So. California SW GTI (grid tie) System with aging AGM batteries--what next

Mark S.
Mark S. Registered Users Posts: 9
I'm a new member here as of today, but an old solar enthusiast, and I appear to have a multi-faceted problem that started up recently - thought to share it with you folks.

System Specs:

Installed in early 2002: Grid-Tie PV system consisting of (40) 120 watt mono-crystalline panels two Trace/Xantrex SW4048 inverters with GTI (Grid-Tie Interface Units). I only have one inverter active at present. The PV array is arranged into 10 groups of 4 panels each, wired in series for a 48 volt system. The 10 groups run through two Trace TCB-10 combiner boxes (5 circuits on each TCB), run through a GFI unit and then to two C-40 charge controllers. The battery bank consists of (4) Concorde PVX-12255 batteries. Also, a small 400 watt 48 volt wind generator connects directly to the battery bank (no load or diversion controller). BTW, the wind genny doesn't get too much action here.

The system typically pumps good voltages back into the grid during daylight hours and the C-40's consistently read upwards of 52 ~ 53 v DC.

The problem:

Over the passed few months, I started noticing that transmitted RF signals from my Amateur radio equipment was interfering with the solar system, 'tripping' the system, especially at night. This has never been a problem up until now. When the RF signals are transmitted, the load side of the inverter's output would start pulsing, and more recently shut down ('trip') completely after a few minutes. The inverter's outages will last for a minimum of 5 minutes, and as long as up to 15 minutes before restoring full power.

Note, I do not loose AC from the grid, as standard grid connected loads remain energized, but it seems that the transmitted RF signals somehow cause the inverter to either disconnect from the grid and start pulling from the battery bank or from some other condition that I haven't determined yet.

During one of the night time power outages, I checked the voltage readings at the C-40's and on the inverter meter, and noticed low voltage readings (approx 37 v DC). When I checked the individual batteries during normal daylight conditions, I found three of them were reading 14 ~ 15 v DC and one was reading low at 8.7 v DC.

Tracking down the root cause:

I removed a 25 year-old linear power supply that was powering the radio equipment. This power supply unit was powered from the load side the inverter providing emergency back-up capabilities. My thinking was that the linear power supply might be absorbing RFI, so for testing purposes, I replaced it with a different (and newer) switching-type power supply ... this appears to have cured the 'tripping' problem.

I cleaned off a fair amount of semi-conductive dust from the regulator circuit board ans all the components in the linear power supply and found one leg of an MOV had come undone from across the 120 v AC input legs. The MOV was soldered back into place and the power supply's output readjusted to 14 v DC. I will reinstall the power supply in it's original location in the near future to determine if the loose MOV and/or the conductive dust accumulation was the cause for the 'tripping', but regardless, I still have a failing battery in the bank to deal with.

Being that the system's batteries are at least eight years old now, I'm not sure if I should purchase a single new AGM battery to replace the failing battery or to purchase an entire new bank of four. (?)

I have read that some AGM's can remain in service for up to twenty years if they are maintained properly ... has me wondering what the true track record is for AGM's ... what's a good load test for AGM's?

Another option I'm contemplating, is to just Grid-Tie the system and not maintain back-up batteries - and tie in my 6500 watt auto-start generator to the SW4048 for those emergency grid outages (?)

All replies welcome ...

Mark S.
So Cal


  • BB.
    BB. Super Moderators, Administrators Posts: 33,439 admin
    Re: So. California SW GTI (grid tie) System with aging AGM batteries--what next


    I moved your post to its own thread... It is usually easier to have all of the facts, questions, and answers in one thread for your system.

    Sort of depends on how far you want to take the system...

    Battery Bank wise--it is possible that the shorted/failed battery may have put too much charging voltage across the other AGM's in the string (15 volts?). Depending on how long--it is possible to wear out the catalyst and/or vent a bit of hydrogen during this period.

    Here is a nice thread started by Dapdan about recovering capacity/life on good quality AGM batteries.

    Perhaps load testing your remaining batteries, and possibly replacing the catalyst caps (if everything else looks OK) + 1x replacement battery may be worth a try...

    However, the catalyst caps are not cheap, and you still are working with 8 year old batteries.... :confused:

    Regarding rewiring to an Off-Grid system and use backup generator/inverter battery set... Your existing inverters are getting long in the tooth (10-15 years is probably a good life). A simple GT inverter may end up being more efficient and less maintenance issues (at least no more batteries).

    For a sanity check--how much Grid Tie Power are you feeding back to the utility. A strait GT system (Using PVWatts website) for Los Angeles, and 4,800 watts of solar panels:
    "Station Identification"
    "Lat (deg N):", 33.93
    "Long (deg W):", 118.40
    "Elev (m): ", 32
    "PV System Specifications"
    "DC Rating:"," 4.2 kW"
    "DC to AC Derate Factor:"," 0.770"
    "AC Rating:"," 3.2 kW"
    "Array Type: Fixed Tilt"
    "Array Tilt:"," 33.9"
    "Array Azimuth:","180.0"

    "Energy Specifications"
    "Cost of Electricity:","12.5 cents/kWh"

    "Month", "Solar Radiation (kWh/m^2/day)", "AC Energy (kWh)", "Energy Value ($)"
    1, 4.44, 421, 52.62
    2, 5.35, 463, 57.88
    3, 5.62, 532, 66.50
    4, 6.06, 548, 68.50
    5, 6.19, 576, 72.00
    6, 6.17, 552, 69.00
    7, 6.48, 596, 74.50
    8, 6.68, 611, 76.38
    9, 5.78, 511, 63.88
    10, 5.43, 504, 63.00
    11, 4.84, 440, 55.00
    12, 4.46, 418, 52.25
    "Year", 5.63, 6173, 771.62

    If your existing system output is similar to the above predictions--then it is working well for you and is probably not worth the expense, rewiring of the solar arrays, permits, new GT Inverter ($3-$4k or so) to upgrade your installation. One battery or one bank every 8+ years is not bad.

    If/when your inverters fail on you--then you have the choice of going pure Grid Tied or replacing your SW+C40's with a Xantrex XW Hybrid Inverter + MPPT Solar Charge Controllers. Plus a new ~600 AH 48 volt battery bank.

    What is it you are looking for (simplicity, low costs, low maintenance, OK with backup generator)?

    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: So. California SW GTI (grid tie) System with aging AGM batteries--what next

    My guess:
    As batteries age and loose capacity, when you transmit, it loads down the inverter, which draws from the weak batteries, and imposes a 60Hz/120Hz ripple load on top of the DC, which likely drives the volt and trip point sensors in inverter and charge controllers, nuts.
    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

    gen: ,

  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: So. California SW GTI (grid tie) System with aging AGM batteries--what next

    i don't think the rf was the root cause of the trouble, but may have had a hand in it as just a set of circumstances that happened. we may never know for sure as it is difficult to keep rf out of the electronics due to the strength of the rf field. some filtering may help, but it was just the battery's time. the others are probably not far behind, but one just does not know when. to mix a new battery with the old ones is a risk as that new battery will be drawn down to the level of the old ones. maybe you could get a new set of batteries and keep the old bank for something else, ham radio maybe?
    i have seen my sb50 kick on with 100w of rf as it engages the relay and led as if it is working with pv input, but i use the battery for tx/rx and not a power supply. i can only make suggestions as you may need to think this one through. no easy answers.
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: So. California SW GTI (grid tie) System with aging AGM batteries--what next

    You might get by with just replacing the one battery but age of batteries is against you. I am betting the one that drops to 8.7 vdc under load has high resistance cells due to positive grid corrosion. The three others may have gotten enough current to keep them charged. Grid corrosion is accelerated due to overcharging.

    You can try changing just the one bad battery but check the voltage balance after they are well charged and in float mode. If the voltage on each battery varies more then 0.2 vdc then other three batteries may have to also be replace. Set float and bulk voltage setting on SW's per the battery manf. recommended levels. If you set PV controller to 0.2v higher then SW bulk setting you are covered whether the SW is in bulk or float mode. Just keep in mind that if SW's go offline for some reason you lose their battery voltage clamping. In this case don't let PV charge long period at the higher bulk voltage.

    AGM cells usually die with high resistance cells, not shorted.

    The SW's will clamp the battery at the SW's bulk and float voltages. The PV controller needs to be set higher. SW's will pull off any excess above SW's set regulation voltage.

    You also have too much panel wattage to be stuck with a PWM controller. You are wasting some of PV available power. MPPT controller would improve yield.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: So. California SW GTI (grid tie) System with aging AGM batteries--what next

    also, if you can see if you can load test the remaining batteries per concorde's specs as this may help you make up your mind if you should take the risk of 1 new battery or not.
  • Mark S.
    Mark S. Registered Users Posts: 9
    Re: So. California SW GTI (grid tie) System with aging AGM batteries--what next

    Sorry its taken me so long to respond ... I was recuperating from surgery when I posted my original questions and had plenty of time to sit at the PC. Since I got back to work in May, it's been real busy.

    As for the bad battery, I did replace a single battery in late March or early April 2010, and all is still looking good so far.

    To help with the HAM radio RFI issues, I had recently contacted Bob Rumsey of Balun Designs. I was planning on ordering a 1:1 current balun for my G5RV antenna - thought it would help to reduce RF on the property. We discussed the RF getting into the 'Solar System' and Bob recommended placing ferrite filter chokes on the DC wiring in my combiner boxes. Bob stated that the PV array acts like an antenna and absorbs RF when they aren't generating power. I am happy to report that the chokes corrected that problem!

    BTW, I purchased a balun (model # 1115dt) ... they build nice, high quality baluns if any of you might be in the market.

    Thanks for all the replies ...
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: So. California SW GTI (grid tie) System with aging AGM batteries--what next

    good to hear you're doing ok now and i understand of going from one extreme to another with the first of lots of time on your hands to 'i need a vacation'. glad you made it back here. on that particular antenna going from unbalanced coax to a balanced ladder line always necessitates a balun. i like to skip the coax and run ladder line the whole way. most tuners are not properly set up for keeping the balance even though they can still transform the impedance.

    good luck with your battery bank.