Active cooling for battery pack

userix

10 mW
Joined
Jan 4, 2011
Messages
34
I’m trying to develop An active liquid cooling solution for my electric kart build which runs on a 44ah 13s lipo pack. It’s gets pretty warm during extended heavy use (heavy acceleration and hard pulls).

So far I have a custom water block put together using multiple small water blocks lined up together. I’m stuck trying to figure out the best way to connect all of them in series. The problem is the barb of the adjacent block are so close to each other that simply connecting tubing together will cause a kink in the lines. The connections made need to be flat in the same plane as the cooling plate in order to properly fit under the battery pack (the small ruler on the right side shows the confines of the space this water block plus tubing connections need to fit in):
B50-E628-D-129-F-47-E9-BAD0-09-DBD1-F2-D4-C1.jpg
,


therefore I can’t do a alternating connection method to alleviate the kinking issue as such because the overlap of the tubing will kink on top of each other once the battery pack sits on top of them:
4-B5049-D5-EE12-46-AE-B7-B4-CE5282-CC4-C1-C.jpg
.

What would be a way to connect two barb fittings that are this close to each other?
 
userix said:
I’m trying to develop An active liquid cooling solution for my electric kart build which runs on a 44ah 13s lipo pack. It’s gets pretty warm during extended heavy use (heavy acceleration and hard pulls).
Sorry I don't have a better solution. Overclocker's idea is solid.

Just wondering, did you consider just using more battery in place of the added weight and volume of your cooling solution? By the time you add a pump and radiator, you might just be better off adding more cells in parallel (lower IR = less heat produced, and more range and less stress on the cells as a bonus). Just curious.
 
serious_sam said:
Just wondering, did you consider just using more battery in place of the added weight and volume of your cooling solution? By the time you add a pump and radiator, you might just be better off adding more cells in parallel (lower IR = less heat produced, and more range and less stress on the cells as a bonus). Just curious.

This.

If your cells are getting hot, it means they're working too hard, and you need more of them, or you need cells with higher current capacity. Cooling them down with a complicated add-on system addresses the symptom but not the problem.

If the point is to have a cooling system just because that’s what race cars have, well... carry on.
 
The trend in EV's is to have a cooling/warming system in the battery, and to also use very low-resistance cells. The main reason is to permit very high charge rates.

I'd also look at 52V as a system voltage, under the perspective that higher volts might allow lower amps.

You might consider Nissan Leaf cells. The modules are 2S, so you only need 7 modules to make 14S. They can provide an exceptional amount of amps.

Best of luck with the cooling experiments...
 
serious_sam said:
userix said:
I’m trying to develop An active liquid cooling solution for my electric kart build which runs on a 44ah 13s lipo pack. It’s gets pretty warm during extended heavy use (heavy acceleration and hard pulls).
Sorry I don't have a better solution. Overclocker's idea is solid.

Just wondering, did you consider just using more battery in place of the added weight and volume of your cooling solution? By the time you add a pump and radiator, you might just be better off adding more cells in parallel (lower IR = less heat produced, and more range and less stress on the cells as a bonus). Just curious.


Yeah, I am thinking about upgrading the battery pack as well, but I want to try out this cooling system I am working on. I do have a radiator and pump integrated setup used for overclocked computers and it's pretty compact and lightweight. The lipo pack I have now does pack a good punch and performs consistently great all to way down to 3.6V per cell. The reason it gets hot is I ride the kart for almost an hour, zipping around my neighborhood (with repeated decent hill climbs). I do long spirited riding around for fun, repeated hard pulls and full speed sprints. So I know with prolonged use duration (~1hr), heat will be generated with any pack of reasonable size.

Making a larger pack will result in a larger footprint, which I already have limited space on my kart setup and I would need to redesign my battery mount situation completely. Plus it's already pretty light given the battery setup I have now. It's a 44Ah, 13s Lipo pack with cells having 0.8 mOhms per cell, which it decent internal resistance.

As it is now, I get the fun I want from ripping around for an hour (~20mi range) and I figure the active cooling will help keep the battery cooler than it is now at the end of the ride.
 
Overclocker said:
get some vinyl tubing and a metal coil spring that'll fit inside. dip in boiling water and form the bend you want. let cool in cold water. voila permanent bend no kink

So I just use the tubing I have now and fit a bending spring inside and follows the steps you mentioned?

LeftieBiker said:
Copper U fittings.

How would I secure the copper pipe to the barb fittings?
 
How would I secure the copper pipe to the barb fittings?

Depending on what they are made from, you may be able to carefully stretch the copper over them and use sealant, or do the same and solder them. I'm assuming that the fittings aren't plastic.
 
Do you want to connect adjacent nipples like the red lines, or non-adjacent like the clear tubing picture? Either way, find a cooperative auto parts person and look for an existing molded rubber hose that will work as-is or cut to suit. Hose clamps to taste, spring-type probably better than worm drive. Hose manufacturer Gates has online catalogs you can browse.
 
I am assuming that you are using 6 mm tubing & fittings ?
Are those nipples in the heat sinks screww in ?
Look for “John Guest” tubing and fittings ... plastic , push fit PVC flex tubing system used for leak proof pressure plumbing.
 
userix said:
I do have a radiator and pump integrated setup used for overclocked computers and it's pretty compact and lightweight.
Have you calculated how much heat you are creating vs how much heat a PC cooling radiator can move?
 
Solid plan all round. Even low impedance cells will heat up if you thrash them hard enough.
But remember this - the only reason to use a cooling system is to extend the useful life of the battery. Cooling will only add weight and complexity, and the risk of a coolant leak is real and ever-present. Battery packs and water = expensive failures.

What do your packs look like? And how will they be mated to the cooling plates? Electrical isolation is paramount. Also consider the thermal conductivity of a pouch cell is greatest along the long axis - it's pretty much solid copper and aluminium through there. I used a thermal epoxy to mount the cells to the plates on the Prelude, but I was let down by a failed seal which leaked water into the space between the cells and the baseplates, causing corrosion.
 
Back
Top