Undercharged Battery Bank? Low Specific Gravity

Skipmca

I live in the mountains in Colorado.  Temps have been quite variable for the past month with nighttime temps averaging about 35 degrees.  A couple of nights it dropped to the 10s. I have a new battery bank (in service for about 5 weeks) of 12 Rolls 6-volt series 500 in 3 strings parallel/serial for 24 volts and a Flexmax 60 CC.  The batteries are outside in three separate vented battery boxes surrounded by R-39 insulation inside a plywood box.  The battery temp sensor hooked up to a Conext SW 4024 Inverter shows an on-averge battery temp of 50-53 degrees F. 

The float voltage is set to 27.8 and the absorb level is 30.8.  The CC logs show that the max voltage over the last 5 days averaged about 26.5 volts.  During the same period, the voltage has been dropping overnight to 23 to 23.5 volts and once to very close to cutoff at 22.5 volts.  I checked the SG on all of the batteries.  The SG of each of the cells in each battery are close, but the readings are very low: 1.10 to 1.15. The water level is where it should be, about 1/4" below the battery tube and above the top of the cells.

The CC logs appear to show that the system has not been absorbing for several weeks.   I'm guessing the bank is being severely undercharged. All battery connections are firm and not over tightened. The average load is probably 300-400 watts during the day with occasionally up to 1800 or so for brief periods (microwave, hair dryer).  At night the load is probably 100-150 watts.

My questions are:
Will absorbing raise the SG and with it the SOC?  I plugged in my generator boost and the voltage eventually rose to 27.8, but still did not go into absorb mode.  Is it possible that absorb mode (if that is what is necessary to raise the SG) would kick in if I lower the float voltage setting?  Is the low SOC damaging my new batteries?  I will take the whole bank offline and run my house on generator until I fix this if necessary.

Thanks for any help you can give and let me know if I need to supply more information.

Skipmca

BB.

Welcome to the forum Skip,

Just to make sure, you have an accurate DC volt meter and are measuring actual battery bus voltage and it never exceeds 27.8 volts?

Assuming that is true, your battery bank is simply not receiving enough charging current and/or hours on charge.

A solar charge controller should charge maximum available current until your battery bus voltage hits ~30.8 volts, hold it there for something like 6 hours (2-6 hours, in your case, lots of time because the battery is severely under charged). And after it hold the Absorb voltage for 2-6 hours, drop back to 27.8 volts float.

We probably need to know more about your system:

• Battery bank AH and Voltage (is this a 357 AH @ 20 hour @ 6 volt battery, or a 24 volt @ 1,071 AH battery bank?)
• Charge controller Outback FM 60
• Solar array (XYZ panels, xxx Watt, YY Vmp, ZZ Imp), number of panels and configuration.
• Where roughly is the system (nearest big city)
• Array free of shadows/shading during most of the day
• What is your genset and DC battery charger type/size/configuration

This is a very big (and nice) battery bank--I fear you simply have too small of Solar array (Wattage) and/or the array is miss-configured (improper Vmp-array).

Do you have  a DC current clamp type meter so you can verify charging current to the battery bank during the day?

https://www.amazon.com/gp/product/B019CY4FB4 (something like this?)

At this point, your battery bank sounds severely undercharged and will be damaged (if not ruined) if not properly brought back to >90% state of charge quickly. The longer it stays at low charge, the faster it will sulfate.

Do not fill the cells any more (1/2 full, plates at least covered for now). If you do get vigorous charging voltage/current, the batteries will heat, gas form, and possibly puke out electrolyte during charging (gas in battery, electrolyte expanding as it gets hot, electrolyte spitting out caps/top of cell/making mess).

-Bill

706jim

I'm assuming you're charging this bank with solar panels. If so, what is the array capacity?

mcgivor

Judging by the 60A controller it would appear the PV capacity is not sufficient to provide sufficient charging for the battery capacity, even in perfect conditions if full time off grid. The best course of action would be to separate each parrallel string, charge each string seperately to float, then equalize, when all done reconnect in parallel, during this process there should be no dischargeing whatsoever. 

Hopefully they will recover, it will take time but there are no other options. The single string approach will ensure each receives a ballanced charge, the time taken will be almost the same.

Skipmca

Thanks all!

- Battery bank: 428 AH @ 20 hour @ 6 volt

- 5 Solar World Sunmodule SW 320 XL mono panels

• In parallel
• Max power 320w
• Vmp 36.7
• Imp 8.78
• South Park Colorado 65 miles from Colorado Springs 
• Off-grid / No shadowing

I have a multi-meter. I will check the bus voltage.  That would be the positive and negative leads going into the inverter from the batteries, correct?

Looking into a clamp type meter.

What is meant by 'genset and DC battery charger type/size/configuration'? I just have a 7500 watt propane generator plugged into the inverter.  Apologies for my lack of understanding of some to the terms.

Skip

BB.

Ok, assuming that is a single 6 volts @ 428 AmpHour battery in 4x series for 24 volt by 3 parallel strings for 1,284 AH battery bank. That is a very large battery bank for 24 volts.

In general, for a flooded cell lead acid battery bank, you would want 5% to 13% (to 20%) rate of charge. 5% is low, but can work for a weekend/(sunny) seasonal cabin or emergency backup. For for full time off grid/following the battery mfg. recommendations, you really need a 10%+ rate of charge (a larger battery bank needs a large solar array to properly recharge the battery bank)... And example would be:

• 1,284 Amp*Hour * 29.0 volts charging * 1/0.77 solar panel and controller derating * 0.05 rate of charge = 2,418 Watt array minimum
• 1,284 Amp*Hour * 29.0 volts charging * 1/0.77 solar panel and controller derating * 0.10 rate of charge = 4,836 Watt array nominal
• 1,284 Amp*Hour * 29.0 volts charging * 1/0.77 solar panel and controller derating * 0.13 rate of charge = 6,287 Watt array typical "cost effective" maximum
• 1,284 Amp*Hour * 29.0 volts charging * 1/0.77 solar panel and controller derating * 0.20 rate of charge = 9,672 typical absolute maximum battery charging current (FLA)

Your solar array:

• 320 Watt array * 5 panels = 1,600 Watt array

So, from a battery bank charging current point of view, you have about 2/3rds the minimum solar array you should have for proper charging.

And if you where using 60 amp MPPT charge controllers (typically would suggest >= 80 amp controllers):

• 4,836 Watt nominal array * 1/29.0 volts charging * 0.77 panel+controller derating * 1/60 amp MPPT controller = 2.1 parallel 60 amp MPPT charge controllers (plus two array circuits)

Normally, for an MPPT charge controller, you would want to see 2x of your present solar panels in series, and 2 or 3 parallel strings to get your Vmp array voltage higher. MPPT controllers need higher input voltage (~1.3x * 1/0.8 Vmp-array-hot derating = ~1.6 Vbatt charging voltage) to work optimally:

• 29.0 volts charging * 1.6 Array voltage uprate for MPPT charging on hot day = 46.4 volts minimum nominal Vmp-array-std voltage

Your array is Vmp~37.7 volts and not great for charging a 24 volt battery bank on hot days (Vmp falls as cells/panels heat up).

For AC inverter-charger, the charging current from a generator would typically be around 10-20% of battery bank AH capacity... Or:

• 1,284 AmpHour * 0.10 rate of charge = 128.4 amps typical
• 1,284 AmpHour * 0.10 rate of charge = 256.8 Amps max

Then there is the load on your system... Guessing from your first post:

• 400 Watts * 12 hours per day = 4,800 WH per day
• 150 Watts * 12 hours per night = 1800 WH per night
• Total ~ 6,600 WH per day 

If you assume 2 days of backup power (no sun) and 50% maximum battery discharge, the "optimum" battery bank would be:

• 6,600 WH per day * 1/24 volt bank * 1/0.85 AC inverter eff * 2 days storage * 1/0.50 max discharge = 1,294 AH @ 24 volt battery bank...

Your bank is sized well for your (my guess) loads. I would really be suggesting a 48 volt battery bank (a bank over ~800 AH should look at next step up in voltage):

• 1,294 AH / 2 (going from 24 to 48 volt bank) = 647 AH @ 48 volt battery bank

For various reasons, this generally works better (smaller diameter copper cables and 2x more solar panels per MPPT charge controller because of the higher battery bank voltage). But there are some issues too with the higher voltage wiring and solar array that make things a bit more complex.

And then there is sizing the solar array based on your loads and where you live... Colorado Springs, fixed array, facing south
http://www.solarelectricityhandbook.com/solar-irradiance.html

Colorado Springs
Average Solar Insolation figures

Measured in kWh/m2/day onto a solar panel set at a 51° angle from vertical:
(For best year-round performance)

Jan	Feb	Mar	Apr	May	Jun
4.55	4.97	5.65	5.77	5.90	6.12
Jul	Aug	Sep	Oct	Nov	Dec
5.81	5.64	5.99	5.76	4.73	4.36

So, the average harvest from your 1,600 Watt array would be:

• 1,600 Watt array * 0.52 off grid AC system eff * 4.73 hours of sun (November) = 3,935 WattHours per average November day

And this appears not enough hours of sun/solar panels to sustain your daily loads... If you where to up the array to a 10% array or 4,836 Watts:

• 4,836 Watt array * 0.52 off grid eff * 4.73 hours of Nov sun = 11,894 WattHours per Nov day from 10% array

And, for the most part, for base (required daily loads), I suggest that you plan on using ~50-65% of your predicted harvest (allow for some clouds, quicker recharging after bad weather, and general reduction in genset usage):

• 11,894 Watt*Hour harvest (10% charging and November sun) * 0.65 base load fudge factor = 7,731 WH "base load" from 10% November array

So, lots of numbers and assumptions... But my guess from 1,300 miles away, is you need to add dramatically to your present solar array, rewire the array to 2 panels (of the same type) in series for higher Vmp-array, and add a second charge (probably 80 amp or larger) MPPT solar charge controller.

Longer term changing to a 48 volt battery bank and larger AH batteries would be a suggestion--Especially if you want to upsize the system in the future.

Before you make any changes--I always suggest making some paper/back of the envelope designs first--Always cheaper to make mistakes on paper vs after you bought the hardware and did the installation/work.

My thoughts and guesstimates, which may or may not be accurate/correct for your needs. Your thoughts?

-Bill

BB.

DC battery bus voltages matter everywhere... Too small of copper cable can cause high voltage drops from battery bank to AC inverter (one example). But for the batteries, actual battery voltage on the terminals is what matters to the batteries themselves (too much voltage drop, dirty/loose/corroded connections, etc. all affect battery voltage).

And you can measure the voltage on each battery in your system. The voltage on all batteries should be very close to the same (on the order of a 1/10th of a volt). If they do not match, then there may be wiring/battery/cell issues which need to be investigated... Or just a good "equalization" charge (which is hard with your present small/lower voltage solar array).

There are also optimum ways of cabling battery bank wiring... Here is a good explination:

http://smartgauge.co.uk/batt_con.html

-Bill

Estragon

How far are the panels from the controller, and what size wire from combiner to controller?  Have you measured voltage at the controller PV in terminals when the panels are in full sun?

Estragon

I would measure both PV In and DC (battery) out charge controller terminal voltages.  I'm wondering about the PVIN in particular though, as I'm wondering if there's enough voltage "overhead" for the controller to charge properly.

jonr

> I plugged in my generator boost and the voltage eventually rose to 27.8, but still did not go into absorb mode.

This shouldn't happen and you need it resolved asap.  Do you have an inverter/charger that will charge batteries from the generator?  What model?

Skipmca

jonr: Yes, the Conext 4024 is an inverter/charger.

Estragon: The panels are 2-3 feet from the CC with 6AWG cable. I measured the voltage coming into the CC at the combiners: 40.3 volts.  It has been overcast/snowing the past two days however.  Probably won't be full sun until 2 days from now.

BB: I will eventually get the panels and CC upgraded, but given the costs, I'm looking for short-term solution.  Using your figures, if I remove one string from the bank, I can get a 10%+ rate of charge with one additional panel giving me a 1,920 watt array.  I can run my house on the two strings.  The bank I had before was 625AH and that was adequate.  24 volts would also be ok for that size bank, agreed?

Someone suggested that I charge the batteries one string at a time to get them to full charge.  That seems like a good suggestion.  After that I could remove one of the strings.  If I do that, I would move the 4 batteries to my storage shed for a stable ambient temp.  I could keep fully charged batteries in storage for a few months, I believe.  What do you think?

BB.

Regarding storage of FLA batteries... Yes fully charge and equalize the batteries. Store in a cool place and either float charge them or recharge them ~1 day out of every month (self discharge). That is for ~75F batteries... For colder batteries, self discharge is less. For hotter batteries, self discharge is higher.

As I understand, your battery bank is 6 volt @ 428 AH... 2x parallel strings would be 24 volts at 856 AH:

• 856 AH * 29 Volts * 1/0.77 panel+controller deratings * 0.10 rate of charge = 3223 Watt array for 10% rate of charge for 2 strings of batteries...

Yes, I would add one panel (at least) to the existing string so you can setup 2s x 3p (6 panels total). The typical peak available charging current would be:

• 6 * 320 Watt panels * 0.77 panel+controller deraing * 1/29 volts charging = ~51 amps peak... Your 60 Amp controller would be fine
• 60 amps controller * 29 volts charging * 1/0.77 panel+controller derate = 2260 max over paneling of controller suggested
• 2,260 Watts max / 320 Watts per panel = 7.1 panels max

So, the 6x total panels for your 60 Amp controller is about the maximum cost effective array--Although, making it 8 panels in total (2s x 4p) would not be the worst thing in the world (MPPT controllers will safely limit their maximum output current--To 60 Amps in this case. You will have your output capped by the controller during the middle of clear/sunny/cool days--But at other times, the more panels will give you more overall harvest per day without spending $$$ on additional controllers.

And to add to the conversation, if you went to a 48 volt bank, the same 60 amp MPPT controller could have a 2x 2,260 Watt array (double the power support, with the same single controller). However, it will require some further research... Vmp-std rises in cold temperatures. 3 panels in series (3*Vmp-std=3*36.7volts=) 110 Volts Vmp-array-std... That is probably too high of voltage array for an FM 60 in Colorado (again, need to check the specs and string sizing).

And placing only 2x 320 Watt panels in series--Not really great for charging a 48 volt battery bank (59 volts charging * 1.6 MPPT and heat fudge factor = 94 Volt Vmp-std array voltage suggested).

In any case, still need a lot more solar (or generator support) to keep even 2/3rd bank "happy" at 10% rate of charge.

A 4kWatt AC inverter will draw (at maximum sustained power rating):

• 4,000 Watts * 1/0.85 AC inverter eff * 1/21.5 battery cutoff voltage = 219 Amps

As you see, that requires some pretty heavy cable (and breaker/fuse) to sustain that level of 24 VDC current at 100% rated power. And why we suggest going to next higher voltage battery bus for large loads (and/or large AH battery banks).

-Bill

Skipmca

Bill:
Thanks so much for the time and effort you have invested.

Since I have a Conext SW Inverter/Charger, I checked the settings for Absorption settings.  I found the Absorb voltage was set higher (32v) on the Inverter than the absorb voltage on the FM60 (30.8).   I changed that value to 30.8 and the Inverter changed to Absorb mode with the generator providing AC in.  I'm hoping that will bring the SG up, but I don't think it will make it to 30.8.  Tomorrow, I will disconnect two of the strings and see if the CC will charge the one string adequately.

BB.

Skip,

It will also depend on how much energy you are pulling concurenlty with charging... To properly/fully charge the battery bank, a 5% rate of charge is probably needed (as a minimum).

If you have other loads at the same time and charging with solar, the rate of charge to the battery will be less...

You may need to fire up the genset to recharge the battery bank (or dramatically reduce loads when the sun is up).

As a rough starting point, a wild guess would be approximately 3 kWH of charging for every 1 gallon of propane, or in AH:

• 1,000 Watt*Hour / 29 volts = 34.5 AH per kWH
• 34.5 AH * 3 = 104 AH per 1 gallon of propane

If your bank is at 50% state of charge, again very roughly:

• 856 AH bank * 50% = 428 AH to be recharged (2x parallel strings)
• 428 AH to recharge / 104 AH per gallon = 4.1 gallons of propane to fill 2 parallel strings from 50% to 100%

It is actually more complex than that, but it does give you very rough idea of what it costs to run the genset to recharge your bank.

My numbers may be wrong, because I think I am understanding a different bank AH capacity than you are telling me when pulling 1 string of batteries...

-Bill

Skipmca

No worries on the cost.  I'm protecting a hefty investment in those batteries!  Got kids and grandkids coming for Thanksgiving and I sure would like to get the system running well for that!

Thanks again for all of the help.  

mcgivor

Getting the batteries fully charged should be the top priority, the longer it's delayed the more damaging it is, as sulfation becomes more established. This would  mean using the generator exclusively during this process, not ideal but sometimes it takes drastic measures, the batteries are probably the most expensive part of the system so worth the sacrifice.

Once each string is fully charged then equalized connect in parallel, then use energy conservatively until additional PV is installed,  use generator support to supplement the solar as required so as not to repeat the undercharging which caused the problem in the first place.

Personally I wouldn't remove a string as this would cause imbalances when it is installed at a later date, but that's just an opinion.

Once completed, setting the inverter's low battery cut off higher to a higher value to prevent deep discharge should be considered if regular generator useage is not desired.

By the way what are your loads? 

Skipmca

mcgivor:
Thanks for your suggestion regarding separating the strings to charge them.  I will be doing that and then reconnecting only two strings until I can upgrade the panels and CC.

jonr

+1 on using the generator to do an equalization.   And using it on an ongoing basis to supplement the solar.   And for splitting the string while doing this - not because a low "% of capacity"  won't eventually charge a battery, but because low charge rates can't stir the electrolyte (I'd be interested if anyone doesn't agree that this is the reason for % charge recommendations).

Estragon

As I understand it, high charge rates in bulk don't stir electrolyte much.  Most of the heating and gassing that does the mixing is near the end of absorb at lower current.

mike95490

For the most efficient use of fuel, run the generator early in the morning, so that Bulk is completed by 10am, when solar should be sufficient to take over Absorb for a couple hours, leaving a couple more hours for EQ.  This loads your generator fully, rather than burning fuel for just a slight charging.