Learning Forum

[ajaynsv]

Hello everyone,

I am a student of mechanical engineering and I am working on bearing simulations. I am doing transient analysis of multibody system i.e bearing. I am using Augmented Lagrange formulation with normal stiffness as 0.1.

Still, the solution is not converging. 

[peteroznewman]

1) ANSYS staff are not permitted to download attachments. Please insert images directly into your reply so they can see them.

2) What is the goal of your simulation?

[peteroznewman]

Does one of the rollers have a crack in the geometry?

It will be a very long simulation if you have to spin the bearing up from zero velocity.

I would recommend removing all revolute joints, adding a cage, and relying on contact only.

[ajaynsv]

Thanks, peter

No, none of the rollers has crack geometry in it. I will add it later when the simulation will run without a crack.

can you elaborate what do you actually mean by zero velocity? As the bearing is running at a particular speed so I have taken two steps and in tabular data, I have taken 0 t0 300 rad/s and in 2nd step 300 to 300.

Ok, I will with that.

 

 

[ajaynsv]

sir,

I am not getting the response as I want. The effect due to rotation is not coming in the simulation. I am attaching the file(*.wbpj). Kindly look into this and help me.

https://drive.google.com/open?id=10X-_GiVYRLQcTZBPexbbZucSthrmscZK

Thanks

[peteroznewman]

The .wbpj file on its own is useless.

Please use File > Archive to create a .wbpz file. That is how you share ANSYS models.

[ajaynsv]

https://drive.google.com/open?id=1_tArm0YObRAfv8BHCF5Oj2hwhEPbdadc

[ajaynsv]

Sir, did you find any solution for my problem?

[peteroznewman]

You used an Inertial Load called Rotational Velocity. You don't understand what this does. It applies a radial acceleration load to the mass in the model from the axis that is proportional to the radius and the rotational velocity squared. It does not rotate a body.

You can apply a moment to rotationally accelerate the mass of the inner ring.

There is no cage to keep the rollers spaced evenly.

I simplified the contacts, I reduced the bearing load temporarily to see if that helps it converge. I improved the mesh.

This model converges for a few steps, so I am uploading it for you to look at.

[ajaynsv]

Thank you, sir, I do not find any attachment in your reply.

[peteroznewman]

I let the attached model run for nearly 12 hours on 15 cores on my computer.  It computed 799 iterations to generate 59 converged steps to cover a simulated time of 5.4 milliseconds. In the attached zip file, you can see a movie of the rollers falling due to the gravity load I added.  There is not enough moment on the inner ring to get that to rotate.

Considering that it will take even longer to compute once you add the cage and a crack, I suggest that is is not a fruitful method of studying a crack in a roller or race.

[ajaynsv]

Thank you, sir, for your effort. I have found a similar problem please have a look.

/forum/forums/topic/bearing-vibration-analysis-in-explicit-dynamics/?order=all#comment-545b0b1d-b5a3-4c2d-99e0-a9c6004bf93c

just the difference is I am using roller bearing and the defect is present on the ball instead of the outer race. I think with some simplification it can be done like giving independent motion to each roller with the same angular frequency etc.

Please help me to do this sir.

[ajaynsv]

Sir, please help me with it I am completely blocked and I don't know how to proceed further.

[ajaynsv]

Hello Sir,

Could you please have a look at the attachment and tell me the possible reason for this.

https://iitk-my.sharepoint.com/g/personal/ajayky_iitk_ac_in/EVF5UhUAAg9BiIYnAeIfwH4BLtIPSBEC-W86-CYQQrFt8g?e=HtaFN5

 

[ajaynsv]

https://iitk-my.sharepoint.com/g/personal/ajayky_iitk_ac_in/EVF5UhUAAg9BiIYnAeIfwH4BLtIPSBEC-W86-CYQQrFt8g?e=eLX5dc

[ajaynsv]

https://iitk-my.sharepoint.com/ : v : / g/personal/ajayky_iitk_ac_in/EVF5UhUAAg9BiIYnAeIfwH4BLtIPSBEC-W86-CYQQrFt8g?e=eLX5dc

Sir, change the link after .com/     to : v : /  without any space

[Ajiket_Patil]

Hi, Ajay..did you complete the transient bearing analysis ?.