Closed RemDelaporteMathurin closed 7 months ago
Hi @RemDelaporteMathurin,
Could you uncomment the lines 97 and 199 in the input file (https://github.com/idaholab/TMAP8/blob/devel/test/tests/val-2b/val-2b.i) and comment out 98 and 198 and rerun? We had used different parameters for generating the figure used in documentation; those parameters allow getting a more accurate solution but are not ideal for testing purpose, so those parameters are commented out in the input file. Once you reactivate those parameters you should be able to reproduce the figure shown in documentation.
I haven't run TMAP8 yet I just plotted the scripts available in the repo. I understand the differences for testing purposes.
On a side note, I noticed that there was an extra 15s for the high pressure stage:
These extra 15s don't appear in the temperature expression: https://github.com/idaholab/TMAP8/blob/894c92aac2961569ad4b3cb6d52aaaa202d1407e/test/tests/val-2b/val-2b.i#L159
Running this case with FESTIM I realised that these extra 15s were needed to obtain the results shown on the Figure in the docs. This should be clearly stated in the description of the test case otherwise people may struggle to reproduce the results (unless they go and dig in the input files like I did 😄 )
I will add that information in the documentation. Thanks for your suggestion.
@singhgp4321 as discussed I ran the val-2b.i script by uncommenting/commenting the lines you suggested.
Unfortunately it didn't go through as it diverged at time step 362 (time 9446.46s). It cut the stepsize several times after eventually being killed because the stepsize was too small.
Any idea?
Could you try decreasing the penalty value slightly? Instead of 0.09 may be try 0.085 or 0.08. Ideally you would want the penalty value to be high for accuracy but low enough for converging.
Using 0.08 seems to have done the trick
Closing this issue as it seems you have successfully reproduced the solution described in the documentation.
Closing this issue as it seems you have successfully reproduced the solution described in the documentation.
I haven't checked quantitatively, however I would suggest updating the documentation to avoid any confusion as it is a bit misleading as it is.
Hi all!
Been having a look at the V&V recently and wanted to plot the results in the val 2b folder.
In the documentation, the following figure is displayed:
When trying to plot the data in the val 2b folder (stored in the csv files, either this one or that one), the line doesn't seem to match (see below)
To obtain this, I scaled the experimental data csv by 1e15 and the TMAP8 data by 2*1e20 as done in your plotting script.
which one is correct (ie. produced by the val-2b.i script)?