Fix bonds fail

Dear community,
I try to figure out which maximum time step + fix bonds setup I can apply to my polymer molecular dynamics in Forcite. At some point (time step = 4 fs + fix bonds) the calculation set up dies with following error output, std.err: "dynamics calculation - Constraint solver failed to converge (function/property "Compute") at -e line 25.", resp. .txt: "Dynamics calculation - Constraint solver failed to converge". Do you know how to solve this problem? FYI: The xsd-document contains round about 20 000 atoms (a few MMA chains). Is it a too big model for the Constraint solver? Is there any chance for an efficient workaround?

Thanks in advance!

Best regards

Attached you will find:
the set-up of the calculation: .txt
And a picture of the MatStudio log

Answers 2
RA
RA 2023-05-04

Hi Nils,


The error means that the Rattle algorithm failed to meet the tolerance for length (1E-4 Å) and/or velocity (1E-6 Å/ps) within the maximum number of iterations (250).


I would check which bonds are fixed at 4 fs (by pressing the Calculate button on the Forcite Dynamics dialog). For example, constraining all bonds in a ring can be difficult to converge.


Also check that multiple time stepping is switched off. You may find that multiple time stepping with a default time step gives better performance than fixing bonds.


Regards,
Reinier

ND
ND 2023-05-09

Dear Reinier,

thank you very much for your fast answer and the multiple time stepping hint, it works fine for me!

Regards,
Nils