LiFePO4 battery heater project (makes sure battery doesn't freeze)

[mess]

TL;DR Summary

I am in the design stages of a simple project to make sure my campervan battery never freezes in the cold Canadian winters(-40 is my goal), so that the solar array can continue to charge it

I think PID is overkill (but would consider it). I am thinking of setting up an array of thermistors around the battery to make sure that they are all above T? Maybe two on top, two on the sides. The battery is about 8"x20".

I am building an r10 insulation box around the battery, which will contain the heating elements, sensors and battery. The box can seal itself with a lid.

I ordered two 3d printer heated beds (they have their own sensors) to put under the battery, but also a heating wire ( 12K Floor Warm Heating Cable 33ohm/m Carbon Fiber Heating Wires) that I can wrap around it (not sure which to go with yet), my thought is to just turn them on at a set amperage so that they never go above a max temp (~40c), and shut them down when the sensor array is over 10c, and keep cycling.

Any feedback on my design direction before I build this thing?

 

[anorlunda]

https://batteryuniversity.com/ tells me that lithium batteries freeze at -20C, so you are well below that.

I think one temperature sensor would be enough. I would set the on/off thermostat to hold -10C or even 0C to give me margin. Then the demands on smart temperature control are reduced. That's pretty much what you describe except that -40C to +10C seems like an excessively wide range.

The mechanical engineers may be able to help you with heat loss calculations for the box with R10. If it is difficult to calculate, can you just experiment? Turn on the heat tape and measure the outdoor and in-box low temperatures for the night.

Is the box exposed to the sky, or is it enclosed?

 

[Baluncore]

Any feedback on my design direction before I build this thing?

If battery is cold, PV solar energy goes to the battery heaters.
If battery is hot, energy goes to the charging circuit.
The battery may still freeze at night in winter, but it will not be charged while frozen.

 

[DaveE]

Of course you want insulation. So, I agree with @Baluncore, you don't need lots of sensors. The temperate inside should be pretty uniform. I also agree that you don't need to be too precise on the setpoint temperature, as long as it's above freezing; this will also allow you to not be as careful about temperature sensing.

Normally I hate electromechanical solutions when electronics can do the job. But in your case I would seriously consider a couple of bi-metallic temperature switches to control a switch between heaters and chargers.

Also, give some thought to where the sensors are compared to the heaters so you aren't cycling on/off constantly.

This strikes me as a perfect application for a "good enough" solution vs. precision.

 

[hutchphd]

This all sounds foolish to me. How about using some phase change magic..

An R10 box with $Area=1m^2$ will lose 4W in a temperature differential of 40K. By my calculation it takes 3.34x10^5 J to freeze ONE LITER of water. It would take 30 hr to freeze a ONE LITER bottle of water in your box at -40C outside. 60 hr for two liters. Until then you're at 0 C inside.

The very same water bottles will keep the batteries happily cool while they are charging, particularly if they are somewhat frozen. Not as much fun as PID control, but a lot more sensible

Somebody should check my numbers of course.

 

[Tom.G]

Summary:: I am in the design stages of a simple project to make sure my campervan battery never freezes in the cold Canadian winters(-40 is my goal), so that the solar array can continue to charge it

never go above a max temp (~40c)

That seems a much higher temperature than needed.

There are also available heaters with built-in thermostats at a fixed temperature and those that have essentially thermisters that control power (resistance) based on their temperature. My personal preference would be a mechanical thermostat for complete disconnection when not needed.

(by the way, some have interpreted ~40C as minus 40C, rather than the intended approx 40C. understandable because the Summary has a minus 40 for ambient!)

Cheers,
Tom

 

[Baluncore]

(by the way, some have interpreted ~40C as minus 40C, rather than the intended approx 40C. understandable because the Summary has a minus 40 for ambient!)

I keep getting confused if it is −40 F or −40 C.

 

[mess]

https://batteryuniversity.com/ tells me that lithium batteries freeze at -20C, so you are well below that.

I think one temperature sensor would be enough. I would set the on/off thermostat to hold -10C or even 0C to give me margin. Then the demands on smart temperature control are reduced. That's pretty much what you describe except that -40C to +10C seems like an excessively wide range.

The mechanical engineers may be able to help you with heat loss calculations for the box with R10. If it is difficult to calculate, can you just experiment? Turn on the heat tape and measure the outdoor and in-box low temperatures for the night.

Is the box exposed to the sky, or is it enclosed?

While freezing is a concern, charging below zero is a totally different concern. In this battery chemistry it creates dendrites and overtime these can severely reduce the capacity of the battery.

 

[mess]

This all sounds foolish to me. How about using some phase change magic..

An R10 box with $Area=1m^2$ will lose 4W in a temperature differential of 40K. By my calculation it takes 3.34x10^5 J to freeze ONE LITER of water. It would take 30 hr to freeze a ONE LITER bottle of water in your box at -40C outside. 60 hr for two liters. Until then you're at 0 C inside.

The very same water bottles will keep the batteries happily cool while they are charging, particularly if they are somewhat frozen. Not as much fun as PID control, but a lot more sensible

Somebody should check my numbers of course.

and if i leave my camper parked for 2 weeks on its own in my backyard? This is part of what I am trying to solve also. But great outside of the box thinking with the water though! It just won't last long enough.

 

[anorlunda]

He lives in Canada everyone. He means -40 day and night for weeks at a time.

I think you better run the experiment. If too much heat leaks out, the battery will need more energy to heat itself at night than the solar panel makes during the day.

It sounds like just one battery 8"x20". Is that correct.

A passive solar hot water system might serve to keep the battery warm 24x7. But the size and complexity grows expotentially.

Is your campervan heated? Can you store the battery inside? When I used to live in Potsdam, NY we had -40 degree temperatures too. -45F was the coldest. I recall that several people there brought their car batteries inside at night and put it under the covers in bed with them. True, no kidding.

 

[sysprog]

A Kat's 22200 electro-thermal battery wrap is rated at 80W. When you're away from a power source, if you use a 100AH secondary battery that's in a heated part of the camper, then you can run the heat wrap overnight from an inverter that's connected to that battery, and so keep the starter battery warm.

I like the idea of a solar panel array on the rooftop, but it's in my view a good supplement for, not a good substitute for, a 6-phase 300A alternator.

 

[anorlunda]

The OP didn't give his solar capacity. But with a single battery, my guess is that his daily production is on the order of 1 kWh. That has to cover both his house load and the heating of the battery compartment.

Don't forget that the battery box needs heat 24 hours per day; not just at night.

 

[hutchphd]

Don't forget that the battery box needs heat 24 hours per day; not just at night.

Hence my call for using the happily calibrated transition temperature and large enthalpy of fusion (heat of melting) in a bolus (2L say) of water. When the batteries are charging, the liberated heat in the batteries will melt the water and store that waste heat at 0C until it is required. Nature does most of the work for you here.
If necessary a small backup 4W heater with a -5C cut-on could be added to the mix. I have lived in the Maine woods and have seen -40C but only overnight for a few hours and for a few days running. The worst daily average T might be -20C so 1L of enthalpy water will store 60 hr of grace (see my previous post)

This all sounds foolish to me. How about using some phase change magic..

There are times when more technology is not he answer. And it will burn you up if you are not careful..

 

[sysprog]

The OP didn't give his solar capacity. But with a single battery, my guess is that his daily production is on the order of 1 kWh. That has to cover both his house load and the heating of the battery compartment.

Don't forget that the battery box needs heat 24 hours per day; not just at night.

Running the heat wrap for 8 hours after dark with no running of the engine would deplete 80W * 8h = 640Wh of the 1200Wh of a 12V 100Ah battery, which could be replenished in about an hour by a 1kW solar panel array connected to a 75A charger with the heat wrap continuing to draw 80W $-$ reckoning on 5 hours per day of good sunlight, with enough batteries, you could store about 4 kWh per day, depending on what other electrical loads you were running. On gloomier days, some time with the engine idling would top off your batteries, and of course you wouldn't need to run the heat wrap with the engine running.

 

[anorlunda]

When the batteries are charging, the liberated heat in the batteries will melt the water and store that waste heat at 0C until it is required. Nature does most of the work for you here.

If necessary a small backup 4W heater

I'm a bit skeptical about both of those assertions. Do you have data on the heat loss charging lithium batteries?

A 4W heater could probably be managed, but if it needs 40W or more, it becomes unrealistic. That's why I encouraged the OP to experiment with his box to actually measure the thermal losses.

I found one source that says:

https://www.powertechsystems.eu/home/tech-corner/lithium-ion-battery-advantages/
Lithium batteries charge at nearly 100% efficiency, compared to the 85% efficiency of most lead acid batteries.

If that's true, then the losses during charging would not keep the box warm.

 

[hutchphd]

The 4W number is the heat loss from a $1 m^2$ surface area box with R10 AT 40K temperature differential. My data for charging is nearly nil and probably crucially depends upon the quality of the charger. I note the batteries I use for RC stuff are sometimes very mildly warm after charging. The internal resistances are amazingly for those batteries I agree.

 

[mess]

The OP didn't give his solar capacity. But with a single battery, my guess is that his daily production is on the order of 1 kWh. That has to cover both his house load and the heating of the battery compartment.

Don't forget that the battery box needs heat 24 hours per day; not just at night.

I have a 700watt array now. I get about 8 hours of sun a day(this time of year), often overcast

I'm a bit skeptical about both of those assertions. Do you have data on the heat loss charging lithium batteries?

A 4W heater could probably be managed, but if it needs 40W or more, it becomes unrealistic. That's why I encouraged the OP to experiment with his box to actually measure the thermal losses.

I found one source that says:If that's true, then the losses during charging would not keep the box warm.

The BMS on the battery will not let the charge start if the battery is under 0. So a heater is required.