[cabinetcooler]

Hello colorful people of CFD Forum!

I have been struggling with an analysis recently.

I have a fully enclosed cabinet with heat boxes inside. They all have different heat dissipation rates. Total load is defined and known. I put two heat exchanger inside of it to cool the system and make sure the air circulates inside the cabinet. I have my total load; my heat exchanger attributes defined as porous media and fans with necessary data within.

What I was struggling is that; any time I run analysis; I cannot get the stabilized solution for inside temperature of enclosed system. They just don't converge.

I defined my heat exchangers as negative heat load; But I am not really certain with the correct value. I have 850 W heat load and I defined -600 W load for each exchanger. Flow part is not problematic. It converges earlier iterations but temperatures just don't.

What am I doing wrong? My approach is correct or? Can you give me a hand with this problem?

Thanks in advance.

Cheers,

AC

[Boris_M]

Hi CabinetCooler,

I like the colorful people part :-)

Ok, so if I understand you right, you have a cabinet with a heat source somewhere of 850W and two heat exchanger you use a negative heat source of each -600W?
Plus a fan to move the air.

Is this fully internal without any outflow of air?
Is this adiabatic or do you have any heat transfer to the outside?
So I assume it is an internal fan, correct?

I'm just wondering about the heat loads if you have less heat generated than you use to cool it and the flow is only internal without any exchange of air to the outside via an opening, then you are bound to have a problem.
There is a heat loss of -350W that is cooling something. If you have a heat transfer rate on the outside or an airflow coming in, then the heat is taken from there as well.

It might depend on the configuration of the flow in the geometry that it is unstable but it is hard to tell.
You could try to use the results as an initial condition for a transient simulation to see if the phenomena are just due to the steady state settings but in transient, it will smoothen. Some phenomena are simply not steady state and can cause issues with convergence.

Hope that helps,
Boris

[evrenykn]

Hi cabinetcooler,

If you have still come across the problem, can we get clear heat loads from you to help?

Best Regards,
Evren