I think that a split line might work to give you a selectable feature w/o changing model geometry.

I have no Idea how to make cosmos give you cylndrical coords.

Nick
I love materials science!

Nick


Thanks for the tip! That did the trick...

I was able to use the split line command to draw a box onto the surface I'm interested in and it created an edge that was queryable (is that a word?) and I was also able to apply a mesh control to this defined region and increase the mesh only in that area, where I have large stress gradients.

I am using this to simulate readings from a strain gage, so for optimization I need a repeatable dense-mesh region that I can query to get the stress vs position curve. It was easy in Pro Mechanica, and it appears as though it should work in Cosmosworks as well. I'm currently helping my company decide on an FEA package but I wanted to be as un-biased as possible because I was using Mechanica for 5 years, so it's tempting to sway in that direction, but on the other hand, I love learning new software.

Thanks again! If anyone can answer my earlier question regarding using a cylindrical coordinate system that would be much appreciated. Sometimes there is need to place a strain gage on a curved surface, and sometimes the gage is oriented 45 degrees (shear direction) so I would like to be able to query a defined region and graph the data. Usually to get shear stress you query the Sxy, but on a curved surface  with a cylinrical coordinate system you would need to look at the stress in the Z-T  (45 degrees between the upward Z-axis and the tangent to the curved surface)

Easily done in Mechanica.

Have you tried posting in the Solidworks forums under subscription support. Someone from Solidworks may chime in there, sometimes.

Jason

No, it's currently not critical... I'm just evaluating the software for my company for two weeks and I was just trying to replicate EVERYTHING I did from my previous job as far as functionality in the program is concerned...

I most likely will not need to do this stuff here. I used to design a lot of hermetically sealed strain gage based force and torque transducers which in many cases involved placing strain gages on a curved surface in shear...

The odds of this scenario occuring here where I'm dealing mainly with flat plates for pressure transducers is slim. I'm sure the software can handle it if I ever need it.

Let us know what your finally thoughts are. I've always wondered how Cosmos stacks up against other FEA programs out there.

Jason

No problem. Once I get a chance to play with the 2006 version I'll have a more informed opinion.

Hi,
perhaps this thread is already "closed" but I try to give my 2-pence thought about cylindrical system:
1- CosmosWorks, up to 2004 version (the last I really know), did NOT have ANY functionality for cylindrical CSYS. I seem to know it was so also on v.2005, but I know nothing about v.2006.
2- Of course it's annoying, but the transformation from cartesian to cylindrical can be very easily done with Excel
3- You can use cylindrical clipping planes to inquire the "inside" results in a "cylindrical-CSYS fashion". In v.2004, the "probe" didn't feel the inside nodes even if the model was clipped, but most probably this has been modified in v.2005 / v.2006.

Regards

I spoke to the Cosmosworks reps about this when they gave their presentation. I guess you can look at points relative to a cylindrical c-sys if you pre-select an axis before defining the plot. Unfortunately, the axis units do not change to R-theta-Z, they remain ay X-Y-Z but they are actually in cylindrical coordinates...You just have to (I think) follow the right hand rule and call X=Radial, Z=Z and Y=theta.

I told the reps that this is not good and they took note of it. I have actually seen them put in features that I suggested in the past (they do listen to their customers). They get especially irritated when I tell them "Gee, Pro-Mechanica can do it!".