LFePO4 in remote cold/snowy install

[v_green57]

Per subject line.

I have a small wireless network in a remote area which uses solar power at all the sites. I have been using AGM batteries since I set things up in 2008, replacing them as they reach EOL.

Recently, the AGM at one of the sites starting showing its age, so I thought this might be a chance to evolve to LiFePO4 and try it out. The site isn't important enough that if I make mistakes during the learning experience and stuff goes down for awhile somebody will complain. It only hosts an ultrasonic water level indicator for a tank there and provides a wireless relay for data to one cabin which won't see any occupancy until next Spring.

I spent quite a bit of time here and on other sites educating myself on Lithium before pulling the trigger on a Weize 12V/100aH battery last October. It would replace an old Interstate DCM0090 90Ah AGM. Ran it at my shop for a couple months off a 250W panel and an EPEVER CC programmed per Weize's instructions. I was pretty impressed.

Took the battery up to the site and installed it over Thanksgiving week. We see snow and single digit low temps there so I went with the beehive heating pad and 12V thermostat from Amazon route. This worked well - keeping the battery at 40-42 deg. The battery was reaching "float" (I know that's not really how LiFePO4 works, but it's what I call it) almost every day. The load is a constant 15W.

At the end of December, we had a weather event that resulted in a week of snow and little to no sun. The Weize kept powering through for 6 days when the BMS finally cut it off due to low voltage. We have not had that many days of no sun since I got involved there in 2004.

Because of that the heater stopped working, so we had to do a "boots on the ground" visit to jump the battery with 12V and reset the BMS after warming it above 32.

My question is this:

How do you configure a remote Li install that can restart itself if LVD cuts off the heater and the battery temp goes below freezing? I gave some thought to running the heater directly off the panel array using a DC-DC converter, but that wouldn't run at night.

I know the obvious solution is to install so much battery capacity that this will never happen. That gets expensive.

Are there other ways to approach this? I really would like to make the move to Li, but the result needs to be as bomb-proof as an AGM install. Having a LVD event with those was annoying, but as soon as the snow melted off the panel array, they would self-restore. Snowshoeing over a mile to a hilltop site to kick-start a Li battery that has been BMS'd does NOT appeal to me.

Thanks for any advice you all can give.

 

[Bobert]

Per subject line.

I have a small wireless network in a remote area which uses solar power at all the sites. I have been using AGM batteries since I set things up in 2008, replacing them as they reach EOL.

Recently, the AGM at one of the sites starting showing its age, so I thought this might be a chance to evolve to LiFePO4 and try it out. The site isn't important enough that if I make mistakes during the learning experience and stuff goes down for awhile somebody will complain. It only hosts an ultrasonic water level indicator for a tank there and provides a wireless relay for data to one cabin which won't see any occupancy until next Spring.

I spent quite a bit of time here and on other sites educating myself on Lithium before pulling the trigger on a Weize 12V/100aH battery last October. It would replace an old Interstate DCM0090 90Ah AGM. Ran it at my shop for a couple months off a 250W panel and an EPEVER CC programmed per Weize's instructions. I was pretty impressed.

Took the battery up to the site and installed it over Thanksgiving week. We see snow and single digit low temps there so I went with the beehive heating pad and 12V thermostat from Amazon route. This worked well - keeping the battery at 40-42 deg. The battery was reaching "float" (I know that's not really how LiFePO4 works, but it's what I call it) almost every day. The load is a constant 15W.

At the end of December, we had a weather event that resulted in a week of snow and little to no sun. The Weize kept powering through for 6 days when the BMS finally cut it off due to low voltage. We have not had that many days of no sun since I got involved there in 2004.

Because of that the heater stopped working, so we had to do a "boots on the ground" visit to jump the battery with 12V and reset the BMS after warming it above 32.

My question is this:

How do you configure a remote Li install that can restart itself if LVD cuts off the heater and the battery temp goes below freezing? I gave some thought to running the heater directly off the panel array using a DC-DC converter, but that wouldn't run at night.

I know the obvious solution is to install so much battery capacity that this will never happen. That gets expensive.

Are there other ways to approach this? I really would like to make the move to Li, but the result needs to be as bomb-proof as an AGM install. Having a LVD event with those was annoying, but as soon as the snow melted off the panel array, they would self-restore. Snowshoeing over a mile to a hilltop site to kick-start a Li battery that has been BMS'd does NOT appeal to me.

Thanks for any advice you all can give.

I would suggest super insulating the battery in winter so the heater won’t have to run as much extending its power considerably. Remember the agm battery would have been flattened under the same circumstances and probably quite a bit sooner than the lifepo4 probably ruining the battery instead of just shutting it off.

 

[v_green57]

Bobert:

Thanks for the response.

I probably should have mentioned that I learned the value of keeping battery temps from wildly fluctuating by carefully insulating years ago. The Weize is installed underground in a 18" dia. vertical piece of poly storm drain pipe about 3' down. On top of it are two pieces of 1" foam board and a lid on top of the "shaft". The inside of the pipe is lined with silvered mylar bubble wrap insulation and the pieces of foam board have the same on the downward facing side. See attached images.

I install ALL my batteries in non-metallic enclosures partially underground. In the winter, it keeps them at least 10-20 deg. warmer than ambient and the reverse is true in summer. I regularly get banks of L-16's to last 8-10 years this way.

An AGM install doesn't run the batteries into the ground after my CC LVD's the load because they don't need the heater that LI does ;-)

 

[45North]

If your system is not "mission critical" then why not just live with it being down for a few days every few years, or so?
I assume it can "reboot" itself eventually when the sun comes back.

 

[v_green57]

If your system is not "mission critical" then why not just live with it being down for a few days every few years, or so?
I assume it can "reboot" itself eventually when the sun comes back.

The electronics aren't the problem, they will come up fine, it's the battery itself.

If the Li gets low enough that either the CC LVD or the internal BMS kills the load, the heater shuts down and the battery temp can drop below 32. The BMS won't reconnect until the temp rises above 32. But that won't happen unless the heater is powered up. Which can't happen until the battery is warmer, the BMS reconnects, and power becomes available. If I wait for the battery to warm naturally, things could be down for weeks until the snow melts.

Catch-22.

This behavior is what I'm trying to find a way to get around.

 

[sunshine_eggo]

Victron charge controller.

Once initially configured for 12V, the MPPT will be powered by PV and provide voltage/current to the battery even if the BMS disconnects.

 

[v_green57]

But what if the battery temp. is below 32?

I have no problem supplying power to the battery. But the internal BMS has disconnected due to low temp. and the battery can't accept it.

 

[45North]

But what if the battery temp. is below 32?

I have no problem supplying power to the battery. But the internal BMS has disconnected due to low temp. and the battery can't accept it.

So you need to find a way for the panels to heat the battery till it's above freezing and can accept a charge again.

 

[v_green57]

So you need to find a way for the panels to heat the battery till it's above freezing and can accept a charge again.

Exactly.

Except that won't work at night, when temps. are at their lowest and the heater would be needed the most.

The assumption would have to be made that during a sunny day, a direct from the panels connection to the heater would warm things enough that the battery would then get enough charge to carry it through the next night via another heater connected to the battery.

After a "no charge" period long enough to cause LVD in the first place, the battery will be very low and probably need a couple days good charge to get back to normal. Running the heater overnight might cause LVD again.

It's looking like excess capacity over what I am used to needing might be the way out. If I win the lottery ($$$).

Or just deciding that Li needs too much manual intervention and/or a climate controlled environment and is not a viable solution for unattended systems subject to weather extremes.

In other words, stick with AGM.

 

[45North]

Yes, that's the big advantage of lead acid batteries here in the great white north.
And they've been starting our cars reliably for decades!

 

[sunshine_eggo]

But what if the battery temp. is below 32?

I have no problem supplying power to the battery. But the internal BMS has disconnected due to low temp. and the battery can't accept it.

Will the BMS not use incoming charge to power the heater? It's my understanding that's how heated batteries work. A set of FETS that prevents charge. A set of FETS that prevents discharge and a relay that diverts power to the heater.

 

[v_green57]

Will the BMS not use incoming charge to power the heater? It's my understanding that's how heated batteries work. A set of FETS that prevents charge. A set of FETS that prevents discharge and a relay that diverts power to the heater.

s_e:

Unfortunately, the WEIZE self-heated battery came out a short time after I bought the non-heated one.

Based on the good info you all have provided here, some of my own thinking, and several cups of coffee this morning, I have the basics of an idea that might work. Will write it up later and post it for further review.

 

[v_green57]

Okay here goes.

In addition to the existing battery-powered heater system, install another heater and thermostat powered directly off a separate solar panel. I have lots of old ones so this is not a problem. Run it off a small DC-DC converter if the panel VOC is too high. Instead of setting it to kick on at 40 deg as most of us are doing, set it to keep things in the enclosure at 50-55 or more. Because it's connected to a completely different power source, this won't deplete the battery system. It can run all day full-bore if necessary. And keeping temps higher than actually needed will mean that the enclosure temp will be relatively high when the sun goes down and the heat stops. I already have a concrete paving stone under the WEIZE and can add more dense stuff (bricks) to increase thermal mass and retain heat in the enclosure.

Cost of hardware (heater / thermostat) is $22. DC-DC converter another few $$. BFD.

Keep the existing heating thermostat/heater that is connected to the battery set at 40. Maybe even 35. It'll only come on if there's not enough heat from the 55 deg system to keep things from getting that low overnight.

Use the LVD function of the charge controller to cut off the load at a higher voltage than usual - say close to 11.5V. This will leave a substantial reserve in the Li batt - hopefully it'll be enough that the next day when the sun comes out it won't have been BMS'd. According to the WEIZE docs, that happens at 10V.

I was talking to a friend of mine who is helping figure this out and he says we have "lead-acid brains" from using that tech for over twenty years. We need to push the reset button and think much, much differently with Li. I would really like to figure this out as it seems that the world is making the switch whether we like it or not. Li batteries have many advantages over lead-acid (you know what they are and I won't repeat them here) that make it worthwhile to figure them out.

I will say that BMS makes it seem kinda like whack-a-mole sometimes. You get one situation that caused a disconnect figured out, and whatever you did to correct that causes another...

But I am a stubborn S.O.B. and I will win. I am smarter than a BMS.

 

[Bobert]

Bobert:

Thanks for the response.

I probably should have mentioned that I learned the value of keeping battery temps from wildly fluctuating by carefully insulating years ago. The Weize is installed underground in a 18" dia. vertical piece of poly storm drain pipe about 3' down. On top of it are two pieces of 1" foam board and a lid on top of the "shaft". The inside of the pipe is lined with silvered mylar bubble wrap insulation and the pieces of foam board have the same on the downward facing side. See attached images.

I install ALL my batteries in non-metallic enclosures partially underground. In the winter, it keeps them at least 10-20 deg. warmer than ambient and the reverse is true in summer. I regularly get banks of L-16's to last 8-10 years this way.

An AGM install doesn't run the batteries into the ground after my CC LVD's the load because they don't need the heater that LI does ;-)

You can improve the design by insulating the ground roughly 2’ out from from the perimeter of the enclosure of the battery with 2” or thicker polyiso or rigid foam. This should prevent the ground from freezing especially in cold conditions. Thus keeping your battery above freezing.

 

[v_green57]

You can improve the design by insulating the ground roughly 2’ out from from the perimeter of the enclosure of the battery with 2” or thicker polyiso or rigid foam. This should prevent the ground from freezing especially in cold conditions. Thus keeping your battery above freezing.

Yep, that would certainly help. In conjunction with a heater, of course. I wouldn't want to count solely on Mother Nature keeping things above 32. She has a vendetta against us solar users.

 

[Rocketman]

Here is an idea that may help… (but slightly expensive)

Victron Smart Battery Protect (BP-65). Use that to operate the battery heater. You can use the BP65 to cut-off the battery heater at any voltage you choose - say 12.5v then program it to restart at any assigned voltage- say 12.6v.

That way you keep the battery off the bottom of the charge voltage.

If it’s cold and the bp65 hits the cut-off 12.5v the heater stops, the next morning the sun comes out, the BMS prevents charging (make sure of the low temp charging cutoff), but the heater starts up, hopefully it warms up enough to start charging before the sun sets.

Good Luck with your project

 

[NEpowerandlight]

A seedling heating mat draws about 20 watts and could be plugged into WiFi outlet plug which allows for control from anywhere.

You can get both for around $40