Battery Pack Cooling - No Heat Transfer in Analysis despite enabling it

[Anonymous]

Hello,

I am trying to model a battery pack with 50 cells. The cells are surrounded by air, which is stationary. The Air contacts a housing, which encases a fluid mixture of 50/50 water and ethylene glycol ("cooling fluid"). The cooling fluid enters the inlet at 12 mm/s, 15C and exits at 0 pressure. Ambient is 35C.

The batteries have 3W applied to each. I applied this as a 'total heat generation' of 150W and selected all 50 cells. Within the Physics tab, I enabled heat transfer, auto forced convection, radiation, and flow. When the analysis runs, there is no heat transfer at all. The only thing the solver is doing is finding the velocity magnitude of the cooling fluid. The cells aren't heating up in any way. Why not? I had no errors or alerts on startup. I modeled the air in CAD instead of using the built in external volume. Is this the issue? I'm just confused as to why the batteries are staying at 15C with no heat up. I only applied 15C to the inlet. No other temperature initial or boundary conditions.

Please help! I can provide more details and will be actively looking for replies. I need a solution ASAP! Thank you for your time in advance!

Mark

[Amal.C · Accepted solution]

Hi Mark,

Thanks for reaching out! 🙂 It looks like an interesting project you are working on!

If you are turning Auto-forced convection On (flow and heat transfer are solved separately when the flow does not depend on the temperature distribution), then CFD will go through these steps:

1. The analysis runs as Flow-only with heat transfer disabled until either the analysis converges (as determined by the Automatic Convergence Assessment) OR the prescribed number of iterations is complete.
2. The flow calculation is disabled and heat transfer is enabled automatically. The analysis runs an additional 250 iterations or until the solution reaches convergence

Did you wait till the end? 

May I ask why are you activating the Auto-forced convection? 

PS: If the 50 cells do have separate volumes i.e your model does have 50 different parts and you select all while applying a total heat generation of 150W, this means that each volume has a total heat generation of 150 W instead of 3W. 

I hope this helps.

Thanks,

Amal

[Jon.Wilde]

Simple question to start 🙂

Do you have heat transfer enabled within Solve->Physics?

Are the 50 cells a single volume? If not, you would have assigned 150W to each volume you selected.

How do the coolant inlet and outlet get into the housing? Somehow through the air volume? I ask to be sure they are not also somehow attached to the air.

How have you modelled the 35c ambient? It sounds like you either need to consider modelling the natural convection around the unit (more on that if you need it) or to remove the external air and utilise film coefficients on the external solid surface to allow heat to escape (5W/m2/K at 35C should be close).

To answer your other question. Modelling the air in CAD or CFD would be fine 🙂

Hope that helps,

Jon

[Anonymous]

Hi Amal,

Thanks so much for the reply. It looks like I was impatient (or didn't know what to look for). I let the solver run overnight and came back to finished results. I didn't realize that they'd run separately like that. I also was able to figure out that I applied 150W to each cell when I was troubleshooting. Oops!

I enabled auto-forced convection because I don't know any better. I'm a very new user of Autodesk CFD. Since the cooling fluid is flowing and "forced", that's just the button I hit. Is that wrong? Would my results be different if I deselected Auto Forced Convection but kept everything else the same?

Thanks again for the help!

Mark

[Anonymous]

Hi Jon,

Thanks for your reply. Yes, I have heat transfer enabled. It looks like the solver went in 2 steps: flow then heat transfer. The 50 cells are separate volumes, so I was incorrectly applying 150W to each. That's been fixed. I was able to let the solver run overnight and I have some results now.

The cooling fluid flows within the housing. The air contacts the "outside" housing and the outside of the battery cells. So the heat transfers as follows: from the battery, to the air, to the housing, and then to the cooling fluid which is flowing.

I modeled the 35C ambient via a film coefficient of 4 W/m2K on the outside of the entire housing system, with the reference temperature set to 35C. Should I also apply an initial condition of 35C to all volumes?

Thanks again for your help!

Mark

[Amal.C]

Hi Mark,

Thanks for your feedback 🙂 

Turning Auto-forced convection On runs the simulation faster. Otherwise, for your case, I would keep it off especially if the model is not big in size and in mesh count. 

Regarding your question about the boundary condition, we were not able to correctly figure out how your model is. If you have time, can you please draw a quick scheme or make a screenshot of the model so that we can better advise you on the right setup? 

Thanks,

Amal

[Anonymous]

Amal,

The size is very large, and I'm running on my local PC. So I think it might be best to leave Auto Forced on.

I'm only able to show you a sliver of the part, but you can get the overall understanding of it with the attached photo.

The lid is removed so you can see the inside. Dark blue is the cooling fluid. Light blue is the housing. White is air. The cells are yellow.

I have a film coefficient of 4 w/m2K on the outside of the light blue housing. The cells each have 3W. The air is stationary. The inlet is shown, which is 15C and has a uniform velocity profile. Outlet is not shown, but has 0 pressure.

I am varying the inlet velocity to manipulate flow rate, and seeing how cool I can keep the batteries.

Thanks!

Mark

[Amal.C]

Hi Mark,

Everything looks great here 😉 Well done! 

Please allow me to mark my post as Solution to close this post. 

Have fun while simulating your model 🙂 

Cheers,

Amal

[Anonymous]

Thanks Amal!