diy solar

diy solar

Cooled battery compartment helpful for LiFePO4 in 100F+/38C+ humid conditions? Also peltier and refrigerator experimental data.

jmcroy

New Member
Joined
Jan 20, 2020
Messages
6
Necessary / helpful to keep battery compartment cool if ambient is 100 F but low C-rates on charge/discharge on the LiFePO4 batteries? This would be for an off-grid solar install in South Georgia, USA, where it is hot and humid. It seems I could double or triple the life by not letting the battery compartment get so hot. Has anyone experience with this? I don't think much self-heating will occur, as I should be pushing low C-rates. But I'm worried about the ambient temperature's effect on battery longevity. ..I've also wondered if I could boost the life of the inverter and other electronics by not exposing them to 100F+ south Georgia heat.

Also, in trying to answer my question as to whether it was even possible, I did some tests on my own using a 12V peltier mini-fridge to cool 1 cubic foot of volume and a (I think it's 4.7 cu ft.) mini-fridge that uses a compressor. My conclusion was that the (compressor) refrigerator was superior for cooling in-terms of energy efficiency, which is a big motivation of mine in this small system. Consumed 15W to keep compressor fridge 49 F at 70 F ambient. Consumed 35W to keep (my much smaller volume) other compartment 80-81 F at 110 F ambient temps. The data I collected, along with some explanation (copied/pasted from another post) is below, if you have a use for it.

Let me know if I made a mistake. I am learning.




Info on test:

IMG_8412.JPG

I took a cheap 12V peltier fridge and rigged up a compartment out of an old styrofoam cooler. 12V fridge was affixed to the top of the cooler and sealed-up such that the 12V peltier could cool the approximately 1 cubic foot volume that was the styrofoam cooler. styrofoam cooler then placed inside of an insulated dog house where I used hair driver to keep ambient temp (in the dog house) 110 F+. I gave power to the peltier with temp control loop to try to get to 25 C / 77 F and watched the temps as hairdrier was applied. An hour after starting to heat-up the doghouse, the temp increased inside of the 1 cubic foot volume to 27 C, after which the temp controller kicked-on and stayed on, trying to cool the volume with peltier back to 25 C / 77 F. For the next 6 hours, dog house temp I kept at 110 F+, and the 1 ft cubed volume stayed at 80 - 81 F for the duration. I think the system was at steady state, so I concluded that it was possible to cool such a volume to ~ 81 F when it's 110 F outside. Peltier power consumption I measured to be 35W. This was 1 peltier with heat sink and fan - 35W to run it all.

So I concluded 35W of peltier could likely adequately cool such a 1 cubic foot volume to 25 C / 77 F, if it was 100F - 105F outside.

I did something similar with a mini-fridge (but ambient temps were ~70 F instead of 110 F. I think fridge is about 4 cubic feet in volume. Inside and outside of fridge started at 70 F. After starting the fridge, the compressor ran for 27 min to get the temp in the fridge down to 49 F. While compressor on, fridge consumed ~ 115W. After the initial cool-down to get the fridge to setpoint, I just let it cycle on and off as-needed to maintain the temperature inside of the fridge. The next few hours it cycled a handful of times, and I concluded I could maintain about a 20 F dT (fridge 49 F ambient 69-70 F) with 15 W on-average of power. So the fridge would actually only run for like 5 minutes, then it would hit setpoint and turn off for like 30 minutes plus. 1 complete cycle I said was about 37 min. 5 min/37 min = 0.13 and some change. So that's 13% or whatever it was that the fridge was actually on and consuming power. 115W * 0.13 and some change gave me 15W.

Insulation I'm sure was better in the fridge than my cheap styrofoam cooler. And ambient temperatures were different between the 2 trials. But the test confirms what I've been reading - and that's that vap-compression is more energy efficient than peltier. Surge loads are of course to be considered - no surge loads for peltier, which is a plus if system size constraints.

so in summary, this is what I concluded / learned:
peltier consuming 35W continuous can maintain 1 cubic foot of volume to 80-81 F with 110F ambient temps. peltier had heat sink and fan too. 35W accounts for all of it.
compressor/mini-fridge can maintain 49 F with 69-70 F ambient temps of a 4 - 5 cubic foot volume with 15W on-average power. Fridge only needed to run a few minutes per hour (5 minutes out of every 37 minutes) to maintain temps.
THEREFORE - fridge is probably better in-terms of energy efficiency.


more pics:
IMG_8412.JPGIMG_8385.JPGIMG_8401.JPGIMG_8407.JPGIMG_8887.JPGIMG_8888.JPGIMG_8889.JPGIMG_8890.JPGIMG_8388.JPGIMG_8891.JPG
 
Thank you for this great research. Did you run into condensation issues with the peltier on the surface or on the inside? I was thinking of lining up my box with a thermal pad to keep the cold in better. Have the peltiers hooked up behind an aluminum plate. Heat sink used from the computers that have copper heat pipes for better cooling, and a fan attached to draw the heat away from the cool sink. Rather than styrofoam, I was thinking wood, because I would not dare put 90kg battery system into a Styrofoam container.

@BiduleOhm has helped me greatly design it. Now I just need to do it at some point ?. 35watts for 6 hours is acceptable to me, if I can increase their life from 5 years to 10 years of abuse. Condensation is the only thing I am wondering about for now, at 25~35 degrees temperature, I doubt we should see any?

Did you implement this for your battery pack?
 
I am giving this a try also. Plan to have condensation drain in bottom of box. I have this. I'm planning maybe a water cooled unit. https://www.amazon.com/gp/product/B07PK3V834/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1

Depending on the orientation of your box, you might want to think of a heatsink with a fan in this seting? If i use the normal heatsink, I would have to mount the fan on the top of the heatsink, rather than the traditional sides. I plan on putting mine on the side. Have to think how to cut the squares. Let me know how things go for you (y).
 
Great write up and well written. I would like to see the same test run with a heat source inside the cooler and refrigerator. I think the refrigerator would remove heat more efficiently than the peltiers. The peltiers are used in small wine coolers, within living areas (family rooms with wine bars) because they are quieter than a compressor of a mini refrigerator. Keep up the great work, I live in South Carolina and have the same concerns with heat and humidity. My batteries are on the floor of my garage but lately the temps are in the mid 90's and 1000% humidity (just feels that way)
 
I would like to see the same test run with a heat source inside the cooler and refrigerator.

Where do I report Peltier abuse? Jokes aside, if a significant heat source is present inside, I doubt the peltiers will be able to keep up. I was thinking of creating a similar box for my inverter as well. Than I changed the idea to do it for inflow air vents of the inverter. If cool air is constantly coming inside (30 degrees celcius ~40 degrees celcius) while the ambient temperature is greater outside, the inverter should run a bit cooler, right?
 
what you need first is a real enclosure with proper isolation.
i mean at least 5 inches thick styrofoam, with each side having metallized air tight cover. the box must also have an air tight join.
then you are sure that no moisture will come in , and no air (and calories) will move in.
then only you can find exactly how much calories your battery is expelling and size the peltier to at least keep up + necessary for real cooling.
the advantage with compressor is it is so powerful (even a small one) that you are sure it can keep up.

a small chest fridge could be a good start , provided it offers already all what you need, just small modification is needed.
 
A cheap mini fridge (with condenser NOT under the metal skin) combined with another 2" of XPS foam works for modest sized packs. The next step would be to make a custom enclosure. There are kits for making custom fridges for boats, take one of those evap/compressor kits, add an internal fan. Since compressor units have a coefficient of performance around 2-3, you will save energy over other methods. Small fridges typically have 75-200W of cooling capacity. A full sized fridge will often be 300-500W of cooling capacity. Though some are higher. If you know a local HVAC tech, cutting up a fridge, and installing the components with a custom thermostat is pretty straightforward.
 
I am only interested in cooling the LiFePo battery bank. The need is to protect the bank from ambient heat if vehicle is in hot sun. There will be outside air flow to the hot side of the peltier. The cold side ill be inside the battery box. The unit I posted worked very well during tests with the van in full sun. I have options to use a water cooled unit, but this one looks like a flux-capacitor.
 
I have been thinking of doing the same thing, but instead of using a styrofoam box, I was thinking of using a cheap igloo cooler.
That being said, I was thinking of going a different direction and connecting to the cars a/c ducting to the battery box.
 
I have been thinking the same thing, about heat on LiFePO4 batteries. We are on the island of Luzon in The Philippines where heat and humidity are.... well let's just say it makes Florida in the summer feel cool! Has anyone tried one of these micro dc a/c units? Mini a/c 12/24/48 volt? It has inlet and outlet that use a hose or pvc to connect and puts out between 500 and 1800 btu. I was thinking of making a box out of 2 layers of Hardiboard with aluminum sheet or styrofoam insulation between them and a pipe coming out to allow some of the air to enter the small fans I use to blow air on the inverters.
That's about it.. the dog
 
If you have a local AC repair person, they can probably take a fridge and make a small AC unit which will cool a battery pack pretty easy. In the Philippines, I bet the labor and material would be cheaper than buying one of those DC native micro chillers.
 
Thanks for the suggestion. Yes labor is ridiculous here! Our top staff electrician makes $12.US a day and regular semi trained workers about $7. One of our commercial tenants is a refrigeration company they could probably build it for a lot less! I never thought of them......
That's about it. the dog
 
Following - I'm in Dubai and summer heat is pretty insane.
A compact battery cooler would be interesting. It doesn't have to keep them cold - If it can keep the batteries below 35c, then that would be great.
Mine are just powering a fridge, so low c rate solar charging & discharge. When the engine is running the maximum charge C rate is 0.25 which is still low, but the vehicle AC is also running.
 
I'm wondering if a water cooling option might be appropriate using the fridge? Put a copper loop inside the fridge, a small pump to circulate the water/coolant and a couple of loops around an insulated battery box? A thermostat inside the battery box that controls the pump would probably be necessary.
 
Great idea, I bet that would work, instead of just a couple of coils why not try a heater core out of a car, one in the fridge, one in the box, they would transfer cold real efficient. If you put antifreeze in the system you could put it in the freezer if the fridge isn't cold enough. That's about it. the dog
 
Has anyone actually done this yet? either for a portable RV/CAR or home storage?

I too live in a very hot area (35-50deg C) and I am considering a deep freezer with temp sensor set to switch on/off to keep it at 25deg. I must admit I did not consider much of the humidity issue. I assumed if you sealed it well (put weights on the lid and seal up cabling humidity shouldnt be an issue. My main concerns is if it is a stupid idea for some reason, what the heat generation would be of the batteries and would the deep freezer keep up?

I would be looking at 20kw of 48v with room for more in server rack style batteries and a max peak discharge of 5kw but a lot less on average. I think only about 10-18kwh used per day for our house (on grid import) as we have 7kw of solar on the roof (that would expand).

I think that at 40-50deg C ambient there could be a big impact on how long the batteries would last.
 
Well I finally did something. I chose the easiest route. I closed in and insulated my power room. We installed a 1hp window air conditioner set on a timer, it comes on at 8AM and goes off at 5PM then it goes on for 1/2 hour 4 times a night. It keeps the room down to about 30-32C even when the outside is 40C with 90% humidity. I decided that the inverters might like it cool as well as the batteries. Now the system actually seems to work better than before.
 
I am a big proponent of having heat generating equipment in a separate compartment from the batteries when in a hot climate. The energy required to keep a LFP pack cool is nearly an order of magnitude less than what is required for inverters, charge controllers etc. In many cases a moderately well insulated battery pack will average out the daily temps on its own (high 40 low 25C, so the pack is an average of 32C), with little or no additional cooling. While a pack in a ventilated enclosure with inverters etc can easily reach or exceed peak ambient.
 
Back
Top