[JMDag2004]

Something strange happened when I restarted a simulation while using reactingFoam. I don't believe it had anything to do with the solver being used as I understand "time-step continuity errors" are more relevant to the PISO algorithm. I am building a combustion simulation and everything was running perfect. The simulation has a rather refined mesh and due to the nature of the simulation I have to maintain a very small time step CFL < 0.1. This obviously necessitates parallel processing. When I complete a quality simulation I will eventually need to run a mesh convergence study, which will take the needed computing resources beyond my laptop. The next step will be to use a high powered computing center, TACC if you are familiar. I have a hyperthreaded quad core and run in parallel using 8 processors. After running for 8 hours I decided to give my laptop a break and used the "writeNow" functionality looking to pick up the simulation the following day. Before stopping the simulation,the time step errs were extremely low. The overall err was on the order of 10^-11, residuals looked good, temp/pressure/vel/enthalpy/specie mass fractions all looked good before stopping the simulation. Upon restart, the overall time step err jumped to roughly 0.01 while flow variables looked the same. After two more hours or so the min temp dropped to 200 without tripping the janaf temp warning. I wanted to see if anyone had experienced anything like this before because I will almost assuredly have to stop and start the simulation multiple times using TACC time allocation and wanted to see if there is something I might be able to do to prevent these problems from happening again. I stopped the simulation again because I haven't been able to look closely at flow variables after the temp dropped to see what kind of impact this has had.

[Howard]

Originally Posted by JMDag2004

Something strange happened when I restarted a simulation while using reactingFoam. I don't believe it had anything to do with the solver being used as I understand "time-step continuity errors" are more relevant to the PISO algorithm. I am building a combustion simulation and everything was running perfect. The simulation has a rather refined mesh and due to the nature of the simulation I have to maintain a very small time step CFL < 0.1. This obviously necessitates parallel processing. When I complete a quality simulation I will eventually need to run a mesh convergence study, which will take the needed computing resources beyond my laptop. The next step will be to use a high powered computing center, TACC if you are familiar. I have a hyperthreaded quad core and run in parallel using 8 processors. After running for 8 hours I decided to give my laptop a break and used the "writeNow" functionality looking to pick up the simulation the following day. Before stopping the simulation,the time step errs were extremely low. The overall err was on the order of 10^-11, residuals looked good, temp/pressure/vel/enthalpy/specie mass fractions all looked good before stopping the simulation. Upon restart, the overall time step err jumped to roughly 0.01 while flow variables looked the same. After two more hours or so the min temp dropped to 200 without tripping the janaf temp warning. I wanted to see if anyone had experienced anything like this before because I will almost assuredly have to stop and start the simulation multiple times using TACC time allocation and wanted to see if there is something I might be able to do to prevent these problems from happening again. I stopped the simulation again because I haven't been able to look closely at flow variables after the temp dropped to see what kind of impact this has had.

Have u solved the problem?