Can't reach float; dead cell

terrynew · October 2020

Hi, all. It's been 8 years since I've been on the forum, so before I ask my question let me catch you up with my tale of woe. Exactly ten years ago this month I moved into a new house with a new solar package: 2560 W of panels (12x230W), 800 Ah of 6V Rolls/Surrette batteries (two parallel strings of 8 in a 48 VDC setup), an Outback Flexmax 80 charge controller, and a Magnum 4448 inverter (4400 W at 48 V) with an ME-ARC remote panel.

Right from the start I had trouble keeping the batteries charged. Part of it was likely my hesitation to run the generator so often in winter (eastern Ontario, Canada), but I also think my package was badly matched. Despite long discussions on this forum and with the various manufacturers and two local solar installers, I destroyed my batteries TWICE yet amazingly Surrette replaced them under warranty. This all happened over two maddening years. And all while my wife was dying. So I decided in late 2012 to throw in the towel -- sort of.

Instead of selling everything for a big financial loss, I went hybrid: I hooked up to the local power grid (just across the street, so cheap; I wanted off-grid for independence), but I kept the solar package and wired it completely isolated from the grid system. It's hooked up to just six isolated wall outlets, and I use it for my biggest loads (fridge, freezer, HRV) year-round. This way my hydro-electric bill is super low, I can recharge the batteries from the grid instead of a generator, and in summer I can use the surplus power for assorted appliances thru a solar outlet in the kitchen.

That worked for seven years. But my wife passed away and I guess I didn't pay enough attention to the maintenance after that. Last year I noticed my charge controller (CC) log told me I wasn't reaching Float even on long sunny days; I normally get a few hours of float when the Absorb finished, and I would equalize if I had a week with no floating days. But my Inverter/charger's remote showed I was getting Absorb voltage (59 V) so I didn't take any action. I was away this past winter with the system shut down, and then COVID took my attention away from the problem. I'm now looking to sell the house, so I want to get the system back in working order. So here's my problem (sorry about the long intro, but I think it was important): My refractometer tells me most of my 48 2-volt cells (16 batteries, 3 x 2 V) have an SG around 1240 +/-5. But one is dead, reading 0000. I assume it's shorted? Is this why I can't reach Float?

I have a switch to employ both or either of my two 8-battery strings. If I absorb from the grid with just my string with the dead cell connected (I'll call that A-string), it absorbs and the inverter/charger's remote display says its in Float state, but my CC still says it's got 0 hours of Float that day. (Does the CC's log only reflect solar charging, not inverter/charger charging?) If I then Equalize from the grid on A-string, it doesn't reach my preset 64 V, staying around 56 V. But the current goes up to 63 Amps! I didn't notice that the first time, and after several hours the batteries were all hot and bubbly and the utility room had lots of hydrogen smell in it despite the temp-controlled fan being on. I had hoped the high current might repair the shorted cell, but no luck.

Should I just run my system off the B-string of 8 good batteries, forever more, taking me down to 400 Ah from 800? Is it practical to buy a 6V 400 Ah battery to replace the one with the dead cell, or is that a no-no? Should it be new or is a used one better? Does it have to be a Surrette?

I haven't tried a B-string Equalize off the grid to see if it has normal low currents, but that's my next action, unless you folks can suggest something else. It would be a drag on my selling the house if I can't figure this out.

Thanks so much for listening to me. Wishing you all a good day,
...Terry

BB. · October 2020

Terry, our condolences on losing your wife.... Everything else is secondary in life while and after that happens.

If you are selling your home--Then try to find a used (good) battery to replace the bad one. It is possible to put a couple smaller Golf Cart batteries in parallel (or series parallel) to limp along for another year or 3... Or, just buy 1x replacement battery.

It sounds like the A string cell failed shorted... You end up with one string as a 44 volt battery bank, and the second as a 48 volt bank. The 44 volt battery bank is going to pull all of the charging current and prevent your charger from "seeing" a 100% state of charge bank.

The A bank will get over charged (basically it is running an EQ with way too much current)--And the B string is just sitting there twiddling it thumbs (not really getting fully charged).

Checking the cells with a hydrometer--And see if you have any other bad/weak cells in either string. The A string is fully charged and EQ
'ed at this point (at least the remaining "good batteries" are...). The B string is probably not--And disconnecting the A string, and then fully charging the B string and EQ as necessary to confirm all cells are good or not.

Using a Voltmeter on each battery--Check and log the voltages (charging, discharging, resting) and look for "differences". For example, if the A bad cell is shorted--That one battery will read significantly low (up to 2 volts low).

If it turns out you have more than one cell/battery failing--Then you are left with the choice of replacing several batteries, replacing the whole bank, or piecing together a single bank to get the system working properly--And at least you have a good show piece for any buyer...

The problem is that off grid solar power, especially with grid connected power, you have to find a buyer that sees that as a plus for the property... There are lots of buyers that see anything solar as a negative (alls these wires running everywhere).

I (little real estate experience, and no off grid/battery/GT solar home sales) suggest that any money you put into the system will not be returned in a higher sales price... Getting a single string running reliably is probably the best you got going for you right now. I would be hard pressed for you to buy two new strings of batteries unless you had a signed sales contract in-hand that made it worth while for you to replace the bank.

Placing two golf cart batteries in parallel to replace one 400 Amp @ 6 volt battery--That may be worth it to you for get a few more years of life from the existing battery bank... However, I would worry that it would make many buyers skittish and hurt the price of your home.

You do want to match the old and new batteries relatively closely. You want the same charging voltage and current. If you can find a used (matching) battery for a good price--Not a bad way to save some money. A new Surrette (matching model) should work fine for your needs. It is not the cheapest solution, but at least you can get the system running quickly and move on for now.

If you are placing the home up for sale--Find a good agent (assuming that is the custom in your area) and work with them.$400 to $800 for a single replacement battery (guessing) is a small fraction of the value of your home... If it "cleans up" your system and lets is run as it has for the last 8+ years--Just bite the costs and move forwards. The existing bank--May have 6+ years of life left (with possibly another weak battery or two down the road). But nobody really can give you exact numbers. But it has your system working well again for not that much money. And you can work with your agent on any other issues to address before placing the home on the market.

What ever you do--Need to stop "boiling" the rest of the batteries in A bank--And you need to get B bank charged/EQ'ed as needed. Connecting B bank by itself for a week or so and get it charged/EQ'ed is a good start. And you use the hydrometer (if flooded cell) and voltmeter to make sure that the batteries in B bank are in good shape and they are equalized (SG should be within ~0.015 to 0.030 between 'high and low' cells in string--If there is a big difference then EQ the bank/string).

Note that EQ current should be around 5% to 2.5% of the AH rating of the bank... A 400 AH string * 5% = 20 Amps EQ current max... The 800 AH bank ~40 Amps max EQ (you do EQ after the string/bank is near/at 100% charged). Until you replace the bad battery--Keep the strings separate (running on B string until bad A battery is replaced). At this point you are "cooking" A string, and undercharging B string--Neither is good for battery health.

Warmest regards,
-Bill

Photowhit · October 2020

I would go with the best string you can make, equalize off solar and/or the grid and don't worry about it, it will likely do most of what you have been doing if it's in good shape, on those periods of sunny days, and only reduce your capacity on those long periods of cloudy days.

terrynew · October 2020

Thanks, Bill and Photowhit, for your good advice, and for your detailed explanations, Bill. I'll look for a used battery to get both strings up and running, and just use the good B-string for now. But even with that good string alone, I still have trouble:

Last night I charged the B-string from the grid, reaching float overnight after running through my inverter/charger's 7.5 hrs absorbing program (at 59 V; the remote showed 44 A). Then at dawn today, I started equalizing from the grid. But a half-hour into the 4-hr EQ program, my remote said it was charging at 56 V and 62 A! Of course I cancelled the EQ. I checked all SGs using my refractometer, and most were 1235-1250, with one 1270 and one 1215. There were no 0-readings (shorted cells). So why was I getting the dangerously high current, just as I had been with the A-string that had one dead cell? You wouldn't expect that to be due to the 1215 cell, would you? I could replace that battery with one from the now-inactive A-string if so. It's more likely something else is wrong, correct? Are there any settings on my inverter/charger or CC that might be causing this?

Incidentally, the B-string batteries were still boiling (slowly cooling down) when I measured their SGs this morning, with no input and no loads after turning off the grid charging. Should that absorbing (59 V) also have had a lower current than the 44 A I was seeing? Absorbing shouldn't cause boiling, yes? Same question: what could be causing the high current if there are no dead cells?

Thanks, all, for your thoughts... Terry

mike95490 · October 2020

You need to understand the difference between charging batteries GASSING and overheated batteries Boiling.

About noontime, in my battery shed, it sounds like a room full of snakes all hissing. That's 40 batteries gassing as they are being charged. They get about 2 degrees warmer than ambient and consume several gallons of water a month - all very normal.
If there was enough power to get them boiling, cases would be melting.
But you have something else wrong. As batteries heat up, the temperature sensor should be cutting back on the voltage.

You should not start an EQ cycle at dawn, you batteries are low, and need to be Bulk-Absorbed-Float first to get them fully charged, and then you start EQ. Don't wait several hours between cycles, go directly to EQ

What was your voltage and amps at the end of Absorb ?

You can't Absorb at 59V 44A and then EQ at 56V 62A. Those numbers make no sense. Were any loads running off the inverter ?
Why is Abs voltage higher than your EQ voltage ?

terrynew · October 2020

You ask important questions, Mike. Thank you for pointing to the temperature sensor as one possible cause of the high current. Anyone know how I can test if my temperature sensor is working?

I started my EQ at dawn when the inverter/charger had finished its absorb cycle and was keeping the battery at float, 50 V and 0 A (no loads other than the inverter/charger itself). The actual absorb voltage (49 V) was higher than the actual EQ voltage (56 V) because something was stopping the inverter/charger from reaching its EQ setting, which was 64 V. Perhaps that was due to the temperature sensor cutting back the voltage, implying my problem isn't the sensor. So again I'm hoping to find out why I'm getting such high currents. Anyone?

BB. · October 2020

You have one battery/cell that is near full charge (1.270--Full charge is near 1.265 to 1.275 at 25C/75F)... The rest are quite undercharged.

https://rollsbattery.com/public/docs/user_manual/Rolls_Battery_Manual.pdf (page 15)

*Check specific gravity regularly to confirm charging parameters are properly programmed, avoiding possible over/undercharge, resulting in possible cell damage or battery failure. Special order batteries may be shipped dry (acid shipped separately). To activate these batteries, start by removing the vent caps. Using approved battery grade electrolyte (1.265), slowly fill each cell until the liquid level reached halfway between the plates and the bottom of the vent well tube. (See Figure 8.) It is important not to over fill the cells as the acid will expand upon charging. If the cells are too full, the acid will overflow and require cleanup. Allow electrolyte to saturate into the plates and separators for at least 90 minutes.

The temperature of the electrolyte will rise and the specific gravity will drop. Once this is complete, place the batteries on charge at the finishing rate (5%-10% of the 20 hour rate). The rate may be increased if the battery does not begin to bubble and gas. Do not let the cell temperature exceed 52°C (125°F).

If the temperature becomes excessive or the cells begin to gas vigorously, reduce the rate of charge. Continue charging until the cell (or cells) reaches within .005 points of the specific gravity of the filling electrolyte corrected for 25°C (77°F). We recommend allowing the charge to continue for an additional 60 minutes to insure no further rise in specific gravity. A longer lower charge rate is better than a higher heavier charge.

FIGURE 7: Temperature Sensor TABLE 3: Specific Gravity vs State of Charge
% Charge Specific Gravity* (SG)
100% 1.255 - 1.275
75% 1.215 - 1.235
50% 1.180 - 1.200
25% 1.155 - 1.165
0% 1.110 - 1.130

NOTE: Specific Gravity is dependent on the electrolyte temperature. These values are for a temperature of 25°C (77°F). To adjust, add/subtract 0.003 for every 5°C (10°F) increase/decrease.

Very roughly, the battery, when charged at the proper voltage setpoint, is "full" when the charging current drops to 1% or less rate of charge (400 AH string, that would be 4 amps or less) for a "good" battery bank (no shorted cells, cell SG within ~0.015 to 0.030 sg units from high to low).

And once battery bank is fully charged, EQ around 5% (to 2.5%) of bank AH capacity (raise or lower EQ voltage as needed). 400 AH then 20-10 amps EQ current.

Use a voltmeter to measure the charging voltage across each battery--Ideally, you want to see the voltages to be very close to each other (more or less, a 0.010 volt per cell difference is equivalent to a 0.010 SG unit difference... A 6 volt/3 cell battery that is +0.030 volts higher than its neighbor is around ~0.010 sg units per cell higher.

Anyway... You may be at the point where you need to match cells/batteries into one string--And attempt to get that string working correctly.

Then focus on another method to see if you can recover the rest of the batteries... If you have a 6 volt (or 12 volt) battery charger--Charge each battery (or "matched" pair) separately... A big issue is that you are "cooking" (and potentially damaging) the good batteries while trying to bring up the weak cells in a single 48 volt string.

A battery is "recovering" if the sg continues to rise while charging (10% rate of charge, EQing at 5% rate of charge) when taking the SG reading every 1 hour... If you do not see a rise from one reading to the next (be sure to take temperature into account) between 1 hour readings--Then that is the "new" 100% state of charge SG measurement.

Depending on the "health" of the weak cells--It is possible/probable that some of those will never recover.

It is very possible that you have multiple bad batteries (batteries with one or more "bad" cells). When you had the one string with a "dead" (shorted?) cell in parallel with the second string... The first string the "good" batteries were "cooked" from overcharging, and the second string was undercharged and cells started to sulfate (hardening of lead sulfate into hard/dark crystals) resulting in a permanent loss of capacity (and inability to bring SG back to the ~1.265 sg range).

Make sure plates are covered with electrolyte before charging (plates exposed to air are quickly damaged/oxidized).. And only fill 1/2 way or so to prevent overflowing as batteries get hot/start gassing. Once the battery is fully charged (and warm), then you can fill rest of the way.

-Bill

Photowhit · October 2020

terrynew said:

Incidentally, the B-string batteries were still boiling (slowly cooling down) when I measured their SGs this morning, with no input and no loads after turning off the grid charging. Should that absorbing (59 V) also have had a lower current than the 44 A I was seeing? Absorbing shouldn't cause boiling, yes? Same question: what could be causing the high current if there are no dead cells?

So yes the batteries should 'out gas' while absorbing. The amount of distilled water used is often an indicator of a healthy system. If you don't need to add some every couple months, you likely have poor settings.

terrynew said:
I started my EQ at dawn when the inverter/charger had finished its absorb cycle and was keeping the battery at float, 50 V and 0 A (no loads other than the inverter/charger itself). The actual absorb voltage (49 V) was higher than the actual EQ voltage (56 V) because something was stopping the inverter/charger from reaching its EQ setting, which was 64 V. Perhaps that was due to the temperature sensor cutting back the voltage, implying my problem isn't the sensor. So again I'm hoping to find out why I'm getting such high currents. Anyone?

Somethings not right here...

I hate voltage numbers, but basically a flooded lead acid 12 volt battery that is fully charged is roughly 12.7 volts, x4 for a 48 volt battery bank = 50.8.

50 volts is not a proper 'float' voltage, usually around 54.4-54.8. 50 volts actually doesn't represent a fully charged battery.

Absorb voltage of 49 is likely a type-o and should be written 59. Having just looked this up for Rolls recently, Rolls suggests 60 volts.

I would guess there is a poor setting in your inverter/charger. I would suggest it's obvious you didn't start with a fully charged battery.

terrynew · October 2020

My mistake, Photowhit, I said the absorb V (both actual and setting) were 49, when I meant to say 59. So I think my inverter/charger is working properly when absorb-charging. But I suspect that my charge controller may not be working correctly. Something I mentioned in my first post happens all the time, including today: it's been cloudless all day, and yet the CC says I never went into float from absorb. Agreeing with that, the inverter/charger's display at noon in full sun never went above 54 V. So why can't a full day of sunshine take my batteries up to absorb voltage, let alone keep it there for a few hours to go into Float?

Bill, thanks again for all your detailed material. May I ask why I should measure the voltage of each battery while charging, when I can measure the SG directly with my refractometer?

Bill, I agree that I need to match batteries/cells into one good string. Do you mean picking batteries with cells that are within 0.015 SG units?

Yikes, to repair the bad batteries I balk at the amount of work you suggest (charging each battery separately and reading its SG every hour to see if it increases). But I'll try when I can (deer hunting season starts soon). I suspect the 100% SG level of most of my cells is .1240-ish. Should I charge at the slower 5 A or use the 15A setting on my charger?

Interesting technique, only filling cells half-way to avoid fluid loss. I usually fill to 1/4", and do get fluid loss on some of my batteries every time I have a full day of sun (or grid charging), so I suspect over the years that I've lost a notable amount of electrolyte.

Happy Halloween, all.
...Terry

Vic · October 2020

Hi Terry,

Very sorry to hear of the passing of your wife. Condolences to you and your family.

FLA batteries that need a good charge and Corrective EQ can need a large accumulated number of Ah to get done, as much as can be done.

The A and B banks seem to really need a full charge and that EQ. The large variation in the SGs of the cells can make the job of TEmp COmpensation of the charge voltage difficult, as one would expect a fairly large variation in the cell/battery temps, so, just where does the Battery Temp Sensor (BTS) belong?

Several things do not make sense -- with charge/EQ voltages, vs charge currents, but you probably do not have the Mate 3 and FNDC Thingies to measure actual battery current, etc.

The BTS for OB electronics is a 10 K Ohm nominal NTC Thermistor -- at 25 degrees C, it has 10 K Ohms or resistance, which diminishes as its temperature increases. The BTS uses four conductor phone cable, and the standard phone RJ connector. IIRC, the two connections on one end of the plug are connected together on one lead of the thermistor, and the other two plug connectors are connected to the other NTC lead. There is a decoder ring that can be used to translate from a "count value" in the FM 80, to determine the temperature that the FM 80 is reading. I have that table of values, and will try posting that, later. But, forget just where one finds that count reading on the FM.
EDIT: This is too far into the weeds, but the Decoder Ring for Outback RTS Count vs RTS Temperature equivalence is in BB Bill Post #10, here:
https://forum.solar-electric.com/discussion/comment/355066#Comment_355066

Perhaps by now, OB has changed the FM firmware to allow reading the BTS temperature equivalence.

BB Bills suggestion of using an external 6V (ideally), or perhaps a 12 V battery (two 6V batts in series), OH good, somehow, turned on underline, and am clueless of how to turn it off. [turned off underlining by selecting text and used ^u to toggle. -Bill B.]

Anyway, Good Luck, FWIW

BB. · November 2020

The check every hour... Don't have to do that. If the sg levels are low (1.23x or so)--It will take time to get up to the ideal ~1.265+/- sg level...

At this point, doing it to one or two pair (6 or 12 volt) batteries at a time--You will quickly see which ones are are not rising (i.e., you measure 1.230 initially, and a few hours later, you see 1.230). The idea is that if you see the sg numbers rising between readings, the cells are doing what they are supposed to do (getting charged, and SG numbers eventually approaching 1.265 or so).

The reason to measure battery resting voltage--Is a quick check on the health of the 3 cells overall (in each 6 volt battery). Easier and quicker than measuring SG for each cell and a quick sanity check on the whole bank... If 7 batteries read "the same" voltage, and one battery does not, then that one battery needs to be looked at and see what is going on (quick check of SG per cell, etc.). All 8 series connected batteries (8 x 6 volt = 48 volt battery bus) are "seeing" the same charging and discharging current. When you see one (or two) batteries that are "different" (high voltage reading or low voltage reading)--It generally means you need to take a detailed look at each cell (SG), cabling (especially when you have two or more paralleled strings).

A battery with high voltage during charging--Perhaps an open cell, or starting to sulfate (you may see one or more cells with low SG values).

A battery with low voltage during charging--Perhaps a shorted cell (2 cells with normal/high SG, one cell with low SG--Two cells "gassing", one cell perhaps not, etc.).

Using the voltmeter per battery under charging/discharging/resting is to help you quickly catch any cells/batteries that are "going south" before damage has been done...

For example, you may have a bad cable/connection... All the batteries are low in String B, compared to string A during charging (and String B is not getting proper charging current while String A is getting charged). A bad cable connection or open cell can prevent the batteries in string B from properly charging/discharging--And eventually they can sulfate if left under charged for days/weeks at a time--While String A is working its little heart out.

Charging current---

5% * 400 AH (per string or per battery) = 20 Amps.
2.5%*400AH=10 Amps....

If you want to doing things "quickly", the 20 amps (max) is fine--But you do not want to leave the battery(ies) unattended for full day... At 20 Amps, the batteries can overheat when EQing (and all cells are "balanced", or 'controlled' overcharge).

If you are going to leave for 12 hours--Then I would suggest closer to 10 amps maximum. The batteries will still EQ (may take 2x longer), but are less likely to overheat when left unattended for extended EQing (many hours at a time).

EQ charging should not be done for 24 hours at a time (example) unattended--Many batteries will reach their maximum temprature limits after ~12 hours of EQ @ 5%. Heat (and lots of excessive gassing) is not good for the batteries and (in rare cases), batteries can meltdown/charging gasses can explode (hydrogen+oxygen bubbles during EQ/excessive charging) from excessive temperature/heat buildup.

Once you do a couple batteries, you will see how it all works out for your batteries and their current state of health.

Typically "bad" cells/batteries show up pretty quickly during charging (SG stays low, one cell not bubbling) for shorted cells. Or very little bubbling, and low charging current for "open" cells. And in either case, one or more cells will probably see little improvement in SG readings.

-Bill

Photowhit · November 2020

terrynew said:

So why can't a full day of sunshine take my batteries up to absorb voltage, let alone keep it there for a few hours to go into Float?

Well lets look at your system, you have;

terrynew said:

2560 W of panels (12x230W), 800 Ah of 6v Rolls/Surrette batteries (two parallel strings of 8 in a 48 VDC setup

So you have 2560 watts feeding a battery bank of 800ah x 48volts, so The array has a maximum potential of generating 2560watts /48 v or 53 amps, and a more realistic potential of 2560 x .75 (Normal Operating Cell Temperature value of about 75%) = 1920 watts / 59 volts (charging voltage) = 32.5 amps. About 4% of the capacity of your battery bank during the prime hours about 2 hours on either side of solar noon, somewhat less during the rest of the day.

So if you have depleted the battery 30% over night (nights are getting longer) It will take longer to fully charge than you have direct sun light.

In general, I'd always prefer an over paneled system, at 4-5% of your battery capacity, it would require careful monitoring and generous generator use for 24/7 occupancy. Rolls allows up to 20% charging rate! So you could realistically have 5x the array you currently have.

I have a battery 1/3 the capacity and a larger array...

mike95490 · November 2020

Also, with such a low charge rate, you will seldom generate enough solar power to effectively bubble or gas the cells to stir them enough to de-stratify them.
You need to spend some mornings with a 4 hour generator run - say 6am - 10am charging at whatever the highest amp rate you can manage. and then let the solar attempt the absorb and maybe get to float. After 3 days of this regime, the batteries should be good enough to try a real EQ
And have distilled on hand to be able to keep the plates covered.

terrynew · November 2020

Hi all. Sorry for the 2 week absence -- freezer still empty of venison. :-( Let's see now: Vic, I agree that several things do not make sense, but to me that includes all your numbers and letters. (Sorry, non-techie.) As to your point about the states of the different batteries affecting charging, perhaps a better solution than BTS work is charging the batteries separately as Bill recommends. Those different states are also why I'll hold off on Mike's morning generator runs for now also (but that input will help once I have healthy batteries, so thanks, Mike.).

Photowhit, thanks for the calculations showing I'm underpaneled. Bought from a bad vendor, I guess. The one # I didn't quite follow was how the 32.5 Amps amounts to 4% of my 800 Ah battery bank over the 2 prime hours at noon -- Q1 is could you break that calculation down for me, please? Oh, and Q2 is could you also please explain how you made the big jump from that to "So if you have depleted the battery 30% over night (nights are getting longer) It will take longer to fully charge than you have direct sun light"?

Since I'm hoping to sell my house and don't want to double my #panels, but I can't change from a 48 V system, Q3 is does it make sense that going down to 400 Ah (just my 'B' string) is a viable solution, giving me half the power but keeping the batteries charged?

Q4 is does that require that I can get all eight batteries up to SGs near 1265, rather than the 1235-1260 range they're in now (avg 1245)?

So I guess I'll start Bill's recommendation of going offline and charging each B-string battery separately on my 6V-15A charger, until the SGs on each cell stop rising after a few hours. Thank you again, Bill, for the patience to explain things in such detail. You guys all rock.
...Terry

BB. · November 2020

I will try a bit here...

32.5 amps typical maximum available charging current (at 59 volts from your array) / 800 AH battery bank = 0.040625 ~ 4.1% rate of charge

More or less, 5% is the minimum suggested rate of charge for a solar power system with a lightly cycled system (weekends, sunny weather, etc.). 10% is the typical minimum recommended rate of charge for a deep cycle lead acid battery (for proper chemical reactions) and to quickly recharge a daily use system. 13% rate of charge is a typical "cost effective maximum" rate of charge for most solar power systems... You can go to 20% or even 25% rate of charge max--But that can be hard on battery banks and is only suggested for systems with a relatively small battery bank and lots of daytime (sunny weather) loads.

For gensets, suggest around 10% to 20% rate of charge.... Note that FLA batteries usually last longer with a 10% max rate of charge.

Note that is current to the battery bank... If you have a 5% load (800 AH * 0.05 = 40 amps @ 48 volts) during the daytime (say irrigation system)--And a 10% rate of charge solar array--That is really only 10% solar array - 5% load = 5% available for charging the battery bank.

FLA batteries are not 'great' batteries for solar... They are heavy, need water, and take time to charge, however they are cheap (LiFePO4 are much better, but much more expensive and some other issues to understand).

Say you have 5% rate of charge, most of your loads are at night (sun is used to charge, batteries to supply loads at night)...

4% * 800 AH = 32 Amp charging current
30% * 800 AH = 240 Amp*Hours of used capacity
240 AH / 32 Amps charging = 7.5 Hours of full sun per day (most places are around 3-6 hours per day of full sun per day, and worst during winter)
And it takes about 2 hours of absorb time for a 25% (or less) discharged battery). For a 50% or greater discharged battery, it takes around 6 hours of absorb time...
7.5 hours of "full sun" per day + 3 hours absorb for a ~30% discharged battery bank = 10.5 hours of full sun per day

Using a fixed south facing array (we can get better numbers for better solar estimates knowing where you are are "exactly"--But good enough for now):
http://www.solarelectricityhandbook.com/solar-irradiance.html

Ottawa
Average Solar Insolation figures

Measured in kWh/m2/day onto a solar panel set at a 45° angle:
(For best year-round performance)

Jan	Feb	Mar	Apr	May	Jun
2.63	3.66	4.48	4.74	4.75	4.92
Jul	Aug	Sep	Oct	Nov	Dec
5.04	4.83	4.23	3.19	2.30	2.09

With a 4% array--Need close to 10.5 hours per day of "full sun", and in October, looking at 3.19 hours of sun per day (average). Or ~3+ days to fully charge after a 30% discharge (70% state of charge) solar system...

The issue with series strings of batteries--If you have one cell (or one battery) with high resistance (open cell), it reduces the charging current for the whole string--And can overheat the one "high resistance" cell... And cause the rest of the cells to undercharge.

Again with series batteries--A "shorted" cell can cause your 48 volt bank to look like a 46 volt bank, and cause more charging current to to to the shorted cell/string... Over charging the rest of the cells/batteries in the string.

With Shorted cells in parallel strings--The shorted "cell string" will draw most of the charging current, and leave the "good string" of batteries under charged and supplying (or trying) 48 volts to the loads, while the shorted string is supplying 46 volts and almost no current to the loads.

At this point--Suggesting get one good string of batteries so you can limp along (reduced power usage)...And with one string you are now at 8.2% rate of charge--Which is much better for the surviving string.

Batteries only store energy. Having a larger bank does almost no good without having more charging current (more solar, more genset runtime).

In the olden days, when solar panels were $10-$30 per Watt--People were told to add more batteries to get more solar power--Not really a good suggestion back then either (when panels were expensive, fuel and batteries were "cheap"). Today, solar panels are "cheap" (~$0.50 to $1.00 per Watt for large format solar panels) and batteries+fuel is expensive--Adding more panels is almost always the "no-brainer" to add power to a solar power system (suggest typically 10% to 13% rate of charge minimum--Add more panels to lessen genset runtime in winter--Although, in the far north--No sun, no solar, need genset+fuel anyway).

Clear as mud?

-Bill

Photowhit · November 2020

Bill does a nice job. I popped in a couple times, but haven't the time to run numbers, but Yes, you would have a better balanced system with 400ah of battery bank.

terrynew · December 2020

Thanks, Bill, once again for that detailed explanation, and Photowhit too. I'm angry at my supplier for giving me such an underpowered charging system ten years ago, and disappointed in myself for not doing the work myself to check his figures. I guess he just wanted to sell me a bunch of expensive batteries? Who knows. In any case, I've now switched to just one 400 Ah string of batteries (8x6V).

It took me a couple of weeks to isolate each of my 16 batteries and give it 12+ hours of 15A charging until its SG was no longer rising. I've hooked up the best 8 of those batteries into my new single string. But the 8 years of undercharging has payed a toll, and those best 8 only have one battery with all 3 cells at 1265. The other 7 best batteries are 1250-1260.

So imagine my dismay when I powered up the new system and two sunny days in a row, with no loads, failed to even reach my 59 V absorb level!! They got up to 57 V max. I understand that the 8% rate of charge which my system can provide is less than the 10% recommended minimum, but why don't I even reach Absorb voltage on an unloaded system?? Confused and depressed.

Photowhit · December 2020

terrynew said:

...In any case, I've now switched to just one 400 Ah string of batteries (8x6V).

It took me a couple of weeks to isolate each of my 16 batteries and give it 12+ hours of 15A charging until its SG was no longer rising. I've hooked up the best 8 of those batteries into my new single string. But the 8 years of undercharging has payed a toll, and those best 8 only have one battery with all 3 cells at 1265. The other 7 best batteries are 1250-1260.

How high a voltage were you equalizing at? I think Rolls suggests a very high voltage compared to other battery systems.

I have looked it up recently of others, but I think it's 64 volts for a 48 volt system. After many years of undercharging they might be just too sulfated, but if you haven't tried at the recommended voltage, might give it a go. I've not got time to look it up this morning, I've got to go to town, but can look it up later, if you haven't found it.

8 years isn't the worse service from a Flooded Lead Acid (FLA) battery bank, I would expect 10 maybe 12 from a quality Rolls L16 battery bank, but I usually use 7-10 for L16's as expected battery life.

BB. · December 2020

You need to look at the details during charging... When the voltage stops rising, is the current also falling (i..e, the charge controller is probably limiting the charging voltage) or is the charging current still high (battery bank limiting the voltage rise--Possibly a bad/shorted cell???)...

-Bill

Dave Angelini · December 2020

An EQ needs to really stay at a solid voltage the whole time. Below is a graph of a charge and an EQ.
If you can't maintain the voltage you need to use a genset. This is of course the wrong time of year for doing corrective EQ's. A battery system need regular preventative EQ's and summer is the time for fixing problems like this. Clouds can ruin the effort of doing long EQ's and you probably need 5 to 25 hours (not all at once) of EQ if the SG is bad. How far off are you?

I did not read this thread so I may have missed something. Good Luck! The Solstice is next week!

Image: https://us.v-cdn.net/6024911/uploads/editor/nh/1m7tehsggja0.png

Clouds below! No Bueno !!

terrynew · December 2020

Thanks for your replies, guys. Photowhit: I wasn't equalizing the string, I was hooking up each battery individually to a 6 V 15A battery charger for 12+ hrs (until there was no further SG rise).

Bill: I checked the SGs of each cell multiple times during each battery's 12+ hr charging, and can confirm that none of the 8 batteries in my new string has a cell below 1250.

Dave: I don't need to use a genset to externally charge my batteries, as my house has two isolated and independent power systems feeding separate AC outlets - I'm hooked up to the community's hydro lines for most of my outlets and my solar power feeds other AC outlets close to my fridge, freezer and HRV. So I can plug into the grid or into the batteries, depending on the battery status. And I can flip a breaker to charge my string of batteries from the grid (thru the inverter/charger, as if it were coming from a genset).

So I take it from all three of you that the best action for me now is to do a grid-sourced Equalize on my new string of batteries. I've just started a bulk charge, and once it gets into Float then I'll start an Equalize from my inverter. Ahh, but wait, even as I type this ten minutes after starting the bulk charge, I see that my inverter/charger's remote display is telling me it is now Absorb charging. That's good, except it's doing so at 57 V and 62 Amps!! So I appear to be back to a dangerously-high-current problem that I described way back in the first post of this thread. Yet I have no dead cells in this string. What would you do at this point, please?

terrynew · December 2020

I stopped the grid-fed 62 A absorb charging, by the way. Right thing to do?

BB. · December 2020

At 1.250, the battery bank (or at least that cell) is still somewhere between 80-90% state of charge (roughly, assuming temperature corrected specific gravity readings).

Take your voltmeter and measure (and log) the voltage on each battery when under your 62 amp charge... Of course, all 8 batteries should add up to your bus voltage.

I have seen/read here weird things about DMMs with weak/near dead batteries. They would give consistent readings across a voltage, but it would be consistently off. And at different voltages, they would read low or high too (different than the other voltage). The readings ended up being 'impossible' (such as the X batteries in series voltages did not add up to the bus voltage). It seemed that (at least those) DMMs did not say they had a "dead" battery(ies).

Look for other issues too.. One (or a couple) cells gassing more (or less) than the rest of the cells.

Are the batteries getting hot.

Usually "sulfated" batteries have higher resistance and lower SG readings than "good" batteries... So seeing high charging current and low bus voltage does not seem to indicate an sulfation failure.

SG is related to battery resting voltage... Single-cell open circuit voltage = specific gravity + 0.845

That means if all your cells were at 1.250 (old/sulfated)... And 1.265 was original full charge then

0.015 loss of SG * 24 cells in series = 0.36 volt drop in resting bank voltage (and should you drop nominal charging voltage by 0.36 volts to prevent too much gassing??? Not sure).

If the battery bank is not getting too hot, and the cells are not "bubbling" (aka rolling boil) but "fizzing", it seems that it is taking more charge--If you do not see the SG rising in any/all cells--That that is your new 100% state of charge.

Rolls/Surrette batteries are known to benefit from cycling/Several EQ cycles over time.

Since you really have nothing to lose--Try cycling them to 75% or down to 50% state of charge. 25% discharge * 400 AH string = 100 AH, or 50% discharge = 200 AH discharge.

First, see if the bank supports your loads (i.e., one to two days of running home of battery bank--Or whatever you "need"). If the bank does not support your needs/expectations (and all the cell SGs are in line, voltage across each 6 volt battery is consistent with the rest of the batteries, i.e., nothing stands out as bad)--Then the bank is a candidate for replacement.

Then go through the whole recharge to >90% and EQ cycle again.... This make take days or even a couple weeks or so....

If you have done this several times and the battery bank still does not support your needs--Then not much more you can do for that bank.

-Bill

Dave Angelini · December 2020

Test leads can do some weird diode affects also. Anything the user touches we use to say in metrology....

Photowhit · December 2020

terrynew said:

Thanks for your replies, guys. Photowhit: I wasn't equalizing the string, I was hooking up each battery individually to a 6 V 15A battery charger for 12+ hrs (until there was no further SG rise).

So what are you charging with?
This sounds like an Auto battery charger. They aren't usually designed for deep cycle batteries, but some are and there are others that might be used to present 15 amps, but voltage is a major factor in battery charging.

Just so we have points of reference, you are currently charging through the Magnum 4448?

terrynew said:

... once it gets into Float then I'll start an Equalize from my inverter. Ahh, but wait, even as I type this ten minutes after starting the bulk charge, I see that my inverter/charger's remote display is telling me it is now Absorb charging. That's good, except it's doing so at 57 V and 62 Amps!! So I appear to be back to a dangerously-high-current problem that I described way back in the first post of this thread. Yet I have no dead cells in this string. What would you do at this point, please?

This sounds like the preset charging voltages for you Magnum are still being used. 57 volts would be a max voltage for AGM's. Perhaps setup by you or original setup, expecting the batteries to be used in standby mode and not cycled. Presets are usually done for AGM batteries, since over charging will destroy them quickly.

Rolls recommends around 60 volts for a bulk/absorb setting and 64 volts for equalizing (check for your specific battery, but large 6 volt (L16 size) are likely this range. You can search your battery type add manual and perhaps pdf to search.

So Voltage has a lot to do with charging Flooded Lead Acid (FLA)batteries. I think I've been through this before, but here are the basic stages of battery charging. Generally people suggest waiting until the battery is completely charged before starting an equalizing cycle. This time of year, I admit to doing much as you are and starting the cycle an hour or 2 after the battery reaches Absorb. Once I can see the battery absorbing falling to below 5% of capacity.

Here are the stages of charging and how it works with FLA batteries;

During charging, there are basically 3 stages of charging, Bulk, Absorb, and Float.

BULK;

First thing when charging starts you will be in bulk, the voltage rises from what ever the system voltage was to a set point, around 14.5 volts. At that point the Charge controller stops the voltage from rising. Higher voltage can damage sealed batteries.

ABSORB;

Once the battery hits the preset point the charge controller keeps it at that point. Your batteries are roughly 80% full. Flooded batteries will start accepting less current at 80-85% full AGM/Sealed may go a little longer before accepting less current.

On many controllers you can set this point, Some will have different presets for Flooded, and sealed batteries, or flooded, AGM, and sealed batteries.

The charge controller has a couple ways to know when to switch to float, Most inexpensive Charge controller are just timed for 1.5-2 hours. Some will also see less current flowing through the charge controller and shut it down when minimal current is flowing through the controller. On more expensive charge controller. You can set battery capacity to give the Controller a better idea of when to stop. you can also set a longer Absorb time. Or set 'end amps' a amount of amps flowing through the charge controller to stop Absorb and switch to the final stage.

FLOAT;

Once the Controller has determined the battery is fully charged it reduces the voltage to a point where very little current is flowing to the battery. This will prevent the battery from over charging and heating up.

While in 'Float' the charge controller watch for voltage drop, which would indicate a load. If the voltage begins to drop the charge controller will allow as much current to flow from the panels/array to compensate and maintain the voltage. If the voltage can be maintained, the load will in essence be running directly off the array/solar. If the voltage drops below the preset float voltage, the controller may start a whole new cycle if it stays there for a period of time.

So equalizing is a controlled over charging of the battery to help recombine sulfates into solutions (what we read as changes in the Specific Gravity / SG) It also helps reduce stratification in tall batteries and lightly cycled batteries. Here's Trojan's discussion on Equalizing.

Trojan Tips 5 - Equalization is Key to Extending Battery Life and Performance - YouTube

BB. · December 2020

As a recap--Terry had 2x parallel strings of 400 AH Rolls batteries @ 48 volts.

In his two strings, he had several bad cells/batteries. Took the bank down and recharged each 6 volt battery one at a time. Batteries that did not reach absorb or showed other signs of failure were set aside... After all the batteries were gone through, took the "best" 8x 6 volt batteries and got the one string (400 AH) running again. Originally around 4% rate of charge, with 1x string now at 8% rate of charge.

Plus now has grid power.

-Bill

terrynew · December 2020

Good recap, Bill. So my current issue (sorry) is, why do I get 60 A and higher during absorb and equalize charging through my Magnum 4448 inverter/charger (which is being fed by the grid)? The batteries boil after a few hours of this, and I fear leaving them for the Magnum's absorb setting of 7.5 hrs at 59.2 V or its EQ setting of 64 V at 4 hrs.

Another related mystery is, why does the Magnum when charging from the grid switch from Bulk to Absorb charging at 57 V instead of its 64 V?

And the third mystery is, why do I fail to climb to Absorb charging during solar charging through my Outback FM80 charge controller, with no loads, even after two full (winter) days of sun with panels tilted at 60 degrees and the FM80 set to Absorb at 59 V? My Magnum says the voltage never gets above 57 VDC, and the FM80 log shows 0 hrs absorb time those days. This happens after I've confirmed all 8 batteries (24 cells) have SGs around 1250-1260.

(Photowhit, you're right, I charged each individual battery using an MotoMaster auto battery charger. It had a Deep Cycle battery setting, 6V and 15 A, which I confirmed while charging each battery by using a clamp ammeter.)

Tomorrow I'll follow Bill's suggestion and report back: "Take your voltmeter and measure (and log) the voltage on each battery when under your [Absorbing] 62 amp charge... Of course, all 8 batteries should add up to your bus voltage."

Bill, your second suggestion is to "Try cycling them to 75% or down to 50% state of charge. 25% discharge * 400 AH string = 100 AH, or 50% discharge = 200 AH discharge." and "Then go through the whole recharge to >90% and EQ cycle again.... This make take days or even a couple weeks or so." My question here is, how do I know when I've cycled them down to 50-75% (100-200 Ah)? I can plug in my loads (HRV, fridge, freezer) and observe the VDC drop on my Magnum's remote display, but I can't see where I observe the Ah consumed.

You guys have a ton of patience with me, and I'm very VERY grateful. Thanks...Terry

Vic · December 2020

Hi terrynew,

Regarding the FM 80 showing NO Absorb time, it is possible that your batteries are cool, or cold, which would significantly raise the voltage needed to reach Absorb. 48 V LA batteries will be compensated, 0.12V, for every degree C below the 25 C reference temperature that the RTS measures as the battery temperature.

Have read this entire Thread, and it was asked if you are using the OB Remote Temp Sensor (RTS), but do not remember the answer, so the above may be bogus.

AND, your quality Hydrometer is a great tool to measure actual temp-compensated SG (at least of your few Pilot Cells).

FWIW, Good Luck, Vic

terrynew · December 2020

Thanks for the suggestions, Vic. I do indeed have the RTS, and the battery box is indoors and room temperature (in fact dangerously hot when charging at those high currents I mentioned). I use a refractometer rather than a hydrometer, as I found the former to be more accurate and less messy.
Take care... Terry

Vic · December 2020

OK, Thanks Terry. Good Luck, Vic

terrynew · December 2020

Hi, all. Bill had recommended I get the voltages across each each of the 8 batteries while I was Absorb charging at high currents. I did this today, while the remote monitor was reading 58 V and 64 A. The individual voltages ranged from 6.84 to 7.64 V, with most at 7.1 V, and this totals 58.1 V. This is a normal voltage spread for 8-year-old batteries with SGs averaging 1250, right?

I still don't know why my inverter/charger charges at such dangerously high currents. I've just emailed Magnum TechSupp to ask, but I'm past warranty life so they may not answer. Any ideas?

By the way, I have a correction. In my Dec 19 post, my second mystery was "why does the Magnum when charging from the grid switch from Bulk to Absorb charging at 57 V instead of its 64 V?" That 64 V should have read 59 V, my Absorb voltage setting. It's still an open question. Do both questions imply a problem with my inverter/generator, not my batteries? (Incidentally, after a couple of hours absorbing, the current drops to 42 A but the voltage is still 57-58, not the expected 59.)
...Terry

Can't reach float; dead cell

Comments

Ottawa
Average Solar Insolation figures

Categories

Can't reach float; dead cell

Comments

OttawaAverage Solar Insolation figures

Categories

Ottawa
Average Solar Insolation figures