Discussion Forum

Hi

Indeed if you are beginning, it is confusing to read the 3.5 doc and use v4, and particularly 4.0 as this was really the first "early" release of the new Comsol environment. Even the few tousand of pages of doc was not finished. So my first recommendation is to upgrasde asap to 4.1 and add latest patch, pls be aware hat v4.2 is to come soon, I hope within 1-2 months, but as I'm not with COMSOL I have no inside info, so I'm not sure. Anyhow if you have an active maintenance you will receive the info.

If you come from an engineering environment with ANSYS experience, the first time with COMSOL is often derouting. Here you have a PDE solver tool that applies "physics", but any sort of physics. So you must be prepared to write some more equations, but with the gain that you can postprocess about anything, and link in any physics, even add your private "physics" what you can not do in ANSYS, at least not in the same way.

Torque load is one of the cases where you need to add some equation writing, just as measurement of angles (solid body angles). The two dimensions are related to the curl of the deformations field U. The formula you are referring to is the 3.5 notation of the Rx torque expressed as a vector product of the (x,y,z) in m coordiante of any point considered by your formula with the Tangential stress vector (Tax,Tay,Taz) in Pa. By integrating this over an boundary = surface in 3D you end up with a total moment in m*Pa*m^2=N*m units .

In v4 you have also the new Rigid Body coordiante which givces you access to rottion (but derived via quaternion, hence it's valid also for large deformation rigid body angles, not only for the traditional small deformation angles). To this "boundary" node yopu may add forces or torques directly in N*m. What you need to know is that thes eforces and torques are distributed evenly over the boundary (as an average density value) AND that the selected boundary is made rigid i.e. no deformations (which might overstress your system). "Soft Boundary loads are planned for v4.2, next release.

Another difference, in 3.5 you add a suffix "_smsld" to all variable names to identify the application mode = physics you are using. In v4 you write that as prefix "solid." This is a major difference, plus several variable names has changed. It's not obvious to find rapidly the link (for doc update 4.2 I hope ;) My way is a) to turn on the equation view in V4 see the preferences, and to index the doc pdf files for extensive search functions.

THen if you study and repeat the explanations of V3.5 you should manage to redo it in v4 as is adapting the name changes. Or you could try an import if yopu have the v3.5, but be aware the import was still rather limited in v4.0

Finally, pls understand when you select a boundary and add a formula for a load, you add normally in "Pa" a density, this density is either a constant, or might be a field depending on (x,y,z) or t (or p, T ...) This density is "distributed" onto the mesh elements that are related to the boundary such thatan element at location x1,y1,z1, gets tthe corresponding load as defined at that location and with any other related variable



--
Good luck
Ivar

Ivar! Thank you very much...I'll try that you said.
Monika