0.463 0.469 0.472 0.474M_crit[-] 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000M_rel [-] 0.765 0.837 0.914 0.985 1.050 1.109 1.163 1.213 1.260beta_fl° -55.19 -59.65 -61.79 -63.21 -64.34 -65.33 -66.23 -67.08 -67.90beta_bl° -43.03 -46.77 -50.17 -52.99 -55.37 -57.46 -59.30 -60.92 -62.35incidence 12.16 12.88 11.62 10.22 8.98 7.86 6.93 6.16 5.55p [bar] 0.8706 0.8778 0.8730 0.8669 0.8614 0.8571 0.8540 0.8519 0.8510t [K] 282.38 283.04 282.60 282.04 281.54 281.14 280.84 280.66 280.57rho[kg/m3] 1.0742 1.0806 1.0764 1.0710 1.0661 1.0623 1.0595 1.0577 1.0569A*/A 0.9485 0.9758 0.9935 0.9998 0.9980 0.9907 0.9799 0.9666 0.9517m/m_max 0.8661 0.8545 0.8633 0.8664 0.8725 0.8753 0.8747 0.8749 0.8536m_prime 75.46 84.55 95.91 106.88 116.97 126.15 134.46 141.98 148.76m_pr_max 87.12 98.94 111.09 123.37 134.07 144.11 153.70 162.27 174.26ch_ratio 0.8662 0.8545 0.8634 0.8664 0.8725 0.8753 0.8748 0.8749 0.8537Flow parameters - print_level=2eps [°] 12.82 14.30 12.54 10.34 8.13 6.01 4.00 2.08 0.23psi [°] 77.18 75.70 77.46 79.66 81.87 83.99 86.00 87.92 89.77c [m/s] 146.87 142.25 145.30 149.18 152.57 155.22 157.11 158.32 158.87w [m/s] 257.28 281.50 307.34 330.98 352.35 371.83 389.80 406.55 422.31u [m/s] 211.24 242.91 270.83 295.46 317.60 337.88 356.73 374.45 391.29wu [m/s] -211.24 -242.91 -270.83 -295.46 -317.60 -337.88 -356.73 -374.45 -391.29M_abs [-] 0.437 0.423 0.432 0.444 0.455 0.463 0.469 0.472 0.474alpha_fl° 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00pt [bar] 0.9921 0.9921 0.9921 0.9921 0.9921 0.9921 0.9921 0.9921 0.9921tt [K] 293.00 293.00 293.00 293.00 293.00 293.00 293.00 293.00 293.00s [J/kgK] 2.29 2.29 2.29 2.29 2.29 2.29 2.29 2.29 2.29ds[J/kgK] 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00ewb [-] 0.0500 0.0500 0.0500 0.0500 0.0500 0.0500 0.0500 0.0500 0.0500Loading parameters - print_level=3w_s [m/s] 287.34 314.82 342.24 366.60 388.17 407.50 425.12 441.51 456.87w_p [m/s] 227.21 248.18 272.45 295.36 316.53 336.16 354.47 371.59 387.74M_ws [-] 0.855 0.936 1.018 1.091 1.157 1.215 1.268 1.318 1.364M_wp [-] 0.676 0.738 0.810 0.879 0.943 1.002 1.058 1.109 1.157C_btob [-] 0.234 0.237 0.227 0.215 0.203 0.192 0.181 0.172 0.164

12.2.2.10. Specification of the Text Data Files (*.txt)

The text data files include lines of text in ASCII format which are intended as input for various softwarepackages for producing plots of the converged results. The text files can be used as input data forTecplot. Standard layout (.lay ) files for Tecplot have been prepared which allow typical diagrams tobe produced on the basis of this data. The information is structured in such a way that it can just aseasily be used by Excel or some other similar program with an appropriate interface.

In fact, after running the program, several standard .txt files for Tecplot are produced: one coveringthe whole flow field and one for each blade row. The flow field data is written in the file prefix.txt ,and the blade-row data in the files row_01_prefix.txt , row_02_prefix.txt , and so on, wherebythe number of the files refers to the blade rows numbered from the start of the computational domain.


Vista TF

You only need to provide the name of one of these files in the vista_tf.fil file, because the othersare automatically generated and numbered by the program.

Some standard Tecplot layout files have been prepared that operate on the .txt files to producetypical plots that are needed during the design of a component. For example, the files used for radialcompressor impellers are called: flowfield_2d_v_2_0.lay and rc_impeller_row1.lay . Thefirst of these prepares a plot showing the 2D meridional channel with contours of constant parameters(such as meridional velocity, swirl velocity, static temperature, Mach number, cm_error and so on)and the second prepares blade loading plots for a typical blade row. The layout file (.lay ) determinesthe format of the plot and the ASCII .txt file contains the data to be plotted. The format and thescales can also be changed on-line within Tecplot by clicking on the screen. Other Layout files for othercases (ac = axial compressor, hp = hydraulic pump, ht= hydraulic turbine, rt = radial turbine, at= axialturbine) are also available and these can be easily customized (scales, parameters and so on).

The standard Tecplot layout files (.lay ) have all been written to work on the data in text files namedimpeller.txt and row_01_impeller.txt , and so on. The first dataset, impeller.txt , containsinformation for the contour plots showing the meridional velocity and other parameters projected onthe meridional plane and the second, row_01_impeller.txt , contains the various blade loadingparameters for an individual blade row. If the case concerned has several blade rows then there is alsoa file called row_02_impeller.txt , row_03_impeller.txt , and so on; there is one file for eachblade row. A new layout file is needed for each blade row. The letters “impeller” in this name rely onthe fact that the prefix for the .txt file has this name in the vista_tf.fil file.

Other prefixes will produce .txt files containing the prefix as specified in the vista_tf.fil file. Inthis case, the layout files need to be modified; the layout files can be opened in a text editor; the nameof the .txt file is on the second line. This name can then be changed to match the name given as aprefix in the vista_tf.fil file. Alternatively, the same layout files can be used for several cases ifall data that needs to be plotted always retains the prefix name impeller ; the text output file fromvista_tf always has the name impeller.txt in the vista_tf.fil file.

Generally, a good strategy is to set up the .aero , .cor , .con , and .txt files with a certain prefix inthe vista_tf file and to let these keep the same names for all runs on a particular case because theynormally do not change as the design progresses. It then remains necessary to switch the .geo filesaround to look at different impellers, or else the prefix for the .geo file can be retained and the casescan be distinguished by simply putting them into different directories to distinguish the different cases.

The data in the prefix.txt file is in approximately the following format:

TITLEFlow field data for whole flow field from hub to shroud for all q-os

Variables = '"r" "z" "cm" "cu" "cr" "cz" "p" "t" "pt" "tt" "s" "h" "q_o" "error" "choke_ratio""M_rel" "M_abs" "alpha_flow" "beta_flow" "1/rc" "beta"'

"r" Radius coordinate"z" Axial coordinate"cm" Meridional velocity"cu" Swirl velocity"cr" Radial velocity"cz" Axial velocity"p" Static pressure"t" Static temperature"pt" Total pressure"tt" Total temperature"s" Entropy"h" Static enthalpy“q_o” an integer identifying the particular type of q-o“error” Meridional velocity error in %"choke_ratio" choke ratio (ratio of local mass flow to choke flow at this location)




"M_rel" Relative Mach number"M_abs" Absolute Mach number"alpha_flow" Absolute flow angle [°]"beta_flow" Relative flow angle [°]"1/rc" Curvature ( inverse of the radium of curvature)"beta” Blade angle [°]

The data in the row_01_prefix.txt file is approximately in the following format:

TITLEData on hub, mean and tip streamlines for all blade rows from LE to TEand data along leading edge and trailing edge of all blade rows

VARIABLES =VARIABLES = "i_type", "f_bl", "M", "M_s", "M_p", "w", "w_s", "w_p",..."p", "p_s", "p_p", "c_btob", "c_htos", "beta_bl", "beta_fl", "dep_angle", "gamma_in", ...gamma_out", "f_qo", "de_haller", "cp_ideal", "incidence", "deviation", "lambda", ..."df_lieblein", "zweifel" "c_lift", "c_zw""f_bl" Fraction of meridional distance along the blade"M" Mean mid-passage Mach number (relative to blade)"M_s" Suction side Mach number"M_p" Pressure side Mach number"w" Mean mid-passage velocity (relative to blade)"w_s" Suction surface velocity"w_p" Pressure surface velocity"p" Static pressure in mid-channel between two blades"p_s" Suction surface static pressure"p_p" Pressure surface static pressure"c_btob" Blade to blade loading parameter (c_btob = (w_s-w_p)/w, that is the difference between suction surface and pressure surface velocities divided by the mid-channel velocity"c_htos" Hub to shroud loading parameter (c_htos = (cm_shroud – cm_hub)/ cm_mean, that is the difference between the meridional velocity on the casing and the hub divided by the mean velocity."beta_bl" Blade angle in degrees [°]"beta_fl" Flow angle in degrees [°]"dep_angle" Departure angle in degrees [°]"gamma_in" Blending function at blade inlet"gamma_out" Blending function at blade outlet"f_qo" fractional distance along q-o"de_haller" De Haller number (de_haller = w2/w1, that is outlet/ inlet relative velocity"cp_ideal" Ideal static pressure recovery coefficient ( Cpideal = 1 – (de_haller)**2 )"incidence" Incidence at the leading edge [°]"deviation" Deviation at the trailing edge[°]"lambda" Work coefficient ( lambda = (h2-h1)/u**2, that is Deltah / u squared)"df_lieblein" Lieblein Diffusion Factor (see below)"zweifel" Zweifel loading parameter (see below)"c_lift" Lift coefficient (see below)"c_zw" Lift coefficient times solidity (see below)

Note that the values of the last four parameters have been included mainly for axial blade rows. Eachof them includes the blade solidity (ratio of chord to spacing) in its definition. In radial machines, thespacing and solidity change with radius along the streamline. There is no generally agreed method tocalculate these parameters for radial machines, so the mean value of the spacing has been selected inthe following definitions used in Vista TF. In any case, caution is suggested in the use of these parametersfor radial machine blade rows because the experimental basis for limit values has generally been derivedfrom axial machines.

Lieblein diffusion factor (df_lieblein)

Zweifel coefficient (zw)


Vista TF

Lift coefficient (CL)

where

Zweifel number (c_zw)

12.2.2.11. Specification of the CFD-Post Output Files (*.csv)

The text data files *prefix.csv include lines of text in ASCII Comma Separated Variable format files(.csv ) which are intended as input for CFD-Post for plotting purposes, using the user surface dataformat. For details of this format, see the documentation provided with CFD-Post.

In the current .csv files, the following data is available for each grid node:

[Name]Vista-TF[Data]X [m], Y [m], Z [m], Cm [m/s], Cu [m/s], Cr [m/s], Cz [m/s], p [bar], t [K], s [J kg^-1 K^-1],h [J/kg], q_o [], error [], M_rel[], M_abs[], rc [m^-1]

The file global_prefix.csv contains a list of parameters, whereby these are specified as:

Name 1 = value1 [units]Name 2 = value2 [units]

Separated by comment lines which begin with “#”.

In addition, there are four additional files produced for each blade row from 1 to n:

row_0n_hub_prefix.csvrow_0n_mean_prefix.csvrow_0n_tip_prefix.csvrow_0n_loading_prefix.csv

These contain essentially similar information to the row_0n_prefix.txt files described above, butthis is split into four separate files: one for the hub, one for the mean span, one for the tip, and one forthe loading parameters.

12.2.2.12. Specification of Convergence History Data File (*.hst)

The convergence history data file includes lines of text in ASCII format that show the convergence ofthe simulation. This contains details of the convergence of the main iterative procedures, and extensivedetails of the terms in the radial equilibrium equation for each stream tube and calculating plane. It israre for this to be examined in any depth, but this can be useful to identify problems if the solutionfails to converge.




Section1: Input Data

At the start of the .hst file, the input data is recorded. This is essentially in the same format as the inputfiles. At the start of this file, there is a list of the names of the input data files that were used. This inform-ation can be useful if an error occurs in the input data because the .hst file records only the data thathas been successfully read into the program.

Section 2: Convergence Data

At every iteration, a summary of the iteration progress is provided. It looks like this:

History of the main iteration loop---------------------------------- it_main: 1 error_cm(%): 100.000 at i_qo: 0 j_sl: 0 it_mass: 11 at i_qo: 1 it_main: 2 error_cm(%): 38.940 at i_qo:10 j_sl:17 it_mass: 11 at i_qo: 1 it_main: 3 error_cm(%): 30.036 at i_qo:10 j_sl:17 it_mass: 11 at i_qo: 3 it_main: 4 error_cm(%): 24.600 at i_qo:10 j_sl:17 it_mass: 11 at i_qo: 4 it_main: 5 error_cm(%): 20.939 at i_qo:10 j_sl:17 it_mass: 11 at i_qo: 5 it_main: 6 error_cm(%): 18.306 at i_qo:10 j_sl:17 it_mass: 11 at i_qo: 6 it_main: 7 error_cm(%): 16.403 at i_qo:10 j_sl:17 it_mass: 11 at i_qo: 6

.....

it_main: 284 error_cm(%): 0.013 at i_qo: 7 j_sl:17 it_mass: 1 at i_qo: 1 it_main: 285 error_cm(%): 0.012 at i_qo: 7 j_sl:17 it_mass: 1 at i_qo: 1 it_main: 286 error_cm(%): 0.011 at i_qo: 7 j_sl:17 it_mass: 1 at i_qo: 1 it_main: 287 error_cm(%): 0.010 at i_qo: 7 j_sl:17 it_mass: 1 at i_qo: 1 it_main: 288 error_cm(%): 0.009 at i_qo: 7 j_sl:17 it_mass: 1 at i_qo: 1

Section 3: Streamline Curvature Solution Data

At every quasi-orthogonal, details of the terms in the radial equilibrium equation for each stream tubeare provided. This can be useful in identifying errors and also helpful to determine the magnitude ofthe terms in the equations. “rhs” is the value of the right hand side of the radial equilibrium equation,giving the square of the gradient of the meridional velocity.

The streamline curvature solution data looks like this:

Quasi-orthogonal - i = 5 n_blade n_curve i_type i_row i_spool 9 8 3 0 0Radial equilibrium parameters -print_level = 4rhs 1049.90 1059.86 871.10 684.48 510.68 349.39 199.63 55.28ret 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00ret_dh 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00ret_tds 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00ret_drcu 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00sct 1049.90 1059.86 871.10 684.48 510.68 349.39 199.63 55.28sct_rc 641.48 681.97 558.43 431.70 315.91 211.42 118.25 34.87sct_dcm 408.43 377.89 312.67 252.78 194.78 137.97 81.38 20.40bft 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00dft 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00cm_error% -0.0010 0.0002 -0.0003 -0.0007 -0.0007 -0.0003 0.0003 0.0010f_mixing 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.00001/rc 2.853 4.466 4.153 3.505 2.773 2.016 1.261 0.517 -0.209cm 137.936 145.590 152.396 157.708 161.716 164.594 166.473 167.450 167.568rcu 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000drcu/dm 47.323 53.336 48.562 35.419 15.376 -11.226 -43.868 -81.313-123.649s 1.885 1.885 1.885 1.885 1.885 1.885 1.885 1.885 1.885bl_block 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

12.2.3. Software Limitations

There are several potential sources of errors and uncertainties in all CFD simulations, all of which areextremely relevant in turbomachinery applications using a throughflow program, such as Vista TF. Thestandard categorization of such errors is into the following groups:


Vista TF

• Numerical errors

• Model errors

• Application uncertainties

• User errors

• Software errors

and these are briefly described next.

Numerical errors decrease the quality of simulations and can be reduced but not eliminated. The controlof numerical error is largely a matter of adequate grid design, appropriate numerical discretization, andadequate level of convergence. An example in Vista TF is the approximation of the streamlines by apiecewise parabola through 3 points. In a throughflow calculation the so-called model errors probablyoutweigh all other sources of error. These are related to the fact that the equations that are solved donot really describe the real flow particularly adequately (in this case the solution is for inviscid, circum-ferentially-averaged mean values on widely spaced grid lines). Application uncertainties may be onlypartly in your control and are related to the detail with which aspects of the geometry are accuratelyknown or specified. Typical examples would be a lack of knowledge about the fillet radii in the geometryfor a blade row calculation, or not having correct information about the inlet boundary conditions fora specific calculation. User errors relate to incorrect use of the program, such as making use of incorrectcontrol parameters to control the calculation. With regard to software errors, every reasonable precautionto ensure the accuracy and reliability of the Vista TF program has been taken. However, when usingthe program, especially for a critical design, you should first complete an appropriate validation andcalibration process. A good turbomachinery analysis procedure dictates that any program, includingVista TF, must be thoroughly tested with non-critical data before there is any reliance on it.

12.2.4. Streamline Curvature Throughflow Theory

Vista TF is a general purpose streamline curvature throughflow program for the analysis of all types ofturbomachinery, but with special emphasis on single stage centrifugal machines, such as radial pumps,turbines, and compressors, which can usually not be calculated by other throughflow programs.

The program has the ability to compute radial blade rows in both compressors and turbines, and theswirling flows in the radial channels of such machines. This is possible because the blade and channelgeometry is defined in a very general way independent of the meridional streamline direction.

This section of the documentation is intended for readers who are not familiar with throughflowmethods; it provides a general introduction to the streamline curvature axisymmetric meridionalthroughflow calculation method. Some details are given in Casey and Robinson (2008) [5] (p. 283). Thissection includes a list of technical papers on the subject.

The following topics are discussed:12.2.4.1.The Equations12.2.4.2.The Mean Stream Surface12.2.4.3.The Grid12.2.4.4. Ductflow and Throughflow12.2.4.5. Iterative Solution Procedure12.2.4.6. Initial Estimate12.2.4.7. Radial Equilibrium Equation12.2.4.8. Combination of Velocity Gradient and Continuity Equations12.2.4.9. Relaxation Factors




12.2.4.10. Streamline curvature12.2.4.11. Equations for Enthalpy and Angular Momentum12.2.4.12. Boundary Conditions12.2.4.13. Empirical Data12.2.4.14. Blade-to-blade Solution12.2.4.15. Spanwise Mixing12.2.4.16. Streamline Curvature Throughflow Theory: Bibliography

12.2.4.1. The Equations

The equations that are solved are:

• The continuity equation

• The energy equation (Euler equation of turbomachinery)

• A suitable equation of state

• The inviscid momentum equation for the flow on the mean axisymmetric stream surface

Although these equations are inviscid and do not include frictional forces, the effect of the losses areincluded by empirical changes to the entropy in the equation of state, such that the final solution hasa density and pressure field consistent with the presence of losses in the flow.

12.2.4.2. The Mean Stream Surface

This blade-like surface may be regarded as being obtained by averaging all flow properties in the cir-cumferential direction, or as being the flow on a mean stream surface between the blade rows, whoseorientation is roughly determined by the blade camber surface. The flow blockage caused by the bladesis roughly fixed by the circumferential blade thickness of the blades. It is often known as the S2 meridi-onal stream surface, in contrast to the S1 blade-to-blade surface, and in fact corresponds to the so-called S2m surface in the theory of Wu (1952) [22] (p. 283). See Figure 12.2: S1 and S2 Stream Surfacesin the Theory of Wu (1952) (p. 277). Note that it has an arbitrary orientation in the circumferential directionand that the flow in the mean stream surface is determined by the slope angles of the surface to theaxial and the radial directions, both of which need to be specified in the geometrical input data.


Vista TF

Figure 12.2: S1 and S2 Stream Surfaces in the Theory of Wu (1952)

12.2.4.3. The Grid

The grid for the calculation is based on the streamlines of the mean circumferentially averaged flow inthe meridional direction, and fixed calculating stations which are roughly normal to the streamlines.The meridional streamline grid is not fixed, apart from the hub and shroud streamlines on the annuluswalls, but changes continually during the iterations so that location of the final streamlines is a resultof the solution. The fixed calculating stations are orientated with the blade row leading and trailingedges and are often known as quasi-orthogonals because they are normally nearly orthogonal to thestreamlines. The quasi-orthogonals can be in duct regions, that is in the blade-free space upstream anddownstream of blade rows, at the leading and trailing edges of the blade rows (which actually are alsogenerally considered to be in the duct region) and internally within the blade rows. See Figure 12.3: Quasi-orthogonal Calculating Stations in an Axial Compressor Stage Calculation (p. 278). The computationalresults are available only at the points where the streamlines cross the quasi-orthogonals. In comparisonwith modern CFD methods, a very crude grid is used with typically 9 to 17 streamlines and 3 to 15quasi-orthogonals per blade row, giving 50 to 300 nodes per blade row in comparison to 100 000 to500 000 in modern CFD simulations.




Figure 12.3: Quasi-orthogonal Calculating Stations in an Axial Compressor Stage Calculation

In many early streamline curvature programs, the quasi-orthogonals were generally straight radial linesfrom hub to casing placed roughly at the location of the leading and trailing edges, but given the moreextensive use of lean and sweep in modern designs, and the natural leading edge and trailing edgecurvature arising from the variable stagger across the span in most rotors, curved quasi-orthogonals ofarbitrary orientation are more appropriate. The use of a grid involving meridional streamlines, all ofwhich proceed smoothly from the inlet plane to the outlet plane of the calculation domain, assumesthat there is no reverse flow, and thus operation points with reverse flow in the mean meridional flow(even at the design point) cannot be calculated by this method.

12.2.4.4. Ductflow and Throughflow

Earlier methods, which did not include blade internal calculating planes, are often known as duct flow

methods because only planes in the duct regions between the blades, and at the leading and trailingedges, are used. Throughflow refers to methods in which internal planes within each blade row are in-cluded as shown in Figure 12.3: Quasi-orthogonal Calculating Stations in an Axial Compressor StageCalculation (p. 278). In throughflow methods, the component of the blade force acting along the directionof the quasi-orthogonal can be included in the meridional throughflow equations, so that some globalaspects of the effect of the blade lean on the meridional flow can be included. This is not possible induct flow methods without some empirical adjustment of the pressure gradients.

12.2.4.5. Iterative Solution Procedure

The solution method is iterative in terms of several variables (primarily the meridional velocity, but alsothe density, streamline location, and other variables), all of which progressively converge to a finalconverged solution within a fairly complex structure of three nested iterations. In this convergenceprocedure, many variables lag behind the main iteration for changes in the meridional velocity, so thatduring one iteration leading to an update of the meridional velocity, many parameters are treated asconstant and are then updated based on the new estimate of the meridional velocity prior to the next


Vista TF

iteration. It is assumed that this has no effect on the solution provided that a converged solution forall variables is reached.

12.2.4.6. Initial Estimate

The initial solution generally comprises a first guess of streamline positions (typically by dividing theflow path into equal areas at each quasi-orthogonal) and flow variables at all the grid points (that is,the junction points between the variable streamlines and the fixed quasi-orthogonals). This is thensuccessively refined as a result of each iteration. Once a first estimate of the streamline positions is inplace then various terms in the radial equilibrium equation, such as the curvature of the streamlines orthe rate of change of meridional velocity along the streamlines, can be determined.

12.2.4.7. Radial Equilibrium Equation

The momentum equation on the mean stream surface is a form of the radial equilibrium equation,giving a velocity gradient equation for the gradient of the meridional velocity along a calculating station,or quasi-orthogonal:

This equation is given in many text books (see Cumpsty (1989) [6] (p. 283)) and is not repeated here. Itrelates the velocity gradient to the shape and the current positions of the streamlines, to the orientationof the mean stream surface, and to the flow parameters and their gradients from the previous iteration.In its simplest form, this equation is the well-known simple radial equilibrium equation of turbomachineryflows, which gives the relationship between the radial gradient of the axial velocity and the gradientof the swirl in the radial direction, as follows:

In its general form the radial equilibrium equation takes into account flow gradients, streamline curvature,and radial flows, and allows a general arbitrary orientation of the quasi-orthogonal lines. The key differ-ence between the simple radial equilibrium equation and the more general equation is the inclusionof the streamline curvature terms, and it is these terms that give their name to the method. There areseveral forms of this velocity gradient equation, but the one used here follows the method of Denton(1978) [7] (p. 283), but takes into account the blade force terms as described in chapter 3.4 of the bookby Cumpsty (1989) [6] (p. 283).

12.2.4.8. Combination of Velocity Gradient and Continuity Equations

The velocity gradient equation is solved in combination with a method for finding the correct velocitylevel on the mean streamline that ensures that the velocity along the quasi-orthogonal satisfies thecontinuity equation giving the correct mass flow across the quasi-orthogonal. See Figure 12.4: Continuityand the Velocity Gradient Used in Determining the Meridional Velocity Level (p. 280). The meridionalvelocity on the mean streamline is specified in the innermost iteration, integrated across the flowchannel with the help of the meridional velocity gradient, and then continually updated until the massflow is correct. This requires some care to ensure that the method converges for all mass flows up tothe choking mass flow. In this iteration, the density is also needed in the continuity equation. In factthere are two values of the density that can be used, corresponding to supersonic or subsonic flow. Inthis method, no attempt is made to distinguish these during the innermost iteration; the density is




simply taken as constant at the value from the previous iteration, and automatically takes on the super-sonic or subsonic value as the iteration converges.

Figure 12.4: Continuity and the Velocity Gradient Used in Determining the Meridional Velocity

Level

12.2.4.9. Relaxation Factors

The meridional velocity distribution along the quasi-orthogonal determines the spacing of the streamlinesof the meridional flow, and hence the meridional streamline positions. These positions are continuallyupdated as the program converges. Generally, it is not acceptable to use the newly calculated positionsof the streamlines directly from one iteration to the next, but a damping factor (often much less thanunity) is required to factor the streamline shifts to obtain convergence. A big weakness of streamlinecurvature methods is that the required damping factor becomes very small when the quasi-orthogonallines are closely spaced, and so improved accuracy through more calculating planes causes a large in-crease in calculating time. For this reason, a relatively coarse grid (compared to CFD computations) isused, leading to calculations that take just a few seconds on a modern PC. The problem with closelyspaced quasi-orthogonals is related to the large errors in the estimated streamline curvature when asmall error in the streamline position is present. It is for this reason that many methods do not includeblade internal calculation stations because these automatically decrease the spacing between adjacentquasi-orthogonals.

12.2.4.10. Streamline curvature

The streamline positions can be used to interpolate new blade element data appropriate to their currentlocation and to find the slopes, curvatures, and derivatives of the flow parameters along the streamlines.This data is needed in the radial equilibrium equation (velocity gradient equation; see Radial EquilibriumEquation (p. 279)). The accuracy and stability of streamline curvature methods is related to the predictionof the curvatures of the streamlines. Several different numerical methods for this have been examined.In the current method, the curvatures are calculated with a parabolic approximation through three ad-jacent points along the streamline.

12.2.4.11. Equations for Enthalpy and Angular Momentum

In the region between blade rows, the total enthalpy and angular momentum may be considered tobe convected along the meridional streamlines from the previous station. The entropy is also convectedbut rises due to the additional losses between the calculating stations. In a blade row, the changes inmomentum and enthalpy are calculated from the Euler equation on the assumption that the flow followsthe mean stream surface. The mean blade stream surface is roughly oriented in alignment with thecamber surface of the blade, but methods are needed for finding the true fluid flow direction takinginto account the incidence and deviation of the flow from the mean blade surface. This is generallydealt with using empirical correlations for outlet angle, deviation, or slip factor.


Vista TF

12.2.4.12. Boundary Conditions

At the inlet plane (first quasi-orthogonal) you are generally required to specify the variation of totalpressure, total temperature, and angular momentum or flow angle, together with the gas data and in-formation on the mass flow at the outlet plane. For calculations with choked flows, it is not possible tocalculate with the mass flow, making it necessary to calculate with specified outlet static pressure, suchthat the mass flow is a result of the simulation.

12.2.4.13. Empirical Data

Additional empirical methods are used to provide data for the loss production, for the boundary layerblockage, and for the deviation of the flow direction from the mean blade camber surface, so that theeffect of viscosity can be taken into account. In fact Denton (1978) [7] (p. 283) argues that “in many ap-plications, throughflow calculations are little more than vehicles for inclusion of empiricism in the formof loss, deviation and blockage correlations, and their accuracy is determined by the accuracy of thecorrelations rather than the numerics”. The three main effects of the empirical data are:

• In the equation of state, a change in the entropy leads to a pressure loss for a given value of the totalenthalpy.

• In the continuity equation, the blockage due to the boundary layer displacement effect leads to a highervalue of the meridional velocity.

• In the momentum equation, the deviation of the flow from the blade direction leads to a change in thecalculated swirl velocity.

Thus, although the basic inviscid equation of motion used by the method is inherently incapable ofpredicting entropy rises through the machine, some effects of losses can be included. There are numerouspossible combinations of data for the empirical information, based on various definitions of loss coeffi-cients, dissipation coefficients, efficiencies, and so on, and this leads to the largest source of confusionin the data preparation for throughflow programs, and a large and complex array of branching “IF”statements within a typical throughflow program to deal with the alternatives. In making this program,a single form of loss definition has been included based on the entropy rise.

12.2.4.14. Blade-to-blade Solution

The mean stream surface provides the flow field in the meridional plane through the turbomachine.Generally the engineer also needs information on the blade surface velocity and Mach number distribu-tions, so the meridional solution needs to be combined with a blade-to-blade method to find bladesurface velocities. This may be either a simple approximation or a more accurate solution of the blade-to-blade flow equations. The more accurate methods enable the exact location of the mean streamsurface of the flow to be determined at different blade-to-blade planes, so this can also lead to a furtheriteration in which the mean stream surface is no longer considered as fixed. Such iterative solutionsare often known as S1/S2 solutions (the S1 surface being the blade-to-blade surface and the S2 surfacebeing the meridional surface. See Figure 12.2: S1 and S2 Stream Surfaces in the Theory of Wu(1952) (p. 277)). This technique has generally been replaced these days by fully 3D CFD solutions withmixing planes, but is still of historical interest. The simpler methods generally use a linear approximationfor the velocity variation from suction surface to pressure surface. These can be expected to producesensible results only when a sufficient number of internal blade row calculating stations are includedin the grid, probably 5 as the minimum.




12.2.4.15. Spanwise Mixing

A major shortcoming of the basic streamline curvature method is the neglect of the spanwise transportof angular momentum, energy, and losses in the direction normal to the streamlines. By definition, athroughflow program is based on the assumption that the flow remains in concentric streamtubes asit passes through the turbomachine, and no mass transfer occurs across the meridional streamlineswhich are the streamtube boundaries. In a duct region of a throughflow calculation, enthalpy, angularmomentum (swirl), and entropy are taken to be conserved along the meridional streamlines. In reality,there are several mechanisms that lead to an apparent spanwise transport of fluid relative to flow onthe mean streamlines, as follows:

• Non-axisymmetric blade-to-blade stream surfaces as a result of streamwise vorticity being shed by theblades (stream-surface twist)

• Secondary flows in the end-wall boundary layers and in the blade boundary layers

• Wake momentum transport downstream of blade rows

• Tip clearance flows with tip clearance vortices

• Turbulent diffusion

As a result of not modeling these effects, throughflow programs in which realistic loss levels are specifiedfor the end-wall regions often predict unrealistic profiles of the loss distribution after several stagesbecause there is no mechanism for the high losses generated near the end walls to be mixed out. Thesimplest approach to deal with this problem is to specify unrealistic loss distributions across the spanthat avoid high levels in the end walls. In fact, in preliminary design calculations, this is often adequate.A useful approximation is to specify a mean-line value of loss coefficient or efficiency and to assumethat the entropy generated by the losses is the same on each stream-tube. This approximates a completemixing of the entropy distribution across the span.

Several more sophisticated methods to include more detailed physics of these mixing processes havebeen attempted so that realistic loss distributions can be specified. The common approach is to modelthe spanwise mixing as turbulent diffusion, even though some of the effects are due to deterministicflow features. Even with throughflow programs in which spanwise mixing is incorporated, it is stillgenerally not possible to fully incorporate the very high loss levels close to the end walls. Vista TF includesthese effects as turbulent diffusion across the streamlines.

12.2.4.16. Streamline Curvature Throughflow Theory: Bibliography

Bibliography

[1] G. G. Adkins and L. H. Smith. “Spanwise mixing in axial flow turbomachines”. Trans ASME Journal of

Engineering for Power , Vol. 104. pg. 97-110. 1982.

[2] P. M. Came. “Streamline curvature throughflow analysis”. Proc. First European Turbomachinery Conference,

VDI Berichte 1185. pg. 291. 1995.

[3] M. V. Casey and O. Hugentobler. “The prediction of the performance of an axial compressor stage with

variable stagger stator vanes”. VDI Berichte Nr. 706. pg. 213-227. 1988.

[4] M. V. Casey and P. Roth. “A streamline curvature throughflow method for radial turbocompressors”. I.

Mech. E. Conference C57/84. 1984.


Vista TF

[5] M. V. Casey and C. J. Robinson. “A new streamline curvature throughflow code for radial turbomachinery”.ASME TURBOEXPO 2008, ASME GT2008-50187. Berlin. 2008.

[6] N. A. Cumpsty. Compressor aerodynamics. Longman Scientific. New York. 1989.

[7] J. D. Denton. “Throughflow calculations for axial flow turbines”. Trans ASME, Journal of Engineering for

Power, Vol 100. 1978.

[8] J. D. Denton and C. Hirsch. “Throughflow calculations in axial turbomachines”. AGARD Advisory Report

No. 175 AGARD-AR-175. 1981.

[9] S. J. Gallimore. “Spanwise mixing in multistage axial flow compressors: part II throughflow calculations

including mixing”. Trans. ASME, Journal of turbomachinery, Vol. 108. pg. 10-16. 1986.

[10] I. K. Jennions and P. Stow. “The quasi-three-dimensional turbomachinery blade design system, Part I:

Throughflow analysis, Part II: Computerized system”. Transactions of the ASME, Journal of Engineering

for Gas Turbines and Power Vol. 107. pg. 308-16. 1985.

[11] I. K. Jennions and P. Stow. “The importance of circumferential non-uniformities in a passage averaged

quasi-three-dimensional turbomachinery design system”. Transactions of the ASME, Journal of En-

gineering for Gas Turbines and Power Vol. 108. pg. 240-5. 1986.

[12] K. I. Lewis. “Spanwise transport in axial-flow turbines: Part 2 - Throughflow calculations including

spanwise mixing”. Trans. ASME, Journal of turbomachinery, Vol. 116. pg. 187-193. 1984.

[13] B. Liu, S. Chen, and H. F. Martin. “A primary variable throughflow code and its application to last stage

reverse flow in LP steam turbine”. Paper: IJPGC2000-15010, Proc. Int. Joint Power GenerationConference, July 23-26. Miami Beach, Florida. 2000.

[14] H. Marsh. “A digital computer program for the through-flow fluid mechanics in an arbitrary turbomachine

using a matrix method”. Aeronautical Research Council R and M 3509. 1968.

[15] C. Hirsch and J. D. Denton. “Throughflow calculations in axial turbomachines”. AGARD Advisory report

No. 175, AGARD-AR-175. 1981.

[16] K. I. Lewis. “Spanwise transport in axial flow turbines: part 2 - throughflow calculations including

spanwise transport”. Trans. of ASME, Journal of Turbomachinery, Vol. 116. pg. 187-193. 1994.

[17] R. A. Novak. “Streamline curvature computing procedures for fluid flow problems”. Trans. of ASME,

Journal of Engineering for power, Vol 89. pg. 478-490. 1967.

[18] M. Schobeiri. Turbomachinery Flow Physics and Dynamic Performance. Springer. Berlin. 2005.

[19] L. H. Smith. “The radial equilibrium equation of turbomachinery”. Trans. ASME Journal of Engineeringfor Power, Vol 88. pg. 1-12. 1966.

[20] D. H. Wilkinson. “Streamline curvature methods for calculating the flow in turbomachines”. Report No

W/M(3F). English Electric. Whetstone, England. pg. 1591. 1969.

[21] D. H. Wilkinson. “Stability, convergence, and accuracy of two-dimensional streamline curvature methods

using quasi-orthogonals”. paper 35, I.Mech.E. Convention. Glasgow. 1970.

[22] C. H. Wu. “A general theory of three-dimensional flow in subsonic, and supersonic turbomachines of

axial, radial and mixed-flow types”. Trans. ASME, Nov. 1952. pg. 1363-1380. 1952.




[23] C. C. Yeoh and J. B. Young. “Non-equilibrium throughflow analyses of low-pressure, wet steam turbines”.ASME J. Eng. for Gas Turbines & Power, Vol 106. pg. 716-724. 1984.

12.2.5. Appendices

The following appendix topics are available:12.2.5.1. Appendix A: A Note on Sign Convention for Angles and Velocities in Vista TF12.2.5.2. Appendix B: Example of a Control Data File (*.con)12.2.5.3. Appendix C: Example of a Geometry Data File (*.geo) for a Radial Impeller12.2.5.4. Appendix D: Example of an Aerodynamic Data File (*.aer)12.2.5.5. Appendix E: Examples of Correlations Data Files (*.cor)12.2.5.6. Appendix F:Troubleshooting12.2.5.7. Appendix G:The RTZTtoGEO Program12.2.5.8. Appendix H: Example of a Real Gas Property Data File (*.rgp)

12.2.5.1. Appendix A: A Note on Sign Convention for Angles and Velocities in Vista TF

When looking along the axis of the machine in the direction of flow we have a cylindrical coordinatesystem (r, theta, z) or (r, θ, z). The positive θ direction is the clockwise direction when viewed along themachine axis from the inlet. The angular coordinate (theta) is taken as positive in the clockwise directionand negative in the anti-clockwise direction. The normal case considered by Vista TF if the speed ofrotation is specified as a positive value is clockwise rotation when looking along the axis in the directionof the flow so that the angular coordinate increases in the rotational direction. See Figure 12.5: CoordinateSystem used by Vista TF (p. 284) below.

Figure 12.5: Coordinate System used by Vista TF


Vista TF

Blade lean angles are then defined from the axial and radial directions in a similar way using the positivetheta direction as a basis. The rate of change of the angular coordinate with the meridional directionthen defines the sign of the blade angle, and this applies for the blade angles, flow angles and alsoapplies to the blade lean angles; see below. As you move along an axial rotor blade in the axial directionof a clockwise rotating machine then the blade wrap angle theta steadily decreases as the blade issloped backwards against the rotational direction. This also applies for radial impeller outlet angles,which are generally also backswept, both compressors and turbines. In a stator vane the angular co-ordinate increases as you step along the blade in the meridional direction.

Figure 12.6: Sign Convention for Blade Angles in Vista TF

For example, in an axial compressor rotor with clockwise rotation direction where the blades lean backfrom the direction of rotation, theta becomes more negative as you move along the blade from theleading edge (LE) to the trailing edge (TE). In a compressor stator, theta becomes more positive as youmove along the blade. This rule works for axial, radial, and mixed flow compressors, provided the meri-dional direction is used as a basis. In a turbine stator theta also increases positively from LE to TE and,in a turbine rotor, theta decreases from LE to TE. There are some exceptions to this rule, related to highcamber at leading edges to adapt the flow to the incoming flow direction, as shown in Figure 12.6: SignConvention for Blade Angles in Vista TF (p. 285) for a turbine stator. This can also occur in blades withleading edge recamber in ventilator blades in channels of high curvature.

Figure 12.7: Sign Convention for Angles in Vista TF (p. 286) shows the angle definitions used in Vista TF;these are consistent with the blade angle definitions. The sign convention for velocity values is that anaxial velocity component in the positive axial direction is positive, a radial component in the positiveradial direction is positive, and a circumferential component in the positive angular direction (theta) ispositive. Note that the rotational speed has an associated sign, so that clockwise rotation is a positiverotational speed and anti-clockwise rotation is negative.




Figure 12.7: Sign Convention for Angles in Vista TF

As examples of this notation, Figure 12.8: Sign Convention for Blade and Flow Angles in Vista TF for aClockwise Turbine (p. 286) and Figure 12.9: Sign Convention for Blade and Flow Angles in Vista TF for aClockwise Compressor (p. 287) show the velocity triangles with blade angles and flow angles for com-pressors and turbines rotating in the clockwise rotational sense.

Figure 12.8: Sign Convention for Blade and Flow Angles in Vista TF for a Clockwise Turbine


Vista TF

Figure 12.9: Sign Convention for Blade and Flow Angles in Vista TF for a Clockwise Compressor

To demonstrate that this notation can be used for anti-clockwise rotation and for counter-rotating bladerows, Figure 12.10: Sign Convention for Blade and Flow Angles in Vista TF for an Anti-clockwise Tur-bine (p. 287), Figure 12.11: Sign Convention for Blade and Flow Angles in Vista TF for an Anti-clockwiseCompressor (p. 288), and Figure 12.12: Sign Convention for Blade and Flow Angles in Vista TF for aCounter-rotating Compressor (p. 289) show these cases.

Figure 12.10: Sign Convention for Blade and Flow Angles in Vista TF for an Anti-clockwise Turbine




Figure 12.11: Sign Convention for Blade and Flow Angles in Vista TF for an Anti-clockwise

Compressor

Note that the rule for blade angles is applied independently of the rotational direction of the bladerows. For example, if the blade rotational speed is defined as negative then the blade angles of acompressor rotor would be positive as the angle theta increases in the flow direction, as shown in Fig-ure 12.11: Sign Convention for Blade and Flow Angles in Vista TF for an Anti-clockwise Compressor (p. 288).

Figure 12.12: Sign Convention for Blade and Flow Angles in Vista TF for a Counter-rotating Com-pressor (p. 289) shows an example of a counter-rotating compressor with two rotors, the second rotatingin the reverse direction.

Note

Vista TF uses the rules outlined above to identify the type of machine from the geometricaldata of the blade angles in the geometry input file, so you do not have to specify this in theinput data.


Vista TF

Figure 12.12: Sign Convention for Blade and Flow Angles in Vista TF for a Counter-rotating

Compressor

12.2.5.1.1. Definition of Blade Lean Angles

Consider a point q on a radial impeller blade camber surface, as shown in Figure 12.13: Definition of

Angles (p. 290). We can define the blade lean angles, and as the angle made by the blade fiberswith a radial line and the angle made with an axial line, as follows:

In order to assist the understanding of the angles, , , and ε, the analogy to the angles known asyaw, roll, and pitch of a sailing boat might be useful.




Figure 12.13: Definition of Angles

Notes on :

• This is the blade lean to the radial direction.

• This angle is zero for a blade comprising purely radial blade elements, so is generally close to zero foraxial blade rotor rows (which tend to have no lean) and centrifugal impeller leading edges or radial turbinetrailing edges.

• It is non-zero for blades with lean. If, when moving along the blade in the radial direction, the lean isagainst the direction of rotation then it is negative and lean in the direction of rotation is positive.

• A radial compressor impeller with a purely radial blade at impeller outlet (no backsweep) would have

=0° and corresponds to the back-sweep angle in a typical back-swept radial impeller. For an impeller

with 30° of back-sweep, would be negative (-30°) at the trailing edge. Note that, in some sign conven-tions, the back-sweep angle would be given a positive value, but in Vista TF, it is negative.

• In a the diffuser of a radial pump or compressor stage, this angle corresponds to the diffuser blade angleand would be typically between 60° and 70° at the leading edge, and would decrease through the bladerow.

Notes on :

• This is the blade angle measured from the axial direction in the direction of rotation.

• In an axial blade element (or close to the leading edge of a typical radial impeller or the trailing edge ofa radial turbine impeller), this is the blade camber angle. At the inlet to an axial blade row, this is theblade inlet angle; at the outlet, it is the blade outlet angle.

• In a typical rotor blade, it is negative and, in a typical stator, it is positive. Note that the common exceptionswould be turbine rotors with a low degree of reaction and compressor rotor roots with very high turning.


Vista TF

• In a compressor where the flow is turned by the blade rows towards the axial direction, the absolute valuesdecrease from the leading edge to the trailing edge whereas in a turbine where more swirl is added tothe flow, the absolute values increase from the leading edge to the trailing edge.

• For a blade with purely axial blade elements (as in a typical 2D diffuser and at a radial impeller outlet withno rake angle), this angle is zero.

12.2.5.1.2. Definition of Meridional Streamline Inclination Angle or Pitch Angle

The angle is the inclination of a meridional streamline to the axial direction:

which on the hub and casing walls becomes the meridional slope angle of the walls.

12.2.5.1.3. Definition of Blade Angle

Internally the program uses the angles and to calculate , which is the effective blade angle

measured from the meridional direction. is defined as:

Notes on :

• The wrap angle is taken as positive in the direction of rotation. This is consistent with the flow angledefinition used by the program.

• This angle is zero for a blade in which the wrap angle does not change in the meridional direction, such

as an axial strut in an axial channel or a radial strut in a radial channel, where .

• If the wrap angle increases in the meridional direction then the blade angle is positive. If the wrap angledecreases in the meridional direction then the blade angle is negative.

• Note that the actual value of the blade angle as seen by the flow is dependent on the lean angle of themeridional streamline, and is not an absolute fixed value for a certain blade. Thus a change in the meridi-onal direction of the flow at an impeller outlet with lean changes the value of the blade angle.

12.2.5.2. Appendix B: Example of a Control Data File (*.con)

The following is a described example of a control data file:

Three lines to identify the run (maximum 72 characters per line):




PCA stageControl parameters for typical radial impeller5 May 2008

Two lines for the integer control parameters:

n_sl max_it_main max_it_mass9 500 10

Two lines for integer control parameters for output data and the use of the restart file:

i_print_plane i_print_level i_progress i_display i_restart i_interface 4 5 0 0 0 0

Two lines for integer control parameters for various models and reference parameters:

i_expert i_flow i_fluid i_inbc i_mass i_mix i_ree 0 1 0 0 0 0 0

Two lines for convergence and damping factors:

damp_sc damp_vl cm_start tolerance_cm tolerance_mass grad_ree 0.25 0.25 0.2 0.01 0.005 1.00

12.2.5.3. Appendix C: Example of a Geometry Data File (*.geo) for a Radial Impeller

The following is a described example of a geometry data file:

Three lines to identify the run:

PCA Stage from .rtzt fileThickness converted to tangential thickness10 blades and 1 splitter

Two lines for the number of quasi-orthogonal lines:

n_qo scale23 1.00000

A section that defines the quasi-orthogonal type and end points:

i r_hub r_shr z_hub z_shr n_blade n_curve i_type i_row i_spool1 0.00069 0.05594 -0.04959 -0.04959 0 1 1 0 02 0.00224 0.05215 -0.03682 -0.03757 0 1 1 0 03 0.00702 0.04941 -0.02481 -0.02527 0 1 1 0 04 0.01189 0.04776 -0.01284 -0.01277 0 1 1 0 05 0.01357 0.04720 -0.00004 -0.00018 10 5 3 0 06 0.01447 0.04728 0.00570 0.00371 10 5 3 0 07 0.01609 0.04758 0.01128 0.00759 10 5 3 0 08 0.01832 0.04809 0.01664 0.01145 10 5 3 0 09 0.02108 0.04884 0.02175 0.01526 20 5 3 0 010 0.02463 0.04993 0.02700 0.01930 20 5 3 0 011 0.02867 0.05132 0.03188 0.02325 20 5 3 0 012 0.03314 0.05305 0.03637 0.02706 20 5 3 0 013 0.03799 0.05513 0.04045 0.03068 20 5 3 0 014 0.04318 0.05759 0.04408 0.03407 20 5 3 0 015 0.04866 0.06040 0.04726 0.03717 20 5 3 0 016 0.05440 0.06358 0.04995 0.03988 20 5 3 0 017 0.06035 0.06709 0.05214 0.04216 20 5 3 0 018 0.06646 0.07087 0.05379 0.04393 20 5 3 0 019 0.07271 0.07488 0.05485 0.04514 20 5 3 0 020 0.07902 0.07901 0.05538 0.04575 20 5 3 0 021 0.08033 0.08032 0.05544 0.04585 0 1 1 0 022 0.08164 0.08164 0.05548 0.04592 0 1 1 0 023 0.08295 0.08295 0.05550 0.04597 0 1 1 0 0

A section that defines the quasi-orthogonals and the blade geometry:


Vista TF

i j r_qo theta_qo z_qo thu_qo gamma_r_qo gamma_z_qo5 1 0.01357 0.32039 -0.00004 0.00255 -0.18038 -23.156545 2 0.02198 0.26684 -0.00008 0.00244 -0.36249 -32.520255 3 0.03038 0.21411 -0.00015 0.00251 -0.34266 -41.271895 4 0.03879 0.26814 -0.00016 0.00243 -0.00979 -49.732305 5 0.04720 0.29630 -0.00018 0.00234 -0.18913 -57.432246 1 0.01447 -9.69106 0.00570 0.00269 1.88713 -23.089476 2 0.02267 -8.14852 0.00518 0.00256 0.41685 -31.316986 3 0.03087 -7.40827 0.00464 0.00243 -0.99339 -39.399576 4 0.03908 -6.96808 0.00417 0.00230 -2.04619 -47.449926 5 0.04728 -6.75726 0.00371 0.00218 -3.95792 -54.902817 1 0.01609 -18.29413 0.01128 0.00262 3.70754 -22.858217 2 0.02396 -15.68568 0.01033 0.00248 1.15036 -30.472417 3 0.03183 -14.35239 0.00936 0.00234 -1.75718 -37.836967 4 0.03970 -13.58089 0.00846 0.00220 -3.70752 -45.357647 5 0.04758 -13.16589 0.00759 0.00206 -6.69381 -52.398178 1 0.01832 -25.28463 0.01664 0.00254 4.70103 -22.816468 2 0.02576 -22.28703 0.01531 0.00241 1.41759 -29.881468 3 0.03320 -20.61659 0.01398 0.00227 -2.35690 -36.526128 4 0.04065 -19.60303 0.01270 0.00212 -5.03581 -43.434548 5 0.04809 -18.99898 0.01145 0.00196 -8.73336 -50.000399 1 0.02108 -30.92988 0.02175 0.00248 4.74907 -23.146109 2 0.02802 -28.02047 0.02011 0.00234 1.11611 -29.563309 3 0.03496 -26.23881 0.01846 0.00220 -2.97754 -35.524979 4 0.04190 -25.07781 0.01685 0.00204 -6.21556 -41.740889 5 0.04884 -24.32487 0.01526 0.00188 -10.40615 -47.7916010 1 0.02463 -36.00322 0.02700 0.00242 3.85677 -23.9725410 2 0.03096 -33.42982 0.02507 0.00228 0.15779 -29.5665810 3 0.03728 -31.67995 0.02313 0.00213 -3.84908 -34.8360010 4 0.04361 -30.43321 0.02121 0.00197 -7.49971 -40.2443210 5 0.04993 -29.55790 0.01930 0.00182 -12.07824 -45.6758611 1 0.02867 -40.35207 0.03188 0.00236 2.23718 -25.2645011 2 0.03435 -38.17139 0.02974 0.00222 -1.42539 -29.8948111 3 0.04001 -36.55263 0.02757 0.00207 -5.15749 -34.4920411 4 0.04567 -35.29039 0.02542 0.00192 -8.89269 -39.1210211 5 0.05132 -34.35094 0.02325 0.00177 -13.85353 -43.8584112 1 0.03314 -44.26280 0.03637 0.00232 0.10221 -26.9747412 2 0.03816 -42.42247 0.03407 0.00217 -3.48507 -30.5620812 3 0.04314 -40.96080 0.03175 0.00203 -6.83731 -34.5293212 4 0.04811 -39.72235 0.02942 0.00188 -10.45941 -38.4080012 5 0.05305 -38.76700 0.02706 0.00173 -15.90754 -42.2855813 1 0.03799 -47.91006 0.04045 0.00228 -2.40966 -29.0783213 2 0.04233 -46.32055 0.03805 0.00214 -5.94640 -31.5610413 3 0.04663 -45.00031 0.03562 0.00200 -8.93117 -34.8894713 4 0.05090 -43.80060 0.03317 0.00185 -12.32638 -38.0738013 5 0.05513 -42.85755 0.03068 0.00171 -18.25871 -40.8899714 1 0.04318 -51.39712 0.04408 0.00226 -5.26535 -31.5263414 2 0.04685 -49.96856 0.04163 0.00212 -8.83099 -32.8249614 3 0.05047 -48.75090 0.03914 0.00198 -11.46588 -35.5458514 4 0.05405 -47.58476 0.03663 0.00184 -14.51275 -38.1537414 5 0.05759 -46.66554 0.03407 0.00170 -20.92644 -39.6154915 1 0.04866 -54.78108 0.04726 0.00224 -8.53323 -34.2014915 2 0.05168 -53.43857 0.04478 0.00211 -12.12520 -34.3099915 3 0.05464 -52.27642 0.04227 0.00197 -14.44340 -36.4535115 4 0.05755 -51.13094 0.03975 0.00184 -17.13253 -38.5572715 5 0.06040 -50.22395 0.03717 0.00170 -23.67359 -38.4039716 1 0.05440 -58.09399 0.04995 0.00224 -12.36035 -36.9130616 2 0.05678 -56.77929 0.04747 0.00211 -15.80725 -36.0202016 3 0.05910 -55.62589 0.04497 0.00198 -17.90732 -37.5502516 4 0.06137 -54.47843 0.04245 0.00185 -20.39385 -39.1142516 5 0.06358 -53.55910 0.03988 0.00172 -26.48486 -37.3714817 1 0.06035 -61.35160 0.05214 0.00227 -16.87176 -39.3144017 2 0.06211 -60.02598 0.04967 0.00214 -19.92706 -37.8809917 3 0.06382 -58.83913 0.04719 0.00201 -21.91242 -38.6909817 4 0.06548 -57.66538 0.04469 0.00188 -24.33144 -39.5717717 5 0.06709 -56.69062 0.04216 0.00175 -29.09953 -36.7604718 1 0.06646 -64.56377 0.05379 0.00231 -22.10267 -40.9656618 2 0.06763 -63.20368 0.05134 0.00220 -24.62539 -39.6160118 3 0.06875 -61.94636 0.04888 0.00207 -26.48127 -39.6530418 4 0.06983 -60.72039 0.04641 0.00195 -28.81198 -39.7095018 5 0.07087 -59.63480 0.04393 0.00181 -31.57298 -37.0659219 1 0.07271 -67.72621 0.05485 0.00241 -27.77629 -41.34374




19 2 0.07330 -66.32376 0.05243 0.00231 -29.91973 -40.6214119 3 0.07384 -64.96857 0.05000 0.00218 -31.51785 -40.0610619 4 0.07437 -63.66663 0.04757 0.00202 -33.54181 -39.5478319 5 0.07488 -62.40934 0.04514 0.00194 -34.26011 -38.7280020 1 0.07902 -70.87669 0.05538 0.00263 -32.69635 -39.9227920 2 0.07904 -69.41995 0.05297 0.00260 -34.43144 -39.8309720 3 0.07902 -67.94505 0.05056 0.00239 -36.32393 -39.9147020 4 0.07901 -66.52396 0.04815 0.00205 -37.25527 -39.7831820 5 0.07901 -65.01019 0.04575 0.00216 -37.35707 -41.84796

A section for additional geometry data:

j throat throat_pos clearance max_thick te_thick dummy1 dummy2 dummy31 0.00000 0.00000 0.00000 0.00000 0.00000 0.00 0.00 0.002 0.00000 0.00000 0.00000 0.00000 0.00000 0.00 0.00 0.003 0.00000 0.00000 0.00000 0.00000 0.00000 0.00 0.00 0.004 0.00000 0.00000 0.00000 0.00000 0.00000 0.00 0.00 0.005 0.00000 0.00000 0.00000 0.00000 0.00000 0.00 0.00 0.00

12.2.5.4. Appendix D: Example of an Aerodynamic Data File (*.aer)

The following is a described example of an aerodynamic data file:

Three lines to identify the aerodynamic data:

Radial stage at design flowideal gas5 May 2008

Two lines for the reference aerodynamic parameters, which depend on the value of i_flow in the.con file. Assuming i_flow = 1:

ref_n ref_mass ref_d60000.0 1.00 0.15804

Two lines for the Reynolds number:

Re_ref0.0

Two lines for the fluid data, which depends on the value of i_fluid in the .con file. Assumingi_fluid = 1 (for ideal gas):

cp_gas gamma_gas1005.2 1.40

Two lines for the number of points on the inlet boundary where flow conditions are specified:

n_inbc1

Lines for the fraction of the inlet boundary where flow conditions are specified (n_inbc values):

f_inbc0.0

Lines for the pressure on the inlet boundary (n_inbc values, which depend on i_bc in the .con file).Assuming i_bc = 0:

pt_inbc98000.0

Lines for the temperature on the inlet boundary (n_inbc values, which depend on i_bc in the .confile). Assuming i_bc = 0:


Vista TF

tt_inbc293.0

Lines for swirl on the inlet boundary (n_inbc values, which depend on i_bc in the .con file). Assumingi_bc = 0:

rcu_inbc0.00

Two lines for the mixing model and blade transparency model parameters:

eddy f_bl_le f_bl_te0.0001 0.0 0.0

12.2.5.5. Appendix E: Examples of Correlations Data Files (*.cor)

EXAMPLE 1

The following example of a correlations data file is for a radial compressor. For most radial compressoror pump impeller calculations, the only parameter that changes is the small-scale polytropic efficiency.Note that a single value of the empirical information for flow outlet angle is specified (k_dev = 3 impliesthe use of a slip factor, and because k_dev is specified as 0.0 the program calculates the slip factorfrom the Wiesner correlation).


Typical correlations file for radial compressorsmall-scale efficiency, slip factor of 0.0 (so Wiesner) and no blockage5 May 2008

Two lines for integer data:

i_loss i_dev i_ewb1 1 0

Two lines for loss input data:

n_loss_sl n_loss_qo1 1

Two lines for empirical loss data:

i_qo_loss k_loss f_loss loss1 1 0.500 0.86

Two lines for deviation input data:

n_dev_sl n_dev_qo1 1

Two lines for empirical deviation data:

i_qo_dev k_dev f_dev dev1 3 0.500 0.00

Two lines for blockage input data:

n_ewb_sl n_ewb_qo1 1

Two lines for empirical blockage data:

i_qo_ewb k_ewb f_ewb ewb1 1 0.500 0.00




EXAMPLE 2

The following example of a correlations data file is for a nine-stage radial compressor with a vanelessdiffuser and a return channel downstream of each impeller except the ninth. The losses in this case arespecified as a single polytropic efficiency for everything, and the blockage is zero. But for the deviationof the flow from the blade angle at the trailing edge it is sensible to use a slip factor of 100 for theimpeller and a deviation angle for the stator vanes. In this case 0.0 is specified as the slip factor so theWiesner slip factor is used, and 5.00° is specified for the return channel deviation angle. k_dev controlswhether to use the slip factor or deviation, and i_qo specifies the calculating station at which the slipfactor or deviation is applied. A sensible rule would be to make the value of i_qo equal to the numberof the q-o at the trailing edge, but the program can accept any value between the previous trailingedge and the current one. If there is a vaned diffuser with a different deviation angle then an additionalline with k_dev set to 1 would be needed between each impeller and return channel line.


Typical correlations file for multistage radial compressorEfficiency, slip factor in impeller, deviation in stators, no blockage18 September 2008





Two lines for empirical loss data:

i_qo_loss k_loss f_loss loss1 1 0.500 0.80



Lines for empirical deviation data:

i_qo_dev k_dev f_dev dev1 3 0.500 0.0045 1 0.500 5.0065 3 0.500 0.095 1 0.500 5.00115 3 0.500 0.0145 1 0.500 5.00165 3 0.500 0.0195 1 0.500 5.00215 3 0.500 0.0245 1 0.500 5.00265 3 0.500 0.0295 1 0.500 5.00315 3 0.500 0.0345 1 0.500 5.00365 3 0.500 0.0395 1 0.500 5.00415 3 0.500 0.0




Vista TF



EXAMPLE 3

The following example of a correlations data file is for a radial turbine stage with an inlet nozzle andan impeller. In the example shown, the losses are different for the two components. The deviation angleis specified by use of the cosine rule for both components (k_dev = 5) but a correction to this (2.00°)is applied in the rotor.


Typical correlations file for radial turbine stageEfficiency, deviation as mod to cosine rule and no blockage18 September 2008





Three lines for empirical loss data:

i_qo_loss k_loss f_loss loss1 1 0.500 0.8835 1 0.500 0.75




i_qo_dev k_dev f_dev dev1 5 0.500 0.0035 5 0.500 2.00





EXAMPLE 4

In all of the above examples, a constant value of the empirical information is applied on all streamlines,so only a single value is specified. An additional feature is that the program also allows this data to varyacross the span for each blade row. Then there would be an additional line for each fraction of thespan. This is shown with regard to the losses, which specifies the efficiency at 0%, 25%, 50%, 75% and100% of the span.





Radial compressor impeller with spanwise variation of etaEfficiency, slip factor for impeller with no blockage18 September 2008





Lines for empirical loss data:

i_qo_loss k_loss f_loss loss1 1 0.000 0.771 1 0.250 0.791 1 0.500 0.801 1 0.750 0.791 1 1.000 0.77




i_qo_dev k_dev f_dev dev1 3 0.500 0.00





EXAMPLE 5

The following example of a correlations data file shows the use of a blade oriented specification for theKacker Okapuu correlations for an axial turbine calculation. No blockage is specified.


Typical correlations file for axial turbine - no blockageKacker-Okapuu10 September 2008




n_loss_bladerow dummy1 1

Two lines for empirical deviation data (i_loss = 2):


Vista TF

i_blrow i_loss_type f_loss(1) f_loss(2) f_loss(3) f_loss(4) f_loss(5) f_loss(6)1 1 1.0 1.0 1.0 1.0 1.0 1.0


n_dev_bladerow dummy1 1


i_blrow i_dev_type f_dev(1) f_dev(2) f_dev(3) f_dev(4) f_dev(5) f_dev(6)1 1 0.0 0.0 0.0 0.0 0.0 0.0




i_qo k_ewb f_ewb ewb1 1 0.500 0.00

EXAMPLE 6

The following example of a correlations data file shows the use of a blade oriented specification for theMiller-Wright correlations for an axial compressor calculation.


Typical correlations file for multistage axial compressorMiller Wright Correlations10 September 2008




n_loss_bladerow dummy1 1

Two lines for empirical loss data (i_loss = 2):

i_blrow i_loss_type f_loss(1) f_loss(2) f_loss(3) f_loss(4) f_loss(5) f_loss(6)1 11 1.0 1.0 1.0 1.0 1.0 1.0

Two lines for deviation input data (i_dev = 2):

n_dev_bladerow dummy1 1

Lines for empirical deviation data (i_dev = 2):

i_blrow i_dev_type f_dev(1) f_dev(2) f_dev(3) f_dev(4) f_dev(5) f_dev(6)1 11 0.0 0.0 0.0 0.0 0.0 0.0


n_ewb_bladerow dummy1 1





i_blrow i_ewb_type f_ewb(1) ... ... .... ... f_ewb(6)1 11 1.0 1.0 1.0 1.0 1.0 1.0

12.2.5.6. Appendix F: Troubleshooting

The Vista TF program and numerical method is extremely robust, especially in unchoked flows andwhen running from an existing restart file. There should generally be no numerical problems with radialcompressor, radial turbine, and radial pump calculations, which are the first applications for which theprogram has been released. Nevertheless convergence problems and unexpected program errors canstill occur. The most common problems are format errors in the input data or numerical problems in-herent to the streamline curvature method or associated with choking. The numerical errors cansometimes result in a failure of the solution to converge or even a complete breakdown as the compiledprogram encounters a FORTRAN error, such as the square root of a negative number, or a floating pointoverflow. The most common of these possible errors are trapped and reported by the program. It ishoped that all such errors can be removed in a later version. The compiler of course will report theseerrors, but this is not of much use to the user. The most likely cause of such errors is an input data error.The sections below outline typical problems and suggest solutions for these with some advice on howto deal with input format and data errors.

If the solution converges to the user-specified level of convergence then the program exits normallywith a value of 0 in the command line (CALL EXIT(0)). If the solution has not converged but the maximumnumber of iterations has been reached, then a value of 1 is returned. If the program has identified someserious difficulty which has an appropriate trap, then the program exits with a negative return valueand prints an appropriate error message. Only the most common errors have been supplied with a trapagainst such errors. If the program identifies a feature of the input data which it determines may leadto some difficulties (rotational direction, grid aspect ratio, inconsistent flow and speed data) it prints awarning to the screen and the output file. This information may help to correct the data errors.

Some examples of warnings and error messages are as follows:

WARNING: Anti-clockwise rotation

Vista TF suspects that the first rotor blade is counter-rotating

Possible cures:

(a)Change value of i_flow to be negative (b) Change geometry from left-handed to right-handed or vice versa

WARNING: Wilkinson damping factor used Damping factor damp_sc reduced by Vista TF to value suggested by Wilkinson

WARNING: error increasing -relaxation factors reduced 5%

ERROR: Error noted in subroutine throughflow_manager

ERROR TYPE: error_max > 400%

ERROR PARAMETERS: -10 10 17 139 0 0

0.2100E+16 0.5600E+00 0.7250E+00 0.2905E+01 0.0000E+00

Problem: Near to choke on quasi-orthogonal: 17 on iteration: 139Inlet mass flow: 0.56000Local estimated choke mass flow: 0.72497Local effective relative Mach number: 2.90494Possible cures:

(a)


Vista TF

Remove this q-o from domain (b) Check consistency of flow and speed

12.2.5.6.1. Input-output Errors

If the program fails to run at all, then this is usually a sign that some of the formatted input data is notcorrect (no empty lines where these were expected, or an empty line where data was expected, orsimilar formatting problems). The screen monitor information then says that the file has broken downin a particular input subroutine; this provides a clue for identifying which file has an incorrect format.Most straightforward errors caused by new users or new cases are related to the data input files, whichare fairly rigid with regards to the lines of data and which do not accept an empty line where a line ofdata should be. No built-in consistency checks of the input data are made, so errors of this nature caneasily occur, especially at the beginning of a completely new calculation. To help avoid this type of error,you can copy lines from similar existing input files.

The program might report the name of the input file that contains an error. Should such errors happen,you should examine the .hst file where the input data is recorded. From this, you can determine whichfile has not been correctly read or might have errors, and also which part of that file has been successfullyread.

If the program reads all the input files, starts successfully but then fails before the iterations begin, thenthis is often a problem related to the specified flow and speed conditions in the input data or thegeometry specification, and these should be checked. These errors are usually related to errors in theunits of the specified data (flow conditions, boundary conditions, geometry, empirical data, and so on).A typical user error of this type is that the diameter is needed as a reference value for the size in theflow information (.aer), but the user specifies a radius because the coordinates in the geometry file(.geo) are radius values. Another common error with users is to specify the geometry in inches ratherthan meters. Another typical user error of this type is that the pressure is specified in bars whereas it

is expected to be in N/m2. Other errors may be related to the fact that the specified flow conditionsimply choked flow or reverse flow. A useful consistency check of the flow data is made where the fol-lowing information is printed:

Vista TF: Axial compressor calculation--------------------------------------Estimated mean flow coefficient cm/u at first rotor inlet = 0.4567Estimated mean Mach number cm/a at first rotor inlet = 0.4426

These values may be used to identify if the mass flow, speed and other data have been specified sensibly,as typical values of these parameters for each type of machine will be known.

If the program runs, but breaks down after only a few iterations, then this can often be a sign that someof the input data is still not correct because the program is generally very reliable. Here the strategy isto reduce the value of the maximum number of iterations to a lower value than that at which the flowbreaks down (reduce the value of the parameter max_it_main in file .con ), to rerun the case, andthen to examine the results in the .hst and .out files. The errors are usually related to mistakes inthe specified data (such as flow conditions, boundary conditions, geometry, and empirical data) ratherthan mistakes in the numerical method, and these data problems can then often be identified from un-converged results or values provided in the .hst , .txt and .out files. It can also be useful at thisstage to examine the plot files of the initial geometry set up by the program, which is namedtest_prefix.txt or test_prefix.csv . This can often identify aspects of the geometry or flowdata that are inconsistent.

If the program runs, but the results are extremely unexpected, such that a pump impeller or a compressorstator blade row has been interpreted by the program as a turbine row, then the blade angle definition




should be checked. The program attempts to identify the type of blade row from the geometry specified,but in some cases the rules that have been programmed may fail. For example, a compressor stator isidentified as a stator blade row which turns the absolute flow towards the meridional direction, that isthe blade angle decreases from the leading edge to the trailing edge through the blade row. In somespecial cases, such as an axial compressor stator with a degree of reaction above 100% or rotor bladeswith extensive local leading edge recamber, the blade may have other blade forms. In these cases it ispossible to specify additional data in the geometry file (see parameter i_row below) to overwrite theprogram’s own attempts at blade type identification. Geometry produced by BladeEditor contains bladetype information provided that, in the blade feature within BladeEditor, a blade type (rotor or stator)is specified.

12.2.5.6.2. Convergence

The program has several different internal convergence checks, but a converged solution can be achievedonly if all of the internal loops also converge, so only one convergence criterion really needs to bechecked, which is the convergence of the meridional velocity.

The most important convergence check is the maximum error of the local meridional velocity anywherein the flow field, expressed as a percentage of the local meridional velocity and denoted as error_cm(%) on the screen and in the output files. During the iteration process, the maximum value of thechange in the meridional velocity in the whole flowfield between one iteration and the next (delta_cm%) and the location of the maximum error is printed onto the screen every 10 iterations and into thehistory file for each iteration. The local value of this error is also printed in the output file and into the.csv and .txt plot files. To achieve convergence near machine accuracy, use a maximum value of0.01% as a convergence criterion. Note that this corresponds to a maximum residual error in CFD ter-minology of 0.0001, that is 1E-04, which is a much more stringent criterion than the RMS residual erroroften used in CFD programs of 1E-04. The value of the local error is output for examination with plotsoftware and, for cases that have not converged, it is worthwhile to examine the location of the max-imum as this may identify the location of the problem.

It is important to remember, however, that numerical methods have many sources of error. In athroughflow calculation, the so-called model errors, related to the fact that the equations we are solvingdo not really describe the real flow particularly adequately (in this case we solve for inviscid, circumfer-entially-averaged mean values on widely spaced grid lines) probably outweigh all other sources of error.A solution that is converged to a maximum error in meridional velocity of 0.5% is likely no worse interms of its agreement with reality than a solution that converges to 0.01% or lower. So calculationsthat converge to 0.5% can also be considered to be converged for practical engineering purposes.

The second convergence check occurs in the innermost mass-flow iteration loop where the programmonitors the number of loops required to solve the continuity and radial equilibrium equations on eachcalculating station. The maximum value of the number of mass flow iterations and the location of thequasi-orthogonal where this occurred is also printed onto the screen and into the .hst file. You canspecify the maximum number of loops for the internal mass-flow iteration loop with the max_it_massparameter in the control file. The recommended value for this parameter is 10. Early in the run, theprogram usually stops the internal loops when the value of max_it_mass is reached. Later, as conver-gence is approached, less then 10 internal loops are generally required. A typical converged solutionmay require only 1 or 2 internal loops. The error in the mass flow should be tighter than that for themeridional velocity; a value of 0.001% is currently recommended.

The third convergence check occurs in an iteration to a specified pressure ratio and is related to theconvergence of the inlet mass flow to a final value and the convergence of the specified target pressuresat the trailing edges to a fixed value. These are written onto the output file as error_p and er-ror_mass respectively. The same limit for the mass flow convergence is used as given above and the


Vista TF

target pressure convergence is set internally within the program to be tighter than this. Experienceshows that it is important in simulations to a specified pressure ratio to ensure that tight tolerances onthese parameters are given, otherwise the program modifies the target pressures on the basis of inad-equately converged data and divergence may result.

In calculations with real gas equations (release 13.0 onwards) an additional convergence criterion isincluded related to the change in the real gas factor of the gas. The maximum error in the real gasfactor is written to the .out file.

In some cases with closely spaced calculating stations (high aspect ratio grid) it has been identifiedthat, although the simulation converges to a relatively low level of the maximum error in the meridionalvelocity of 0.1% relatively quickly, it does not always continue to converge below this level of error. Inthese cases you have several options. Firstly, it might be sensible to accept this level of convergenceand continue to optimize the geometry of the machine. In some cases, the simulation that is not perfectlyconverged may indicate the existence of an unwanted flow feature as a cause of the poor convergence.Such features include: anything that creates a tendency for the flow to reverse direction; an extremelyhigh curvature in the meridional channel; a poor grid. A second approach is to make use of othermodels that are built-in for the damping factors within the program, by reducing the value of thestreamline curvature damping factor damp_sc and the velocity level damping factor damp_vl . Insome cases it may be helpful to reduce both the value of damp_sc and damp_vl to smaller valuesthan the standard values of 0.25 and 0.40.

The most common failure for the program to converge is related to difficulties in the streamline curvaturecalculation causing divergence of the maximum error in the meridional velocity distribution. If theprogram identifies a trend for the results to start to diverge then it automatically decreases the dampingfactors to avoid divergence by causing more damping of the solution. If the errors continue to increase,then further reduction of the relaxation factors tends to freeze the unconverged iteration at the statewhere the problem was identified so that there is at least no unexpected exit from the program evenif the simulation does not converge. Because of this feature it is not advisable simply to increase thenumber of iterations in the hope that the simulation will converge. A better strategy is to calculate witha large number, say 2000 iterations, and if the simulation does not converge, restart the calculationfrom the restart file to reset the damping factors to sensible values. This strategy often works in difficultcases.

The program's internal numerical fix of reducing the damping factors when the error diverges is reportedin the history file. If this fix does not work, then the calculation of the streamline curvatures may ultimatelyfail, although this only occurs in simulations that have otherwise started to have serious numericalproblems. The ultimate failure here tends to be an error in the subroutine pero , called from subroutinecurvature , or in subroutine parabola , called from subroutine streamlines . Both are, in themselves,generally robust. Subroutine pero is a modified interpolation routine along the lines of the so-calledAKIMA splines. The breakdown is related to the numerical difficulty of calculating curvatures in a flowwhen the streamlines are no longer smooth and the spacing of the calculating planes is small. Subroutine"parabola" attempts to fit a parabola through internal data in the program and is also robust until seriousproblems occur. These errors have mostly been trapped such that an error message is printed and theprogram exits the calculation without breaking down. This error is trapped to avoid a catastrophicbreakdown, but the program stops and reports that the streamlines are too close together.

In cases where such problems occur, it is often worthwhile simply to run the simulation again from therestart file generated from an earlier unconverged simulation or with a different initial estimate of theflowfield (which is controlled through the initial value of cm_start in the control file), because startingfrom better initial conditions may clear the problem found in the initial calculation. In choked flows itis generally better to start with a lower value of cm_start , as this will not be choked. In some casesit may be worthwhile for you to decrease the relaxation factors (increase the damping) rather than to




let the program try to do this automatically. If this does not work then again a useful strategy is to reducethe maximum number of iterations to a lower value, say 50 to 100 (parameter max_it_main in .confile), and recalculate and examine the unconverged results. It may also help to run the simulation re-peatedly with such a low limit on the maximum number of iterations because each successive run tendsto get to a lower value of the maximum error. If this fails then it is useful to examine the .hst and.out files and plots of the unconverged results. These files and plots can be used to identify aspectsof the design that, when improved, enable convergence to be achieved.

The program also writes the values of the local error and the local choke ratio into the .txt file so itis possible to examine this and quickly locate the location where the problems are occurring. A valueof unity for the choke ratio implies that the flow is choked and a value above unity indicates that thelocal mass flow is above the choking mass flow of the streamtube. It may be necessary to have somefundamental understanding of how the turbomachine operates in order to identify how to removethese problems. Further comments on choking are given in Choking (p. 304).

The program prints an error message if it reaches a state where the maximum change in meridionalvelocity from one iteration to the next is more than 400% and then it closes down the calculation.Should this occur it is then recommended that you repeat the calculation with the maximum numberof iterations set at a value lower than that where the program stopped (see the history of iterations inthe output file) and then examine the results for this unconverged point. This can help you to identifythe problem.

12.2.5.6.3. Reverse Flow

Another common failure mode for the program where it has difficulties converging is related to difficultiesin calculating reverse flow in the meridional plane; such flow is not possible with the streamline curvaturemethod. If the program observes that reverse flow occurs in the meridional plane, then it attemptstemporary fixes to enable the iterations to proceed in the hope that the problem will be cleared. Oneparticularly important fix (that is not reported) is to avoid negative values of the meridional velocity bysetting any negative values to 5% of the mid-span value. This may be used in early iterations and laterno longer be needed, but if the final “converged” result contains such values, it is not really a validsolution. This can often be identified by a difficulty in convergence at a particular location in the grid.The background to this problem is a fundamental difficulty related to the physics, whereby a stronglyswirling flow in an annular duct may separate at the hub or shroud, and the streamline curvaturemethod is inherently unable to calculate such a reverse flow. This problem becomes more serious withcalculations of low hub-to-tip ratio. The use of a limit on the meridional velocity on any streamline isreported in the output.

12.2.5.6.4. Choking

It should be mentioned that the throughflow method is not particularly suitable for choked blade rowsbecause the mean stream surface equations average out the flow in the circumferential direction andare therefore not aware of high Mach numbers on the suction surface of blades. In addition, any shocksthat may be present in turbomachinery flows are generally not oriented in the circumferential directionso are smeared out in the circumferential averaging of the flow to determine the mean stream surface.Furthermore, the basic method is inviscid and this precludes the existence of strong shocks.

Nevertheless, despite these serious limitations, an attempt has been made to model choking in theblade rows so that, in combination with correlations, the maximum flow and the additional losses relatedto shocks are taken into account in the overall predicted performance. In this way, the program includesaspects of choking that are compatible with the level of empiricism of typical 1D calculation methods,and may even be more successful than these because the variation of Mach number over the span istaken into account. This is useful in a program intended for design purposes because it helps choking


Vista TF

problems to be identified at a relatively early stage in the design process, and aids the understandingof the axial and radial matching of the blade rows as the rotational speed varies.

All aspects of special calculations for choking flows are hidden. Setting the parameter i_expert =0 causes the program to examine choking but not limit the mass flow if choking is found to occur. Thisis the most robust way to run the program and is recommended for beginners. Expert users can usei_expert = 1 , which limits the mass flow according to various choke models.

Choking is strongly related to the throat areas between two blade rows, so any real estimate of thechoking flow should make use of accurate estimates of the throat areas. In Vista TF the shortest straight-line distance between two blade rows can be specified in the geometry file for each of the input sections,and if the values are not specified (that is, a value of zero is given) then Vista TF makes its own crudeestimate of the throat areas based on its limited knowledge of the blade geometry. This estimate is, atthe moment, too crude to be used accurately in the calculation of choked blade rows, and may leadto a value for the choked mass flow that is incorrect. It may be worthwhile, in some situations, to runVista TF with no values specified for the throat and then examine the program´s own estimate of thethroat areas, which are in the output file, and then run with slightly larger or smaller values specifiedin the geometry file, depending on the real throat areas or the choking mass flow if the latter is known.

One method that tends to work in the most difficult cases is to remove any calculating planes in theneighbourhood of the throat, as suggested by Denton (1978). Local details of the flow calculation arelost, but the calculation can then be made to converge up to the limiting mass flow.

When you specify a mass flow rate, you must ensure that it is not greater than the choking value. Theprogram may generate warning messages if the specified mass flow exceeds the choking value.

The output parameter choke_ratio is the ratio of the local mass flux to the maximum possible atthat location. A value above unity is not a realistic solution but if the choke parameter i_expert isset to zero then such solutions can be generated. This option is available because the program issometimes more robust under this operating mode than when i_choke is set to 1 and the full chokingmodels are used. Note that in some multistage computations the actual specified mass flow may bebelow that required to actually choke the machine, but during the iterations, individual blade rows maystill become choked. The program becomes less robust as choking is approached.

In some high-speed situations where it is difficult to obtain convergence with a specified mass flow,the restart file can be used to store results for a converged operating point at a lower speed and lowerflow, and then the required operating condition can be obtained by starting from the restart file withnew flow conditions slowly stepping to the required operating point. In a similar way, it is advisable tofirst set up a simulation close to the design flow, before attempting to move towards a higher flowwith a higher risk of choking.

You can run a simulation with a specified pressure ratio. For a choked blade row, the shift to a specifiedpressure ratio is a more physical approach than specifying the mass flow because, in this case, thesolution is indeterminate. In turbines with steep characteristics (little variation in mass flow with pressureratio) this is reliable. In compressors with flat operating characteristics (large variation in mass flow withpressure ratio) this approach may be more unstable so it is not recommended for calculations close tothe expected surge line.

Iteration to a defined pressure ratio makes use of the so-called “target pressure” ratio method of Denton(1978). This requires the program to make a first guess of the pressure at each trailing edge of themachine. The algorithm currently incorporated makes a crude estimate of this but it has been foundthat this may not be sufficient to secure convergence, especially for compressors. For this reason, youhave the option to define the first guess of the pressure at each trailing edge (set i_flow = 6 insteadof i_flow = 5 ). Even when using this option, the iteration to pressure ratio for multistage compressors




is still very sensitive, and convergence is not guaranteed. If iteration to pressure ratio is used for chokedblade rows, then an accurate estimate of the throat widths needs to be provided in the geometry fileand the value of i_expert should be set to 1 so that the limits to the mass flow at choke are imposed.If the throat area feature is calculated by BladeEditor then the throat widths will be calculated and addedto the geometry file.

12.2.5.6.5. Computational Grid

As in most numerical methods, a finer grid leading to a closer spacing of the quasi-orthogonal linesand streamlines will lead to a higher numerical accuracy of the simulation. Closer spacing of the quasi-orthogonal lines for streamline curvature calculations can, however, cause instabilities in the convergenceprocess. This can be overcome by increasing the numerical damping (lowering the relaxation factors)but this causes an increase in computational time, so a compromise is generally needed.

The relationship between the needed relaxation factors and the aspect ratio of the computational gridin a throughflow calculation was theoretically derived in a classical paper by Wilkinson in 1970 (seeStreamline Curvature Throughflow Theory (p. 275)). His studies showed that close spacing of the quasi-orthogonal lines required more damping. The problem is related to the fact that if the calculating stationsare closer together then a small displacement of a streamline can lead to a large curvature. In this caseit is only possible to take a small part of the new solution forward each iteration and so the number ofiterations increases. Wilkinson derived an equation of the following form to calculate the optimum re-laxation factor for a certain grid aspect ratio:

He also showed that the value given here as k2 is actually a function of the Mach number, the flow

angle, and of the method used to calculate the curvature of the streamlines. The aspect ratio of thegrid (h/∆m) in this equation is the ratio of the calculating station length to the meridional spacing ofthe grid lines. The equation above was used in older versions of Vista TF to determine the streamlinecurvature relaxation factor with the numerical values of k1 = 0.5 and k2 = 20/96. The value was based

on the largest aspect ratio that can be found in the domain. This is then further reduced if the calculationsshow any tendency to diverge during the iterations. The largest aspect ratio in the whole grid was used,whereby the meridional spacing is the shortest meridional distance between the two adjacent gridlines.

A value of 0.01 for this relaxation factor corresponds to an aspect ratio of around 15 and will lead tolong calculation times because only 1% of the new solution can be taken into account and 99% is carriedforward from the earlier solution. In order to reduce an error in the initial solution to 0.01% of its initialvalue then the number of iterations required would be

A value of 0.001 for this relaxation factor (aspect ratio near to 50) would lead to ten times as many iter-ations. Values of the aspect ratio larger than 15 are therefore not recommended but in cases wherethis is unavoidable (such as low aspect ratio blades with high span and small chord with internal calcu-lating stations) then the program still converges but at a slower rate.

A different stability scheme has been incorporated in Vista TF. In this, the relaxation factor ratio is cal-culated as:


Vista TF

Where alpha is the absolute flow angle in ducts and stator rows and the relative flow angle in rotorrows. In the new scheme, the relaxation factor can vary from quasi-orthogonal to quasi-orthogonal sothat regions of the grid with small aspect ratios are not penalized by a locally poor grid spacing elsewhere.In addition to this improvement, experience with many difficult cases has been incorporated in the se-lection of the maximum relaxation factor. With this new model, you should not need to adjust thedamping factors because the program does this automatically.

All flow gradients in the radial equilibrium equation and the curvature of the streamlines in the solutionare determined by a piecewise parabolic interpolation through three points. This is a very rapid numer-ical procedure but experience shows that this can cause errors in the estimation of the curvature of thehub and casing if the quasi-orthogonal spacing is too wide. For the case of an axial to radial bend withcircular arc meridional wall contours the error in curvature is of the order 2.5% on a grid with 7 quasi-orthogonal lines (a quasi-orthogonal placed every 15° around the bend). This decreases to 0.2% for 19quasi-orthogonals (a quasi-orthogonal every 5°). This suggests that typical radial impellers with an axialinducer should be calculated with around 15 quasi-orthogonals in the bladed region. Moreover, thebasic assumptions of the meridional throughflow method (for example: no frictional forces, axisymmetricflow, no spanwise mixing) certainly cause larger errors than this error in the curvature estimate.

It should be noted that decreasing the convergence tolerance to low values does not eradicate thiserror, so that for typical engineering applications a tolerance of 0.1% on cm is generally adequate becausethe curvature calculation is not more accurate than this. A lower value is however generally used becausethis confirms that numerical convergence has really happened.

Using 17 streamlines across the span is recommended. The use of such a large number of streamlinesacross the span brings little improvement in the accuracy when compared with experimental datacompared to using only 9, but nevertheless clearly leads to fewer numerical errors in the integrationof the radial equilibrium equation. 17 and 9 streamlines have the advantage that the streamlines splitthe flow uniformly into 16 or 8 streamtubes.

The limitations mentioned above on the calculation of flow gradients and curvature also determine towhat extent details of steps in the wall geometry can be taken into account. Steps, kinks and wells inthe meridional contour need to be omitted as these features are typically of a sub-grid size. Generallyattempts to improve the accuracy by using finer grids with more details are not made in streamlinecurvature calculations because this is counterproductive.

In axial turbomachinery calculations with quasi-orthogonals at the leading edge and trailing edge planesand with flare on the hub and casing walls, inaccuracies will arise because the kinks in the contourscannot be accurately modeled.

12.2.5.6.6. Other Numerical Issues

Computing a new solution from the restart file of an existing solution, even when no changes havebeen made in the input data, may require 50 or more iterations to re-converge to the solution on therestart file. This effect is due to the fact that the restart file stores only a small part of the informationrelated to the original solution, and many fine details of the solution have to be recalculated.

In running the program from a cold start (without a restart file) it is useful to think carefully about theinitial value of cm_start that is specified for the control file. This determines the meridional velocitylevel of the initial solution and it has been experienced that an initial value too far from reality may




cause the solution to breakdown. The value that should be specified for cm_start is the flow coefficient(cm/u) of the machine concerned, whereby the reference velocity used is the velocity determined bythe reference diameter and reference speed (ref_d and ref_n ). In flows with a risk of choking, it isbetter to use a low value of cm_start to ensure that the initial conditions are not close to choke.When a converged solution is available as a restart file, it should be used because, although it does notnecessarily reduce the number of iterations needed, it does make the solution process more robust.

The restart file is continually updated by each converged computation, but also if the solution reachesthe limit of the maximum number of iterations without converging. In a case where the solution is ac-tually diverging, the non-converged restart file might be a worse starting condition than the program´sown initial guess, so it can be useful to discard this and start again. If a converged solution has beenreached then it is sometimes useful to store the converged restart file for a better starting solution ifsomething later goes wrong during calculations at other operating points.

As with all numerical methods, the use of unconverged results for design decisions is extremely dangerousand is not recommended. Nevertheless, examination of unconverged results can often be extremelyuseful for identifying numerical issues. In some cases examination of unconverged results may indicatefeatures of the design that can be changed to avoid such problems by modification of the geometry.

One measure that has been incorporated and which may be useful during debugging of a poor calcu-lation is to set the parameter grad_ree equal to zero. This overrides the radial equilibrium equationand leads to a spanwise constant meridional velocity (see below). This can be useful in identifying errorswhere flow data is inconsistent with the geometry being calculated. Another alternative that can beused with turbines is to specify a near-zero swirl velocity on the mid-streamline at the outlet of the lastrotor and let the program determine the mass flow and pressure ratio consistent with this (“iterationto outlet swirl”). This has unfortunately not been thoroughly tested on enough cases to guarantee thatit will always work.

If all else fails then set i_ree = 3 in the control file. Under these circumstances the value specifiedas grad_ree in the control file is then used to reduce the meridional velocity gradient as follows:

This can be extremely useful for debugging, because it can allow the program to avoid failing due tohigh spanwise velocity gradients. In this way, it effectively becomes a mean-line program with nospanwise variation in meridional velocity (if grad_ree = 0.0 ). Other parameters such as the bladespeed still vary across the span, so it is not exactly a mean-line program. Most cases converge underthese conditions and this ensures that the axial matching along the mean streamline of the blade rowsis approximately correct. When converged it may be possible to approach a solution with the correctradial distributions by slowly relaxing the value of grad_ree towards unity. During this process, thefeatures of the velocity gradients that cause trouble slowly become part of the calculation and this canthen help to identify the problem.

Cases that do not converge well are usually either poor designs, or are good designs operating a longway from their design point. Non-convergence generally means that the design is a long way fromsatisfying the condition of radial equilibrium. If the flow solution converges well then this usually meansthat the flow is automatically close to radial equilibrium and this will generally indicate a higher-qualitymachine.


Vista TF

12.2.5.6.7. Using real gas files

Considerable effort has been made to ensure that the real gas implementation is as robust as otheraspects of the program. Nevertheless, errors in the definition data for the real gas can cause stabilityproblems, as they may lead to unrealistic values of the gas properties. In some cases it may be worthwhileto run a particular case using ideal gas equations (with an estimated ratio of specific heats and an es-timated specific heat at constant pressure). The restart file for a converged simulation of this type canthen be used for subsequent calculations with real gas properties.

12.2.5.7. Appendix G: The RTZTtoGEO Program

The ANSYS .rtzt and the Vista TF .geo Geometry Definition File

The data in the .rtzt file is the meanline coordinate data as produced by the program BladeGen ona number of layers from hub to shroud, with a format as described in Description of the RTZT File (p. 312).There are several differences to the .rtzt option in BladeGen and the geometry actually needed bya throughflow program, and for this reason the RTZTtoGEO program has been written to convert thisdata into a more suitable format, rather than working directly with the .rtzt file.

The first difference is related to the need for a throughflow program to work with blade angles, in thecase of Vista TF with the blade lean angles gamma_r and gamma_z (see Definition of Blade LeanAngles (p. 289)). Throughflow programs work on relatively coarse grids and so it is sensible to providethe programs with the blade angles rather than just the blade coordinates. Otherwise the blade angleswould have to be calculated by interpolation and differentiation on a coarse grid which will lead toerrors. In axial duct flow programs with planes just at the leading and trailing edges, this is done byspecifying the axial blade angles or flow angles at the inlet and the outlet. In Vista TF, which includesthe blade force in the solution and is designed for radial machines, both the axial and the radial leanangles (gamma_r and gamma_z) have to be defined and, because the program includes internal planes,these are defined throughout the blade, not just at the edges.

The second key difference is that RTZT concentrates on the information required on each stream-surface(or layer) at a certain percentage of the span of the blade, or on blade sections or layers. Throughflowprograms concentrate on the information in a plane at right angles to this and require informationalong equally spaced quasi-orthogonal lines at fixed meridional positions through the blade rows, andupstream and downstream. In .rtzt files, the number of points and spacing in the meridional directionis not the same on each layer. In a throughflow program, it is the location of the layers on each quasi-orthogonal line that may be different, but the points along the streamline must be evenly spaced forall layers.

The throughflow program also needs additional data upstream and downstream of a blade row todefine the position of the quasi-orthogonal in a region of duct. This is not the case in the current .rtztfile, because different levels of information are available on each layer in this region, and curved quasi-orthogonals outside of the bladed regions are difficult to define. In some of the cases tested with highlycurved leading edges, it was necessary to modify the .geo file by hand to avoid problems with thequasi-orthogonals upstream or downstream of the leading edge crossing the leading or trailing edge.

Conversion of a .rtzt File into a .geo File

A program, RTZTtoGEO, can be used to generate the geometrical input data file (file extension .geo )for the Vista TF streamline curvature program from the BladeGen RTZT output format (file extension.rtzt ). The RTZTtoGEO program can be obtained from PCA, although the preferred method to generatethe .geo file is to use the BladeEditor VistaTFExport feature (see Export to Vista TF (.geo) (p. 78)). Noguarantee can be given that the RTZTtoGEO program will work on all cases. This program can be used




to merge several separate .rtzt files from BladeGen into a single .geo file for multiple-blade-rowsimulations with Vista TF.

For a single blade row, the program makes use of three files:

• One input data file is needed: RTZT geometrical data file (.rtzt ).

• The program creates one output file: Vista TF geometrical data file (.geo ).

• The input and output file names are specified by you in an auxiliary data file that must be called rtzt.fil .It must contain the necessary file names in the following order and form:

Number of blade rows 1RTZT datafile name impeller.rtztVista TF geometry datafile name impeller.geo

For multiple blade rows, additional .rtzt files are needed for each blade row and the associated .geofiles are merged into a single .geo file for the whole domain. The number of individual blade rows isgiven in the first line and the names of the individual .rtzt files follow this.

For n blade rows the program makes use of n+2 files:

• n input data files are needed: n RTZT geometrical data files (.rtzt ).

• The program creates one output file: Vista TF geometrical data file (.geo ).

• The input and output file names are specified by you in an auxiliary data file that must be called rtzt.fil .It must contain the necessary filenames in the following order and form:

Number of blade rows nRTZT datafile name Prefix1.rtztRTZT datafile name Prefix2.rtzt: :RTZT datafile name Prefixn.rtztVista TF geometry datafile name impeller.geo

Note that the different .rtzt files must be generated from BladeGen files that join up to each otherconsistently with no gaps. If the domains between adjacent blade rows as defined by the .rtzt filesmeet exactly at the boundary of the domains then the program includes a single quasi-orthogonal atthis boundary.

Running the RTZTtoGEO Program

The file impeller.rtzt and the file rtztgeo.fil should be in the same directory as the executableof the compiled program files. Starting the program produces the .geo file from the .rtzt file. If the.geo file already exists, there is a prompt as to whether this should be overwritten.

The program prompts you to identify whether the blade row is a compressor (pump) or turbine, andwhether it is a rotor or stator. In addition the program suggests the number of quasi-orthogonals up-stream of the leading edge and downstream of the trailing edge but you can change these numbersif this seems appropriate, simply by replying with n (n = “no”) to the prompt and then suggesting thenew number of quasi-orthogonals to be used. If you reply with a y (y= “yes”) then the automatic choiceof the program is used, which is fairly sensible in most cases.


Vista TF

Note

The limit on the number of points in each layer of the .rtzt files is 1000. The limit on thenumber of layers is 15. The maximum number of blade rows is 20. Up to 3000 quasi-ortho-gonals can be included in the final .geo file.

There are options to modify the number of quasi-orthogonals in the blade row and to change the ori-entation of the calculating grid so that this is oriented in the positive z direction. The change in orient-ation is needed because in some BladeGen files the flow travels in the direction of the negative z axis,and Vista TF assumes that the flow is traveling in the direction of the positive x axis. Without this switchthe hub and shroud contours become reversed. The option to modify the number of quasi-orthogonalsin a blade row has been included so that RTZTtoGEO can better cope with axial blades with short chordsrelative to radial machines with long chords. The program makes its own choice of the number of quasi-orthogonals needed, based on the aspect ratio of the blade; this is more or less consistent with thestability requirements of the Vista TF program. The rule used leads to 16 quasi-orthogonals in most ra-dial compressors. You can change this number if this seems appropriate, simply by replying with n (n= “no”) to the prompt and then suggesting the new number of quasi-orthogonals to be used. If youreply with a y (y= “yes”) then the automatic choice of the program is used, which is fairly sensible inmost cases.

Merging Several .rtzt Files into a Single .geo File

Vista TF can run a computation for up to 20 blade rows. BladeGen can currently only be used to definethe blade rows individually so each .rtzt file contains only a single blade row. It is possible to mergethese into a single .geo file automatically using Vista RTZTtoGEO, as explained above. The procedurebelow describes how to merge individual .geo files by hand, just in case the automatic option fails.

1. Run RTZTtoGEO for the first blade row.

2. Set up a Vista TF calculation for this blade row with all the appropriate input files and get this to con-verge.

3. Run RTZTtoGEO for the next blade row.

4. Edit the .geo file of the first blade row to include the additional information for the downstream bladerow, as follows:

1. Change the title section as appropriate (section 1).

2. Add new quasi-orthogonals in section 3 (copy from the .geo file of the next blade row and addto the .geo file of the first blade row). This may require modifications by hand in the duct regionbetween the blade rows because the quasi-orthogonals have to step smoothly from inlet to outletand not overlap. Usually this means deleting some quasi-orthogonals downstream of the first bladerow and upstream of following blade row where the domains would overlap. Note that the num-bering of the quasi-orthogonals does not have to be changed because the program does not readthis information. The numbering of each quasi-orthogonal can then be left as originally output bythe RTZTtoGEO program. Be careful in this process not to add any extra empty lines to the .geo




file between the different lines or to take any away. Note that the type and orientation of the bladerow should be correct automatically (with i_type 2 for stators and 3 for rotors) and that thenumber of upstream and downstream quasi-orthogonals can be modified so that this process iseasier.

3. Do the same for the blade definition section (section 4), whereby only those lines within the bladerow need to be copied in.

4. At the end of the .geo file, the details in section 5 need to be modified to give throat areas forthe blade row that has been added (which is actually zero because the ,rtzt file does not providethis information. Thus if there are two blade rows each defined with five sections then there wouldbe 10 lines at the end of the .geo file. These extra lines can also be cut and pasted from the .geofiles for the appropriate blade row if you want.

5. Count the number of quasi-orthogonals that are left and change this in section 2. If you are luckyand have not had to delete any quasi-orthogonals, this is straightforward.

5. Edit the .cor file of the first blade row to include additional downstream blade row losses, deviation,and blockage data, noting that the location of the deviation is related to the number of the quasi-or-thogonal and not the blade row. Note that the aerodynamic and control data files (.aer and .con )do not usually need to be changed in this process.

6. Delete the restart .rst file because the restart capability only works if there has been no change tothe number of quasi-orthogonals. Vista TF recognizes this and recreates a new restart file automatically.

7. Run Vista TF, and repeat from step 3 for each additional blade row. Note that experienced users mightmanage to make the modifications to the .geo file and the .cor file for several blade rows togetherin 1 step, but doing it step-by-step gives better control of where any errors have been introduced.

Description of the RTZT File

This section describes the generic RTZT data file format for BladeGen. The file is an ASCII file that usesspace separation between values.

Note

Angular values must be in radians.

Example file:

text enclosed in is a data itemtext enclosed in () is a commenttext enclosed in [] is optional

number of bladesnumber of splitters (0 is main blade only)(for each blade, main and splitter) pitch fraction (Ignored for main blade) number of layers [N] [T] (Normal or Tangential Thickness Flag)

(for each layer) span fraction number of points [a][t][b] (for each point)

r theta z thickness


Vista TF

: :

12.2.5.8. Appendix H: Example of a Real Gas Property Data File (*.rgp)

Example 1

The first example given here is for air.

Section 1: 3 lines of characters identifying the run (max 72 characters/line)

Real gas data for airFirst test case of real gas data6 January 2010

Section 2: Name of gas (72 characters)

gas_nameair

Section 3: Molecular mass and/or Gas constant

MW (kg/kmol) Gas_R (J/kg/K)28.97 287.1

Section 4: Critical point parameters and acentric factor

Pc (Pa) Tc (K) Vc (m3/kg) gas_omega (-)3758000.0 132.3 0.002857 0.033

Section 5: Temperature limits of piecewise specific heat curves

T_min (K) T_max(K) order_T_poly (max 8)100.0 1000.0 8

Section 6: Coefficients of cp_polynomial for lower temperature region (T_min < T < T_mid)

A1 A2 A3 A4 A5 A6 A7 A81.161482e+003 -2.368819e+000 1.485511e-002 -5.034909e-005 9.928569e-008 -1.111097e-010 6.540196e-014 -1.573588e-017





wb_bm

Documents

ansys cfx

ansys turbogrid

ansys workbench journaling

trademark usedby ansys

proprietary products

software license agreementthat

geometry sources

imported bladegen geometry