[Wien] The appropriate number of plane waves

[Stefaan Cottenier]

>      I recently calculate the structure of super cell. The number of 
> generated plane waves is over 150,000 so the compute cost much time 
> (over 12 hours per cycle in SCF). And the results are not convergence.
> 
>      Could you give some advice to me. Thanks.

Test convergence in a small cell (similarly for the k-mesh), then do the
supercell immediately with the correct number of plane waves (and with a
properly 'reduced' k-mesh).

Stefaan