[gmx-users] NVT and NVE

[Markus Miettinen]

Hi Jes,


  to me it seems that you have rlist = rvdw:

> nstlist                  = 10
> rlist                    = 1.0
> rvdw                     = 1.0


  This will lead to a loss of energy, because between list updates
  two particles originally at distance, say, 1.01 from each other
  can easily move within rvdw from each other, but they will not
  "see" each other until the lists are updated again.

  In other words, you have particles moving within the attractive
  LJ-potential region without giving them the speed-up that the
  attractive potential should provide. This mistake equals loss of
  energy, and becomes apparent in NVE. In NVT the thermostat
  (artificially!) fixes the energy loss.

  The correct way is to set rlist > rvdw.


Cheers,
markus.

--
Halli delendum est.

On Apr 3, 2008, at 13:00 , gmx-users-request at gromacs.org wrote:

> Date: 3 Apr 2008 08:12:40 -0000
> From: "JMandumpal" <jesbman at rediffmail.com>
> Subject: Re: Re: [gmx-users] NVT and NVE
> To: "Discussion list for GROMACS users" <gmx-users at gromacs.org>
> Message-ID: <20080403081240.1402.qmail at f4mail-234-118.rediffmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Dear David,
>
> I post the input file for the NVE run.
>
> I use SHAKE, Would it be better if I use LINCS?
>
> title                    = NVE
> cpp                      = /usr/bin/cpp
> ; RUN CONTROL PARAMETERS
> integrator               = md
> ; Start time and timestep in ps
> tinit                    = 500
> dt                       = 0.0015;
> nsteps                   = 360000; 540 ps
> ; For exact run continuation or redoing part of a run
> init_step                = 0
> ; number of steps for center of mass motion removal
> nstcomm                  = 1
> ; group(s) for center of mass motion removal
> comm-grps                = SOL
> ; OUTPUT CONTROL OPTIONS
> ; Output frequency for coords (x), velocities (v) and forces (f)
> nstxout                  = 0
> nstvout                  = 0
> nstfout                  = 0
> ; Checkpointing helps you continue after crashes
> nstcheckpoint            = 1000
> ; Output frequency for energies to log file and energy file
> nstlog                   = 100000
> nstenergy                = 1000
> ; Output frequency and precision for xtc file
> nstxtcout                = 0
> xtc-precision            = 1000
> ; select multiple groups. By default all atoms will be written.
> ; NEIGHBORSEARCHING PARAMETERS
> nstlist                  = 10
> ; ns algorithm (simple or grid)
> ns_type                  = grid
> ; Periodic boundary conditions:
> pbc                      = xyz
> ; nblist cut-off
> rlist                    = 1.0
> ; OPTIONS FOR ELECTROSTATICS AND VDW
> ; Method for doing electrostatics
> coulombtype              = pme
> rcoulomb                 = 1.0
> ewald_rtol               = 1e-5  ; since erfc(sigma*rcutoff) 
> =ewald_rtol
> optimize_fft             = yes
> ; Relative dielectric constant for the medium
> epsilon_r                = 1  ; for water
> ; Method for doing Van der Waals
> vdw-type                 = cut-off
> rvdw                     = 1.0
> ; Spacing for the PME/PPPM FFT grid
> fourierspacing           = 0.12 ;
> ; EWALD/PME/PPPM parameters
> pme_order                = 4
> ewald_geometry           = 3d
> epsilon_surface          = 0
> ; OPTIONS FOR WEAK COUPLING ALGORITHMS
> ; Temperature coupling
> Tcoupl                   = no
> Pcoupl                   = no
> ; GENERATE VELOCITIES FOR STARTUP RUN
> gen_vel                  = no
> gen-temp                 = 300
> gen-seed                 = 173529
> ; OPTIONS FOR BONDS
> constraints              = all-bonds
> constraint_algorithm     = shake
> shake_tol                = 0.0001
> morse                    = no
>
>
> rgds,
> Jes
>
>
>
> On Fri, 28 Mar 2008 David van der Spoel wrote :
>> JMandumpal wrote:
>>> Dear list,
>>>
>>>         These are the energy I obtained running 60ps NVE of 500  
>>> TIP5P water molecules- I still wonder why the energy has not been  
>>> converged even having run for 160 ps in total (100 NVT + 60 NVE).
>>>
>>>
>> it is difficult to run NVE.
>> what constraint settings etc. do you use and what cutoffs.
>>
>>>   100.000000  -15887.623027
>>>   101.500000  -15887.961065
>>>   103.000000  -15894.289055
>>>   104.500000  -15898.915859
>>>   106.000000  -15903.562196
>>>   107.500000  -15906.540627
>>>   109.000000  -15912.662455
>>>   110.500000  -15915.571468
>>>   112.000000  -15917.889520
>>>   113.500000  -15923.955612
>>>   115.000000  -15929.135491
>>>   116.500000  -15932.507824
>>>   118.000000  -15937.882055
>>>   119.500000  -15941.562218
>>>   121.000000  -15945.644608
>>>   122.500000  -15949.063632
>>>   124.000000  -15953.389791
>>>   125.500000  -15957.524802
>>>   127.000000  -15962.299775
>>>   128.500000  -15967.579640
>>>   130.000000  -15972.576889
>>>   131.500000  -15975.427574
>>>   133.000000  -15980.852975
>>>   134.500000  -15985.584201
>>>   136.000000  -15988.488715
>>>   137.500000  -15993.295989
>>>   139.000000  -15997.867767
>>>   140.500000  -16000.630051
>>>   142.000000  -16005.100451
>>>   143.500000  -16009.649201
>>>   145.000000  -16014.842764
>>>   146.500000  -16019.498233
>>>   148.000000  -16022.975477
>>>   149.500000  -16027.677441
>>>   151.000000  -16032.965475
>>>   152.500000  -16036.043976
>>>   154.000000  -16039.844112
>>>   155.500000  -16044.836132
>>>   157.000000  -16047.992186
>>>   158.500000  -16052.196926
>>>   160.000000  -16056.946593
>>>
>>>
>>> -----------------------
>>> I corrected these variables to:
>>> tinit                    = 100
>>>
>>> init_step                = 0
>>>
>>> Where could be the problem, then? Or should I run for even longer?
>>>
>>> regards,
>>> Jes
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: http://www.gromacs.org/pipermail/gmx-users/att