A common source of uncertainty in molecular dynamics is deciding when equilibration is finished. During NPT equilibration, pressure can fluctuate widely, so a noisy pressure trace does not automatically mean something has gone wrong—or that the system is ready for production. A more useful question is whether the system’s density has settled into a stable regime.
NPT means the number of particles, pressure, and temperature are controlled. This stage follows energy minimization and NVT equilibration: after reducing unfavorable contacts and stabilizing temperature, the goal is to let density adjust. Continuing too soon can leave density drifting; continuing indefinitely because pressure is not perfectly steady can waste time.
Look for a density plateau, not a perfectly flat line
In practice, inspect the density over simulation time and ask whether its running behavior has reached a plateau near the value appropriate for your system. Individual values may still fluctuate. The aim is not a motionless trace, but a stable regime that gives you enough confidence to move on.
The tutorial’s example reaches about 1030 kg/m3. That is close to the experimental value of 1000 kg/m3; for the SPC/E water model, the expected density is about 1008 kg/m3. These numbers are an illustration, not a universal target: interpret density in the context of your system and model.
Pressure remains useful to inspect, but do not treat every fluctuation as a verdict. The documentation notes that pressure can vary widely during equilibration or simulation. If density is still trending rather than settling, the system likely needs more equilibration time.
A practical check before production
- Start from the previous stage. Use the GRO file or batch project produced by minimization, NVT equilibration, or an earlier NPT run. The auto-fill control can select the previous successful project for you.
- Keep temperature coupling consistent. Reuse the temperature-coupling choices from NVT unless there is a clear reason to change them. If velocity generation is enabled, make sure its temperature is consistent too.
- Begin with a reasonable run, then assess. The tutorial suggests that 100 ps is typically a good start, while emphasizing that the required time depends on system size and contents. Treat this as a starting point, not a guaranteed endpoint.
- Use the plots to decide whether to continue. GROMACS Wizard generates pressure and density plots after the job finishes and results are loaded. If density has not stabilized, launch another NPT equilibration using the current NPT results as the input.
This iterative approach turns equilibration into a checkable decision rather than a guess: run, inspect the density trend, and continue if it has not reached a sufficiently stable regime. Once density is stable enough for your system, proceed to production molecular dynamics.
Running and reviewing the job
GROMACS Wizard can generate inputs, run locally, or launch a cloud computation. For a local run, progress is available in the Output window and through Local jobs. When the job finishes, you can import the full trajectory, only the last frame, or a selected range. If you cancel trajectory import, the results are not deleted and the plot can still be generated. Results are also available in the configured results folder; NPT run folders use the suffix _npt.
For most projects, the tutorial describes exponential-relaxation pressure coupling with the stochastic c-rescale term and a time constant of about 5 ps as a common choice. The advanced parameters can be adjusted when needed, and the default values can be restored with Reset. The right settings still depend on the project, so preserve the temperature-coupling consistency and use the density trend to evaluate the outcome.
For the complete interface walkthrough, see the GROMACS Wizard NPT equilibration tutorial.
Note: SAMSON and all SAMSON Extensions are free for non-commercial use. Get SAMSON.