[Wien] Failure in mini

Jorissen Kevin Kevin.Jorissen at ua.ac.be
Tue Aug 5 14:24:15 CEST 2003


Hello Michael,
 
this is not a crash in mini.  As you know, mini is just a script for launching several scf-calculations one after the other, and moving the atoms in between two calculations.
Your error occurred INSIDE of one scf-calculation.
The '***' in your help-file, and the sudden charge move suggest that this particular scf-calculation diverged in some way.  Two possibilities :
* It's just a numerical divergence.  In this case, you will have to restart the calculation with modified input (eg, different mixing parameter).  So you keep the current case.struct, modify some input, and restart min.
* It is a mini-divergence after all.  Mini has moved your atoms in such a way that they got into an unphysical configuration.  In this case you can't keep the current struct-file, but you'll have to get back to a previous one, and modify some of the parameters in case.inM (probably decrease them), and then restart mini.
 
Hope this helps.  Good luck!
 
Kevin.
 

	-----Original Message----- 
	From: Michael Gurnett [mailto:michael.gurnett at kau.se] 
	Sent: Mon 8/4/2003 1:26 PM 
	To: Wien at zeus.theochem.tuwien.ac.at 
	Cc: 
	Subject: [Wien] Failure in mini
	
	

	I was running mini at was at the 9th scf cycle when after or during
	lapw2c the following error occurred
	
	Input/Output Error 148: Invalid character
	
	   In Procedure: outp
	        At Line: 181
	
	      Statement: Formatted READ
	           Unit: 1040
	   Connected To: ./Germanium.help070
	           Form: Formatted
	         Access: Sequential
	Records Read   : 105
	Records Written: 0
	
	Current I/O Buffer:
	
	  L= 2   97.06107  ********** 14692.603********************-13727.608
	                   !
	
	
	End of diagnostics
	
	
	What is the cause behind this error. The charge distance did suddenly
	increase the step before. I was running it with a mix of 0.2. The system
	is a 50 atom slab using 10 k points.
	
	Hope someone can help
	
	Michael
	
	_______________________________________________
	Wien mailing list
	Wien at zeus.theochem.tuwien.ac.at
	http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
	

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 5822 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20030805/7948320e/attachment.bin


More information about the Wien mailing list