I want to use Prepomax to simulate the meshing of bevel gears and obtain the simulation results of gear contact stress and spoke plate vibration displacement during the meshing process. The specific working conditions are the driving wheel speed of 15000rpm and the driven wheel load of 2Nm. I have a few questions to ask:
1.Should I use the dynamic module of prepomax and choose explicit or implicit?
2.How should the constant speed of the driving wheel be applied?
3.How to set the contact between the driving wheel and the driven wheel?
I thought that you have this mostly worked out: Bevel gear dynamic simulation
Dynamic is a good idea but it really depends on the case. Implicit should be faster but prone to convergence issues. You don’t have to worry about convergence with explicit but it’s terribly slow for large models (especially in CalculiX).
Constant speed can be simulated as a boundary condition on the reference point. Contact can have default settings for now (unless the stiffness needs to be varied, especially in explicit). Just make sure that all the surfaces that may come into contact are selected for contact pair(s).
Bevel gear dynamic simulationI set both gears as rigid bodies here, and the result is that the stress values at all positions on the gears are the same. Later, I set the driven wheel as a flexible body and the driving wheel continued to be a rigid body. In the simulation results, the stress value of the driven wheel was too high, reaching several thousand or even tens of thousands of MPA. The driven wheel gear seemed to have twisted instead of rotated.
Make sure that it’s free to rotate if you want to simulate their motion. Just release the proper DOF at the rigid body reference point.
1 Like
First of all, this is a very hard problem to solve. In order to understand the problem and since everything is moving at a constant speed, I would first try the static analysis. There you can add the centrifugal force using the appropriate load.
This is the simulation setting I just made (the simulation results cannot be saved in this PMX file, so I used an AVI file to display the simulation results). The results show that the gear rotation amplitude is very small, but the speed I set is 15000rpm, and I don’t know where the problem is
I will have a try，but I donnot know if static analysis simulate the spoke plate vibration displacement during the meshing process.
You set it as an applied moment. It would be better to use angular displacement BC with amplitude. Currently, you have the amplitude selected for a boundary condition with only 0 or free DOFs so it doesn’t work.
I have used angular displacement BC with amplitude.
Yes, but the amplitude values are multiplied by the magnitudes specified in the boundary condition definition. So they can be only multiplied by zero here (UR1 has no BC at all).
I misunderstood the amplitude usage. Is there a range requirement for the value of UR1
Rotational DOFs can take any values in radians but they should be reasonable. It’s better to start with a smaller value and see how it works. And you can use the amplitude to control the rotation (no need for a prescribed moment then).
Thank you for your reply. I will now readjust the settings and start the simulation.Also, I would like to know if prepomax can be set to a constant speed.
CalculiX doesn’t have velocity BCs but you can achieve the same with displacement BCs and (again) amplitudes.
Is there any method to accelerate the calculation speed of prepomax?
Use Pardiso, it’s the fastest solver. Make sure that you have max number of CPUs your machine can use set in PrePoMax. Avoid explicit dynamics or use mass scaling if possible (for quasi-static simulations). Try with coarser meshes first.
Yes, set there the number of cores your processor has.
I set a reference point on the gear spoke to measure vibration displacement, but the results were all 0. Is my method wrong
Does this reference point 1 move or rotate at all ? Only then some of its displacement component outputs will be non-zero.