[gmx-users] 3xRe: Impact velocity

[David van der Spoel]

Janne Hirvi wrote:
> Hello David and other gmx-experts!
> 
> I havent yet been able to extract thermal motion contribution for temperature
> (without translation) even I tested trjconv with fit (translation) option as
> you adviced. Actually, I was able to remove the center of mass translation of
> the droplet (tested with g_traj -ox) but however analyze still gives the center
> of mass velocity which contributes to temperature (g_traj with -ov and -ot)? 

Maybe trjconv does not modify the center of mass velocity.

What you can do is from your original trajectory calculate the center of 
mass temperature with g_traj and subtract that from the normal T. You 
may have to be careful with degrees of freedom.

> 
> I am wondering how thermal motion contribution for temperature is extracted by
> g_energy in case of acceleration combined with temperature coupling, while
> g_traj seems to include also translational motion. At least g_traj temperature
> is higher and almost equal to reference temperature + temperature increase due
> to translational energy(0.5*m(total)*v(COM)^2)). So, I think g_traj temperature
> is calculated directly from the velocities
> (0.5*SUM(m(i)*v(i)^2)=0.5*N(df)*k*T), but are (1) these velocities scaled to
> remove the center of mass velocity to get temperature given by g_energy or is
> (2) translational energy contribution directly substracted from the total
> kinetic energy? I tested the latter method and got pretty good agreement, but
> not perfect, between g_energy and g_traj temperatures but is it the exact
> method? 

g_energy does not do any processing of data, just reading the energy 
file and printing it.
g_traj calculates T using above equation.

> 
> 
> Thanks for your time,
> 
> Janne
> 
> 
>> Janne Hirvi wrote:
>>> Thanks for your quick replies David and Bob!
>>>
>>> I was icefishing all the weekend and wasnt able to comment on your
>> suggestions
>>> earlier but you are absolutely right that I should use NVE-ensemble and
>> not
>>> count overall center of mass (COM) velocity to temperature. 
>>>
>>> NVT-ensemble was just used on testing purposes so that I could understand
>>> something what was happening but actual problem conserned temperature and
>>> velocities. If I generate extra velocity to the center of mass of the
>> droplet
>>> (in equilibrium at 300K) either by my own code or using accelerating
>> force,
>>> temperature analyze programs in Gromacs (g_energy and g_traj) show that
>>> temperature is much higher due to translation of COM. However, it would be
>> nice
>>> to somehow detect temperature due to only termal motions (without
>> translation)
>>> so that I could observe possible heat transfer in the collision and make
>> sure
>>> that there is no energy drift in NVE ensemble for other reasons (There will
>> be
>>> some problems while I have used plain cut-offs for LJ-interactions in
>> earlier
>>> NVT-simulations without impact velocity in other words just placing the
>> droplet
>>> on a surface). 
>>>
>>> Any comments or is it the best option to extract COM velocity with g_traj
>> and
>>> just make own code which substracts this translational temperature from
>> total
>>> temperature?
>>>
>>> Thanks,
>>>
>>> Janne
>>>
>>>
>>>> If you wanted to study a collision, why do it at constant temperature?
>>>> Like David said, maybe you should equilibrate the droplet at 300K, add
>>>> a center of mass velocity to the droplet, and then study the collision
>>>> at constant energy. The real collision probably doesn't take place at
>>>> constant temperature anyway. The surface probably heats up a little
>>>> upon colliding with the water droplet. Anyway, Just my thoughts...
>>>> Bob
>>>>
>>>> On 2/16/07, Janne Hirvi <janne.hirvi at joensuu.fi> wrote:
>>>>> Hello!
>>>>>
>>>>> I am trying to generate center of mass velocity to a water droplet so
>> that
>>>> I
>>>>> could study the effect of impact velocity in the collision with a solid
>>>>> surface.
>>>>>
>>>>> First, I simply tried to generate extra velocity and compensate the
>>>> increase in
>>>>> temperature by decreasing other velocity components. However at the
>>>> beginning
>>>>> of the collision simulation temperature increased by value which
>>>> corresponds
>>>>> approximately to the temperature increase due to extra velocity. This is
>>>>> probably due to poor equilibrium when other velocities are scaled.
>>>> Temperature
>>>>> coupling will equilibriate temperature but at the same time generated
>>>> center of
>>>>> mass velocity decreases.
>>>>>
>>>>> Then I tried to to accelerate a single droplet in vacuum with
>> temperature
>>>>> coupling so that I could extract "equilibrium" droplets in 300K with
>>>> wanted
>>>>> velocities and tranfer these droplets on the studied surfaces. However I
>>>> still
>>>>> have a problem with temperature. When I use g_energy, droplet
>> temperature
>>>> is
>>>>> wanted 300K but g_traj informs that temperature is much higher when I
>> take
>>>> into
>>>>> account constrained freedoms by multiplying with 1.5 (rigid SPC).
>>>> Temperature
>>>>> given by g_traj actually equals to temperature given by g_energy when I
>>>>> transfer the droplet on a surface for collision simulation and remove
>>>>> accelaration.
>>>>>
>>>>> I am not sure what is happening and would be pleased if I get some
>>>> clarification
>>>>> and maybe other suggestions how to generate center of mass velocity for
>>>>> specific temperature.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Janne
>> You can take out the overall motion by trjconv using a fit.
>>
>> -- 
>> David.
>> ________________________________________________________________________
>> David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
>> Dept. of Cell and Molecular Biology, Uppsala University.
>> Husargatan 3, Box 596,  	75124 Uppsala, Sweden
>> phone:	46 18 471 4205		fax: 46 18 511 755
>> spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php


-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++