how much shoud we trust SolidWorks?

hi.


I am presenting a topic about how much should we trust SolidWorks?


I built a model vehicle three months ago, which is a little complicated and has never appeared in reality, and so far, the COSMOSMotion has kept telling me that this kind of vehicle is NOT able to move.


Should I trust it? I mean, should I believe that this kind of vehicle is unable to move at all in reality? I still doubt that Solidworks fails to make it move just because it is too complicated and then it is NOT able to find the solutions to the dynamical equations. Anyway, SolidWorks is nothing more than a program environment, how shoud it be equal to the reality?


what's your opinion?


Cheers

I would trust the stress analysis and load bearing info that Cosmos gives.
However if you belive you have new technology or are creating locomotion through a unconditional means then i would hope that the theory at least is not disproven by cosmos. Untill it exists and can be tested physically it is just a theory. It cannot be proven either way untill it actually exists. Cosmos is one of the most advanced programs Edit- (One of the more widely used and generally accepted not necesarilly the most advanced)-end edit
in the field of what it does, certainly it can be trusted, however you may have discovered soemthing new.
Edited by: skateboardkid

I would have to contest that Cosmos is one of the most advanced programs... before COSMOS I worked with ANSYS a lot, and I wasn't too impressed with what COSMOS had to offer by comparison. IMHO, of course...


But the results of any analysis softwareshould not be taken at face value. The underlying constraints and boundary conditions must be thoroughly understood. The results should be considered an estimate until verified by testing. Personally, I would never design something that is safety critical solely on theresults of any FEA software. I would need some experimental or legacy data to back it up.


There's also the limitations of the computer running the analysis to consider. If your mesh, for example, is too coarse, then you are diverging fromreality; however, if it is too fine, then it won't solve.

cheers!


In my opion, Solidworks, no matter how advanced it is, it is only a software built with computer languages, and therefore, it can't avoid being formulized.


To make myself clear, i would give an example. To model a car, Solidworks will build an equation for each wheel, then a group of 4 equations is required for a normal car with 4 wheels. And then, based on this, if the car happen to have more than 4 wheels, it will then add one more dimension to the equation group for each extra wheel. No doubt Solidworks can successfully model a normal car and even an 8-wheeled car.


However, if a car gets 8 wheels, and the chassis is cut into two halves each of which has four wheels with it. These two halves are then connected via a joint. Try to ask SolidWorks to model this 8-wheeled car, it could probably say that this car is NOT able to move.


I have not tested this one, but it is similar to the problem i am facing. I just want to make a point that SolidWorks, since it is formulized, it could appear to be stupid in some case. Therefore, if we are testing something new, which has not yet shown up in reality, in SolidWorks, then we are told by it that this kind of machine is not able to work, we should still doubt that.

I mostly agree with SSLaser. You have to understand how FEA works before you can say how accurate any FEA program is.



I have some documents that give examples comparing Cosmosworks to
various "exact" or classical solutions. I'll see if I can find them and
post a link to them here later.

Hi<?:namespace prefix = o ns = "urn:schemas-microsoft-comfficeffice" />
I use SW & CW works in some of my projects, you know i think your dubiety has a good reasons, that i experienced them in some cases, if you try to make a report from your analysisresults in CW; you give a message that it say:
Note:
"Do not base your design decisions solely on the data presented in this report. Use this information in conjunction with experimental data and practical experience. Field testing is mandatory to validate your final design. COSMOSWorks helps you reduce your time-to-market by reducing but not eliminating field tests. "
I

As an engineer, Cosmosworks is a tool to be used with engineering judgment.

The message from Simulation merely indicates that YOU as an engineer are responsible for your designs.

As for this message not appearing in other FEA programs, I can remember learning ALGOR in the 90s and the instructor purposely defining a model that indicated failure and asking us what was wrong here.

Remember any software is a tool and should be used as such.