AGM batteries voltage drop .

CsolarVI
CsolarVI Registered Users Posts: 20 ✭✭
I am fairly new to Off grid system I live in the US Virgin Islands and last September we got hit with two category five hurricane’s.I purchase off grid systems. For my home in October . Living off grid is new to me however batteries and systems similar to this are is not running large mechanical / electrical equipment and owning boats With large battery banks. And having to watch and maintain your batteries it comes natural to me. However I’ve been having issues with my battery banks not sure exactly what it is but basically in a nutshell. I am not getting the amount of amp hours I should be getting. Let me explain .
The system consist of
36 solar panels at 290 watts each, for 10,440 watts STC on my array.
The battery bank consists of 24 DC400-6 AGMs in three strings of eight, in parallel at the bank with big 4/0 cables
Backup generator is a 22kW Generac, charging through a dual inverter and auto-starting at 48.0 volts.

Solar Charging Specs:
Absorb Amps Solar: 142 Amps
Absorb Volts: 58.8 Volts
Absorb Done Time: 3.0 Hours
Absorb End Amps: 25 Amps
Float: 54.6 Volts
EQ Volts: 58.8 Volts
EQ Time: 0.0 Hours

Generator Charging Specs:
Absorb Amps Solar: 120 Amps
Absorb Volts: 58.0 Volts
Absorb Done Time: 3.6 Hours
Absorb End Amps: 25 Amps
Float: 54.0 Volts
EQ Volts: 58.8 Volts
EQ Time: 0.0 Hours
I’m also connected to the grid as a back up.
Now my state of charge does not match my voltage of my battery bank The BMK Records that I am at 70% state of charge however my voltage at open circuit and and at rest for six hours is 47.6 .that’s 30% SOC . I also have a web monitoring kit that record every 30 seconds voltage amperage temperature etc. this is how I know for a fact my voltages and state of charge over the months. The system was capable of running my entire house through the night the loads are exactly the same as it was and my battery bank would never get below 48.8 even with a big load “ water pump turning on”. now my system cannot make it to half of the night without my battery bank falling below 46 V and the BMK reading 70% SOC with a load of 36 amps. -250 ah used. My usage is around 480ah by 8am My usable amp hours is around 622
Bad battery right ?
On a sidenote From the time I got the batteries they have return to 100% SOC According to the BMK and voltage Practically every day.
What I have done so far
I have” fully charged “. The battery banks disconnected all batteries allowed them to sit open circuit for 24 hours Tested the voltage on all batteries they all maintain the voltage of 6.4
We have tried charging the batteries running multiple absorb cycles at 58.8 V as recommended by battery manufacturer we have tried charging the batteries running multiple absorb cycles at 58.8 V as recommended by battery manufacturer.
I have checked and all battery connections. inspected all wires all terminals. I have changed out that the BMK confirm the voltage and amperage recorded by it using a DC multimeter and amp probe This problem started when the batteries were only 5 months old. I’ve been dealing with this and trying to troubleshoot this for months with no answers.
Any suggestions would be greatly appreciated
«1

Comments

  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    A few thoughts...

    Are the charging voltages temperature compensated, with remote sensors attached to the bank from each charging device, and if so, is it possible the sensor(s) are giving false readings (eg sun hitting the sensor while most of the bank is shaded, or vice versa)?

    Have you checked individual battery voltages under significant load to see if any sag more than others?

    Do you have a DC clamp meter to check current to/from individual strings?
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    edited August 2018 #3
    I do have a nice fluke meter capable of reading DC amps all three strings Are giveing equal load +/- 1 amp
    Yes they are sensor is attached to each bank 4 in total. The two from midnight solar charge controller are in the middle of the banks where they recommend them to be installed on the side of the battery. and two for Magnum inverter/ charger on the negative posts none are exposed to a heat. Source of any kind.
    The compensator is set at 3.5 mV
    I have loaded each battery individually with a carbon pile load tester capable of doing 1000 Amp DC. loaded them to Half the rated cold cranking amps which is 750amps all test exactly the same. Don’t recall what the numbers were just that they were exactly the same .
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    edited August 2018 #4
    I assume you've looked closely at logs from around that 5 month mark for anything unusual (like you were away for a few days and guests ran down to a low state of charge, but didn't fess up).

    Does the solar usually end absorb with end amps, or does it normally time out at three hours? If end amps, about how long in absorb? Do you use a wbjr?
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    edited August 2018 #5
     Yes very closely. 
      The way the system is set up and programmed was  once I hit 60% state of charge it switches over to the grid and charges the battery banks back up to 96%  If the grid it is not there i.e. power outage and state of charge gets to 55% the generator will come on and charge batteries back to 96% so in theory we should never get below 55% if I am there or not.  And prior to this upset we never had to use the grid or gen in this way
     Just in testing now I have. The grid coming in at 75%  gen at 70%  

     We were using  time on the charge with the midnight’s.  3 hours to absorb 
      Then we went to ending amps, Now we are back to time to begin the float cycle all under the direction of the battery company 
       As for the BMK it’s a  ME-BMK from magnum  sign with a 500 amp shunt 
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    Using the 3hr timed absorb, how much current is going to the bank near the end of absorb? If ~ the 2%/25a end-amps, have you noticed about how long the absorb takes to drop current to about that?
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    Also, you've likely already checked voltage at each string vs what the classics think it is while charging in bulk? Do you have an offset value set on the classics?
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
     Yes   All the voltage are exactly the same on each bank  each bank is also parallel with a 04 wire And each bank has a 04 going to termination plate on the + and - 
     The offset is -0.2 on one midnight  And -0.3 on the other 
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
     three hours of absorb time I’m usually about  7-9 amps And by the end of the day I am floating at 2 A
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    Do you know about how long it takes after the transition from bulk to absorb to reach ~25a?
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Sadly, with the maintenance free AGM batteries, you cannot check the SG of the electrolyte.   If they used to go all night, and now they don't,  they have reduced capacity.

    With AGM, 2 things can reduce capacity in a year,  over-charging and off-gassing electrolyte, or chronic undercharge resulting in sulfation.

     With AGM, you cannot rely on voltage, you have to measure the recharge ending amps, to tell if they are completely charged, and you must synchronize all charging sources to be accurate with this. (Grid Charger, Magnum inverter/ charger, Midnight chargers, Generator)  If one senses the batteries full, but the other keeps on charging and gassing your electrolyte away ...

    When you get new batteries, I suggest you get Flooded batteries, and use 24,  L-16, 2V , 1100ah batteries in one string.   You can monitor them, check the SG, see if you are getting full charge with your settings.   Once you have the system up and running, you can record and compare how it's working compared to the old, damaged ones.   

    With parallel strings of AGM batteries, it's very hard to balance the current levels when charging and discharging, and the "best" string does most of the work, till it's damaged and then all the strings are goners.  So using 2V 1100ah cells is the ideal string to replace what you have, and you have a few less interconnects to mess with.

    For economy you should be running off the Grid, with batteries in Float, and using the inverters in Load Shave/Grid Support mode to reduce grid consumption.  Batteries should only come online for use if the grid fails.

    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
     Yes   around two hours 
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    Mike95490 Power in the US Virgin Islands is $.41 a kilowatt you don’t want to know what my power bill was . This is why I purchased a system like this the grid is not reliable and very expensive.   There are $12,000 worth of batteries  to just go out and buy new batteries without figuring out what happened to these.  Just doesn’t make sense these batteries were never abused I have records showing system,  voltage usage, state of charge .  I went by the settings the  battery manufacturer gave me  how can this be . 

  • Marc Kurth
    Marc Kurth Solar Expert Posts: 1,142 ✭✭✭✭
    What are the date code ranges on these batteries?
    Have you checked for hot spots?

    I always have more questions than answers. That's the nature of life.
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    12 are K173108GS  
    8 are K173209GS 
    As far as hotspots go t as far as hotspots go I have a FLIR thermal imaging camera that I’ve taking photographs from day one at least one every couple of months there are nothing  unusual.  Never recorded anything above 91F on the image. 
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    You mentioned this started at about the 5mo mark, so around March of this year?

    Did things deriorate in a fairly short period of time (days), or more steadily (over weeks/months)? Is apparent capacity still deteriorating, or have things stabilized since March?
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    Yes  the middle of March   We believe that happened over one or two days . Seems to have stabilized since. 
  • SolarMusher
    SolarMusher Solar Expert Posts: 176 ✭✭✭
    CsolarVI said:
     Yes very closely. 
      The way the system is set up and programmed was  once I hit 60% state of charge it switches over to the grid and charges the battery banks back up to 96%  If the grid it is not there i.e. power outage and state of charge gets to 55% the generator will come on and charge batteries back to 96% so in theory we should never get below 55% if I am there or not. 

    I could be wrong, but if you're talking about a Magnum PAE system, it doesn't work that way. It can be set to gen OR grid connect but not both. If the grid is down and the gen is started from AGS, it will use batteries only and won't pass any current from the gen until the Low Battery Disconnect setting is met (waiting for the grid).
    Who told you that?
    From memories,
    Erik
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    edited August 2018 #19
    Do a load test on each individual, fully charged  battery at a C20 rate for 5 hours.  Record idle/rested voltages before and after.  Re-assemble according to capacity (most similar batteries in a series string).

    > reading DC amps all three strings,   Are giving equal load +/- 1 amp 
    Not surprising, parallel balance problems are over-blown.

    Also check voltage on each individual battery (checking for series balance issues) while it's in the string.  There are ways to fix this if  they vary.

    I am available for custom hardware/firmware development

  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    IMHO (others here with more experience with AGMs may be able to help more), the most likely problem is an overcharge relative to battery temperature. Even in the March V.I. heat, hardened sulfation from undercharging would take a while. Venting gas would be a sudden event.

    The root cause of the overcharge could be a bad temp sensor. Although it may have been a transient problem (eg loose connection), it might be worth checking the consistency of currently reported temp values on charging devices vs actual battery temp.

    You mentioned "the two sensors" from classics. Mine have a single sensor, and are wired and configured for "follow me", in which when one controller goes to absorb or float, the other follows. There's a single temp sensor, which is read and used for compensation by both. It sounds like each of yours has a separate sensor, and operates independently. As I suspect your voltages would always be compensated down in your climate, a bad or misconfigured sensor on one of the classics could have been the problem.

    If you have classic log data from March, it might be interesting to compare logged high battery voltages vs estimated target voltage adjusted for ambient high temps for the same period.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    As I said, there's only 2 ways AGM loose capacity and won't run through the night like they used to.
    Electrolyte loss.
    Sulfation from chronic undercharge.

    Unless you have a setting that changed somewhere, you are looking at a new battery bank to replace a failed bank.  Sorry to break the bad news to you.

    Using batteries to replace grid power - even at $0.41 kwh, is going to cost you more than simply conserving energy and saving power, batteries cost more than $0.41 kwh

    With AGM, it's really hard to judge their state of charge without counting amps in vs out.

    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    Yes my system does work this way once the system hits 60% state of charge it allows grid power in. Or Pass through  If there is no grid power and the batteries get to 55% state of charge the AGS Starts my.  22K generator  then my automatic transfer switch switches over to generator and power comes and runs the house and charges the batteries .  It is that simple. 
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    Mike95490.  Due to the hurricanes are grid is extremely unreliable most of the time will go days without power from the grid as for the AGM  they were not my first choice I actually did not want them but they were part of a kit from wholesale solar Where I got my system and I tried to get crown flooded  But obviously couldn’t  
     On a sidenote my last power bill without the solar system was $1400 for one month you do the math. It’s totally worth it.
  • Marc Kurth
    Marc Kurth Solar Expert Posts: 1,142 ✭✭✭✭
    Something just isn't making sense,

    I am very familiar with those batteries - I currently have around 600 of them on projects, several going back 5-6 years. Many of those sites are industrial and very rough duty by nature - and I have seen (2) batteries develop a problem in that time span. Generally they are 1-2 strings of 8 batteries in off grid 48v systems.

    Have you taken this to Fullriver's engineering guys?

    Marc
    I always have more questions than answers. That's the nature of life.
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    Marc kurth   I have spoken with fullriver several times they have sent me one replacement battery. Still nothing changed when I installed the new battery  battery replaced is still holding  a charge of 6.3v after weeks of Open circuit 
    They have given me series of test to run one including a device that I am waiting to get my hands on that will load test an entire bank at 75 amps. And I am to time how long it takes to get to 42v.  This device is very expensive  to buy  I cannot find one to rent  as of yet . 
     
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    CsolarVI said:
    Marc kurth   I have spoken with fullriver several times they have sent me one replacement battery. Still nothing changed when I installed the new battery  battery replaced is still holding  a charge of 6.3v after weeks of Open circuit 
    They have given me series of test to run one including a device that I am waiting to get my hands on that will load test an entire bank at 75 amps. And I am to time how long it takes to get to 42v.  This device is very expensive  to buy  I cannot find one to rent  as of yet . 
     
    You already have a great load test device, your inverter and microwave oven, along with a blow dryer. 
     Shut the generator down, switch things on  and start measuring.   Does your inverter readout what the loads on it are ?
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭
    I have done this On several different occasions. comparing voltage on each bank and each battery under lode and at Various states of charge Comparing. The inverter is reading to the amperage reading it all checks out.  However there manufacturers want a more official testing method . 
      Found once the load is applied and I have drawn a proximally 200Amps From the banks The voltage starts the drop dramatically at this point it’s not A linear curve more like a steep drop in voltage. 
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭

    CsolarVI said:
    I have done this On several different occasions. comparing voltage on each bank and each battery under lode and at Various states of charge Comparing. The inverter is reading to the amperage reading it all checks out.  However there manufacturers want a more official testing method . 
      Found once the load is applied and I have drawn a proximally 200Amps From the banks The voltage starts the drop dramatically at this point it’s not A linear curve more like a steep drop in voltage. 
    Lost capacity would cause a sharp drop in voltage as described, during charging  the voltage will most likely climb rapidly with a short absorption period before transition to float, text book indicators.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    Steep voltage drop usually occurs around 80-90% DOD.  Once you are doing that, they aren't going to last much longer.

    Besides improper charging, high temperature is a battery killer.

    I am available for custom hardware/firmware development

  • Marc Kurth
    Marc Kurth Solar Expert Posts: 1,142 ✭✭✭✭
    CsolarVI said:
    Marc kurth   I have spoken with fullriver several times they have sent me one replacement battery. Still nothing changed when I installed the new battery  battery replaced is still holding  a charge of 6.3v after weeks of Open circuit 
    They have given me series of test to run one including a device that I am waiting to get my hands on that will load test an entire bank at 75 amps. And I am to time how long it takes to get to 42v.  This device is very expensive  to buy  I cannot find one to rent  as of yet . 
     

    Who are you working with at Fullriver?
    Marc
    I always have more questions than answers. That's the nature of life.
  • CsolarVI
    CsolarVI Registered Users Posts: 20 ✭✭

    I have work with Matthew Karlovic Technical Support Engineer and Aaron Plew