SAP2000 deflection error when bar is meshed or divided into parts

[Ashok Sinha Karanam]

Hi,

To validate SAP2000 trail version v14.2.4, I have created a simply supported beam and assigned a pointed frame-load at its mid span. I got certain displaement, x1, which is precisely matching the value WL3/48EI.

Now I had meshed the beam using a maximum mesh size of span/100 and then assigned a joint load of same value at its mid point and rerun the analysis. To my surprise I got a different midspan deflection x2 which is slightly higher than x1 and the difference between x1 and x2 is considerable.

I rerun the anlaysis with different solvers and different mesh sizes, and also instead of meshing I divided the frame into multiple frames, but the result didn't change.

Can anybody explain the reason behind this.

 

[JoshPlumSE]

Could it be shear deformation? What's your span to depth ratio?

 

[Ashok Sinha Karanam]

Hi JoshPlum,

It is an I section having span to depth ratio as 10.077.

Other characteristics are as follows.

Simply supported.
Span 13.1m
Material=STEEL Shape="I/Wide Flange" t3=1.3 t2=0.55 tf=0.025 tw=0.01 t2b=0.55 tfb=0.025 Area=0.04 TorsConst=5.97967083333334E-06 I33=1.28052083333333E-02 I22=6.93333333333334E-04 AS2=0.013 AS3=2.29166666666667E-02 S33=1.97003205128205E-02 S22=2.52121212121212E-03 Z33=0.0214375 Z22=0.0038125 R33=0.565800502238495 R22=0.131656117720877 Color=Cyan FromFile=No AMod=1 A2Mod=1 A3Mod=1 JMod=1 I2Mod=1 I3Mod=1 MMod=1 WMod=1
All dimension in meters.

 

[JoshPlumSE]

The reason why I asked about shear deformation is because I have seen the same question with RISA a number of times and the cause was usually shear deformation. With RISA shear deformation only gets calculated at Joint locations. Therefore, if you use two joints to model a simply supported beam, then you don't see the shear deformation. If you sub-divide it, then you will see the shear deformation. With RISA it was very easy to turn off the effect of shear deformation. But, if SAP has that ability, I don't know where it would be done.

Another idea could be related to stiffness adjustments for the "direct analysis" method from AISC. Be sure to turn off these stiffness adjustments if you are mostly interested in deflection.

 

[JasonMcKee71]

This should not change. Shear deformation effects don’t affect the results of frames structures, as their stiffness matrix is derived analytically and not numerically. Consequently another setting may result with n that behavior, for instance the distribution of the loads to the adjacent frames.

Jason McKee
proud R&D Manager of
Cross Section Analysis & Design
Software for the structural design of cross sections
Moment Curvature Analysis
Interaction Diagrams
Reinforcement Design etc.