[Wien] RE : Re: mini_lapw !!!!!!!!!

[Stefaan Cottenier]

I noticed before (but never really went into it) that sometimes the 
'forces small' message is written without convergence having been 
reached. This is what happened in your case. Neglect that message, and 
when your restart the relaxation procedure, make sure your spheres are 
small enough (and basis set etc. are good enough for the force accuracy 
you need).

Stefaan


nadia iles wrote:

>Before runinig mini_lapw, in the scf files the forces
>along z axis are in mRy/a.u:
>at1 =0
>at2=0
>at3=10.115
>at4=12.071
>at5=-87.748
>at6=27.757
>
>As mentionned in optimisation notes in the wien web
>site, when Forces are greater than 5 mRy/a.u, then we
>must relax our structure. After relaxation in the file
>case.scf_mini, the forces became:
>  
>at1 =0
>at2=0
>at3=18.313
>at4=10.01
>at5=-42.295
>at6=0.577
>
>and I obtain the previous error message and warning,
>if my atoms don't need to be relaxed, why did I found
>great forces values???????? 
>
>
>
>
>
>
>--- Stefaan Cottenier
><Stefaan.Cottenier at fys.kuleuven.be> a écrit :
>
>  
>
>>>The
>>>file error is: mini.error, it contains : "Stop in
>>>mini. Forces small" . 
>>>
>>>      
>>>
>>If that 'error' message is true, then you can be
>>happy: your structure 
>>is sufficiently relaxed, as the final forces are
>>sufficiently small.
>>
>>    
>>
>>>at the end of the scf_mini file
>>>there is 2 warning:
>>>1- Step size reduced due to overlapping spheres --
>>>check RMT
>>>2- You might have inconsistant Forces/energies.
>>> 
>>>
>>>      
>>>
>>Apparently two atoms came rather close to each other
>>during the 
>>relaxation procedure, and a 'collision' of the
>>spheres was avoided by 
>>moving them not so much. Nevertheless, this seems to
>>have resulted in 
>>small forces. Check with smaller spheres, and
>>observe whether the forces 
>>remain small. If they do, you should not further
>>relax (assuming that 
>>basis set, k-point sampling etc. are OK to yield
>>good forces).
>>
>>Stefaan
>>
>>    
>>
>>>I have changed RMTs but I'm waiting the scf cycles
>>>      
>>>
>>to
>>    
>>
>>>be finished to execute mini_lapw to relax the
>>>      
>>>
>>surface
>>    
>>
>>>atoms.
>>>Can someone help to know the problems in my
>>>calculations??? should I dispalece manualy atoms
>>>      
>>>
>>from
>>    
>>
>>>their original position to ensure the
>>>      
>>>
>>convergence???
>>    
>>
>>>advanced thanks.
>>>ILES NADIA.
>>>Laboratooire de Physique des Couches minces et
>>>Matériaux pour l'Electronique.
>>>
>>>
>>>
>>>
>>>
>>>	
>>>
>>>	
>>>		
>>>      
>>>
>>___________________________________________________________________________
>>
>>    
>>
>>>Découvrez une nouvelle façon d'obtenir des réponses
>>>      
>>>
>>à toutes vos questions ! 
>>    
>>
>>>Profitez des connaissances, des opinions et des
>>>      
>>>
>>expériences des internautes sur Yahoo!
>>Questions/Réponses 
>>    
>>
>>>http://fr.answers.yahoo.com
>>>_______________________________________________
>>>Wien mailing list
>>>Wien at zeus.theochem.tuwien.ac.at
>>>      
>>>
>>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>    
>>
>>> 
>>>
>>>      
>>>
>>Disclaimer:
>>http://www.kuleuven.be/cwis/email_disclaimer.htm
>>
>>_______________________________________________
>>Wien mailing list
>>Wien at zeus.theochem.tuwien.ac.at
>>
>>    
>>
>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>  
>
>
>
>
>	
>
>	
>		
>___________________________________________________________________________ 
>Découvrez une nouvelle façon d'obtenir des réponses à toutes vos questions ! 
>Profitez des connaissances, des opinions et des expériences des internautes sur Yahoo! Questions/Réponses 
>http://fr.answers.yahoo.com
>_______________________________________________
>Wien mailing list
>Wien at zeus.theochem.tuwien.ac.at
>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>  
>


Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm