CCL: ESP fit and symmetry in G98

["Petr Toman" <toman %! at !% imc.cas.cz>]

Dear CCL members,
 I am calculating the atomic charges of some symmetric molecules by means
 of the Merz-Kollman fitting. The conformations as well as the density
 matrices are symmetric, but the resulting MK charges are completely
 asymmetric (the differences are more than 0.2 !!!).
 Does anybody know the reason?
 It seems that using a higher point density by means of IOp(6/41=10,6/42=17)
 helps a bit.
 I tried also IOp(6/41=20,6/42=35) but in this case an error occurred:
   287560 points will be used for fitting atomic charges
  Fitting point charges to eletrostatic potential
  Convergence failure in SVDSol.
  Error termination via Lnk1e in /software/g98-a7/sgi_65/g98/l602.exe.
 Does anybody know how can I control the convergence in SVDSol (maybe some
 limit of the number of steps)?
 Thank you in advance for any help.
 Yours,
 Petr Toman