3D contact problems: using ANSYS or ABAQUS

[paulliu]

I have chosen ANSYS to solve 3D/2D gear contact problems,
and found it's pretty slow to build the model and obtain
the solutions. Neverthless, the solutions obtained were
pretty accurate.

So I wonder whether ABAQUS can do better. I choose ANSYS
last year because it's cheaper than ABAQUS (7K difference
for a bought-out license plus parasolid translator).

I evaluated COSMOS for such problems, and found that it
cannot be used for solving gear contact problems.

 

[fkmeyers]

Have you looked at the contact that you can solve with NEiNastran? Check out

http://www.nenastran.com/newnoran/animation.php

We use this and love it. We also have ABAQUS and it is not bad either but it is much easier to set up in NEiNastran and the results are very accurate (as good as ABAQUS).

 

[gwolf]

(7K difference
for a bought-out license plus parasolid translator).

You get what you pay for.

 

[GBor]

I would agree that you may want to look at NEiNastran. The greatest advantage (fkmeyers correct me if I'm wrong) is that NEiNastran uses FEMAP for its pre and post processor, and FEMAP can write out an ANSYS file, so if you like the ANSYS processor, but not the pre-processor...

I generally don't think of NEiNastran in the same league with ANSYS and ABAQUS, although for most problems it seems to hold its own. Algor has been used for gear contact...admittedly I can't vouch for the accuracy, but the person that did the contact problem seemed to think it was acceptable. I usually think of Algor, COSMOS, and NEiNastran as being competitors in the mid-range cost, but the COSMOS contact engine isn't, in my opinion, as developed as Algor and NEiNastran.

Garland E. Borowski, PE

 

[paulliu]

Thank you for the replies. I'll try to apply
NEINatran on some contact problems to see if
it's better on this type of problems.

Regarding the prices of FEA software, the higher price
reallly does not mean high quality of software, but
more on the management and distributors. Actually
some of current developers of ANSYS worked for
ABAQUS before.

I still want to hear the comparison of ANSYS and ABAQUS
on contact problems.

 

[feajob]

Hello,

I would like to know your feedback regarding MSC.Marc too. I am using MSC.Marc right now, but I used to work with ABAQUS. I prefer ABAQUS, because they have perfect documentations. I never found the same quality of documentations in the other FE softwares.

AAY

http://www.geocities.com/fea_tek/

 

[cloche]

I'm late on this topic, but...
In my company we use ANSYS almost for everything, it has been benchmarked against prototype results a number of times and we are sure we can obtain extremely accurate results. This is also true for contact problems. I recognize that these problems are very difficult to set up, but the advantage is that you have 100% control on what you are doing.
I once tried an evaluation of ALGOR Multiphysics MES, and I think it's extremely interesting when you have coupled structural + explicit dynamics.
CosmosWorks is another planet and can absolutely not be compared to high-end systems. The built-in contact algorythm gives NO control upon its behaviour.