Hi -
In addition to having a larger cut-off (8 or -even better- 9 Angstroms)
you should also increase the shake tolerance to tol = 0.0000005.
Together with an 1fs integration step, this will probably lead to an
approximate temperature drift of 1K per 100ns.
best,
George
On 12/22/2014 01:04 AM, Daniel Roe wrote:
> Hi,
>
> Constant energy simulations will inevitably experience some drift due
> to things like round-off errors etc over the course of the simulation
> (it shouldn't be very significant though, probably less than 1 K over
> your simulation time). I think the issue in your case is that a cutoff
> of 6.0 is way too small. What are the dimensions of your box?
>
> -Dan
>
> On Sun, Dec 21, 2014 at 10:48 PM, Robin Jain <robinjain.chem.gmail.com> wrote:
>> Dear all,
>> I have equilibrated my MD simulation using NPT. Now for the calculation i
>> used NVE ensemble , for this i used following input -
>>
>> &cntrl
>> imin=0
>> irest=1,
>> ntx=7,
>> ntb=1,
>> cut=6.0,
>> ntr=0.
>> ntc=2,
>> ntf=2,
>> ntt=0,
>> ntp=0,
>> nstlim=4000000,
>> dt=0.0005,
>> ntpr=1000,
>> ntwx=1000,
>> ntwr=1000, ntwe=1000,
>> ioutfm=1
>>
>> after run, i check the energy and temperature graph and it show that the
>> energy and temperature are decreasing as the md job proceeding. Now i want
>> to know that what is wrong with this and what are the possible solution for
>> this bug.
>> Thanking you in advance.
>>
>>
>>
>> --
>> Robin Jain
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Dec 22 2014 - 09:00:02 PST
