Learning Forum

[ashishkumar.gupta]

Hi All,

What are the limitations for using Remote Displacement for 3D Ball/Sphere (Solid Deformable)?

What are the alternatives for my case in Transient Structural for smooth simulations and physical Results? 

P.s. I created Remote point first, Then I used Remote Displacement.

[ashishkumar.gupta]

Hello, could someone please help me out here?

[Ashish Khemka]

Hello,

What is the error message you see when using the outer surface of the ball for remote point definition? I am assuming that you are defining contacts as well using the same surface and it might be conflicting with the remote point definition. For the first approach - it seems you are using half-symmetry. What is the reason that you doubt that results might be incorrect?

Regards,

Ashish Khemka

 

[ashishkumar.gupta]

Yes, I am using half symmetry. since I have applied Remote Displacement on the Plane surface of Half ball, I am not able to see any rotation.

I guess the right way would be apply Remote Displacement on the curved surface of the ball.

 

Should I switch to Full Ball Model instead of this Half Ball model?

or the Current Half Ball Model with Remote displacement on the plane surface of Ball is correct?

[Ashish Khemka]

Hi,

Thanks for providing the update. Instead of remote displacement - you may try creating a body-to-ground revolute joint for balls and define a frictional contact between ball and groove. To move the balls apply a joint load for the revolute joint.

Regards,

Ashish Khemka

[ashishkumar.gupta]

 

Hi Ashish, That doesnt support my theory of Contact Mechanics. I am doing 2 way FSI.

 

So only option is Remote Displacement. But The only doubt I have on which surface should the remote displacement be applied? Also, is Half Ball Model fine? 

Please update on this ....thank you!!

 

[Ashish Khemka]

Yes, using the half-ball model with remote displacement (free rotation about z and displacement in z=0) should be fine.

Regards,

Ashish Khemka

[ashishkumar.gupta]

Like Remote Disp. applied on plane surface? then why do I not see any rotaional movement of ball?

[Ashish Khemka]

What are the loads applied? I do not see any rotational load being applied to the ball.

Regards,

Ashish Khemka

[ashishkumar.gupta]

 

The rotational Velocity are applied to Inner Race (+ 1000 RPM) and Outer Race (-658 RPM) …which means Ball can rotate around its own axis in the same position (corr. constraints are applied in form of Remote Disp.)

 

[ashishkumar.gupta]

Hi, could u please confirm?

[Ashish Khemka]

Hi,

Rotational Velocity accounts for the structural effects of a part spinning at a constant rate. Instead of that you can define revolute joints (body to ground) for inner and outer races.

Regards,

Ashish Khemka

[ashishkumar.gupta]

 

so u wanna say Rotational velocity would not give me Rotational Effects or I would not able be able to Rotation? 

Ball Bearing has no Joint effect …I do not think  revolute joints would fit here…..It would not create the contact stresses I want ….

Maybe could you please confirm with ur colleagues? 

 

[ashishkumar.gupta]

 

Also, I have only modeled 40° out of 360° ….which is 1 ball with part of IR and OR…. All bodies are free to deform in XY plane....but No deformation & only rotation around Z axis

My ball bearing should Rotate in Ansys Mechanical (Transient) and this will cause Fluid to move in Ansys Fluent (Transient – Sliding Mesh Tech with Dynamic Mesh) —–> 2 way FSI. [Mech – Fluent -System Coupling ]

So my B.C. should be:

Inner Race:  Revolute Joint (Body to Ground ) +1000 RPM

Outer Race:  Revolute Joint (Body to Ground ) -658 RPM

Ball:  Remote Disp. (Rz= Free) around own axis.

Between Ball and Races, there is fluid (in Ansys Fleunt) which will transfer Pressure forces to Mechanical body …..so contact is kept ‘Frictionless’ as there will never be Metal-Metal conatct

Right?

 

 

[Ashish Khemka]

Hi,

I think with revolute joints for the races and contact between the ball and races you should see the ball rotating. Meanwhile please wait for other members to comment:

 

Regards,

Ashish Khemka

 

 

[ashishkumar.gupta]

Thank you Ashish for the informative description with pictures.

In my Case, Ball and Race are not touching each other. which means Fluid is present and it is modled in Fluent ...so 2 way FSI.

do you think Revolute joint will work?

[Ashish Khemka]

Hi Ashish. let me ask team to comment more on this.

Thanks,

Ashish Khemka

[ashishkumar.gupta]

Hi Ashish, any update from the team??

[ashishkumar.gupta]

ok. I'll wait. Thank you!

Btw The Revolute joint worked for seeing the rotational movements. 

But I hv read 'Revolute joint' does not produce Real contact Mechanics Results. I could not find any tough evidence for that....

[Ashish Khemka]

Hi Ashish,

I have defined revolute joint for races but the ball is free to move between the races due to frictional contact. Movement of ball seems realistic but for races you will need to check if they only rotate or have any other loads as well.

Regards,

Ashish Khemka