Low Voltage Disconnect

Hello. I have an AIMS 1250 watt modified sine wave inverter that is connected to a 12V, 105 amp/hour Concorde Sunextender battery, all metered through a Prostar 15 amp Morningstar controller. The battery is charged by a 130 watt Kyocera solar panel (it delivers this power at 16-17 volts). The inverter is connected to the battery by six-foot long #4 AWG inverter cables.

The load on the inverter is six 19 watt compact fluorescent bulbs (each with the equivalent of 75 watts incandescent). When I turn the system on (whether starting the inverter and load together or starting the inverter first and then turning on the lights), the inverter "whines", it's voltage meter drops, and then within about ten seconds, the battery voltage drops to less than 10V and the system shuts down. The battery is always fully charged before this happens. The Morningstar controller is registering this as a "low voltage disconnect", which, I assume, means the drain on the battery is too great, causing the voltage of the battery to drop below a critical level.

Is this caused because the total wattage draw of the bulbs (114 watts) is too great, or is it the fact that the wattage draw is OK, but (as an engineer friend told me) the "fluorescent" bulbs themselves require too much of an initial power drain on the battery? Or is it the fact that even though the bulbs are only 19 watts, their incandescent equivalent is 75 watts, so I'm really drawing 450 watts off the system? Or is it because the inverter cables are too "thin"? Or is the inverter itself :? :?the culprit? Or is there some other problem? The electrical connections are all proper, and sometimes the system works without any problem, and sometimes it just doesn't. I assume the battery has enough power (1200 watts) to power these lights (at 114 watts) for 5-6 hours, but it never seems to happen.

Would adding another battery solve the problem, and if so, should I add it in parallel, to increase the amperage available, or in series, to increase the voltage of the system? Or should I change all the fluorescent bulbs to incandescent, which of course would then increase the wattage draw to 450 watts? Or do I need to do both? Do I need to upgrade the inverter cables? Or do I need a pure sine wave inverter?

I called Morningstar about this, and they told me NOT to connect the inverter to the "load" leads of the controller, that sometimes this could cause the system to shut down. I did in fact have the inverter connected that way, but when I removed the connections, it did not solve the problem.

I've even tried just starting some of the bulbs, or even one, while leaving the rest off, but the problem still happens. Sometimes the lights will come on, but then go out within about 60-90 minutes, which doesn't seem right if there is only a draw of about 200 watts total during that period of time.

Any suggestions would be appreciated!

Thanks,
Lou Dalessandro

Comments

  • halfcrazy
    halfcrazy Solar Expert Posts: 720 ✭✭✭
    Re: Low Voltage Disconnect

    i had a xantrex ux612 do the same thing as soon as you turn it on full load on batterys and they almost instantly drop to low voltage. in my case it was a bad inverter dont know what it was inside but something musta gone bad?
  • Roderick
    Roderick Solar Expert Posts: 253 ✭✭
    Re: Low Voltage Disconnect

    If you have the means to measure the input current going to the inverter, that might help as a debugging measure.

    As a ballpark, let's say you're drawing 120 watts AC, and the inverter has horrible efficiency at 50%. Then you would expect to draw 240 watts at the input, or 12 V * 20 Amps. If I saw more than 20 A being drawn, I would suspect the inverter. If I saw less than 20 A, I would suspect the battery If I saw less than 10A, I would suspect my test equipment.

    If you aren't able to borrow or get the equipment to measure current, you could always put (say) a 15 or 20A fuse in series with the inverter input, and see if it blows.
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    also consider battery health, does it hold its normal healthy voltage (12.6 or 12.7 in he morning if unused all night?

    can you try tuning on the lights one at a time? it just might be too much startup draw all at once for your lonesome battery. definiyely do not switch to incandecents. i have never hard of CF lights having any kind of notable startup surge but that doesnt mean it is not happening.

    definitely hard to say, suggestion to measure draw is definitely going to help you narrow it down. another alternative: can you add up the rated wattage (forget out equivalent incandecent light output ratings entirely!) of all bulbs, and find a similar sized load to run and see if it does the same thing it on your sine wave inverter?

    also as an alternative to not measuring the amp draw: if you have something that runs off 12v directly (not using the inverter) similarly sized load to determine if the batetry voltage drops with that too? if THIS runs ok your inverter is incredibly inefficient AND/OR the combined load of the inverter and bulbs is too great, which would be my first guess with the one battery.

    : oh try removing one light at a time from the load to see how many you have to take off unil the prob goes away.
    expanding he battery bank in parallel to stay at 12v, would help but there are issue of mixing old and new batts, and determining if the one panel will still be enough to charge both.

    back to battery health: is it fully charging regularly and keeping its voltage? a weak batt drops its voltage quicker.
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    OK - thanks! More stupid questions. 1 - when measuring the "input current going to the inverter", do I measure between the battery and the inverter, or between the AC load and the inverter?

    Stupid question number 2 - the Morningstar charge controller has a load measuring capability and is already measuring a load of roughly 1-1.5 amps drawn on the battery (this includes AC load plus the inverter). The POS and NEG terminals of the inverter are directly connected to the appropriate "load" input terminals on the charge controller. By your analysis, this would indicate that the testing equipment itself is faulty, so, what other device could I use to measure the input current (besides a fuse)?

    The controller also is connected to the battery, and indicates 13.7-14.5Vdc when the battery is fully charged, then it stays at "Pulse Width Mode", which I suppose is a trickle charge. Finally, the controller is also connected to the PV panel - when the battery is fully charged, the amperage input from the panel registers as "0" (probably because no further charge is needed for the battery) and when load is applied to the battery, the charge conroller will read anywhere from 1-5 amps from the PV panel, depending, of course, on AC load, time of day and amount of sunlight available. Up till now, I've had no reason to suspect the charge controller itself.

    Stupid three - the battery sat for about 2-3 months before being used and charged, although it seems to hold a charge OK. Does this matter?

    Finally, #4 -if I get a second battery, do I ditch the first one, or is there that much of a problem mixing old and new batteries?

    Thanks,
    Lou Dalessandro

    Roderick wrote:
    If you have the means to measure the input current going to the inverter, that might help as a debugging measure.

    As a ballpark, let's say you're drawing 120 watts AC, and the inverter has horrible efficiency at 50%.  Then you would expect to draw 240 watts at the input, or 12 V * 20 Amps.  If I saw more than 20 A being drawn, I would suspect the inverter.  If I saw less than 20 A, I would suspect the battery  If I saw less than 10A, I would suspect my test equipment.

    If you aren't able to borrow or get the equipment to measure current, you could always put (say) a 15 or 20A fuse in series with the inverter input, and see if it blows.
  • rplarry
    rplarry Solar Expert Posts: 203 ✭✭
    Re: Low Voltage Disconnect

    Lou
    You have gotten some good advice, the only other thing I would try would be to take the battery somewhere where it can be load tested. that is about the only way to truly diagnose the battery's health. From what you are discribing it sure sounds like a faulty battery. Many automotive places that sell batteries have the test equipment to load test your battry and they usually do it for free hoping to sell you a new battery. If your battery checks out as bad then do not just add another battery to it. Get a new battery.
    Larry
  • SolarJohn
    SolarJohn Solar Expert Posts: 202 ✭✭
    Re: Low Voltage Disconnect

    You said: 
    sometimes the system works without any problem, and sometimes it just doesn't.
     

    That is a good clue, and I think it points away from the battery as the most likely source of the problem.  If you're sure that the battery is fully charged before each test, I believe you should get 4 to 5 hours out of it with the 6-cf light load before the state of charge drops to 50% or so.

    I would start troubleshooting by replacing the load.  Since your cf bulbs total 114 watts, test by using two 60-watt bulbs instead.  If the problem doesn't occur after several tests, one or more of the cf bulbs is the culprit.

    You may have already done this, but connect the inverter directly to the batteries, not to the charge controller.  Double-check all of your connections, and use a digital voltmeter to make readings.  Take readings when the load is on.  Look for crimped on connectors that are heating up, as that would be a sign of a poor connection, and would account for the voltage drop. 

    If this problem persists, I would suspect the inverter.  You may be able to send it away for testing and repair.  I remember an ad for a company that specializes in inverter repair, but I don't have it handy.  Try Googling "Inverter Repair", and see what happens.

    Good Luck!
    John
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Low Voltage Disconnect

    the load from the cfls should be in the neighborhood of 10amps from the batteries with it actually being slightly over 10amps due to the efficiency level of the inverter you have adding a few amps. do confirm that the draw is in and about that figure to rule out any partial shorts in the inverter. the controller measures the current from the pvs and not loads from the batteries as far as i know. if you wish you can run your the inverter and cfls connected off of your car's electrical system. if that battery goes down to low voltage quickly, you know the inverter is at fault. the car battery will then need a good charge or jump so be prepared.
    now the batteries you have may be no good anyway as you don't know the condition they were in by the time you gave them a charge. if the batteries have a provision to view the water levels then please pop it off to look. if you see the plates above the water(electrolyte) level then those batteries have sulphated and aren't long for this world even if you fill them up with distilled water or not. the filling with the water may give you back some life, but count on having to replace them. those batteries could still be bad even with good water levels and a load test would show their condition.
  • Brock
    Brock Solar Expert Posts: 639 ✭✭✭✭
    Re: Low Voltage Disconnect

    Wait a minute. IS the inverter directly connected to the batteries? It should be connected to the battery with at least a #4 wire, straight from the inverter to the battery. You didn't connect it through the Morning star did you?

    Another quick thing to try. Put a regular volt meter on the battery and take a reading. Then as you turn on the lights watch that voltage and let us know what it is. If your getting a low voltage tone on the inverter and the battery is sitting at say 12v or better your wire from the battery is to small or has a bad connection.
    3kw solar PV, 4 LiFePO4 100a, xw 6048, Honda eu2000i, iota DLS-54-13, Tesla 3, Leaf, Volt, 4 ton horizontal geothermal, grid tied - Green Bay, WI
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Low Voltage Disconnect

    Are your CF lights happy running on a non-sine inverter ? Maybe the inverter is having trouble driving them because of some weird interaction with the micro-ballast inside each bulb ? I like the idea of a couple 60W incandescent bulbs as a test.
    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 ,

  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: Low Voltage Disconnect

    All batteries suffer an initial voltage "dip" when first loaded known as the "coup de fouet", or "crack of the whip". The battery voltage drops at first, but then recovers somewhat as the electrolyte mixes inside the battery.

    That's a big inverter for a 12 V x 105 Ah battery and your load. I suspect that the inverter's start-up surge -- whether loaded or not -- is causing the battery voltage to drop enough to trigger the low voltage disconnect. Indifferent battery care will have reduced capacity and compounded the problem.

    An appropriate size inverter for your 6 bulbs x 19 W / bulb = 114 W would be ~300 W.

    HTH,
    Jim / crewzer
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect
    crewzer wrote:
    An appropriate size inverter for your 6 bulbs x 19 W / bulb = 114 W would be ~300 W.
    HTH,Jim / crewzer

    I was going to suggest that too, as 1250 watt inverter is WAY overkill, with a huge idle current draw compared to a 300 watt inverter. I've used many different kinds of Cf's over the years and never had a problem with any of them on MSW. In fact I have an old 150 watt Statpower, with only a 20 ma idle draw, always on, ready to supply my Cf's day or night. Bigger inverters are switched and dedicated to specific loads.
    Wayne
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    This makes sense. Even testing with two regular 60-watt incandescent bulbs causes the same problem, so I don't think it's the CF bulbs.

    The battery voltage recovers quickly after these events, so I think the battery is OK too - it just needs to be free of this large inverter.

    I've ordered another battery, so now I'll have 210 Ah to fool around with. I was thinking of purchasing a smaller inverter, a Samlex 600 watt pure sine wave inverter to go with these batteries. Is this too big of an inverter for the two 105 Ah batteries?

    Finally, what is the best type of battery-inverter cable to go with the Samlex inverter? I think this particular inverter accepts "stud" connectors on the inverter end. Also, what is the best gauge cable? I already have #4 AWG gauge cable on my current inverter, and would rather not have to change that if I don't have to. However, both ends of these cables are ring connectors - is there an "adapter device"  that would change one ring connector end to a stud connector end so as to allow the Samlex inverter to utilize these cables?

    Thanks for your help!
    Lou Dalessandro
    crewzer wrote:
    All batteries suffer an initial voltage "dip" when first loaded known as the "coup de fouet", or "crack of the whip". The battery voltage drops at first, but then recovers somewhat as the electrolyte mixes inside the battery.

    That's a big inverter for a 12 V x 105 Ah battery and your load. I suspect that the inverter's start-up surge -- whether loaded or not -- is causing the battery voltage to drop enough to trigger the low voltage disconnect. Indifferent battery care will have reduced capacity and compounded the problem.

    An appropriate size inverter for your 6 bulbs x 19 W / bulb = 114 W would be ~300 W.

    HTH,
    Jim / crewzer
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    i just put one in, the manual i THINK says 4/0 , (i used 6guage with much bigger loads than you with no probs, so even if im wrong about the manual you should be fine with 4). it comes with what i think is called a post adapter so you'll need to cut off the inverter end of those cables to put into it, or you might get something easily enough at the hardware store to convert to a post adapter. cutting them off, strip back a little and crimping them on was pretty easy. i dont have the crimp tool but a vise worked pretty good ;) i just looked and couldnt find a picture for you, sorry
    arthuritis wrote:
    Finally, what is the best type of battery-inverter cable to go with the Samlex inverter? I think this particular inverter accepts "stud" connectors on the inverter end. Also, what is the best gauge cable? I already have #4 AWG gauge cable on my current inverter, and would rather not have to change that if I don't have to. However, both ends of these cables are ring connectors - is there an "adapter device" that would change one ring connector end to a stud connector end so as to allow the Samlex inverter to utilize these cables?
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Low Voltage Disconnect

    they should be specifying a length that accompanies the gauge number used. if the wire isn't thick enough you could parallel a like wire with the present ones to allow more current to flow with the same voltage drop.
    as to the 2 105ah paralleled batteries being used with a 600w inverter i think you'll be ok with the loads you've got in mind now. you don't want to utilize all 600w of inverter capability as this would be greater than 600w/12v=50amps. that is a large % of your battery capacity. 50/210=about 25% which is a high draw rate putting you at 50% depth of discharge(dod) in 2 hours or less with more emphasis on less. your lights represent less than one fourth of that draw making them a good match to your current batteries. the extra inverter capacity you could use to temporarily run other loads in addition to the lights, but will degrade the amount of time left on the batteries for running the lights. if planning larger future loads to use on the inverter then my suggestion is to get a 3rd battery or even more depending on how close to the inverter capacity you go.
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    Just a question - do you usually have all 6 lights on at the same time? Or never all on at the same time? Do you have loads other than lights that you want to run off this proposed 600 watt Pure Sine inverter? Are these other loads sensitive to MSW?
    What I'm getting at is do you need an inverter that big and do you realize that the idle current is usually much larger with a pure sine inverter than with a MSW?

    Reason I'm asking is I don't know your situation and with your battery capacity, trying to figure out if you need an inverter that big and power hungry. If you need it, that's fine, but if you don't, it's like putting a 747 engine in a car.
    All the best
    Wayne
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    the idle draw when on is something like 7w and definitely a few watts even off so while the inverter would do the job you might get a battery switch too (im waiting for mine to come in monday actually)

    if your system might grow its a good choice
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    Actually, I've come to realize that a 300-400watt MSW inverter would probably be best - and a lot cheaper than a pure sine wave inverter.

    All six lights will be on at once (114watt draw), but no other loads will be applied, so I guess the samller MSW inverter will be enough.

    One final question - what's the best gauge battery interconnector cable to connect two 105Ah 12 volt batteries? They'll be in the same battery box.

    Thanks,
    Lou Dalessandro
    Just a question - do you usually have all 6 lights on at the same time? Or never all on at the same time? Do you have loads other than lights that you want to run off this proposed 600 watt Pure Sine inverter? Are these other loads sensitive to MSW?
    What I'm getting at is do you need an inverter that big and do you realize that the idle current is usually much larger with a pure sine inverter than with a MSW?

    Reason I'm asking is I don't know your situation and with your battery capacity, trying to figure out if you need an inverter that big and power hungry. If you need it, that's fine, but if you don't, it's like putting a 747 engine in a car.
    All the best
    Wayne
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    actually saw today on the mx60 that the batteries were floating at a nice minimal 20w (inc'd the web server computer,)flipping on the inverter with no loads jumped 20w! it did settle down to around 16w but still, ouch. sometimes its hard to get exact numbers without a battery meter or measuring the amps going from the mx60 display, but when the batts are very healthy like this and floating at a low wattage its pretty close.
    mattl wrote:
    the idle draw when on is something like 7w and definitely a few watts even off so while the inverter would do the job you might get a battery switch too (im waiting for mine to come in monday actually)

    if your system might grow its a good choice
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect
    mattl wrote:
    flipping on the inverter with no loads (and consumption) jumped 20w! it did settle down to around 16w but still, ouch.


    I find it interesting that my 10 or more year old Statpower 150 watt MSW inverters using older technology, use only 20 ma on idle. I also have a couple of 250 watt, same age Statpower inverters with about 30 ma idle. Thats less than one half watt!! That's why I can leave them on 24/7 and not worry about the batteries. Granted they were expensive, but all inverters were back then.
    My question is why, with much newer designs, must the idle current be so outrageously high?
    I have inverters that prove that isn't necessary.
    Newer Delco brand MSW 200 and 400 watt inverters that I have are PIGS when it comes to idle current, at about 5 watts!
    What's going on?
    Wayne
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    maybe the small , low priced models are to blame. i am sure that your better ones must have som logic to sense when to sleep. i am not sure what this one uses when its switched off but i think it must be a couple watts which i still find unacceptable.. still it was cheap. something i do know about this one is that there is a large capacitor that starts chargig when you power it on, so that manual warned me (i dont know what that is/does except use power obviously) when hooking it up the manual warns of a spark, i was pretty nervous but it was small. ;)

    once i hookup my battery switch which is coming in tomorrow ill be able t see wht it uses when "off".
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    Mine don't have a sleep function, they're always on, always full power available. I guess just better made by people who understood the importance of not wasting power. That and the fact that most small MSW's are now, like you suggested, very cheaply made in China for pennies. Pity.
    Wayne
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    Update on the situation.

    I replaced the AIMS 1250 watt inverter with a 400 watt Vector MSW inverter. It works fine, and interestingly enough, now I'm getting a true amperage reading from the six lights via the charge controller - they draw 9.6 amps (plus 0.5 amps for the inverter). So, for some reason, the first inverter's connection to the charge controller was not allowing the true reading of the load - the controller only registered 1.5 amp draw at the most when the old inverter was used. My guess is that the controller never sensed the true amperage draw on the battery, so the battery gradually discharged "right under the controller's nose". Is this interpretation correct?

    This brings up the next question - can a discharged battery ever be salvaged? This is a Sunextender 105Ah battery, and although I can recharge it to 14.5 volts, if I put any load on it at all at night (without amperage coming through the solar panel) it immediately discharges. Is this because the amperage is effectively gone from the battery, and can never be replaced, or is it because the amperage is OK, but the battery just can't hold a charge to deliver the amperage? During sunlight hours, I can get the battery to register 14.5 volts (full charge), and the controller indicates no amperage coming through from the solar panel - probably because the controller recognizes the battery has a full charge, and hence will not put any more amps into the battery. Am I reading this situation right?

    So, again, is this battery salvagable, or should I just ante up for a new one?

    Thanks,
    Lou Dalessandro


    Mine don't have a sleep function, they're always on, always full power available. I guess just better made by people who understood the importance of not wasting power. That and the fact that most small MSW's are now, like you suggested, very cheaply made in China for pennies. Pity.
    Wayne
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Low Voltage Disconnect

    it's a shame, but i believe it's possible the first inverter truely drew too much current due to being defective and not just a bigger inverter drawing more. this strained the battery as 10v is a half volt lower than being dead or 100% dod. also know that 14.5v is an overcharge. the proper range is 14.2-14.4v and as usual i recommend the bts for the controller. you could get the battery load tested to be sure, but i think it is dead.
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    Thanks. Just for clarification, when you say that 10V is a half volt lower than 100% dod, what did you mean? Also, what is "dod" - is that "dead on demand". Finally, what is "bts" for the controller?

    Lou Dalessandro
    niel wrote:
    it's a shame, but i believe it's possible the first inverter truely drew too much current due to being defective and not just a bigger inverter drawing more. this strained the battery as 10v is a half volt lower than being dead or 100% dod. also know that 14.5v is an overcharge. the proper range is 14.2-14.4v and as usual i recommend the bts for the controller. you could get the battery load tested to be sure, but i think it is dead.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Low Voltage Disconnect

    "DOD"=Depth of Discharge....

    DOD=1-SOC (State of Charge)
    0%=fully charged
    20%=80% full
    50%=1/2 full
    80%=20% full
    100% DOD = 0% charge

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Brock
    Brock Solar Expert Posts: 639 ✭✭✭✭
    Re: Low Voltage Disconnect

    10.5v is considered a "dead" or completely empty battery, but this is under no load so that can make s big difference.

    I, unfortunately, would also think the battery is toast.  My experience with batteries has shown that is how they behave when they are all used up.  They basically act like a smaller and smaller battery until the point where they aren’t useful anymore and often need more and more power to charge them back to “full”.
    3kw solar PV, 4 LiFePO4 100a, xw 6048, Honda eu2000i, iota DLS-54-13, Tesla 3, Leaf, Volt, 4 ton horizontal geothermal, grid tied - Green Bay, WI
  • System2
    System2 Posts: 6,290 admin
    Re: Low Voltage Disconnect

    bts = battery temperature sensor. batts need more voltage in the cold vs room temp in the summer, as an for example.