[gmx-users] 2016.3 only gmx grompp error on NVT step for ions

[Sarai G]

Greetings and Salutations,

​
 1qs21-test_log
<https://drive.google.com/file/d/1ZcrtdKW3TJnjIY9OpsyLG4P3zEFqkZLj/view?usp=drive_web>
​​
 1qs21_nvt.mdp
<https://drive.google.com/file/d/1x3Mb5OswZ5UH76nkB6CkrNAIBzHTPMLX/view?usp=drive_web>
​​
 qs21_testNVT.mpg
<https://drive.google.com/file/d/1SR9PvY-QC5y0myzJi4C8hC4o9ocPPuok/view?usp=drive_web>
​I was trying to create a reference system of water, some QS21, and K
cations with all-bond constraints (which is a needed setting) only when I
got the following error at the NVT step after energy minimization was
successful;

WARNING 1 [file Case5-1.top, line 14]:
  There are atoms at both ends of an angle, connected by constraints and
  with masses that differ by more than a factor of 13. This means that
  there are likely dynamic modes that are only very weakly coupled. To
  ensure good equipartitioning, you need to either not use constraints on
  all bonds (but, if possible, only on bonds involving hydrogens) or use
  integrator = sd or decrease one or more tolerances:
  verlet-buffer-tolerance <= 0.0001, LINCS iterations >= 2, LINCS order >=
  4 or SHAKE tolerance <= 5e-06

I tried many things to fix it including remaking the system and going
through the force field files. Everything came out clean and the error only
appeared in GROMACS 2016.3.I used -maxwarn 1 to test how the system looked
in VMD, and the ion(s) were moving around fine. From what I have found,
there appears to be no problem causing this error.

Has anyone ran into the problem before?


-- 
*I fancy days of yore, quote the Raven nevermore.*