[molpro-user] Integral thresholds for CCSD(T)

[Mikael Johansson]

Hello All!

I'm having trouble converging the CCSD in a CCSD(T) calculation. I  
managed to converge the CCSD doing a direct calculation, and  
tightening the thresholds using:

{ccsd
direct, THR_DTRAF=1.D-12, THR_DKEXT=1.D-12
start,4000.2
save,4000.2}

which gave the following convergence:
  ITER.    ENERGY CHANGE        DEN1      VAR(S)    VAR(P)
   11       -0.00000033     0.00000000  0.85D-10  0.29D-10
   12        0.00000017     0.00000000  0.30D-10  0.90D-11

(With default evaluation thresholds, VAR(S) got stuck at ~0.1D-08 and  
VAR(P) at ~0.2D-06)

But when restarting for the (T) part, I couldn't find a keyword to  
control any evaluation thresholds, do they exist? Any way, the restart  
does not converge the CCSD:

  ITER.    ENERGY CHANGE        DEN1      VAR(S)    VAR(P)
   33        0.00000001    -0.00000002  0.15D-11  0.53D-08
   34        0.00000000    -0.00000003  0.12D-11  0.58D-08

The VAR(P) doesn't go below ~0.5D-08. So the question is, is there  
anything to tune to try to get the CCSD converged? If not, is the (T)  
evaluation expected to give reliable results if I just loosen the  
convergence threshold of the CCSD part?

The total CCSD energies are essentially the same for the converged  
calculation and the non-direct CCSD that refuses to converge, that is,  
-462.44965903 and -462.44965917, respectively.

Any help would be appreciated, I'd really like to get the (T) of this  
system :-)

Have a nice day,
     Mikael J.
     http://www.iki.fi/~mpjohans/


----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.