Slow to converge, or failing?

As I mentioned above it only shows 1,0 thru 1,8 no sign of step 2.

Do I have to do something special to make sure results get saved??

Can you share image of cvg file.
From the Temp directory in the installation directory.
And frd file size

Think I’ve finally got it working - have to press Results button on Monitor Window or stuff gets lost.

Thanks for help everyone - I’m sure this must all seem obvious to those in the know.

:slight_smile:
The Temp directory holds the frd results file which can be opened directly and independently.

Arghhh… my new attempt (still only 2 parts) won’t even converge on the initial pre-load on the bolts.

And it seemed so close after so much help…

Finally manged to get a convergence - BUT the stress plot doesn’t seem to reflect the pressure loads applied on the flow channels.

Surely a uniform pressure should give surface stress equal to the pressure? - OK that would be normal stress - is there some way to view that? (actually 14.26MPa isn’t bad for 15MPa applied - there’ll be some bending no doubt).

Am I missing something else. I’m finding it rather unwieldy to manipulate the model and the section view - is there a simple way to get section view normal to a principal axis?

How do I probe for stress on a face?

How do I view residual clamping pressure between plates (to determine if the process fluid is likely to leak out)? I’d have thought S33, but it doesn’t look promising.

Tools → Query → Surface (there’s also a toolbar button for Query) but I usually use the Vertex/Node mode instead to probe more precisely.

CPRESS might be what you are looking for but you may have to request it in field output if you don’t have it already.

If you have contact then CPRESS and COPEN contour and set range +1 to -1 with 2 colours

Did you right-click the section view icon?

yes, this allows you to define a section and move it along your vector

If you’re iterating to find pretension and ensure process fluid is not leaking, then it can be quicker to use tied constraint and look at S33 (in your case) to get the setup close before doing the lengthier analysis with contact. This will converge quickly and tell you if you’re miles away as a first pass ot if you’ve got lots of different bolting placement solutions. I may have already mentioned this in the thread, apologies if I’m repeating.

I saw the control for section view - I just don’t understand how I’m supposed to use it to define direction. What is the ā€˜plane point’ ?

I have a contact in the model - but no contact shows up in the results tree… (Ah - so I have to specifically add that before running the analysis - not obvious).

Is there a list of names/abbreviations ? I searched both PrePoMax and Calculix documents for COPEN - no result. I can guess some, but would prefer to be sure.

It’s just a way of defining a plane by specifying a point and a normal direction for the plane.

https://www.dhondt.de/ccx_2.22.pdf#subsection.6.13
https://www.dhondt.de/ccx_2.22.pdf#subsection.7.20

COPEN is part of CDIS - you request the latter to get the former (and CSLIP for tangential displacement). Same in Abaqus.

The section view is defined by a plane. The plane is defined by a direction - normal and a point. The point is not very important if you use the slider to move the section view plane.

?
image

From the calculix manual ccx_2.22.pdf (dhondt.de)