openinverter forum

The lexus UX300E has active air cooling whereby it circulates air across the modules and returns to air to water heat exchangers that I believe are plumbed into the vehicle AC lines and solenoid controlled. The pack is well sealed and so there is a reduced risk of external contamination or moisture.

Just thinking out loud here, but has anyone looked into a basic immersion cooling system? If the battery box does not contain the contractors or other electronics and is water tight, then what if it was simply filled with transformer oil? this alone may give enough thermal mass and conductivity to allow Johan to charge at the rate he wants one time per day. Transformer oil is stable, and non corrosive but im not sure how it would interact with what ever the module manufactures are using for isolation between cells. However if you have iso monitoring then this would not likely be a safety issue. Also the battery box could be "bunded" for safety against leaks.

There are of course more suitable fluids available but they are big $$$.

Just a thought, and I have not seen it tried yet. It's on my list to try out, but may be a long time off yet.

barracuda816 wrote: ↑Mon Jul 29, 2024 9:26 am Just thinking out loud here, but has anyone looked into a basic immersion cooling system? If the battery box does not contain the contractors or other electronics and is water tight, then what if it was simply filled with transformer oil? this alone may give enough thermal mass and conductivity to allow Johan to charge at the rate he wants one time per day. Transformer oil is stable, and non corrosive but im not sure how it would interact with what ever the module manufactures are using for isolation between cells. However if you have iso monitoring then this would not likely be a safety issue. Also the battery box could be "bunded" for safety against leaks.

There are of course more suitable fluids available but they are big $$$.

Just a thought, and I have not seen it tried yet. It's on my list to try out, but may be a long time off yet.

I think it was DIY tested and fnally rejected for some kind of active liquid cooled solution.
https://www.diyelectriccar.com/threads/ ... ar.198305/

Also any servicing would be greatly hindered by oil and goo all over the elements you needed to service...

I still think for "normal" DIY without special tools air cooling of some kind is best option. Then one would need to adapt fast charging limits in BMS. Engineering is allways a game of compromises.

So far I like the vortex tube best as it would require the least modification of the box. Second best would be exit vents in the back. Peltier might help but I can only access the lower 4 modules without a major teardown.

Oil wouldn't work as the box is tight to reject spray water but not really sealed well. Plus the only access possibility is removing the bottom cover - go figure

Tested traction control: viewtopic.php?t=6018

Small update on battery cooling. The fan proved not very effective so now I added air scoops. They can be closed off to prevent debris and water getting in on everyday driving and opened on longer trips with rapid charging. Test is pending.

And finally installed this one: These are 3 NMC 60Ah batteries. So they will sit at 9.5V when empty and at 12V when full.

I got a cheap CC/CV buck converter from ebay, supposedly up to 20A. It is a synchronous design, so not just a diode on the low side but a FET for both high and lowside. It still gets quite warm (70C) even though I set it to just 13A. The high side is always closed when there is no input voltage, meaning you get the full battery voltage out of the "input". So that is boot strapping sorted. I set charge end voltage to 12V or 4V per cell.

It is super small compared to the "80 Ah" lead acid battery and probably has twice the usable capacity. Now I also have space to mount some more terminal blocks on the DIN rail and get rid of my dodgy 12V junction box that just floats around before the battery compartment.

There are downsides of course. It is no longer a low tech solution, so if the buck converter breaks I have no 12V and the car is dead. Also it being an NMC battery it requires a BMS to keep it balanced and to prevent deep discharge.

I could have also gone for 4 NMC cells but was hesitant to run the cars systems at 16V. Also the Outlander DC-DC only outputs constant 14.3V. And if you wonder why these? Because I have them.

I still have the lead acid battery in the trunk in case something goes up in smoke
I also "hardened" the buck converter a bit by gluing down the inductor and properly bolting the FETs to their heatsink, they weren't sitting flush and one had a solder blob or something on its back. Finally soldered the on/off switch permanently on as these tend to fail over time when getting moist.

I have used 4S (used 18650s) on the motorbike for a few years now as the Kelly controller needed it. Plus, it's drain only / charged alongside the main battery. Considering downsizing it and just live with the 50-60% SOC with the DCDC's output voltage

marcexec wrote: ↑Mon Apr 07, 2025 1:47 pm I have used 4S (used 18650s) on the motorbike for a few years now as the Kelly controller needed it. Plus, it's drain only / charged alongside the main battery. Considering downsizing it and just live with the 50-60% SOC with the DCDC's output voltage

Yes we'd be at 3.6V per cell. Maybe I can trick the Outlander DC/DC to generate a bit more voltage by spoofing the sense input a bit.
With V2G I know have a use case that pulls about 0.5A for longer amounts of time without running the DC/DC.

Today I rewired the various 12V circuits: GND, permanent 12V, ignition, charge OR ignition, rapid charge, charge OR rapid charge, GND

Diodes in the back. Just see something is bent here...

Got a chance to test my cooling scoops yesterday on a 350 km trip. Outside temp was 7°C so quite favorable. I drove 250 km then charged from 25 to 65% with 200A. Battery heated up to 44.5°C. I opened the scoops and after 50 km the battery had cooled down to 37°C. Then it started raining so I closed the scoops again. Now on the remaining 50 km the battery did not cool down at all!
So it looks they work rather well. Next test is a trip to 400 km away Dresden in June, so less delta T.

Just found this thread and read all through the last few pages trying to learn of others using air cooling on batteries designed for liquid coldplates. Tough spot you're in since you can't get air between your layers! I don't have any useful ideas for that, unless you can afford 10mm. I found someone selling 3/8" thick embedded pipe coldplates (copper tube in alu plate) on ebay for $90usd. i can send the link if interested.

One industrial solution i haven't seen in a car yet is evaporative cooling. If you're feeling experimental you could consider that technology. I think of it as a middle ground between air conditioning and simple forced air cooling. It's more energy efficient than AC but not as effective. You make one side of a metal wall wet, blow air over it, the evaporation cools the metal. Air on the other side of the wall is cooled without touching the humidity and sent to the electronics. Not sure how antifreeze effects this process, but i'd assume that's been figured out. Marvair makes these for industrial applications.

Or the poor man's cooling idea: pour water all over your battery box once in a while!

I just began designing my battery boxes and there are more complexities and costs to water cooling than i had thought. My pros/cons list is pointing me towards air cooling for simplicity and better packaging!