[gmx-users] Big problems with constrained pull simulations

[Semen Esilevsky]

Dear All,
Could you help me to understand the strange results of
the constrained pull simulations.
I want to measure the force acting on the particular
ion from the protein along z-axis. What I did is the
normal pos-restrained simulation with the pull code
added.
The ion was defined as a pull group 1 and the protein
as a reference group. Pull type was set to
"constrained" along 0.0 0.0 1.0.
1) The first strange thing: the file bla-bla-bla.pdo
determined as output is ignored and the output is
written to default pull.pdo.
2) When I had a look at the force I was shocked:
instead of fluctuating but well directed attraction to
the protein I have a qusi-periodical noisy sinusoidal
line oscillating around zero with large period (~10-20
ps). I tried to remove pos. restrains - nothing. Tried
to move the ion (the force HAVE TO depend on its
position dramatically) - again nothing, the same wild
sinusoidal line...

Do somebody know what's wrong? How to make the pull
code work correctly (or probably it is not working at
all)?

Sincerely,
Semen

__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com