FLA guys, is my battery bank overcharging?

[Grateful Ed]

My battery bank was quite hot and gassing off in the middle of the night pumping out a fair amount of acid, I let the load continue to draw and waited until the morning to turn the system off (convenient to still have the grid). I disconnected the bank from the inverters via the breakers to let them chill out, they were running at about 100°+, mopped up the acid and let them cool down on their own. It has been nearly 24 hours and they are still lightly bubbling. I ignorantly had not protected the terminals when I installed it and have been cleaning corrosion off topically and intended to do what I am now which is disconnect everything, clean and de-acidify it and then reassemble properly. The only info I found regarding the continued gassing in the night was that maybe I had a bad cell and the other batteries were charging back to it but I disconnected one that was working away and it read 1.280 on the end cells and 1.275 in the middle. I didn't check the cell voltage yet but am relieved it looks happy. I will be recording every cell as I clean them. I have been periodically searching and still have not been able to get any real feel for when or how often I should do an equalization charge and stuff like that.

I have had the system on 100% since august of last year and had to charge the batteries externally twice over the winter due to extended dark periods of 14+ days but things seemed to work perfectly. I didn't record it but I think when I was charging them from the grid on new years eve each inverter was running at about 40A at 110V, they are rated as 100A charge controllers (@48v I would assume), they were cranking when I went to bed and all charged up by morning. I checked the levels monthly and barely put any water in it until last week when it took nearly a quart for the whole set. In these several days of blazing sun they appear to have boiled out a lot more. I have three freezers and two refrigerators running, a 1/2hp 220V well pump and use arbitrary power for the on demand hot water heater, microwave and toaster oven and have used my welders and other items periodically but during the day the panels are putting out enough to cover any of that so I think I am not leaning on the batteries much and as I designed it generally have all that juice stored for those extended dark periods.

It appears that they have been overcharging since the summer came with much longer sun exposure. I had thought the inverters would shunt the load when the batteries were topped regardless of how much sun they had. My 32 405W panels generate anywhere from 25 - 50A at 160 - 180V on good days, most bright winter days the batteries would be topped by 11am, by 2pm with light cloud cover.

I have 2 phocos 6.5kW inverters (same chassis as MPP or EG4) for split phase 220V, 24 Crown 390Ah 6V batteries in three parallel series of 8 for a 48V storage set. The inverters are set to factory specs for FLA charging voltages which looked pretty vanilla to me, but learning this stuff by the seat of your pants is a challenge. Guess I'll learn some more now.

I am thinking I need to make sure the batteries are balanced and then set my charging up with different values for summer and winter with a periodic equalization charge which I believe the phocos software accommodates.

Any help on any of this would be appreciated thanks. here's a pic of how its laid out, before I finished the series wiring and put the covers on everything.

 

[sunshine_eggo]

1. Being at night, it implies they were bubbling and burping acid while discharging. Where they charging or discharging?
2. Are your batteries at a steady 77°F? If not, are you using temperature compensation for charging?
3. Chargers should be properly programmed for the correct absorption voltage, absorption time and float voltage. Are they correctly programmed?
4. A sure sign of overcharging is a frequent need to replenish electrolyte by adding distilled water. Is this the case?
5. At less than 100% SoC, water should only be added to sufficiently cover the plates. Once the batteries are fully charged, they can be filled to the proper level with distilled water. Was this not done correctly (water added to fill level when battery was not fully charged)?
6. I count 24 6V batteries. Do all 72 cells have sufficient electrolyte to keep the plates submerged?

 

[Grateful Ed]

Hi thanks, if anything they were discharging as there was no power source, the grid was disconnected, however the biggest load would be the water pump which draws 5A @110 on each inverter for about 2 to 2.5 minutes. I doubt that was on as I have large pressure tanks and there was no water usage going on. The only loads would have been the freezers and fridges which don't run often or draw much, they were hot and boiling ass at 3am.

The batteries usually stay around 80° in the insulated box, I was worried they would be too cold in the basement so I super insulated, maybe overkill. Thanks for mentioning the temp concern I saw no reference to altering the voltage for the temp in any of the docs. However the temps in the basement only vary by 5 or 10 degrees between winter and summer so I would attribute any temp increase to charging.

As stated throughout the fall and winter months they literally used tablespoons of water and suddenly they used more, there were no big nighttime gassing events I knew of. I kind of thought the inverters' overcharge limits would remain steady even if they got more sun but I am just figuring this stuff out.

They have never been low on water, the differences between the cells was very small, maybe one would be 1/8" lower than the others.

I was just down there checking voltages and they are all disconnected and still bubbling very lightly a full 24 hours after the whole system was shut down, nothing in or out, normal? I haven't experienced this before. Thanks again.

 

[time2roll]

Excessive 'boiling' is usually over voltage. This could be equalization too high or a shorted cell. Time to get out the voltmeter and measure the actual voltage on the set and the voltage of each battery.

 

[sunshine_eggo]

Excessive 'boiling' is usually over voltage. This could be equalization too high or a shorted cell. Time to get out the voltmeter and measure the actual voltage on the set and the voltage of each battery.

Agree.

Important to do this while system voltage is at peak. You may find that imbalance between the 6V has some of them at excessively high voltage.

Once that's done, wait until after you've been in float for two hours and measure the specific gravity of all 72 cells.

Hi thanks, if anything they were discharging as there was no power source, the grid was disconnected, however the biggest load would be the water pump which draws 5A @110 on each inverter for about 2 to 2.5 minutes. I doubt that was on as I have large pressure tanks and there was no water usage going on. The only loads would have been the freezers and fridges which don't run often or draw much, they were hot and boiling ass at 3am.

This is VERY concerning. I have no theory on why this happened. Normally, I would assume the chargers initiated an automatic EQ charge for a few hours, but if grid was disconnected, this didn't happen (you're SURE)

The batteries usually stay around 80° in the insulated box, I was worried they would be too cold in the basement so I super insulated, maybe overkill. Thanks for mentioning the temp concern I saw no reference to altering the voltage for the temp in any of the docs. However the temps in the basement only vary by 5 or 10 degrees between winter and summer so I would attribute any temp increase to charging.

If insulated, they are likely getting warmer than ambient, so temp compensation is likely needed, and the charge voltage should be lower.

I was just down there checking voltages and they are all disconnected and still bubbling very lightly a full 24 hours after the whole system was shut down, nothing in or out, normal? I haven't experienced this before. Thanks again.

Definitely not normal. Are you 1000% certain there is nothing charging them?

 

[Grateful Ed]

voltages of each battery range from 6.34 on one and the rest are 6.37 - 6.38 with one 6.39

 

[sunshine_eggo]

voltages of each battery range from 6.34 on one and the rest are 6.37 - 6.38 with one 6.39

Not particularly relevant. Those are resting voltages even below float. You need to know what they are at absorption voltage - probably around 57.6V.

 

[Grateful Ed]

The grid is on a transfer switch feeding the same main panel so it cannot be on at the same time, when its on I disable the AC out and can feed the AC in and it will charge. Like I said there are tiny intermittent bubbles coming up right now when they are physically disconnected and if you listen you can hear the others, I assure you I was equally freaked out at 3 am with them boiling.

 

[Grateful Ed]

well I thought those resting voltages at least indicated no dead cells, the default settings on my inverters in the manual are boost 58.4V, float 55.2 and overcharge limit 66V

I'll have to finish cleaning and reassembling to turn it on and check. Its pretty sunny so I should get some good numbers. give me a couple hours.

 

[Grateful Ed]

Ok all cleaned up, turned it on and it showed 80% or better (the graphic indicating level is lame) and it went right into float, but the sun was flagging, it was still pushing 51.4V but I wanted to see the full monty so I flipped the grid on again and really cheated and left the AC out on but it was simply disconnected by the transfer switch and so the grid and the panels were then charging and it went right to 55.76, the setting is actually 56V, I let it run for an hour with the box closed and they were 79°. I finally found some Crown specs and they recommend a 54V float so I may do that but there was an awful lot of crud and the connections might have had an effect. I see they recommend a monthly equalization so I will do one tomorrow and perhaps put it on a schedule in the inverter.

I did have to put almost a half gallon of water in though but for 24 units that's not really much, this is my first summer with the system, looks like I'll be making a still shortly to make my own distilled water, and maybe some other stuff, ha ha ha.

I'll follow up again, especially if it has any crazy 3am menopause symptoms again like my ex used to have, ha ha ha cheers and thanks so much for your help.

sure looks a lot better anyway

 

[sunshine_eggo]

What did the 6V measure when you hit near 56V?

 

[Grateful Ed]

Crap too late, I switched to batteries for the night, they are running as the sole power source at 6.474, how's that sound?
I'll check them in the morning although they start charging about 5:30 before I get up ha ha ha

 

[time2roll]

Yes a bad cell would generally drop the voltage fast under load. If all are above 6 volts as the sun is rising it would seem the batteries are good.

Any chance an equalization charge started in the middle of that episode? Consider dropping the 66 volt protection to 60.

 

[Hedges]

We expect bubbling during charging, not during discharging. Confirm there were no sources of power available?

24, 6V batteries. What is the configuration, 8s3p?
If so, one string with a bad cell might be getting current dumped into it by others, overcharging its good cells.

Did you confirm all similar voltage?
Take hygrometer readings? (Sunshine mentioned specific gravity)

Were the batteries over-filled with water?
(I used to fill car batteries before charging. Now I understand only ensure plates covered, otherwise leave lots of room, top off to proper level after charging.)

 

[Grateful Ed]

Thanks, they are 3 parallel series of 8 6V to create a 48V bank. It appears the resting voltage is very close hopefully indicating no bad cells, I will do specific gravity on all cells tomorrow. I have kept the batteries filled perhaps to excess. the water level has never been low

as mentioned there is no possibility of a power source, complete isolation

 

[Hedges]

Seems OK, then.

Are these the exact batteries you have?

Crown CR-390

Crown Battery has been a high quality battery manufacturing group for over 90 years. When it comes to the best, Crown is the one to turn to for reliability and performance.

batteryguys.com

Links included to detailed specs, try to charge according to precise voltage, current, time values given.
FLA wants a minimum charge current to stir it; your system may or may not maintain that if there are loads too.

And equalize as recommended with voltage or specific gravity get imbalanced.

 

[sunshine_eggo]

Crowns are a bit finicky.

 

[Grateful Ed]

Thanks for those spec links. Things are looking copacetic, this morning before the sun is on the panels they are handling the load and charging at 52.5V, the batteries are all 6.65 - 6.7V and 74 - 79°. I am going to do a cell check in a little while as I had added water last night so I couldn't then and will set my float down to 54 like the specs say and any other settings. Lots of sun today so it will be a good day to do an equalization charge and I'll set that for once a month like it tells you to. I have to assume that the corrosion was so bad on some terminals that it caused this episode. Still not really understanding it but there must've been some massive imbalance. perhaps the 24 hours of sitting idle created its own equalizing as all the batteries were very close when I cleaned them up. Maybe I've never paid that close attention to it but you could still hear some bubbles in nearly all the batteries before I turned it on last night, not boiling but steady periodic bubbles with no power and no load. I'll post an update on the final specs when I'm finished. Getting a bit better educated as I should. Thanks for your help and rest assured if anything crops up I'll be yelling ha ha ha.

 

[sunshine_eggo]

Thanks for those spec links. Things are looking copacetic, this morning before the sun is on the panels they are handling the load and charging at 52.5V, the batteries are all 6.65 - 6.7V and 74 - 79°. I am going to do a cell check in a little while as I had added water last night

Water should be added in two cases:

1) when the lead plates are exposed - add just enough to cover them.
2) when the battery is fully charged and has been in float for two hours to "top them off" to the "full" level.

Based on the voltages you reported, I would not have concluded the batteries were actually fully charged, and filling beyond #1 is not appropriate. Crowns tend to require an extended absorption time.

SG readings need to be taken AFTER a full charge cycle, the batteries have been in float for two hours. They hydrometer is used by intaking and expelling electrolyte three times before taking a reading to help ensure the electrolyte is well mixed.

so I couldn't then and will set my float down to 54 like the specs say and any other settings. Lots of sun today so it will be a good day to do an equalization charge

An equalization charge should never be conducted until a normal full charge is attained.

and I'll set that for once a month like it tells you to.

Depends on where you look. The datasheet says once a week. The battery manual indicates different schedules.

Additionally, the battery manual indicates you should be confirming > 1.265 SG monthly.

I have to assume that the corrosion was so bad on some terminals that it caused this episode.

I see no mechanism for this conclusion. Corrosion/filth on top of the battery tends to create a potential path of self-discharge and increases the likelihood of cell contamination when the batteries are opened to check SG and/or fill.

Still not really understanding it but there must've been some massive imbalance.

Not in evidence based on the 6V voltages.

perhaps the 24 hours of sitting idle created its own equalizing

There is no mechanism for this to happen. Equalization occurs via controlled over-charging due to increasing coulombic inefficiency as SoC increases and charge shedding, i.e., a battery at 95% takes on less charge from 5A than a battery at 80%. When a battery is at 100%, it sheds the low current charge as heat while the lower SoC batteries in the string continue to take on charge, so that eventually all batteries are at 100%.

as all the batteries were very close when I cleaned them up. Maybe I've never paid that close attention to it but you could still hear some bubbles in nearly all the batteries before I turned it on last night, not boiling but steady periodic bubbles with no power and no load. I'll post an update on the final specs when I'm finished. Getting a bit better educated as I should. Thanks for your help and rest assured if anything crops up I'll be yelling ha ha ha.

FLA batteries are a pain in the ass. If you're a little OCD, and you maintain them faithfully in accordance with the manufacturer's recommendation, you can expect them to deliver an exceptional good service life long after AGM would have failed. If you fall short, they will force you to replace them way earlier than your wallet will be pleased with.

Bubbling during discharge just isn't a thing, so this is causing a lot of head scratching.

 

[Grateful Ed]

thanks for that, all points taken, I will not stop trying to find out what the hell happened, I have never heard of FLA self immolating. and have never seen bubbling except when charging?!? over 100° gassing like mad at 3am with zero power is the strangest thing I ever heard of.

the illustration in the crown pdf shows the water just below the collars, like 1/4", that is way over the plates, if they are supposed to be lower I have certainly had them overfilled, but that's how full they were when they shipped and probably why I get so much acid burping out. When I checked them monthly I just reflexively added a tiny bit if it looked any lower than the last time I looked. But doing the math I have put about 2 1/2 gallons in in 10 months, that is about 4oz per cell.

its 11am, the lame icon is showing full, no more charging activity, it is in float at 56.7V, the batteries are 7.10 - 7.19V, so I will wait until 2 or 3 this afternoon and check the gravity, does that sound right? I will report back with a full list of SoC

Then I can initiate an equalization charge if I need one which I would expect I would as I've never done one.

the crown battery safety pdf I have says equalization is primarily dependent on daily DOD, saying that below 50% DOD it should be done more often, biweekly, over 50% it should be done monthly. So are they saying that lighter use creates more imbalance? My common load is very light, I really don't think the batteries discharge much at all, according to my kill-o-watt my appliances each use about 1kW in 24 hours, I have five that most likely use a total of about 2 or 3 kW overnight and even on most cloudy days the system does not draw at all from the batteries after 10am and the charging activity icon is usually full by mid to late afternoon. I have to find more detailed specs on my inverters, they have a low voltage warning but it doesn't say what that figure is. they will shut off if it gets below 50% which it never has and I can boost that number so it shuts off sooner.

the "equalizing" I was envisioning was what you mention when one cell is weak power flows to it, I know its not the same as the charging technique but I figured if one or two of the batteries were less charged than others they would suck juice from others over time while sitting

 

[Zwy]

the "equalizing" I was envisioning was what you mention when one cell is weak power flows to it,

No, EQ will charge until all cells are at the same specific gravity.

The idea is to remove any plate sulfation. It can reverse the process of sulfation by separating the lead and sulphuric acid. Hard sulfation that can not be removed will be blown off the lead plates exposing new lead. Measuring specific gravity indicates the amount of sulphuric acid in the electrolyte.

I know its not the same as the charging technique but I figured if one or two of the batteries were less charged than others they would suck juice from others over time while sitting

Only in parallel. In series this will not occur.

As for the boiling problem, I've seen this occur when cell plates short internally due to sulfation. If you have not performed EQ for a year, most likely this is the cause. While an EQ charge for an extended time might remove this sulfation, it will blow hard sulfation off the plates and you will have less lead and acid available so capacity can be reduced. The sulfated plate material blown off will settle at the bottom of the battery.

How often does absorption charge complete in winter? If full absorption is completed daily, then EQ intervals can be extended. If not often, the EQ will have to be performed often to reverse soft sulfation and prevent it from becoming hard sulfation.

 

[Hedges]

Their use of "below" and "DoD" might be mixed up or cause confusion.
I think my Sunny Island track how many equivalent full cycles and equalize based on that.
If your voltage and specific gravity are close between cells, then equalization shouldn't be needed.

A heavy bubbling charge should be good for it. I haven't read your manual, but I think about 0.12C is desired charge for some/most FLA.
Some inverter/chargers (including AIO) and some SCC coupled with battery shunt can provide a programmed charge rate, harvesting PV for that plus loads.

Try disconnecting all but one 8s string of batteries. That way no others can dump into it. See if it still bubbles at night.
That's after bringing all three in parallel to a full charge + some hours absorption.
The two disconnected strings can settle for a day, then you check voltage and specific gravity.

If you have a DC clamp ammeter, then you can check direction of current flow in-situ. Also balance of current flow.

If you can't complete absorption charge daily in winter, or at least each time the sun comes out, then lead-acid is not for you.
Run a generator if necessary to get a full charge early in the day, switch off most loads, let PV hold absorption for the several hours required.

There is no reason to use lead-acid batteries in a way that doesn't properly charge them, except in an emergency. Doing so will kill them, then you won't be using them. So better to live without loads and have FLA batteries there for when you need them, than to use the loads until you don't have usable FLA batteries and can't use your loads.

My system with AGM has oversized PV (enough for 0.6C) but undersized batteries.
It uses the concept of "load shed" to disconnect heavy loads at 70% DoD, keep critical loads running until sun comes up and recharge has reached 50% DoD.

A similar concept could support regular full/absorption charge, but it has to be predictive not just reactive. Begin with manual control, and generator if needed.

Look what kromc5 said about enough PV for overcast days (he has a big system and many loads)

Very new to solar but started purchasing items, not sure of the final specs of the whole system.

How tall is the top of the array? What angle? I was hoping for close to 30 degrees, but it's right at 21. When I calculate, it only shows a loss of around 400 watts. The back is almost 11 feet, and the ground slopes downward from the back as well. My concern was just wind loads; I added...

diysolarforum.com

 

[sunshine_eggo]

its 11am, the lame icon is showing full, no more charging activity, it is in float at 56.7V, the batteries are 7.10 - 7.19V,

0.1V difference across the 6V @ absorption voltage is a pretty good sign. 7.11 is ideal, so you're likely overcharging the peak batteries a bit, but within reason.

so I will wait until 2 or 3 this afternoon and check the gravity, does that sound right? I will report back with a full list of SoC

56.7 is the absorption voltage, so it should hold this for about 2-4 hours. Then it should drop to the float voltage of 54V. Once it is in float for two hours, the electrolyte should be relatively stabilized - there's a lag in SG accuracy after 100% SoC is reached.

Then I can initiate an equalization charge if I need one which I would expect I would as I've never done one.

I would hold off until you have a handle on the SG, plus it's probably going to take you a solid 30-60 minutes to measure and record all 72 SG values. A helper recording the values makes it far less tedious.

Crown recommends equalization with a generator to ensure constant current is provided.

the crown battery safety pdf I have says equalization is primarily dependent on daily DOD, saying that below 50% DOD it should be done more often, biweekly, over 50% it should be done monthly.

That's not speaking to peak loads, it's speaking to capacity utilization. If you're using > 50% of your capacity every day, then every two weeks. If you're using < 50% of your capacity daily, once a month.

So are they saying that lighter use creates more imbalance?

Opposite. DoD is depth of discharge. A DoD of 60% means you're USING 60% of the battery's capacity with only 40% remaining. DoD of 30% means you're only using 30% with 70% remaining.

My common load is very light, I really don't think the batteries discharge much at all, according to my kill-o-watt my appliances each use about 1kW in 24 hours, I have five that most likely use a total of about 2 or 3 kW overnight and even on most cloudy days the system does not draw at all from the batteries after 10am and the charging activity icon is usually full by mid to late afternoon. I have to find more detailed specs on my inverters, they have a low voltage warning but it doesn't say what that figure is. they will shut off if it gets below 50% which it never has and I can boost that number so it shuts off sooner.

You really should invest in a quality battery monitor. As a smurf, I must recommend the Victron smartshunt.

the "equalizing" I was envisioning was what you mention when one cell is weak power flows to it, I know its not the same as the charging technique but I figured if one or two of the batteries were less charged than others they would suck juice from others over time while sitting

No such mechanism. All 8S 6V in a string always experience exactly the same current flow through all cells identically and simultaneously. There is no mechanism for current to bypass or target specific cells. Your 3P batteries may experience different currents due to wiring imbalanced, but the 8S 6V in each experience the same current as their string-mates.

 

[Grateful Ed]

my float is currently set to 55.2 I didn't lower it yet, and it has been there for well over an hour, I don't want to turn it off yet so as soon as I am done with the measurements which I'll wait another couple hours to do then I'll set it to 54 like the crown specs.

to the comments on winter, most days, it will go into absorption/float cycle around noon, much sooner if its sunny, only the worst multiday storms will keep it from charging completely, even with a light coat of snow I can pull 3 to 4A @140V to each inverter and the panels are set to 60° winter angle permanently and shed quickly, my load doesn't change seasonally, I heat exclusively with wood, maybe use a few more lights but I have big pressure tanks so its easy to use water more sparingly and the pump won't run, then the chargers only compete with the fridges and freezers which I will have fewer of this winter, and an occasional 300W for the propane on-demand water heater

if the grid is on I use that to charge, smoother current than a generator no?

All my wiring to the batteries is exactly the same length. I will try to notice if it bubbles at rest again and see if I can find a current.

here's why I was confused with the DOD, this a typo or editorial mistake then right? D being depth means lower is less higher is more, "below 50% DOD" would literally mean a lower number and that would be less DOD, a smaller load, "50% and above DOD" would be a higher number, more DOD and a bigger load.


and this is the water diagram, that is where I keep it, it has never been lower than an eight to quarter inch below this, some months it didn't need any at all

 

[Zwy]

my float is currently set to 55.2 I didn't lower it yet, and it has been there for well over an hour, I don't want to turn it off yet so as soon as I am done with the measurements which I'll wait another couple hours to do then I'll set it to 54 like the crown specs.

to the comments on winter, most days, it will go into absorption/float cycle around noon, much sooner if its sunny,

What are the charging amps thru the afternoon? As full absorption is reached, the amps will continue to fall. Any loads that will pull from the bank and charging starts over again. While some chargers have a time limit and some battery manufacturers might have a time limit (max of 4 hours as an example), the charging amps really indicate whether full absorption is reached.

I've always recommended splitting banks in 2, one half being charged thru absorption, the other in use, then switch each half. This ensures reaching full absorption charge on a regular basis.

Not reaching full absorption on a regular basis (daily) will kill lead acid faster than anything.

only the worst multiday storms will keep it from charging completely, even with a light coat of snow I can pull 3 to 4A @140V to each inverter and the panels are set to 60° winter angle permanently and shed quickly, my load doesn't change seasonally, I heat exclusively with wood, maybe use a few more lights but I have big pressure tanks so its easy to use water more sparingly and the pump won't run, then the chargers only compete with the fridges and freezers which I will have fewer of this winter, and an occasional 300W for the propane on-demand water heater

if the grid is on I use that to charge, smoother current than a generator no?

All my wiring to the batteries is exactly the same length. I will try to notice if it bubbles at rest again and see if I can find a current.

here's why I was confused with the DOD, this a typo or editorial mistake then right? D being depth means lower is less higher is more, "below 50% DOD" would literally mean a lower number and that would be less DOD, a smaller load, "50% and above DOD" would be a higher number, more DOD and a bigger load.

It is stated correctly, if you discharge over 50%, then bi weekly. If discharge is under 50%, then monthly.

DOD is for Depth of Discharge

How deep the discharge cycle is.

View attachment 219936
and this is the water diagram, that is where I keep it, it has never been lower than an eight to quarter inch below this, some months it didn't need any at all

View attachment 219939