Pressure scaling more than 1%

User discussions
1

Dear all,

I encountered this error:
Warning: Pressure scaling more than 1%. This may mean your system
is not yet equilibrated. Use of Parrinello-Rahman pressure coupling during
equilibration can lead to simulation instability, and is discouraged.

I saw gromacs explanation to solve this error.

But my question is that, does this error mostly assigned to NPT or prvious steps may bring about this error?

I apperciate your responses in advance,
Best

2

Without knowing what you’ve done to prepare the system, it’s hard to say specifically, but the Parrinello-Rahman barostat is prone to large swings in systems that are not adequately equilibrated already so you may want to try using the Berendsen method for the barostat first, then switching to P-R afterwards.

3

npt.mdp (2.7 KB)

Dear Prof. Lemkul,
Thanks for your responses.

Will the npt.mdp that I enclosed work?
I am performing the simulation of DNA-ligand in water.
And, could you please explain how to switch from Berendsen to Par_R?
I know only how to use a npt.mdp.

I appreciate your help.
Best

4

There is an option “pcoupl = Berendsen” in your npt.mdp, you can use this file to equilebrated the system first, then switching to “pcoupl = Parrinello-Rahman”

5

Excuse me, in the 2023 version of gromacs, I first used pcoupl = Berendsen to simulate 100ps in the pre-balance, and then when I used pcoupl = Parrinello-Rahman, an error still occurred: pressure scaling more than 1%. This may mean your system is not yet equilibrated. Use of Parrinello-Rahman pressure coupling during equilibration can lead to simulation instability, and is discouraged. Is this related to the 2023 version of gromacs I use?

7

Equilibrating for 100ps with the Berendsen barostat might not be enough. It depends on the system. However, after the c-rescale barostat was added, it is recommended to use that both for equilibration and production simulations.