Lithium auxiliary battery and cold weather charging

[67cj5]

Interesting idea. Offhand I dunno the numbers for it, and wont bother looking it up.
But off the hop, I would think to use electricity to directly create heat.
Would be better efficient to directly heat resistive heating elements.

As Crazy as the Idea is I went and tested one hooked up to a Fluke DMM back earlier this year and It actually used less Ah that to Heat up than it did to Cool down, I also tried the Dometic TCX 35L and it also used less power to heat up than it did to cool down, On cooling It was using about 7 to 9 watts per hour when hooked up to 240v and it gave almost the same results running on 12v DC, using it set to 4 and 5 LEDs setting, It has 7 LEDs that are selectable from 0 to 7 So in the 1 to 3 setting there is no reason why it should not run for a few days when set that Low.

Hope that helps.

 

[luthj]

Peltier coolers have a limited hot/cold differenetial. Their Coefficient of Performance is usually less than 1. So 1 watt of cooling uses 1 watt of power, and dissipates 2 watts from the hot side. Some units can be run in reverse.

The max temperature differential when moving any appreciable amount of power is about 40C. If used for heating in the winter, you will run into icing/frosting of the cold side (exterior side). You can do the math, but the power consumption would be considerable.



If you are handy with microcontroller programming, you could write a basic control program for heating. It would run a defrost cycle by reversing polarity to the peltier device every so often. This type of approach, assuming the hot/cold temps are within the devices operating range, would allow heating of the battery box with less power consumption than a straight restive heating element.

The typical single stage coolers are about 0.3-0.7. So in theory you could get 1.3-1.7W of heating from 1W of input power. Though you would need a fan, which would reduce that somewhat.

 

[67cj5]

Well According to Dometic The TCX35L uses 8w of power and when I tested it it actually used between 5.5 to 9w per hour in an ambient temp of 27*c Their low power draw is such because they actually cycle on and off like a fridge does and it will cool down to 30*c / 50*f below the Ambient I saw mine hit -1*c /30*f in an ambient temp of 27*c / 80.6*f with nothing inside it and they will heat up to around 65*c, I saw mine hit 57/58*c, Plus it has 7 Cooling settings and 7 heating settings, These coolers are the best performing 12v coolers on the market bar non. The figure I got were close enough to what Dometic claim, and in fact they are some what nicer to use than a fridge due to only weighing 11kgs.

 

[shade]

Peltier coolers have a limited hot/cold differenetial. Their Coefficient of Performance is usually less than 1. So 1 watt of cooling uses 1 watt of power, and dissipates 2 watts from the hot side. Some units can be run in reverse.

The max temperature differential when moving any appreciable amount of power is about 40C. If used for heating in the winter, you will run into icing/frosting of the cold side (exterior side). You can do the math, but the power consumption would be considerable.

View attachment 545724

If you are handy with microcontroller programming, you could write a basic control program for heating. It would run a defrost cycle by reversing polarity to the peltier device every so often. This type of approach, assuming the hot/cold temps are within the devices operating range, would allow heating of the battery box with less power consumption than a straight restive heating element.

The typical single stage coolers are about 0.3-0.7. So in theory you could get 1.3-1.7W of heating from 1W of input power. Though you would need a fan, which would reduce that somewhat.

I'm sure it could be made to work. Heck, just find a TEC fridge and put the battery in it. Some control hacking might be needed to set it to the right temperature range. With a DIY solution, I wonder if condensation would be a problem during defrost cycles.

I'd rather stick with a thermostat controlled resistance heating pad system and be done with it. With good enough insulation, I'm not sure the possible power savings of a TEC solution would justify the trouble.

 

[john61ct]

Most TEC are **much** less energy efficient than a compressor fridge.

Would love to find out how that Dometic is different.

I belief the maximum temperature delta is still pretty limited.

 

[67cj5]

Most TEC are **much** less energy efficient than a compressor fridge.

Would love to find out how that Dometic is different.

I belief the maximum temperature delta is still pretty limited.

John here is a picture of it while it was cooling down Reading 27.0*c Ambient with an internal Temp of 1.0*c

 

[john61ct]

Pretty sure that is unique.

Most such coolers just plain svck

 

[67cj5]

Pretty sure that is unique.

Most such coolers just plain svck

When I tested it on 1 to 5 LED's it was cycling just like a fridge where it would cool down and then shut off and it will keep doing that and as it gets cool it will then switch in to it's ECO mode where the fans slow down because it has what appears to be 3 speeds that it automatically selects, Once it is cool it will start up on the mid speed and then it will drop to the very slow speed and then shut off and it will keep doing that just switching on and off as it see's fit which is how it manages to use so little power.

 

[john61ct]

When I tested it on 1 to 5 LED's

Sorry what does that mean?

 

[67cj5]

Sorry what does that mean?

If you look at that picture I posted it has 7 LED's lit up, with every press of the power button the LED's will light up one at a time and once all 7 are Lit if you press it again it will go back to One LED, Each LED is a power setting and then there is another button that changes it from Cooling to Heating and the 7 LED's become Heating Levels.

When it is running in the ECO mode settings it is almost impossible to hear unless you place your head within a foot to 18" away from it.

hope that helps.

 

[hour]

Some dranks lead to tearing my entire shebang apart to install a false floor tonight. This is a 120w 120mmx120mm 12v (1212121212121212) heating pad for 3d printers. Did a lot of testing, have 4 different silicone heaters now but picked this one. While its dimensions don't match the footprint of the cells perfectly (you can get spot-on from China if willing to wait, and in whatever wattage), using thick aluminum is helpful. There's no hotspot. I added some cheap silicone baking sheet directly beneath the cells though for added neat factor. Funny, the 3pk of baking sheets is one of my better idiot purchases of 2019. Nonskid, transfers heat well, good base for electronic doo-dad modules that have pins poking out the back necessitating spacers to screw-mount.

This heater is more powerful than it needs to be so I'm controlling it with a microcontroller. You could do this with a $5 Arduino nano, that's how i mocked it...
It applies the perfect voltage to maintain the temperature. Like exactly. It'll take some fiddling to see what the floor needs to be at to ensure the probe+bms+bmv are satisfied, but that's no big deal. FWIW it took 4w to keep the plate @ 120*F in 66*F ambient after a brief couple of minutes to get thermistor to temp. Plate was wonderfully warm a few minutes after that... open air test. This thing should use very little power if all goes to plan and is sealed well. Low n slow. Biggest expeh nse is the aluminum. Biggest suck is cutting said aluminum. This is like a $50 project for an area the size of 1 to 4 drop in replacements.

This is one of the 12w silicone heat pads I tested. It wasn't adhesive backed and didn't have a thermistor included. Who knows, maybe that would have been adequate given enough insulation.



But this one was tougher, more powerful, had a thermistor that was giving me good readings, and stickyAFbacked





Fabulous stuff



And then padded, stuffed, and locked in all directions once top bar was installed, pre-BMS + fuse box floor





Testing underway

 

[Alloy]

Wow that's great. Only thing I wonder about is letting air circulate around the cells.

 

[shade]

I think he should fill it with mineral oil. It's the only way to be sure.

 

[luthj]

If you are keeping the pack warm continuously, air circulation isn't a big deal, the pack will conduct heat best from the bottom up, as the anode and cathode sheets will be oriented that way. In the situation where a pack needs to be warmed from freezing, having the sensor opposite the heater ensures the entire pack will be above the minimum charging temperature.

The only major concern is overheating localized to the heating pad. This is easy enough to test for. Turn the heater one continuously in cool ambients for say an hour or two. If the temperature doesn't exceed approximately 160F, you should be good. The test length may need to be adjusted depending on how long the pack needs to warm up.

 

[shade]

If you are keeping the pack warm continuously, air circulation isn't a big deal, the pack will conduct heat best from the bottom up, as the anode and cathode sheets will be oriented that way. In the situation where a pack needs to be warmed from freezing, having the sensor opposite the heater ensures the entire pack will be above the minimum charging temperature.

The only major concern is overheating localized to the heating pad. This is easy enough to test for. Turn the heater one continuously in cool ambients for say an hour or two. If the temperature doesn't exceed approximately 160F, you should be good. The test length may need to be adjusted depending on how long the pack needs to warm up.

I want to see that box filled with oil. You aren't helping by offering useful advice.