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.

Which variable(s) denote fluid shear on a surface in 3D

Please login with a confirmed email address before reporting spam

Hi!

I think you can define shear in 2D as: spf.T_stressx*tx + spf.T_stressy*ty

However, this doesn't seem to work in 3D: spf.T_stressx*t1x + spf.T_stressy*t1y + spf.T_stressz*t1z

If I use that in 3D, the resulting model is pressure-sensitive, leading me to believe that I am not getting just tangent components. Using t2x, t2y, and t2z does't work either (what the difference between these?).

Anyone know the proper description of shear forces in 3D?

Thank you!

5 Replies Last Post Jun 21, 2017, 2:44 p.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 7 years ago Jun 20, 2017, 3:49 p.m. EDT
Updated: 7 years ago Jun 20, 2017, 10:32 p.m. EDT
Just an FYI in case anyone else needs to know this: It appears that in X and Y you use t1, but in Z you must use t2. The documentation doesn't really describe the difference between the two, just saying that t1 exists on edges and both t1 and t2 exist on surfaces.

So, the final equation expressions that seem to provide shear stress along all axis:

spf.T_stressx*t1x
spf.T_stressy*t1y
spf.T_stressz*t2z

However, I'm far from sure this is correct in theory -- it just seems to work for my system.
Just an FYI in case anyone else needs to know this: It appears that in X and Y you use t1, but in Z you must use t2. The documentation doesn't really describe the difference between the two, just saying that t1 exists on edges and both t1 and t2 exist on surfaces. So, the final equation expressions that seem to provide shear stress along all axis: spf.T_stressx*t1x spf.T_stressy*t1y spf.T_stressz*t2z However, I'm far from sure this is correct in theory -- it just seems to work for my system.

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 7 years ago Jun 20, 2017, 4:49 p.m. EDT
Hi James,
In 3D, on an edge t1 is the edge's tangential vector and on surfaces t1 and t2 are the surface's tangential vectors. Their x,y, and z components are t1x, t1y, t1z and t2x, t2y and t2z respectively. See Reference Manual, version 5.3, section entitled "Predefined and Built- In Variables", and particularly page 261.
Best,
Jeff
Hi James, In 3D, on an edge t1 is the edge's tangential vector and on surfaces t1 and t2 are the surface's tangential vectors. Their x,y, and z components are t1x, t1y, t1z and t2x, t2y and t2z respectively. See Reference Manual, version 5.3, section entitled "Predefined and Built- In Variables", and particularly page 261. Best, Jeff

Please login with a confirmed email address before reporting spam

Posted: 7 years ago Jun 20, 2017, 9:36 p.m. EDT
Thanks for the information, but I am not clear on what the first tangent and second tangent are. Are they aligned with particular axis?
Thanks for the information, but I am not clear on what the first tangent and second tangent are. Are they aligned with particular axis?

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 7 years ago Jun 21, 2017, 9:32 a.m. EDT
Hello James,

t1 and t2 are literally two vectors that are tangential to the surface at the point in question. How they are computed from the parametrization of the surface is explained on the page I listed earlier. In the general case, they are not aligned with a particular axis, which is why they will usually have three non-zero components along the x, y and z axes, denoted, for t1, t1x, t1y and t1z (and same idea for t2) as mentioned above. You can use an Arrow Surface plot to visualize these vectors using those component names, see attached screenshot showing t1 in red and t2 in green for a cone.
Perpendicular to both t1 and t2 is n, the outward-pointing normal vector, with components nx, ny and nz.
Should you need further assistance, please contact COMSOL's support team at support@comsol.com , as this is stretching to the limits of my knowledge in that field.
Best,
Jeff
Hello James, t1 and t2 are literally two vectors that are tangential to the surface at the point in question. How they are computed from the parametrization of the surface is explained on the page I listed earlier. In the general case, they are not aligned with a particular axis, which is why they will usually have three non-zero components along the x, y and z axes, denoted, for t1, t1x, t1y and t1z (and same idea for t2) as mentioned above. You can use an Arrow Surface plot to visualize these vectors using those component names, see attached screenshot showing t1 in red and t2 in green for a cone. Perpendicular to both t1 and t2 is n, the outward-pointing normal vector, with components nx, ny and nz. Should you need further assistance, please contact COMSOL's support team at support@comsol.com , as this is stretching to the limits of my knowledge in that field. Best, Jeff


Please login with a confirmed email address before reporting spam

Posted: 7 years ago Jun 21, 2017, 2:44 p.m. EDT
Thank you Jeff. I'm new to COMSOL and it's hard for me to know where I am not understanding the physics, versus where I am not understanding COMSOL's terminology and variable definitions. Very frustrating initially, but I guess any program this complex is like that. I appreciate the help!
Thank you Jeff. I'm new to COMSOL and it's hard for me to know where I am not understanding the physics, versus where I am not understanding COMSOL's terminology and variable definitions. Very frustrating initially, but I guess any program this complex is like that. I appreciate the help!

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.