[CP2K:2103] Re: small Jump in Conserved Quantity

Francesco Ragone francesc... at gmail.com
Tue Jun 2 15:04:17 UTC 2009


Thank you very much Teo,
fast and clear as always!

Ciao
Francesco

Teodoro Laino wrote:
> Hi,
>
> When you restart you still have the keyword CENTER_COORDINATES in the 
> topology section of your input file.
> Therefore your system will be centered and since codes in PW (QS is 
> partially one of them) are not translational invariant you will get 
> slightly different energies (if you move the system around the box).. 
> the situation may even be more severe since you use the wavelet decoupling.
> The jump will be smaller if you use a larger cutoff., with tuckerman 
> decoupling for example, while with wavelet you may still experience 
> large jumps.
> Two option for you:
> 1) live with this jump (is anyway small)
> 2) disable the center_coordinates in the topology section
>
> Teo
>
> Francesco wrote:
>   
>> Hi to all,
>> following a NVE simulation of an isolated molecule, I found that after
>> the restart there's a jump in the Conserved Quantity ( I obtain jumps
>> like this each time I restart the simulation).
>>
>> #  Step Nr.          Time[fs]               Kin.[a.u.]
>> Temp[K]                Pot.[a.u.]             Cons Qty[a.u.]
>>       2040         1020.000000         0.060956952
>> 188.712354820      -350.782198702      -350.721241749
>>       2041         1020.500000         0.060918552
>> 188.593475153      -350.782159564      -350.721241011
>>       2042         1021.000000         0.060888428
>> 188.500213394      -350.782128677      -350.721240249
>>       2043         1021.500000         0.060871701
>> 188.448430001      -350.782111378      -350.721239677
>>       2044         1022.000000         0.060872854
>> 188.452001719      -350.782111983      -350.721239129
>>       2045         1022.500000         0.060895749
>> 188.522878866      -350.782134430      -350.721238681
>>       2046         1023.000000         0.060943620
>> 188.671078602      -350.782181953      -350.721238333
>>       2047         1023.500000         0.061019026
>> 188.904523257      -350.782256965      -350.721237939
>>       2048         1024.000000         0.061123792
>> 189.228861046      -350.782361468      -350.721237677
>>
>>  
>> RESTART
>>
>>       2041         1020.500000         0.060949314
>> 188.688708063      -350.782442958      -350.721493644
>>       2042         1021.000000         0.060941947
>> 188.665901432      -350.782435584      -350.721493636
>>       2043         1021.500000         0.060938872
>> 188.656381373      -350.782432557      -350.721493684
>>       2044         1022.000000         0.060944036
>> 188.672366867      -350.782437451      -350.721493415
>>       2045         1022.500000         0.060961411
>> 188.726156528      -350.782454251      -350.721492841
>>       2046         1023.000000         0.060994791
>> 188.829498101      -350.782486835      -350.721492043
>>       2047         1023.500000         0.061047420
>> 188.992425479      -350.782538667      -350.721491247
>>       2048         1024.000000         0.061122123
>> 189.223695729      -350.782613042      -350.721490918
>>
>> Is this "jump" a round off problem?
>> Is it small (I mean negligible)?
>>
>> Attached the input file
>> ----
>> In the input file I've chosen to print the ener file for each step,
>> while the position, velocity, wave functions and restart each 20 step
>> ---
>>
>>
>> &EXT_RESTART
>>    RESTART_FILE_NAME RX1bis-RX2TS-1.restart
>> &END EXT_RESTART
>> &MOTION
>>   &MD
>>    COMVEL_TOL 1E-08
>>    ANGVEL_ZERO T
>>    ENSEMBLE NVE
>>    STEPS 3000
>>    TIMESTEP 0.5
>> !   TEMPERATURE 300.0
>> !   &THERMOSTAT
>> !      TYPE NOSE
>> !      REGION GLOBAL
>> !      &NOSE
>> !        LENGTH 4
>> !        TIMECON 20
>> !      &END NOSE
>> !   &END THERMOSTAT
>>   &END MD
>>   &PRINT
>>     &TRAJECTORY
>>       &EACH
>>         MD 20
>>       &END EACH
>>     &END TRAJECTORY
>>     &VELOCITIES
>>       &EACH
>>         MD 20
>>       &END EACH
>>     &END VELOCITIES
>>     &RESTART
>>       &EACH
>>         MD 20
>>       &END EACH
>>     &END RESTART
>>   &END PRINT
>>   &GEO_OPT
>>   &END GEO_OPT
>> &END MOTION
>> &FORCE_EVAL
>>    METHOD QS
>>    &DFT
>>       &POISSON
>>        POISSON_SOLVER WAVELET
>>        PERIODIC NONE
>>      &END POISSON
>>      BASIS_SET_FILE_NAME BASIS_NEW
>>      POTENTIAL_FILE_NAME POT_NEW
>>      WFN_RESTART_FILE_NAME RX1bis-RX2TS-RESTART.wfn
>>      MULTIPLICITY 1
>>      CHARGE 0
>>     &MGRID
>>       CUTOFF 300
>>     &END MGRID
>>     &QS
>>       EXTRAPOLATION ASPC
>>       EXTRAPOLATION_ORDER 3
>>     &END QS
>>     &SCF
>>       EPS_SCF 1.0E-6
>>       SCF_GUESS RESTART
>>       MAX_SCF 500
>>       &PRINT
>>         &RESTART
>>           &EACH
>>             MD 20
>>           &END EACH
>>         &END RESTART
>>       &END PRINT
>>       &OT
>>         MINIMIZER DIIS
>>         N_DIIS 7
>>         PRECONDITIONER FULL_SINGLE_INVERSE
>>       &END OT
>>       &OUTER_SCF
>>         EPS_SCF 1.0E-4
>>         MAX_SCF 4
>>       &END OUTER_SCF
>>     &END SCF
>>     &XC
>>       &XC_FUNCTIONAL PBE
>>       &END XC_FUNCTIONAL
>>     &END XC
>>   &END DFT
>>   &SUBSYS
>>     &TOPOLOGY
>>        CENTER_COORDINATES T
>>     &END TOPOLOGY
>>     &CELL
>>       ABC 20.0 20.0 20.0
>>       PERIODIC NONE
>>     &END CELL
>>     &COORD
>>     &END COORD
>>     &VELOCITY
>>     &END VELOCITY
>>     &KIND H
>>       BASIS_SET DZVP-GTH-PBE
>>       POTENTIAL GTH-PBE-q1
>>     &END KIND
>>     &KIND C
>>       BASIS_SET DZVP-GTH-PBE
>>       POTENTIAL GTH-PBE-q4
>>     &END KIND
>>     &KIND N
>>      BASIS_SET DZVP-GTH-PBE
>>      POTENTIAL GTH-PBE-q5
>>     &END KIND
>>     &KIND Cl
>>      BASIS_SET DZVP-GTH-BLYP
>>      POTENTIAL GTH-PBE-q7
>>     &END KIND
>>   &END SUBSYS
>> &END FORCE_EVAL
>> &GLOBAL
>>   PROJECT_NAME RX1bis-RX2TS
>>   RUN_TYPE MD
>>   PRINT_LEVEL MEDIUM
>> &END GLOBAL
>>     
>>   
>>     
>
>
> >
>
>   




More information about the CP2K-user mailing list