Microstran - Automatic calculation of steel member effective length

[Settingsun]

Hi,
I'm hoping someone can clarify how Microstran uses the Elastic Critical Load analysis to determine compression effective length in its steel design module.

I have a 125*5 square hollow section truss chord (Ixx = Iyy = 5.44E6 mm4) that has a maximum axial compression of 7.73 kN in a particular load case. The elastic critical load analysis gives a buckling load factor of 55.38 for this load case. Looking at the buckling shape from the elastic critical load analysis, the buckling mode is about the Y axis (out of the truss plane) between rafters at centres that vary but a bit under 5m maximum.

I thought the design module would use an axial force of 7.73 kN * 55.38 = 428 kN to calculate an effective length of 5.01m, for both X and Y axis buckling (since Ixx = Iyy). However it's actually using Lx = 38.0m and Ly = 6.8m. The chord is about 14m long. Lx corresponds to buckling in the plane of the truss in which diagonals intersect the chord at 1.5m centres.

Because of the long effective length, the steel design module give a fail result for this chord due to low member capacity in compression. I'm struggling to understand how this could be the case since the buckling load factor was 55.38.

Appreciate your thoughts.

Edit: The load case is earthquake and the chord axial force is a little odd, going from compression to tension to compression. Could the program be averaging the axial load along the member length for calculation of effective length, then using the maximum compression for the design check? Still not sure that would explain different lengths for the X & Y axes.

 

[rowingengineer]

Can you print the design report to pdfso we can see what the capacties are.

http://www.nceng.com.au/

"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning."

 

[Settingsun]

Here's the steel design report along with a picture of the buckling load.

I may have checked the wrong box somewhere so will play with the options when I have a bit more time, but help also appreciated.

 

[SethGuthrie]

The Lx and Ly values you noted are clearly on the report. These are user determined values in Microstran based on the member lateral restraint data. Get there via Design menu - Design data - Input/edit then check the Restraint Data dialog. Search the help for "Lateral Restraint Data" to read more.

 

[Settingsun]

The Lx and Ly values were calculated by Microstran using the elastic critical load analysis results and the restraint data. I played a bit last night and removed the restraint data. (I had previously entered them for manual effective length calculation and not deleted them when switching to ECL method, assuming they would be ignored). The result was that both Lx and Ly are now 5.01m and the member passes the design check, both as expected.

So, it turns out that nominating restraints may reduce the calculated member capacity when using the ECL method. I haven't got to the bottom of it yet, but I suspect that Microstran used a small axial force to calculate the effective length, then used the largest axial force for the design check. Inconsistent on the conservative side if so.

 

[SethGuthrie]

I asked one of my colleagues in Australia for any additional insight and he said:
I think this relates to this excerpt from the Microstran Manual and also Bentley Communities article:

Results from ECL Analysis
If results are available from an elastic critical load (ECL) analysis, you may enter “ECL” in the kx and ky fields to have effective lengths computed from the elastic buckling load factors (c). The results from an ECL analysis will only be applicable if the model includes all member restraints relevant to column action. If you change column restraints during member design ECL analysis results may not be correct. In this case, repeat the ECL analysis after any change to the column restraints. ECL analysis results will be most useful for structures modelled as plane frames where all out-of-plane action is restrained. ECL analysis results must be available for all the load cases nominated for design. Effective length factors for out-of-plane column buckling cannot be computed if the structure has been modelled as a plane frame. You may manually input effective length factors if you do not wish to use the values derived from the ECL analysis.

https://communities.bentley.com/pro...149.microstran-elastic-critical-load-analysis

At the bottom of the article it specifically discusses the difference between Frame Buckling as calculated by ECL and Traditional Effective Length Factor.