Beam Deformations

[SparWeb]

I cannot understand the behaviour I am seeing.

Attached is a screen-capture of the undeformed structure.  Essentially a network of beams secured with pin joints.  The group of beams is symmetrical and I've zoomed into a quarter of it.  Note the unit load at C2.

In a second post, I will attach the deformed structure I'm seeing.  I can make no sense of it.

[SparWeb]

Follow-up posting with the deformed appearance.

[SparWeb]

Below are the List Output Query values.

Node 958 is marked "C2" and Node 421 is marked "A2" in the screen-shots above.

List Output Query
Node 958
  Output Set       14 : NX NASTRAN Case 1
        Value =     0.000195636  |  Output Vector        2 : T1 Translation
        Value =    -0.000133336  |  Output Vector        3 : T2 Translation
        Value =       0.0101353  |  Output Vector        4 : T3 Translation
 
Node 421
  Output Set       14 : NX NASTRAN Case 1
        Value =    0.0000840457  |  Output Vector        2 : T1 Translation
        Value =     0.000153157  |  Output Vector        3 : T2 Translation
        Value =      0.00123497  |  Output Vector        4 : T3 Translation

[Stuart_S]

Why not try running a free free modes to understand how the entire sytems is reacting and ensure it has 6 rigid bodies then assess the next few structural modes.

[SparWeb]

I cannot determine the relevance of your comment to my query.

This is a static analysis, where the load is applied at one point in one direction.

My concern is that FEMAP is not displaying a deformed view consistent with the NX NASTRAN output.

[T_Giampietro]

FEMAP scales the deformation in the view. The user can always set this to "Actual Deformations"  in the PostProcessing Toolbox. Perhaps this is what you're seeing?

Set to scaled:

Set to "Actual Deformations"

[SparWeb]

Both % scale and actual deformations exaggerate the deflections of the wrong beam.

AFAIK there is no way to exaggerate the deflections of ONE component at a time.

[Stuart_S]

Apologies if I’m going over old ground.

But by your own admission “I cannot understand the behaviour I am seeing” would suggest that the model is behaving unexpectantly - not that there is a graphical error (at first glance, without looking assessing the numbers).

The free free modes is used to establish baseline validity /sanity check on the FEM to ensure it is behaving as expected.

If you don’t want to run a modal, and If the uncertainly is in the model, check the F06 for SPC=OLOAD, what nodes/ if any are being autospc’d if AUTOSPC PARAM is on.

If that’s all good, then Femap is the culprit.

[SparWeb]

Hi Stuart.

No old ground here - I only recently took the intro course and signed up to the forum today.  If anything is unclear, it's probably my clumsy explanations that don't fit the terminology you expect to use.

The behaviour I'm concerned about is the graphical presentation of the deformed view.

Note that the upper beam deforms with no load, and the lower beam does not deform when a load is applied to it.  That makes no sense.

Furthermore, the output data from the NASTRAN solution that I posted in #3 confirms that the deformation of the lower beam should be 10x greater than the upper beam - exactly what I would expect given the load and where it is placed on the lower beam.  But the opposite is shown on screen.

That's the confusing behaviour I'm referring to.

PS I haven't learned the F06 "language" yet so I can only respond to your suggestion robotically.

The string "SPC=OLOAD" is not present in the F06 file.

I was able to find reaction to the input load vector here:

0                                                  OLOAD    RESULTANT       
  SUBCASE/    LOAD
  DAREA ID    TYPE       T1            T2            T3            R1            R2            R3
0        1     FX    0.000000E+00     ----          ----          ----       0.000000E+00  0.000000E+00                             
               FY       ----       0.000000E+00     ----       0.000000E+00     ----       0.000000E+00                             
               FZ       ----          ----       1.000000E+02  1.800000E+03 -3.062500E+04     ----                                  
               MX       ----          ----          ----       0.000000E+00     ----          ----                                  
               MY       ----          ----          ----          ----       0.000000E+00     ----                                  
               MZ       ----          ----          ----          ----          ----       0.000000E+00                             
             TOTALS  0.000000E+00  0.000000E+00  1.000000E+02  1.800000E+03 -3.062500E+04  0.000000E+00                             
 *** USER INFORMATION MESSAGE - SINGULARITIES FOUND USING EIGENVALUE METHOD
 ***       0 SINGULARITIES FOUND         0 SINGULARITIES ELIMINATED

 I do not know why there are values on the FZ line other than "1.00000E+02".  I didn't put any other loads on the model.

[Stuart_S]

It is a peculiar one! Without running the model locally it’s hard to give further advice. Attach the dbf/dat if you can.

On the F06 one. Essentially search for OLOAD and note down the applied as you have done - then search for SPC and do the same. These values should equal zero or a very very small value depending on the epsilon value (rounding error).