Problem solved - fit npt problem, volume not changing correctly

Dear Axel, Andrew, Steve, Ray,

Just want to update you that we finally figured out what the problem was:

there is a conflict between ‘minimize’ and ‘fix shake’, which has been clearly descried at the minimize document but we did not pay enough attention to.

http://lammps.sandia.gov/doc/minimize.html

Two solutions:

a. have a reasonable initial simulation box and skip the minimization;

b. add ‘run 0’ in the control file after assigning the initial velocity, and then re-assign the velocity. This seems to solve the conflict between ‘minimize’ and ‘fix shake’. See the lines below

velocity sds create 300.0 293288 dist gaussian mom yes rot yes
velocity water create 300.0 293288 dist gaussian mom yes rot yes

run 0
velocity sds create 300.0 293288 dist gaussian mom yes rot yes
velocity water create 300.0 293288 dist gaussian mom yes rot yes

minimize 1.0e-4 1.0e-6 100 1000

min_style cg

Both methods are working for our system. We do not have to change anything else.

This is a good lesson on two: post a complete input control file & read the command documentation again and again…

Many thanks for your time!

Paul