Re: [Wien] Problems of calculating the effective U and setting shift during the supercell procedure

Please note that this procedure is limited and gives only some estimate of U (usually a fairly large one). NiO: No, you don't need a shift. This shift is for surfaces or when generating heterostructures. Fe3Al: Why do you have 12 atoms ? Is this not an FCC structure and one f.u. is enough.

Re: [Wien] HF with HDLO?

Hi, No, HDLOs do not work with the hf module. You can use a HELO (high-energy LO) instead, but be careful with the setting of its energy parameter. I'll mention it in the UG and put a stop in the scripts. Best regards Peter Blaha Am 16.11.2022 um 11:48 schrieb Mikhail Nestoklon via Wien:

Re: [Wien] Semicore band ranges too large

For small C spheres, the setting of an LO for C-2s may lead to problems, in particular when one is not yet self-consistent. If you set iqtlsave to zero for some time, it is fine, but at the end of the scf you should switch it back and check if the error is still present. Also check for qtl-b