SCF-Error

129 views
Skip to first unread message

thamim m

unread,
Dec 3, 2021, 2:05:35 AM12/3/21
to NWChem Forum
Hi everyone,

Greetings of the day. I am a new user of NWchem. I was trying to do DFT optimisation calculation and ended up with an error like this - Received an Error in Communication: (6291456) 0:grpwrite: insuff eafsize:
add input line set grid:eaf_size_in_dbl 6304257. 
Even i include the key word set grid:eaf_size_in_dbl  8500000 in my input line but still am getting error. Can any one help me to sort out this problem.

Thanks in advance.
M.Thamim


output.109292.nwchem-test
Input file.txt

Edoardo Aprà

unread,
Dec 3, 2021, 2:08:06 AM12/3/21
to NWChem Forum
Try the following

...
 set grid:eaf_size_in_dbl     12565433
task dft optimize

thamim m

unread,
Dec 3, 2021, 10:33:02 AM12/3/21
to NWChem Forum
Dear  Edoardo Aprà,
I try this above mentioned key-word in my input line but again am getting error like : WARNING : Found    23 linear dependencies
Looking forward your response
Thanks in advance.
M.Thamim
output.109564.nwchem-test

Edoardo Aprà

unread,
Dec 3, 2021, 1:23:52 PM12/3/21
to NWChem Forum
The warning message found in the output file is not an error. It is just an indication of potential numerical instability. To avoid this issue, you can switch from the default cartesian to spherical basis set as shown in the input snippet below.

basis spherical
  * library 6-311++G**
end


thamim m

unread,
Dec 4, 2021, 10:29:12 AM12/4/21
to NWChem Forum
Dear  Edoardo Aprà,
Again am getting the warning message like : WARNING : Found    23 linear dependencies
Looking forward your response
Thanks in advance.
M.Thamim
output.109699.nwchem-test

Edoardo Aprà

unread,
Dec 6, 2021, 7:56:20 PM12/6/21
to NWChem Forum
As I mentioned in the Dec 3 posting, that's a warning, not an error message.

If you are trying to speed up the calculation you might want to change the input file in the following way
...
dft
  direct
  xc b3lyp
end

set quickguess t

task dft

thamim m

unread,
Dec 7, 2021, 10:15:22 AM12/7/21
to NWChem Forum
Dear  Edoardo Aprà,
I tried all the way and followed all your instruction, but still my input is not optimised. Herewith i attached the output files for your kind perusal
Looking forward your response
Thanks in advance.
M.Thamim
output.110161.nwchem-test
output.110160.nwchem-test
Reply all
Reply to author
Forward
0 new messages