Calculate viscosity

Hello, I encountered the following problem while using the RNEMD method to calculate viscosity.There are strange points during the calculation process. Its dVx is significantly different from other points, which affects the overall viscosity calculation. Can you help me answer this?

Strange point
Step Temp v_visc v_dVx
1510000 903.4817 14.979769 0.15275676
1520000 906.6025 7.0123448 0.32627508
1530000 894.13296 -472.46546 -0.004842172
1540000 917.36048 5.4157615 0.42241344
1550000 900.49248 4.1208301 0.55501266
1560000 902.45437 8.3593848 0.27363739

Hi @Lty,

Not really. This is more like a science question than a LAMMPS related question. Did you try using other methods to compare your results? Did you look into the literature to see how other people do this kind of simulation? What are your parameters and do they fit your system? Does the methods fit your system? Have you looked at a trajectory of your simulation to see if nothing weird happened? Did you double check your input script? Did you check that the version you use is up to date as some related bugs might have been corrected?

These are all leads you should follow in order to pinpoint why you get a value different compared to what you would expect, and how to deal with it. If you have issues with the commands you use or think there is a bug in the code, it might be appropriate to elaborate further here. Else you might want to make a more detailed post in the “Science” category, but this part of the forum is unfortunately less active.

Thank you very much for your answer. The method should be applicable to my system. I tried to change the Nevery number in the fix viscosity command, but I didn’t get the desired result. Perhaps as you said, I can try using other methods to compare my results.