[molpro-user] CCSD(T) RHF not converging

[Kirk Peterson]

Hi Jörg,

yes, the input below should work - Molpro is one of the few codes I  
know that will do this seamlessly.

good luck!

-Kirk

On Jul 29, 2009, at 2:15 AM, Jörg Saßmannshausen wrote:

> Hi Kirk,
>
> thanks for the reply.
> I am currently trying it with the
> hf,maxit=300
> lets see if that is working.
> The MP2 calculation was performed with a different program and a  
> different
> basis set. That seemed to have worked without much problems.
>
> I am not sure about the HF/6-31G* before I am doing the HF/cc-pVTZ
> calculation. I actually have tried that (with a different program)  
> but I was
> unable to read the MO (which probably makes sense if one thinks  
> about it).
> So the input would be something like that:
>
> basis=6-31G*
> hf
> basis=cc-pVTZ
> hf
> ccsd(t)
>
> or?
>
> Lets see if it is picking up the 300 iterations this time and if it is
> converging.
>
> All the best from Glasgow!
>
> Jörg
>
> On Tuesday 28 July 2009 17:22:50 Kirk Peterson wrote:
>> Jörg,
>>
>> in terms of the maxit option try either
>>
>> hf,maxit=100
>>
>> or
>>
>> hf;maxit,100
>>
>> You mention the structure was optimized at the MP2 level.  Was this
>> with Molpro and did you not see
>> HF convergence problems there?  You could perhaps try put a HF/6-31G*
>> calculation in front of  your
>> cc-pVTZ one to see if that will generate better starting orbitals.
>>
>> regards,
>>
>> Kirk
>>
>> On Jul 28, 2009, at 3:16 AM, Jörg Saßmannshausen wrote:
>>> Dear Sarandis,
>>>
>>> thanks for the swift reply.
>>>
>>> I am trying it now as a serial run, see if that makes a difference.
>>>
>>> What I cannot work out is, despite setting
>>> maxit=300
>>> or
>>> gmaxit=300
>>> I still get
>>> MAX. NUMBER OF ITERATIONS:       60
>>> in the
>>> 1PROGRAM * RHF-SCF (CLOSED SHELL)
>>> section of the output. Anybody an idea what I am doing wrong here?
>>>
>>> All the best
>>>
>>> Jörg
>>>
>>> On Tuesday 28 July 2009 10:15:23 you wrote:
>>>> Dear  Jörg, hello
>>>>
>>>> I am having a similar problem with ccsd(t) calculations using
>>>>
>>>> 2006.1 molpro running in 8 parallel processors. In the past,
>>>>
>>>> the same input was ok for sequential run on a different computer.
>>>>
>>>> I was wondering whether you experience the same problem if
>>>>
>>>> you run it on a sequential mode? I think by the end of today,
>>>>
>>>> I will be able to see whether the problem with my input file  
>>>> appears
>>>>
>>>> only in the parallel version ...
>>>>
>>>> With best wishes,
>>>> Sarandis
>>>
>>> --
>>> *************************************************************
>>> Jörg Saßmannshausen
>>> Research Fellow
>>> University of Strathclyde
>>> Department of Pure and Applied Chemistry
>>> 295 Cathedral St.
>>> Glasgow
>>> G1 1XL
>>>
>>> email: jorg.sassmannshausen at strath.ac.uk
>>> web: http://sassy.formativ.net
>>>
>>> Please avoid sending me Word or PowerPoint attachments.
>>> See http://www.gnu.org/philosophy/no-word-attachments.html
>>> _______________________________________________
>>> Molpro-user mailing list
>>> Molpro-user at molpro.net
>>> http://www.molpro.net/mailman/listinfo/molpro-user
>
>
>
> -- 
> *************************************************************
> Jörg Saßmannshausen
> Research Fellow
> University of Strathclyde
> Department of Pure and Applied Chemistry
> 295 Cathedral St.
> Glasgow
> G1 1XL
>
> email: jorg.sassmannshausen at strath.ac.uk
> web: http://sassy.formativ.net
>
> Please avoid sending me Word or PowerPoint attachments.
> See http://www.gnu.org/philosophy/no-word-attachments.html
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user