Compute temp/region gives lower temperature in regions comparison to the whole domian

I’m simulating water using SPCE model and I used regions to compute the spatial temp profile. the water box is thermostated by Noose hoover at 300 K. While a temperature of specific bin printed out via temp/region, it gives some temperature around 200 K or so, What is the problem with compute temp region?

no problem. this is to be expected. you are using shake
to keep your waters rigid. that removes three of the nine
degrees of freedom that you have with three atoms. check
out this segment of the documentation:

Thanks Alex, Is there any correlation between N (number for extra) and the number of molecules in the region/bin? Because I tried 3, -3 , there was no change to the temperature.

Thanks Alex, Is there any correlation between N (number for extra) and the
number of molecules in the region/bin? Because I tried 3, -3 , there was no
change to the temperature.

of *course* there is a correlation between the number of DOFs removed
and the number of molecules. i explained it in my last e-mail.

however, compute temp/region does care about molecules and the extra
parameter is for *total* degrees of freedom to be removed. since 1/3rd of
the DOFs is removed, why don't you just multiple the result by 1.5 and
be happy? think about this for a bit, look at the formula in the documentation
and pick up a statistical mechanics text book and re-read the part about
the relation between macroscopic properties like temperature and statistical
mechanical ensembles. ...and wait for the spark.

axel.