[gmx-users] error about x of the xxx bonded interactions could not be calculated ....

[Chang Woon Jang]

Dear Gromacs Users,

   I am simulating coarse grained polymer systems with tabulated potential
(A B C D types).

To obtain pair potentials, Iterative Boltzmann Inversion is used with
Gromacs.

After 14 step (each step runs 1 ns (1000000) with time step 1 fs - 0.001),
the simulation had failed with the following error.

x of xxx bonded interactions could not be calculated ....

It sounds like Blowing Up problem. Therefore, I reduced the time step to
0.0008 (0.8 fs), then the simulation failed at 164 step.

I do not think that I started 1) a bad structure due to I equilibrated for
120 ns at atomistic level, 2) inappropriate pressure coupling because I use
NVT ensemble.

I am not sure what the following tip is in the gromacs website.

"*your position restraints are to coordinates too different from those
present in the system,*"


I am using gromacs 5.0.8-dev-20151014-1f04b58

Do you have any advice for solving this problem?

I still keep decreasing the time step now at 0.0006.

Thank you.


Best regards,
Changwoon Jang,