[molpro-user] GRADIENT NORM NOT ZERO!

[Li, Jun]

Dear MOLPRO colleagues,

After numerical frequencies calculations (e.g. ccsd(t)), sometime we got
the following warning:

Gradient norm at reference geometry: 0.11426D-01 
 ?WARNING: GRADIENT NORM NOT ZERO!

Of course the imaginary frequencies would then look strange.  This
happens no matter how tight the geometry is converged in previous
geometry optimization  (e.g. grad changed from the default 3d-4 to
1d-6).  Does anyone has a solution to this problem?

Thanks a lot!

Jun Li
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20050806/0ba6a7f9/attachment.html>