Material overhaul

A forum to discuss the implementation of a good Materials system in FreeCAD
knipknap
Posts: 106
Joined: Sat Mar 05, 2022 1:21 pm

Re: Material overhaul

Post by knipknap »

grandcross wrote: Sat Aug 26, 2023 8:22 pm One thing this tool will enable is shared catalogs as web service or database back ends. This will allow data to be shared in ways I don't currently envision. Material models support inheritance, so having "AluminumX" and a derived version "AluminumXFoundInTheScrapHeapOfUnknownOrigin" with the second having measured values, or values derived through experimentation on your lathe/CNC/milling machine, is possible.
Ahh, those features combined sound like a really good solution for the use case! Looking forward to using this!
keithsloan52
Veteran
Posts: 2682
Joined: Mon Feb 27, 2012 5:31 pm

Re: Material overhaul

Post by keithsloan52 »

One thing has come up with Materials in the GDML workbench that may apply to other workbenches or there may already be a better solution.

GDML objects are implemented as Python Features and have materials that are currently implemented as enumerate properties, where the enumerate list is the list of possible materials.

Now the issue is with a facility that introduces a list of extra materials. What I think is needed is that a Python Feature could have a enumerate property that has a document wide list, such that the document wide list could be changed and this would be reflected in all the python feature objects.
Last edited by keithsloan52 on Fri Sep 01, 2023 1:36 am, edited 2 times in total.
User avatar
Shalmeneser
Veteran
Posts: 8168
Joined: Wed Dec 23, 2020 12:04 am
Location: Fr

Re: Material overhaul

Post by Shalmeneser »

@keithsloan52 : problem with your post above :mrgreen:
User avatar
NewJoker
Veteran
Posts: 2331
Joined: Sun Oct 11, 2020 7:49 pm

Re: Material overhaul

Post by NewJoker »

grd wrote: Tue Aug 22, 2023 11:31 pm I can be wrong but I think that the only safe material for a FEM calculation is a metal.
I just want to correct this. With a large number of material models available today, one can analyze pretty much any material with sufficient accuracy (I've even seen a book describing someone's study involving the modeling of a carrot). Of course, some materials are harder to model than others. But rubbers and other polymers can be simulated rather easily using hyperelasticity (and viscoelasticity if needed). There are also special material models for concrete, ceramics, soil and even tissues. Not to mention user-defined material models that can be highly specialized. Composites are common in FEA as well - commercial software usually offers some features for the modeling of laminates.
grd
Posts: 322
Joined: Wed Apr 13, 2022 5:13 am
Location: Eindhoven, The Netherlands

Re: Material overhaul

Post by grd »

NewJoker wrote: Sun Sep 24, 2023 6:31 pm I just want to correct this....
A FEM calculation is always a model. Please keep that in mind.
About Nim. Release 2.0 is just out of the door. Here is Nim in 100 seconds and a Nim package. There are Qt and OCCT packages.
User avatar
NewJoker
Veteran
Posts: 2331
Joined: Sun Oct 11, 2020 7:49 pm

Re: Material overhaul

Post by NewJoker »

grd wrote: Sun Sep 24, 2023 7:25 pm A FEM calculation is always a model. Please keep that in mind.
Sure, there's always some inaccuracy depending on the level of approximation and source of data (proper material modeling requires some physical testing as well). But with good software and correct approach, one can achieve an accuracy of a few percent even for highly complex and nonlinear cases.
Post Reply