cancel
Showing results for 
Search instead for 
Did you mean: 

Contact Issues

Experimenter
Experimenter

Dear All,

 

I resolved the tapered issue with the modify option now I have an even more puzzling problem. I am testing the model to see if the contact I set up is working, turns out only one part of the Geometry is sticking and I have not a clue why, I checked all my regions and the directions but for some reason it does not want to work, can anybody shGeometryGeometryConnectorsConnectorsPlatesPlatesDeformed showing contact for no reasonDeformed showing contact for no reasoned some light about this issue? I have uploaded some pictures of what the part is doing.

 

Thank you.

 

Oscar Escobar

4 REPLIES

Re: Contact Issues

Solution Partner Phenom Solution Partner Phenom
Solution Partner Phenom

Dear Oscar,

It´s clear you have a "penetration" problem caused by the prescribed OFFSET in shell elements, you can investigate in this direction: check in the contact property if the shell element z-offset is included in the contact solution o not (By default, ZOFFSET=0 and shell z-offsets are included. ZOFFSET=1 will prevent them from being included).

 

I don´t like at all to mix OFFSET & surface-to-surface contact, instead I suggest to play always with midsurfacing and use the "clasic" meshing strategies, my favorite one is using another Shell element instead of rigid elements.

 

winLIFE-shell-meshing.png

 

Best regards,

Blas.

Blas Molero Hidalgo, Ingeniero Industrial, Director
IBERISA • 48004 BILBAO (SPAIN)
WEB: http://www.iberisa.com
Blog Femap-NX Nastran: http://iberisa.wordpress.com/

Re: Contact Issues

Phenom
Phenom

Perhaps you should try to reverse one or both the connectors' Master vs Slave, given your centre plate is used for both sets of connectors.  There could be an issue if the same sets of nodes are the targets for two sets of connectors.   Also, I assume you have both positive and negative regions defined for the centre plate?  You probably need both.  If you have more than one glue property, the seach distance must bridge the midsurface-to-midsurface gap for both glue properties.

Lastly, when troubleshooting, use the "Generate Contact Preview file" option on the glue property.  This creates a simple nastran .bdf of the glue/contact interactions which have been found.  You can import that into a new model (File -> Import -> Analysis Model) to see what NX Nastran found as interfaces from the setup of your glue properties, regions and connectors.

Re: Contact Issues

Experimenter
Experimenter
Dear Blas,

Thank you very much for your response they have given me more insight as to how contact works and better practices, I started to try using the method suggested and then the GTAC group who was also looking at this issue had an enlightenment moment and suggested I increased the number of iterations in the property card, when that was done, then the plate came off the other surface the way it is supposed to.

Thank you,

Oscar Escobar

Re: Contact Issues

Experimenter
Experimenter
Dear EndZ,

Thank you very much for your response I did try changing the master slave to begin with and even created a model that did not have offsets I figure maybe that created the issue, the GTAC team was able to figure out the issue, turns out that the default Max Iterations of 20 on the contact property card were not enough to make it converge and that is why it was creating issues when displacing it, I bumped up that Max Iteration number and it was a done deal, the Butt-strap came of the other plate the way it is supposed to.

Thank you.


Oscar Escobar