Help! Mysterious load completely draining batteries

Zachrey
Zachrey Registered Users Posts: 6 ✭
in Off Grid Solar & Battery Systems #1
Hi Everyone!

I am trying to help out a friend who has a simple, pretty standard off-grid solar setup that suddenly developed a severe drainage issue. Been working fine for years...

Here is the setup:

PV
one BP SX150B and two BP3150 (identical specs Vmp 34.5V, Imp 4.35A) and two Photowatt 1000-24 (Vmp 34.4V, Imp 2.9A)

Charge Controller
Morningstar ProStar 30 (PS-30M)

Batteries
6 Energizer 29HM (12V batteries) in 24V config (125AH per battery for a total of 375AH)

Inverter
Trace DR Series Inverter/Charger


I used my DC clamp ammeter and went through the whole system and everything works as expected. I was able to determine the duty cycle of the Sunfrost, the only DC load as far as I can tell. The Inverter was in search mode the whole time and there was nothing demanding AC power the whole time I was there.

I got to the setup around 12:30PM and Vbatt was only 25.9V with 19 amps of solar coming in so I knew the batteries had been hammered pretty good.

We ran the generator into the Trace Inv/Chgr for a bit to give them an extra boost and the next morning the power was out (LVD must've kicked in). Looked at all the possible loads (100 watt TV, some CFL lights, a 50 watt fan and that was about it). Even if the TV was left on all night, there's no way it could drop Vbatt below the LVD cutoff if batts were at 100%. Sunfrost pulls about 32AH, lights about 6AH, TV about 12AH (one movie), Fan about 16AH and the only thing that was out of the ordinary was a constant 400mA draw on the DC line to house for about 10AH (assuming 24/7load). This totals to about 75AH/day, which is about what the panels produce each day.

Can a sunfrost draw 400mA when the compressors are off?

About the only thing I can think of is to have a sleep-over in front of the system to see what could take out the batteries like this! Can the PV panels discharge the batteries to the tune of 600 watts for 12 hours?!

Any suggestions most appreciated!

Does anybody know of a decent clamp-on datalogger for monitoring batteries?
· Share on Twitter

Comments

  • YehoshuaAgapao
    YehoshuaAgapao Solar Expert Posts: 280 ✭✭
    #2
    Re: Help! Mysterious load completely draining batteries

    How old are the batteries? The other guys will probably will want you the check SGs and voltages on the batteries. My intuition would be a failed cell in one of the batteries. A bad battery/cell would show up quickly when a load is placed on them. I have not yet experience battery failure yet in a series/parallel solar setup (just dead car batteries where a shorted cell wipes out the whole battery) so the other guys will be more experienced with this.
    · Share on Twitter
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    #3
    Re: Help! Mysterious load completely draining batteries

    i agree that it could be a bad cell or the batteries themselves have just become to old cyclic wise. there still could be a phantom load on it so start isolating things to rule stuff out, but odds are there isn't a phantom load that all of a sudden popped up so suspect the batteries first.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #4
    Re: Help! Mysterious load completely draining batteries

    Welcome to the forum.

    The ProStar is a PWM type controller, so the only current will be what the panels produce as Imp. You've got three @ 4.35 Amps for 13 plus two @ 2.9 Amps for 5.8 = total 18.8 Amps. Voltages on all panels are high enough to charge a 24 Volt system. So far so good.

    However: 18.8 Amps * 100 / 375 Amp hours = a peak charge rate of just barely 5%. Add in loads that are inevitably drawing at the same time and it doesn't even make the minimum rate for those batteries. This is called chronic deficit charging. It does not show up right away; it takes time. In this case a few years. Since we don't know what the loads actually are it's hard to say how bad the situation is. If the sun doesn't shine everyday it will be much worse.

    But you must now deal with it. The Energizer HM are flooded cell Marine/RV type, no? Get a hydrometer, check the specific gravity. Even if that is good and the Voltage reads "normal" as well the chances are they are sulphated and now useless (will no longer hold a charge for a long time against loads due to reduced capacity). There are also three parallel strings of them, possibly not wired with equal length connectors to common positive and negative connection points.

    For that much battery there ought to be about 1312 Watts of panel on a PWM controller of 40+ Amps. Better yet would be an MPPT controller as the array size and battery capacity would justify it. You could have about 1200 Watts then and still manage a decent charge rate. As it is there's 450 Watts of one and 200 Watts of the other for 650 Watts total, about half the amount of panel that much battery needs.

    Right now find the four best batteries of the bunch by charging, equalizing, and testing. Remove the other two. This will improve the charging to 7.5% peak. When that still doesn't work replace the batteries with four GC2's (220 Amp hour 6 Volt) wired as per Smart Gauge method #2: http://www.smartgauge.co.uk/batt_con.html

    Absolutely get some real world numbers on power usage. Use a Kill-A-Watt to measure all the 120 VAC loads. Otherwise you'd have to install a battery monitor to measure actual Amps in and out of the bank.
    · Share on Twitter
  • solarix
    solarix Solar Expert Posts: 713 ✭✭
    #5
    Re: Help! Mysterious load completely draining batteries

    Let me think. There are batteries in the system. Therefore, with high probability, the batteries are the problem.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #6
    Re: Help! Mysterious load completely draining batteries
    solarix wrote: »
    Let me think. There are batteries in the system. Therefore, with high probability, the batteries are the problem.

    You mean like they're dead after a few years of being used and recharged with very minimal power? :p
    · Share on Twitter
  • Zachrey
    Zachrey Registered Users Posts: 6 ✭
    #7
    Re: Help! Mysterious load completely draining batteries

    WOW!!

    Thanks everyone for all the questions!! The batteries are about 6 months old AND he replaced all the batteries two days ago and got the same problem!
    · Share on Twitter
  • Zachrey
    Zachrey Registered Users Posts: 6 ✭
    #8
    Re: Help! Mysterious load completely draining batteries

    Shalom, Yehoshua,

    The batteries were only about 6 months old. And he replaced them all a couple of days ago with the same results. He runs the generator every so often to keep them up.
    · Share on Twitter
  • solar_dave
    solar_dave Solar Expert Posts: 2,397 ✭✭✭✭
    #9
    Re: Help! Mysterious load completely draining batteries

    Certainly sounds like time for a DC clamp on AMP meter to see what the load is like. If they are topped up at day end by SG readings (or maybe by resting voltage with no loads if you are without a hydrometer) then you do have a load problem.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #10
    Re: Help! Mysterious load completely draining batteries

    Then I would suspect a major wiring fault if 375 Amp hours of 24 Volt battery is being drawn down with "no loads". That is in fact 4.5 kW hours of battery potential there; it should run the place for a couple of days without any charging at all.

    First thing to check is the batteries themselves, even though new. Take specific gravity readings of all cells and write them down. Make sure the batteries really are fully charged at the end of the day, then check them all again the next morning. If you see one string is lower or higher than the other two then current is not being shared equally. That could be incorrect (see Smart Gauge) or undersized wiring.

    Yes, things like the Sunfrost do draw power even when off. You can check this with a DC clamp-on Ammeter (available from Sears if you're in the US). Make sure it is for DC: an AC meter will not work.

    If the ProStar is not fried it is impossible for the panels to discharge the batteries through the panels at night. You can check this by covering the panels or waiting until dark and see if there's any current flow through the PV leads. Also watch the Voltage when charging from the PV and see how high it gets: if the charge controller is shorted the Voltage will climb past the Absorb set point and may go up near Vmp; a sure sign that no regulation of charge is occurring.

    You can also install a battery monitor like this: http://www.solar-electric.com/tr20mosy.html

    Keep in mind that my original evaluation is still 100% valid: there is not enough panel for that much battery. Chances are that reducing the amount of battery will actually improve things by allowing the remaining capacity to be more fully charged.

    BTW 24 Volts under load is a perfectly acceptable LVD for a 24 Volt system: it will insure the batteries don't go below 50% (remove loads, Voltage will rebound a bit) while maximizing usable capacity.
    · Share on Twitter
  • Zachrey
    Zachrey Registered Users Posts: 6 ✭
    #11
    Re: Help! Mysterious load completely draining batteries

    Thank you Cariboocoot,

    Do you know of a battery monitor for sale? That would be preferable but I'll cobble one together with an Arduino and USB interface if I have to. The dataloggers I've seen so far do not measure current with a clamp-on or do not have enough memory for several days worth of data or cost WAAAYYY too much money.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #12
    Re: Help! Mysterious load completely draining batteries

    They're expensive things all right. At this point you may be able to do without, though.

    You need to undertake the "disconnect and inspect" analysis first. It's a tedious process, but highly accurate for finding problems and doesn't cost much more than time. You simply disconnect everything from the batteries, then add things on one at a time until the problem reappears. Yes, it will take days under the circumstances (if you can't find the current draw with the ammeter). Yes, it will be a pain in the anatomy to keep the place going (split the bank up; let part of it run things while the other part supplies power to find the fault).

    But all the battery monitor could tell you is how much power is being used, not where it's going. If it shows a number that there is not much power being used, then something is wrong on "the other side" (i.e. the batteries are being discharged somehow at the bank, not via loads on the other side of the shunt).
    · Share on Twitter
  • Zachrey
    Zachrey Registered Users Posts: 6 ✭
    #13
    Re: Help! Mysterious load completely draining batteries

    God bless all y'all!

    I'm going to dive into some cisterns now but will be back later today to come up with a plan of action for tomorrow. It's a 2.5 hour drive to this guys place so I'd like to make it productive! :-D

    You have have given me a number of things to look for already!
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,613 admin
    #14
    Re: Help! Mysterious load completely draining batteries

    Get an "inexpensive" DC Current Clamp Meter (this one from Sears is "good enough"--@$60 and it is an AC/DC full DMM). Note that DC Current Clamp meters have a +/- sign--So they can report on the direction of current (is the battery net charging or net discharging).

    Note that DC clamp meters "drift" over time (typically minutes) and if your are after "accurate readings" you have to rezero them.

    The Sears DC clamp meter has a weired zeroing interface. Basically, with the meter disconnected, push the "ZERO" button. On the LCD you will see a "ZERO" status or not... If you do not see a "ZERO" status, press the button again and it will not be "zeroed". Pushing the button again turns off zero and you see the uncorrected output. (i.e., pressing zero once does not "zero" the meter, it toggles the uncorrected/corrected mode--very confusing at first--why would you want an uncorrected DC current reading????).

    Other things to think about:
    • Good quality glass hydrometer+Thermometer (temperature correction)
    • Some sort of log book (keep track of specific gravity/resting/operational voltages per cell/battery)
    • A "known good" DMM (digital multi-meter) or, ideally, a DC current clamp meter like above (check meter on your car's electrical system first)

    And the optional stuff (some are the same things, different vendors):
    For flooded cell batteries... Always have a hydrometer (if 2.5 hour drive, get two or three--One will always roll off a table and break within the first year). Bring a gallon of distilled water... Before using the hydrometer, rinse it out. And after using rinse with distilled water twice before putting away (there is always "scum" in the battery cell that will cause the glass float to get sticky over time).

    For sealed batteries (AGM, etc.), and very handy for everyone else (especially for spouses/kids/guests) is the Battery Monitor. Much easier for folks understand the current state of battery bank (>75% SOC, everything fine; 50%-75% SOC think about starting genset if next morning if poor weather forecast; below 50% SOC start generator and ask for help; approaching 20% SOC--turn off all power and ask for help).

    My justification for spending $$$ for meters/tools is that the battery bank can be killed in just a few days of poor operating practices. A good meter and understanding how to use it can easily save a battery bank from early death/murder.

    But, most of us here have murdered more than our fair share of batteries--Especially the first time we went off grid/setup a battery bank. It appears to be a right of passage.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • Zachrey
    Zachrey Registered Users Posts: 6 ✭
    #15
    Re: Help! Mysterious load completely draining batteries

    Just got done vacuuming the cisterns! Ugh! "Oh it'll only take several hours!" FAMOUS LAST WORDS!

    OK,

    I have an Extech Dc clamp meter and am very happy with it. I can clamp on to panel leads and battery leads. Just plain fun to use and a great tool for solar!

    http://www.shopextech.com/product/extech-ex730-800a-ac-dc-clamp-meter

    BTW, the setup has a Bogart Engineering TM-2020 trimetric. I programmed it but the 200 foot long cable has twisted pairs of very small gauge so the shunt voltage is hopelessly lost in noise and IR drop. Ammeter readings make no sense. Need to redo that puppy with some 14 gauge wire or maybe coax!

    Now I gotta sleep and then a long drive!

    Zac
    · Share on Twitter
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    #16
    Re: Help! Mysterious load completely draining batteries
    Zachrey wrote: »
    ... Now I gotta sleep and then a long drive!

    Isn't that "miles to go before I sleep" ??
    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 ,

    · Share on Twitter
  • bmet
    bmet Solar Expert Posts: 630 ✭✭
    #17
    Re: Help! Mysterious load completely draining batteries
    Zachrey wrote: »
    Just got done vacuuming the cisterns! Ugh!

    Sorry you had to do that. I have 3 cisterns of different ages, and all of them have a bowl-shaped bottom with a drain run-off exiting the very bottom of the bowl. All I had to do was scrub with the oversized straw broom to push the deposits out.
    · Share on Twitter
  • Photowhit
    Photowhit Solar Expert Posts: 6,006 ✭✭✭✭✭
    #18
    Re: Help! Mysterious load completely draining batteries
    Zachrey wrote: »
    BTW, the setup has a Bogart Engineering TM-2020 trimetric. I programmed it but the 200 foot long cable has twisted pairs of very small gauge so the shunt voltage is hopelessly lost in noise and IR drop. Ammeter readings make no sense. Need to redo that puppy with some 14 gauge wire or maybe coax!

    Might mount it closer for a bit more info for this current situation...
    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.
    · Share on Twitter
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    #19
    Re: Help! Mysterious load completely draining batteries
    Zachrey wrote: »
    ... but the 200 foot long cable has twisted pairs of very small gauge so the shunt voltage is hopelessly lost in noise and IR drop.

    IR drop is only a factor when there is current!
    Since you are measuring voltage across the shunt, and should be using a high input impedance device to measure that voltage, there should be no IR drop.
    All kinds of noise from capacitive and inductive coupling, but no IR drop.
    Larger wires should not make a big difference, but going to coax should. Make sure that you do not ground both ends of the coax shield, since you do not want it carrying current.
    The circuit which senses the shunt voltage should have a differential input, so that any common mode voltage will not affect the measurement. But if the common mode voltage threshold is exceeded, you will get unreliable measurement.
    SMA SB 3000, old BP panels.
    · Share on Twitter
  • NorthGuy
    NorthGuy Solar Expert Posts: 1,913 ✭✭
    #20
    Re: Help! Mysterious load completely draining batteries
    inetdog wrote: »
    IR drop is only a factor when there is current!

    I experimented with cables of various length up to 20ft (not 200ft) and there was no degradation in precision of voltage measurement with 0.01mV precision. I have 0.1 mOhm shunt, so it translates to 0.1A. Chances are that if my measurements were more precise, it would've worked too. That's way better than the error of the shunt itself.
    · Share on Twitter
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    #21
    Re: Help! Mysterious load completely draining batteries
    Photowhit wrote: »
    Might mount it closer for a bit more info for this current situation...
    +1

    Also, even with twisted pairs, you need to be careful if you are running the shunt sense wires in the same conduit with either DC or AC wiring. The steady state DC will not be a problem, but spikes and noise on the DC caused by CC or inverter can be. Same with AC. If you have to run AC in the same conduit, you should twist the AC wires too if possible.

    Finally, make sure that there is a good metallic ground connection between the shunt location and the meter location OTHER THAN the shunt wires themselves. Do not depend on ground rods for this purpose.
    SMA SB 3000, old BP panels.
    · Share on Twitter

Categories