[molpro-user] CI VECTOR FOR STATE 1 DOES NOT OVERLAP SUFFICIENTLY WITH REFERENCE VECTORS

[Dawid das]

Dear molpro experts,

Before writing this e-mail, I have browsed mailing list for similar
problems and I found them, but no answer was given.
For some of geometries of the same model I get following error:

 Requested states
   1-  3     1     2     3

 Current NROOT
   1-  3     1     2     3

 CI VECTOR FOR STATE 1 DOES NOT OVERLAP SUFFICIENTLY WITH REFERENCE VECTORS

 It may help to increase NSTATI, e.g., use OPTION,NSTATI=4

 Rotated S(REF|INT)
                   1             2             3
         1     0.0004266    -0.0009188    -0.0015962
         2    -0.0009188     0.0054027    -0.0051402
         3    -0.0015962    -0.0051402     0.9108664
 ? Error
 ? Insufficient overlap
 ? The problem occurs in cihdia

Increasing nstati does not do anything good.
Is it there any idea why MRCISD calculations fail for some of the
geometries of S0 state
of my model, why for geometries obtained at different levels of theory they
do fine?
Is that intrinsic error of mrci directive?

Best wishes,

Dawid Grabarek
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20141127/78340261/attachment.html>