[molpro-user] Error: Overflow in exponential generator

Klaus Doll doll at theochem.uni-stuttgart.de
Fri Jul 15 16:37:59 CEST 2016


Dear Andrew,
when I run your input with Molpro 2015, the first "hf" command results 
in a non-converged calculation.
In the output, in the RHF-SCF section you find that the calculation 
stops after 60 cycles
(which is also the  "MAX. NUMBER OF ITERATIONS:       60"), and there is 
a message

" No convergence "

at the end of the RHF-SCF section. The following UHF-SCF does not 
converge either.
You have to converge these calculations first.
Also check whether the orbital guess is correct, where it  says:

  "Orbital guess generated from atomic densities. Full valence 
occupancy:   13   5   5   1

  Molecular orbital dump at record        2100.2

  Initial alpha occupancy:  12   5   5   1
  Initial beta  occupancy:  11   4   4   0
  Wave function symmetry:    1  "


best wishes
Klaus



On 12.07.2016 07:47, Andrew Rosen wrote:
> Hello,
>
> I am running a CASPT2 calculation on FeS. However, I receive the 
> following error message:
> zz= 4933.00251596958 nterm= 10001
> ? Overflow in exponential generator
> ? The problem occurs in muint2:updui
>
> Any idea what this indicates and how it can be resolved?
>
> separator
> Here is my input:
>
> ***, casscf_caspt2_fes
> gprint, orbitals, civector
> memory, 125, m
> set, charge=0
> set, spin=4
> c=[9,3,3,0]
> o=[7+c(1),3+c(2),3+c(3),1+c(4)] !o-c = active orbitals
> geomtyp=xyz
> geometry={
> Fe 0.000000 0.000000 0.000000
> S 1.960000 0.000000 0.000000
> }
> basis=aug-cc-pVTZ
> hf
> uhf
> {casscf
> closed,c(1),c(2),c(3),c(4)
> occ,o(1),o(2),o(3),o(4)
> }
> {multi
> closed,c(1),c(2),c(3),c(4)
> occ,o(1),o(2),o(3),o(4)
> pspace,10
> }
> {rs2,xms=1,mix=1}
> put,molden,casscf_caspt2_fes.molden
>
> separator
> Here is the relevant part of my output:
>
>  Setting scratch file implementation to df
> 1PROGRAM * MULTI (Direct Multiconfiguration SCF) Authors: P.J. 
> Knowles, H.-J. Werner (1984) S.T. Elbert (1988)
>  Number of closed-shell orbitals: 15 ( 9 3 3 0 )
>  Number of active orbitals: 14 ( 7 3 3 1 )
>  Number of external orbitals: 114 ( 42 28 28 16 )
>  Molecular orbitals read from record 2140.2 Type=MCSCF/NATURAL (state 
> 1.1)
>  Wavefunction dump at record 2141.2
>  Convergence thresholds 0.10E-01 (gradient) 0.10E-05 (energy) 0.10E-02 
> (step length)
>  Number of orbital rotations: 1105 ( 81 Core/Active 546 Core/Virtual 0 
> Active/Active 478 Active/Virtual)
>  Total number of variables: 1105
>  ITER. MIC NCI NEG ENERGY(VAR) ENERGY(PROJ) ENERGY CHANGE GRAD(0) 
> GRAD(ORB) GRAD(CI) STEP TIME
>    1 200 0 0 -1636.32478886 -2278.59498333 -642.27019448 
> 4.5848038915.49091256 0.00000000 0.25D+01 3.97
>    2 201 0 0 -1617.43568430 -1784.54250389 -167.10681958 14.20444960 
> 3.28767730 0.00000000 0.16D+01 8.37
>    3 201 0 0 -1619.98983143 -1757.81472638 -137.82489494 9.06267977 
> 0.38160069 0.00000000 0.17D+01 13.26
>    4 201 0 0 -1624.46404613 -1773.31008295 -148.84603682 9.13703354 
> 1.01409046 0.00000000 0.16D+01 17.71
>    5 201 0 0 -1623.15928098 -1859.47718813 -236.31790715 9.26978464 
> 1.87010289 0.00000000 0.12D+01 22.32
>    6 201 0 0 -1627.94731713 -1924.82071840 -296.87340127 6.20277339 
> 2.36648215 0.00000000 0.11D+01 26.75
>    7 201 0 0 -1623.32553269 -1876.60182015 -253.27628746 5.82296938 
> 2.40385291 0.00000000 0.12D+01 32.43
>    8 201 0 0 -1625.17070007 -1856.06673244 -230.89603237 9.47014702 
> 2.05460418 0.00000000 0.19D+01 37.07
>    9 201 0 0 -1567.67748756 -1888.79006525 -321.11257769 *********** 
> 2.94358214 0.00000000 0.16D+01 41.65
>   10 201 0 0 -1625.58361683 -1856.13712752 -230.55351068 8.40778505 
> 3.92290580 0.00000000 0.15D+01 46.05
>  zz= 4933.00251596958 nterm= 10001
> * ? Error *
> * ? Overflow in exponential generator *
> * ? The problem occurs in muint2:updui *
>
>
>
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20160715/5ef7071a/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 92 bytes
Desc: not available
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20160715/5ef7071a/attachment.png>


More information about the Molpro-user mailing list