Open mbercx opened 1 year ago
Something like that, I am not a 100% sure on the details either. Here is the example output:
WARNING: integrated charge= 0.00003902, expected= 114.00000000
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine electrons (1):
charge is wrong
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
stopping ...
Just noticed this beautiful copy-pasta:
https://github.com/aiidateam/aiida-quantumespresso/blob/7a303f9e10ff6dd9eae652b4f2d8ad2c482022d6/src/aiida_quantumespresso/calculations/pw.py#L155-L156
That apparently has managed to avoid being spotted for over 4 years, since it was added in https://github.com/aiidateam/aiida-quantumespresso/commit/4fc9e12826ed79ce5fe4c04d79259b8b9a3f82d4. ^^
@sphuber do you remember what "charge is wrong" means, specifically? I'm assuming that QE knows the number of valence electrons, and when the integrated charge density doesn't match this number this error pops up?