[edmund32]

Hi, I'm sorry if this is a stupid question, but what are good values of rho, nu, mu and sigma in constant/transportProperties to describe a confined plasma in a fusion reactor? I currently have them all set to one just to figure out what's going on but I'm not really sure how to modify these to increase the realism, I mean I've been looking at papers and found values of resistivity ~1e-5, viscosity ~1e-6 (Stanier 2013, Zhang 2019) but I'm not sure how to set them when the permeability is set to 1. Thanks!

[HPE]

'Good' in terms of? 'Good' for what? IMHO, the question needs further refinement.

[edmund32]

Yes, I'm sorry I'm fairly new to this, I'll spend some time reading papers and try to formulate this better, but what I mean is I've used the same parameters of permeability,density,conductivity and viscosity (all set equal to 1) to describe liquid metal flow in the Hartmann and Shercliff tutorials that I'm trying to use now to describe plasma, and I feel like I should vary them for this vastly different state of matter, but I'm not completely sure how, I mean, can I just set them to mu=1,rho=1,sigma=10^4,nu=10^-4 to include the effects of the greater conductivity and much smaller viscosity? I'm not really sure how to calculate these when the permeability and density are set to 1, and setting the permeability to that of vacuum and the density to fusion reactor values (10^18 m^-3) seems to break things.

[edmund32]

One thing I would like to ask: when I set the permeability to 4*pi*1e-7 kgms**-2A**-2, the kinematic viscosity to 10m**2s**-1, the density to 1e-6 kgm**-3 and the conductivity to 1e5 s**3A**2kg**-1m**-3, I have to set the timestep to 1e-10 just so the simulation doesn't become unstable. All of these are acceptable parameters for fusion plasma calculations, and the required timestep should be of the order of 10**-8s for an even finer mesh from the Alfven speed of this plasma, why is it so low? And why are the output pressures when I do manage to keep it running on the order of 1e8? Is it just because they're kinematic pressures so dividing by such a small density causes them to be huge? I get similarly large pressures when I start reducing these parameters to the orders of rho=1e-3,nu=1,mu=1.26e-3,sigma=1e2 and I'm wondering if it would be better to just keep these parameters set to unity.

[itaniOmar]

Originally Posted by edmund32

One thing I would like to ask: when I set the permeability to 4*pi*1e-7 kgms**-2A**-2, the kinematic viscosity to 10m**2s**-1, the density to 1e-6 kgm**-3 and the conductivity to 1e5 s**3A**2kg**-1m**-3, I have to set the timestep to 1e-10 just so the simulation doesn't become unstable. All of these are acceptable parameters for fusion plasma calculations, and the required timestep should be of the order of 10**-8s for an even finer mesh from the Alfven speed of this plasma, why is it so low? And why are the output pressures when I do manage to keep it running on the order of 1e8? Is it just because they're kinematic pressures so dividing by such a small density causes them to be huge? I get similarly large pressures when I start reducing these parameters to the orders of rho=1e-3,nu=1,mu=1.26e-3,sigma=1e2 and I'm wondering if it would be better to just keep these parameters set to unity.

Hello,
Did you ever find an answer to ur question because i am not able to understand this solver.Each time i change the transport properties from a unitary value to anything realisitc like for example the properties of liquid sodium it doesn't give me converging solution so it would be greatly appreciated if you can respond.