[CPMD-list] Pseudopotential/cutoff/box-size

[Pradip Kumar Biswas]

Dear all,

i was trying a test run on external potential using a water-dimer: 
One molecule in the box and the other comes as an external potential.
 I used H_SG_LDA & O_SG_LDA and H_MT_PBE & O_MT_PBE. With a cutoff of
120 and boxsize of 20, I find arround 30-40 fold difference in the
forces on the atoms: also the energies differ a lot (given below).

**For MT_PBE, I get a warning message that XC-functionals are inconsistent
for O_MT_PBE. 

But should that lead to a 30-40 fold difference in force? 

Can anybody give me an idea, why the two pseudo-potential differ that
much? Where can I find a fair description on the pseudo-potential/cutoff/box-size
and their prudent use. 

Thanks in advance.
Pradip.

for MT_PBE:

 (K+E1+L+N+X)           TOTAL ENERGY =          -15.11096543 A.U.
 (K)                             KINETIC ENERGY =            7.78105945 A.U.
 (E1=A-S+R)  ELECTROSTATIC ENERGY =          -11.81976842 A.U.
 (S)                                              ESELF =           12.63317221 A.U.
 (R)                                                 ESR =             .76626227 A.U.
 (L)    LOCAL PSEUDOPOTENTIAL ENERGY =           -7.53254590 A.U.
 (N)      N-L PSEUDOPOTENTIAL ENERGY =             .00000000 A.U.
 (X)     EXCHANGE-CORRELATION ENERGY =           -3.53971056 A.U.


for SG_LDA:

 (K+E1+L+N+X)           TOTAL ENERGY =          -17.12839327 A.U.
 (K)                             KINETIC ENERGY =           13.55607066 A.U.
 (E1=A-S+R)  ELECTROSTATIC ENERGY =          -11.34709110 A.U.
 (S)                                              ESELF =           12.63317221 A.U.
 (R)                                                 ESR =             .87906879 A.U.
 (L)  LOCAL PSEUDOPOTENTIAL ENERGY =          -16.37995374 A.U.
 (N)      N-L PSEUDOPOTENTIAL ENERGY =            1.16261164 A.U.
 (X)  EXCHANGE-CORRELATION ENERGY =           -4.12003073 A.U.
 

--
Universidade do Vale do Paraíba - UNIVAP.
http://www.univap.br/