Trouble shooting 9+ year old system...

MadJack
MadJack Solar Expert Posts: 47 ✭✭
1. I'm in Maine, Off-Grid
2. Purchased designed 24v system from NAWS 2012, self installed, ground mount. (Outback components inside house)
OB 2435 Inverter, Mate, FlexMax 80-amp charge controller, and the 2" dial battery monitor.
3. 9 years acceptable use on 8 - 6 volt L16RE-b batteries
4. 2020 we purchased a Generac-off-grid/standby 6000 watt generator, connected via

Issue last fall with low battery cut-off/system shut down going from 50% state of charge to 60, then 70, etc...
[that's when we'd typically start the generator]
Initial thought was battery end of life situation, after re-reading manual and attempting to trouble shoot the system.

January I purchased 4-12 volt AGM 4D batteries, used, date of manufacture December 20 to July 21.(standby use only)

Once I had them installed, we ran through the manuals to adjust for the battery change to AGM. Over the past two months the system is shutting down with higher state of charge (meter) while batteries are holding 12.6 to 13.3 volts.

Question: Could the issue be the inverter, Mate, or charge controller going bad?

I welcome this forums thoughts and ideas...
Thanks in advance!

(I have been a member here for years, I just haven't posted much)
Tagged:

Comments

  • MadJack
    MadJack Solar Expert Posts: 47 ✭✭
    Addendum: Our daily Amp-Hour intake (according to the charge controller readings) have gone from 3.6 - 8.6 kwh per day, down to no more than 3 kwh.
    And the charge controllers Bulk Charge time seems much less... (seems to almost always be in Absorb or Float) and we can't find [in the manual] where to change settings for this.
  • wellbuilt
    wellbuilt Solar Expert Posts: 763 ✭✭✭✭
     My system is 5 years old now . 
      I had a similar problem , I did not do any firmware upgrade and my controller  was  going thru bulk charge very fast 
     then it would hit absorb and just go to float in a hour the absorb time was set for 3 hours . 
       I would go back to factory settings , and reprogram every thing .
        If that did not fix the problem  go on line and install all the new firm wear . I’m sure you are missing up dates . 
          Every thing started working as it should after that . 
     Now my only problem I have is after the inverter update I lost some charging power  I went from 1900watts  to 1700 watts for no reason ? 
      AWS will help you  , Logan is really good .
    Out back  flex power one  with out back 3648 inverter fm80 charge controler  flex net  mate 16 gc215 battery’s 4425 Watts solar .
  • MadJack
    MadJack Solar Expert Posts: 47 ✭✭
    Thanks WellBuilt.
    I'll have to find a cable took hook it to the Web, I guess. And I'll give AWS a call.
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    Lots of things...
    AGM batteries can come in a standby type as well as a daily cycling, since they were used, and kept at float, they may be design for stand by use...

    Did you get the charging voltage from the manufacturer? Sounds a bit low to me, Soome will have 2 different charging voltages, 1 for batteries that are cycled daily and one for batteries that will be in float/ held in reserve.

    Worth 'load testing the AGM batteries, and see if you have a bad battery. It's much easier to tell with flooded batteries but reading the SG. AGM tend to have a higher resting voltage. As always VOLTAGE IS AN AWFUL WAY TO CHECK SOC!!!
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • MadJack
    MadJack Solar Expert Posts: 47 ✭✭
    edited March 2022 #6
    My issue is definitely NOT battery related.
    Several emails with RAVI at Outback, it's looking like the Inverter/charger refuses to charge batteries from generator AC-in. (Eco-Gen 6000w Generac)
    Settings checked, (so far), Adv/FX/Inverter reset to 'defaults' but unknown if it even took...
    But... I am still only collecting HALF our normal kilowatt-hours for the first week of March. My records show as much as 9.6 kwH readings in early March with snow-cover on the ground. Even so-so days (of sun) we usually draw 4.5 kwH.
    Recently we barely reach the 3kwH for a full day of bright sunshine.
    I'm going to clean, grease, re tighten all connections today. I am still super happy my panels are ground-mounted!!!
    I'd gladly Venmo someone a 'thank you' if someone knows these components and wouldn't mind helping to diagnose...
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    edited March 2022 #7
    Have you added an "new" loads to your DC bus?

    In times past, we had one member that had "battery desulfator" on his battery bank. And the "electrical noise" from the desulfator was enough to cause one Outback charge controller to dramatically reduce its harvest.

    Turned off the desulfator, and the Outback MPPT controller started harvesting correctly again (he had a second Outback, either on a second system, or the same system--Don't remember, but one controller was definitely impacted by the desulfator).

    I always recommend a "Star" wiring connection for charge controllers (wire run from battery bus directly to charge controller). Daisy chaining (battery bus to controller A to controller B to controller C ) charge controllers can cause "cross talk" between controllers and confuse them (erratic battery bus voltage readings confusing the charge controller processor?).

    Also, how is your solar array configured (A series panels by B parallel panels)? If you have parallel strings and you have per string circuit breakers, you can turn off one string at a time and see how well they are well they are sharing current.

    Or you can get an AC-DC current clamp DMM and measure each solar panel string. DC capable current clamp meters are very easy, quck, and safe to use when debugging battery systems (also works for AC power too):

    https://www.amazon.com/gp/product/B00O1Q2HOQ (inexpensive--Good enough for our needs)
    https://www.amazon.com/gp/product/B019CY4FB4 (mid priced meter)

    The above are just examples of DC Capable Current Clamp DMS... Check around and see what works best for you. There are AC only clamp meters--They are good units too... But AC only clamp meters do not work for DC power systems.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • MadJack
    MadJack Solar Expert Posts: 47 ✭✭
    This is the current readings on my Outback FLEXMAX 80 charge controller after 8 hours in full sun. 
    A typical sunny March day would read 8.6 to 9.1 kwH.
    No faults, no warnings, that I can find. (but with over 400 screens on the Mate it's hard to know)
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    My issue is trying to understand:
    • Solar Panel / Array configuration (series/parallel/etc.) to see if one string is not carrying its share of current.
    • Battery bus voltage (currently 25.5 volts)
    • Charge controller "state" (yours says float)
    • What is the system DC loads/current at this point (are DC loads greater than Solar Array output power?)
    • What is the battery current (charging/discharging/etc.)...
    Generally, I would expect "float" to be around around 27.2 Volts. And battery resting voltage (full charge, no loads, not charging) around 26.5 Volts.

    Charging should be around 28.8 volts (in Absorb) and held for 2-6 hours (longer time for deeper discharge).
    • Solar Array: 83 volts * 4.5 Amps = 374 Watts
    • Controller output: 25.5 volts * 13.4 Amps = 342 Watts
    • Controller eff: 342 W/ 374 Watts = 0.91 = 91% efficiency (makes sense for lower power levels)
    • Mode: Float
    It looks like the controller "thinks" your battery bank was charged to 100%... And In Float mode. Actual battery bus voltage (if float voltage was set correctly) suggest solar panel power in to charge controller is equal to power out to DC loads (and battery current is near zero Amps).

    For off grid power systems--The total daily harvest (kWH) is based both on your daily loads and the amount of sun/solar available.

    If you are using less power (and have "enough sun"), then it looks like you are using less power right now?

    If the batteries are "under charging" when you have good sun--The possibly one solar array string is not supplying its share of current?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • MadJack
    MadJack Solar Expert Posts: 47 ✭✭
    edited March 2022 #10
    BB, what do you consider a "DC Bus"?
    Nothing has been added or removed from the system since 2012.

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    The DC bus is you Battery Bank 24 VDC +/- Terminals (usually the central point where all DC loads connect if not the battery terminals themselves).

    The DC Bus should be about the same voltage as the battery terminals (within a few 0.1 volts).

    Each load (AC inverter), sources (solar charger) add and subtract their current(s) to the DC Bus. The battery bank itself is the "buffer" or giant capacitor) that supplies current when need (bus below ~25.6 volts) and bank discharging. Or absorbing current when over ~27.2 volts (float, bulk, or absorb charging).

    What is the maximum battery bank voltage during the day (absorb setpoint of ~28.8 volts for AGM) and float around 27.2 Volts (aka solar array has more Watts available then the DC loads and battery bank charging are taking)?

    How low is the battery bank voltage falling to in the morning (before sunrise)? What is the load current.

    Do you know if the battery bank is not getting fully charged? How do you know?

    I don't know your loads--kWH per day. If you are charging at 3.x kWH per day and discharging at 9.x kWH per day--That is going to quickly discharge your battery bank over a few days ("deficit charging")?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • MadJack
    MadJack Solar Expert Posts: 47 ✭✭
    PV on the ground mount rack is 3 strings of 3 panels each.
    Multi meter test on each string (full direct sun) is typically 98 volts to 105 volts,  currently at 83 volts in late day sun.

    All strings are equal output.

    Multi meter test (Volts) at inside panel shut-off is same as above.

    Your post above...from the word "Generally" is spot on.
    My issue is that my Outback VFX 3524 Inverter is shutting down power to the house within two hours of no sun, with a status "error" (red light in inverter).

    ***even with minimal house loads***
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    edited March 2022 #13
    Unfortunately, you cannot really measure per string "current" with just a volt meter (I have done it, but it is much easier with a DC current clamp meter.

    If you have a combiner box with circuit breakers--You can turn off one breaker at a time and see how much the array current drops each time. Or you can even throw a dark tarp or cardboard over one or two panel in a string, and see how much the array current drops. Do these tests with lots of sun. (or you can turn on/expose one string at a time and check the current).

    AC+DC Current Clamp DMMs:

    https://www.amazon.com/gp/product/B00O1Q2HOQ (inexpensive--Good enough for most our needs)
    https://www.amazon.com/gp/product/B019CY4FB4 (mid-priced current clamp DMM)

    You could measure Voc with a meter and a good or bad string will read the same voltage. Volt meters draw almost no current and so if there is a poor connection/weak cell/etc.) somewhere, you need a lot of current to get a lot of voltage drop (Volts=Current*Resistance).

    At the moment, I cannot tell if it is the Solar Array not outputting full/rated current (you should see Imp in the 50% to 100% range at noon on a clear/cool day). Or it could be the solar charge controller (terminating charge too soon, not reaching 28.8 volt absorb setpoint and holding for 2-6 hours). It could also be the battery bank having problems (high internal resistance, not fully charged, etc.).

    When the inverter faults, what is the voltage reading (both on the "DC bus" and the DC input terminals) on the AC inverter (looking for poor electrical connection from battery bank to DC input terminals on inverter).

    What is your array size (i.e., 9x 200 watt panels or what)?

    Things I would do:
    • Reboot the Charge controller. 1. turn off solar panel power (or do at night). 2. turn off DC bus power to the charge controller. 3. Wait 1 minute. 4. Connect DC bus and let boot. and 5. reconnect solar panels (if you connect solar panels first, it can confuse the charge controller or even damage it--DC bus needs to be powered before solar panel connection).
    • Get a DC current clamp meter and check the array current (AM bright sun) and check the current for each array string. They should be very close to each other (10% or less good. 50% or greater difference, string wiring or panel problems). You have 3 "identical" strings--So doing A/B/C testing is really quick and accurate. It is unlikely that all panels have the "same fault").
    • Review the settings for the MPPT charge controller that all voltage/timer/etc. settings are correct. You could factory reset the controller and reprogram too (possible corrupted database).
    • Checking the bus voltage early evening--It is possible that the AC inverter is faulting by reading the "wrong" battery bank voltage--But having multiple problems (array+inverter, etc.) is unlikely with "new" problems.
    I am not quite sure... Do you have the original FLA batteries in parallel with the new AGM batteries? Or did you replace the old batteries with AGM instead.

    With the battery bank--Check and log each battery's voltage (resting, during "normal to heavy charging" and during "normal or heavy loads".

    You have 4x identical batteries--And ideally the voltage across each battery should match the rest during operating conditions... If you see one battery with very high or very low voltage--You need to figure out why.

    "Standby use only" batteries will "last OK" if shallow discharging (to 85% or so State of charge)--But they are designed for UPS installations (only discharged if there is an AC main power outage)... If cycled to 50% or less SoC daily, you can end up with very short cycle life (a handful of "months?).

    Your batteries, if going down to (for example) 23.0 volts after a little evening use--They either severely discharged, or not long for this world... With your genset--Are they charging as one would expect?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • MadJack
    MadJack Solar Expert Posts: 47 ✭✭
    edited March 2022 #14
    Just so I can understand what you are saying...
    WHAT IS a DC BUS? Nothing in my original NAWS purchase is called a DC Bus. And there is nothing I can find in any of the set & manuals lists a DC Bus.
    Is there another name for it?

    And yes, I have 9 235 watt panels.

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    The DC Bus is where all the cables (charge controller, AC inverter, battery bank, backup battery charger, other DC loads all meet).

    It may be real (large hunks of copper or brass with lots of wires coming to it. Or it may be simply all the wires bolted to the + and - terminals of your batteries.

    It is where you will find your 24 VDC connection points. A random example:


    This is an example of a physical bus bar. A chunk of metal with screws to hold various wires. 1-2 heavy power cables, and smaller wires going elsewhere.

    They can also be Swiss cheesed with hold to take raw wire ends:

    Or just wires stacked on battery teriminals:


    Just slang for where all of your DC connections are made (to battery/charger/etc.).

    Regarding your array. 9x 235 Watt panels (3s * 3p). The typical to maximum current I would expect to for your array (full sun/cool clear day):
    • 9 * 235 Watts * 0.77 panel+controller derating =1,629 Watt "best case" a few times a year:
    • 1,629 Watt best harvest / 29.0 volts charging = 56 Amps best case harvest to 24 volt battery bank
    • 112 Amp best case * 1/2 = 28 Amps "best case" random middle of day to 24 volt battery bank
    And each array string. Guessing Vmp~36 volts per panel:
    • 235 Watt panel / 36 volts Vmp = 6.5 amps per string "best case" current
    • 6.5 amps per string / 2 = 3.3 amps per series string "average case" current
    • 3 * 6.5 amps Imp = 19.5 Amps "best case" current from array
    • 3* 3.3 amps Imp "average worst case" = 9.9 Amps worst case "good" current from array
    • And your photo is 4.5 amps from the array (may or may not be working correctly--too many variables for me to guess right now)
    Depending on where you live and season, array tilt, etc... Your best case average harvest assuming at least 3 hours of sun per day (need your location):
    • 9 * 235 Watt panels = 2,115 Watt array
    • 2,115 Watt array * 0.52 off grid AC system eff * 3.0 hours of sun per day = 3,299 Watt*Hours per day early spring / late fall harvest estimate
    • MPPT controller output * 2,115 Watt array * 0.77 panel+controller deratings = 4,886 WH per day controller harvest (best case ~ 3 hours of sun per day)
    8.6 kWH per day output from your Outback with this array would be on the order of 5.2 hours of sun per day (pretty sunny weather/location).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    MadJack said:
    PV on the ground mount rack is 3 strings of 3 panels each.
    Multi meter test on each string (full direct sun) is typically 98 volts to 105 volts, currently at 83 volts in late day sun.
    Testing the strings at the terminal when not connected to the system will give you the VOC (Voltage Open Circuit) This will typically be 120% - 125% of VMP (Voltage under load); You can do the math from the VOC and VMP from the back of your panel. Here are a couple from the back of other panels;


     
    It looks to me like your settings are wrong. What setting did the battery manufacture say to set the batteries for?

    Your current settings appear to be 25.5 volts for Float, That is lower than any AGM battery I have seen. Many AGM batteries will hold 13+ volts at rest fully charged.



    At "Float" the battery bank is no longer accepting any current, but actually should be slightly higher than the normal resting fully charged voltage. 

    This is the voltage suggested for some 4D batteries Northern Arizona Wind and Sun sells, along with a link to their sepc sheet;



    So in a system with cycling (not in float use) the absorb setting should be 14.6-14.8 volts (or in series 29.2 - 29.6 volts) and the Float set at 13.6 - 13.8 (27.2 - 27.6 volts). This is well above the 25.5 volts your Float level is set!

    For this model, it is suggested not to let the batteries fall below 60% State of Charge (SOC) or they may not recover. I will again suggest you load test your batteries! Since the batteries are falling too low to support your inverter (do we know this setting?) They may well be below this SOC daily!

    Link to the specs of the above 4D AGM battery;

    45965.pdf (solar-electric.com)

    Here is a charge with AGM and Flooded lead acid batteries by Voltage,

    Battery state of charge for Wet and AGM - EcoSoch Solar

    Note that you are 'floating' at roughly 80% as the voltage falls slowly without a load. This setting is only allowing the charge controller to keep your battery at 80% SOC. 

    This is a basic description of how charge controller work;

    The voltage you are seeing is the system voltage and not the battery voltage. If you are connected to charging or a load it will effect the system voltage.

    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.

    The system voltage drop you see at night when the sun goes down is the charge controller moving into a resting mode with no energy to contribute to the system.

    The morning voltage may reflect a load present that is effecting the voltage level. With sealed batteries, you would want to disconnect the battery from the system and allow it to 'rest' for a while to get an accurate idea of it's SOC (State Of Charge) from the voltage.
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    You ask about the 'Bus' or a Busbar, often if you have a breaker box you make all your connections there; All your connections from charge controller and batteries are connect to a busbar for each.

    Here is mine, you can follow the large battery cables to connections to Bussbar(sorry) labeled, nearby.



    I've had 3 charge controllers connected in this without issue, though the 2 Midnite Classics are connected. I've hung a couple others Schneider and a C60 (Xantrex) without noticing issues, though I've played with the settings a bit, using them for auto equalizing, when my battery was younger. My early Midnite's seem to have issue with.




    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • MadJack
    MadJack Solar Expert Posts: 47 ✭✭
    Thank you Bill! 
    All DC connections here have specific connecting locations on each component, so a 'buss bar' is not needed.

    As for my PV panels, they are 24 volt (29v max).

    Photowhit, I will reread your post (and great looking info) in the morning with a fresh mind.

    I appreciate all of your efforts and will let you know as soon as I know anything more.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Regarding 24 volt panels... There are really two general types that are called 24 volt panels... One has Vmp~30 volts. The other has Vmp~36 volts.

    The details do matter here... I am trying to guess at what you should be seeing/looking for. Voltage and current levels will be "wrong" by 20% if I guess wrong (my luck, my guesses are not always correct).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    edited March 2022 #20
    In Reply #8, I read the charge state, as: MPPT Float.  Have never used an FM CC, but it seems that, once in the current charge cycle, it had reached Vfloat, but was unable to maintain it, so, the present Vfloat setting may not be as some had noted  ...  IMO, FWIW, etc, Vic
    Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH X2@48V, 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.