contact modeling

[folle76]

Hello everyone,
someone might see to "resolve" the attached exemplifying file. I tried to insert a contact interface, but the solutor (depending on the number of elements) gives me one of the following two errors:

1) elements not sufficiently bound;
2) does not "make contact between the two bodies and there is therefore no tension in one of them.

Thank you all. ah, it is made with wf 5.0. I can post it for 4.0.

 

[folle76]

mumble... place an example that opens with wf 4.0 and here seems to work. undesirable behavior is worse than stably unwanted behavior. Well, if you can give me some tips, or I'll be grateful anyway.

 

[Matteo]

Here I am, I hope to take a look at him in the afternoon! Hi.

 

[Matteo]

Unfortunately with prom wf5 I can not help you, the study where work has not renewed maintenance and we are stopped at wf4. I only modified the definition of contact: for the solutor is much easier to converge by indicating contact surface pairs (in this case only a couple) rather than generically "solid" bodies.

Now I'll do it again with neinastran.

 

[Matteo]

here is the result with neinastran: consider that calculations with contacts are always very complex and that you will meet with "mathematic" (convergence) questions as well as engineering.

 

[Matteo]

ps: I was not too much to check the post processing settings and the measuring units.

 

[folle76]

ps: I was not too much to check the post processing settings and the measuring units.

Thank you. with the Straus7 I took the theoretical manual that illustrated the tricks to improve convergence. with promechanica and neinastran is there no manual? However I tell you that the promechanic postprocessor wf 5.0 doesn't seem much changed from wf 4.0, indeed for nothing. I think I'll make a demo on our real case from some innastran dealer. a photo illustrates the tension study of a hook coupled with the chain (we had to decrease the tension peak at a point because we did not exceed the fatigue tests) (in rhinestone you have to insert variable stiffness links to simulate contact), another photo one of our products. the elaboration of the results in the post processor for us is fundamental. cmq between the purchase of proes, of the new management this year we have already spent a package in software (fundamental, for charity).

 

[Matteo]

Thank you. with the Straus7 I took the theoretical manual that illustrated the tricks to improve convergence. with promechanica and neinastran is there no manual? However I tell you that the promechanic postprocessor wf 5.0 doesn't seem much changed from wf 4.0, indeed for nothing. I think I'll make a demo on our real case from some innastran dealer. a photo illustrates the tension study of a hook coupled with the chain (we had to decrease the tension peak at a point because we did not exceed the fatigue tests) (in rhinestone you have to insert variable stiffness links to simulate contact), another photo one of our products. the elaboration of the results in the post processor for us is fundamental. cmq between the purchase of proes, of the new management this year we have already spent a package in software (fundamental, for charity).

to write via mp, but I do it here for transparency. If you want a direct contact with the team of neinastran I can give you the references of the most qualified people.

ps: I know that hook. Consider that such a problem requires hours and hours of calculation of "mostruous" workstations (regardless of the solutor) :smile:

 

[folle76]

Well, obviously the chain ring is schematized with beam elements...and then cmq I positioned the hook in the "presumption" configuration of balance (almost static simulation). the deformations are minimal and then we were interested in a comparative result (among different geometric features) rather than the absolute correct values. Of course, that image is used for mercenary purposes.

 

[Matteo]

the effect is beautiful in fact! Congratulations!