Good morning,
I am trying to change from running my simulation in Marc or Abaqus to PrePoMax/Calculix.
I think I already know how to correctly write the input file to make it work, since I already achieved making work a couple of them.
However, I have a problem in the convergence of 2D plane stress models for my rack and pinion simulations.
I want to block x,y of pinion and y axis and rotation of the rack.
I apply torque to the pinion and a velocity to the rack.
I wan unable to make to converge this and i would like to know if there is any way to make it work please.
However, I do not want to control my simulation applying a rotation instead of a torque. Indeed, I reached to make it work if I apply a rotation to the pinion… but however the results is very strange since the teeth do not deform as expected…
Hi, I think the BC were not properly applied at the reference points, I modified the file in a prepomax v 2.1.0 file and now is “as expected”. You should check the details, for plane stress the thickness is necessary but maybe this problem is better suited for plane strain, check the displacement BC, I just modified everything so it would work, not to make it physically meaningful. Also the contact surface were redefined as Calc_em mentioned.
I am going to carefully check your comments.
Concerning plane stress or plane strain, I have always done these kind of simulation as plane stress in Abaqus or Marc.
I wanted to also check the .pmx file you shared, but how should I do to properly open the file you uploaded:
You can do the same in ccx. You only need to have the appropriate thickness called out. Don’t be discouraged if it takes you much longer to find the right contact parameters for your simulations in ccx; this is one of the disadvantages with respect to Abaqus or Marc. Marc is so powerful for nonlinear analysis and handling contact, even better than Abauqs, in my experience.
hello, yes, should work the same way just changing enforced rotation by a moment. However I tried and got a wrong result (and non convergence) as the rigid body seems to be distorting the elements. but I think is due to linear element formulation (please read CCX manual on that point, not as good as abaqus by far) so maybe you could update your model to parabolic elements and then should work.
PS: in the previous model there was a mistake, since T3 translation should not be set zero as is by default in a plane stress model, change it setting R3 to zero, then the rack/gear will not rotate.
Thank you for you messege @jbr !! I’ve always employed marc ,ad really think it is very powerful, but I am sure ccx can handle it too, that is why I want to try and achieve it. However, I see that perhaps there is knowledge that I do not have…
Thanks for the help provided!
Excuse me I do not understand the change you are talking about at this point, could you please send me an image of th point you are suggesting to change?
I mean the boundary condition for the rack/gear at RP-2, should be U1=value,U2=0,UR3=0. Before it was U1=value, U2=0, U3=0.
Results will not change very much, but is not consistent.
