[Pw_forum] "charge is wrong"

Francesca Costanzo costanzo at ms.fci.unibo.it
Wed Apr 23 18:15:58 CEST 2008


Dear Stefano,
after
      number of scf cycles    =  14
      number of bfgs steps    =  12
where the energy were going down the run was interrupted without error.

Then, I took the last optimized geometry starting "from_scratch" and
the error "charge is wrong" appears after
      number of scf cycles    =   1
      number of bfgs steps    =   0

the only possibility that I see is that it took the old si.save file
without overwrite on it, although I used the keyword "from_scratch".
Is it possible?
 		Thanks anywyay,
 			 Francesca


On Wed, 23 Apr 2008, Stefano de Gironcoli wrote:

> Dear Framcesca,
>    when does the error occurs ?
>    before/after the first scf iteration? just after an ionic move ?
> after restarting an interrupted run ?
>    stefano
>
> Francesca Costanzo wrote:
>> Dear all,
>> I have a problem when I run the optimization geometry of
>> terthiophene on Si(001).
>>
>> The error looks like:
>> -------------------------------------------------------------------------
>> WARNING: integrated charge=   476.79997686, expected=   490.00000000
>>
>>
>> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>       from electrons : error #         1
>>       charge is wrong
>>
>> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>
>>       stopping ...
>> --------------------------------------------------------------
>> I have read in your forum that this error can be generated by a wrong
>> pseudopotential.
>> I don't think is that the problem
>> since  I have already optimized with the same conditions and the
>> same pseudopotentials the dithiophene molecule on the Si(001) surface
>> and the surface Si(001) such as.
>> May you suggest to me what should I pay attention to ?
>> best regards,
>>  		Francesca Costanzo
>>
>>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>



More information about the Pw_forum mailing list