Restarting jobs from CHGCAR failing

Message

boris · #1 Post by **boris** » Mon Aug 10, 2009 8:35 pm

Dear vasp admin and users

I would like to know the reasons why restarting jobs from CHGCAR files would fail.

My job successfuly ends and the CHGCAR file seems complete.

However, when I set ICHARG=1 and restart the job after moving the CONTCAR file to the POSCAR file, the calculation sometimes does not restart with the CHGCAR file and calculate the charge density of overlapping atoms, which is a problem to me.

Do you have any idea of what could be the reasons of the CHGCAR file not to be taken into account?

Thanks for future answers.

Regards,

Boris Dorado.

dcurulla · #2 Post by **dcurulla** » Mon Aug 10, 2009 10:26 pm

Hi Boris,
What is the value of ISTART? Is it ISTART=0? Maybe, I telling nosense (I haven't tried it myself), but you may need to set ISTART=1.
Cheers,

boris · #3 Post by **boris** » Tue Aug 11, 2009 7:51 am

Thanks for answering.

To my knowledge, the keyword ISTART determines whether the WAVECAR file is read.

As I do not print the WAVECAR file in my calculations, I thus set ISTART=0.

But since restarting from the CHGCAR file sometimes fails, I guess I'll have to print the WAVECAR file and calculate the charge density from the initial wave functions.

But I would have liked to know why my restarts were failing.

B.

#4 Post by **admin** » Mon Oct 12, 2009 3:15 pm

please check whether the basis functions (NGX NGY NGZ) are the same for the old and the new run. If the cell shape or volume have changed (cell relaxation runs) , you will not be able to continue from CHGCAR most probably