Thermal Expansion Coefficient Conversion Error

v2.0.0: Thermal expansion coefficients entered in °C units are shown as °F units when imported to a model with imperial units, but the values are not converted.

Can you be more specific about the steps to reproduce the problem?

  • Create a file with mm, ton, s, °C units.
  • import and mesh a simple step file.
  • Create Material-A with the following properties.

  • Open a material library and save the material.
  • Create another file with in, lb, s, °F units.
  • Import and mesh a simple step file.
  • Open the material library and assign Material-A to the file.
  • In the new file, open the material edit dialog, and the properties are as follows.

  • Although the new unit labels are shown. Only the “Zero temperature” has been converted to the new units.

While preparing this post, I remembered a similar bug previously resolved.

I was able to reproduce this bug. Thank you.

1 Like

I think I fixed the problem. In the following days, I will release a new developer version where you can test it out.

Thank you for the quick response @Matej.

Please try the new version 2.0.3 if it is working as it should.

I am still getting some conversion errors, but I have not yet determined exactly what is and what is not working properly. I will do some experimenting over the coming week and try to clearly define what is happening.

Thank you for looking into this.

Please test on the materials that were originally in the library and the ones you add in version 2.0.3. Materials added in previous versions might have problems. To fix those, go to each value field, delete it, and overwrite it with the same value.

I did a few different tests:

  • I created a material with imperial units then opened it in another file with metric units. The conversions were all correct.

  • I created a material with metric units then opened it in another file with imperial units. The conversions were all correct.

  • I opened two previously defined materials from the default material library in both imperial units and metric units. The conversions were all correct.

  • I opened two previously defined materials from my custom material library. Most of the conversions were correct, but the following items were not correct. 1. For one material with temperature dependent properties, temperature for the density was converted incorrectly, and the thermal conductivity coefficients and temperatures were not converted. 2. For another material, yield stress (under plasticity) was not converted.

I deleted the incorrect properties from the “Selected” box of the material editor then added the properties back in and pasted the previously entered values. For the first material, the temperature for the density still was incorrectly converted. For both materials, the first entry for the temperature dependent values was converted, but the other values in the temperature dependent table were not converted.

I did not enter temperature curves for the tests with newly entered materials, so I do not know if the entries below the first one would be converted correctly.

Thank you for testing all the scenarios. In the case of using old materials, there is no solution. The problem was that, previously, when the user entered a value, PrePoMax attached a unit to it, but this unit was not saved in the library. If the user entered a value with the unit, both were saved. So, there is no way of correctly restoring old material values. And this cannot be fixed.

Can you reproduce this behavior also with newly added materials? Adding/removing properties or something else?

Thank you for the update. I will create a new custom material library then recreate and check my materials as needed.

I created new temperature dependent materials in both imperial and metric units and saved them in both the default library and my custom library. When I opened each one in the editor in the other unit system, the first temperature data point for each property was converted, but the properties at additional temperatures were not converted. In other words, for each property, only the first point on the temperature curve was converted.

Thank you for working on this. Unfortunately, for now, our ASME standards list material properties with imperial units, but usually the FEA results must be reported with metric units.

So this is still not working?

It doesn’t appear to be working. As far as I can tell though the only remaining problem is with temperature dependent properties beyond the first entry.

Ok, I will look into it.