PC GAMESS/Firefly-related discussion club

Learn how to ask questions correctly

Alex Granovsky
gran@classic.chem.msu.su

note, this forum is international - so please! use only English letters to name files you want to archive and attach.

Your input is not very informative because it looks like
it is artificially post-assembled using results of
at least two different attempts to calculate hessian...

If you look onto dipole moments (the last row of the each
subsection of $vib group), there is a sudden change of the
sign of the dipole moment at some point... My guess is that
you used coord=cart and restarted hessian calculations at
that point. However, coord=cart should not be normally used
unless you really know why exactly you need to use them.
One of the problem here is that coord=cart reorient your system.
However, the same coordinate frame must be used for the
entire calculations. I therefore guess that at some point
the molecule was reoriented... and this makes calculations
more or less senseless (or, if you know how exactly it was
reoriented you can manually rotate gradient and dipole moment
data to the common frame).

Thus, the output files would be much more informative -
as I doubt if everybody here will repeat the (relatively
costly) calculations from the scratch.

regards,
Alex Granovsky


On Fri Nov 13 '09 0:28am, Alexandr wrote
----------------------------------------
>  Hi,
>   I count hessian and frequencies.
>   The molecule structure has been preliminary optimised in the same basis. Components of Energy gradient too have turned out small.
>   But among components dipole derivative tenzor there were big unreal numbers.
>   As result - 4 imaginary values of frequencies.
>   It is wrong.
>   It is failure of the program which is reproduced.
>   What to do?
>   I attach files INPUT and OUTPUT.
>   Thanks in advance!
>
>

[ This message was edited on Fri Nov 13 '09 at 2:34am by the author ]