this isn't the right way to do it, there is no real control over how
long the heating takes. The Langevin thermostat gets things up to
temp0 almost immediately. This may indeed be what you want, I just
meant to say that if you want to "heat" a system (as part of an
equilibration procedure), this isn't the way to do it. Instead, you
should use the weight change section to gradually modify the bath
temperature (temp0) over time. See the tutorials for more details.
On Fri, Jun 29, 2012 at 7:43 AM, Vishal Kumar Jaiswal
<vishal3990.gmail.com> wrote:
> I meant gradual heating not constant heating
>
> On Fri, Jun 29, 2012 at 1:42 PM, Vishal Kumar Jaiswal
> <vishal3990.gmail.com>wrote:
>
>> Carlos,
>> I think the options in Brandon's input file WILL perform a constant
>> heating . I did a similar simulation ( heating from 0K to 300K with the
>> same inputs and the temp reached to 300k in about 5.4 picoseconds )
>>
>> 0.000 0.00
>> 0.200 62.83
>> 0.400 105.09
>> 0.600 137.55
>> 0.800 164.46
>> 1.000 182.86
>> 1.200 205.31
>> 1.400 224.02
>> 1.600 228.46
>> 1.800 241.73
>> 2.000 253.39
>> 2.200 262.48
>> 2.400 260.94
>> 2.600 269.21
>> 2.800 266.20
>> 3.000 276.66
>> 3.200 284.18
>> 3.400 283.72
>> 3.600 283.19
>> 3.800 292.74
>> 4.000 287.69
>> 4.200 291.04
>> 4.400 290.79
>> 4.600 291.18
>> 4.800 289.07
>> 5.000 299.28
>> 5.200 296.97
>> 5.400 301.01
>>
>>
>> Vishal
>>
>>
>>
>> On Fri, Jun 29, 2012 at 12:42 PM, Carlos Simmerling <
>> carlos.simmerling.gmail.com> wrote:
>>
>>> seems ok- but be aware that this is NOT going to be gradual heating.
>>> you've assigned a thermostate of 300K, which means the T will go to
>>> 300K very quickly. if you want to heat 0-300K, you need to slowly
>>> change temp0. tempi just sets initial temperature and the thermostat
>>> kicks in immediately.
>>>
>>> can you paste the end of the sander mdout- from the last energy report
>>> to the end?
>>>
>>> On Fri, Jun 29, 2012 at 6:39 AM, Brandon Sim <simpavid.gmail.com> wrote:
>>> > Initial MD equilibration: 20 ps, at constant volume, 0-300 K
>>> > &cntrl
>>> > imin=0,
>>> > irest=0,
>>> > ntx=1,
>>> > igb=0,
>>> > scee=1.2,
>>> > ntb=1,
>>> > cut=10,
>>> > nstlim=10000,
>>> > dt=0.002,
>>> > ntc=2,
>>> > ntf=2,
>>> > ntt=3,
>>> > gamma_ln=1.0,
>>> > tempi=0.0,
>>> > temp0=300.0
>>> > ntpr=100,
>>> > ntwx=100,
>>> > /
>>> >
>>> > On Fri, Jun 29, 2012 at 6:24 AM, Carlos Simmerling <
>>> > carlos.simmerling.gmail.com> wrote:
>>> >
>>> >> can you reply with a copy/paste of your input file?
>>> >>
>>> >> On Fri, Jun 29, 2012 at 6:14 AM, Brandon Sim <simpavid.gmail.com>
>>> wrote:
>>> >> > Dear Vishal,
>>> >> >
>>> >> > Sorry - I mistyped. I meant 7 picoseconds (i have timestep of 0.002
>>> >> > picoseconds with SHAKE on).
>>> >> >
>>> >> > Has this problem occurred before to anyone?
>>> >> >
>>> >> > By the way, the .mdcrd file wrote was written out in the appropriate
>>> >> > format, with box size at the end, anyway..
>>> >> >
>>> >> > On Fri, Jun 29, 2012 at 5:46 AM, Vishal Kumar Jaiswal
>>> >> > <vishal3990.gmail.com>wrote:
>>> >> >
>>> >> >> Hi Brandon
>>> >> >>
>>> >> >> No this is not the expected behavior ,the simulation should
>>> complete the
>>> >> >> number of steps ( nstlim)
>>> >> >> By the way, with 3500 steps you are covering 7 nanoseconds which
>>> >> means a
>>> >> >> step size of 2 picoseconds. i do not think it is acceptable.
>>> >> >> you should have timestep of 0.002 picoseconds ( if you restrict
>>> hydrogen
>>> >> >> movement by SHAKE , NTC = 2) or else use timestep of 0.001
>>> picosecond.
>>> >> >>
>>> >> >> Hope this helps
>>> >> >>
>>> >> >> Vishal
>>> >> >>
>>> >> >> On Fri, Jun 29, 2012 at 11:17 AM, Brandon Sim <simpavid.gmail.com>
>>> >> wrote:
>>> >> >>
>>> >> >> > Dear Amber users,
>>> >> >> >
>>> >> >> > Does sander stop equilibration when temp0 is reached, even when
>>> the
>>> >> >> number
>>> >> >> > of steps has not gotten to nstlim?
>>> >> >> >
>>> >> >> > I ran three simulations with nstlim=10000, but all three ended at
>>> step
>>> >> >> > 3500, resulting in 7 nanoseconds of trajectory and 35 frames (I
>>> wrote
>>> >> a
>>> >> >> > frame each 100 steps).
>>> >> >> >
>>> >> >> > The temperature had already gotten to ~300 K , which was my
>>> temp0. Is
>>> >> >> this
>>> >> >> > expected behavior, is my script wrong somewhere?
>>> >> >> >
>>> >> >> > Thanks,
>>> >> >> > Brandon
>>> >> >> > _______________________________________________
>>> >> >> > 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
>>> >> >>
>>> >> > _______________________________________________
>>> >> > 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
>>> >>
>>> > _______________________________________________
>>> > 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
>>>
>>
>>
> _______________________________________________
> 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 Fri Jun 29 2012 - 05:00:03 PDT
