Page 1 of 1

Density went too small error using PAW pp

Posted: Tue Apr 24, 2012 12:09 pm
by njuxyh
Hi all:

i am a beginner. i have found a warning (the red part below) in out file when i do calcualtion of ground state within paw framework.
i have increase the ecut to 30 ha, and the pawecutdg to 100 .but the warning still exists.
i think the warning maybe related paw pp. i tried two version : 6c_lda.paw and 6c.lda.atompaw.
does anyone cound give me a hints?

thank you very much !

FFT (fine) grid used in SCF cycle:

getcut: wavevector= 0.0000 0.0000 0.0000 ngfft= 45 45 240
ecut(hartree)= 100.000 => boxcut(ratio)= 2.01453
scfcv : before setvtr, energies%e_hartree= 0.000000000000000E+000

ewald : nr and ng are 4 and 48
setvtr : istep,n1xccc,moved_rhor= 1 1 0
mkdenpos : enter
xc_denpos= 1.000000000000000E-014

mkdenpos : WARNING -
Density went too small (lower than xc_denpos) at 86960 points
and was set to xc_denpos= 1.00E-14. Lowest was -0.72E-06.
Likely due to too low boxcut or too low ecut for pseudopotential core charge.
scfcv : after setvtr, energies%e_hartree= 67.2531949233842

Re: Density went too small error using PAW pp

Posted: Fri Jun 22, 2012 5:04 pm
by CarmineAutieri
Hi

You find a DOS negative, but the negative value of the DOS is very small.
If the DOS was large, then you must pay attention because it is possible that you have a problem.
You can neglect this WARNING.

Best
Carmine