Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Fluid Shear Stress in Thin Film Fluid Module

Please login with a confirmed email address before reporting spam

Hello,

I'm currently working with the journal bearing model (Comsol 4.2) from the model library and I'm trying to calculate the shear stress within the fluid film with is an essential for the torque calculation. Unfortunatly the shear stress is not one of the output values of this module.

I tried the surface integration and using the simple formula:
tau = eta*du/dh
with dh = tffs.dh
and eta = constant value

I used du = d(tffs.uy,y)+d(tffs.ux,x)

Unfortunately the results don't seem logical. I believe my expression for du or my usage of the integration (under results, derived values) is wrong.

Please help!

Thanks,
Toan

3 Replies Last Post 2012年5月18日 GMT-4 14:31
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年5月17日 GMT-4 15:25
Hi

I'm not by my WS tonight ;) so I cannot check, but I do believe you have uxx and uyy predefined in COMSOL, turn on the equation view and check yourself (the list is unsorted xyou need to go through all values)

furthermore, if you use these higher order derivatives, it's worth to use a higher order discretization 3rd or 4th too.

Only situation where it might not work is with "thin films" as these are simulated by quations and not mesh elements. Furthermore do not forget to have a few mesh elements along the direction of integration, otherwise the FEM method cannot really define a higher order derivative with reasonable precision

--
Good luck
Ivar
Hi I'm not by my WS tonight ;) so I cannot check, but I do believe you have uxx and uyy predefined in COMSOL, turn on the equation view and check yourself (the list is unsorted xyou need to go through all values) furthermore, if you use these higher order derivatives, it's worth to use a higher order discretization 3rd or 4th too. Only situation where it might not work is with "thin films" as these are simulated by quations and not mesh elements. Furthermore do not forget to have a few mesh elements along the direction of integration, otherwise the FEM method cannot really define a higher order derivative with reasonable precision -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年5月18日 GMT-4 03:28
Thank you!

I've checked the equation view but did not find uxx and uyy but I found tffs.U which is a lot easier to work with. I have added the real journal bearing geometry and a bunch of boundary conditions, etc. Nevertheless the calculated values do not agree at all with my test rig experiments. Next step would be checking if the real and calculated pressure agree.

Fluid film shear stress seems to be harder to calculated as I expected :)

Anyway thanks again, and if you have another idea, I'm more than happy to try it.
Toan
Thank you! I've checked the equation view but did not find uxx and uyy but I found tffs.U which is a lot easier to work with. I have added the real journal bearing geometry and a bunch of boundary conditions, etc. Nevertheless the calculated values do not agree at all with my test rig experiments. Next step would be checking if the real and calculated pressure agree. Fluid film shear stress seems to be harder to calculated as I expected :) Anyway thanks again, and if you have another idea, I'm more than happy to try it. Toan

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年5月18日 GMT-4 14:31
Hi

you are right in CFD there is no 2nd deriative of u, as us is already a velocity (thats OK for solid where u is a displacement)
and in tffs you are in a thin film physics, only solving for "p" and you impose _u_ the velocity of the "thin domain boundaries. Ths means that your shear stress can only be obtained by equations, fitting into the equations used for the thin film bearing.
Normally I would expect something like a parabolic film velocity (at rest), with close to non-slip "0" velocity at theach contact surface. Knowing the average thickness "h" and the relative velocity of one boundary w.r.t. the other one should be able to get an approximation for the velocity profile, and then derive the sher from the gradients of this profile. Or have I missed something ?

I would propose to dig further into the docs referred to in the COMSOL user guide for TTFS

--
Good luck
Ivar
Hi you are right in CFD there is no 2nd deriative of u, as us is already a velocity (thats OK for solid where u is a displacement) and in tffs you are in a thin film physics, only solving for "p" and you impose _u_ the velocity of the "thin domain boundaries. Ths means that your shear stress can only be obtained by equations, fitting into the equations used for the thin film bearing. Normally I would expect something like a parabolic film velocity (at rest), with close to non-slip "0" velocity at theach contact surface. Knowing the average thickness "h" and the relative velocity of one boundary w.r.t. the other one should be able to get an approximation for the velocity profile, and then derive the sher from the gradients of this profile. Or have I missed something ? I would propose to dig further into the docs referred to in the COMSOL user guide for TTFS -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.