Non-zero vacuum charge in v0.95

[foobar]

Hi,
using v0.95, i get non-zero vacuum charge for a neutral system computed with quantum espresso trying US/NC/PAW PBE potentials.
the cube file was created using pp.x. While i first thought that it was a problem with the fft grid, it turned out, that the old version of bader (0.28) gives the correct result. BTW, the number of electrons was correct in both versions.

[graeme]

Please check v0.95a which has a bug fix related to the vacuum change density.

[foobar]

v0.95a gives also wrong results.

[graeme]

Please post the charge density files and try to describe the error as clearly as you can.

If you are saying that the problem is that the program gives you a non-zero vacuum charge density, then this might not be a problem. By default, the analysis will consider any cells with a charge density less than 0.001 e/Ang^3 into the vacuum region. This is useful for the analysis of gas phase species. If you would like to analyze all cells, you can use the "-vac off" setting or reduce the charge density value that is considered vacuum: "-vac 1e-5" , for example.

[foobar]

Thank you very much for your reply. That was problem, in my case this led to a charge of roughly two e.

[graeme]

Ok, great. I should probably lower the default vacuum charge or leave this option off by default.