Find unglued volumes

[laserablaatio]

Hi all,
I recently found an error in an ansys model of mine:
parts which were not glued to each other. This seems to
happen quite easily and is not so straightforward to find.

Do you have any tricks to find unglued volumes (which touch each other)? Any scripts?
In principle, one should be able to dig out all the areas which overlap with each other but are not glued together...

Thanks for your comments.

 

[meshparts]

I think it should be much more easy to find out, why those volumes are not glued. Normaly it should be no problem to glue volumes witch share areas. The only idea that comes into my mind why the glue operation should not work, is that the volumes intersect. So check the tolerance BTOL and the dimensions of the volumes.

Regards,
Alex

 

[cbrn]

Hi,
I think the most straightforward way to check if volumes which should be glued are not, is to run a modal analysis and see if / where there are "independent" modes in several parts of the model.

Hope this helps,
regards

 

[doatis]

This macro will go through and find areas that share more than one volume. Very useful for checking the result of a vglue operation:

/nopr
_asel=
_atmp=
cm,_vsl,volume
cm,_abfr,area
cm,_lbfr,line
cm,_kbfr,kp
allsel,below,volume
*get,_vcnt,volume,,count
*get,_amax,area,,num,max
*dim,_asel,,_amax
*dim,_atmp,,_amax
_vv = -1
*do,i,1,_vcnt
_vv = vlnext(_vv)
vsel,s,,,_vv,,,1
*vget,_atmp(1),area,1,asel
*vmask,_atmp(1)
*voper,_asel(1),_atmp(1),add,_asel(1)
cmsel,s,_vsl
*enddo
*vfill,_atmp(1),ramp,-1,0
*voper,_asel(1),_asel(1),add,_atmp(1)
*vput,_asel(1),area,1,asel
cm,acom,area
cmsel,a,_lbfr
cmsel,a,_kbfr

/pnum,line,0
/pnum,area,1
/num,1
/gtype,all,node,0
/gtype,all,elem,0
/gtype,all,keyp,0
/gtype,all,line,1
/gtype,all,area,1
/gtype,all,volu,0
gplot

cmsel,s,_vsl
/go

Enjoy!

 

[laserablaatio]

Thank you very much!!