First post, seeking knowledge.

WayneH

First post.
My off grid system is 9 years old.  My batteries are not happy this winter.  I think.
I need to get my situation under control. 48 volt, 24 Rolls Surette S-550.  Magna, Midnight,
4590 watts of solar in the Southwest.  Ambient temperature power shed. Thanks if this is something 
that you understand. 

706jim

24 batteries? 5k panels? If so, probably not enough solar to charge them fully. But of note, your batteries are 9 years old. Mine too and some of them have failed after 9 years.

BB.

Welcome to the forum Wayne!

I always like to start with your loads--And ensure that your loads are as efficient/conservation minded as possible... It is almost always cheaper to conserve energy than to generate it.

Since I don't have your loads, and getting 9 years from your present FLA (flooded lead acid?) battery bank is pretty good--I will make some estimates how I would suggest using the battery bank and sizing the components and power expectations using are (relatively conservative) rules of thumbs (rule of thumbs, rules of thumb?).

Nominally, starting with your loads that then suggests the battery bank sizing--The battery bank is the "heart" of your system. With what we know of yours, I would base it on powering your loads for two days of "no sun" (bad weather), and 50% max planned discharge (for longer battery life, but you could go down to 30 to even 20% State of charge on occasion--As long as the bank is quickly recharged by sun/genset the next day).

24x @ 6 volt @ 554 AH FLA batteries(?)--8 batteries in series for 48 volts, and 3 parallel strings for a 1,662 AH @ 48 volt battery bank.

The suggested daily Watt*Hours usage (2 days, 50% max discharge) with 120/240 VAC inverter:

• 1,662 AH * 48 volts * 0.85 typical AC inverter eff * 1/2 days of storage * 0.50 max discharge = 16,952 WH per day

That is a pretty good sized system!

And charging--Typically suggest 5% minimum rate of charge (based on battery bank 20 Hour AH capacity)--Typically for Emergency backup and/or non-winter usage (lots of sun for charging). 10-13% is a good range to start for full time off grid:

• 1,662 AH * 58 volts charging * 1/0.77 panel and charger deratings * 0.05 rate of charge = 6,259 Watt array minimum
• 1,662 AH * 58 volts charging * 1/0.77 panel and charger deratings * 0.10 rate of charge = 12,519 Watt array nominal
• 1,662 AH * 58 volts charging * 1/0.77 panel and charger deratings * 0.13 rate of charge = 16,275 Watt array "typical cost effective" maximum

But we also need to calculate the array based on your daily load and amount of sun (and power usage by season--Dark winters, more energy usage? Or hot summers with A/C, etc.).

Guessing near Denver Colorado, using a fixed array facing south, tilted to 40 degrees. AC system eff with FLA batteries around 52% (losses 48%). Tilted to your lattitude for over all yearly best harvest. 1 kWatt as a simply array sizing point for data:
https://pvwatts.nrel.gov/pvwatts.php


Assuming you don't have shading issues (trees, mountains, etc.)--Start with winter December 4.57 hours of sun per day (something like 20+ year December averages). And the 1,6952 WH per day energy usage. The array needed for "December break even" harvest:

• 16,952 WH per day * 1/0.52 off grid AC system eff * 1/4.57 hours of sun per day December = 6,863 Watt array December "break even" array

Generally if this is a full time off grid home and you want to minimize genset usage, then you might want to only use 50% to 65% of your "predicted" daily harvest to account for a string of bad weather and/or some days with heavier loads (summer A/C, Irrigation pumping, etc.):

• 6,863 Watt array December "break even" array / 0.65 solar "fudge factor" = 10,558 Watt array December with "fudge factor"

Or, another way to look at what you have today... Your 4,590 Watt solar array in December would produce (on average):

• 4,590 Watt array * 0.52 AC FLA solar system eff * 4.57 hours of sun per day (Dec average) = 10,908 Watt*Hours per day December average

That you have a (very) large battery bank, and are really running a (relatively) small solar array--And your batteries have lasted this long--You are probably not cycling the battery bank that deeply, so that your present array can keep the bank relatively well charged. Also, if you are in a (on average) a cool climate, keeping your battery bank "colder" also helps (more or less, for every 10C/18F below 25C/77F room temperature, your batteries will age 1/2 as fast--I.e., last 2x longer).

In general, these days batteries are expensive and solar arrays are cheap--It is usually a good idea to "over panel" your system to keep the battery bank "happy" (typical cycling to 75% State of Charge, and avoid lots of deep cycling below 50% SoC). And you get the battery bank to near 100% SoC at least once per week. And EQ charging roughly once per month (or when battery electrolyte specific gravity between high and low cells exceeds around 0.015 to 0.030 differences--Then EQ to bring up weak/low cells).

There are still lots of details to discuss/understand--But this is probably a long enough post to get the discussion started about the details of your energy and system needs.

-Bill

Dave Angelini

Jim and Bill have great advice. If you try and save a 9 year bank you will spend alot of time and still have the problem of multiple strings of batteries.

I would say build a new bank with one string of about 1200 AH batteries or consider LFP.

WayneH

Location is South of Durango near the state line.  No shading.  During 8 months of the year battery is fully charged by 11am.  Part of my
confusion is that all battery SOC meters disagree.  Per the Magna remote never been below 92%, per the Midnite 150 display never been
below 80%.  Stand alone Midnite SOC meter some winter mornings as low as 50 to 60%  (I do not trust it).  No matter what all meters read
100% by end of day.  Generator use as necessary.  Currently 4:30am Magna reads 95%  Midnite 150 reads 85 to 87% and stand alone reads around 60.  Ambient temperature 10 degrees or so.  Two hours generator brings it to 100.  Drops until mid morning  and back to 100
by noon to 1:00.   Perhaps best would be to get some specific gravity readings to compare to meters.  I have one cell that quit consuming
water several years ago, another showed reduced use at winter solstice service this year (50%)  All other cells very consistent. Battery 
voltage on a cold morning (zero) at 4:30 as low as 48.2.  Usage:  1400 sq ft modular with extra insulation at build.  Two adults with conservative power use.  But TV has gotten larger and is on 14hrs a day, refrigerator is aging, small chest freezer added last year.  Primary heating with wood to minimize forced air furnace use.  I need to figure out which meter to trust, and possibly proceed with greater depth of discharge.  I do expect to replace the batteries before next winter.  Currently considering lead acid vs Lithium Iron.  Also confused as to size of Lithium bank comparable to what I have.  Also considering heating power shed for Lithium.  I think that either my charge controller
or inverter should be storing useful information regarding power use but I am unsure how to retrieve data.
I have been running by guess and by golly and faced with current battery costs must inject reality.  Thanks to all for the help.
Bill - I understand what you are saying, but am trying to digest it and fit in my numbers.

WayneH

Bill - I plugged my numbers and location into the PVWatts calculator and see solar radiation numbers matching the days production
amounts on my Midnite 150.  The biggest reason that I think that I have a problem is my am voltage readings.  Single digits temperature
and voltage of 48.2 10 hours of darkness and use from 60.4v ending generator charging at 25 degrees.  Maybe I will sleep in over the
weekend and see what the situation is at 9:00am.  I may have spent many years over reacting.

ELYNN4

Per Rolls…If your batteries are truly 25 degrees your ending charging voltage should be 63+ volts to compensate for cold temperature.  You should be looking for a charge return rate of 2%.  Even with these settings it’s hard to get your batteries fully charged - you have to burn a lot of fuel.

Good luck - hope you can squeeze more life out of them.

BB.

The "gold standard" for FLA batteries is the Temperature corrected Specific Gravity of each cell. For a couple of examples:

https://www.solar-electric.com/midnite-solar-battery-hydrometer.html

https://www.amazon.com/s?k=hydrometer+battery+acid

As always, rise the hydrometer with distilled water after testing to ensure the floats don't get sticky from dried electrolyte. Also, round glass hydrometers are easy to roll off a table and break. Make sure they will not roll as you place them down.

I don't know if you have it (or if your MPPT Charge Controller has the firmware/hardware--Age?)--Midnite has a nice "Whizbang Jr." current shunt accessory (attaches to a current shunt) to actually measure the current (Amps*Hours) into/out of your battery bank... That will help you both monitor the actual battery drain, and how much is being restored to your battery bank (at lower States of Charge, the Amp*Hour almost equals AH in for Lead Acid batteries--The "inefficiency" happens more or less at the last 90-100% SoC part of the cycle--I.e., gassing, or something like 1.25x AH recharged vs AH consumed).

Current Shunt (precision power resistor) sold separately.

https://www.solar-electric.com/midnite-solar-whiz-bang-jr-current-sense-module.html

As ELYNN4 says--The "end of charge" current for an FLA battery bank is something like 2% of 20 Hour capacity:

• 1,662 AH bank * 0.02 rate of charge = 33 Amps (Charging Current where FLA is near 100% full)

That you can recharge your bank in 2 hours with a genset... Typically (very rough numbers) below 80% SoC, you can charge the battery bank at "maximum" current (somewhere between 10% and 20% Current). As you go from 80% to 90% SoC, the battery bank will increase its terminal voltage and reduce charging current (Absorb Stage charging). And above 90%, the charging current should slowly drop to ~2% rate of charge.

Using a Hydrometer and a current shunt in the battery circuit, typically eliminates much of the guesswork for charging FLA batteries.

Pretty much, to get to 80% SoC, the AH to raise SoC is very linear (i.e., full 100 amps * 1 hours = 100 AH increase in battery SoC). As you get to 80-100% SoC, the charging current slows down and (again very roughly) your charger should "hold" the charging voltage (58-60 volts depending on battery brand, temperature, actually charging current, and such) for roughly 2-6 hours... 2 hours if the battery is shallow discharge (10-15%?). And deep discharge (down towards 50% SoC) would require holding Absorb voltage for ~6 Hours.

In the end, the "ideal" charging termination is that ~2% rate of charge (FLA batteries are near full).

The "Typical" generator charging is done from low SoC to 80% or so in the "morning"... And let the solar finish the charging the balance of the day (or next sunny day). This is generally the most efficient use of genset fuel/run time).

You do not have to charge to 100% every day--And many will argue that is "hard on FLA batteries". Instead, as long as you are cycling the battery daily--You only need to reach near 100% once per Week (one Rolls? engineer suggests even reach 100% once per month).

More or less, FLA batteries sulfate when they "sit" (storage) for days-weeks+ at 75% SoC or less. And they don't sulfate (quickly) if they are actively cycling.

One suggestion if you are shallow cycling your bank (only to 90% SoC loading, and back to 100% by end of day)--That you stop charging until and use the bank until you are below ~80% SoC (even if this takes a couple days), then re-enable charging. Rolls/Surette battery are designed for deep cycling and "using" the battery bank to 80% or less SoC at least once a month (or more often if trying to "recover" a poorly perforating bank--Followed by full charging + EQ charging can help recover capacity.

As always, follow the manual instructions:

https://solar-electric.com/lib/wind-sun/surrette-manual.pdf

I am not a battery expert--So others here can add their suggestions too. 

Other possibilities--Measure the voltage of each battery during (relatively) heavy loads and charging currents (Log readings). At some point, you can possibly weed out the "weak batteries" and put two strings back together and see how they work for you--Stretch out their lives. FLA batteries (the unused "good" batteries) would need to be recharged for ~24 hours every 30 days--Or they can begin to quickly sulfate (the joy of maintaining "spare" FLA batteries).

In general, most Lead Acid batteries are killed by under charging, over discharging, and letting set for months without charging. 

LiFePO4 batteries can be a good step up... No electrolyte to check. More efficient (nearly 98% efficient vs 80% or so for FLA). And they charge quickly (no "absorb" time needed).

Other hints for FLA batteries--Is you can have one array facing South East, and a second array facing South West--That gives you more hours in the day for solar harvest (FLA batteries really do better with "more time on charge"). LiFePO4 batteries do not need the "virtual array" and charge quickly to 100% capacity.

I suggest that you design your FLA bank to use 50%-100% SoC range (for longest cell life). With Lithium, in theory you can run from 10% to 100% range (90% of capacity)--Although you may get longer Lithium cycle life if you operate around 20% to 80% SoC--Or 60% of capacity.

The really nice thing with Lithium is they recharge quickly and efficiently (solar or genset)--So if you have some shading (trees/mountains/buildings/power lines, etc.) you can get by with a "smaller" AH Lithium bank vs the 2 days/50% discharge suggested for typical FLA installation.

I don't have any big system Lithium experience--So others here can help better than I.

Don't forget that it is recommended to use a BMS (battery management system) with Lithium, and you keep them above (roughly) 45F or so (see battery specs). Lithium do not like to be freezing (FLA/AGM batteries do better than Li Ion in cold). In hot weather, Lithium do much better than FLA (FLA self heat during charging--Li Ion almost no self heating).

-Bill

WayneH

I greatly appreciate all of the thoughts and information.  I believe that the worst of the cold weather is behind us.  (Famous last words)
I will continue my operations until Spring.  By March 21 servicing I will dust off my refractometer, purchase the Midnite hydrometer, probably purchase the best glass lab grade hydrometer, create some sort of record keeping whiteboard for the 72 cells.  I will access, reread, and cross reference manuals for the charge controller, inverter charger, and batteries.  I will install a whizbang junior as appropriate
and now that I have high speed internet I will update firmware as appropriate.  My concerns were generated during the shortest days and coldest temperatures of the year with simultaneous snow coverage of the panels.   Thanks again for all the help.

ELYNN4

I have been using a refractometer and have found that it works best if I store it with the batteries so that it is at the same temperature as the battery acid.  The readings become less accurate ate the temperature of the refractometer drifts away from the temperature of the acid.  The are supposed to be temperature compensating but the mass of the refractometer quickly overwhelms the mass of the acid and will give you a falls reading.  I’ve had good with the midnite hydrometer.

WayneH

Received Midnite hydrometer and had to try it.  Took a sample of 3rd battery down, center cell, across the bank when it meters said 100%, late this afternoon. 

12.80  12.75  12.55  12.50  12.50  12.60

Looks like a significant amount of equalize charging to start with.