I'm using "pair_style lj/cut/coul/long/tip4p 1 2 1 1 0.1250 0.0 10.0" with "compute pair test reduce ave fx fy fx". The LJ cutoff is set to 0.0 so I can see just the electrostatic contribution to the force. Group test is just a single charged particle; not part of a tip4p molecule. The compute is showing near zero forces (~2e-314). I know the tip4p pair style doesn't support some of the other force calculation computes - like group/group. Is compute reduce not getting the forces from pppm_tip4p?
I'm not clear on what your're doing. But if your
compute reduce is simply getting fx,fy,fz on
a single particle, then you should be able
to check if it is doing the right thing, by also
dumping the forces on that particle into
a dump file and checking the force values