[gmx-users] bond length deviation

[Luke Czapla]

Hi David,

  Your advice with handling Glucose with GLCA was very helpful.  We have a
problem with two simulations we are running.  One is a PDB file taken from
the protein databank of the crystal structure of our protein with glucose.
The other is a simple 4 peptide chain of glycine in a box with a Chloride
ion.  Both seem to get this fatal error about the Bond Length deviating by
more than half about 50-150ps into the simulation.  
	I was wondering how this is so with the harmonic potential and
using only the standard residues with ffgmx.  Maybe we have done something
to create this problem; we used a procedure similar to the gmxdemo file.

thank you,
Luke