Dear all,
my problem is, that I have changed my input-script, which based on
Kremer and Grest. I would like to define some new bonds to create
something like a loop. First I used a chain of 11 atoms and 10 bonds and
9 angles. And I only used 1 bond-style fene. Now, I have added 1 bond
and 1 bond-type, and give this bond another style (harmonic). Before I
have done this, LAMMPS worked. Now it crashes and tells me that bond
atoms 10 11 are missing. Do I have to change some other parameters. In
the data-file I have changed the number of bonds and bond-types. So I
thought I have changed everything.
Thank you and best regards.
Ulrike
bead_spring.lam.ex7 (4.17 KB)
chain7.N10.input (1.02 KB)
Please read the Section_errors.html of the manual
for your error message. You have given LAMMPS bad
input for a system that blows up. There is nothing
LAMMPS can do about it - you have to fix the problem.
Steve
Dear all,
thank you for answering my question. I read the error section of the
manual. But at this point is no information, how I could avoid this
problem?I don't understand, why this happens. I only have defined a new
bond. Before this, it has worked. So what do I have to do, to make it
work? Do I have to change the cutoff of the pair-style, or what else?
The reason, why I have to define a new bond is, that I would like to
model a chain of atoms and define some looping. I thought this would be
possible, if I define a new bond and use the bond_style hybrid with the
fene-potential for short bonds and the harmonic-potential for long
bonds. Was this a bad idea? Do I have to change something else?
Thank you for helping.
Bye
Ulrike
bead_spring.lam.ex7 (4.17 KB)
chain7.N10.input (1.02 KB)
The problem is that your system lilkely runs
with bad dynamics and atoms move too far
and bonds stretch too far and the system blows up.
Just because you "add a bond" doesn't mean
you did it in a reasonable way. Check the dynamics
and thermodynamics of your system and find
out what is going wrong.
Steve