CMS2 vs. CTRQ difference problem and Not having the applied torque till the end of the simulation p. 1

[Yunus9696]

Hello everyone

I'm trying to run a 2D-axisymmetric simulation of a threaded pin in a drilled box, to see the effects of screw insertion torque and axial loading according to the example of Abaqus documentation "Axisymmetric analysis of a threaded connection" :

http://130.149.89.49:2080/v6.14/books/exa/default.htm?startat=ch01s01aex20.html

I apply the twist (torque) and axial displacement (loading) in two subsequent steps (step 3 and 4 respectively).
there are 2 problems with my results:
1. the values of CTRQ (Maximum load transferred) and CMS2 (the amount of torque about the screw axis due to frictional stress) do not coincide after the torque application and there is a huge difference.
2.Based on history output, once the torque application step is done the CMS2 value drops, however I expect that the amount of torque stay the same (since we need to apply the same torque unscrew the screw, from a physical point of view).

How could I configure the settings to have the effect of the torque till the end of my simulation and also get a reasonable amount of CMS2?

Kind regards,
Yunus.

PS: the History output figures are kindly attached .

 

[FEA way]

If your model is tightly based on the documentation example, finding any differences between these two models (in terms of analysis features, not geometry) may reveal the cause of the problem. Was something in your model done differently ?

 

[Yunus9696]

Yah, I also tried to debug this problem by comparing two models but the only differences I could find are:

1. Geometries: Abaqus documentation model is consisted of two long threaded pipes (inner and outer pipe) with internal presure while my model is a small dental implant without any internal pressure.
2. Abaqus documentation model has a pressure penetration option to check pipe leakage from the threaded location.
3. Materials
4. My right side of the model is fixed to prevent threads get apart from the implant but in the Abaqus documentation model the first threads of the box (the outer pipe) can move away from the pin (the inner pipe) threads.
5. The order of torque and axial load application: In My model I first apply the twist and propagated it to the axial load application step, while in Abaqus documentation model the axial load is first applied and it is propagated to the torque application step.

None of these differences seems to be related to, at least, huge difference between CMS2 and CTRQ values in my model as I illustrated in the attached pictures with red color.

Thanks for your comments,
Yunus.

 

[FEA way]

Your mesh is very coarse, it might be better to refine it, especially at the sharp thread tips.

The screenshot that you attached to your first post shows history output from Abaqus documentation example, right ? Can you share the equivalent plots from your model ?

 

[Yunus9696]

Dear FEA way
Thanks for your kind attention and sorry for being late since my laptop was broken for a couple of weeks.

Here I kindly attache the equivalent plots however I changed my mind about 2D axisymmetric modeling and want to do the simulation in 3D.

However I'm still curious about this problem...

Thanks again.

 

[FEA way]

CMS2 is very low here, pretty much zero. Maybe it's a wrong component ? Also, make sure that contact works properly - check field outputs such as CPRESS, COPEN and CSTATUS.

 

[Yunus9696]

Dear FEA way
Thanks for your advice.
You are right, I was checking the wrong component since the "Abaqus documentation Axisymmetric connection..." example's coordinate system is Rectangular and my model has a Cylindrical coordinate system, so CMS2 has different orientations in these two problems.

Now the CMS2 and CTRQ values are closer together but still not the same.
I'll search for the reasons and share my findings here.

Thanks again,
Best regards,
Yunus.