[QE-users] Reconstruction problem in hp.x

[Vahid Askarpour]

Dear QE Community,

I am running Hubbard U on a relaxed slab of 96 atoms with a 25Angstrom vacuum. I generated all the response function matrices (using sum_pertq) but when using compute_hp to collect all the chi0 and chi matrices, I get the following error:

   Many lines similar to the ones below

    Missing chi element for: na=   288  nb=    89  dist=  21.669460


     Missing chi element for: na=   288  nb=    91  dist=  19.080347


     Missing chi element for: na=   288  nb=   161  dist=   7.914550


     Missing chi element for: na=   288  nb=   163  dist=  18.450123


     Missing chi element for: na=   288  nb=   233  dist=  12.788988


     Missing chi element for: na=   288  nb=   235  dist=   7.619240

Possible solutions:
     1. Relax better the structure (in order to have more accurate inter-atomic distances)
     2. Increase the value of the parameter eps_dist in PW/src/ldaU.f90,
        then recompile the pw.x and hp.x codes, and re-run the HP postprocessing step
        by setting compute_hp=.true. in the HP input.

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine reconstruct_full_chi (1):
     Reconstruction problem: some chi were not found
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

The structure is relaxed and the forces are quite small with zero stress on the cell.

I increased eps_dist to 0.1 and still get the above error. 

Any suggestion as to how to avoid this error is deeply appreciated.

Thank you,
Vahid


Vahid Askarpour
Department of physics and atmospheric science
Dalhousie University
Halifax, NS
CANADA