The ABINIT Users Mailing List ( CLOSED )

[mamikon Gulian <mamikon.armen.gulian@gmail.com>]

It appears this problem is due to the position of the tube. If it is centered in the unit cell, the error results. If in ABINIT I shift it so that the tube is centered around the z-axis of the unit cell DP works fine afterwards. Strange.

On Thu, Jan 7, 2010 at 7:45 PM, <mamikon.armen.gulian@gmail.com> wrote:

Hello,

I have a problem running the DP-Code for a Nanotube following a KSS file
calculation with ABINIT.

ABINIT automatically selects a FFT grid of 72x72x24 for my system during the
KSS file calculation.
When I try to use DP on the KSS file I get the error:
 divisor mesh  5042 2521 6
 selecting fft mesh size
 alias =  2
 **ERROR allowed_fftsize
STOP error allowed fft_size

Which is very strange given that ABINIT selected the appropriate FFT Grid
(ngfft) size itself.

One thing to note is that when I calculate with bulk Al instead of a NT ABINIT
takes an FFT Grid size of 12x12x12, but DP, although it works then, seems to
take a size of 9x9x9:
 divisor mesh  1 1 1
 selecting fft mesh size
 alias =  2
 mesh size selected:   9 x  9 x  9
 total size, nr =    729
 fft tables setup

So I guess DP calculates the optimal FFT size independently of the ngfft
parameter of the KSS file.
This would mean that DP for some reason can't get it in the case of the
Nanotube even though ABINIT can.

Is this interpretation correct? Any suggestions on how to fix this? I know
others have used DP to calculate with NT's.

Thanks,
-Mamikon