diy solar

diy solar

Rate of cooling for lifepo4?

Bobert

Solar Enthusiast
Joined
Jan 15, 2022
Messages
1,047
Is anyone aware of a test that has been done to find out how long it takes for lifepo4 batteries to cool down? I have noticed quite a number of posts that mention concerns about damaging lifepo4 batteries because the outside temperatures will be below freezing. I use lifepo4 batteries in an rv application and I have noticed that many RVers panic when the temperatures go below freezing because they don’t want their water to freeze. The reality is that it takes a very long time to freeze water in a hose at 32 degrees. Some lifepo4 batteries that I have seen torn apart appear to have multiple layers of plastic and air space som batteries probably have nearly an r value of 2 between the inside of the cell and the outside world. Has anyone ever studied the length of time it takes to cool a lifepo4 battery and conversely how long it takes to warm it up. Also how much amp draw is required to hold even battery temperatures as it gets colder? (Amp draw produces heat)
 
This doesn't address all of your questions but you might find one of Will's recent videos relevant:

 
I searched on
"lifepo4 specific heat"
but it was a bit technical for practical use.

If you can indirectly monitor the internal temperature of the battery, say by periodically measuring performance [e.g., internal resistance during discharge] you can figure out the thermal time constant of the battery.

I'd soak it for hours in a warm temperature (say 74F) to stabilize it, then soak it for hours at a significantly cooler temp (say 38F, a fridge temperature) while monitoring internal temperature.

I’d say it would take 4 hrs to cool from 74F to 38F, depending on battery size.

BTW, it never reaches the 38F, it just gets closer & closer. The temp follows an exponential curve. The time constant is when it goes 63.2% of the way from the initial temp to the final temp.

An electrical analogy is an RC network, with the voltage on the capacitor representing temperature.
 
Closest thing to get directly to internal cell temperature is the negative terminal of cell. It has all the copper foil electrode connections spot welded to the cell terminal so it conducts heat from core of cell.

I doubt you will have R=2 as side of first layer is just a couple of layers of thin plastic from metal can of cell. Maybe for a Winston cell with stepped thick plastic housing.

Top of double pouch 2.png
 
Last edited:
Last winter I had 2, 12 volt 200AH metal cased lifepo4s in a unheated compartment. A 5 watt 12 volt heating pad under each connected to a thermostat set to 60 deg F with sensor on top of cells. Whole thing wrapped with 2 layers of 1 inch R 5 foam. Mini split and fridge was running varying from 0 watts to 1000 watts with 560 watts average. At 25 deg F morning low temperature. Thermostat indicated battery temperature was 50 degF. At 5 deg F morning low thermostat indicated battery temperature was 40 deg F. Thermostat cycle differential was 3 degF.
I moved batteries to under the bed and remove the heater pads. Trailer floor has R 19 insulation and walls have R 8. With trailer temperature at 62 deg F. Batteries temperature get down to 45 at 5 deg F over night low and high of about 32 deg.
Charging at 1000 watts peak through the day only increases the battery temperature about 4 deg.
Also remember the colder they are the slower you are supposed to charge.
 
Let’s say you leave the batt. out all night & in the morning the case temp. is 40F.

Then you bring it into a 74F house at 9 AM.
There will be a 74F-40F = 34F increase.

After a half hour the case is at 40F + 22F = 62F.
22 is 63.2% of the 34F difference.
The thermal time constant of this batt. is a half hour.

Generally, 5 times constants later, at 11:30AM, you will have reached 74F
 
It is often talked about not charging below 0 degs F, but LFP cell impedance goes up as cell temp drops. As a cell ages, cell impedance rises so will have even a greater terminal voltage slump with load current at cold temps. As cell impedance gets greater, there will be less extractable capacity at moderate load current before low cell voltage triggers. At 0.2 C(A) load current, at -4 degs F, you will only get about 25% of cell rated capacity.

You will get better performance if you can keep cells above 50 degs F.

If your BMS has cell resistance measurement feature, you can use it as an indicator of when cells are getting too cold. For BMS cell resistance measurements you usually need greater than 0.1 C(A) load current to get reasonable accurate reading of cell resistance.

For packed together prismatic cells, center cells are going to stay warmer. Since cell temp effects impedance of cells, it also means matching of cells gets worse. With load current, greater cell mismatch drives faster unbalancing of state of charge of cells.

Rs vs Temp and SOC for LiFePO4 cell.png
 
This won’t help you much, but here is my temps from a pass through RV storage I heated one night:
2D2592C5-1FDD-4EEF-8E65-35104494FCBC.png
The night I took the readngs I did nto turn the heat on and I found out the next day the RV manufacturer put a heating vent in the pass through storage.
 
Closest thing to get directly to internal cell temperature is the negative terminal of cell. It has all the copper foil electrode connections spot welded to the cell terminal so it conducts heat from core of cell.

I doubt you will have R=2 as side of first layer is just a couple of layers of thin plastic from metal can of cell. Maybe for a Winston cell with stepped thick plastic housing.

View attachment 85439
I was thinking of a manufactured battery with a plastic case. 1 layer of pcv + airspace + plastic + tape + the cells themselves.
 
Let’s say you leave the batt. out all night & in the morning the case temp. is 40F.

Then you bring it into a 74F house at 9 AM.
There will be a 74F-40F = 34F increase.

After a half hour the case is at 40F + 22F = 62F.
22 is 63.2% of the 34F difference.
The thermal time constant of this batt. is a half hour.

Generally, 5 times constants later, at 11:30AM, you will have reached 74F
Does this apply equally to Batteries made from cylindrical cells?
 
Cylindrical cells should transfer heat slower, more volume per enclosed surface area.
Finned cells, faster heat transfer.

Thermal resistance is measured in deg. C/watt. A 100w bulb the size of a basketball may hardly get warm.
 
Cylindrical cells should transfer heat slower, more volume per enclosed surface area.
Finned cells, faster heat transfer
Even though the rectangle Prism Cells are stacked one on top of the other side to side, you think the round cells cool worstr? I would have thought more cooling area exposed to the ait on the cylinder cells.
 
Cylindrical cells should transfer heat slower, more volume per enclosed surface area.
Finned cells, faster heat transfer.

Thermal resistance is measured in deg. C/watt. A 100w bulb the size of a basketball may hardly get warm.
in addition to the larger volume to surface area the cylindrical cells are usually wrapped up in packs that have air spaces where the curves don’t fit together. That should insulate the cells slightly.
 
Prism vs. cylinder. . . If the surface area/volume is more, then more heat transfer.

Air spaces. . .if the air is not moving [dead air] then, yes, it insulates.

And if the air moves by convection currents, the cells will heat or cool equally.

Heat transfer is by conduction, convection & radiation.
With colder walls in a house on a colder day, you will feel cooler with the same inside air temperature because your body then loses more heat by radiation, given that the other two stay the same.
 
If there’s 8 people in a cold room, when they are nice and cosey, close together, they are warmer. Standing apart, they are cold. even colder when the wind blows. Perhaps the circular cells with space between them are different than people close together.
 
I'm interested in cooling cells, so I read a few of those technical papers and decided that LiFePO4 cells transfer heat at roughly the same rate as water. I don't remember which paper I saw that made me decide that, so I can't quote it here. In any event, I plan to build a cooler that would keep the same volume (as the battery) of water cool inside a very well insulated box.
 
Back
Top