Frozen flooded batteries

[gardner]

I have an off-grid installation with a pair of US Battery 230Ah golf cart batteries. It's remote and I can only visit once in a while in the winter. I do get SMS messages from the system 4 times a day. The problem is that the solar panels get covered with snow and the system stops charging for months at a time. The loads are tiny -- a few dozen mA -- but it's very cold and over the course of a few months the SOC gets down to the point that the batteries can freeze at -25C or so. Freezing, obviously, really knocks the capacity out of them and they never regain their original vigour, if they survive at all.

Any wisdom about recovering/reconditioning a battery that has been frozen? What sort of actual damage occurs? Would replacing the electrolyte or doing a long equalization charge likely help?

I am looking for some ideas to safeguard the batteries against freezing. One idea is to locate a small 20W panel in a location that can't accumulate snow and use that to provide a topping charge.

If I have to replace my batteries in the spring, would switching to AGM help? I don't know how those stand up to freezing, but if they are significantly superior, I would give them a go.

Any other ideas?

Thanks

[Sunking]

If the battery Specific Gravity got low enough for the batteries to freeze means they have been in a discharged state for a very long time and have sulfated. Meaning your batteries are toast. No way to revive sulfated batteries. In extreme cold climates you should be looking at using AGM batteries. But even AGM batteries will dioe if left in a discharged state. They just do not freeze and crack the battery jars and let the electrolyte leak out.

[gardner]

The day before they froze they had an OCV of 11.7V at a temperature of -25C -- that is approx 40% SOC or so. They were by no means fully charged, but not totally discharged either. Even a 80% charged FLA will freeze at around -40C.

[Willy T]

Raise the panels to a near vertical position for the winter is what most do with a remote unmonitored site. They won't be snow covered.

[Sunking]

The day before they froze they had an OCV of 11.7V at a temperature of -25C -- that is approx 40% SOC or so. They were by no means fully charged, but not totally discharged either. Even a 80% charged FLA will freeze at around -40C.

No 11.7 volts is roughly 20% SOC. They should never be allowed to go below 50% or 12.1 volts. Click on the chart.

battery-state-of-charge.jpg

[gardner]

That chart is for +25C. At low temperatures the OCV for a given SOC drops. Below -15C it drops quite steeply.

[gardner]

Raise the panels to a near vertical position for the winter

It is tricky to get up on the roof to dink with the panels, but that is something I've considered.

I am mulling buying a 50W panel that I can mount in a south-facing window for the winter.

[Sunking]

You are below 50%.

[Sunking]

It is tricky to get up on the roof to dink with the panels, but that is something I've considered.

I am mulling buying a 50W panel that I can mount in a south-facing window for the winter.

That will turn that panel into a 5 watt panel.

[gardner]

You are below 50%.

Yes, I agree. My estimate was ~40% SOC.

Either way, the batteries are either badly degraded or totally destroyed and I won't know until temps climb above -12C, which will be in effing June at the rate we're going here.

[gardner]

That will turn that panel into a 5 watt panel.

Maybe not 5, but I do understand that it will definitely not deliver 50W. However the goal is only to provide a float charge to counter the self discharge. I'm not sure I need more than a few watts. I have some experimenting to do, I guess.

[Wy_White_Wolf]

Add one panel wired to a seperate charge controller that is mounted vertally on a south facing wall or on a post mount. That way you don't have to get on the roof and mess with the array.

WWW

[gardner]

If the battery Specific Gravity got low enough for the batteries to freeze means they have been in a discharged state for a very long time

When I first read this I knew that it was wrong as general guidance. I didn't want to just start a pissing match, but I've done a bit of digging and wanted to put forward some more specific information about the subject of the freezing point of sulphuric acid electrolyte. There are a few useful sources, but this is about the best primary information:

http://www.unionbattery.com/images/2.6.pdf

It includes a useful table of freezing point by specific gravity covering a wide range of acid concentrations from 0% to well above the concentrations found in FLA batteries. Interestingly the freezing point varies in a complex curve. The data covers the conventional ranges of concentration found in SLA batteries pretty well. Combining this with the conventional specific-gravity vs SOC ranges yields this relationship:

SOC	Specific Gravity	Freezing Celsius
100%	1.255 - 1.275	-56.7 - -70.9
75%	1.215 - 1.235	-35.1 - -44.8
50%	1.180 - 1.200	-22.4 - -29.0
25%	1.155 - 1.165	-16.2 - -18.5
0%	1.110 - 1.130	-9.3 - -11.8

In a chart this looks like:

FloodedLeadAcidFreezing.png

I cheated a little in this data since the conventional specific-gravity vs SOC data uses S-G at 25C whereas the freezing temperature references S-G at 15.5C. If I go back and calculate in the S-G temperature corrections I'll follow up.

The net however is the fact that a fully charged FLA is unlikely to freeze outside of Antarctica. Critical in my case though, is that a partially discharged -- even a 50% discharged -- battery can freeze in the weather we put up with here. It can easily drop to -30C in many locations in North America, and a battery charged up to about 60% SOC could be vulnerable.

A FLA does not have to be totally discharged nor sulphated, and may in fact be in quite reasonable condition, and still can be vulnerable to freezing in winter conditions.

[inetdog]

A FLA does not have to be totally discharged nor sulphated, and may in fact be in quite reasonable condition, and still can be vulnerable to freezing in winter conditions.

If you follow the general rule of thumb for design of off-grid systems, the batteries will rarely get below 80%, so the 50% freezing point is very conservative.
The other major factor is to take into account is that whenever the batteries are being actively charged or discharged there will be significant internal heating.
That means that as long as you are charging each day, going down to 50% occasionally may not be that big a risk, particularly if you insulate the battery bank in the winter to retain the internally generated heat.

[Sunking]

When I first read this I knew that it was wrong as general guidance.

Please do tell what is wrong with the statement. Your charts backs the statement up. You have to let your batteries discharge pretty deep and ignore them before they freeze. Keep them charged properly and daily and it is almost impossible for them to freeze. Secondly if you are in an extreme climate, use AGM batteries.

Your batteries froze because you abused them. Pretty simple. If your system was designed correctly, they would have likely never frozen unless the temps drop below -50 and if that is the case you should be using AGM batteries. What is so hard to understand? .