Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

LOADADD for SPCD load after pretension in non-linear analysis with contact

Hello experts,

I'm looking for advice on how to properly use LoadAdd for these different load types. The documentation states "SPCD load SIDs cannot be referenced by LOADADD data". Still, combining a pretension force with fixed displacement is desired here. So a simple cube with friction surface is created to check what is possible. A single element block pushes on the plate, then the plate moves sideways:

Two variations of the above model (optistruct solver decks) are attached:

- "SPCD_Sub2.fem" > SPCD is applied in subcase 2 (via Loadadd). This doesnt work in subcase 2 (the plate does not move).

- "SPCD_Sub1_and_Sub2.fem" > The same SPCD is referenced in both cases (first one directly, second one via loadadd). It works in both subcases as shown in the graph below.

Graph showing reaction (friction) force as function of plate displacement:

Question:

The expected (and desired) behavior in "SPCD_Sub2.fem" is first pretension, then displacement in a separate subcase. However I can only get it working if the SPCD is active in both of them. Is there a proper way to separate the different loads to have Sub1 with only pretension and Sub2 with only SPCD added on top of that pretension?

Community-Pasted-Attachment-1727776648874.png

SPCD_Sub1_and_Sub2.fem

SPCD_Sub2.fem

Find more posts tagged with

English

OptiStruct

v2023.1

Accepted answers

All comments

Rajashri_Saha

Hi Wesley,

I just went through your study and understood that in 1st step you want to push the block by applying force in Y-dir and in 2nd step you want to continue from 1st and along with that applying enforced displacement in -Z direction. You can do this by taking your 1st model "SPCD_Sub2.fem" and remove the loadadd collector, appy only "Loads_spcd". Attached is the updated file.

When reviewed "SPCD_Sub1_and_Sub2.fem", you have applied enforced disp twice in 2nd loadstep.

Thanks

Rajashri

SPCD_Sub2_spcd.fem

Wesley_Peijnenburg2

Hi Wesley,

I just went through your study and understood that in 1st step you want to push the block by applying force in Y-dir and in 2nd step you want to continue from 1st and along with that applying enforced displacement in -Z direction. You can do this by taking your 1st model "SPCD_Sub2.fem" and remove the loadadd collector, appy only "Loads_spcd". Attached is the updated file.

When reviewed "SPCD_Sub1_and_Sub2.fem", you have applied enforced disp twice in 2nd loadstep.

Thanks

Rajashri

Hi Rajashiri,

Thanks for checking the model. I downloaded "SPCD_Sub2.fem" taht you provided and ran it, the results show that the force from step 1 fades to zero during step 2:

However, the goal of this model is to keep the "pretension/force" from step 1 active and constant during step 2. I'm expecting to see a (mostly) horizontal line in the upper graph with only the first iteration ramping up to that constant value of 1N (10% friction force). Do you know a good way to do this?

- Wesley

Wesley_Peijnenburg2

Hi Rajashiri,

Thanks for checking the model. I downloaded "SPCD_Sub2.fem" taht you provided and ran it, the results show that the force from step 1 fades to zero during step 2:

However, the goal of this model is to keep the "pretension/force" from step 1 active and constant during step 2. I'm expecting to see a (mostly) horizontal line in the upper graph with only the first iteration ramping up to that constant value of 1N (10% friction force). Do you know a good way to do this?

- Wesley

Just got it working without the loadadd, however it requires the initial force to be duplicated into the second loadcase. See attached fem file.

This is the desired behavior, but I am still wondering if copy pasting is the recommended way to do this?

2024-10-02_10-35-28_run_now.fem

Rajashri_Saha

Just got it working without the loadadd, however it requires the initial force to be duplicated into the second loadcase. See attached fem file.

This is the desired behavior, but I am still wondering if copy pasting is the recommended way to do this?

I think you got the correct work around, If the load needs to be active in 2nd step then you need to have it applied in 2nd step too.