diy solar

diy solar

Battery life in freezing winters

TimC

WI/UP Border
Joined
Apr 17, 2020
Messages
139
After four winters my camper batteries are going strong. I have eight 18aHr AGM batteries and they are always on active charge 365 days per year. I don't take them out of the camper in the winter and the camper is unheated and sitting outdoors.

I occasionally read that batteries should not be kept in a freezing environment for long. Why are my batteries holding up so well if that is the case? Are these statements just about batteries not kept on a charger?
 
I am asking because I was recently looking at a home backup system (gave up on that idea for now). With that system I was hoping to keep the batteries in an underground bunker sort of thing so they could be near the solar array. The thought was the wiring sizes would be much smaller running 115v across the yard vs 12 or 24 volt.

I was advised to never do that.
 
After four winters my camper batteries are going strong. I have eight 18aHr AGM batteries and they are always on active charge 365 days per year. I don't take them out of the camper in the winter and the camper is unheated and sitting outdoors.

I occasionally read that batteries should not be kept in a freezing environment for long. Why are my batteries holding up so well if that is the case? Are these statements just about batteries not kept on a charger?
Where are you and how cold are the winters where you are?
 
WI/MI border (45.7). Average temp in January/Feb is probably 25-30 degrees. Occasionally below zero for six or eight days. So... cold. If your handle is what I think it is you are probably familiar with that kind of cold.

In garage might be ten degrees warmer on average.
 
As long as lead-acid batteries are fully charged, they won't freeze till the temperature drops to ~ -40F. The sulfanization reaction is accelerated by warmer temperatures, so they sulfanate more slowly when cold. Storing lead-acid batteries cold is good for them, assuming they stay fully charged, which seems to be what you are saying.

I think the worry is batteries that sit for very long periods of time unattended slowly discharge, so by the time winter comes around, the state of charge is not high enough to keep them from freezing.
 
I'm in a similar situation with our cabin.

Sounds like even though these will be in an unheted cbin, but still connected to the panels/charge controller, they will be fine?

We only visit every 3-4 weeks in winter.

200 watts in panels, 200ah of AGM batteries.
 
My solar powered instruments are outside all year and must work completely unattended.
300mA consumption, 100W panel and 150Ah FLA.
Luckily, when it freezes below -10°C, the sky is cloudless and usually I have got a sunny day before, so i can expect a 50% charge at least.
But it is frequently just limit...
 
Last edited:
300mA consumption, 100W panel and 150Ah FLA.
Luckily, when it freezes below -10°C, the sky is cloudless and usually I have got a sunny day before, so i can expect a 50% charge at least.
.
I would suggest you have larger panel for charging your battery. Just getting to 50% sounds very inadequate to me. Doing the math, I'd suggest

(150Ah) X (1/10C) X (13Vcharging) X (1.25fudge factor) = 244W.
(150Ah) X (1/8C) X (13Vcharging) X (1.25fudge factor) = 305W

So, I'd recommend either adding at least 1 additional panel, or better two, or replacing the 100W panel with a 250W-300W grid-tie panel. Your controller will decide what to do. If PWM, then you have to stick with 12V panels. If MPPT, the higher voltage grid-tie is perfect. I can get a 250W grid-tie cheaper than you can get a single 12V panel. Actually, I might be able to get both a MPPT controller, and a 250W grid-tie panel cheaper than you can get two 12V panels.
 
I would suggest you have larger panel...
I am limited by the panel size as well.
Since the instruments are placed on someone else's property I cannot dig a concrete baseplate, every additional surface is increasing the risk of the station being turmoiled and/or vandalized.
Already today, the cost of the mechanics/housing/installation are dominating in the overall bill, with bigger panels this cost will explode.

Falling a few times a year below 50% at times, together with low temperatures, is something that the battery accepts gracefully.
A couple of sunny days and the battery is up to 100% again...
All solar road monitoring systems work that way.
 
I am asking because I was recently looking at a home backup system (gave up on that idea for now). With that system I was hoping to keep the batteries in an underground bunker sort of thing so they could be near the solar array. The thought was the wiring sizes would be much smaller running 115v across the yard vs 12 or 24 volt.

I was advised to never do that.

You have other options. Depending on the size of your array, if you use an MPPT controller, you can series your panels and thus run high voltage DC instead of AC.

As has been stated, the concern is state of charge and temperature. Most manufacturers recommend batteries be kept at 60% SoC or higher when subjected to freezing temperatures.

One of the most important things is to apply temperature compensation to your charging parameters. Charge parameters are for 25°C, and a typical lead acid 12V changes -0.03V for every °C. If at 0°C, an absorption voltage of 14.4V would actually be 14.4V + (-.03 * -25) = 15.15V.

If you were charging on a hot day at 35°C, absorption would be 14.4V + (-.03 * +10) = 14.1V.

The risk of no compensation in cold weather is undercharging; hot weather risk is overcharging.

Lastly, "going strong" is relative. Batteries can behave quite perfectly in their application until they very suddenly do not. It's probably worth pulling one of the many batteries out of the array and testing it for capacity.
 
Doing the math, I'd suggest:
(150Ah) X (1/10C) X (13Vcharging) X (1.25fudge factor) = 244W.
Wrong math for running off-grid! You can completely forget 1/10C:

You must dimension the battery to be able to run the load for many days without any meaningful solar contribution.
1/100C will give you ~48h between 75% charge and 40% cutoff. That is absolutely not enough!
1/250C would give you ~5 days. That is barely enough to survive a week of muddy weather.
1/500C is sound enough, not excluding a few cases of power outages during winter time.
 
Wrong math for running off-grid! You can completely forget 1/10C:

You must dimension the battery to be able to run the load for many days without any meaningful solar contribution.
1/100C will give you ~48h between 75% charge and 40% cutoff. That is absolutely not enough!
1/250C would give you ~5 days. That is barely enough to survive a week of muddy weather.
1/500C is sound enough, not excluding a few cases of power outages during winter time.

It's absolutely the right math for off-grid and lead-acid batteries. Period.

It's the wrong math for your tightly constrained off-grid w/lead-acid design that will cause premature failure of the battery vs. the right math. AGM start to consume themselves below 80% SoC and charging at less than .05C (maybe higher, depending on manufacturer) encourages sulfation DURING charge with permanent capacity loss.

Your setup is fine provided you understand that routine discharges below 80% SoC with low current charging will shorten the life of the batteries, though their lives should still be measured in "years."
 
Lead Acid_Electrolyte Freezing point.png

AGM's are electrolyte staved (just soaked in glass mat) and have higher acid concentration so they freeze at lower temps then above chart.
With less electrolyte, even if they freeze there is less expansion (frozen water) so less likely to do damage.

The reason for freezing temp change with state of charge is during discharge, sulfuric acid is converted to water with sulfate deposited on plates, diluting the acid concentration in electrolyte.
 
Last edited:
It's absolutely the right math for off-grid and lead-acid batteries. Period.
Period? Hmm, you are not showing a lot of openness.
OK, I spoke about discharging, not charging my misunderstanding.

But, if you take the 0.1C bulk charging to be the bible's minimum, it's not the right math either:
Then you will have to include the fact, that a solar panel will only deliver its nominal power during a few peak hours of summer and less than 10% in winter, so you will end up with a ridiculous over-paneling at > 2KW (for a load of 0.3W, remember), from which you will be clipping >90% away nine months of the year.

Like everything, off grid solar is frequently a matter of compromises.

Indeed 0.1C is the recommended bulk charging current, mainly with dumb AC chargers that have no equalization.
Less than 0.1C is acceptable as well for a constrained solar configuration.
Yes, under 0.1C, the electrolyte may get some stratification and the plates sulfation, which can be partly offset by an equalization from time to time.
On the other side the shallow discharge and the fact that all these negative conditions occur at lower temperatures are limiting the adverse effects.
Most of the time, the battery will quasi float at 80%-95% SoC, only at some weeks with bad weather it gets below, but then we speak about ~ 20 of those cycles in a year. Nothing to worry about, should I?
 
Not sure if you missed it, but I listed 0.05C :)

And you don't want to go equalizing AGM. They don't take kindly to it.

You listed many of the challenges, and you're absolutely right. For the days of good sun where you're barely discharging the batteries overnight, it doesn't matter if you're under-paneled as you're likely into absorption very quickly.

The only concern is frequent deep discharges and limited charges that keep the battery below 80% - you'll see reduced battery life, but given the excess capacity, they will still likely have a long useful life.
 
But, if you take the 0.1C bulk charging to be the bible's minimum, it's not the right math either:
Then you will have to include the fact, that a solar panel will only deliver its nominal power during a few peak hours of summer and less than 10% in winter, so you will end up with a ridiculous over-paneling at > 2KW (for a load of 0.3W, remember), from which you will be clipping >90% away nine months of the year.
Whenever I give recommendations, I try to adhere to what the manufacturers themselves state. I wouldn't try to arrogantly think I know better than the people that actually make the products.

Here's a snip from the Trojan battery I myself use, where you can see the manufacturer themselves state that 13% C (which is about 1/8th C) is the charging maximum. I am assuming though that most other manufacturers have similar values. I also checked the Rolls website, and they also had come up with 1/8th.
1607807415385.png

In regards to 10% output in winter, I think that is a bit silly. Are you talking about the winter angle, or just more cloudy days in winter? If you're only getting 10% in winter, I'd suggest that something is wrong? I myself commissioned a new battery bank last year, and utilizing a 1000W rotating array, I generated 6.1kWh the first day doing the commissioning charge on Feb. 8th. So, that works out to be 6.1sunhours in February. Umm, I'm in the northern hemisphere, so I think Februrary is winter, isn't it? Building an array frame that can be adjusted to the winter angle I think helped a lot.

Now where are you coming up with the "over-paneling at > 2KW" number from? I believe I stated a maximum of 305W?

Perhaps the problem here is reading comprehension? I could suggest slowing down your reading speed to allow thoughts to properly sink in.
 
Whenever I give recommendations, I try to adhere to what the manufacturers themselves state. I wouldn't try to arrogantly think I know better than the people that actually make the products.
...
Perhaps the problem here is reading comprehension? I could suggest slowing down your reading speed to allow thoughts to properly sink in.
Please don't confuse maximum and minimum. So long to reading comprehension and arrogance.
In regards to 10% output in winter, I think that is a bit silly. ...
No. Where I live at 51° latitude, it is even optimistic. That is what you can expect in December. Under optimal slope.
1607845571486.png
Whenever I plan a solar unit, I try to adhere to what established databases state. I wouldn't try to arrogantly think I know better from one single commissioning day than researchers that crunched millions of data days into an official database.
From that point of view, planning a full battery charge at C/10 every day is just surrealistic. One must deal with compromises:
1607847865046.png
These are calculations realized according to IEE1562, which is the official standard for Off Grid Solar Powered Systems.
You can buy the IEEE1562 standards if you want to discuss further.
Building an array frame that can be adjusted to the winter angle I think helped a lot.
Of course, my panel slope is fixed-optimized for winter operation: in summer I have excess energy anyway.
 
Last edited:
I'm in Germany, but in NRW we have an "english" climate...
In general, at the same latitude, the european irradiance is much lower than behind the Rocky Mountains.
 
Last edited:
Back
Top