cosmosexpress stress analysis software for accuracy

[mechwarior]

Hello,
Does anyone have any problems with cosmosexpress
stress analysis software for accuracy.?

I ran a test simple stress analysis on a round solid
rod with applied pressure at 5000 psi on one end and
fully supported at the other end. So the maximum
stress should be the same as applied pressure but the max. stress according to
cosmosexpress was 6800 psi.

So something is not right, I may be missing some setting I don't know about or
the software??

If anyone know the answer to this or a way to tweak it to have accurate result as temp. fix, please let me know

 

[crisb]

Where is the maximum stress? Are boundary conditions at the "fully fixed" end causing the problem? Perhaps this should only be fixed in one direction with just enough restraints in the other directions to prevent rigid body motion?

 

[mechwarior]

The Max. stress showed around the OD of the part.

I'm just wondering how reliable is this software.
It may have limited restraint feature. If you have cosmoexpress, please try this simple problem on yours see if you get the same results.

 

[MSDw]

is it a finite element program , and what types of elements is it using?

 

[GBor]

There have been several threads on this. In general, I think most analysts find it dangerous at best. Upgrade to one of the higher Cosmos packages and you may have something, otherwise, use Cosmos Express for trend and load path information.

Garland E. Borowski, PE
Borowski Engineering & Analytical Services, Inc.

http://www.borowskiengineering.com

Lower Alabama SolidWorks Users Group

http://laswug.borowskiengineering.com

 

[corus]

If you fully restrain one end then you're preventing possion's effect from occuring. I'd therefore expect the maximum to be about 6500 psi. Change the restraint to one that acts in the direction of the force and two others to prevent translation/rotation (or use symmety).

corus

 

[johnhors]

mechwarior - there is nothing wrong with your software, as crisb and corus have correctly pointed out, you have over-restrained your model, a common over-sight by many analysts who are blissfully unaware of the false or misleading results that this generates.

for a more complete explanation see:-

http://www.etm-project.com/Roshaz_Art_1a.htm

 

[ERT]

I agree with GBor.

CosmosExpress is very limited in the type of problems it can analyze accurately. It does not have the Restraint options that are required to analyse practical problems as you cannot fix each direction independently.

I recommend using an FEA software that supports restraints that can control specific directions. CosmosWorks linear static capability is a good starting point.

Of course, any software is dangerous if not used properly.

ERT

http://www.akeng.com

 

[mechwarior]

Okay thanks for the feed back.....