[Pw_forum] bfgs error

[brsahu at physics.utexas.edu]

Dear PWSCF users,

I encountered an error in a calculation (carbon slabs, ultrasoft PP,  
ecutwfc = 30 Ry and ecutrho = 250 Ry)

CASE: energy_new > energy_old

      new trust radius        =       0.0000495950 bohr

      trust_radius < trust_radius_min

      resetting bfgs history


   
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
      from bfgs : error #         1
      bfgs history already reset at previous step
   
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

      stopping ...

The solution suggested in the mailing list (scf convergence criteria  
can not be greater than the force
criteria) is contrary to my inputs

as

  Total force =     0.003851     Total SCF correction =     0.000054


      entering subroutine stress ...

           total   stress  (Ry/bohr**3)                   (kbar)     P=   -8.86
   -0.00015635   0.00000000   0.00000000        -23.00      0.00      0.00
    0.00000000  -0.00001219  -0.00000281          0.00     -1.79     -0.41
    0.00000000  -0.00000281  -0.00001204          0.00     -0.41     -1.77


      number of scf cycles    =  61
      number of bfgs steps    =  47

      energy old              =    -370.1139086454 Ry

      energy new              =    -370.1138810384 Ry

and in the input file I have

forc_conv_thr = 1.0D-3
conv_thr =  1.0d-9

etot_conv_thr is default

any suggestions?

Sahu
UT-Austin, USA