Simulating a cooling coil

[DRoam]

I'm looking for some help simulating the effects of a cooling coil that we have in our system. I have two questions:

1) How do I model the cooling coil volume (a rectangular block) as both a flow resistance AND a heat exchanger?

As far as I know I can only assign one of those two materials to the heat exchanger volume.

2) How do I set up the heat exchanger material?

I have the datasheet for our coils which includes the cooling-fluid flowrate and the cooling-fluid entering temperature. I entered the first of those in the Properties > Flow > Value field of the material editor. I entered the second under the Properties > Heat Transfer > Temperature field. Am I correct so far?

The Heat Capacity field is where I'm most confused. Our cooling coil datasheet has a rating for "Capacity" which says "187,853 BTU/hr". These are not the same units as what CFD is asking for (BTU/hr/°R). I thought possibly our coil manufacturer had listed the cooling capacity at the max rated temperature difference between the incoming cooling-fluid and incoming hot air. So I divided 187,853 BTU/hr by the max rated temperature difference (85 °R), and came out with 2153 BTU/hr/°R. I entered this value in the "Heat Capacity" field for my material, and then tried to run my simulation. But I got this error in the message window:

"Error - physically unrealistic heat exchanger device (Property ID 1). Heat Capacity 0.598569 exceeds (MDot*Cp)hot 0.00100364"

What does this mean? Am I misinterpreting what the "Heat Capacity" field is asking for? Or did our coil supplier give us a crazy cooling capacity? I don't have a sense of what is a reasonable cooling capacity for a cooling coil, so it's possible our supplier made a typo and I'm using the correct property but an impractical value.

Any help is very appreciated. Thanks!

[Jon.Wilde]

Hi,

Let me try to help:

1. It depends what you want to assign. More often than not, we can use a resistance material with a negative heat generation applied to the volume
2. Have you seen the help pages about heat exchangers and how they need to be assigned? They really need to be outside of the model, or surrounded by a suppressed volume so that CFD thinks they are

Does that help?

Thanks,

Jon

[DRoam · Accepted solution]

Jon, thank you for the reply. Unfortunately, we need to use a heat exchanger because we need the negative heat generation to be a function of the difference between the working fluid temperature and the tube-side fluid temperature--like the Heat Exchanger material does.

Also, the Heat Exchanger really needs to be internal to the flow volumes, because we're driving our flow through it (and the rest of our system) using a centrifugal fan. Which brings up another problem, namely that the Heat Exchanger material also requires specifying a constant flow rate for the Heat Exchanger to push the fluid at. It can't be used passively.

The above issues prompted me to create a request for a passive heat exchanger in the CFD IdeaStation. The request includes the ability to control how the heat exchanger responds to variations in flow rate. This way, we can have a true passive Heat Exchanger which generates/removes heat based on both temperature difference AND flow rate. The idea is located here: IdeaStation: Add a Passive Heat Exchanger. Anyone else who would like to have this functionality, please go the the page and give the Idea a Kudo.

In the meantime, we've had to settle for making a conservative assumption to simulate the cooling coil's behavior--namely, giving the outlet face of where the cooling coil is a temperature boundary condition of the temperature the coil is rated for. This is conservative because most of the time the coil will be taking in (and therefore outputting) air which is colder than what it's rated for. This is far from ideal but it at least allows us to compare our designs, which is our main goal at this stage.

[Jon.Wilde]

Thank you for the update, I voted your idea up! 🙂

I think that your solution is the best possible right now, nice work getting to that point!

Thanks,

Jon

[DRoam]

Thanks very much!