SCF convergence is reached?

Total energy, geometry optimization, DFT+U, spin....

Moderator: bguster

Locked
proffess
Posts: 5
Joined: Wed Jan 13, 2010 4:39 pm

SCF convergence is reached?

Post by proffess » Wed May 24, 2017 8:27 pm

Hello,

I'm looking into my GS optimization (abinit version 8.2.3) and noticed a strange thing:

......
ETOT 35 -137.39514562659 -1.165E-12 7.789E-07 2.893E-09
ETOT 36 -137.39514562659 -1.421E-13 8.315E-06 2.096E-09
ETOT 37 -137.39514562659 -9.095E-13 6.728E-07 3.449E-09
ETOT 38 -137.39514562659 1.222E-12 7.455E-06 2.727E-09
ETOT 39 -137.39514562659 -2.274E-13 5.791E-07 3.138E-09

Cartesian components of stress tensor (hartree/bohr^3)
sigma(1 1)= 1.44637027E-07 sigma(3 2)= 0.00000000E+00
sigma(2 2)= 1.44637027E-07 sigma(3 1)= 0.00000000E+00
sigma(3 3)= 1.44637027E-07 sigma(2 1)= 0.00000000E+00


scprqt: WARNING -
nstep= 50 was not enough SCF cycles to converge;
potential residual= 3.138E-09 exceeds tolvrs= 1.000E-12

Last SCF step is #39, and after that it says "50 steps is not enough". Did it skip last 11 steps or what?

Thanks,

Locked