Hi,
I have simulated a contact analysis of a chain in PrePoMax. I have done the analysis with reference to a Journal Paper. I found some difference in the results. Could anyone please help me to fix the issue. I have attached the document link and my model with this.
Thank you
WeTransfer - Send Large Files & Share Photos Online - Up to 2GB Free - WE TRANSFER link for model
Could you specify differences ?
There is no output in the Results Tab.
Do you have problems with the contact ?
Cheers
Boundary conditions are not fully described in the article but it seems that they applied fixed constraint to one end and pulled the other end with a given force. In your case, there are additional BCs applied to whole parts (and thus also regions involved in contact which can be quite risky).
The authors of the article included plasticity while your material model is linear elastic. Also, they performed nonlinear structural analysis and thus likely enabled geometric nonlinearity which is off in your model.
Try refining the mesh, especially in the contact region.
You may also have to adjust the contact settings. Defaults are not always sufficient.
Finally, check this tutorial with a similar case: PrePoMax (CalculiX FEA) - Tutorial 14 - Chain links - YouTube
Thank you for replying.Is it possible to consider plasticity of material in prepomax…
Sure, check this video: PrePoMax (CalculiX FEA) - Tutorial 9 - Elasto-plastic bending of a beam - YouTube
look carefully on contact constants. very often I see that hard contact overestimates stress (by it’s definition) and normally I use linear contact, but you have to estimate your K constant (it is described in calculix documentation) for 5 to 50 times the modulus of elasticity of adjacent materials (if using meters) or 5000 to 50000 if using milimeters
this way i got better results when comparing to analitical results
I think you need to run the analysis again…
What adjustment should I make in contact settings. Could you please help?
Thank you for replying
I have updated my model as per your suggestions. Please check…
WeTransfer - Send Large Files & Share Photos Online - Up to 2GB Free - Model Link
It’s much better now. However, the authors of that article didn’t use perfect plasticity, as you did. They specified the tangent modulus and you should convert it to an additional stress-plastic strain point which is not so straightforward.
You may also have to refine the mesh in the contact area - do a mesh convergence study.
Finally, there is a chance that the contact property will have to be adjusted but Lucas already explained this.
But with the current informations given, the analysis failed
Right, there were some results in this new file so I thought that it runs well but now, after submitting the analysis, I see that there are convergence issues from the beginning. This likely means that initial rigid body motions occur. It’s not easy to eliminate them in such simulations. You would have to apply proper minimal constraints (BCs), just not to whole parts like before - try fixing individual nodes. It would be much easier to utilize the symmetry of the model (like in the YouTube tutorial linked above) but on the other hand, this won’t work for the case of the rotated link if you want to study it next.
Dear PrePoMax Users,
I use the PrePoMax discourse group for the first time (I hoped to solve my problem myself), so do apologize, if I have not done everything correctly writing the topic. After watching several times the lectures on youtube and reading the PrePoMax instruction I’m trying to set up a disc brake simulation, but after several defeats I decided to to start at from the very beginning i.e. to simulate the contact between the disc and pad and right at the beginning I get a strange behavior:
and for LGEOM – the pad penetrates the disc
Maybe a I should add, that I have at the start distance of about 2mm between the disc and the pad – – so no contact from the beginning.
If anyone could help me to solve that problem I would appreciate such help very much ?
Best Regards,
And what does it look like with the deformation set to True scale instead of Automatic ? Make sure that cintact pairs are defined correctly.
This thread on the CalculiX forum can be helpful: How to simulate heat generation in disc brakes in Calculix? - CalculiX
Dear Sir,
I do thank you Sir for the hint very much, indeed I’ve changed the scale to TRUE and it seems completely different – the start gap of 2 mm disappears and no penetration of the disc and deformation of the pad is to be seen – so do hope it is OK.
As I’ve tried to solve the problem myself, I was looking everywhere only not in the scale factor in post-processing mode, so do appreciate the received help very, very much, as it allows me to move forward.
Thank you Sir for the hint regarding the disc brake, that source of information is new for me – I was carefully studied all your tutorials (e.g. the one Rotating Dis) and the one from Mr Matej Borovinsek and also some some for Abaqus e.g.
Abaqus Dynamic Explicit:Disk Brake Analysis:Step by Step - YouTube
Disc brake coupled temp-displacement analysis tutorial III/IV - YouTube
Now I will carefully read and try to implement the one proposed by you Sir and will continue my efforts.
Once more very great and loud - Thank you Sir for the help (and such a quick one).
Best Regards,
Dear PrePoMax Users,
following the advice I’ve download the INP file from the DROPBOX but i think it is not complete and at least can not run it “out of the box”, so I’ve concentrated on my problem:
deactivated the pressure force from the PAD
tried to start the rotation of the discourse (following the youtube tutorial “Rotating disc”)
Something is happening (as I put the animation the colors of the disc are changing), but how I can be sure that the Disc is rotating – is it possible to see the rotation ?
In my simulation the moving PAD is one STEP and the rotating DISC another step – how can I start those two steps simultaneously – should I use the keyword PERTURBATION ?
Any help / advice are most welcome and very appreciated.
Best Regards,
That input file is complete and works but doesn’t fully converge so some modifications should be made but it can be a good starting point.
You can switch between steps and increments (frames) in the results using the drop list and player buttons on the right side of the top toolbar. There you can also start an animation to see the movement. The perturbation parameter is not necessary, the second step uses the state of the model from the end of the first step and that’s how it should work here.
Dear Sir,
I do thank you Sir very much for the quick and such optimistic answer - your suggestions are invaluable for me.
.
My problem is that at least I think to understand the youtube tutorials but have immediately problems when I try to switch to my problem.
E.g. the simulation yesterday and today are based each on a single separate step:
1 step (static step): - Pad axial movement and contact to Disc
1 step (static step): - Disc rotation
And as you Sir said - it looked good as a starting point.
Now I have combined those steps into 1 step (coupled temperature-displacement step), introduced the initial temperature of 20 deg. C, transposed all settings from steps above (switched all stresses OFF and only temperature ON – to spare time) and deactivated the both steps above.
The effect is:
the pad does not move any more
the color of the disc (and the pad) changes homogeneously
the temperature value drops
But what is most surprising for me is the fact, that the pad does not move any more and I do not know if there is a contact or not, and why those temperature changes (and why it is actually dropping).
I do apologize Sir for taking your time but any, suggestion are most warmly welcome
Dear Sir,
sorry, please do not answer my questions above - I’ve just noticed that I’ve forget to set the friction parameter in the contact tab.
I will do this, recalculate,and then kindly ask for your Sir opinion.
Best Regards,
