Sub model vs full model
Hi,
I'm trying to use a submodel to get a deeper understanding of a complex part. I tried matching the reaction forces of the two models and there was a factor difference? See below my approach used. I used 2D Shell models for the full model and 3D Tet mesh for the sub model. The flange face and SPC's are in the same location.
Is this approach correct?
What is the best way to verify the reaction forces are translated correctly?
Sub model SPCM on flange face and pinned constraints at the hinges. 
Sub model load step - ULS case 3
Load step 3 from full model - ULS case 3
Best Answer
-
I attached a small tutorial I made a couple of years ago. It is using the classic interface, and is using disp rather than force but the principles are the same, Hope it helps, there is an example with it too.
Your original approach should work too though, not sure what the issue is there, if you can share (even a simplified version) then I could take a look for you?
1
Altair EmployeeAnswers
-
nice study and really nice usage of submodel.
as far as I understand, in submodel, we map the displacement map from global run to submodel.
So, in theory the displacement is captured and applied accurately to your new model.
If the reaction force is not accurate, I would assume that your stiffness of both models is too different from each other (or at least the load path).
Maybe you could calculate the FBD and check the forces as well, and maybe trying applying forces intead of displacements, as a double check.
0 -
Hi Adriano,
Thank you, I do see the value in this technique I just need to get confidence in the results.
The FBD and full model lines up perfectly. I also added the self weight of the sub model and played around with std SPC's at the pivot points and SPCM at all the constraint locations, but the reactions remains the same.
First Image - Shows: Sub model RF without grav, Full model with RF, Submodel with GRAV
But when you look at the displacement they are identical. See image two, DISP at flange face. 50 mm rotating in the correct direction.
I can sere how stiffness may alter the reactions a bit, but not by this much, can I do a submodel with the loads instead like suggested?
0 -
Casper Kruger said:
Hi Adriano,
Thank you, I do see the value in this technique I just need to get confidence in the results.
The FBD and full model lines up perfectly. I also added the self weight of the sub model and played around with std SPC's at the pivot points and SPCM at all the constraint locations, but the reactions remains the same.
First Image - Shows: Sub model RF without grav, Full model with RF, Submodel with GRAV
But when you look at the displacement they are identical. See image two, DISP at flange face. 50 mm rotating in the correct direction.
I can sere how stiffness may alter the reactions a bit, but not by this much, can I do a submodel with the loads instead like suggested?
submodel in OS is based on displacements.
For loads you would need to go through the FDB tools in HM/HV to map loads from one to another.
honestly i don't know what might be the cause of this difference.
I'd test it in a small model first to make sure if this is reproducible.
0 -
Hi Ardriano,
Is there any examples that shows how to export FBD results to another sub model?
Free Body Diagrams (FBD) (altair.com) - This link is not very clear how this works with a sub model.
Any guidance will be appreciated. I created GPForce outputs for the interface element of the submodel. The H3D output did not work and I am now trying OP2. The FBD tool in Hyperworks does not pick up the results, is there an up to date example running through the FBD tool and sub modelling?
0 -
I attached a small tutorial I made a couple of years ago. It is using the classic interface, and is using disp rather than force but the principles are the same, Hope it helps, there is an example with it too.
Your original approach should work too though, not sure what the issue is there, if you can share (even a simplified version) then I could take a look for you?
1 -
Is there a tutorial available (preferably in video format) that can teach me how to properly run a submodel analysis?
0 -
Hi,
The models I am using are of significant weight. The sub model alone weighs 90Ton. When I map the loads from the main model to the sub model, do I still have to allow for the self weight of the sub model or will this be allowed for in the displacement mapping?
The main model has all the self weights in as Mass at the COG.
Thank you.
Casper
0 -
HI
Will FBD method supports material Non-linearity ?0