[KsaMe]

Hi everyone,

I want to analize densities variations of a gas that is inside a chamber which has 2 rectangle hot plates (see picture 1). The plates are at a temperature of 1200K and the walls of the chamber are considered perfect insulators (heat flux = 0 as boundary condition). The gas inside is a mixture of 95% argon and 5% molecular hydrogen at an initial temperature of 300K.

I expect the density in the region between the plates decrease at first and then return to initial value. This because initially the air between the plates get rapidly hot (1200K) so this air has more temperature and more pressure than the air outside this region (surrounding regions). Due to the pressure difference this gas will flow out, carrying particles to the outside, and then decreasing its density. On the other hand, the surrounding region , which is cooler, receive this particles increasing its density and pressure. This process continues until the pressure of both regions get equal. Nonetheless there are heat conduction through the air, so the surrounding regions also get hotter with time, and if one wait a certain time it will reach 1200K. So now both regions has equal temperature, BUT the surrounding has more density, and hence more pressure. So now the flow will be into the space between the plates to restore the initial density.

When I did the simulation I considered the gas as an incompressible-ideal-gas, and was a mistake because it doesn't allow accumulation of particles, and I expect that initially particles will be acumulated in the surrounding regions of the plates. In fact it showed that the density of the chamber went down with time, clearly violating conservation of particles (see pictures 2 and 3). So then I changed the model of the gas to "Ideal-gas" and unfortunately the solution didn't converge (see picture 4).

I would really appreciate if someone could help me with some insight to solve this problem!

Best regards,

KSaMe