May 2014

Hydrogen safety: Large turbo-generators

ENERGY-TECHDedicated to the Engineering, Operations & Maintenance of Electric Power Plants

In Association with the ASME Power Division

MAY 2014

www.energy-tech.comA WoodwardBizMedia Publication

Transmitter Troubles 10 • Compressor Surge 14 • ASME: Harmonic Impacts 20

CUTS_Brush_Problem_Ad_ENTECH_1-13_FIN.indd 1 12/31/12 11:39 AM

May 2014 ENERGY-TECH.com 3

FEAtUrEs

6 Hydrogen safety for large turbo-generatorsBy Steve Kilmartin, E/One Utility Systems

10 The trouble with transmittersBy Mike Schultz and Stace Smith, The Asset Performance Group

CoLUMNs

14 Maintenance MattersCentrifugal compressor surge and compressed air systemsBy DeVon A. Washington, Ph.D., Consumers Energy

26 Machine DoctorTilting pad bearing modifications rotordynamic analysisBy Patrick J. Smith, Energy-Tech contributor

AsME FEAtUrE

20 Impact of harmonics from distributed energy resources on distribution networks By Reza Arghandeh, UC Berkeley, Berkeley, Calif.; Ahmet Onen, Virginia Tech, Blacksburg, Va.; Jaesung Jung, Virginia Tech, Blacksburg, Va.; Danling Cheng, EDD Inc., Blacksburg, Va.; Robert Broadwater, Virginia Tech, Blacksburg, Va.; and Virgilio Centeno, Virginia Tech, Blacksburg, Va.

iNdUstrY NotEs

4 Editor’s Note and Calendar

30 Advertisers’ Index

31 Energy Showcase

oN tHE WEB

Did you miss our April webinar, Proper level control for feedwater heaters, with Mike Catapano and Eric Svensson? Download it for free at www.energy-tech.com and find other great webinars and technical papers too.

Cover photo contributed by E/One Utility Systems.A division of Woodward Communications, Inc.

ENERGY-TECH

P.O. Box 388 • Dubuque, IA 52004-0388800.977.0474 • Fax: 563.588.3848Email: [email protected]

Energy-Tech (ISSN# 2330-0191) is published monthly in print and digital format except in January and July, when it is published in digital format only by WoodwardBizMedia, a division of Woodward Communications, Inc. WoodwardBizMedia assumes no responsibility for inaccuracies, errors or advertising content. Entire contents © 2014 WoodwardBizMedia. All rights reserved; reproduction in whole or in part without permission is prohibited.

Printed in the U.S.A.

Group PublisherKaren Ruden – [email protected] ManagerRandy Rodgers – [email protected] EditorAndrea Hauser – [email protected]

Editorial Board ([email protected])Kris Brandt – Rockwell AutomationBill Moore – Director, Technical Service, National Electric CoilRam Madugula – Executive Vice President, Power Engineers Collaborative, LLCKuda Mutama – Engineering Manager, TS Power Plant

Editorial views expressed within do not necessarily reflect those of Energy-Tech magazine or WoodwardBizMedia.

Advertising Sales ExecutivesTim Koehler – [email protected] Hoppensteadt – [email protected] Gross – [email protected]

Creative/Production ManagerHobie Wood – [email protected] ArtistValerie Vorwald – [email protected]

Address CorrectionPostmaster: Send address correction to: Energy-Tech, P.O. Box 388, Dubuque, IA 52004-0388Subscription InformationEnergy-Tech is mailed free to all qualified requesters. To subscribe, go to www.energy-tech.com or contact Linda Flannery at [email protected] InformationFor media kits, contact Energy-Tech at 800.977.0474, www.energy-tech.com or [email protected] SubmissionSend press releases to: Editorial Dept., Energy-Tech, P.O. Box 388, Dubuque, IA 52004-0388 Ph 563.588.3857 • Fax 563.588.3848 email: [email protected] SubmissionSend advertising submissions to: Energy-Tech, 801 Bluff Street, Dubuque, Iowa 52001E-mail: [email protected].

4 ENERGY-TECH.com May 2014

The cold conundrumWinter demand calls more plants into service –

so what happens next year?

Editor’s NotE CALENdAr

We are finally enjoying spring in the Midwest – and not a moment too soon. Everybody was getting a little edgy.

Now we can turn off the furnace, crack the windows and open the utility bills showing how much energy it took to keep our homes warm during the weeks of negative temps. It isn’t pretty, but it would’ve been worse if there hadn’t been any power at all.

In fact, there was so much demand that utilities had to call older coal-fired plants into service to meet it. Plants

that in a few months or the coming year will be closed due to new emissions regulations.

This begs the question of where that energy will come from during next year’s frigid temps. Will utilities still be able to meet demand, and what happens if they can’t? Rolling blackouts are never fun, but at -20°F, they’re dangerous.

Maybe this is the painful part of change – the part that requires growth out of the old standard. Maybe it will spur more energy efficiency upgrades in homes and buildings, or more careful monitoring of the thermostat.

And while coal remains the largest source of fuel for electric generation, cheap natural gas is close on its heels, accounting for more than 50 percent of new generating capacity in 2013, according to the U.S. Energy Information Administration. Whether natural gas will remain cheap is a debatable point, though, as well as whether we have enough to make it a primary fuel for our power needs.

Reading about all of this reinforces the importance of energy efficiency to me – an aspect of our country’s power use that gets very little press. It isn’t very exciting to talk about new energy efficient light bulbs, Energy Star appliances or double-paned windows, but it all adds up to less energy demand and less likelihood of not having enough for everyone.

It also means those winter heating bills might not be quite so painful, and that’s a pretty exciting thought to me. What do you think about energy efficiency incentives? Are they worth it? Have you applied some in your home?

Email me at [email protected] and let me know, in the meantime, thanks for reading.

Andrea Hauser

May 19-22, 2014Introduction to Machinery VibrationsKnoxville, Tenn.www.vi-institute.org

June 11-13, 20142014 Vibration Institute Training ConferenceSan Antonio, Texaswww.vi-institute.org

June 11-13, 20143rd Natural Gas Vehicles USA Conference &

ExhibitionHouston, Texasngvevent.com

July 21-25, 2014Rotor Dynamics and ModelingSyria, Va.www.vi-institute.org

July 28-31, 2014ASME 2014 Power ConferenceBaltimore, Md. www.asmeconferences.org/power2014

Aug. 19-22, 2014Balancing of Rotating MachineryHouston, Texaswww.vi-institute.org

Sept. 16-19, 2014Machinery Vibration AnalysisSalem, Mass.www.vi-institute.org

Nov. 11-14, 2014Advanced Vibration ControlSyria, Va.www.vi-institute.org

Submit your events by emailing [email protected].

July 28-31, 2014Baltimore, Maryland

Hyatt Regency Baltimore on the Inner Harbor

Mark your calendar now for the BEST technical conference in power generation

Media SponsorMedia Sponsor

2014 Pre-Conference WorkshopsTurbines and GeneratorsSunday, July 27 – Monday, July 28This two day ASME course covers the steam turbines and generators design, components, thermal cycle, performance, operation and maintenance.

The Total Condenser Performance™ WorkshopMonday, July 28 • 8:00am- 12 noon Condenser effi ciency and reliability through effective cleaning, testing and inspection. Maximize MW output, minimize risk of condenser related forced outages, and reduce CO2 emissions.

Critical Power Plant Pumps — Performance Reliability & OptimizationMonday, July 28 • 1:00pm- 5:00pm Basic understanding of pump hydraulics, standard materials, manufacturing & assembly and pump operation. Guidelines for monitoring the pump’s performance and methods of improving & optimizing the pump’s operational reliability.

Heat Rate Assessment: “Looking for the ‘Small Stuff’ to Optimize Plant Performance?”Monday, July 28 • 1:00pm- 5:00pmHeat rate best practices to optimize power plant effi ciency and fuel cost reduction. Provide training to optimize equipment and system effi ciencies. Advanced Pattern Recognition technology will be presented for integration with the thermal performance application for detecting and resolving anomalies that may result in catastrophic equipment failure and/or increased fuel costs.

For complete workshop descriptions and registration, go to:

www.asmeconferences.org/POWER2014/Workshops.cfm

Engage in operating, maintenance, and equipment issues with experts inside the power plant and fi nd out the latest methods and cutting-edge technology to improve your power plant’s operation.

Sign up for ASME Power 2014’s pre-conference workshops before June 8 to take advantage of Early Bird Discount Rates.

2-day workshop July 27-28

Early bird rate $600(after June 7th – $675)

4 hour workshops July 28

Early bird rate $175(each; after June 7th – $225)


FEAtUrEs

By Steve Kilmartin, E/One Utility Systems

The need for hydrogenCentral power stations have been producing electric power

and supplying it to a customer base since the early 1880s. As the demand for power increased, so did the physical size of air cooled electric generators — more megawatts required more iron. In the early 1930s, it became apparent that a better method of cooling these large turbo generators was required and the first hydrogen cooled generator was introduced.

Hydrogen replaced air as a cooling agent principally because of its low density and superior cooling properties. Hydrogen is the lightest known gas and has the lowest density of any stable gas. Because hydrogen is one quarter the density of air, wind resistant losses are greatly reduced and efficiencies are gained. Hydrogen also has a thermal conductivity of nearly 7x that of air, resulting in much better heat transfer through forced con-vention. (Table 1)

Using hydrogen as a cooling medium has additional benefits. Because the generator case is a sealed pressure vessel, the internal components are less likely to be affected by outside contami-nants. Also, pressurized hydrogen will suppress partial discharge and increase the amount of voltage required to cause a compo-nent breakdown.

When used properly, hydrogen is 14x more efficient than air in removing heat and greatly reduces windage friction losses. However, a mixture of hydrogen in air of between 5 percent and 75 percent becomes very explosive. Also, the generator effi-ciency is reduced as the purity of the hydrogen drops. To ensure safe and efficient operation of the generator, the hydrogen purity should be maintained above 97 percent.

From bottle to caseThe safe operation of hydrogen cooled generators requires

proper practices be followed in order to minimize the chances

Hydrogen safety for large turbo-generators


of hydrogen and air ever mixing. The key to safe operation is to never allow a flammable mixture to exist, prevent hydrogen leaks and to eliminate the possible potential sources of ignition.

Three different gases are required to maintain and operate hydrogen cooled generators. The generator will be filled with air during maintenance; an inert gas is used as an intermediate gas so that air and hydrogen do not mix; and hydrogen is used during normal operation.

The air used must be clean and dry and typically comes from the power plant’s instrument air supply system. Some plants will install an air dryer between the plant instrument air supply and the generator to ensure a clean, dry air supply.

The inert gas used to prevent the mixture of hydrogen and air has traditionally been carbon dioxide (CO

2); however, some

power plants and OEMs are now using nitrogen or argon. If CO

2 is used as the intermediate gas, care should be taken to

minimize the chances of freezing and thermal shock to the gen-erator. Some utilities will incorporate a CO

2 vaporizer, which

prevents the CO2 from freezing. Having a sufficient amount of

CO2 available on site is very important in order to purge every

generator, should a situation require emergency purging of each generator.

Pure, dry hydrogen (H2) is used during normal operation of

the generator. Whether the H2 comes from bottles, bulk tank or

an on-site hydrogen generator, the H2 should be 99.997 percent

pure with a dew point of less than -10°C.

A gas manifold system controls the flow of gases from the bulk supplies to the generator case. (Figure 2) The manifold also vents gases from the generator. Most gas manifolds incorporate a mechanical system that prevents hydrogen and air from being supplied to the generator at the same time.

Gassing and degassingAfter maintenance, the generator case will be filled with

air. One of the last tasks performed, prior to gassing the gen-erator, is performing a leak test using pressurized air. During this test it is important to include as much of the generator gas system as possible to ensure that the auxiliary systems also are

Table 1

Table 1 – Hydrogen’s Advantages As Cooling Agent

Cooling Medium Molecular Weight

Specific Heat Capacity Density

Air 28.95 1.00 1.00

Hydrogen 2.02 14.30 0.07

Hydrogen at 30 psig 14.30 0.21

Hydrogen at 45 psig 14.30 0.26

Water 4.18 1000.00

FEAtUrEs

©2014

OUR ERROR-FREE COMMITMENT

Call today to consult with a on-site service expert. 888-345-5015 • 916-288-6100 • www.tritool.com

“Tri Tool’s performance, with their commitments to Safety, Human Per-

formance, and Quality was exceptional. Their personnel and equipment reflected the

Standards of Excellence expected by the Site.”

• Fossil and Boiler Maintenance• Specialty Machining and Welding Services• Consulting Services for Outage Planning• Custom Engineering Projects• Equipment Rental

For on-site service, you can count on Tri Tool for reliable and out-standing project execution. Our Service technicians are highly skilled and trained to be experts in field services. With our culture of safety, integrity and hard work we consistently deliver error-free project results.


free from leaks. Because hydrogen is much more difficult to contain vs. air, checking for leaks after admitting hydrogen into the generator is important. If a standard operating procedure (SOP) specifically for checking for hydrogen leaks is not avail-able, one should be created.

As mentioned, the generator will be filled with air after maintenance and ensuring that the air is completely removed prior to filling with hydrogen is critical. This is accomplished by replacing the air with an inert gas.

There are two gas distribution pipes, called headers, within the generator. One header is located at the top of the genera-tor and one header is located at the bottom of the generator. For the procedure described in this article, the top header will be used for the admission of hydrogen and air and the bot-tom header will be used for the admission of carbon dioxide. Depending on the header that is being used to admit gas, the other header will be used for a vent.

To remove air from the generator case, the top header is open to vent and the bottom header is connected to the car-

bon dioxide supply. Carbon dioxide is slowly admitted into the generator case in order to maintain a blanket between the heavier carbon dioxide gas and lighter air. Once the air has been completely removed from the generator case, the bottom header is connected to vent and the top header is connected to the hydrogen supply. Hydrogen is then admit-ted slowly into the generator case in order to maintain the blanket between the heavier carbon dioxide gas and lighter hydrogen. The gassing process is now complete and the gen-erator is ready for operation.

Hydrogen sealsThe hydrogen seals are two of the most critical components

to contain the hydrogen in the generator case and keep air out of the generator case. One seal is on the exciter end of the gen-erator and one is on the turbine end. Each seal uses pressurized oil to create a seal between the generator enclosure and the rotating field. The seal oil for an auxiliary skid is supplied at a higher pressure than the hydrogen case pressure to ensure the

FEAtUrEs

Figure 2. P&ID


oil flows in the direction of the generator case. A labyrinth seal prevents oil from entering the generator. A crucial component in the seal oil system is the differential pressure regulator, which maintains the proper differential pressure between the seal oil pressure and hydrogen case pressure. Properly operating seal oil systems will help prevent the possibility of an explosive mixture of hydrogen and air from occurring inside the generator case.

Safe generator operationWhen proper procedures are followed, hydrogen-cooled

synchronous electric generators are generally reliable and safe. However, not following standard operating procedures (SOP) can lead to catastrophic results. Recent events illustrate that not following SOPs results in extensive damage to the plant, equip-ment and loss of life.

All generators are equipped with instrumentation to mon-itor vital signs such as voltage, amperage, vars, temperature and vibrations. On the hydrogen system, monitoring case pressure, fan differential pressure, hydrogen supply pressure, carbon diox-ide supply pressure, hydrogen purity, dew point and overheating are common. It is important to note that any instrumentation used in a hydrogen environment, also referred to as a hazardous area, should be third party certified to be used in such an envi-ronment. Case pressure and hydrogen purity are critical mea-surements to assist in the safe operation of the generator. It also is very important to accurately monitor hydrogen usage, which might be accomplished using a mass flow transmitter installed in the hydrogen supply line or by monitoring the number of bottles of hydrogen used daily. Hydrogen usage will be a key indicator should a hydrogen leak develop.

As noted, when the purity of hydrogen drops below 74 per-cent and the contaminating gas is air, the mixture is very explo-sive. Hydrogen purity must be monitored continuously and one should never assume that because pure hydrogen is being put into the generator, it will remain pure.

The above graphic indicates practices that should be followed to ensure the safe operation of hydrogen cooled generators.

During a generator’s life, meeting evolving safety standards for hazardous areas and integrating technologies into cost-ef-fective packages that allow installation work to be accomplished within the more dominant, critical path build and turbine out-age periods remain critical endeavors. As always, safety and risk mitigation should remain at the top of the priority list, which includes production and performance efficiencies, as well as environmental considerations. ~

Steve Kilmartin is the director of products and markets for E/One’s Utility Systems business. Considered a leading expert in the field of generator monitoring and maintenance, he has authored numerous papers, including work as principal investigator for the “EPRI-Turbine-Generator Auxiliary Systems, Volume 3: Generator Hydrogen System Maintenance Guide.” Kilmartin began his career with E/One in 1988 as an instrumentation specialist. Prior to that, he worked in the instrument shop at GE, and also as an applications engineer at Mechanical Technology Incorporated in New York. His career has taken him in and around large machines, and around the world for more than 30 years. You may contact him by emailing [email protected].

FEAtUrEs

Safe Operating Practices• A detailed Standard Operating Procedure (SOP) should be available for purging the generator• A detailed Standard Operating Procedure (SOP) should be available for locating leaks, qualifying the hydrogen leak and

repairing the hydrogen leak• The hydrogen system Piping and Instrumentation Drawing (P&ID) should be accurate and up to date• All hydrogen auxiliary system components should be properly labeled and agree with the P&ID• All hydrogen auxiliary system instrumentation should be operating properly and calibrated• All hydrogen auxiliary system instrumentation should be third party certified to be used in a hazardous area• Continuous monitoring of hydrogen case pressure and hydrogen purity• Continuous monitoring of hydrogen usage• Always maintain a sufficient supply of carbon dioxide to purge every generator on site, should the need arise• The generator and hydrogen auxiliary system should be pressure tested with air or CO

2 prior to pressurizing with hydrogen

• A manifold system should be used that will prevent the possibility of an air/hydrogen mixture• Always maintain a positive hydrogen pressure inside the generator to prevent air from entering • No welding should be done on the hydrogen system or seal oil system while there is hydrogen in the generator• Low explosive limit (LEL)detectors should be incorporated in areas where hydrogen could accumulate

AreShaft

CurrentsDestroying

Your Machinery?

Failure to properly ground rotating

equipment can result in expensive bearing, seal, &

gear damage.SOHRE TURBOMACHINERY® INC.

128 Main Street, P.O. Box 1099 Monson Massachusetts, USA 01057-1099

PH: 413.267.0590 • 800.207.2195 • FX: [email protected] • www.sohreturbo.com


FEAtUrEs

The trouble with transmittersProper maintenance requires focusing on what the device does, not what it is

By Mike Schultz and Stace Smith, The Asset Performance Group

Maintenance, at its most basic level, is a set of tasks that preserves the functions of a piece of equipment at a desired standard of performance.

We take the car in for an oil change to preserve the anti-wear capabilities of the engine so it can continue to get us from point A to point B at a certain speed and in a certain amount of time. We flush the sediment from a domestic hot water heater to preserve its ability to heat water to a desired temperature within a certain timeframe at a desired efficiency and noise level. We calibrate the output of a temperature trans-mitter to preserve its ability to send an accurate signal that is used by other equipment and operations staff to control pro-cesses, trigger safety and environmental controls and warn that intervention or repair is required.

It’s only when we fully understand the functions we’re preserving that we can take the appropriate steps to maintain those functions. The ‘s’ on ‘func-tions’ is intentional. One piece of equipment doesn’t necessarily equal one function. In fact, for one piece of equipment there could be multiple functions, multiple ways it could fail, mul-tiple consequences of failure and — you guessed it — multiple tasks to prevent the failure.

A transmitter, in fact, is a perfect example to use to explore this concept. Not only have transmitters become increasingly common in industry, especially as automation increases, but if not maintained properly, they can be a cause of catastrophic failures that could result in costly downtime, environmental degradation and even loss of life. And they are among the best examples of equipment with multiple functions, failure effects and failure consequences.

In this article we will discuss why simply “calibrating the transmitter” because a work order tells us to isn’t enough when it comes to a transmitter maintenance program. To develop a quality transmitter maintenance routine, we have to understand each of the transmitter’s functions, what happens if each func-tion fails, and what can and should be done — if anything — to reduce the likelihood of each failure to a tolerable level.

The roots of our approachAt APG, we help clients get the best value from their assets

while keeping people and the environment safe. One of the recurring issues we see in industry is transmitters that are incorrectly maintained. We have found that the principles of Reliability Centered Maintenance (RCM), when applied cor-

rectly, can help organizations improve the maintenance of their transmitters and more.

RCM was originally developed for the aviation industry in the 1970s and showed that designers and tradespeople in avi-ation maintenance actually had an incorrect understanding of how equipment behaves, the causes of equipment failures and how to prevent them. The rigorous analysis required by RCM gave them insights that resulted in better maintenance programs.

While we don’t use the principles of RCM to keep planes in the sky, it can help us understand what different systems and pieces of equipment do, how they could fail and what to do about it. Even if you don’t apply formal RCM across your plant, the concepts of RCM will be valuable in helping you understand your equipment and how you can get the most from it.

Introducing our transmitter

The transmitter we use in this article is a temperature transmitter on a water pump-ing system that delivers process water from a reservoir to a downstream process. The water

is heated by a steam coil in the reservoir and must be delivered at a temperature between 70°C and 80°C. Temperatures below 70°C will impact product quality if left for long periods of time. Temperatures above 80°C will immediately compromise quality, might damage process equipment and could result in safety or environmental hazards. If water isn’t delivered to the downstream process, the process will shut down.

The transmitter provides an input signal into the tem-perature control loop and to a circuit designed to alarm at temperatures below 70°C, and both alarm and shut down the process if the temperature exceeds 80°C. The temperature also displays on the operator’s HMI.

The jobs of the transmitterThe first step in developing a world-class maintenance pro-

gram for our temperature transmitter is to identify the various jobs — or functions — it has. This requires a shift in focus from what the transmitter is to what it does.

It becomes easier to identify the functions of a piece of equipment if we think of it as part of a larger system. It’s com-mon maintenance-speak to say, “I’m calibrating the transmitter,” but really we’re calibrating the output of the transmitter, which is an input into a broader system. This system (alarming, for example) is much more important than the transmitter itself.

“It’s only when we fully understand the functions we’re preserving that we can take the appropriate steps to maintain

those functions.”


With this perspective in mind, the 4 jobs (or functions) of the transmitter, as part of a larger system, are:

• To supply process water at between 70°C and 80°C to the downstream process.

• To indicate the temperature of the process water within 2 percent of actual on the HMI.

• To be capable of alarming on the HMI in the event the water temperature drops below 70°C.

• To be capable of alarming on the HMI and shutting down the process in the event the water temperature exceeds 80°C.

How the transmitter can fail, and what to do about it

Now that we know what a transmitter’s jobs are, we can identify the specific ways the transmitter might not meet our per-formance expectations, what consequences that could have and — given those conse-quences — what specific maintenance tasks are worth doing to deal with each conse-quence. As you’ll see, identical failures can have different consequences, depending on the function that has been compromised.

Function 1 – To supply water at between 70°C and 80°C to the downstream process.Failure: Temperature transmitter calibration drifts Effects and consequences: If the tempera-ture transmitter drifts high, it will tell the temperature control system that the water temperature is warmer than it actually is, preventing the control system from maintaining water temperature within the desired range. If the actual water tempera-ture is below 70°C but the transmitter thinks the water is, for example, 75°C, it might not respond fast enough to prevent quality defects in the downstream process. As a result, we would categorize this as an operational consequence for the company, since it affects the bottom line.

If the temperature transmitter drifts low, the temperature control system will behave as if the water temperature is lower than it actually is. If the actual water temperature exceeds 80°C but the transmitter thinks the water is 75°C, the control system might not react in time to prevent quality defects and safety and environmental impacts — which are undeniable operational consequences. Maintenance task: To prevent the transmit-ter from drifting high or low, maintenance should perform an on-condition task to cali-

brate the transmitter on a predetermined interval. On-condition maintenance measures the output of a device, compares that value to a known standard, and adjusts it if the device is out of tolerance. This will ensure the drift doesn’t wander beyond the desired range, thereby avoiding associated manufacturing defects.

Now let’s examine this same failure for the second function.

Function 2 – To indicate the temperature of the process water within 2 percent of actual on the HMI.Failure: Temperature transmitter calibration drifts

FEAtUrEs

VIDEO BORESCOPES

90° Prism & Close-Focustips available!

Ideal for cooling tube inspection!We’ve improved the image quality in the new Hawkeye®

V2 with a higher resolution, more light sensitive camera, delivering bright, crisp, clear images! The new 5” LCD Monitor provides comfortable viewing, and intuitive, easy-to-use controls, allow photo and video capture at the touch of a button! We’ve increased the 4-way articulation range, and improved the feel. It’s still small, lightweight, portable, delivers great image quality, and is priced starting at only $8995. Available in 4 and 6 mm diameters. Optional 90° Prism and Close-Focus adapter tips available.

Made in USA

The NEW Hawkeye® Video Borescope!

Quickly inspect cooling tubesinside heat exchangers, turbineblades, and much more!

Bright, High-Res right, High-Res Bright, High-Res BVideo & PhotosVideo & PhotosLarge Range-of-FocusLarge Range-of-FocusQuality ConstructionQuality Construction4-Way Articulation4-Way Articulation4 & 6 mm Diameters4 & 6 mm DiametersStarting at only $8,995Starting at only $8,995

gradientlens.com800.536.0790

Visit us at:

Booth 4949Booth 4949Las Vegas, NVLas Vegas, NVMay 5 - 8, 2014May 5 - 8, 2014


Effect and consequence: Since the temperature display on the HMI is only for the operators’ information, if the calibration drifts and an incorrect temperature is shown on the HMI, it won’t impact the process. Because the only direct cost to the business is the cost of repairing the out-of-calibration transmit-ter, we’d categorize this as a non-operational consequence. Maintenance task: Because it isn’t impacting the bottom line in a meaningful way, this failure on its own doesn’t justify the cost of performing the on-condition task to calibrate the transmit-ter. Instead, we’d recommend a “no scheduled maintenance” task. But plant staff should be aware that the transmitter has this function, failure, effect and consequence in case something changes. For example, if the operators are now required to control the process based on the temperature reading on the HMI, they would know to adjust their maintenance tasks to ensure nothing is lost or missed.

Hidden failures — what you don’t know can hurt youFunction 3 - To be capable of alarming on the HMI and shutting down the process in the event the water temperature drops below 70°C.

Function 4 - To be capable of alarming on the HMI and shutting down the process in the event the water temperature exceeds 80°C.

Our transmitter is an integral component of the alarm cir-cuit and these two functions are among its most important for one reason: when they fail, the failure is hidden. Failure: High or low water temperature alarm circuit failsEffect and consequence: A hidden failure isn’t evident until you need the device to do its job and it doesn’t. If the low tem-perature or high temperature alarm circuit has failed and the process water temperature stays between 70°C and 80°C, the problem could go undetected indefinitely. If, as a result of a completely separate failure, the temperature drops below 70°C or rises above 80°C, we’ll find out the alarm circuit failed – but only because there have been potentially catastrophic qual-ity, process and safety impacts that could have been prevented if the alarm was working.Maintenance task: Maintenance staff must perform a fail-ure-finding test to validate the operation and set points of the high- and low-alarm and shutdown circuits. If the alarm fails during the test, a subsequent diagnosis will determine if the transmitter is responsible and it will be immediately repaired.

These failure-finding tests might include a mock input test where the temperature is raised or lowered to meet the thresh-old. The test might also confirm the set points by altering the system interface. The testing approach will depend on the importance of the failure and the testing technology available in the system. Failure: Temperature transmitter calibration drifts

Failure of the alarm circuit isn’t the only thing that will cause an alarm failure — the transmitter on its own drifting high or low can also be the culprit.Effect and consequence: If the temperature transmitter drifts low (let’s say by 5°C), the high-level alarm system will think the water temperature is less than it really is. If a completely sep-arate process or equipment upset causes water temperature to exceed 80°C, the alarm won’t activate until 85°C and the pro-cess will shut down too late, with a potential to impact safety, equipment health and the manufactured product. This is a hid-den consequence, since the operating team won’t know about the transmitter failure until the second failure occurs.

If the temperature transmitter drifts high, resulting in a temperature reading that is higher than the actual, there is the potential for the alarm to sound too late to prevent quality problems if a second failure causes the water temperature to drop below 70°C. It’s a less serious hidden consequence, but still worth preventing.Maintenance task: Perform an on-condition task to calibrate the transmitter on a predetermined interval. This task will ensure the drift doesn’t wander beyond the desired range, thereby increasing the availability of this component in the protective system. This will reduce the probability of a multiple

FEAtUrEs

We already calibrate our transmittersHow understanding builds a better maintenance plan

Let’s say your computerized maintenance manage-ment system already regularly generates a “calibrate transmitter” task. Why go through this process to con-firm that you’re already doing the right thing?

The answer is simple. If the maintenance team doesn’t know why it’s doing the calibration task, there can be important consequences.

First, team members who don’t understand the func-tion of the transmitter might think the calibration task is discretionary and prioritize other things. They might perform the task incorrectly. Or they might not perform the task at the right time. In each of these cases, there are potentially dire operational, safety and environmental consequences.

On the flip side, not understanding the function of the transmitter might mean team members perform the calibration task when it isn’t warranted, which might take equipment offline unnecessarily and increase the probability of introducing problems. For example, if your transmitter came with a temperature reading but there isn’t an operational consequence to a high or low read-ing, then there’s no need to spend time calibrating it.

When everyone fully understands the reasons for a maintenance task, it ensures the right work is getting done at the right time.


FEAtUrEs

failure and associated manufacturing defects and increased risk to health and safety.

ConclusionBy focusing on what something does, rather than what

it is, we’re able to stay in touch with the real reason we do maintenance — to preserve function at a desired perfor-mance standard.

In the case of the multi-function transmitter, our anal-ysis has shown that the exact same failure — calibration drift — can have significantly different consequences (some minimal, some potentially catastrophic and some hidden) and, as a result, require different maintenance tasks. We’ve also shown that calibration is a single task that can help preserve three important functions of the transmitter. Finally, we’ve pointed out that calibration on its own isn’t enough. The maintenance team should also be doing a failure-finding test to protect against a potentially catastroph-ic hidden failure in the alarm circuit.

It’s only when we thoroughly under-stand what a device is required to do in the context of the overall process (its function), what causes it to fail to per-form that function, and what happens when that failure occurs (the conse-quence), that we are well positioned to clearly define the most effective and effi-cient plan to identify and prevent poten-tial failures.

Without this understanding, we risk creating maintenance programs that result in too much maintenance — which impacts how efficiently the program uses money and personnel — or too little maintenance — which results in unantic-ipated failures and breakdowns.

What we’re suggesting is contro-versial in some circles. We’d love to hear what you think. Do we need to grow a maintenance program from function on up? Email us at [email protected] to share your thoughts. ~

Mike Schultz is the vice president of Solutions at APG. Schultz has worked in the fields of maintenance and reliability for more than 30 years and is a respected expert on equipment reliability programs. At APG, Schultz performed more than 100 RCM analyses for clients throughout North America and has trained and mentored more than 50 RCM2 facilitators and practitioners. He also teaches courses in PAS 55, Root Cause

Failure Analysis and Predictive Maintenance Technologies, and educates corporate leaders on the fundamentals of reliability. You may contact him by emailing [email protected]. Stace Smith is a reliability specialist at APG. Stace is a certified RCM2 Practitioner and has extensive experience conducting RCM2 and other reliability and maintenance analyses in Power Generation, Transmission and other asset intensive industries in North America. You may contact him by emailing [email protected].

PARTS | SERVICES | REPAIRS

A Division of MD&A

Renewal Parts Maintenance 4485 Glenbrook Rd. | Willoughby, OH 44094

ph. 440-946-0082 | www.RenewalParts.com

Refurbishment

Steam Turbine High Pressure

Valve Actuator


MAiNtENANCE MAttErs

Centrifugal compressor surge and compressed air systems

By DeVon A. Washington, PhD

In a powerhouse, several critical processes depend on a consistent reliable supply of compressed air. Compressed air systems in a power plant can be divided into three major cat-egories: (1) house service air, (2) instrument air and (3) soot blowing air. The air supplied by these systems power pneumat-ic tools and actuators, as well as soot blowers. Providing this essential resource requires the proper selection of a compressor and compressor controls, in addition to a robust system design for the intended application.

This article examines the replacement and upgrade of two reciprocating soot blowing air compressors (SBAC) with two centrifugal compressors, one of which is a 100 percent back-up, for two 160 MWe generating units. Additionally, it surveys modifications to the corresponding soot blowing system.

Once installed, the first centrifugal compressor experienced intermittent periods of surge. It was determined that the inter-mittent surging was induced by an unfavorable piping config-uration and limited receiver tank capacity. A detailed overview is given regarding compressor selection and controls, causes of surge phenomena and key system design modifications imple-mented to eliminate the surge condition.

IntroductionIn the 1980s, to comply with stricter emissions regu-

lations, the power industry began to explore burning fuel blends with low sulfur subbituminous coal and bituminous coal. Subbituminous coal has a relatively low ash-fusion tem-perature and higher heating value (HHV). Burning coal with these properties, in boilers originally designed for bituminous coal, requires a significant increase in soot blowing frequency and duration. (Figure 1) Retrofitting legacy systems originally designed for reciprocating compressors with centrifugal com-pressors can pose several subtle but significant challenges for design and plant engineers. Some of these issues are presented here.

BackgroundThe two generating units under consideration were com-

missioned in 1955 and 1958, respectively; both units share a common soot blowing system. Their nameplate capacities are approximately 160 MWe each. The original soot blowing system comprised furnace-wall and convection-surface soot blowers, two reciprocating compressors, primary and second-ary receiver tanks, and a piping network. (Figure 2) Sustaining optimum boiler performance while burning fuel blends with increased fractions of Powder River Basin (PRB) coal, strained the soot blowing system. Eventually, soot blowing became a

limiting factor in burning fuel blends with larger percentages of PRB coal.

A project was initiated to determine which components in the system should be upgraded to support strategic plans to further increase PRB coal consumption. The scope of work included replacing and upgrading several existing soot blowers, adding waterlance soot blowers, replacing the reciprocating compressors with centrifugal compressors, as well as updating the soot blowing control logic.

The first centrifugal compressor was added to the existing system with no significant changes to the distribution piping or receiver capacity. After commissioning the new compressor, it began to experience several brief intermittent surge events. Examining the performance data logs showed the compressor would momentarily surge, at which time the antisurge con-trols would quickly autocorrect the unit and return it to stable operation. This behavior appeared to be completely random, but during an extended period of time a trend began to devel-op which showed this cyclic behavior was occurring under a definite set of conditions. However, it was unclear whether the surge events were associated with changes in operating, system or ambient conditions; or some combination of the three. A comprehensive investigation was initiated to establish the root cause of the surge events.

Compressors and compressed air systemsClassifications and descriptions

There are two fundamental classifications of compressors: positive displacement and dynamic. Each classification can be subdivided further based on mechanics of motion and meth-

Figure 1. Soot blowing air consumption for fuel blends including PRB coal.


od of work transfer. Reciprocating compressors are a type of positive displacement machine, where work transfer to the gas is accomplished using a piston-cylinder assembly. Centrifugal compressors are dynamic machines, which utilize a rotor assembly to impart work to the gas[1].

The original SBACs were horizontally balanced-opposed, three-stage reciprocating compressors with double-acting pis-tons and direct drive induction motors. The capacity control system was automatic three-step constant-speed unloading. The unloading devices were pneumatic suction valve unload-ers. (Table 1) The compressors also were equipped with two interstage coolers and one aftercooler; the cooling medium was water.

The upgraded SBAC is a three-stage centrifugal compressor with a direct drive induction motor, with constant pressure modulation capacity control. (Table 1) The capacity control components include an inlet butterfly valve (IBV) and bypass valve (BV). Like the original SBACs, the centrifugal com-pressor is also equipped with two interstage coolers and one aftercooler.

Performance and selectionWhen selecting a compressor for a particular application,

the primary parameters to consider are flow rate, efficiency and pressure rise. Other important factors include capital cost, reliability, maintainability and controllability.

In general, reciprocating compressors are selected for appli-cations that require low to moderate flow rates. (Figure 3) The maximum flow rate, which can be achieved by a reciprocat-ing compressor, is a function of cylinder volume, number of throws and motor speed. Reciprocating compressors are capa-ble of generating large discharge pressures and typically require a smaller capital investment. Additionally, they yield higher compression efficiencies during partial load operation. The large number of moving parts inherent to their design reduces reliability and increases maintenance[2-4].

For processes that demand medium to high flow rates at moderate pressures, centrifugal compressors are ideal. (Figure 3)

Table 1 – Compressor Type

Design Specifications Reciprocating Centrifugal

Number of Stages 3 0.3

Discharge Pressure 315 psig 300 psig

Discharge Temperature 100°F 100°F

Capacity 1150 acfm 1925 acfm

Capacity Control Method

Constant-Speed Unloading

Constant Pressure Modulation

Motor Power 350 bhp 1750 bhp

Motor Speed 585 rpm 3575 rpm

Table 1 – Compressor comparison summary

MAiNtENANCE MAttErs A REVOLUTION INBRUSH CHANGING FOR

TURBINE-GENERATORRELIABILITY

It is now possible to easily and safely change brushes onlive equipment! The solution the industry has long awaited is now available from the leader in brush holdertechnology, Fulmer Company.

The NEW Cartridge-Style Plug-In Brush HolderThe NEW Cartridge-Style Plug-In Brush Holder

■ Brushes changed on-line, without service interruption, using a simple changing fixture

■ Direct replacement for many OEM styles with no rigging modification required

■ Longer brush box to provide better brush support and allow the use of a 4” brush, increasing brush life and reducing the frequency of changes

■ Rugged, lightweight and ergonomically built

■ Removable insulated handle with flash guard

3004 Venture CourtWestmoreland Industrial Park IIIExport, Pennsylvania 15632Phone 724-325-7140www.fulmercompany.com

Fulmer CompanyA Wabtec Company

BRUSH CHANGING MADE EASY

THE FULMER

WAY.

FlmrEnrgyTch 1/2 pg-4CWbtch 3/5/12 2:26 PM Page 1


The maximum flow in a centrifugal compressor is dictated by the choke point. Unlike reciprocating compressors, centrifugal compressors have smaller turndowns and are not well suited for low flow, partial load operation. With significantly fewer mov-ing parts, centrifugal compressors are much more reliable and require less maintenance. The improved reliability and reduced maintenance is accompanied by an increase in initial capital investment. Generally, this is associated with the manufacturing of more complex components (impellers)[2-4].

Surge phenomenonSurge is classified as an instability phenomenon, character-

ized by pressure fluctuations and, in the most extreme cases, reverse flow. A precursor to surge is stall, which is caused by the separation of flow from an aerodynamic surface due to an adverse pressure gradient. If this unstable aerodynamic state persists and intensifies, then surge ensues[5]. Surge can be sub-divided into four categories: (1) mild surge, (2) classic surge, (3) modified surge and (4) deep surge. The impact of surge ranges from reduced compressor performance (mild surge) to severe mechanical damage (deep surge). Mild surge is described by small pressure fluctuations with no reversal in flow. Whereas, deep surge is accompanied by large pressure fluctuations and reverse flow is much more likely. A more detailed description of all four categories is provided here[6]. Surge does not occur in reciprocating compressors. The discussion here is centered on the occurrence of surge in centrifugal compressors.

Antisurge controlFor centrifugal compressors there are two common meth-

ods of capacity control: inlet throttling and variable speed. Inlet throttling is used with compressors that have constant speed drives, and variable speed control with compressors that have variable speed drives. The type of surge control required is a function of the type of capacity control in use. Surge control systems are designed to anticipate and prevent surge. Two fun-damental types of surge detection are static and dynamic. Static detection prevents surge by controlling the compressor in such a way that some predefined operating condition is avoided.

MAiNtENANCE MAttErs

Figure 2. An original soot blowing system with reciprocating compressors.

Figure 3. Capacity characteristics of reciprocating and centrifugal compressors

1224 North Utica | Tulsa, OK 74110T: 918-587-6649 | E: [email protected]

W: www.topog-e.com

The Topog-E Gasket Company produces gaskets

to fit every boiler in production today. Topog-E® Series

180 molded rubber gaskets are special for many reasons. Over forty years as the industry standard have

given us feedback as to why!

• Steam pressure vessels

• Hot water heaters• Demineralizers• Steam humidifiers• Water purifiers• Refrigeration units• Liquid treatment

vessels

• Compressed air tanks

• Filtering units• Dryer cans in

paper mills• Water towers• Water softeners• Deaerators• Make-up tanks

Applications:


Dynamic detection is capable of assessing the onset of surge in real-time and avoids surge based on the actual system condi-tions at hand[7].

Although surge control systems based on static detection are more common, dynamic control possess several advantages. Typically, static antisurge controls are designed to maintain a minimum flow through the compressor by either recirculating a fraction of the discharge flow or venting it to the atmo-sphere. The margin between the actual surge condition and the predefined setpoint might be sizeable. While this protects the compressor from surge, there is a significant cost associated with wasted energy due to excessive recirculation or premature venting[7].

Figure 4 shows the performance map for the upgraded compressor. Similar to inlet throttling, surge control here is accomplished by venting a fraction of the discharge to the atmosphere when the operating point reaches the minimum load setpoint.

Even with good antisurge controls there are still condi-tions that make it difficult to prevent surge altogether. Some examples include: insufficient rise to surge, changes in system discharge pressure and rapid excursions in system demand. Rise to surge is a metric used to assess a compressor’s ability to achieve stable operation after some perturbation. These condi-tions can be affected by changes in ambient, system or operat-ing conditions.

System designThe fundamental components in a compressed air system

are: compressors, receiver tanks and distribution piping. The attributes of each component affects the entire system and must be matched appropriately for reliable stable operation.

Receiver tanks serve many purposes in a compressed air system; one of the most important is acting as a buffer between the load and the compressor. For soot blowing applications, pressure pulses are generated in the system due to the starting and stopping of soot blowers during the soot blowing cycle. Receiver tanks also provide additional capacity for situations when demand exceeds the maximum capaci-ty of the compressor. When sized appropriately they can be used to optimize the frequency at which a compressor is loaded and unloaded[8].

For larger systems, usually there is a primary and second-ary receiver tank. The purpose of the primary receiver tank is to allow the compressor to operate within a given range of discharge pressures. The role of the secondary receiver tank is to ensure that an uninterrupted supply of compressed air reaches the end use under all system and operating condi-tions[9].

The configuration of the piping network also is a signif-icant feature of the system. For inlet and discharge piping directly connected to the compressor, typically the original equipment manufacturer (OEM) will provide general guide-lines regarding their arrangement. However, the layout of the remaining system is left to application and plant engineers.

In general, pipe sizes should be selected such that, after the receiver tank, pressure drops are less than 10 percent of the design operating pressure. For soot blowing systems, the over-arching constraint is that each individual soot blower receives the proper supply of air in accordance with OEM specifi-cations. Furthermore, looped headers should be used when possible. This provides an additional pathway for supply air to reach process equipment during periods of peak demand[8].

MAiNtENANCE MAttErs

1.205.824.0004 REVERECONTROL.COM

REVERECONTROL SYSTEMS

From fuel handling systems to switchgear and MCCs, your control systems are only as good as the engineers who design them and the people who build them. Our people combine key industry experience with professional performance to deliver good projects on schedule and on budget.

ENGINEERED TO SERVEPOWERAPPLICATIONS

Control your systems.Control your plant. Control your business.

Figure 4. Centrifugal compressor performance map for constant pressure modulation capacity control.


MAiNtENANCE MAttErs

ConclusionsThe results of the analysis conducted on the modified sys-

tem concluded there was a confluence of issues associated with variations in ambient, operating and system conditions, which led to the surge events experienced by the newly upgraded centrifugal compressor.

Many of the surge events occurred during seasonal changes, particularly when ambient air temperatures were increasing. An increase in ambient air temperature acts to reduce the surge threshold on the performance map, resulting in an insufficient rise to surge. Under these circumstances the compressor can become sensitive to relatively small fluctuations in system demand, which is typical for soot blowing applications, causing the compressor to surge.

The minimum load setpoint (Figure 4) corresponds to a specific discharge pressure. If the discharge pressure changes, then the minimum load setpoint needs to be adjusted accord-ingly. For example, if the discharge pressure was increased and all other parameters held fixed, the compressor turndown would be reduced and surge would occur sooner. For soot

blowing applications, as the types of soot blowers and soot blowing control logic are adjusted, it is important the minimum load setpoint be tuned to match the new operat-ing conditions.

During soot blowing there are rapid changes in system demand that require the controller and the various control compo-nents to respond quickly. These fluctuations in system demand can be attenuated by incorporating appropriately sized receiver tanks into the system.

When the first centrifugal compressor was originally added to the legacy system, there were no notable changes made to the distribution piping or receiver capacity. The approach taken to resolve the issues surrounding the surge events included mod-ifying the piping network and adding addi-tional receiver capacity. The modified system is shown in Figure 5.

The discharge capacity of one upgraded centrifugal compressor is double that of both reciprocating compressors combined. This required that the main distribution header layout be streamlined and the pipe diameter increased to 8˝. The main header for the legacy system was 4˝-6˝ in various locations and had more restrictive piping runs. Additionally, receiver capacity of the system was increased substantially. As a rule of thumb, for every cubic foot of actual discharge, one gallon of storage capacity is desired. The secondary receiver tank shown in Figure 2 has a storage capacity of 3,750 gallons, which was more than adequate for

the legacy compressors. However, at a minimum, the new compressor requires at least 5,000 gallons of storage. Upon analyzing system pressure decay for the modified soot blow-ing process, it was determined that in addition to maintaining the storage capacity of the secondary receiver tank (3,750 gallons), that two new primary receiver tanks would be need-ed; each with a storage capacity of 5,025 gallons. (Figure 5)

Performance maps provided by OEMs characterize com-pressor performance for a specific set of conditions[10]. Ambient conditions and the system configuration in the field can deviate significantly from those outlined in the performance test code. It is important that consideration be given to these differences, as well as their impact on compressor performance. The aforementioned system upgrades eliminated the surging issues. The compressor has maintained stable operation for the past several years continuously, providing compressed air with-out incident to this very day. ~

Figure 5. Upgraded soot blowing system with centrifugal compressors.


AcknowledgementsThe author would like to acknowledge Jon W. Carpenter, senior engineer with Consumers Energy, Generation and Operations, for providing tremendous insight into the many nuances regarding the soot blowing system’s maintenance and operational histories. The author also would like to recognize LeRoy N. Reiss, PE, principal engineer; James B. Lewis, PE, executive engineer (ret.); and Scott D. Thomas, PE, executive manager of Consumers Energy, Engineering Services Department.

References1. D. H. Robison and P. J. Beaty, “Compressor Types,

Classifications, and Applications,” in Proceedings of the 21st Turbomachinery Symposium, College Station, TX, 1992, p. 183.

2. M. P. Boyce, “Principles of Operation and Performance Estimation of Centrifugal Compressors,” in Proceedings of the 22nd Turbomachinery Symposium, College Station, TX, 1993, p. 161.

3. P. Gallick, G. Phillippi, and B. F. Williams, “What’s Correct for My Application - A Centrifugal or Reciprocating Compressor?,” in Proceedings of the 35th Turbomachinery Symposium, College Station, TX, 2006, p. 113.

4. M. O. Khan, “Basic Practices in Compressor Selection,” presented at the International Compressor Engineering Conference, West Lafayette, IN, 1984.

5. D. Japikse, “Stall, Stage Stall, and Surge,” in Proceedings of the 10th Turbomachinery Symposium, College Station, TX, 1981, p. 1.

6. B. de Jager, “Rotating Stall and Surge Control: A Survey,” in Proceedings of the 34th IEEE Conference on Decision and Control, New Orleans, LA, 1995, p. 1857.

7. M. P. Boyce, W. R. Bohanna, R. N. Brown, J. R. Gaston, C. Meher-Homji, R. H. Meier, et al., “Practical Aspects of Centrifugal Compressor Surge and Surge Control,” in Proceedings of the 12th Turbomachinery Symposium, College Station, TX, 1983, p. 147.

8. Fossil Maintenance Applications Center, “Compressed Air System Maintenance Guide, TR-1006677,” EPRI, Palo Alto, CA, 2002.

9. Office of Energy Efficiency and Renewable Energy, “Improving Compressed Air System Performance,” U.S. D.O.E., Washington, DC, 2003.

10. American Society of Mechanical Engineers, “Performance Test Code on Compressors and Exhausters, PTC 10-1997,” ASME, New York, NY, 1997.

Dr. DeVon A. Washington is a senior engineer with Consumers Energy, Engineering Services Department in Jackson, Mich. His primary responsibilities consist of providing technical leadership regarding thermal fluids, turbomachinery and energy conversion systems. You may contact him by emailing [email protected].

MAiNtENANCE MAttErs

We Connect the DotsOver the past 30 years, Structural Integrity has built a team of over

200 industry experts providing comprehensive solutions to the energy industry. We connect the dots from problem to resolution, from regulations to compliance, from nondestructive examination, to engineered solutions, even custom equipment.

You can look to us for our:• Knowledge of power plants, codes, and how

things work.• Extensive experience and leadership.• High quality, hard work, and responsiveness.

Call us today and we’ll connect the dots for you.

( 8 7 7 - 4 S I - P O W E R )8 7 7 - 4 7 4 - 7 6 9 3

Scan the QR Code for more information

www.structint.com/energy-tech

FROMENGINEERINGANALYSIS TO ADVANCED NDE,LOOK TOSTRUCTURAL INTEGRITY

20 ENERGY-TECH.com ASME Power Division Special Section | May 2014

Impact of harmonics from distributed energy resources on

distribution networks By Reza Arghandeh, UC Berkeley, Berkeley, Calif.; Ahmet Onen, Virginia Tech, Blacksburg, Va.; Jaesung Jung,

Virginia Tech, Blacksburg, Va.; Danling Cheng, EDD Inc., Blacksburg, Va.; Robert Broadwater, Virginia Tech, Blacksburg, Va.; and Virgilio Centeno, Virginia Tech, Blacksburg, Va.

AsME FEAtUrE

Introduction and previous worksSmart grid realization involves a steady increase in invert-

er-based components like Distributed Energy Resources (DER), energy storage systems and plug-in electric vehicles. These kinds of renewable DER applications relieve environmental concerns, but raise concerns about harmonic distortion effects on voltage and current waveforms [1]. Harmonics can result in various dam-age and errors in distribution networks [2].

This paper investigates the harmonic propagation from multi-DER sources such as wind, solar and energy storage systems. It also aims to study the interactive effect of harmonic sources on each other via the presented simulations and case studies. Current and voltage harmonic disturbances are quantized via the Total Harmonic Distortion (THD) and other indices.

The paper will discuss harmonics phenomena in distribution networks, the principle factors for harmonic analysis in distribu-tion networks and a discussion of simulation results.

Harmonic issues in distribution networksHarmonics propagation in distribution networks cause con-

ductors to overload with increases in r.m.s currents. Neutral conductors overload, capacitor banks overload and premature aging and protection systems malfunction – all examples of har-monic problems for distribution networks [3]. Voltage harmonic affects shunt connected network components like capacitor banks and electric motors. Current harmonics affect series con-nected components like transformers [14].

Incorrect operation of relays under distorted voltage or current cause unwanted tripping of the supply [9]. Moreover, harmonic distortion can affect energy metering equipment and decrease metering accuracy that can change electricity market transactions [15, 16].

Today’s high level penetrations of distributed energy resourc-es (solar, wind, fuel cell, microturbines, etc.) and the adoption of electric vehicles with a variety of electronic interfaces creates additional concerns for utilities about how harmonic sources interact with each other.

Harmonic characteristics in distribution networksHarmonic assessment analytics

The deviation of perfect sinusoidal voltage and current sig-nal is expressed as distortion. Harmonic distortion is modeled by adding sinusoidal components with higher frequency to the

signal of the fundamental frequency. A distorted voltage and cur-rent waveform are expressed by the Fourier series as given by:

Equation 1

Equation 2

Where: • I

Total and V

Total are the nonsinusoidal current and voltage at

the measurement point• I

h and V

h are current and voltage r.m.s values for the hth

harmonic order

Figure 1. Harmonic analysis algorithm

May 2014 | ASME Power Division Special Section ENERGY-TECH.com 21

AsME FEAtUrE

• θh and φ

h are harmonic current and voltage phase angles

• ω0 is the fundamental angular frequency and h is the har-

monics order

The Individual Harmonic Distortion (IHD) index presents the percentage ratio of current or voltage in the hth harmonic order with respect to the fundamental value. IHDI and IHDV are the IHD for current and voltage respectively, as given by:

Equation 3

Equation 4

Total Harmonic Distortion (THD) is an important index that is usually used for power quality assessment in distribution and transmission systems. It considers the contribution of each harmonic component on the total signal. THD is defined for current and voltage respectively as [19]:

Equation 5

Equation 6

Where: • THDI and THDV are THD values for current and volt-

age, respectively

ASME Power Division: Advanced Energy Systems & Renewable Committee

A Message from the ChairThe Advanced Energy Systems

& Renewable Committee is one of a number of committees in the ASME Power Division. The committee promotes innovative concepts concerning the design, operation, maintenance, testing and improvements to renewable energy technology. These tech-nologies include wind turbines, solar panels, geothermal/hydro-

electric power, Smart Grid and energy storage.One of the committee’s functions is to sponsor the

Advanced Energy Systems & Renewable Energy track in the ASME Power Conference each year. The committee promotes authors to prepare papers in various topics and present them at the annual ASME Power Conference. Each year the committee receives numerous technical papers from authors around the world. These papers are peer reviewed by qualified engineers in their respective fields. After being reviewed, the track chair approves them before the author can present them at the conference.

This year, the Advanced Energy & Renewable Track reviewed 40 Abstracts and accepted more than 30 papers that will be presented at the ASME Power Conference, which will be held in Baltimore, Md., from July 28th – 31st. The number of papers presented in our track has grown exponentially the past few years. The technical aspect of the papers also has evolved.

This month’s ASME feature is an excellent example of the type of material presented during one of our technical session at the 2013 ASME Power Conference in Boston, Mass. We hope that you enjoy it.

John K Fall – PEAmerican Electric PowerColumbus, OH [email protected]

July 28-31, 2014Baltimore, Maryland

www.asmeconferences.org/Power2014

Figure 2. Schematic of distribution network model


• I1 and V

1 are the current and voltage r.m.s. values for the

fundamental frequency

When considering harmonic distortion, the conventional definition of power factor is modified to account for the contri-bution of higher frequencies on the power factor [1]. The modi-fied power factor is called Total Power Factor (TPF). Equation 7 shows the relationship between TPF and THD [3, 20]:

Equation 7

Where: • δ

1 is the angle between voltage and current at the funda-

mental frequency• The cos(δ

1) called displacement power factor and the factor

1/√1+THDI2 is called distortion power factor

Interactive harmonic analysis algorithmFor harmonic propagation assessment, the fundamental fre-

quency power flow needs to be calculated first. The power flow study provides steady-state circuit performance under the nor-mal operating condition.

To calculate the higher frequency voltages and currents, harmonic power flows are calculated after the fundamental fre-quency power flow. To run the harmonic power flow, the circuit is modified to update impedance in distribution network com-ponents for each frequency order.

For the power flow calculation, the Distributed Engineering Workstation software package is used. Figure 1 depicts the algo-rithm for the harmonic analysis.

Case studies and assumptionsIn this paper, a detailed physical-based distribution network is

applied. The circuit model includes unbalanced, multiphase and single-phase loads. The circuit is 13.2 kV, Y-connected with 329 residential and commercial customers. The circuit contains two sets of three-phase harmonic sources representing inverters con-nected to DER sources. It assumed that all three phases have the same current magnitudes. Figure 2 presents the circuit model.

The dominant current and voltage harmonics observed through the simulation are of the 3rd, 5th, 7th, 9th and 11th

orders. Harmonics of higher orders are neglected due to their small values.

The inverter technology and DER type are not considered in this research. The research aim is the harmonic impact study, apart from the harmonic source technology.

Simulations and resultsHarmonic emission in distribu-

tion networks is affected by a num-ber of factors related to the har-

monic sources and distribution network characteristics. In this paper, multi-harmonic source interactions are evaluated based on different values of harmonic source magnitudes and angles. A sensitivity analysis is conducted to investigate the impact of har-monic source magnitude and angle variation on other harmonic sources. Total Power Factor and Total Harmonic Distortion for voltage and current waveforms are considered.

Multi-harmonic source magnitude variationTo show the effects of harmonic current source magnitude

on the THD at the substation, simulation is conducted for dif-ferent current magnitudes in the range from zero to 10 (0, 2, 4, 6, 8, 10) amperes for both harmonic sources.

Figure 3 shows the interactive THDV variation in phase A for two harmonic sources in the circuit. As harmonic current magnitudes increases, the two harmonic sources act together to increase the THDV at the substation.

In Figure 4, the THDI for phase A is presented. Similar to the THDV, the total harmonic distortion for the current wave-form increases with the increase in magnitude of the two har-monic sources.

Figure 5 depicts the total power factor calculated values from Equation 7. The surface in Figure 5 is descending gradually to lower power factor values as harmonic source magnitudes increase. Higher total power factor values mean less loss in the distribution network.

Figures 3-5 show how two harmonic source magnitudes increment interactively to increase the total harmonic distortion and energy loss in distribution networks. Simulation results for phase B and phase C show the similar trend for THDV, THDI and TPF.

Multi-harmonic sources phase angle variationIn systems with a number of harmonic sources, the inject-

ed harmonic currents from harmonic sources have vectorial characteristics. Therefore, the interactive effect of harmonic distortion is not defined just by the magnitude of harmonic sources. To identify the impact of phase angle variations for two source interactive harmonic distortion, the sensitivity analysis is conducted with the phase angles for both harmonic sources being varied as: 0 degrees, 15 degrees, 30 degrees, 45 degrees, 60 degrees, 75 degrees and 90 degrees.

AsME FEAtUrE

Figure 3. Multi-harmonic source THDV for phase A for different current magnitudes

Figure 4. Multi-harmonic source THDI for phase A, for different current magnitudes.


Similar to the previous section, harmonic distortion is evaluated by THDV, THDI and TPF. In this case, the harmonic indices are pre-sented for three phases.

As illustrated in Figure 6, the THDV vs. angle plot for each phase is significantly differ-ent than the other phases. The reasons for this are the different phase loadings, the three phase mutual couplings and network topology. The currents in phase A, B and C are 126.8 amps, 67.3 amps and 73.3 amps, respectively. The Thevinen impedances as seen by harmonic sources are presented in Table 1.

In terms of network topology, the portion of single phase lines and underground cables in the distribution network affects the har-monic emissions because of the capacitance characteristics of such conductors.

In Figure 6-A, the THDV has a minimum value at zero phase angle for both sources. The maximum value (THDV=1.27) is achieved with a 90-degree phase angle in both sources. Figure 6-B has a saddle-shaped surface with the saddle point at a 45-degree phase angle in both sources. Maximum THDV for phase B happens at the 0-degree phase angle in both sources (THDV=1.02). Figure 7-C is similar to a hemispherical plane with its maximum at a 45-degree phase angle for both sources (THDV=1.1).

To show the impact of phase angle variation on harmonic current distortion, Figure 7 presents THDI for each of the three phases. Figure 7-A shows the THDI variation over the harmonic source phase angles. The canyon on the surface shows that at the minimum THDI point (near zero) the two harmonic sources cancel out each other. This observation is crucial for harmonic clearance in systems with multi-harmonic sources. These minimum THDI conditions happen in the following phase angle couples, {(30,90), (45,75), (60,60), (75,45), (90,30)} where the first term is for the first harmonic sources phase angle, and the second term belongs to the second harmonic source phase angle. Figure 7-B illustrates THDI for phase B. The maximum THDI happens at 30 degrees for both sources (THDI=17.3). In Figure 7-C, THDI for phase C is presented. In this phase, maximum THDI occurs at 90 degrees for the two harmonic sources (THDI=16.3).

Figure 8 shows the total power factor (TPF) for phase A. The surface has a semi-hilltop form. Higher total power factors mean less loss, which leads to more profit for the distri-bution network operator.

The maximum total power factor points are presented in Table 2.Figure 9 illustrates TPF for phase B. The TPF plot is similar to a semi-spherical

surface with a minimum value at 30-degree phase angle for both harmonic sources (TPF=0.8406). The maximum TPF happen at (0 degrees, 90 degrees) and (90 degrees, 0 degrees) points with TPF=0.846.

AsME FEAtUrE

- 24 / 7 -EMERGENCY SERVICE

IMMEDIATE DELIVERY

CALL:800-704-200210HP TO 250,000#/hr

250,000#/hr Nebraska 750 psig 750OTTF150,000#/hr Nebraska 1025 psig 900OTTF150,000#/hr Nebraska 750 psig 750OTTF150,000#/hr Nebraska 350 psig115,000#/hr Nebraska 350 psig80,000#/hr Nebraska 750 psig75,000#/hr Nebraska 350 psig60,000#/hr Nebraska 350 psig40,000#/hr Nebraska 350 psig20,000#/hr Erie City 200 psig10-1000HP Firetube 15-600 psig

ALL PRESSURE AND TEMPERATURE COMBINATIONSSUPERHEATED AND SATURATED

RENTAL FLEET OF MOBILETRAILER-MOUNTED BOILERS

75,000#/hr Optimus 750 psig 750OTTF75,000#/hr Nebraska 350 psig60,000#/hr Nebraska 350 psig50,000#/hr Nebraska 500 psig40,000#/hr Nebraska 350 psig30,000#/hr Nebraska 350 psig75-300HP Firetube 15-600 psig

ALL BOILERS ARE COMBINATION GAS/OILENGINEERING • START-UP • FULL LINE OF BOILER

AUXILIARY SUPPORT EQUIPMENT.Electric Generators: 50KW-30,000KW

WEB SITE: www.wabashpower.com847-541-5600 • FAX: 847-541-1279

E-mail: [email protected]

wabash444 Carpenter Avenue, Wheeling, IL 60090

POWEREQUIPMENT CO.

BOILERSSELL•RENT•LEASE

••••••••••••••••••••••••••••••••••

••

••••••••••••••••••••••••••••••••••

••

Figure 6. Multi-harmonic source THDV for phase A (Figure A), phase B (Figure B), phase C (Figure C) for different current phase angles.

Figure 7. Multi-harmonic source THDI for phase A (Figure A), phase B (Figure B), phase C (Figure C) for different current phase angles.

Figure 5. Multi-harmonic source total power factor for phase A, for different current magnitudes.


AsME FEAtUrE

Figure 10 shows TPF values for phase C. It has a saddleshaped surface. The max-imum TPF points occur at (0 degrees, 90 degrees) and (90 degrees, 0 degrees) with TPF=0.865.

The presented observations show that considering Total Power Factor (TPF) in addition to the Total Harmonic Distortion (THD) provides a better picture of multi-harmonic sources, impacts and interactions. The impact of phase angle vari-ation on interactive multi-source harmonic distortion is especially helpful for har-monic distortion minimization and power factor improvement. Any improvement in power factor via the multi-harmonic sources’ control has economic value for distri-bution network operators and stockholders.

ConclusionHarmonics related to DER inverters and similar inverter based components raise

concerns for utilities and customers.Specifically, the interactive impact of harmonic sources on each other needs to be

explored. In this paper, an impact study is carried out to evaluate interactive multi-source harmonic distortion. Based on commonly used harmonic indices, a number of assessments are conducted. The interactive effect of harmonic source magnitudes and angles are considered in terms of THDV, THDI and Total Power Factor.

The sensitivity analysis shows the impact of each harmonic source magnitude and angle on total power factor of the distribution network. Consequently, it helps to find out how the multi-harmonics source interactions impact the distribution network efficiency and operational cost. ~

Editor’s note: This paper, PWR2013-98150, was printed with permission from ASME and was edited from its original format. To purchase this paper in its original format, or find more information, visit the ASME Digital Store at www.asme.org.

References1. F. C. De La Rosa, Harmonics And Power Systems: CRC Press Inc., 2006.2. R. Arghandeh, A. Onen, J. Jung, R. Broadwater, “Harmonic Interaction of Multiple

Distributed Energy Resources in Power Distribution Networks”, Electric Power System Research, Elsevier, vol 105, 2013, pp 124-133.

3. S. Volut, Electrical Installation Guide According to IEC Standards. Franse: Schneider Electric, 2009.

4. M. H. Bollen and I. Gu, Signal processing of power quality disturbances vol. 30: Wiley-IEEE Press, 2006.

Table 1 – Network Thevenin Impedance, as seen by harmonic sources. Table 2 – Maximum TPF points for phase A with different phase angles.

Table 1– Network Thevenin Impedance as Seen by Harmonic Sources

0 1 2

Harmonic Source 1

0 0.8063+ 2.9308j

-0.0102+ 0.0138j

0.0086+ 0.0155j

1 0.0086+ 0.0155j

0.8063+ 2.9308j

-0.0102+ 0.0138j

2 -0.0102+ 0.0138j

0.0086+ 0.0155j

0.8063+ 2.9308j

Harmonic Source 2

0 1.0368+ 3.6547j

-0.0122+ 0.0155j

0.0098+ 0.0179j

1 0.0098+ 0.0179j

1.0368+ 3.6547j

-0.0122+ 0.0155j

2 -0.0122+ 0.0155j

0.0098+ 0.0179j

1.0368+ 3.6547j

Table 2 – Maximum TPF Points for Phase Awith Different Phase Angles

H Source 1. Angle (Deg)

H Source 2. Angle (Deg) Delta Angle TPF

Max TPF 1 30 90 - 60 0.871

Max TPF 2 45 75 - 30 0.871

Max TPF 3 60 60 00 0.871

Max TPF 4 75 45 + 30 0.871

Max TPF 5 90 30 + 60 0.871

Figure 8. Multi-harmonic source total power factor for phase A for different current phase angles.

Figure 9. Multi-harmonic sources total power factor phase B for different current phase angles.

Figure 10. Multi-harmonic sources total power factor phase C for different current phase angles the distribution net-work, “Power Electronics, IEEE Transactions on, vol. 19, pp. 1586-1593, 2004.”


AsME FEAtUrE

5. A. Cataliotti, V. Cosentino, and S. Nuccio, “Static Meters for the Reactive Energy in the Presence of Harmonics: An Experimental Metrological Characterization,” Instrumentation and Measurement, IEEE Transactions on, vol. 58, pp. 2574-2579, 2009.

6. X. Wilsun and L. Yilu, “A method for determining customer and utility harmonic contributions at the point of common coupling,” Power Delivery, IEEE Transactions on, vol. 15, pp. 804-811, 2000.

7. G. J. Wakileh, Power Systems Harmonics Fundamentals, Analysis and Filter Design. Germany: Springer, 2001.

8. L. Cividino, “Power factor, harmonic distortion; causes, effects and consider-ations,” in Telecommunications Energy Conference, 1992. INTELEC ’92., 14th International, 1992, pp. 506-513.

Reza Arghandeh joined the University of California-Berkeley, California Institute of Energy and Environment in 2013. He is also vice-chair of the ASME Advanced Energy Systems committee. He received his Ph.D. in Electrical Engineering and M.S. in Industrial Engineering from Virginia Tech. He also holds a M.S. in Mechanical Engineering from the University of Manchester, UK. You may contact him by emailing [email protected]. Ahmet Onen received his Ph.D. from Virginia Tech – Electrical and Computer Engineering Department in 2014. He received the M.S. degree in Electrical engineering from Clemson University in 2010. He is a member of DEW software developing team for power systems planning and operation. His research interests are distribution system reliability, storm outage and reconfiguration, Distributed Series Reactance (DSR), and Smart Grid optimization, control and economic analysis. You may contact him by emailing [email protected]. Jaesung Jung is the assistant scientist at Brookhaven National Lab. He received the Ph.D. and M.S. degree from the Department of Electrical and Computer Engineering at Virginia Tech and North Carolina State University, respectively. His research interests include the development and deployment of renewable and sustainable energy technologies. You may contact him by emailing [email protected]. Danling Cheng received a B.S. from Huazhong University of Science and Technology, Wuhan, China in 1996, and her Ph.D. and M.S. from Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, Va., in 2002 all in Electrical Enginering. Currently, she is a software engineer in the EDD Inc. She was an instructor in Wuhan University of Technology, Wuhan, China, from 1996-2000. You may contact him by emailing [email protected]. Robert P. Broadwater received the B.S., M.S. and Ph.D. degrees in electrical engineering from

Virginia Polytechnic Institute and State University, where he now is a professor in the department of electrical and computer engineering. You may contact him by emailing [email protected]. Virgilio Centeno received the M.S. and Ph.D. degrees in electrical engineering from Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, in 1988 and 1995, respectively. He worked as a project engineer at Macrodyne Inc., Clifton Park, N.Y., in the development of phasor measurement units from 1991-1997. He joined the faculty of Virginia Tech as a visiting professor in the fall of 1997 and became an associate professor in 2007. You may contact him by emailing [email protected].

(513) 247-5465FAX (513) 247-5462e-mail: [email protected]

9955 International BoulevardCincinnati, Ohio 45246

• Carbon and 316 stainless steel

• 1/2”- 6” 150# and 300# flanged

• 1/2”- 4” 3-piece threaded, socket weld and butt weld

• Solenoid and limit switch for waterproof and explosion-proof environments

Assemblies have been approved by Factory Mutual through rigorous testing and quality certification and are available for delivery when you need them.

Valve Assemblies forLiquid and Gas Shutoff

AT Controls brings you complete FM approved valves for use in applications requiring protection and fuel burning equipment meeting the requirements per FM class 7400, 7420 and 7422.

Supervisory Cocks

SafetyShutoff


Tilting pad bearing modifications rotordynamic analysis

By Patrick J. Smith, Energy-Tech contributor

MACHiNE doCtor

Journal bearings have both static and dynamic charac-teristics. Static aspects include bearing load capacity, side leakage, frictional power loss, oil film temperature and pad temperature. Dynamic characteristics include stiffness and damping properties.

A tilting pad bearing problem was described in the January 2014 Energy-Tech article entitled, “High tilting pad bearing temperatures.” A bearing analysis was performed to understand the bearing behavior, which was discussed in the March 2014 Energy-Tech article entitled “High tilting pad temperature bearing analysis.” The problem was resolved by making some minor bearing changes. However, these changes also affected the dynamic bearing characteris-tics, so rotordynamic analyses were performed based on the original bearing clearance design and the modified bearing clearance design to understand the change in rotordynamic behavior and to determine if there were any rotordynamic issues with the modified bearing design. This article will only focus on the details of the rotordynamic analyses.

IntroductionThis case study pertains to an integrally geared cen-

trifugal compressor driven by a 5,000 HP, 1,794 RPM induction motor. The gearbox consists of a bullgear and three rotors. The stage 1/2 rotor consists of a pinion with overhung impellers mounted at both ends, while the stage 3 rotor consists of a pinion with a single overhung impeller. These rotors are mounted at the horizontal split line. The stage 4/5 rotor is located in the top of the gearbox cover and consists of a pinion with overhung impellers mounted at both ends. The gearbox arrangement for stages 1 to 3 is shown in Figure 1; the stage 4/5 rotor is omitted for clarity.

The rotor that is the subject of this article is the stage 1/2 rotor. It operates at 23,849 RPM and weighs 67.2 kg. The journal bearings are a 5-pad, tilting pad type with spherical seats and a center pivot design. Some basic bearing information is shown in Table 1.

The compressor was commissioned in late 1999 and put into continuous service in early 2000. As discussed in the previous Energy-Tech articles, issues with high temperature of the 2nd stage bearing developed and after performing a detailed bearing analysis, it was concluded that tight bear-ing clearance was the probable cause. The solution was to increase the bearing clearance. However, changing some-thing in the rotor/bearing system will have an impact on the rotordynamics. So as part of the overall engineering evaluation, rotordynamic analyses were performed based

on the original bearing design and the modified bearing design. Note that the 1st stage bearing was not changed.

Rotordynamic analysis From API-617, if a rotor-bearing system is modeled as a

simple mass-spring-damper system, the governing equation of motion for this system can be written as

Equation 1

Where:m = mass of the blockc = viscous damping coefficientk = stiffness of the elastic elementx = displacement of the blockF(t) = force applied to the block (time-dependent function)

Figure 1. The gearbox arrangement for stages 1 to 3.

Figure 2. Effects of decreasing damping on vibration amplification


While this simple model does not reflect the actual rotordynamic behavior of a compressor rotor/bearing sys-tem, it can be used to illustrate the basic concepts. In this example, the forced response is counteracted by the rotor mass and the support’s stiffness and damping characteris-tics. Fluid film bearings provide stiffness and damping, so changing a bearing’s stiffness and damping properties will affect the rotordynamic behavior. The table below shows the differences in the key bearing design parameters and the stiffness and damping coefficients for both the original design and the modified design. Note that when the setup clearance was changed, the radial clearance, or machined in clearance, also was changed in order to maintain a good preload.

The purpose of performing a rotordynamic analysis is to ensure that there are no undesirable rotor/bearing performance issues. The key results from the lateral rotor-dynamic analysis include a damped critical speed analysis and a damped unbalance response analysis. Rotor Bearing Technology & Software Inc. (RBTS) was contracted to per-form the analysis. RBTS is a consulting firm with expertise in rotor dynamics, failure analysis and corrective action determination.

Damped critical speed analysisThe damped critical speed analysis uses bearing damp-

ing and stiffness coefficients to predict the rotor/bearing system natural frequencies and associated mode shapes of vibration. The dynamic coefficients vary with speed, and so the damped critical speed analysis is performed based on the rotor operating at the design speed. The output of the damped critical speed analysis includes all the calculated damped natural frequencies within a speed range of 0 per-cent to 125 percent of the maximum operating speed, and the associated mode shapes and stability values.

A resonance condition exists when the frequency of a harmonic (periodic) forcing function coincides with a nat-ural frequency of the rotor/bearing system. Potential exci-tation sources include rotor unbalance, oil film instability, internal rubs, dynamic seal effects, loose parts, etc. These sources can excite subsynchronous resonant frequencies. When this happens, the forced vibrations resulting from the given exciting mechanism are amplified at the resonant fre-quency. The level of stability or damping refers to the unit’s resistance to these excitations. So the greater the damping, the less the dynamic amplification of the system vibration.

A variety of parameters can be used to evaluate the level of stability or damping. RBTS evaluates stability based on a

MACHiNE doCtor

Table 1 – Differences In Key Bearing Design Parameters For Both Original Design And Modified Design

Case Clearance Rad Clr, in. Setup Clr, in Preload

Results

Stiffness Damping

KXX, lbf/in KYY, lbf/in DXX/lbf/in DYY, lbf/in

Original Design 0.0037 0.0026 0.291 8.97E+05 1.40E+06 6.49E+02 8.20E+02

Modified Nominal 0.0063 0.0044 0.304 5.60E+05 1.07E+06 2.48E+02 3.77E+02

Table 1


parameter called the crit-ical damping ratio. The critical damping coeffi-cient Cc is the damping value required to com-pletely suppress any free vibration of the system. The critical damping ratio is the ratio of the actual damping coeffi-cient divided by the crit-ical damping coefficient, or C/Cc. If the critical damping ratio is negative, the system is said to be unstable. A ratio of zero indicates that there is no damping and the amplifi-cation is infinity. As long as the critical damping ratio is greater than zero,

there is some damping. A critical damping ratio of 1 indi-cates that the system is critically damped and there is no amplification when operating near this resonant frequency. Ratios between 0 and 1 indicate the degree of damping. Figure 2 shows how decreasing damping affects the amplifi-cation of the vibration.

Another stability parameter is called the logarithmic decrement (log dec), the natural logarithm of the ratio of any two successive amplitude peaks in a free harmonic vibration. The log decrement is a measure of how quickly the free vibrations experienced by the rotor system decay. When the log decrement is positive, the system is stable. Conversely, when the log decrement is negative, the system is unstable. According to API-617, if the log dec is less than 0.1, further stability analyses shall be performed. However, there is no industry standard of acceptance criteria on the minimum log dec. Even API-684 states that “… interpre-tation of results and suitability of design to be mutually agreed upon by the vendor and purchaser.”

Tables 2 and 3 show the results of the damped natural frequency analysis for both the original bearing clearance case and the modified bearing clearance case.

What is interesting is that what might be viewed as a relatively minor increase in bearing clearance had a moder-ate impact on the calculated natural frequencies and the sta-bility. Based on the original design case, there are no natural frequencies near the operating speed. With the modified clearance case there are 2 modes (modes 5 and 6) which are close to the operating speed. Fortunately, these modes are very well damped. While there are no stability issues for either case, the magnitude changes show the value in considering this analysis anytime a change is made to some-thing in the rotor/bearing system.

MACHiNE doCtor

Table 2

Table 3

Table 2 – Original Design Bearing Clearances

Mode #Natural

Frequency CPM

Critical Damping

RatioLog Dec Separation

Margin

1 7288 0.0604 0.38 69%

2 9531 0.1013 0.64 60%

3 10395 0.0703 0.44 56%

4 12386 0.0851 0.53 48%

5 54281 0.0558 0.35 128%

6 67580 0.1577 0.99 183%

Table 3 – Modified Design Bearing Clearances

Mode #Natural

Frequency CPM

Critical Damping

RatioLog Dec Separation

Margin

1 6615 0.0654 0.41 72%

2 8535 0.0727 0.46 64%

3 9724 0.0561 0.35 59%

4 11554 0.0851 0.53 52%

5 20336 0.4072 2.56 15%

6 27023 0.4501 2.83 13%

7 36812 0.2838 1.78 54%

8 51447 0.3998 2.51 116%

Figure 3. Mode shape for the modified bearing clearance case for mode 3


MACHiNE doCtor

A mode shape is the deflected shape of a rotor calculated at the particular natural frequency. Each resonant frequency will have a different mode shape. Understanding the mode shapes is valuable for a couple of reasons.

1. The location(s) of the greatest displacement(s) are places that are the most sensitive to excitation forces for that particular mode. So when performing the damped unbalance response analysis, unbalance will be located at these locations(s).

2. Knowing the mode shapes permits an estimation of rotor displacements at close fit areas such as bearing and seal locations, and at vibration probe locations.

The mode shape for mode 3 for the modified bearing clearance case is shown in Figure 3. This mode can be excited by adding unbalance at the locations of greatest dis-placement, locations 1 and 39, as shown.

Unbalance response analysisThe damped unbalance response analysis is the principal

tool used by API to evaluate relevant lateral rotordynamics characteristics, including lateral critical speeds and associat-ed amplification factors. The result of an unbalance response analysis is a calculation of the rotor response (vibration) to a set of applied unbalances as a function of speed. The location, phase and distribution of the unbalances are based on the mode that is being excited. It is important that all critical speeds below the maximum operating speed and the critical speed immediately above the maximum operating speed are analyzed. API also defines how to determine the magnitude of the unbalances that are applied. For this case study, an unbalance of 33.6 gram-mm was added at both impeller locations. These unbalances were in phase with one another.

The results of the unbalance response analysis are dis-played as a plot of vibration vs. shaft rotational speed. This is commonly called a Bode plot. These plots are generated for several axial shaft locations, which typically include bearing locations, shaft displacement probe areas and mid-span. For

this case study, unbalance response plots were generated for the following locations:

• Station 1: 2nd stage impeller• Station 11: 2nd stage bearing • Station 20: Rotor mid-span• Station 29: 1st stage bearing• Station 39: 1st stage impeller

The vibration peaks are the resonant speeds and these peaks are reviewed for the magnitude of the response and the proximity to the operating speed. The magnitude is evaluated based on the amplification factor (AF). The meth-od of calculating the amplification factor is shown on the Bode plot in Figure 4. This also introduces the term called the separation margin (SM), which is the margin of the peak vibration to the operating speed. API defines critical speeds as those resonant speeds with an amplification factor greater than 2.5. If the amplification factor for any critical speed below the operating speed is less than 2.5, no separa-tion margin is required. If the amplification factor for any critical speed below the operating speed is greater than 2.5, the minimum separation margin should be 16 percent, or the value from the equation below, whichever is less.

Figure 4. Figure 5.

Figure 6.


A-T Controls Inc. www.a-tcontrols.com 25

CU Services www.cuservices.net 31

Cutsforth, Inc. www.cutsforth.com 2

EagleBurgmann www.eagleburgmann-ej.com 31

FloScan Instrument Co. Inc. www.floscan.com 31

Fulmer Co. www.fulmercompany.com 15

Gaumer Process www.gaumer.com 31

Gradient Lens Corp. www.gradientlens.com 11

Indeck Power Equipment Co. www.indeck.com 31

MTI Instruments www.mtiinstruments.com 27

Renewal Parts Maintenance www.renewalparts.com 13

Rotork Controls Inc. www.rotork.com 32

Sohre Turbomachinery Inc. www.sohreturbo.com 9

Structural Integrity Associates, Inc. www.structinc.com 19

Topog-E Gasket Co. www.topog-e.com 16

Tri Tool Inc. www.tritool.com 7

Wabash Power Equipment www.wabashpower.com 23

MAY 2014AdVErtisErs’ iNdEX

Equation 2

If the amplification factor for any critical speed above the operating speed is greater than 2.5, the minimum sep-aration margin should be 26 percent or the value from the equation below, whichever is less.

Equation 3

For this case study, the unbalance response plots for both the original bearing clearance case and the modified bearing clearance case are shown in Figures 5 and 6. As shown, for both cases the separation margins are acceptable. However, the unbalance response at similar natural frequen-cies is slightly higher in the modified bearing clearance case. This again shows the value in considering a rotordynamic analysis anytime something in the rotor bearing system is changed.

ConclusionsThe lateral rotordynamics for both the original bearing

clearance case and the modified bearing clearance case are acceptable per API-617 standards. However, the analyses do show that the rotordynamic behavior changes even when making what appears to be a minor bearing clearance change. And for this reason, a lateral rotordynamic analysis should be considered whenever something is changed in the rotor bearing/system. ~

References1. Smith, Patrick J., “High tilting pad bearing temperature”,

Energy-Tech Magazine, January 2014.2. Smith, Patrick J., “High tilting pad temperature bearing

analysis”, Energy-Tech Magazine, March 2014.3. API-684, “Tutorial on the API Standard Paragraphs

Covering Rotor Dynamics and Balancing: An Introduction to Lateral Critical and Train Torsional Analysis and Rotor Balancing”, Second Edition, August 2005, API, Washington DC.

4. API-617, “Axial and Centrifugal Compressors and Expander-compressors for Petroleum, Chemical and Gas Industry Services”, Seventh Edition, API, Washington, DC.

Patrick J. Smith is lead machinery engineer at Air Products & Chemicals in Allentown, Pa., where he provides technical machinery support to the company’s operating air separation, hydrogen processing and cogeneration plants. You may contact him by e-mailing [email protected].

MACHiNE doCtor

Coming in June 2014

Look for the next issue of Energy-Tech

magazine and read about:

• Gas turbines• Cooling towers• Regulations compliance:

Emissions• Turbines: steam• ASME: RAM

Energ

y-tech

show

case

31

ms

TM

DuPont™ Krytox® Oils and Greases.These high-performance fluorinated lubri-cants are derivatives of Teflon® and offer thefollowing advantages: Chemically inert. Widetemperature range (-103°F to 800°F).Compatible with plastics, rubber, ceramics, &metals. Nonflammable. Insoluable in com-mon solvents. No silicones or hydrocarbons.Krytox® is applied to gearboxes, dampers,ductwork valves, steam valves, gaskets,seals, compressors, bearings, boilers,pumps, and Turbine Auxiliary systems. For more information and sample call800.992.2424 or 203.743.4447

DuPontTM Krytox® Channel Partner Since 1991

Krytox Fluorinated Lubricants®

[email protected] www.miller-stephenson.com

BOILERS RENT SALE LEASE

• Rental and Stock Boilers • Generators • Chillers • Deaerators • Boiler Parts • Boiler Services • Combustion Controls • Solid Fuel Applications

24/7 Emergency Service

P 847-541-8300 • F 847-541-9984 [email protected]

www.indeck.com





www.indeck.com





www.indeck.com





www.indeck.com





www.indeck.com





www.indeck.com





www.indeck.com





www.indeck.com





www.indeck.com

Celebrating 50 Years

Fuel Gas Conditioning Systems for all Turbines

713.460.5200www.gaumer.com

Fuel Gas Quality

Problems?

The Energy AnalystMechanical Engineering

Software

CU Services LLCPh: 847-439-2303

[email protected]

Boiler Effi ciency• Steam Turbines• Cooling Towers• HRSG • Condensers • Cogeneration• Heat Exchangers • Fanno Flow • Pipe Networks • Gravity Drain Flow • Steam Heaters • Steam Properties • Space Heating • Piping Pressure • Loss

Gas Turbines• Gas Expanders• Chimneys• Insulation• Gas Compressors• Duct Design• Restriction Orifi ce• Fans• Flash Tanks• Pumps• Psychrometrics• Desuperheaters• Deaerators•

Rotork Controls, Inc.675 Mile Crossing Blvd.Rochester, NY 14624

phone: 585 247 2304email: [email protected]

www.rotork.com

Redefining Flow ControlRedefining Flow Control

Enhance plant-wide per formance withRotork ac tuator and damper-drive solutionsEnhance per formance withRotork solutions

plant-wideac tuator and damper-drive

Rotork provides high-performance, cost-effective valve actuator and damper drive solutions throughout your entire power plant. Solutions that can help you maximize plant performance, minimize maintenance problems, and help you meet stringent safety and environmental regulations.

Hundreds of different valve actuators and damper drives are required throughout a typical power plant facility. Rotork can meet all of your needs by offering motor-operated, fluid power, electro-hydraulic, and manual gear operators. Actuators and drives suited for harsh, rugged operating conditions, inside and out, servicing every size and style of valve from small process control valves to super-large isolation valves as well as rotary, linear, or quarter-turn combustion-air and flue-gas dampers. Furthermore, the advanced technological capabilities of Rotork smart actuators can help you easily upgrade connectivity and provide a smart window to your process that includes important predictive maintenance data to help you improve productivity and efficiency as well as cut maintenance costs.

Whether you are building a new plant or are currently operating a fossil-fuel, nuclear, or green-energy facility, we can help. Rotork actuators and drives are currently servicing hundreds of power plants around the world. Our more-than 50-years experience and provenexpertise can help you, too. Contact us today.

Type K pneumaticrotary damper drive

IQ3 electric actuatorwith smart window technology

SM6000 S2 heavy-dutyelectric damper drive

Redefining Flow Control

May 2014

Documents

virginia tech

energytech magazine

tech issn

asme power division

distributed energy resources

power engineers collaborative

digital format

ts power planteditorial