Battery bank problems

wolfbay
wolfbay Registered Users Posts: 5 ✭✭
Our batteries rapidly loose storage capacity. We have a Xantrex xw6048 inverter and 2 charge controllers and a 48volt system. A new set of outback AGM batteries were installed along with a 15kw generator. The batteries rapidly weakened and the installer said we had a defective battery but outback disagreed and wouldn’t honor the warranty. We were able to enroll in a farm program that paid for most of a new set of batteries. This new installer said the problem was the six string parallel set up with four 12 v batteries in series. His four lead acid batteries (900AH) were 12 volt in series. Now these are 6 months old and they are rapidly weakening. On a sunny day with minimal load or with the generator we can only get a maximum SOC of 1.21. We wonder if there is some problem withthe generator or the inverter that is ruining the batteries.can anyone suggest how to diagnose the problem?
Tagged:

Comments

  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    How low has the bank been drawn down, typically, and ever over the life of the bank?

    Do you know your average daily consumption in kilowatt-hours?

    How much solar is hooked up to the two charge controllers?  Are there any shading issues?

    What absorb voltage and timeout are the controllers and chargers set for?

    A 900ah@12v battery would be huge (~1000lbs), and need heavy equipment to move.  Could they be 900ah total, so 900/4= 225ah each ... 225ah@48v wired in series?
    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
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    @wolfbay asks
    On a sunny day with minimal load or with the generator we can only get a maximum SOC of 1.21. We wonder if there is some problem with the generator or the inverter that is ruining the batteries.can anyone suggest how to diagnose the problem?

    Can you define the SOC of 1.21 statement ? Based on description the batteries are either  225Ah, 12V, configurered in a 4 series  4 parrallel  arrangement, not ideal, or is it 150Ah 4 series  6 parallel. What  charging is involved, assumption would be solar with generator backup, can you describe the system in more detail, there's  not much to work on, for example array capacity, charge controller type, make and settings, inverter charger settings and so forth. 


    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.
  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    Also size and length of wires from controllers and charger to battery bank.  It would help if you could measure battery voltage at the bank end while charging, and compare that reading with what controller/chargers think bank voltage is as well.
    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
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    edited December 2018 #5
    You can have an AH test done to diagnose current battery capacity/health.   You can measure voltage of each battery at rest (after 6+ hours with no charging or load) to verify that each battery is reaching full charge (not reaching full charge will rapidly and permanently weaken batteries).   Ideally you compare the resting voltage to what you measured when the batteries were new and known to be fully charged - not to some value from the Internet.    

    You should also measure the amps to each parallel string (parallel balance, done with heavy load).


    I am available for custom hardware/firmware development

  • wolfbay
    wolfbay Registered Users Posts: 5 ✭✭
    Estragon said:
    How low has the bank been drawn down, typically, and ever over the life of the bank?

    Do you know your average daily consumption in kilowatt-hours?

    How much solar is hooked up to the two charge controllers?  Are there any shading issues?

    What absorb voltage and timeout are the controllers and chargers set for?

    A 900ah@12v battery would be huge (~1000lbs), and need heavy equipment to move.  Could they be 900ah total, so 900/4= 225ah each ... 225ah@48v wired in series?

    usually the batteries are not below 48.5 v in the morning and once they went down to 47.5 and the auto start on gen kicked in. I was told to get a refractometer to measure SOC and in morning usually about 1.19 and when on float in late afternoon about 1.21. They are very big heavy batteries . Four in series and says on each 990AH.
    system is 5.6 kw with no shading issues. We also have four 300 WAtt panels with mini inverters connected to AC side. This time of year we use little power and on a sunny day the batteries are on float by about 11 or 11:30 am. 
    Even when we charge batteries with gen they don’t get near fully charged(1.20-1.21) Not to muddy the waters but I also noticed when gen runs panel indicates 63.5hz. Electric clocks run fast when gen running . Because the gen output is more than 60 could this have damaged the batteries?
     Thanks to all for responding to my post. 


  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    AGM batteries are generally sealed and do not have free electrolyte (it is in a fiberglass mesh).

    If you have a refractometer, you can get a drop of  electrolyte and measure the specific gravity--But that is not usually done.

    Is this how you are measuring S.G.? And to get the "health" (State of Charge) of a battery bank, you need to check each cell's SG... And see how they compare with other cells.

    Did Outback say anything about the 1.21 SG of the battery(cell)? Seems pretty low (usually the SG is closer to 1.265 for "typical" solar storage batteries). 1.21 is not bad in itself (lower SG fill gives longer battery life, but less capacity).

    Have you measured the battery voltage of each battery when charging? Are they call close to the same or is one battery much higher than the rest?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    The genset running at 63.5 Hz should not hurt anything. As long as the batteries are getting proper charging current (and reaching setpoints and times), all is fine.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    When using a generator to charge, you have to be careful about enough hours (need to get through a full absorb once per 3-7 days if batteries go below 75% SOC).

    Temperature compensation is a good idea (doesn't look like you have it).

    You might be able to look up what the fill SG is.   Otherwise, the best you can do is give it a full charge and equalization and then call the SG (and resting voltage) at that point "fully charged".

    I am available for custom hardware/firmware development

  • wolfbay
    wolfbay Registered Users Posts: 5 ✭✭
    edited December 2018 #10
    BB. said:
    AGM batteries are generally sealed and do not have free electrolyte (it is in a fiberglass mesh).

    If you have a refractometer, you can get a drop of  electrolyte and measure the specific gravity--But that is not usually done.

    Is this how you are measuring S.G.? And to get the "health" (State of Charge) of a battery bank, you need to check each cell's SG... And see how they compare with other cells.

    Did Outback say anything about the 1.21 SG of the battery(cell)? Seems pretty low (usually the SG is closer to 1.265 for "typical" solar storage batteries). 1.21 is not bad in itself (lower SG fill gives longer battery life, but less capacity).

    Have you measured the battery voltage of each battery when charging? Are they call close to the same or is one battery much higher than the rest?

    -Bill

    The first set were AGMs 6 strings parallel with 4 in series in each string.
    The installer of the new batteries was confident this was the problem .
    The new batteries(6 months old) are large conventional lead acid 990ah each with 4 in series.
    I’ve taken SP in all the cells and they vary very little. I can’t get the batteries anywhere near fully charged either with the generator or on sunny days with little load and I’m afraid somehow they have lost much of their storage capacity. 
    Thanks for your reply Bill
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    What is your charging voltage setpoint for the flooded cell battery bank? 

    You should be at 59 volts (or possibly higher--Some industrial FLA batteries do need higher voltages). (all at ~75F/25C nominal bank temperature).

    And have you tried to equalize the batteries yet? Yours do not need to be equalized (basically, if the low to high SG is less than ~0.015 to 0.030 units, the cells are "equal"), but 60 to 64 volts held for 1-many hours (if needed) and see if you can get above 1.210 SG...

    Basically check the charging current (something like 2.5 to 5.0% of bank AH capacity at ~60-64 volts--Batteries will "fizz" if EQing--You do not want a "rolling boil"), and monitor bank temperature--If you EQ for many hours straight, even 2.5% EQ current is enough to overheat the battery bank--And you will need to stop EQ and start again the next day to continue EQ.... Once the SG stops increasing (check SG 30-60 minutes), then you are done.

    Once your bank's SG stops rising--That is your new 100% full EQ. We have some posters here who have taken several days of EQ to get the numbers back to "normal". And sometimes, it takes weeks or months of regular EQ (like once per month, not every day EQ) and proper charging to get the battery bank working the way it should.

    Voltages are starting points... You can adjust a bit up or down depending on your needs (perhaps higher in winter with cooler temperatures and less sun, lower in summer in warmer weather and more hours of sun per day).

    More or less, for a full time off grid FLA battery bank, you want to draw a (rough) maximum of 25% of capacity (overnight, on a cloudy day)... Generally, the solar panels under typical conditions, can only bring an FLA bank from 75% to >90% in a single day of sun.

    You should try for >90% SoC once or twice a week nominally. Don't try for 100% every day. That is basically an EQ cycle and this is pretty hard on a FLA battery bank (plate erosion, positive grid corrosion, excess heat, etc. from gassing).

    I am not a battery engineer or expert--Just summarizing what I have read here and elsewhere.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • WaterWheel
    WaterWheel Registered Users Posts: 375 ✭✭✭
    edited December 2018 #12
    On your charge controller what battery bank size do you have programmed in?      Your Schneider charge controllers will end absorb when the amperage drops to a 2% charge rate of the battery bank size you've programmed in the controllers.      If your charge controller thinks you have a larger battery than you actually have the controller will end charging early.      This also applies to the battery size your 6048 is set at for your generator.

    A related thought.

    According to the guys at Rolls Battery they want the battery to be held in absorb at about 2% of battery capacity for an hour before the charger ends absorb.       Some charge controllers allow you to program the controller to hold the battery in that 2% level for a set time before ending absorb but not Schneider controllers.     Schneider controllers go into float as soon as you reach that 2% absorb rate and that 2% level is not programmable. 

    So you have to trick the controller to hold the battery in float for a while after it hits the 2% absorb amperage.      To do this you change the battery bank size to about 65% of your actually battery bank size.       This makes the controller continues to charge the battery for a while longer instead.

    example using your 990 amp hour 48v battery and the controller ending absorb when it thinks your battery bank is only needing a 2% charge rate.

    Let's program your controller to a 990 amp size battery bank.      The controller will end absorb then the amperage required to maintain the set absorb voltage drops to 19.8 amps.    A little early for the guys at Rolls, they want the batteries to stay in absorb for an additional hour.

    So if instead you changed your controllers battery bank size to about 650 amps (about 65% of your battery banks actual size) the controller will hold the batteries in absorb until the absorb amps drop to about 13 amps or roughly another hour.

    But if instead you've got your controller programmed with a larger battery bank than you actually have, say a 1300 amp hr battery instead of the 990 amp/hr battery you actually have your controller will stop absorb when the absorb amperage drops to 26 amps.      This will chronically undercharge your batteries. 

    And as previously stated you really need a battery tempature monitor.       Without it you will be undercharging your batteries in cooler weather.

    Conext XW6848 with PDP, SCP, 80/600 controller, 60/150 controller and Conext battery monitor

    21 SW280 panels on Schletter ground mount

    48v Rolls 6CS 27P

  • wolfbay
    wolfbay Registered Users Posts: 5 ✭✭
    BB. said:
    What is your charging voltage setpoint for the flooded cell battery bank? 

    You should be at 59 volts (or possibly higher--Some industrial FLA batteries do need higher voltages). (all at ~75F/25C nominal bank temperature).

    And have you tried to equalize the batteries yet? Yours do not need to be equalized (basically, if the low to high SG is less than ~0.015 to 0.030 units, the cells are "equal"), but 60 to 64 volts held for 1-many hours (if needed) and see if you can get above 1.210 SG...

    Basically check the charging current (something like 2.5 to 5.0% of bank AH capacity at ~60-64 volts--Batteries will "fizz" if EQing--You do not want a "rolling boil"), and monitor bank temperature--If you EQ for many hours straight, even 2.5% EQ current is enough to overheat the battery bank--And you will need to stop EQ and start again the next day to continue EQ.... Once the SG stops increasing (check SG 30-60 minutes), then you are done.

    Once your bank's SG stops rising--That is your new 100% full EQ. We have some posters here who have taken several days of EQ to get the numbers back to "normal". And sometimes, it takes weeks or months of regular EQ (like once per month, not every day EQ) and proper charging to get the battery bank working the way it should.

    Voltages are starting points... You can adjust a bit up or down depending on your needs (perhaps higher in winter with cooler temperatures and less sun, lower in summer in warmer weather and more hours of sun per day).

    More or less, for a full time off grid FLA battery bank, you want to draw a (rough) maximum of 25% of capacity (overnight, on a cloudy day)... Generally, the solar panels under typical conditions, can only bring an FLA bank from 75% to >90% in a single day of sun.

    You should try for >90% SoC once or twice a week nominally. Don't try for 100% every day. That is basically an EQ cycle and this is pretty hard on a FLA battery bank (plate erosion, positive grid corrosion, excess heat, etc. from gassing).

    I am not a battery engineer or expert--Just summarizing what I have read here and elsewhere.

    -Bill

    Our bulk charging set point is 57.2 . Absorption56.5. Float 56.1 on each charge controller.
    Is the charging voltage set point you refer to the bulk charge set point and raise to 59?
     Thanks bill

  • wolfbay
    wolfbay Registered Users Posts: 5 ✭✭
    On your charge controller what battery bank size do you have programmed in?      Your Schneider charge controllers will end absorb when the amperage drops to a 2% charge rate of the battery bank size you've programmed in the controllers.      If your charge controller thinks you have a larger battery than you actually have the controller will end charging early.      This also applies to the battery size your 6048 is set at for your generator.

    A related thought.

    According to the guys at Rolls Battery they want the battery to be held in absorb at about 2% of battery capacity for an hour before the charger ends absorb.       Some charge controllers allow you to program the controller to hold the battery in that 2% level for a set time before ending absorb but not Schneider controllers.     Schneider controllers go into float as soon as you reach that 2% absorb rate and that 2% level is not programmable. 

    So you have to trick the controller to hold the battery in float for a while after it hits the 2% absorb amperage.      To do this you change the battery bank size to about 65% of your actually battery bank size.       This makes the controller continues to charge the battery for a while longer instead.

    example using your 990 amp hour 48v battery and the controller ending absorb when it thinks your battery bank is only needing a 2% charge rate.

    Let's program your controller to a 990 amp size battery bank.      The controller will end absorb then the amperage required to maintain the set absorb voltage drops to 19.8 amps.    A little early for the guys at Rolls, they want the batteries to stay in absorb for an additional hour.

    So if instead you changed your controllers battery bank size to about 650 amps (about 65% of your battery banks actual size) the controller will hold the batteries in absorb until the absorb amps drop to about 13 amps or roughly another hour.

    But if instead you've got your controller programmed with a larger battery bank than you actually have, say a 1300 amp hr battery instead of the 990 amp/hr battery you actually have your controller will stop absorb when the absorb amperage drops to 26 amps.      This will chronically undercharge your batteries. 

    And as previously stated you really need a battery tempature monitor.       Without it you will be undercharging your batteries in cooler weather.
    I was told to try setting battery bank size as zero and we still can’t get SOC above 1.21. 
    We have a battery temp monitor on one charge controller but need another one.
    Thanks for your reply.
  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    edited December 2018 #15
    Your voltages are too low for a flooded lead-acid bank, they should be ~59V bulk, ~57.6 absorb,  ~54.2 float, not withstanding manufacturer recommendations. The absorption timer should be extended to maximum, for now, to allow end amps to take control, but  program the actual capacity to provide a safety net, the recovery will take a long time without any loads, daily cycling will compromise the situation dramatically, difficult understandably, but nesesary. Once the bank cannot accept any more regular charging then it's equalization, another long process without loads, should it be required.
    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 ✭✭✭✭
    If the battery had a 1.215 initial fill SG (as is sometimes done), then you are doing nothing but damaging the battery trying to get more than that.

    I am available for custom hardware/firmware development

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    The EQ charging (battery at ~100% SoC, then elevated EQ charging voltage), monitoring SG of all cells every ~1 hour (and making sure battery does not overheat from hours of EQ), when the cells stop rising, that is your "new" 100% SoC specific gravity.

    Although, if you do this for once a month) and cycle the battery (generally below 75% SoC at least a few times a month), it is possible that you may see an increase in S.G. over time (industrial type batteries, like big Rolls/Surrette) seem to follow this path fairly often for batteries that have had "issues".

    And yes, for FLA deep cycle batteries, you need ~59.0 volts setpoint for charging. Your AGM value is too low. And extending Absorb time (6 hours or more--i.e., all the sun you can harvest during winter) will help too (as McGivor said).

    It is possible that your batteries were filled to 1.210 SG... Can you ask your battery supplier what their initial fill was?

    Monitor your distilled water usage... For "typical" deep cycle batteries, you should be refilling the cells about every 1-3 months--Indicates good charging practices. Less than one month, dial charging voltage/time back a bit... Over 6 months between fills, crank voltage and time up.

    When checking fill level, just make sure plates are covered (exposed plates will kill battery pretty quickly). Once battery is charged (and usually warm) then fill 3/4 full or so... Nothing like filling when cold, and the bubbling+heat spitting out electrolyte all over the top of the battery case.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    edited December 2018 #18
    Note that if you do have a non-standard SG fill, then all your voltages need to be adjusted downward - significantly. 

    I am available for custom hardware/firmware development

  • Cosuper_Energy
    Cosuper_Energy Registered Users Posts: 6 ✭✭
    wolfbay said:
    Estragon said:
    How low has the bank been drawn down, typically, and ever over the life of the bank?

    Do you know your average daily consumption in kilowatt-hours?

    How much solar is hooked up to the two charge controllers?  Are there any shading issues?

    What absorb voltage and timeout are the controllers and chargers set for?

    A 900ah@12v battery would be huge (~1000lbs), and need heavy equipment to move.  Could they be 900ah total, so 900/4= 225ah each ... 225ah@48v wired in series?

    usually the batteries are not below 48.5 v in the morning and once they went down to 47.5 and the auto start on gen kicked in. I was told to get a refractometer to measure SOC and in morning usually about 1.19 and when on float in late afternoon about 1.21. They are very big heavy batteries . Four in series and says on each 990AH.
    system is 5.6 kw with no shading issues. We also have four 300 WAtt panels with mini inverters connected to AC side. This time of year we use little power and on a sunny day the batteries are on float by about 11 or 11:30 am. 
    Even when we charge batteries with gen they don’t get near fully charged(1.20-1.21) Not to muddy the waters but I also noticed when gen runs panel indicates 63.5hz. Electric clocks run fast when gen running . Because the gen output is more than 60 could this have damaged the batteries?
     Thanks to all for responding to my post. 


    wow, this controller APP is very nice, what the name or what it system, could you introduce this system and APP?
  • tabbycat
    tabbycat Solar Expert Posts: 55 ✭✭✭
    Usually the bulk charging current of FLA's is 10% of the rated bank capacity. For a 900AH bank that would be 90 amps. Is your charging system providing this?