search

i-pi / i-pi

i-PI: a universal force engine
https://ipi-code.org/
221 stars 116 forks source link

Unphysical Temperature Output in Partially-Adiabatic CMD Simulation #336

Open jorgecastro316 opened 4 months ago

jorgecastro316 commented 4 months ago

I am encountering an issue with the output temperature of the centroid normal mode in my partially-adiabatic CMD. The simulation setup utilizes a two-temperature PILE-G thermostat. While the overall simulation runs as expected and the results seem physically reasonable, when the fixcom flag is set to True, the temperature of the centroid normal mode, as obtained by the get_temp function in properties.py, is unphysically high, which seems inconsistent with the rest of the simulation results.

venkatkapil24 commented 4 months ago

is this an issue only with partially adiabatic simulations? Can you check it for standard PIMD? What system are you running?

jorgecastro316 commented 4 months ago

No. It also happens for the centroid mode temperature whenever I use PILE-G thermostat, in single or double temperature mode. I am running MAPI and Carboxilic Acid Fluoride

venkatkapil24 commented 4 months ago

it would be useful t have more info e.g. the centroid and higher normal mode temperatures (expected compared to observed).

mahrossi commented 1 week ago

Hi @jorgecastro316 , could you check whether the problem persists after the latest cleanup? (latest version of i-PI). If so, can you share inputs/outputs from completeness?