'Selective dynamics' problem

Message

b-jour · #1 Post by **b-jour** » Mon Jun 04, 2012 1:33 pm

Hi everyone

I'm new to VASP, trying to relax a structure with "Selective dynamics"
I've tried to run the simulation without the 'selective dynamics' tag and everything seems to work fine.
When the 'selective dynamics' tag is turned on, I get this in the .log file :

------------
POSCAR found : 2 types and 48 ions
No initial positions read in
------------

It seems that I have a trouble with the POSCAR format.

here is the beginning of the POSCAR file:

------------
structure 1
2.5
1 0 0
0 1 0
0 0 11.2
32 16
Selective Dynamics
Direct
0.25 0.25 0.16517857 T T T
0.75 0.75 0.16517857 T T T
0.25 0.75 0.20982143 T T T
0.75 0.25 0.20982143 T T T
0.25 0.25 0.25446429 T T T
0.75 0.75 0.25446429 T T F
0.25 0.75 0.29910714 T T F
0.75 0.25 0.29910714 T T F
------------

Thanks for your precious help

alex · #2 Post by **alex** » Mon Jun 04, 2012 3:40 pm

Try a '.' behind your integers ...

b-jour · #3 Post by **b-jour** » Mon Jun 04, 2012 6:15 pm

Thanks Alex, unfortunately adding the '.' at the end of the integers didn't change anything, it gives to me the same error : "No initial positions read in"

b_lack · #4 Post by **b_lack** » Tue Jun 05, 2012 5:01 am

I'm not too sure about your input is it full input file or not (to prevent too long message), but seem to be 32+16 = 48 ions but you only write 8 ionic positions, wheres the other 40?

May be that error told us that vasp need more input to achieve 32 16 number of ions code, may be you forget to change the number of ions, but this is just may be, ^_^

b-jour · #5 Post by **b-jour** » Tue Jun 05, 2012 9:29 am

Thank you b_lack, indeed i remove the 40 last ionic positions for shorter post. The number of ions is good as it works perfectly without 'selective dynamics' on. I get the structure i wanted as shown by VMD.

b-jour · #6 Post by **b-jour** » Fri Jun 08, 2012 10:05 am

I'm not sure to understand the meaning of this sentence :
Mind: The 'Selective dynamics' input tag is optional: The seventh line supplies the switch between cartesian and direct lattice if the 'Selective dynamics' tag is omitted.

do my problem could come from this?

b_lack · #7 Post by **b_lack** » Fri Jun 08, 2012 11:48 am

I not too sure, ..I never experiences something like that, but I'm curious about your ISIF-tag, which one did you use?

gevolo · #8 Post by **gevolo** » Fri Jun 08, 2012 12:19 pm

Hello, your POSCAR seems fine. You should check first that you have 48 lines of coordinates (quite obvious, but might happen by mistake) and also that there are no any TAB between coords etc, only spaces.
Hope it helps.
George
<span class='smallblacktext'>[ Edited Fri Jun 08 2012, 12:21PM ]</span>

b-jour · #9 Post by **b-jour** » Sun Jun 10, 2012 10:09 pm

Hy!
@b_lack : I use ISIF=0
I choose that to avoid waste of time for calculating the stress tensor. Should I put ISIF=2?

@gevolo : Im actualy looking but their is 48 lines coordinates for sure. And the structure is constructed from a pyhton script where the separators are only spaces.
The weird fact is that without the 'selective dynamics' tag-on VASP was able to read the POSCAR positions.

b_lack · #10 Post by **b_lack** » Mon Jun 11, 2012 1:26 pm

How about try ISIF=2 and see what happen. in case its running may be those are the cause, but in case still not running we need to investigate another things, ^_^

b-jour · #11 Post by **b-jour** » Mon Jun 11, 2012 10:41 pm

Hy everyone,

I want to thank you all for your help, I found the answer, it's really silly and i feel bad for taking your time.
I forgot one F in one of the last lines.
now everything seems to work perfectly

All my apologies for taking your precious and helpful time!