The above simulation looks very interesting. Is there any tutorial on how to setup this kind of simulation?
2 Likes
Not yet (only for standalone CalculiX) but I plan to record such a tutorial. Here’s my YouTube channel with PrePoMax tutorials:
I always include verification with analytical calculations and that’s what usually delays my tutorials but it shouldn’t be so problematic with gears.
For now, I can give you some advice regarding such a simulation:
- try to do it in 2D (plane stress) if possible - meshing will be much easier
- apply local mesh refinement to the contacting teeth
- define a contact pair for all surfaces of the teeth that may come into contact
- apply rigid body constraint to the hub of each gear
- fix one gear (via this rigid body constraint)
- apply rotation or torque to the other gear (again via the rigid body constraint)
Then you can try to increase the complexity of this simulation but the settings suggested above should be a good start.
2 Likes
Thank you for the instruction. I will certainly try it out.
Besides, I am wondering how to simulate a planetary gear set and to evaluate its strength and efficiency, where the planet gears orbit around the sun gear and engage with the ring gear. I guess dynamic contact and kinematic coupling are needed to constrain the motion of planet gears.
That’s much more tricky and it would be nice to see such simulation done with CalculiX.
Dynamic analysis would be the best choice but a static run (“snapshot” from the gear train’s operation cycle) could also suffice in many cases. I would follow the examples from other FEA software - there are some YouTube videos and research papers that you could use as a reference. Unfortunately, there aren’t many of them since gear analyses usually involve spur gear sets.
1 Like
I would like to request it too. I wanted to use PrePoMax for the analysis of various gear profiles, but my models constantly crashes (especially 2D models!) or gives unconsistent results - even when it’s single gear with fixed hub and force applied to its tip 
Something related to this, Prepomax has a wonderfull Advisor, even better than comercial solutions. Maybe @Matej could write a tutorial or add some features in the program that allow to the non programers to create new Advisors for other kind of analysis, this feature would be awesome because could allow more people to create and expand the Advisor cases.
3 Likes
Hello everybody and thanks for your suggestions FEAnalyst,
I would also like to see that tutorial.
ÂżMay I ask you something else meanwhile? I can wait to try it by myself
-When you say: try to do it in 2D (plane stress) if possible - meshing will be much easier.
ÂżDo you mean 3D + Plane Stress?
¿Is it possible to solve it in 2D? I don’t see how to set up Rigid Body plus contact between shell sides.
-When you say: apply rigid body constraint to the hub of each gear and fix one gear (via this rigid body constraint)
I understand one gear hub is made rigid and rotation is then applied. ÂżWhat about the second gear. Did you apply any a brake, inertia, or friction to get the reaction forces?
Thanks in advance.
What I meant by 2D plane stress analysis is shown in this tutorial: PrePoMax (CalculiX FEA) - Tutorial 16 - Plane stress analysis of a U-shaped member - YouTube
To begin with, you can apply fixed boundary condition to the hub of one gear and assign rotation to the other. Then you could try releasing that fixed hub in one DOF to allow for its rotation. But this can be a bit tricky so I would start with the first approach.
I will try to prepare such a video tutorial soon but analytical verification may take a while.
I have advanced straight forward engaging freely the two gears and aplying some friction but knowing how much to brake the second gear to bring the tensions in line with reality seems the tricky part. I will wait for the tutorial.
Thanks again
Hello @FEAnalyst , I have a question. Shouldn’t it be a 2D plane strain? Because you have a constant load applied in every 2D cut of the gear (if their teeth is straight), which I think configure a plane strain case. If it is modelled in 2D plane stress, I think the results might not be accurate, because the constitutive matrix is different in both cases.
Here are some quotes from the research paper “Error Analysis on Finite Element Modeling of Involute Spur Gears” by J.D. Wang and I.M. Howard:
In general, when a gear is subjected to a transmitted load, the stress conditions near the areas of contact and the tooth root are neither plane stress nor plane strain, but are three-dimensional (3D). However, most previous finite element analysis (FEA) models on standard involute gears reduce the problem to two dimensions and many of them provide acceptable approximations.
For gears subject to contact, fracture, and other situations, it is not known if the plane stress or the plane strain assumption is the most appropriate.
The article also cites 6 papers involving plane stress assumption and 4 involving plane strain assumption.
2 Likes
This will be the more expected tutorial in history. 
Thanks for the enlightment, @FEAnalyst .
Maybe a comparison of both cases is a nice approach to the problem.
1 Like
Hello Dear PrePoMax Users!
I am Peter from Hungary. I am very interested in finite element method and I like gears very much. I do finite element as a hobby, but I have little time for it. I have only done Solidworks Simulation so far.
I am glad I found this forum. I will read a lot. I will have 4 days off. 
Sorry about my English, I’m learning it again.
3 Likes
Cool!. ÂżWould you share the step files of those gears?
ÂżWhy do they seem to loose contact at the upper side?
Oh , I just found you alrready share them on GRAB CAD. Thanks a lot.
1 Like
Thank you! 
Yes, you can find my drawings on the GrabCad. I also have the SolidWorks simulations. I have made parametric tooth profiles for SolidWorks and Creo.
@SzPeter79 Did you compare the stresses obtained in the analysis with the values calculated analytically, using the formulas from the thumbnail of that video ?
I have not checked the results shown in the video.
I made a static linear study of tooth contact in 3 different positions. These results have been verified.
2 Likes
Thank you for the reply.
And you probably used torque as input for those linear studies that were compared with analytical calculations ? I’m asking because with applied angle of rotation or rotational velocity, it seems to be impossible to analytically determine the torque necessary for hand calculations and one would have to take it as a reaction moment from the analysis.
1 Like
