Adding freeze protection to Batteries.

[efficientPV]

I think this is what Will did a video on, 65W at 13.5V. Facon 12" x 18" RV Tank Heater Pad, CW-T1218​

 

[HighDesertOffgrid]

I came to this forum for this exact reason. If I can add freeze protection to my (6) Lossigy 12V/200A for a couple hours labor and a few dollars, why would I not do it?
Thanks for posting this topic and your solution!

 

[Texas-Mark]

If I can add freeze protection to my (6) Lossigy 12V/200A for a couple hours labor and a few dollars, why would I not do it?

As long as you understand the issues noted above.

Of course if you are talking about using a heater, then yes, that is a much better option.

 

[HighDesertOffgrid]

As long as you understand the issues noted above.

Of course if you are talking about using a heater, then yes, that is a much better option.

I may or may not understand the issues above. My perception is that the overheat protection may or may not completely disconnect the battery from charge and discharge (depending on the BMS). If, as I suspect, most BMSs completely disconnect the negative connection from the battery to the BMS, losing power would be preferable to instantly ruining batteries. That is my untested hypothesis today.

 

[Texas-Mark]

If, as I suspect, most BMSs completely disconnect the negative connection from the battery to the BMS,

Not necessarily. My BMS (with low temp cutoff) only disconnects the charge MOS on low temp, and you can continue to use it to power things. I believe some?? only disconnect the discharge MOS on high temp cutoff. If that is the case, then this will do nothing to protect it from charging. And even if it does cut off the charging too, then you won't be able to power anything if it is below freezing. Not the ideal situation if you are off grid or need to use the battery at low temps.

If one is dead set on trying this, best to make sure charging is actually disabled on high temp alarm.

Also if multiple batteries are in play, then you need one for each. IMO it is still easier to either heat the batteries or just use a "thermostat" to turn off the charging source.

The attached is from when I did a freezer test to check if the low temp cut-off was working. Note the separate charge/discharge MOS status.

 

[HighDesertOffgrid]

Not necessarily. My BMS (with low temp cutoff) only disconnects...

This is a valuable conversation. Thank you for your input. As I am off-grid for the last 3 years in the Mohave, it is VERY rare that temps dip below 25F in winter. That is short lived daily, and is (so far) negated by the heat my batteries create by being used through the night. I don't relish the idea of cutting my batteries up to test my BMS, but it may be a necessary evil...

 

[McKravitts]

Yes, but you obviously don't know how a BMS or LIfepo4 batteries work.

Here is a quiz for you.
1. Does every BMS's high temp sensor cut off charging when tripped?
2. Does every BMS's high temp sensor cut off discharge when tripped?
3. Is it desirable to cut off discharge at low temps?

Come back when you can answer those.

I read the rules regarding posting comments on this forum and I found nothing granting one member the authority to prohibit another from posting until he/she can answer certain questions.

That said, they are very good questions and worthy of discussion on this forum.

Like all electronic equipment, BMS’s do what some engineer designed them to do and perhaps what type of battery chemistry he/she had in mind at the time. Not to mention the engineer’s skills and knowledge, which varies substantially from engineer to engineer.

And, of course they have to keep the set up task simple enough for end users.

Then there’s the pressure from the non-technically inclined profit seeking members of the organization to make the device cheaper and cheaper to increase sales.

Since many were designed with insufficient shut down protections, it can be inferred that that there are many that shut down, cut off as you say, under some conditions and not under others.

Without having analyzed every BMS on the market, I believe answering NO to your first two questions is a safe bet.

I can’t answer your third question with a simple yes or no.

My guess would be that it depends on several factors.

One is the intended use of the battery. If the battery is used to start a vehicle, it would be very undesirable for a battery to cut off on a cold morning when the driver has to be somewhere soon. Some degradation in the remaining life of the battery might seem like a small price to pay.

In general, the safe discharge rate and safe depth of discharge for any particular battery chemistry must be considered.

I know that Toyota used metal halide chemistry on the batteries it uses in its hybrid cars that are sold in northerly climates. They function much better in cold conditions then Li batteries.

When I said that I used to teach electronics, I didn’t mean to imply that I’m no longer interested in learning more.

I’d like to hear more comments on this subject from anyone with knowledge or interest.

 

[Texas-Mark]

Without having analyzed every BMS on the market, I believe answering NO to your first two questions is a safe bet.

I can’t answer your third question with a simple yes or no.

I rest my case.

I am done with this thread. You obviously do not want to see/acknowledge the flaws with your idea.

 

[efficientPV]

Someone dug a hole they can't get out of.

 

[McKravitts]

This is a valuable conversation. Thank you for your input. As I am off-grid for the last 3 years in the Mohave, it is VERY rare that temps dip below 25F in winter. That is short lived daily, and is (so far) negated by the heat my batteries create by being used through the night. I don't relish the idea of cutting my batteries up to test my BMS, but it may be a necessary evil...

This post is not meant to question your or anyone else’s intelligence, so forgive me if it does.

It is simply in the interest of safety.
Very few manufacturers specify the specific component they use in their sensors.

I just tested a temperature sensor for an EPEVER charge controller. I was surprised to learn that they use a 50 K thermistor. Most HVAC equipment I’ve dealt with use in the 5 to 10 k range. You will need to know the nominal value of the sensor in the battery in order to calculate the necessary parallels resistance.

Make sure you test the resistance at or around 77 degrees Fahrenheit. Most NTC’s have very similar operating curves so using any characteristic curve/chart you find online should get you well into the ball park.

You can add a small rheostat in series to fine tune the resistance of the parallel resistor. Place the temp sensor and added freeze protect thermo into ice water and tune to shut off plus a little margin for error.

 

[HighDesertOffgrid]

This post is not meant to question your or anyone else’s intelligence, so forgive me if it does.

It is simply in the interest of safety.
Very few manufacturers specify the specific component they use in their sensors.

I just tested a temperature sensor for an EPEVER charge controller. I was surprised to learn that they use a 50 K thermistor. Most HVAC equipment I’ve dealt with use in the 5 to 10 k range. You will need to know the nominal value of the sensor in the battery in order to calculate the necessary parallels resistance.

Make sure you test the resistance at or around 77 degrees Fahrenheit. Most NTC’s have very similar operating curves so using any characteristic curve/chart you find online should get you well into the ball park.

You can add a small rheostat in series to fine tune the resistance of the parallel resistor. Place the temp sensor and added freeze protect thermo into ice water and tune to shut off plus a little margin for error.

My particular battery has been review by Lithium Solar(YT) and Will Prowse. During teardown, both commented on the lack of freeze protection. I believe it was mentioned that the overheat protection was a simple thermal switch rather than a thermistor. If so, it would seem to be a simple matter of installing a switch that opens at 32F in series with the existing overheat switch.
As I mentioned earlier, I am in no hurry to experiment on my only source of power, but my curiosity may get the best of me. Too bad the reviewers aren't more scientifically curious as to easy ways to improve these less expensive batteries. Thanks again for the post. It is a worthy topic.

 

[Texas-Mark]

If so, it would seem to be a simple matter of installing a switch that opens at 32F in series with the existing overheat switch.

IMO there is probably a good reason the manufactures did not do that (or this duel purpose senor idea). Again, most likely to prevent a battery from being dead in the water just because it is cold. I would venture to say the vast majority if users do not want to inhibit discharge just because it is below 32 degrees. I know I don't.

 

[Horsefly]

Just some points of clarification here. I'm not asserting that people don't know these things, but some of the comments might suggest some confusion.

1. Almost all of the lower amperage BMS's that we all use here are common port, as opposed to separate port. That means that the charge connection to the BMS is the same as the discharge connection. So any mechanical disconnect that the BMS uses (relay / contactor) will cut off both the charge and discharge.
2. Most of the lower amperage BMS's (<200A, or even at 200A for some) are FET-based, so they usually have separate cut-offs for charge and discharge, even though it is on the same common wire. This can be a great advantage, in that the BMS can cutoff charging (due to cells reaching HVCO or low temperature) and it would still allow discharge. Similarly, a FET-based BMS would probably cut off discharge when a cell gets below LVCO, but would still allow charging.
3. An extension of point #2: A FET-based BMS may be better suited to autonomous operation, since a contactor/relay cutoff may prevent the system from correcting whatever caused the cutoff. For example, if a relay cuts-off the battery due to low voltage, the charger can't charge the battery back up. Similarly, if the battery is cut-off due to prevent over-voltage, the loads can draw down the battery back into normal range.
4. From the videos I've seen, I think Will has been pretty critical of batteries he's torn down and found a high-temp sensor but that didn't have low temp cutoff. I can't remember now, but I think Lithium Solar has also commented the same way.
5. The cell spec sheets from the manufacturer specify a temperature range for both charging and discharging, and I think most of the smart BMS's (those for which the parameters can be set via Bluetooth or RS485) allow both high and low temp cutoffs be defined separately for charge and discharge.
6. It seems to be a point of debate here often, but I think most of us believe the BMS should be only the "last line of defense" of the battery, and should not be used to control charging and discharging limits. That is, other items in the system should generally prevent the BMS from having to cut-off the battery.
7. As I've said before: Keeping LiFePO4 warm isn't particularly hard, and doesn't consume a huge amount of power in most cases. It depends on the individual situation, but I think it would generally make more sense to prevent cells from getting too low in temp, rather than have the BMS or other devices cut-off the battery when it gets close to low temp.

I guess I'll say these are my opinions, but I think they are shared by a large majority of the experienced LiFePO4 folks on the forum.

 

[Leeds]

McKravitts- I designed the very circuit you're describing.

I am using a Tracer 2210AM charge controller which comes with a high temp cutoff but no low temp cutoff. The native thermistor for my device is a 47k ntc and I've bought a $5 thermal switch to pull an 8k resistor into the circuit. If you look up the details on your charge controller you can easily tweak this logic for your hardware/purposes.

Note that because of the huge deadband on the switch I chose this basically locks the system out when there is a chance of damage.

I'm still working on the tweaks but I intend to overcome this lockout condition with a 5W dedicated solar panel driving a 5w heater under the battery with no other logic/thermostats involved. That should both make lockouts less frequent and will provide a means to unlatch themselves faster than waiting for a warm day.

Good luck!

 

[HighDesertOffgrid]

McKravitts- I designed the very circuit you're describing.

I am using a Tracer 2210AM charge controller which comes with a high temp cutoff but no low temp cutoff. The native thermistor for my device is a 47k ntc and I've bought a $5 thermal switch to pull an 8k resistor into the circuit. If you look up the details on your charge controller you can easily tweak this logic for your hardware/purposes.

Note that because of the huge deadband on the switch I chose this basically locks the system out when there is a chance of damage.

I'm still working on the tweaks but I intend to overcome this lockout condition with a 5W dedicated solar panel driving a 5w heater under the battery with no other logic/thermostats involved. That should both make lockouts less frequent and will provide a means to unlatch themselves faster than waiting for a warm day.

Good luck!

View attachment 112128

Other than the 41F switching, I like this.

 

[Leeds]

Other than the 41F switching, I like this.

You can get a 0° switch but these are generally fairly cheap devices so you'd have to test every one.

If it triggers at 31.5°F it's probably still in spec for the part but it would be useless for our purposes

 

[Texas-Mark]

Shutting down the charging source is always a better option, especially if you have multiple batteries. Plus it does not inhibit using the batteries (unless perhaps it is an all-in-one unit that totally shuts down). Using one of those digital thermostats IMO is better than a mechanical "disc" since you can set the deadband as necessary, and they are quite accurate.. Could always use a disc as a failsafe.

I use six of these in various projects.

https://www.amazon.com/dp/B076Y5BXD9

 

[Samsonite801]

Shutting down the charging source is always a better option, especially if you have multiple batteries. Plus it does not inhibit using the batteries (unless perhaps it is an all-in-one unit that totally shuts down). Using one of those digital thermostats IMO is better than a mechanical "disc" since you can set the deadband as necessary, and they are quite accurate.. Could always use a disc as a failsafe.

I use six of these in various projects.

https://www.amazon.com/dp/B076Y5BXD9

Victron charge controllers connected via BMV, Smart Shunt, or Smart Battery Sense, can stop charging based on temperature. This kind of setup does ad another layer of protection. I use the BMV-712 with my two Victron chargers, in addition to the built-in low-temp cutoffs on each of the six OverKill BMS's (as last line of defense).

 

[McKravitts]

Leeds,
That's exactly the circuit I was alluding to.
Nice diagram.
Snap disc and pop switches do have very large dead bands.
If your just switching based on barttery compartment temp or a probe attached to a battery case, there are some very inexpensive chinese thermostats on amazon. RioRand for one.
They claim to have Hysteresis Setting Range as low as .1 C
I've never used one but they are cheap.
Thanks for saving me the trouble of drawing a diagram. Ther are a few members of this forum who still don't understand what I was trying to convey. I'm not disparaging anyone. My communication skills are not great.

 

[Dave911]

Leeds,
That's exactly the circuit I was alluding to.
Nice diagram.
Snap disc and pop switches do have very large dead bands.
If your just switching based on barttery compartment temp or a probe attached to a battery case, there are some very inexpensive chinese thermostats on amazon. RioRand for one.
They claim to have Hysteresis Setting Range as low as .1 C
I've never used one but they are cheap.
Thanks for saving me the trouble of drawing a diagram. Ther are a few members of this forum who still don't understand what I was trying to convey. I'm not disparaging anyone. My communication skills are not great.

Hey McKravitts,
I just found this thread and I understood exactly what you were saying, the first time I read it. So your communication skills seem fine to me.
Has anyone come up with any better solutions? The cheap smart thermostats look appealing to interrupt the charge circuit only. Right now a 100 ah battery without charging freeze protection can be had off Amazon for $167. A battery with freeze protection is at least $20 more and that freeze protection is questionable. Like they mentioned it in some comment on Amazon, but when you get the battery it is never mentioned in the manual. Time to stick the battery in the freezer to see if it really works? I guess so! Honey, I need to empty a shelf in the freezer for a few hours....
....Wife's reaction!

 

[Samsonite801]

Hey McKravitts,
I just found this thread and I understood exactly what you were saying, the first time I read it. So your communication skills seem fine to me.
Has anyone come up with any better solutions? The cheap smart thermostats look appealing to interrupt the charge circuit only. Right now a 100 ah battery without charging freeze protection can be had off Amazon for $167. A battery with freeze protection is at least $20 more and that freeze protection is questionable. Like they mentioned it in some comment on Amazon, but when you get the battery it is never mentioned in the manual. Time to stick the battery in the freezer to see if it really works? I guess so! Honey, I need to empty a shelf in the freezer for a few hours....
....Wife's reaction!

I have a bunch of these little W1209 thermostats for various things, which I like a lot.

W1209 12V -50-110°C Digital Thermostat Temperature Control Switch Sensor + Case 732387203216 | eBay

Discover the precision of the W1209 12V Digital Thermostat Temperature Control Switch Sensor with Plastic Case. This reliable controller operates from -50°C to +110°C, making it ideal for a variety of applications. With an easy-to-use display and tactile buttons, you can effortlessly set...

www.ebay.com

Others here on this forum have used them for heat pads on batteries too.

 

[Dave911]

After some thought I decided to go with one of these controllers:

https://www.amazon.com/gp/product/B07GQPT9VG/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1

Yes, I know, I am splurging at $12 per controller!

I'm adding a $6, 60 amp Automotive 5 pin relay to cut off the charge current. The controllers onboard relay is rated at 20 amps, maybe.

I like this temp controller since it is:

1. Adjustable
2. Fairly cheap
3. Has a wide range power input of 6-32 volts.

The other temp controllers voltage limit is right at 12 volts from what I can tell. My alternator gets up to nearly 14 volts. So for the other low cost temp controllers I would need to add more components.

Low temp cutoff in these inexpensive batteries seems to be an after thought at best. This solution should work all of the time and still allow me to use cheap batteries without a low temp charge cutoff bms.

I just bought two more batteries for use outside at $167 each.