Learning Forum

[shrav007]

I'm trying to model a simple Planetary gear Model for stress analysis in Transient Structural. I have three gears Sun, Planet and Internal gear. I have put a Body- Ground Revolute joint for the Sun [ 1 DOF RZ ], a Body- Ground Planar joint for the Planet since it revolves arround the Sun [ 3 DOF  UX UY RZ ] and Body- Ground Fixed joint for the Internal gear as it remains stationary [ 0 DOF]. 

When I ran the simulation, the planet is revolving but it is not revolving in a circular path but going out of path in the XY Plane. Also the Planet gear is not meshing properly with the Sun and Internal Gear. I tried to promote all the joints to remote points so as to use constraint equation. But I have no idea how to use the constraint equation. Can any one tell me how to do this ?

Thank You

[peteroznewman]

You don't mention the planetary carrier arm. That may be what you are missing.

Please insert an image in your reply and label the parts. Is what you call an Internal Gear the same as a Ring gear?

 

[shrav007]

Yes the internal gear is same as the ring gear. I do not have the carrier in the assembly. Here is the image.

Because the computation time takes a lot of time, I decided not to put the carrier.

[peteroznewman]

Put in the carrier which has a revolute joint to ground at its center. The planet gear has a revolute joint to the carrier.

[shrav007]

If I input the joints as you have suggested, then where should i put the joint loads. I have two joint loads. One is the sun revolute joint which has moment and the other is the planet revolute joint which has rotation from 0 to 60 degrees as its load. What I want is that the with a particular torque of sun gear, I want the planet to revolve from 0 to 60 degrees. With this simulation,  I want to see the stresses developed in the planet gear and the Ring gear. And should I or should I not use a PLANAR JOINT ? Also is there a need to use constraint Equation ?

[peteroznewman]

If the input to this is rotation on the sun gear, what is the output?

The output is the carrier rotation, right?  The planet and ring gears are just there to move the carrier.

Why are you solving in Transient Structural?  What is the output you want from the analysis?

If you want to know the stress in the gear teeth, then you don't want rotation, so you would solve in Static Structural. Apply a moment to the sun gear and have the joint load on the carrier revolute joint to ground be a rotation of 0 degrees. In the solution, you will probe the joint to obtain the reaction moment on the carrier revolute joint and you will have the stress in all the teeth generated by the frictional contact.

If you don't care about stress in the teeth and just want to create an animation of the planet gear rotating, then use a Rigid Dynamics analysis. Input a joint velocity on the sun gear revolute joint and watch the system rotate. This will take a lot less time to compute than solving Transient Structural.

[shrav007]

Thank you for answering my question. I was trying to simulate how the stress propagates along the line of contact in Transient Structural. I was also trying to do the same for the Planetary gear System by trying to make the Planet gear revolve around the ring gear. I also wanted to see the Contact Stresses developed along the line of action. But any way, thank you for sharing what you know. I will try what you mentioned.

 

But If you do know how to do it with actual rotation in Transient Structural, please let me know.

[peteroznewman]

Static Structural will give you the pattern of stress in the parts. Transient Structural is less clear because it is full of transients.

If you want to see parts moving, then just make the sun gear have a non zero displacement and Transient Structural will show that motion.

[shrav007]

I tried what you suggested. I put body-body revolute joint with the planet to carrier and body ground revolute joint to the carrier. I put moment on the sun and a rotation on the carrier to rotate from 0 to 60 degrees as joint load. But the carrier is going faster than the planet gear, thus making the carrier to bend and the planet gear to come out of plane. Here are the results

I also tried to put the actual calculated rotational velocities at the three joints. But whenever the joint has a rotational velocity load, the solution is failing. I tried many other combinations of moment at sun, rotation and rotational velocity at the planet and carrier joints. The above one was the only one which converged.

Can you tell me how to simulate the motion.