Eigenfrequency Z88 solver not supported by FC

[uwestoehr]

I had a careful look how I can use FC with Z88 but cannot find a way to start an eigenfrequency calculation.

In the Z88 software, I can choose between these 4 analysis types:
- linear mechanical
- non-linear mechanical
- Free vibration (which is eigenfrequency)
- steady state thermal

In the FC preferences (I use FC 0.18.15748) I tried both
- first the z88r.exe from the Z88OS V15 package
- second the z88r.exe form the Z88 Aurora V4 software

However, when I input an Z88 solver in FC, I can only use the "Analysis Type" static.

So how can I access Z88's Free vibration feature in FC? Or did I stumble over a bug in FC or is this not yet supported?

[bernd]

The short answer:
It is just not supported.

The long answer:
Some years ago I thought FreeCAD FEM should be able to have more solvers than CalculiX. The best way to do this was just to add another solver IMHO. Since the only solver I knew at that time was Z88, I choose Z88 for this. What I did was implementing the very basic functionality for Z88 and separeated all solver code. Thus for Z88 only constraind fixed, constraint force as well as only one material and so on are supported. Mainly all the stuff what we had in FEM when I came to FEM. Furthermore, result reading is only supported for displacements. Actually I do not know someone using Z88 solver in FreeCAD regulary. I also posted on Z88 forum but noone had any interest. For me it seams Z88 is not widely used outside Bayreuth university.

Feel free to extend Z88 support in FreeCAD ...

hope that helps. Cheers bernd

[uwestoehr]

Feel free to extend Z88 support in FreeCAD ...

I understand. However, I am too new in FC to send sensible patches.

Actually I do not know someone using Z88 solver in FreeCAD regulary. I also posted on Z88 forum but noone had any interest. For me it seams Z88 is not widely used outside Bayreuth university.

This is interesting because I heard about Z88 from companies. Especially the Z88 Arion V2 topology solver the presented on last years' Hannover fair had an impact in companies.
I am an industry guy and to me it appears the opposite: Engineers know and use Z88 while they don't use FC. For me it was the same since FC 0.16 was too unstable for industrial application. But now I am really amazed how stable FC is. But I am the first one from my colleagues investing time to learn FC. In my real life most people use Solidworks or AutoCAD and since they don't know FC, they use these programs for the different simulation tools.

[bernd]

Actually I do not know someone using Z88 solver in FreeCAD regulary. I also posted on Z88 forum but noone had any interest. For me it seams Z88 is not widely used outside Bayreuth university.

This is interesting because I heard about Z88 from companies. Especially the Z88 Arion V2 topology solver the presented on last years' Hannover fair had an impact in companies.
I am an industry guy and to me it appears the opposite: Engineers know and use Z88 while they don't use FC. For me it was the same since FC 0.16 was too unstable for industrial application. But now I am really amazed how stable FC is. But I am the first one from my colleagues investing time to learn FC. In my real life most people use Solidworks or AutoCAD and since they don't know FC, they use these programs for the different simulation tools.

I do not know about companiees. It is just my feeling if I access the Z88 forum. There is just nearly no traffic at all. In conjunction on CalculiX mailing list are only rare posts as well, but one really get the feeling this software is used !

In the regard of FreeCAD, yes FreeCAD is geting better fast, really fast. The funny thing is at work I do use FreeCAD FEM from time to time but quite seldom. I mostly use FreeCAD for other stuff at work. I slided in FEM development some years ago and just never stoped sending PullRequest ...

[Jee-Bee]

I slided in FEM development some years ago and just never stoped sending PullRequest ...

And i'm happy about that

[thschrader]

Engineers know and use Z88 while they don't use FC.

No way sir...
Running a FEM-analysis on FC is 10-times faster then using Z88.
Your Eigenfrequency problem: calculix gives very good results (compared to my test
problem: single beam fixed at one side).
About Z88 Arion: brillant idea from the developer, but the
result cant be used for CAD. You can try that by your own.
Design a model with FC, gmesh it in FC (1. order elements!), export mesh to Arion, run
optimization. You will get an stl wich you cant use for CAD.
Ever tried this?
Z88 as opensource software is not industrial standard. My opinion.
But I dont know there private policy.

@Bernd: keep on running.
Thomas

[Jee-Bee]

according to https://en.wikipedia.org/wiki/Z88_FEM_software Z88 is specially for small and medium size companies... but there are also a few big names in the small list in wikipedia.
but i guess that that kind of companies use multiple solvers to compare the results.

[thschrader]

I also posted on Z88 forum but noone had any interest. For me it seams Z88 is not widely used outside Bayreuth university.

OK, lets remove Z88.

[kkremitzki]

I also posted on Z88 forum but noone had any interest. For me it seams Z88 is not widely used outside Bayreuth university.

OK, lets remove Z88.

That may be premature, Z88 is definitely not unused, it may just need more attention for the integration. For reference take a look at the Debian popcon (popularity contest) data for z88 and calculix-ccx:
https://qa.debian.org/popcon.php?package=z88
https://qa.debian.org/popcon.php?package=calculix-ccx

From the looks of it z88 has about 1/4th the popularity of calculix.

[bernd]

IMHO everybody should use what best fits for him.

Might be Prof. Rieg has some connections in the economy and gives direct support and thus they gone use his software. Thus we do not get noticed about this.

In the regard of debian. I do not belive people really use debian packages of z88 even if they gone install them These packages are over 5 years old, may be even older. In debian is still version 13 of z88os.