By: Rob Beckers
Batteries are poorly understood, even by those living off-grid that depend on them for power. A better understanding of the inner-workings would go a long way in keeping those batteries happy. Given how expensive batteries are, and how they are a major (recurring) expensive of an overall off-grid system, battery care should be high on everyone’s off-grid list.
The best investment you can make is to get a decent hydrometer so you can keep an eye on the specific gravity (SG) values of each cell (keep a log), in particular after a 100% charge when SG is a great indication of battery health. Personally I love the Swiss-made “HydroVolt” from Compaselect; they are cheap, very easy to use and read, no mess, build-in temperature compensation, and very very accurate. They look a little different from the usual ‘turkey baster’ type, the guy that designed it had a really good idea with this one.
If you keep a log of specific gravity values after a full charge you can see if and when you need to equalize the battery bank. Over time the cells of a bank tend to drift apart, some reach 100% charge a little earlier than others and after weeks or months some cells may be lagging. That is when you equalize. For customers that we know are relatively clueless about batteries I normally set up the charge controller to auto-equalize once in 3 months, just so all the cells get pulled back in line every now and then.
If you do measure specific gravity, and keep a log, there’s no need for ‘preventative equalizing’. As long as all the cells keep doing the same thing all is well. Nothing gets hurt either by doing an equalize cycle once every 3 months or so, and some claim it has benefits. If you see one or more cells drift apart from the rest by more than 0.025 – 0.030 when fully charged it is indeed time to equalize. Another reason can be that the batteries get sulphated, over winter for example, and don’t reach “full” specific gravity values of 1.265 any more. When equalizing is needed, do it for 2.5 hours only. Equalizing is hard on the batteries and running an equalize session for 5+ hours just beats the tar out of your batteries. If after a 2.5 hour cycle the issue has not been rectified you can always do another cycle, but wait a few days in between, equalizing will be more effective this way too.
Having a battery log is also extremely helpful if you ever have to make a warranty claim with the manufacturer. Cells sometimes do go bad for no particular reason, and if you can show a log they take you a heck of a lot more serious when you make the claim. I would advise topping up (distilled) water once a month until you have a feel for what they need, and taking SG readings once in two months, and writing those down.
Unless your batteries are in a spot where they see a fairly constant temperature year around, it is important to have a temperature sensor mounted on one of them, and a charge controller that knows how to do auto temperature compensation. This is of particular importance if your batteries see cold temperatures in winter, that is when a higher charge Voltage is needed. The difference can be pretty large: Over here a bank that’s outside in an insulated box can see -10C in winter (when it’s -30C outside), and for a 24V bank that would take the bulk/absorb Voltage up a full 2.1 Volt!
The Voltages I generally use for flooded Trojan batteries are a little bit higher than the manufacturer’s values. This helps to keep them de-sulphated a little better, and make it more likely for them to see a full 100% charge on a regular basis. The trade-off is slightly higher water use:
- Bulk/absorb = 59.5/29.8/14.9 Volt
- Float = 52.8/26.4/13.2 Volt
- Equalize = 63.2/31.6/15.8 Volt for 2.5 hour
The recommended Voltages for Surrette/Rolls batteries are even higher, and these come directly from their battery-care manual:
- Bulk/absorb = 60.0/30.0/15.0 Volt
- Float = 54.0/27.0/13.5 Volt
For the absorb time, how long it should keep the batteries in absorb, I use Rolls’ equation;
Absorb Time = 0.42 x Capacity / Current
Where ‘Capacity’ is the 20 hour Ah rating of the bank, and ‘Current’ is the charge current you have realistically available (not the same as the maximum current you may see once in a blue moon). The result is in hours, and usually quite a bit longer than people expected (yes, with fairly feeble charging sources it can take a long time to get through absorb). For example, a 24 Volt battery bank made from four Surrette S-550 batteries, our most popular model, would have a capacity of 428 Ah @ 20 hours. Say we have 4 panels of 260 Watt, and a 40 Amp charge controller. Realistically, those panels will produce just about 30 Amp in full sun. The absorb time therefore is:
0.42 x 428 / 30 = 6 hours
Yes, a full six hours!! That is what I would set up in the charge controller. This also shows how hard it is to fully charge even a small battery bank within a single (solar) day. Keep in mind that the absorb time comes on top of the time it takes to do the first stage of charging, bulk charging, which accounts of getting the batteries back up to 80% – 85% of full. Absorb only takes care of that last 15% – 20%.
There are other strategies to determine when absorb is over, such as measuring the Amps going in and cutting off when it falls below this ‘End Amps’ setting (generally a value of 2% – 5% of the bank’s capacity), but time is reliable and works well for the most part.
If there is another charging source, such as an inverter that can use a generator to charge the batteries, I set the absorb Voltage lower to save on (fossil) fuels. For those chargers I use 58.8/29.4/14.7 Volt (as long as there is also solar or wind available to charge!).
Wait with watering the batteries until they are being charged and are bubbling vigorously; that will help overfilling since the acid is already at its largest volume (warm and full of bubbles). Watering them when they are cold and not doing anything runs the risk that acid comes pi$$ing over when they are charging. During an equalize is a great time to water!
To get long life out of your batteries it helps to know how people kill otherwise good batteries in short order (and avoid that). The no. 1 way we see people put new batteries in the grave well before their time is by letting them sit at partial charge for long periods of time. Lead-acid (flooded or AGM) MUST, absolutely MUST see a full 100% charge once in (ideally) two weeks, or once a month at the outset. A FULL charge means going all the way through bulk, absorb, and absorb time until they are as full as they are going to get (an SG of around 1.265 is ‘full’ for batteries in good state).
The reason for this is in the chemistry: A full battery has lead on the negative and lead-oxide on the positive plate. During discharge both are turned (in part) into lead-sulphate. The dreaded ‘sulphating’… This is not a problem because that lead-sulphate is initially in a state where it can readily be turned back into lead and lead-oxide again, as long as it’s done soon enough! Let it sit, even if that battery is sitting at nearly full, and lead-sulphate grows as crystals, much like the salt-crystals kids grow, they get larger over time. It does no longer reverse as readily in that state, or at all if enough of the plates is covered. Lead-sulphate does not dissolve in water or sulpheric acid, and it does not conduct electricity. Those sites of the plates covered in lead-sulphate are dead for the world as far as the battery is concerned. They do not participate any more in the reaction.
By the way, this is also what you are measuring with a hydrometer: The chemical process that produces lead-sulphate uses up sulpheric acid, turning it into water. The less sulpheric acid, the smaller the specific gravity, the nearer it gets to just water (SG = 1). So, if after charging part of that lead-sulphate did not reverse back into acid and lead/lead-oxide it means the SG will not bounce back to that of the straight acid as it was put into the battery, and your SG reading will show this.
A major cause of batteries that rarely see a full 100% charge is lack of charging sources. In other words, a large battery bank and little solar PV is a recipe for this type of failure. Unless you have a generator set up to run frequently, and do a full absorb cycle, your charging sources (solar, wind etc.) should be able to bring the battery bank back from an average night’s use to full again by noon or so the next (sunny) day. So what does that mean for solar PV for example? Say we have a 24 Volt bank of S-550 batteries, 428Ah, that is just over 10 kWh in energy storage. A typical day may see 30% discharge the next morning, or 128 Ah (3 kWh) in energy we have to put back in. With 2,000 W of solar panels on the roof a really nice sunny day will see about 65 Amp going in, starting in the morning. By 11 am the batteries will hit absorb, and about 1 in the afternoon they’ll be full! This can work!
The moral of this story is that a bigger battery bank is only better if you can recharge it fast enough. What we have found for our area is that for every 10 kWh in battery bank storage, at least 1,350 Watt in nicely south-facing unshaded solar PV is needed, for a system that’s used summer and winter. So, to keep that string of 4 Surrette S-550 batteries happy requires at least 24 x 428 x 1.35 / 10,000 = 1.4 kW in solar PV! This is somewhat of a minimum requirement, and having a bit more is certainly preferable. For other locations the standard off-grid rule that sizes the battery bank for 3 days of autonomy, and sizes solar PV to make up for daily solar energy needs plus other losses, will work well. Having a larger battery bank than that will get you in trouble, especially in winter.
The no. 2 reason people euthanize otherwise good batteries is by discharging them completely. It may seem straight forward, but it happens more often than I’d like to see, mostly by accident. There are reasons for this. Some very well-known inverter manufacturers have absolutely dreadful battery cut-off Voltages set out-of-the-box. Way to low, and at a point where they do damage. LVCO (Low Voltage Cut Off) should not be below 44.3/22.2/11.1 Volt! That corresponds to just about 80% DOD (Depth Of Discharge) under load. Deep-cycle batteries are great at handling anything up to 80% DOD, but not so great once you go beyond. Combine a deep, deep discharge with letting them sit over the winter and you get the picture. Dead batteries. I had a lady on the phone, in tears, not all that long ago who did just that, killing a brand-new $10,000 AGM battery bank in the process!
No. 3 on the list would be consistently overheating the batteries. The warmer batteries get, the more the chemical processes speed up, including the aging of the plates inside the battery. That’s not a problem if it happens occasionally, but keeping a battery bank at high temperature all the time can take years of life off of them. Make sure there’s half an inch (or more) spacing between batteries, and the battery box has some air flow. Also use a temperature sensor, as this will dial down the charging source (either the inverter or solar charge controller) in case it senses that the batteries are getting hot.
When it comes to having multiple batteries in series and in parallel there are some guidelines to consider. Putting multiple batteries in series increases the Voltage; two 6V batteries in series makes 12 Volt. The Amp-hour rating of the total stays the same. Putting batteries in parallel keeps the Voltage the same; two 6V batteries in parallel still makes 6 Volt, at twice the Amp-hour rating. Usually a battery bank is a combination of multiple batteries in series to reach the needed Voltage, and multiple of those series strings in parallel to increase Amp-hours. Fewer batteries, or strings of batteries, in parallel is generally better. It means fewer filler caps to keep watered and that need measuring of specific gravity. It also makes it easier to keep all those battery cells doing the same thing. Series is no problem, it’s parallel strings that creates issues. Some people prefer two parallel strings of batteries, in case a battery or cell goes bad they still have the other string. Normally the maximum I would advice in parallel battery strings would be three; with four the absolute maximum ever. Even with three parallel strings it becomes difficult to keep all those cells in sync. There’s always a cell that falls short vs. the others, and that will get progressively worse. That means an increased need to equalize the battery bank, and possibly diminished battery life.
My suggestion for those with a substantial investment in batteries is to buy a MidNite Solar MNBCM. They are not very expensive, and show the approximate state-of-charge at a glance. They are (nearly) idiot-proof, no training is needed to be able to read them (put a 1 Amp automotive fuse in the positive line before connecting to the batteries so you don’t burn the house down!). They are not very accurate, but will give you an idea of “what’s left int the tank”, to help determine if those 3 loads of laundry are a good idea at that time or not. What makes them worth every penny is that they have a little red LED that comes on when the batteries did not see a full 100% charge in the last two weeks. When that LED lights up you know that unless there’s lots of sun in the forecast it’s time to fire up the generator…
-Rob-
Thank you Rob, I found your article very informative but find I am wondering how it may relate or what may differ in comparison to my 4, 400×6 volt, wired in series battery bank. Other sites state they can be discharged down to 20% with little detriment to the battery and I see just running the test setup that they get through bulk rather quickly with absorb taking longer. Maybe you could write a book “Solar Power for Idiots” for guys like me, include one as part of your systems. Thanks again for all your help.
When profiling an EVEver EXTRA4215N-XDS2 Charge controller for 2 Trogan 12 V 100/Ahrs batteries in parallel, which option from Pb/acid, Gel, Lithium, or User should be used in the absence of an AGM option. If the answer is the latter, then what values should be used.
Cheers
Hi Ron,
The Trojan batteries you have are AGMs? Your message doesn’t mention that…
The XTRA series inverters has a “sealed” setting as well that’s between Gel and Flooded in its Voltages. That would be a pretty safe setting for AGM batteries. You can always use the User setting and enter your own values too of course, that allows you to use the exact Voltages the battery manufacturer recommends. So the answer to your last questions depends on the values Trojan publishes for your particular type of battery.
-Rob-
Hello Rob
My name is Craig. I have a 48 volt system which consists of 2 strings of 8 6 volt crown cr 350 batteries. With 2 victro mppt 150 85 .batteries are connected in series and in parallel. I cant seam to get an accurate setting for the charge controllers to set the right setting per charge .pls advise me as to what i should set my charging to .even crown battery cant give me a answer .i read your right up and it makes sense to me .and trust me im no genius when it comes to me .but common sense tells me there charging settings are not enough to pack my batteries. I await your response.
Regards
Craig
Hi Craig,
I found a spec sheet for your batteries here: https://cdn.shopify.com/s/files/1/1579/5179/files/CR-350_PowerSheet_D06350_0213.pdf?16021135676618962151
According to that, your cells have a fully-charged specific gravity of 1.275, a little higher than most deep-cycle batteries. So the first step is to read SG values after a full charge, to find out if your batteries are sulphated or not. If they reach 1.270 and up you’re good, if it falls low then a change in charge settings may be in order.
The same spec sheet says the absorb Voltage of these batteries is 56.9 Volt, but that seems very low to me.I would try the Trojan values, of 59.5V for bulk/absorb. The absorb time depends on how much current overall your two charge controllers normally produce. The equation in the article can calculate that (or let me know and I’ll run the numbers).
-Rob-
Hello Rob
I read many of your posting and tutorial and i’m still missing a real basic information regarding the charging. i’ll try to be as clear as possible.
up to what Vdc source can apply to a battery bank for charging if your bank is fully discharged?
I have 2 PMG (sold as 12v system) connected to rectifier of course. I found that both PMG will produced a ball park of 1V per 15rpm. so at high wind i can easily reach 60VOC (open circuit). what happened to my 12v battery bank? in other word, up to what VOC you can apply to your bank? I understand the Ohm law and the Thevenin theorem but that still not totally clear in my mind.
and yes of course if i hook up my PMG to MPPT controler it make more sens but i can see that most of the user are connecting their PMA directly on the battery bank.
hope this is not to much confusing.