Top Banner
International Design Codes 7C. European Codes - Steel Design to Eurocode 3 [EN 1993-1- 1:2005] STAAD.Pro is capable of performing steel design based on the European code EC3 BS EN 1993-1- 1:2005 Eurocode 3: Design of steel structures Part 1.1 General rules and rules for buildings. The implementation of EN1993-1-1:2005 includes the amendments as per CEN corrigenda of February 2006 and April 2009. Design of members per EC3 BS EN 1993-1-1:2005 requires the STAAD Euro Design Codes SELECT Code Pack. 7C.1 General Description 7C.2 Analysis Methodology 7C.3 Material Properties and Load Factors 7C.4 Section Classification 7C.5 Member Design 7C.6 Design Parameters 7C.7 Code Checking 7C.8 Member Selection 7C.9 Tabulated Results of Steel Design International Design Codes European Codes - Steel Design to Eurocode 3 [EN 1993-1- 1:2005] 7C.1 General Description The main steps in performing a design operation are: 1. Selecting the applicable load cases to be considered in the design process. 2. Providing appropriate ‘Parameter’ values if different from the default values. 3. Specify whether to perform code-checking and/or member selection. These operations can be repeated by the user any number of times depending on the design Page 1 of 43 EC3 -EN 1993_TOC 3/4/2014 file://C:\Documents and Settings\agpfm\Configuración local\Temp\~hh1167.htm
43

file___C___Documents and Settings__agpfm__Configuración local__Temp2

Nov 26, 2015

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
  • International Design Codes

    7C. European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    STAAD.Pro is capable of performing steel design based on the European code EC3 BS EN 1993-1-1:2005 Eurocode 3: Design of steel structures Part 1.1 General rules and rules for buildings.

    The implementation of EN1993-1-1:2005 includes the amendments as per CEN corrigenda of February 2006 and April 2009.

    Design of members per EC3 BS EN 1993-1-1:2005 requires the STAAD Euro Design Codes SELECT Code Pack.

    7C.1 General Description

    7C.2 Analysis Methodology

    7C.3 Material Properties and Load Factors

    7C.4 Section Classification

    7C.5 Member Design

    7C.6 Design Parameters

    7C.7 Code Checking

    7C.8 Member Selection

    7C.9 Tabulated Results of Steel Design

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.1 General Description

    The main steps in performing a design operation are:

    1. Selecting the applicable load cases to be considered in the design process. 2. Providing appropriate Parameter values if different from the default values. 3. Specify whether to perform code-checking and/or member selection.

    These operations can be repeated by the user any number of times depending on the design

    Page 1 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • requirements. The Parameters referred to above provide the user with the ability to allocate specific design properties to individual members or member groups considered in the design operation.

    7C.1.1 Eurocode 3 - EN 1993-1-1:2005 (EN 1993)

    The EN 1993 version of Eurocode 3, Design of steel structures, Part 1.1 General rules and rules for buildings (EN 1993) provides design rules applicable to structural steel used in buildings and civil engineering works. It is based on the ultimate limit states philosophy that is common to modern standards. The objective of this method of design is to ensure that possibility of failure is reduced to a negligible level. This is achieved through application of safety factors to both the applied loads and the material properties.

    The code also provides guidelines on the global methods of analysis to be used for calculating internal member forces and moments. STAAD uses the elastic method of analysis which may be used in all cases. Also there are three types of framing referred to in EC3. These are Simple, Continuous, and Semi-continuous which reflect the ability of the joints to developing moments under a specific loading condition. In STAAD only Simple and Continuous joint types can be assumed when carrying out global analysis.

    7C.1.2 National Annex Documents

    Various authorities of the CEN member countries have prepared National Annex Documents to be used with EC3. These documents provide alternative factors for loads and may also provide supplements to the rules in EC3.

    The current version of EC3 (EN 1993)implemented in STAAD adheres to the factors and rules provided in EN 1993-1-1:2005. The current version of STAAD.Pro includes the following National Annexes viz.

    a. British National Annex [NA to BS EN 1993-1-1:2005]

    b. The Dutch National Annex [NEN-EN 1993-1-1/NB] and

    c. Norwegian National Annex [NS-EN 1993-1-1:2005/NA2008]

    d. French National Annex [Annexe Nationale a la NF EN 1993-1-1:2005]

    e. Finnish National Annex [SFS EN 1993-1-1:2005] f. Polish National Annex [PN EN 1993-1-1:2005] g. Singaporean National Annex [SS EN 1993-1-1:2005] h. Belgian National Annex [NBN EN 1993-1-1:2005]

    The choice of a particular National Annex is based on the value of a new NA parameter that is set by the user when specifying the EN 1993 version of Eurocode 3. See "European Codes - National Annexes to Eurocode 3 [EN 1993-1-1:2005]" for a description of the NA parameter.

    7C.1.3 Axes convention in STAAD and EC3

    By default, STAAD defines the major axis of the cross-section as Z-Z and the minor axis as Y-Y. A special case where Z-Z is the minor axis and Y-Y is the major axis is available if the SET Z UP command is used and is discussed in Section 5.5 of the Technical Reference Manual. The longitudinal axis of the member is defined as X and joins the start joint of the member to the end with the same positive direction.

    Page 2 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • EC3, however, defines the principal cross-section axes in reverse to that of STAAD, but the longitudinal axis is defined in the same way. Both of these axes definitions follow the orthogonal right hand rule. See figure below.

    Bear this difference in mind when examining the code-check output from STAAD.

    Figure 7C.1 - Axis convention in STAAD and EC3

    See "Example of a TRACK 2 output" for an example of how this appears when Y is up (default).

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.2 Analysis Methodology

    Elastic analysis method is used to obtain the forces and moments for design. Analysis is done for the primary and combination loading conditions provided by the user. The user is allowed complete flexibility in providing loading specifications and using appropriate load factors to create necessary loading situations.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.3 Material Properties and Load Factors

    The characteristic yield strength of steel used in EC3 (EN 1993) design is based on table 3.1 of the code. Design resistances are obtained by dividing the characteristic value of a particular resistance by the global partial safety factor for the resistance, m. The magnitude of m is based on Cl. 6.1 of

    EN 1993-1-1:2005 and can change depending on the selected National Annex.

    Material coefficients for steel in STAAD take the following default values unless replaced by users numerical values provided in the input file.

    Page 3 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Modulus of Elasticity, E = 205000 N/mm2

    Shear Modulus, G = E/2(1+ )

    Poissons Ratio, = 0.3

    Unit weight, = 76.8 KN/m3

    The magnitude of design loads is dependent on f, the partial safety factor for the action under

    consideration. You are allowed total control in providing applicable values for the factors and their use in various load combinations.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.4 Section Classification

    The occurrence of local buckling of the compression elements of a cross-section prevents the development of full section capacity. It is therefore imperative to establish this possibility prior to determining the section capacities. Cross sections are classified in accordance with their geometrical properties and the stress pattern on the compression elements. For each load case considered in the design process, the program determines the section class and calculates the capacities accordingly. It is worth noting that the section class reported in the design output corresponds to the most critical loadcase among those being considered for design.

    The EC3 (EN 1993) design module in STAAD can design members with all section profiles that are of Class 1, 2, or 3 as defined in section 5.5 of the code. However, the design of members that have a Class 4 section profile are limited to:

    wide flange tee single channel single angle rectangular hollow sections circular hollow sections

    Also built-up user sections that are class 4 sections are not dealt with in the current version of EC3 design in STAAD.Pro, unless they are defined as any of the section types given above.

    The design of laced and battened members is not considered in the current version of EC3 (EN 1993) design module in STAAD.Pro. The current version also does not support the design of tapered section profiles or I-Sections with top and/or bottom plates.

    International Design Codes

    Page 4 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.5 Member Design

    EN 1993-1-1:2005, together with any specified National Annex, is used for code check or selection of all cross sections and shapes listed in Section 7C.4. However, where EN 1993 or the National Annex has not specified a method or values for a specific clause or parameter, STAAD.Pro uses Non-Contradictory Complimentary Information (NCCI) documents as explained in the following corresponding sections.

    The design philosophy and procedural logistics are based on the principles of elastic analysis and ultimate limit state design. Two major failure modes are recognized:

    failure by overstressing failure by stability considerations

    The following sections describe the salient features of the design approach. Members are proportioned to resist the design loads without exceeding the characteristic stresses or capacities. Member selection is done on the basis of selecting the most economic section on the basis of the least weight criteria. It is generally assumed that you (the engineer) will take care of the detailing requirements, such as the provision of stiffeners, and check the local effects like flange buckling, web crippling, etc.

    The design of class 4 (slender) sections is limited to WIDE FLANGE, TEE, SINGLE CHANNEL, SINGLE ANGLE, and RECTANGULAR & CIRCULAR HOLLOW SECTIONS. The effective section properties are evaluated as described in Cl. 6.2.2.5 of the code.

    You are allowed complete control over the design process through the use of the parameters listed in Table 7C.4. Default values of parameters will yield reasonable results in most circumstances. However, you should control the design and verify results through the use of the design parameters.

    7C.5.1 Members Subject to Axial Loads

    7C.5.2 Members Subject to Bending Moments

    7C.5.3 Members Subject to Shear

    7C.5.4 Members Subject to Torsion

    7C.5.5 Members Subject to Combined Forces

    7C.5.6 Design of Slender pipe sections to EN 1993-1-6

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    Page 5 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • 7C.5.1 Members Subject to Axial Loads

    The cross section capacity of tension only members is checked for ultimate limit state as given in Cl. 6.2.3 of the code.

    Compression members will be checked for axial capacity of the cross section in addition to lateral buckling/stability. The cross section capacity will be checked as given in section 6.2.4 of the code.

    Lateral stability of a pure compression member will be checked as per the method given in Cl. 6.3 of the code. The compression member stability will be verified as:

    Where Nb,Rd is the design buckling resistance given by:

    for Class 1, 2, or 3 cross-sections

    for Class 4 cross-sections

    Where:

    is the reduction factor as given in section 6.3.12 of the code. The buckling curves used to evaluate the reduction factor are selected from Table 6.2 of the code based on the cross section type and the steel grade.

    Only the five grades of steel given in table 6.2 will be used when selecting the buckling curve. The steel grade used for this selection is based on the SGR design input parameter (See "Design Parameters"). Even if you have specified a custom yield strength (using the PY parameter), the choice of a buckling curve will be based on the value of SGR parameter.

    Compression members that are susceptible to torsional or torsional flexural buckling are checked for these modes of failure as well. The non-dimensional slenderness T for these members is evaluated

    per Cl. 6.3.1.4 of the EN 1993 code. The maximum slenderness among the flexural buckling slenderness, torsional slenderness, and torsional-flexural slenderness is used to evaluate the reduction factor, , for such members. The elastic torsional buckling load, Ncr,T, and the elastic torsional-

    flexural buckling load, Ncr,TF, are evaluated based on the method given in the NCCI SN001a-EN-

    EU: Critical axial load for torsional and flexural torsional buckling modes (unless otherwise specified by a particular National Annex). The effective length for the members can be controlled using the KZ, KY, LZ and LY parameters. If these parameters are specified, the effective length will be calculated as KZ*LZ for length about the Z-Z axis and KY*LY for length about the Y-Y axis. By default, the effective length will be taken as the member length.

    EN 1993-1-1:2005 does not specifically deal with single angle, double angles, double channels, or Tee sections and does not provide a method to evaluate the slenderness of such members. In these cases, the EC3 (EN 1993) design module of STAAD.Pro uses the methods specified in BS 5950-1:2000 to calculate the slenderness of these members. Cl. 4.7.10 and Table 25 of BS 5950-1:2000 are used in the current version of the Eurocode 3 design module.

    Single Angle Sections

    Page 6 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Angle sections are un-symmetrical and when using BS 5950:2000 table 25 you must consider four axes: two principal, u-u and v-v and two geometric, a-a and b-b. The effective length for the v-v axis, Lvv, is taken as the LVV parameter or LY KY, if not specified. The a-a and b-b axes are determined by which leg of the angle is fixed by the connection and should be specified using the LEG parameter, see section 5B.6 for more information on the LEG parameter. The effective length in the a-a axis is taken as LY KY and the effective length in the b-b axis as LZ KZ.

    The following diagram shows the axes for angles which have been defined with either an ST or RA specification and is connected by its longer leg (i.e., a-a axis is parallel to the longer leg).

    Figure 7C.2 - Axis orientation for single angles

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.5.2 Members Subject to Bending Moments

    The cross section capacity of a member subject to bending is checked as per Cl .6.2.5 of the code. The condition to be satisfied is:

    Where Mc,Rd is the is the design resistance given by:

    for class 1 and 2 cross-sections

    for class 3 cross-sections

    ST angle and USER table angles RA angle

    Page 7 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • for class 4 cross-sections

    Cross sectional bending capacity checks will be done for both major and minor axis bending moments.

    Members subject to major axis bending will also be checked for Lateral Torsional Buckling resistance as per Section 6.3.2 of the code. The design buckling resistance moment Mb,Rd will be

    calculated as:

    Where:

    LT is the reduction factor for lateral torsional buckling. This reduction factor is

    evaluated per Cl. 6.3.2.2 or Cl 6.3.2.3 of the EN 1993 code depending on the section type. For I sections, the program will by default use Cl. 6.3.2.3 to evalute LT and for all

    other sections the program will resort to Cl 6.3.2.2. However, if a particular National Annex has been specified, the program will check if the National Annex expands on Cl.6.3.2.3 (Table 6.5) to include sections other than I sections. If so, the program will use Cl. 6.3.2.3 for the cross-section(s) included in Cl. 6.2.2.3 (or Table 6.5). For all other cases the program will use Cl. 6.3.2.2.

    You have the option to choose the clause to be used to calculate LT through the MTH

    design parameter. Setting MTH to 0 (default value) will cause the program to choose Cl.6.3.2.3 for I Sections and Cl 6.2.3.2 for all other section types. As mentioned above, if the National Annex expands on Cl. 6.3.2.3 to include sections other than I Sections, the program will use Cl. 6.3.2.3 by default.

    When using Cl. 6.3.2.3 to calculate LT, the program will consider the correction factor

    kc (Table 6.6 of EN 1993-1-1:2006) based on the value of the KC parameter in the design input. By default the value of KC will be taken as 1.0. If you want the program to calculate kc, you must explicitly set the value of the KC parameter to zero.

    If the National Annex specifies a different method to calculate kc (e.g. the British, Singapore & Polish NAs), the program will use that method by default even if the KC parameter has not been explicitly set to zero. If the NA method does not deal with a specific condition while working out kc, the program will then fall back to table 6.6 of the code, thus ensuring that kc is considered for the particular NA.

    The non-dimensional slenderness LT (used to evaluate LT) for both the above cases is evaluated as:

    Where:

    Mcr is the elastic critical moment for lateral torsional buckling. EN 1993-1-1 does not

    however specify a method to evaluate Mcr. Hence, the program will make use of the

    Page 8 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • method specified in Annex F of DD ENV 1993-1-1 to evaluate Mcr by default.

    The method specified in Annex F will be used only when the raw EN 1993-1-1:2005 code is used without any National Annex. If a National Annex has been specified, the calculation of Mcr (and LT) will be done based on the specific National Annex. (See

    "European Codes - National Annexes to Eurocode 3 [EN 1993-1-1:2005]" for specific details). If the National Annex does not specify a particular method or specify a reference document, the program will use the NCCI document SN-003a-EN-EU for doubly symmetric sections and SN030a-EN-EU for mono-symmetric sections that are symmetric about their weak axis. For all other sections types the program will use Annex F of DD ENV 1993-1-1 to calculate Mcr. In cases where Annex F does not

    provide an adequate method to evaluate Mcr, such as for Channel sections, the program will resort to the method as per Cl.4.3.6 of BS 5950-1:2000 to calculate the lateral torsional buckling resistance moment (Mb,Rd) for the member.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.5.3 Members Subject to Shear

    The cross section capacity of a member subject to shear is checked as per Cl. 6.2.6 of the code. The condition to be satisfied is:

    Where:

    Vc,Rd is the is the shear design resistance given by:

    Av is the shear area and is worked out for the various section types as given in Cl. 6.2.6

    (3) of the code.

    Shear Buckling

    For sections that are susceptible to shear buckling, the program will perform the shear buckling checks as given in Section 5 of EN 1993-1-5. The shear buckling checks will be done only for I Sections and Channel sections. Shear stresses induced from torsional loads are taken into account while performing torsion checks.

    Web shear buckling is checked in STAAD.Pro V8i (SELECTseries 3) (release 20.07.08) and later.

    The susceptibility of a section to shear buckling will be based on the criteria given in Cl 5.1(2) of EN

    Page 9 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • 1993-1-5 as is as given as follows:

    a. For unstiffened webs, if hw/t > 72/, the section must be checked for shear buckling.

    The design resistance is calculated as:

    Where:

    hw = distance between flanges of an I Section (i.e., depth - 2x flange thickness).

    t = thickness of the web

    = (235/fy), where fy is the yield stress

    = 1.2 for steel grades up to and including S 460 and = 1.0 for other steel grades

    k as defined in sections below

    w is the web contribution factor obtained from Table 5.1 of the EC3 code and is

    evaluated per the following table:

    b. For stiffened webs, if hw/t > 31Ek/, the section must be checked for shear buckling.

    The design resistances considers tension field action of the web and flanges acting as struts in a truss model. This is calculated as:

    Where:

    Vbf,Rd is the flange resistance per Cl.5.4 for a flange not completely utilized by

    bending moment.

    Table 7C.1-Evaluate of wSlenderness Parameter Rigid End Post Non-rigid End Post

    w < 0.83/ 0.83/ w < 1.08 0.83/w 0.83/w

    w > 1.08 1.37/(0.7 + w) 0.83/w

    Page 10 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • bf is the width of the flange which provides the least axial resistance, not to be

    taken greater than 15tf on each side of the web.

    tf is the thickness of the flange which provides the least axial resistance.

    Mf,Rd = Mf,k/M0, the moment of resistance of the cross section consisting of the

    effective area of the flanges only. For a typical I Section or PFD, this is evaluated as btfhw. When an axial load, NEd, is present, the value of Mf,Rd is reduced by

    multiplying by the following factor:

    Af1 and Af2 are the areas of the top and bottom flanges, respectively.

    a = transverse stiffener spacing. The equation of c is likewise used to solve for a sufficient stiffener spacing in the case of demand from loads exceeding the calculated capacity for a specified stiffener spacing.

    The following equation must be satisfied for the web shear buckling check to pass:

    Where:

    VEd is the design shear force.

    The shear forces due to any applied torsion will not be accounted for if the TOR parameter has been specifically set to a value of 0 (i.e., ignore torsion option).

    If the stiffener spacing has not been provided (using the STIFF parameter), then the program assumes that the member end forms a non-rigid post (case c) and proceeds to evaluate the minimum stiffener spacing required.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-

    Page 11 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • 1:2005]

    7C.5.4 Members Subject to Torsion

    This feature requires STAAD.Pro V8i (SELECTseries 2) build 2007.07 or later.

    General

    Eurocode 3 (EN 1993-1-1:2005) gives very limited guidance for the analysis and design of torsion members. While both elastic and plastic analyses are permitted generally, the design analysis methods for torsion discussed within EC3 are primarily based on elastic methods. Also, only the first yield design resistance is specifically discussed for torsion members. Furthermore, there is no guidance on section classification nor on how to allow for the effects of local buckling on the design resistance for combined torsional effects. EC3 also does not specifically deal with members subject to combined bending and torsion and loosely states that the yield criteria (Eqn 6.1 in the code) can be used for elastic verification.

    The method used by STAAD.Pro is therefore based on the SCI publication P057: Design of members subject to combined bending and torsion. Though this publication is based on the British standard BS 5950-1, the principles from this document are applied in the context of Eurocode 3.

    At the time this feature has been implemented in STAAD.Pro, SCI are in the process of updating document P057 to be in accordance with Eurocode 3. Hence this method might be subject to modifications subject to the publication of a newer version of P057. The NCCI document SN007b-EN-EU: Torsion will also be referenced where appropriate.

    Code Basis

    Torsion design in EC3 is given in Cl. 6.2.7 of EN 1993-1-1:2005. Therefore, this clause is used primarily for this implementation.

    EN 1993-1-1:2005 does not deal with members subject to the combined effects of torsion and lateral torsional buckling. However, EN 1993-1-6 considers such a condition in Appendix A. Therefore, STAAD.pro uses Appendix A of EN 1993-1-6 to check for members subject to combined torsion and LTB.

    The following clauses from EC3 are then considered:

    Cl. 6.2.7(1) Cl. 6.2.7(9) Cl. 6.2.7(5) EC-3 -6 App A

    STAAD.Pro does, however, use this clause (6.2.7) to report the output for all torsion checks. Also any distortional deformations and any amplification in the torsional or shear stresses due to distortions will be neglected by the program.

    Clause 6.2.7(1)

    States that for members subject to torsion, the design torsional moment TEd at each cross

    section should satisfy:

    TEd / RRd 1.0

    Page 12 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Where:

    TRd is the design torsional resistance of the cross section.

    This is the primary condition that will need to be satisfied for members subject to torsion. The method for working out the torsional resistance TRd, for the various cases is dealt in the

    following sections.

    Cl. 6.2.7(9)

    States that:

    For combined shear force and torsional moment, the plastic shear resistance accounting for torsional effects should be reduced from Vpl,Rd to Vpl,T,Rd and the

    design shear force should satisfy:

    VEd / Vpl,T,Rd 1.0

    The code also gives means to evaluate Vpl,T,Rd in equations 6.26 to 6.28. These equations,

    however, only deal with I/H sections, Channel sections, and structural hollow sections (RHS, SHS, CHS). Therefore, the application of Cl. 6.2.7(9) is only performed for these section profiles.

    Cl 6.2.7(5)

    States that the yield criteria given in Cl. 6.2.1(5) of EN 1993-1-1:2005 may be used for elastic verification. STAAD.Pro evaluates the stresses due to the various actions on the cross section and applies this yield criterion.

    The program allows for two types of checks for members subject to torsion for EC3 design:

    I. Basic Stress Check: This method is intended to be a simplified stress check for torsional effects. This method will produce the output corresponding to Cl. 6.2.7(5) of EN 1993-1-1.

    II. Detailed Checks: This method will perform a full torsional analysis of the member. All four of the clause checks mentioned earlier will be performed.

    The details of these checks are as described below.

    You have the option to choose the method to be used for a specific member or group of members. This will be facilitated by setting the value of the TORSION. The TORSION parameter set to zero by default, which results in torsion checks only being performed if the member is subject to torsional moments (i.e., for this default setting, the program will ignore torsion checks if there is no torsional moment in the member). Setting the value of the TORSION parameter to three (3) will cause the program to ignore all torsional moments. The detailed output (i.e., TRACK 2) will indicate that torsion has been ignored for that particular member. The details of setting the values to one (1) or two (2) and the corresponding checks performed are as described below. See "Design Parameters" for additional details.

    If the TORSION parameter is set to 1 or 2, the program will perform the appropriate checks even if the member is not subject to torsional moments. In such cases, the program will perform the checks with a value of zero for the torsional moment.

    Page 13 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Basic Torsion Stress Checks

    Detailed Torsion Stress Checks

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    Basic stress check

    This method is used when the TORSION parameter is specified as one (1).

    This method is intended to be a simplified stress check for torsional effects per Cl. 6.2.7(5). Any warping stresses that may develop due to the end conditions will be ignored for this option. The program will consider the forces (including torsion) at various sections along the length of the member and for each section, will calculate the resultant stress (Von Mieses) at various points on the cross section. The location and number of points checked for a cross section will depend on the cross section type and will be as described below.

    The stress check will be performed using equation 6.1 of EN 1993-1-1:2005 as given below:

    Where:

    x,Ed is the longitudinal stress

    z,Ed is the transverse stress and

    Ed is the resultant shear stress.

    Since transverse stresses are very small under normal loading conditions (excluding hydrostatic forces), the term will be negligible and hence is taken as zero.

    x,Ed = x + bz + by = Fx/Ax + Mz/Zz + My/Zy

    Ed = T/J t + VyQ/(Izt) + VzQ/(Iy*t)

    Where:

    T is the torsion at the particular section along the length of the member

    J is the torsion constant

    t is the thickness of the web/flange

    Page 14 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • V is the shear force

    Q is the statical moment about the relevant axis

    I is the second moment of area about the relevant axis

    The stress check as per equation 6.1 is performed at various stress points of a cross section as shown in figures below:

    Shape Section Sketch

    Doubly symmetric wide flange profile

    Pipe profiles

    = tan-1(Mz/My)

    Tube profiles

    Page 15 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • The resultant ratio will be reported under Cl. 6.2.7(5) in the detailed design output.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    Detailed stress check

    This method is used when the TORSION parameter is specified as two (2).

    This method performs a detailed torsional analysis of a member depending on the torsion loading conditions and the support conditions at the member ends. This method is based on the SCI publication P057 and includes any warping stresses (direct warping stresses and warping shear stresses) depending on the end conditions of the member. This implementation considers seven different cases of loading and end conditions as given in publication P057 Section 6. The loading/end conditions for a member are specified by the use of the CMT design parameter (See "Design Parameters" for parameter values and descriptions).

    All the equations used to evaluate the torsional moments and associated stresses are as given in Appendix B of P057. The resultant stresses are evaluated at various sections along the length of the member and the following checks will be performed:

    Clause 6.2.7(1) Torsional resistance of the section.

    In general, the torsion at any section TEd is resolved into two components, viz.

    The pure torsional (St. Venants) moment (Tt,Ed) and

    The warping torsional moment(Tw,Ed)

    Channel profiles

    Page 16 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Therefore,

    TEd = Tt,Ed + Tw,Ed = GJ = EH

    [Ref SCI pub. P057]

    Where:

    and are the first and third derivates of twist ( ), respectively, and depend on the end conditions and loading. These are evaluated from the equations in Annex B of P057 and are based the specified CMT parameter.

    Although the equation given the NCCI document SN007b-EN-EU can be used to evaluate Twrd, the

    NCCI does not give the eqn. to evaluate . Therefore, Annex B of P057 is used.

    The torsional resistance of the section is also considered as the sum of the pure torsion resistance and the warping torsion resistance. The pure torsion resistance (Tt,Rd) and the warping torsional

    resistance (Tw,Rd) are evaluated as:

    For closed sections:

    Tt,Rd = 2 Ac t max

    Where:

    Ac is the area enclosed by the mean perimeter

    t is the max thickness

    max is the max. allowable shear stress = (fy/3)/ m0

    For open sections (I & channel):

    Tt,Rd = max J / t

    Where:

    J is the torsion const

    t is the max thickness.

    Tw,Rd = (fy/ m0) t b2 / 6

    Where:

    b is the width of the section

    t is the thickness of the flange for I- sections; minimum of flange or web thickness channel sections

    Page 17 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • The check according to Cl 6.2.7(1) will then be performed to ensure that the following conditions are satisfied:

    Tt,Ed / Tt,Rd 1

    Tw,Ed / Tw,Rd 1

    TEd / TRd 1

    Clause 6.2.7(9) Plastic shear resistance due to torsion

    STAAD.Pro checks for shear resistance of a section based on Cl. 6.2.6 for EC3 and the plastic shear resistance (in the absence of torsion) is evaluated as:

    Where:

    Av is as pre Cl.6.2.6 (3) for the various sections

    When torsion is present, along with the shear force, the design shear resistance will be reduced to Vpl,T,Rd, where Vpl,T,Rd is evaluated as follows:

    i. For I or H Sections:

    ii. For Channel Sections:

    iii. For Structural Hollow Sections:

    Where

    t,Ed is the shear stress due to direct (St. Venants) torsion and

    w,Ed is the shear stress due to warping torsion.

    The various shear stresses due to torsion t,Ed and w,Ed are evaluated as follows:

    i. For Closed sections:

    Page 18 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • The shear stresses due to warping can be ignored as they will be insignificant and hence:

    t,Ed = TEd/(2Act)

    [Ref NCCI Sn007b-EN-EU]

    Where:

    TEd is the applied torsion,

    Ac is the area delimited by the mean perimeter and

    t is the thickness of the cross section

    w,Ed = 0, since warping is ignored

    ii. For Open sections [I, H, Channel] sections:

    For I and H sections, the web will not be subject to warping stresses and therefore warping shear can be ignored (w,Ed=0).

    The stress due to pure torsion is evaluated as:

    t,Ed = Gt

    [Ref SCI pub. P057]

    Where:

    G is the shear modulus

    is a function depending on the end condition and loading(T). This will be taken from section 6 and Annex B of P057.

    Although the maximum stress is at the thickest section of the profile, the program uses the web thickness for this clause (since the shear capacity is based on the web area) unless the load is parallel to the flanges, in which case the flange thickness is used.

    For channel sections that are free to warp at the supports and, thus, are not subject to warping stresses:

    The warping shear stress is evaluated as:

    w,Ed = ESw / t

    [Ref SCI pub. P057]

    Where:

    E is the elastic modulus,

    Page 19 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Sw is the warping statistical moment and

    is a function depending on the end condition and loading(T). This will be taken from section 6 and Annex B of P057.

    t is the thickness of the element.

    Clause 6.2.7(5) Check for elastic verification of yield

    Eurocode 3 gives yield criterion as per eqn. 6.1 and STAAD.Pro uses the yield criterion given in EC-3. When a member is subject to combined bending and torsion, some degree of interaction occurs between the two effects. The angle of twist caused by torsion is amplified by the bending moments and will induce additional warping moments and torsional shears. Account must also be taken of the additional minor axis moments produced by the major axis moments acting through the torsional deformations, including the amplifications mentioned earlier.

    For members subject to bending and torsion, the stresses are evaluated as follows:

    Direct bending stress (major axis): bz = Mz / Zz

    Direct bending stress (minor axis): by = My / Zy

    Direct stress due to warping: w = EWns

    Direct stress due to twist (min. axis): byt = Myt / Zy

    Direct stress due to axial load (if any): c = P/ A

    Where:

    Mz is the major axis moment & My is the minor axis moment.

    is the differential function based on twist (ref P057 Annex B. & Table 6)

    Wns is the normalized warping function.

    Myt = Mz (see Appendix B of P057 to evaluate )

    Shear stresses due to torsion and/or warping is evaluated as described above for Clause 6.2.7(9).

    Check for yield (capacity checks) is then done according to Eqn 6.1 of EN 1993-1-1:2005, as described for the Basic Stress Check (TORSION = 1):

    Clause EC-3:6 App A Check for combined Torsion and Lateral Torsional buckling

    The interaction check due to the combined effects of bending (including lateral torsional buckling) and torsion will be checked using Annex A of EN 1993-6: 2007. Note that this interaction equation

    Page 20 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • does not include the effects of any axial load.

    At present, SCI advises that no significant work has been published for this case and work is still ongoing. So at present is advisable not to allow for torsion in a member with large axial load.

    Members subject to combined bending and torsion will be checked to satisfy:

    Where:

    Cmz is the equivalent uniform moment factor for bending about the z-z axis, according

    to EN 1993-1-1 Table B.3.

    My,Ed and Mz,Ed are the design values of the maximum moment about the y-y and z-z

    axis, respectively.

    My,Rk and Mz,Rk are the characteristic values of the resistance moment of the cross-

    section about it y-y and z-z axis, respectively, from EN 1993-1-1, Table 6.7.

    My,cr is the elastic critical lateral-torsional buckling moment about the y-y axis.

    Tw,Ed is the design value of the warping torsional moment.

    Tw,Rk is the characteristic value of the warping torsional resistance moment.

    LT is the reduction factor for lateral torsional buckling according to 6.3.2 of EN 1993-

    1-1.

    For all of the above checks the effective length of the member to be used for torsion can be set by using the EFT design parameter.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    Page 21 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • 7C.5.5 Members Subject to Combined Forces

    Members subject to Bending and Axial Force

    When a member is subject to a combined axial load and a bending moment, the program evaluates a reduced moment capacity based on Cl. 6.2.9 of the code. For Class 1, 2, and 3 sections, the program evaluates the reduced moment from the equations given in Cl. 6.2.9.1 of the code. For class 4 sections, the interaction equation given by equation 6.44 are checked.

    In the case of members subject to axial load and biaxial bending, the program will consider the interaction equation 6.41 of the code.

    By default, the program will use the values of the constants and as given in the code for the different sections types. However, you can override these values using the ALPHA and BETA design parameters (See "Design Parameters").

    The program uses the parameter ELB (See "Design Parameters") to override the Cl.6.2.9 checks for combined axial load and bending case. When specfied as 1, the program uses the more general equation 6.2 of EN 1993-1-1, instead.

    Members subject to Bending, Shear, and Axial Force

    When a member is subject to a combined axial load, shear force, and a bending moment, the program evaluates the reduced yield strength as given in Cl 6.2.10 (3) of the code. The reduction in the yield strength is done only when the applied shear force exceeds 50% of the design shear resistance Vpl,Rd. This reduced yield strength is then used to evaluate the reduced moment capacity

    of the section.

    Members subject to Bending and Axial Compression

    The bending resistance of members could be reduced by the presence of a co-existent axial load. This is then checked against the lateral-torsional buckling resistance of the section. The EN 1993 design module in STAAD takes such a scenario into account and performs the necessary checks as per Cl. 6.3.3 of the code.

    Generally, EC3 requires checking cross-section resistance for local capacity and also checking the overall buckling capacity of the member. In the case of members subject to axial tension and bending, there is provision to take the stabilizing effect of the tension load into consideration. This is achieved by modifying the extreme compression fibre stress and calculating an effective applied moment for the section. The checks are done as per Cl. 6.2.9 of the code. In case of a combined axial compressive load and bending moment, the member is checked per the rules in section 6.3.3 of the code. The program checks to ensure that both the interaction equations 6.61 and 6.62 of the code are satisfied. The interaction factors kzz, kyy, kzy & kyz will be evaluated using Annex B of EN 1993-1-1 by default. Hence for the EN 1993-1-1 code in STAAD.Pro (without National Annexes), uses Annex B. The choice between using Annex A and Annex B will be based on the choice specified by a particular National Annex, if used. If the National Annex itself gives a choice between Annex A and Annex B, the program uses Annex B to evaluate the interaction factors.

    EN 1993-1-1:2005 does not specifically deal with single angle, double angles, double channels or Tee sections and does give a method to evaluate the slenderness of such members. In these cases, the Eurocode 3 (EN 1993-1-1) design module of STAAD.Pro uses the methods specified in BS 5950-1:2000 to calculate the slenderness of these members. Cl. 4.7.10 of BS 5950-1:2000 is used in the current version of the EC3 design module. See "Single Angel Sections" for ST and RA angle specifications.

    Page 22 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Laced or battened compression members are not dealt within the current version of EC3 (EN 1993) design module in STAAD.Pro.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.5.6 Design of Slender pipe sections to EN 1993-1-6

    The design of Slender CHS sections is performed per EN 1993-1-6:2007 (hereafter, EC3-6). EC3-6 does not specify additional or modified safety factors. Therefore, the program uses the default safety factors from EN 1993-1-1.

    You can change these values through the GM0, GM1, & GM2 design parameters.

    EC3-6 deals with four types of ultimate limits states: plastic limit state, cyclic capacity limit state, buckling limit state, and fatigue. The following are considered by STAAD.Pro:

    LS1 Plastic limit state: Deals with the condition when the capacity of the structure is exhausted by yielding of the material.

    LS3 Buckling Limit state: Deals with the condition in which the structure (or shell) develops large displacements normal to the shell surface, caused by loss of stability under compressive and/or shear membrane stresses.

    The limit state verification is made based on the Stress design method described in EC3-6. The stress design approach takes into account three categories of stresses:

    Primary stresses: Stresses that are generated for the member to be in equilibrium with the direct imposed loads.

    Secondary stresses: Those that are generated for internal compatibility or for compatibility at supports due to imposed loads or displacements (e.g., temperature, settlement etc.)

    Local stresses: Local stresses generated due to cyclic loading (or fatigue).

    Only the primary stresses are considered the program. The primary stresses considered are those generated due to axial loads, bending, shear and /or a combination of these conditions.

    In the context of slender pipe section design for the Eurocode 3 module, the secondary and local stresses can be neglected since the loads and corresponding stresses dealt with in the design engine are largely direct and shear stresses.

    The local axis coordinate system for a CHS is defined as:

    circumferential around the circumference of the circular cross section ()

    meridional along the length of the member (x)

    normal perpendicular to the tangential plane formed by the circumferential and meridional directions (n)

    Page 23 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • and the corresponding membrane stresses will follow the convention given below:

    Figure 7C.3 - Nomenclature for membrane and transverse stresses in Slender CHS sections

    Stress Design

    Stress checks are made based on the Stress design method as per Section 8.5 of the code. This section deals with the buckling strength of the member (LS3). The principle is to evaluate the membrane stresses due to the applied loads and then compare that to the buckling strength, which is evaluated giving due consideration for local buckling effects.

    The membrane stresses are evaluated as given in Annex A of the code. The pipe section is considered as an unstiffened cylindrical shell.

    i. Meridional Stresses:

    1. Axial load

    Fx = 2rPx

    x = -Fx/(2rt)

    2. Axial stress from bending

    M = r2Px,max

    x = M/(2rt)

    ii. Shear Stress:

    1. Transverse force, V

    V = rP,max

    max = V/(rt)

    2. Shear from torsional moment, M

    Mt = 2r2P

    = Mt/(22r2t)

    Membrane stresses Transverse stresses

    Page 24 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Where:

    r is the radius of the middle surface of the shell wall.

    t is the wall thickness of the cylinder

    Calculation of Axial Buckling Stress

    The buckling strength of A slender pipe section is evaluated using the method given in section 8.5.2 ofEC3-6. The design buckling stresses (buckling resistance) are calculated separately for axial, circumferential, and shear. The circumferential stresses are ignored in STAAD.Pro.

    The naming convention and the coordinate axis used will be as given in the following diagram:

    Figure 7C.4 - Naming convention and coordinate system used for the buckling stress of a slender

    CSH section

    The axial buckling resistance is given by:

    x,Rd = x,Rk/M1

    M1 will have the same default value of 1.0 as in EN 1993-1-1.

    x,Rk is the characteristic buckling strength given by:

    x,Rk = x fyk

    Where:

    x is the meridional buckling reduction factor. x is evaluated per Section 8.5.2(4) of

    EC3-6 and is determined as a function of the relative shell slenderness given by:

    Where:

    x,cr is the elastic buckling critical stress.

    Page 25 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Once the relative slenderness is evaluated, the reduction factor is calculated as follows:

    = 1 when 0

    when 0 < < P

    = /2 when P

    Where:

    p is the plastic limit for slenderness given by:

    The meridional buckling parameters the factors and are evaluated per section D.1.2.2 of EC3-6.

    A Normal fabrication quality will be assumed when evaluating the fabrication quality parameter as given in table D.2 of the code, unless the fabrication quality is set using the FAB design parameter. See "Design Parameters"

    The elastic critical buckling stress, x,cr and the factors and are evaluated per Annex D of EC3-6.

    The details are as given below:

    The CHS section is classified based on the following criteria:

    Where:

    The elastic critical buckling critical stress is evaluated as:

    x,Rcr = 0.605ECx(t/r)

    Where:

    Cx is a factor dependant upon the CHS length classification as described in section

    D.1.2.1 of EC-3-6.

    For a long cylinder, there are two separate methods that can be used to evaluate the Cx

    factor: Eqns D.9/10 and Eqn D.12. Initially the program evaluates Cx based on the

    maximum from equations D.9 and D.10. However, for long cylinders that satisfy the conditions in equation D.11, the program will also work out Cx based on equation D.12

    CHS Length Classification Criteria

    Short 1.7 Medium 1.7 < 0.5 r/t

    Long > 0.5 r/t

    Page 26 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • and then choose the minimum obtained from D.12 and D.9/10.

    Calculation of Shear Buckling Stress

    The shear buckling resistance is given by:

    x,Rd = x,Rk/M1

    M1 will have the same default value of 1.0 as in EN 1993-1-1.

    x,Rk is the characteristic buckling shear strength given by:

    x,Rk = fyk

    Where:

    is the shear buckling reduction factor. will be worked out as given in section 8.5.2

    (4) of En 1993-1-6 and is determined as a function of the relative shell slenderness given by:

    Where:

    x,Rk is the elastic buckling critical stress.

    The reduction factor, , is then evaluated as described for the axial buckling stress, based on the

    same p, , and parameters given in Annex D of EC3-6.

    The CHS section is classified based on the following criteria:

    Where:

    The elastic critical buckling critical stress is evaluated as:

    Where:

    CHS Length Classification Criteria

    Short 10 Medium 10 < 8.7 r/t

    Long > 8.7 r/t

    Page 27 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Cis a factor dependant upon whether the CHS length classification as described in

    section D.1.4.1 of EC-3-6.

    A Normal fabrication quality will be assumed when working out the fabrication quality parameter as given in table D.6 of the code, unless the fabrication quality is set using the FAB design parameter.

    Buckling Strength Verification

    The buckling strength verification will be performed so as to satisfy the following conditions:

    For axial stresses:

    x,Ed x,Rd

    For shear stresses:

    x,Ed x,Rd

    For a combined case of axial and shear stresses acting together, an interaction check will be done according to equation 8.19 of the code as below:

    Where:

    kx and k are the interaction factors as given in section D.1.6 of EN 1993-1-6:

    kx = 1.25 + 0.75 x

    k = 1.75 + 0.25

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.6 Design Parameters

    Design parameters communicate specific design decisions to the program. They are set to default values to begin with and may be altered to suite the particular structure.

    Depending on the model being designed, you may have to change some or all of the parameter default values. Some parameters are unit dependent and when altered, the n setting must be compatible with the active unit specification.

    Page 28 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Table 7C.4 lists all the relevant EC3 parameters together with description and default values.

    Table 7C.2-Steel Design Parameters EC3 EN

    Parameter

    Name

    Default

    ValueDescription

    CODE -

    Must be specified as EN 1993-1-1:2005 to

    invoke design per Eurocode 3:2005 (EN 1993).

    Design Code to follow.

    See section 5.48.1 of the Technical Reference

    Manual.

    ALH 0.5

    The ratio of the distance of the point torque

    (from the start of the member) to the length of

    the member. The default value of 0.5 represents

    torque acting at the mid-span of a

    symmetrically loaded member. Values can

    range from 0 to 1.

    ALPHA 1.0

    Used to input a user defined value for the

    factor in equation 6.41 for combined bending

    and axial force checks.

    BEAM 3

    Parameter to control the number of sections to

    checked along the length of a beam:

    1. Check at location of maximum Mz along

    beam

    2. Check sections with end forces and forces

    at location of BEAM 1.0 check.

    3. Check at every 1/13th point along the

    beam and report the maximum

    BETA 1.0

    Used to input a user defined value for the

    factor in equation 6.41 for combined bending

    and axial force checks.

    C1 1.132

    Corresponds to the C1 factor to be used to calculate Elastic critical moment Mcr as per

    Clause 6.3.2.2

    C2 0.459

    Corresponds to the C2 factor to be used to calculate Elastic critical moment Mcr as per

    Clause 6.3.2.2

    C3 0Corresponds to the C3 factor to be used to calculate Elastic critical moment Mcr as per

    Page 29 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Clause 6.3.2.2

    CAN 0

    Member will be considered as a cantilever type

    member for deflection checks.

    0 indicates that member will not be

    treated as a cantilever member

    1 indicates that the member will be

    treated as a cantilever member

    CMM 1.0

    Indicates type of loading and support conditions

    on member. Used to calculate the C1, C2, and C3 factors to be used in the Mcr calculations.

    Can take a value from 1 to 8.

    Refer to Table 7C.5 for more information on its

    use.

    CMN 1.0

    Indicates the level of End-Restraint.

    1.0 = No fixity

    0.5 = Full fixity

    0.7 = One end free and other end

    fixed

    CMT 1

    Used to indicate the loading and support

    condition for torsion (ref. SCI publication P-

    057).

    Can take a value of 1-7. The values correspond

    to the various cases defined in section 6 and

    App. B of SCI-P-057.

    Refer to Table 7C.6 for more information

    DFF

    0

    (Mandatory

    for deflection

    check,

    TRACK 4.0)

    "Deflection Length" / Max.. allowable local

    deflection

    See Note 1d below.

    DJ1 Start Joint

    of member

    Joint No. denoting starting point for calculation

    of "Deflection Length" . See Note 1 below.

    End Joint of Joint No. denoting end point for calculation of

    Page 30 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • DJ2 member "Deflection Length". See Note 1 below.

    DMAX 100.0 cm Maximum allowable depth for the member.

    DMIN 0 Minimum required depth for the member.

    EFT Member

    Length

    Effective length for torsion. A value of 0

    defaults to the member length.

    ELB 0

    Used to specify the method for combined axial

    load + bending checks

    0. Uses Cl. 6.2.9 of EN 1993-1-1:2005

    1. Uses Cl. 6.2.1(7) - Eqn. 6.2 of EN 1993-

    1-1:2005

    ESTIFF 0

    (For use with the Dutch NA only) Method for

    checking columns forming part of

    (non)/buttressed framework:

    0. Checks per Cl 12.3.1.2.3 of NEN 6770:

    Section 1

    1. Checks per Cl 12.3.1.2.3 of NEN 6770:

    Section 2

    See "Clause 12.3.1.2.3 (NEN 6770):

    Rotation/bending capacity" for additional

    description on this parameter.

    FAB 3

    Used to specify the fabrication class to be used

    to check for slender (Class 4) CHS/pipe

    sections (EN 1993-1-6:2007)

    1. Class A Excellent

    2. Class B High

    3. Class C Normal

    FU 0 Ultimate tensile strength of steel.

    GM0 1.0 Corresponds to the m0 factor in EN 1993-1-

    1:2005

    GM1 1.0 Corresponds to the m1 factor in EN 1993-1-

    1:2005

    GM2 1.25 Corresponds to the m2 factor in EN 1993-1-

    1:2005

    Used to specify the section type to be used for

    designing a General Section from the user

    table. The member will be considered as the

    Page 31 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • GST 0

    specified type with the user defined properties.

    The available options and corresponding values

    are as below:

    0. I-Section

    1. Single Channel

    2. Rectangular Hollow Section

    3. Circular Hollow Section

    4. Angle Section

    5. Tee Section

    This parameter will be ignored if it has been

    assigned to any section other than a General

    Section.

    KC 1.0

    Corresponds to the correction factor as per

    Table 6.6 of EN 1993-1-1:2005. Program will

    calculate kc automatically if this parameter is

    set to 0.

    For the British, Singapore, & Polish NAs, kc

    will be calculated as given in the NA by default.

    KY 1.0

    K factor in local y axis. Used to calculate the

    effective length for slenderness and buckling

    calculations.

    KZ 1.0

    K factor in local z axis. Used to calculate the

    effective length for slenderness and buckling

    calculations.

    LEG 0

    Slenderness values for angles as determined

    from BS 5950-2000 Table 25.

    See "British Codes - Steel Design per

    BS5950:2000"

    LVV Max. value of

    Lyy

    Leg length for Lvv (length about v-v- axis of

    single angle section), as per Lyy. Used for

    slenderness calculations.

    LY Member

    Length

    Compression length in local y axis, Slenderness

    ratio = (KY)*(LY)/(Ryy)

    LZ Member

    Length

    Compression length in local z axis, Slenderness

    ratio = (KZ)*(LZ)/(Rzz)

    Used to select the clause to be used to calculate

    Page 32 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • MTH 0

    the LTB reduction factor, LT. The available

    options and corresponding values are as below:

    0. Use default method based on section type

    (default)

    1. Use Cl.6.3.2.2

    2. Use Cl.6.3.2.3

    By default, the program will use Cl 6.3.2.3 for

    rolled & built-up I-sections and Cl. 6.3.2.2 for

    all other sections. If, however, the specified

    National Annex expands on Cl. 6.3.2.3 to

    include other section types (e.g., the UK NA),

    the program will use Cl. 6.3.2.3 by default for

    that particular section type.

    See "European Codes - National Annexes to

    Eurocode 3 [EN 1993-1-1:2005]" for additional

    details on NA documents.

    MU 0

    To be used with CMM values of 7 and 8. See

    Table 7C.4.

    Currently valid only with the French & Belgian

    NAs.

    NA 0

    Choice of National Annex to be used for EC3

    design. See "European Codes - National

    Annexes to Eurocode 3 [EN 1993-1-1:2005]"

    for values allowed for this parameter.

    (See "National Annex Documents" for more

    information)

    NSF 1.0 Net tension factor for tension capacity

    calculation.

    PLG 0

    To be used to determine whether to include the

    additional interaction checks as per CL. NA.20

    (2) and NA.20(3) of the Polish National Annex.

    This parameter will be applicable only to the

    Polish NA

    PY Yield

    Strength

    The yield strength default value is set based on

    the default value of the SGR parameter.

    Page 33 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • RATIO 1 Permissible ratio of loading to capacity.

    SBLT 0.0

    Indicates if the section is rolled or built-up.

    0.0 = Rolled

    1.0 = Built-up

    SGR 0

    Steel grade as in table 3.1 of EN 1993-1-1:2005

    0.0 - indicates S 235 grade steel

    1.0 - indicates S 275 grade steel

    2.0 - indicates S 355 grade steel

    3.0 - indicates S 420 grade steel

    4.0 - indicates S 460 grade steel

    As EN 1993-1-1:2005 does not provide a

    buckling curve in table 6.2 for grade S 450 steel

    (in Table 3.1 of EN 1993-1-1:2005), the

    program will use the same buckling curves as

    for grade S 460 when calculating the buckling

    resistance as per clause 6.3.

    STIFF

    Member

    Length or

    depth of

    beam,

    whichever is

    lesser

    Distance between transverse stiffener plates,

    used to prevent web shear buckling. If not

    specified or if a value of 0 is provided, the

    program will assume the web is unstiffened.

    TOM 0

    Total torsion for design used for torsion checks.

    Can be used to override the total torsional

    moment to be used for member design.

    TORSION 0

    Method to be used for a specific member or

    group of members:

    0. Perform basic torsion checks if member is

    subject to torsion.

    1. Perform basic stress check (Ignore

    warping effects).

    2. Perform detailed checks (including

    Page 34 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Notes:

    1. CAN, DJ1, and DJ2 Deflection

    a. When performing the deflection check, you can choose between two methods. The first method, defined by a value 0 for the CAN parameter, is based on the local displacement. Local displacement is described in Section 5.44 of the Technical Reference Manual.

    If the CAN parameter is set to 1, the check will be based on cantilever style deflection. Let (DX1, DY1, DZ1) represent the nodal displacements (in global axes) at the node defined by DJ1 (or in the absence of DJ1, the start node of the member). Similarly, (DX2, DY2, DZ2) represent the deflection values at DJ2 or the end node of the member.

    Compute Delta = SQRT((DX2 - DX1)2 + (DY2 - DY1)2 + (DZ2 - DZ1)2)

    Compute Length = distance between DJ1 & DJ2 or, between start node and end node, as the case may be.

    Then, if CAN is specified a value 1, dff = L/Delta

    Ratio due to deflection = DFF/dff

    warping effects).

    3. Ignore all torsion checks

    For options 1 or 2, the program will perform the

    torsion related checked even if torsional

    moment is absent and will use a value of zero

    for the torsional moment.

    TRACK 0

    Specify level of detail in output.

    0. Summary of results only.

    1. Summary with member capacities.

    2. Detailed results.

    4. Deflection check results only.

    UNF 1Unsupported length as a fraction of the actual

    member length.

    UNL Member

    Length

    Unrestrained length of member used in

    calculating the lateral-torsional resistance

    moment of the member.

    ZG +Section

    Depth/2

    Distance of transverse load from shear center. Used to calculate Mcr.

    For Tee sections, ZG will have a default value

    of (+Flange thickness/2)

    Page 35 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • b. If CAN = 0, deflection length is defined as the length that is used for calculation of local deflections within a member. It may be noted that for most cases the Deflection Length will be equal to the length of the member. However, in some situations, the Deflection Length may be different. A straight line joining DJ1 and DJ2 is used as the reference line from which local deflections are measured.

    For example, refer to the figure below where a beam has been modeled using four joints and three members. The Deflection Length for all three members will be equal to the total length of the beam in this case. The parameters DJ1 and DJ2 should be used to model this situation. Thus, for all three members here, DJ1 should be 1 and DJ2 should be 4.

    D = Maximum local deflection for members 1, 2, and 3.

    PARAMETERS

    DFF 300. ALL

    DJ1 1 ALL

    DJ2 4 ALL

    c. If DJ1 and DJ2 are not used, "Deflection Length" will default to the member length and local deflections will be measured from original member line.

    d. It is important to note that unless a DFF value is specified, STAAD will not perform a deflection check. This is in accordance with the fact that there is no default value for DFF (see Table 2B.1).

    e. The above parameters may be used in conjunction with other available parameters for steel design.

    2. CMM Parameter

    The values of CMM for various loading and support conditions are as given below:

    Table 7C.3-Values for the CMM ParameterCMM

    ValueLoading and Support Conditions

    1

    2

    3

    Page 36 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • 3. Checking beam deflection

    With the TRACK parameter set to 4, the members included in a BEAM CHECK command will be checked for the local axis deflection rather than for the stress capacity using the current LOAD LIST.

    If both stress capacity and deflection checks are required, then 2 parameter blocks with code checks are required, one with a TRACK 4 command and one with a TRACK 0, 1 or 2, thus:

    LOAD LIST 1 TO 10

    PARAMETER 1

    CODE EN 1993

    TRACK 2 ALL

    CODE CHECK MEMBER 1

    ***************************

    LOAD LIST 100 TO 110

    PARAMETER 2

    TRACK 4 ALL

    DFF 300 MEMB 1

    DJ1 1 MEMB 1

    DJ2 4 MEMB 1

    4

    5

    6

    7

    varying end moments and uniform loading

    8

    varying end moments and central point load

    Page 37 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • CHECK CODE MEMB 1

    While both sets of code checks will be reported in the output file, only the last code check results are reported in the STAAD.Pro graphical interface.

    4. CMT Parameter

    The values of CMM for various loading and support conditions are as given below:

    For CMT = 2 and CMT = 3, you have the option of specifying the distance at which the concentrated torque acts, measured from the start of the member. This can be done by using the ALH design parameter. The ALH parameter indicates the ratio of the distance of the point torque (from the start of the member) to the length of the member. This parameter will have a default value of 0.5 (i.e., the torque acts at the center of the span) and will accept values ranging from 0 to 1.

    The GB1 parameter that is being used for compression checks in builds preceding this release (STAAD.Pro 2007 build 06) has been removed as this parameter is no longer required in EN 1993-1-1:2005. Hence any legacy files that use GB1 parameter will indicate an error message and you will be required to substitute GB1 with GM1, in accordance with EN 1993-1-1:2005.

    Table 7C.4-Loading and Support Conditions represented by CMT Parameter Values

    CMT Value

    Description Diagram

    1 (Default) : Concentrated Torque at Ends. Ends Torsion fixed and Warping fixed

    2 Concentrated Torque along length of member. Ends Torsion fixed and Warping free

    3 Concentrated Torque along length of member. Ends Torsion fixed and Warping fixed

    4 Uniform Torque in member. Ends Torsion fixed and Warping free

    5 Uniform Torque in member. Ends Torsion fixed and Warping fixed

    6 Concentrated Torque in cantilever. End Torsion fixed and Warping fixed

    7 Uniform Torque in cantilever. End Torsion fixed and Warping fixed

    Page 38 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.7 Code Checking

    The purpose of code checking is to ascertain whether the provided section properties of the members are adequate. The adequacy is checked as per EN 1993-1-1:2005 and a corresponding National Annex (if specified). Code checking is done using the forces and moments at specific sections of the members.

    When code checking is selected, the program calculates and prints whether the members have passed or failed the checks; the critical condition; the value of the ratio of the critical condition (overstressed for value more than 1.0 or any other specified RATIO value); the governing load case, and the location (distance from the start of the member of forces in the member where the critical condition occurs).

    Code checking can be done with any type of steel section listed in Section 2B.4 or any of the user defined sections as described in Section 1.7.3 of the Technical Reference Manual, with the exception of ISECTION. ISECTION has been currently excluded since the option of Tapered section design is currently not supported in the EC3 module. The EC3 (EN 1993) design module does not consider these sections or PRISMATIC sections in its design process.

    Checks for slender sections to EN 1993-1-1 are limited to I-SECTIONS, TEE, SINGLE CHANNEL, SINGLE ANGLE and CIRCULAR & RECTANGULAR HOLLOW SECTIONS.

    Code checking for GENERAL sections can be also done using the EN1993 module. The program will design GENERAL sections as I sections by default. However, you are given the option to choose a section type to be considered while designing the member. Refer to the description of the GST design parameter in Section 7C.6 for details.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.8 Member Selection

    STAAD is capable of performing design operations on specified members. Once an analysis has been performed, the program can select the most economical section, i.e., the lightest section, which fulfills the code requirements for the specified member. The section selected will be of the same type section as originally designated for the member being designed. Member selection can also be constrained by the parameters DMAX and DMIN, which limits the maximum and minimum depth of the members.

    Page 39 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Member selection can be performed with all the types of steel sections with the same limitations as defined in Section 7C.7.

    Selection of members, whose properties are originally input from a user created table, will be limited to sections in the user table.

    Member selection cannot be performed on members whose section properties are input as prismatic or as the limitations specified in Section 7C.7.

    International Design Codes

    European Codes - Steel Design to Eurocode 3 [EN 1993-1-1:2005]

    7C.9 Tabulated Results of Steel Design

    For code checking or member selection, the program produces the results in a tabulated fashion. The items in the output table are explained as follows:

    MEMBER refers to the member number for which the design is performed.

    TABLE refers to steel section name, which has been checked against the steel code or has been selected.

    RESULTS prints whether the member has PASSED or FAILED. If the RESULT is FAIL, there will be an asterisk (*) mark on front of the member.

    CRITICAL COND refers to the clause in EN 1993-1-1:2005 code which governs the design.

    RATIO prints the ratio of the actual stresses to allowable stresses for the critical condition. Normally a value of 1.0 or less will mean the member has passed.

    LOADING provides the load case number, which governed the design.

    FX, MY, and MZ provide the axial force, moment in local Y-axis and the moment in local z-axis respectively. Although STAAD does consider all the member forces and moments (except torsion) to perform design, only FX, MY and MZ are printed since they are the ones which are of interest, in most cases.

    LOCATION specifies the actual distance from the start of the member to the section where design forces govern.

    For a TRACK 2 output, the module will also report all the relevant clause checks that have been performed and will also indicate the critical ratio and the load case that caused the critical ratio as well as the corresponding forces that were used for the respective checks. A TRACK 2 output will also include the various design data used for the calculations such as the section modulii, section class, section capacity etc.

    If an NA parameter (other than 0) has been specified and if the particular National Annex requires

    Page 40 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • additional checks outside those specified in EN 1993-1-1:2005 (e.g., The Dutch National Annex), the respective NA clauses and any associated code clauses will be listed along with the critical ratios and the forces that were used for these clause checks.

    7C.9.1 Example of a TRACK 2 output

    Documentation notes appear in red.

    The results and output follow the axis convention as described in Section 7C.1.3

    Code title & version

    STAAD.PRO CODE CHECKING - BS EN 1993-1-1:2005

    ********************************************

    National Annex used, if any

    NATIONAL ANNEX - NA to BS EN 1993-1-1:2005

    Design engine version

    PROGRAM CODE REVISION V1.9 BS_EC3_2005/1

    ALL UNITS ARE - KN METE (UNLESS OTHERWISE NOTED)

    MEMBER TABLE RESULT/ CRITICAL COND/ RATIO/ LOADING/

    FX MY MZ LOCATION

    =======================================================================

    Member number, section profile & table

    1 ST HD320X127 (EUROPEAN SECTIONS)

    Design status, critical code clause, & critical ratio

    PASS EC-6.3.3-662 0.045 1

    Section forces & critical section location

    25.00 C 5.00 -10.00 0.00

    =======================================================================

    MATERIAL DATA

    Grade of steel = USER

    Modulus of elasticity = 205 kN/mm2

    Design Strength (py) = 275 N/mm2

    SECTION PROPERTIES (units - cm)

    Page 41 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • Member Length = 500.00

    Gross Area = 161.30 Net Area = 161.30

    "z-axis" here refers to bending about Z-Z (when Y is Up), where as EC3 uses the Y-Y axis convention.

    z-axis y-axis

    Moment of inertia : 30820.004 9239.001

    Plastic modulus : 2149.000 939.100

    Elastic modulus : 1926.250 615.933

    Shear Area : 81.998 51.728

    Radius of gyration : 13.823 7.568

    Effective Length : 500.000 500.000

    DESIGN DATA (units - kN,m) EUROCODE NO.3 /2005

    Section class as per Table 5.2

    Section Class : CLASS 1

    Max. cross section capacity (A fy/GM0

    Squash Load : 4435.75

    Axial force/Squash load : 0.006

    Partial safety factors used

    GM0 : 1.00 GM1 : 1.00 GM2 : 1.10

    z-axis y-axis

    Slenderness ratio (KL/r) : 36.2 66.1

    Compression Capacity : 4078.2 3045.5

    Tension Capacity : 4435.8 4435.8

    Moment Capacity : 591.0 258.3

    Reduced Moment Capacity : 591.0 258.3

    Shear Capacity : 1301.9 821.3

    BUCKLING CALCULATIONS (units - kN,m)

    Lateral Torsional Buckling Moment MB = 591.0

    Factor C1 used in Mcr calculations and End restraint factor (corresponds to the CMN design

    Page 42 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm

  • parameters

    co-efficients C1 & K : C1 =2.578 K =1.0, Effective Length= 5.000

    Elastic Critical Moment for LTB, Mcr = 1541.5

    Critical Load For Torsional Buckling, NcrT = 13898.0

    Critical Load For Torsional-Flexural Buckling, NcrTF = 13898.0

    ALL UNITS ARE - KN METE (UNLESS OTHERWISE NOTED)

    MEMBER TABLE RESULT/ CRITICAL COND/ RATIO/ LOADING/

    FX MY MZ LOCATION

    =======================================================================

    CRITICAL LOADS FOR EACH CLAUSE CHECK (units- kN,m):

    Max. ratio, loadcase, & section forces for each clause check

    CLAUSE RATIO LOAD FX VY VZ MZ MY

    EC-6.3.1.1 0.008 1 25.0 0.0 0.0 -10.0 5.0

    EC-6.2.9.1 0.020 1 25.0 0.0 0.0 -10.0 5.0

    EC-6.3.3-661 0.035 1 25.0 0.0 0.0 -10.0 5.0

    EC-6.3.3-662 0.045 1 25.0 0.0 0.0 -10.0 5.0

    EC-6.3.2 LTB 0.017 1 25.0 0.0 0.0 -10.0 5.0

    Torsion and deflections have not been considered in the design.

    _________________________

    ************** END OF TABULATED RESULT OF DESIGN **************

    Page 43 of 43EC3 -EN 1993_TOC

    3/4/2014file://C:\Documents and Settings\agpfm\Configuracin local\Temp\~hh1167.htm