Learning Forum

[mrivero]

Hello everyone, 

I am trying modelling compression of a gasket between two plates caused by bolt pretensions. The assembly consist of two external plates bounded by four bolts (simplified modelling by beam connections), between them there are a gasket and a layer which is thinner than the gasket. The objective is compress the assembly to the gasket's thickness will be equal than the layer's thickness. 

I can not run the simulation because an error message comes up ("The L-2 norm of the residual force overflowed. This may be caused by PRED,ON or birth/dead elements. Please use PRED,OFF and try again.")

I have tried changed the mesh, the contacts, enter the command "PRED,OFF", adjustment... I don't know what to do to solve the problem. 

 

I'd really appreciate any help.

 

Best regards.

 

[Aurojyoti Prusty]

Hi,

Could you please try reducing the contact stiffness which would allow more contact with the elements with larger penetration, if the converegence issue is being caused by contact definitions. You may want to change the numerical damping factor from its default value to see if that helps.

You may refer the following discussions and links 

/forum/forums/topic/why-am-i-seeing-residual-forces-so-huge-here/

/forum/forums/topic/error-static-structural-non-linear-simulation/

https://www.padtinc.com/2012/10/10/overcoming-convergence-difficulties-in-ansys-workbench-mechanical-part-i-using-newton-raphson-residual-information/

https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v232/en/ans_str/Hlp_G_STRUNST.html

I hope this helps

Thank you,

Aurojyoti.

[mrivero]

Hi, Aurojyoti

thanks for your advice. I've tried reducing the normal stiffness (factor = 0,1) and changing the damping factor (value = 1). Unfortunately, it hasn't worked for me. I think that the problem is caused by the gasket because if I supress the body "junta" (which is the gasket one), I manage to solve it. However, if I supress the body "gdl" and I try to solve it with the gasket, the same error message comes up ("The L-2 norm of the residual force overflowed. This may be caused by PRED,ON or birth/dead elements. Please use PRED,OFF and try again."). I don't understand this error, is it a convergence issue caused by contact definitions? 

I don't know how to continue.

I really hope you can help me,

Thanks you.

[peteroznewman]

Did you mesh the gasket body with SOLSH190 elements? There is a special workflow that requires that the gasket body has exactly one element through the thickness. You need to use a Mesh Method of Sweep.  More details are in this discussion.

[mrivero]

Hi, 

now I'm trying to solve it with SOLSH190 elements, thanks for the recommendation. I thought that I had to use INTER195 elements for the gasket. I have a question about the Contact Step Control that I have defined for the contact between gdl and base_sup (top plate): at the beginning of the simulation the contact is near open (there is a initial gap) and lastly this gap will disappear due to the bolt pretensions, but I have to define the contact alive or dead in the contact step control? 

Also, I will let you know if I have new problems with the simulation.

Thanks for your help,

best regards.