Strange second dataset

[Jannis Heinemann]

Hello, during calculations (both mixed and thick) I noticed that for each regular dataset, an additional strange dataset appears in the results manager.

These datasets have the same flow condition label, but empty convergence histories and mostly zero values in the sweep window, except for Alpha, friction-drag-related values, Mach, L/D (although CLtot = 0), WallTime, and residuals. No data is shown in the Load Dist. window and they do not appear in the .csv output.

[Jannis Heinemann]

Here is the corresponding .vsp file, it seems i cannot upload the whole folder containing all of the solution data due to size restrictions

[Rob McDonald]

Thanks for posting this.  It is harmless and you can ignore it.  It'll be fixed in the next version.

Your Reynolds number is set to 10119916 which was getting written out with limited precision as 1.01199e7.  Then, when parsing the data after running VSPAERO, the values were compared and found different.  This caused an extra empty Results set to be created.

I doubt you need Re of 1e7 accurate to the nearest digit, so if you are annoyed by this occurrence, you can round your Re to 1.012e7 (or 1e7) and the problem should go away.

Rob

[Jannis Heinemann]

Thanks for your answer. I changed Re from 10119916 to 1.01199e7 (thus matching the input value to the displayed value), changed nothing else and re-ran the simulation. The strange dataset dissapeared, but I noticed, that this had an what I consider large influence on solution values.  CLtot went from  0.58539 to 0.60272 (3% difference relative to initial value), CDtot from 2.9518e-02 to 2.7043e-02 (8,4% difference relative to initial value). I can't imagine that this is due to the negligible change of Re, which suggests that the solution data is manipulated by the parsing error, would you agree? Thank you and your team by the way for the great community support!

[Rob McDonald]

I suspect something else changed between the two analysis runs.

Can you replicate both cases with only changing the Reynolds number?

Rob

[Jannis Heinemann]

I re-ran both cases best of the first case with Re =  1.01199e7 and they were identical, so the difference was indeed elsewhere. Thank you for helping out!