[Wien] FERMI5 - Error

[John Appleton]

Dear Prof. Blaha,
  I wish to seek your help.
  My calculations keep crashing at lapw2 for 
  TEMP or TEMPS or GAUSS option. This 
  happens in the first SCF step. 
   
  1. Checking uplapw2.error the following message is printed:
   
   'FERMI5' -  emax too low in vector-file
**  testerror: Error in Parallel LAPW2
   
  2. Increasing emax in case.in1 from 2.0 to 10.0
      the problem still persists. I reduced the mixing factor
      to 0.01 and the problem is still there. 
      
   
  3. Checking case.output2up, the following is printed at the end:
   
  TEMP.-SMEARING WITH    0.00500 Ry
    sum over weight(k) (after  renom.) =  NaN
  emin =   -3.06067  emax =    2.49843
   
  4. I went into line 166 of fermi5.f in $WIENROOT/SRC_lapw2 and told print 
  enorm, sumw, fact=enorm/sumw:
    write(*,*)enorm, sumw, fact 
      output was 2.00000000000000       0.000000000000000E+000 Infinity
      so the problem is from sumw.
   
  Could you kindly advise me on what to do? I checked TETRA and it was fine but 
  TEMP is needed for faster convergence ad few k-points.
   
  Thank you
       
   
   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20070514/4929dcb2/attachment.html