Not enough power over night

James Wilson
James Wilson Registered Users Posts: 28 ✭✭
We have finally moved to our new fully solar place in Western North Carolina. :)

We have been noticing that our battery voltage is very low in the mornings, even after a full charge the day before.

We have eight Rolls S-530 batteries wired in two parallel strings to feed our Trace 4024 inverter.

According to our Killowatt workalike meter we are only using under four KVA hours per day and rarely use above 300 VA at any given time. We have kept our Outback MX80 charge controller in equalize for a few weeks now in the hope of performing an equalize. The battery voltage will get up to around 28.2 volts in the afternoon and start bubbling, but not enough to complete an equalize cycle. If I take the charge controller out of equalize the batteries are generally floating at 2:00 pm.

We're in kind of a valley so full sun starts at around 10:00 am and the panels start getting shaded at around 3:30 pm.

By about 9:00 pm the batteries are down around 24.4 volts with about a 150 VA to 200 VA load on them. Next morning they are much lower than I would like - 23.8 to 23.9 volts! (with a 60 VA to 120 VA load)

Of course I'm measuring voltages under a load - it's kinda tricky to shut down the whole place for an hour or three for a real voltage measurement. I'm looking for a battery monitoring system now so I can see what's going in and coming out of the batteries.

Does this sound like this is working correctly? I was trying to size the batteries for at least about two days with no input and get no lower than 50% DOD, but it looks like we're getting too low overnight!

We do seem to make enough power, but we don't seem to be able to store it.

Comments

  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Not enough power over night

    first of all we need to know how big your system is. you know, how many of what pvs are you using and give the voltage configuration you are using for them. ie 24v, 48v, etc. now by the way you talk the batteries appear to be a 24v configuration, but the controller is either a flex max 80 or an mx60. which one do you have and do you have a bts?
    as to this, "We have kept our Outback MX80 charge controller in equalize for a few weeks now in the hope of performing an equalize.", know that no battery system should be kept in eq for that long. most will eq for a few hours typically, but if the batteries are not getting their full charge then do know that eq won't be entered into. you say it does reach the float stage so maybe the controller just isn't programmed correctly for the eq stage. for the benefit of those who are more intimately familiar with the outback controllers please give them the general settings you have on the controller. also check the batteries for their electrolyte levels and specific gravity and let everyone here know what you find.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Not enough power over night

    Do you have a genset you can run 8am - 10am, to bulk the batteries, and then let solar EQ it ?

    May need to play that game a couple of days to fully charge the batteries. You have to get a real charge into the batteries, deep discharges like you are doing, will kill them soon.
    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 ,

  • BB.
    BB. Super Moderators, Administrators Posts: 33,613 admin
    Re: Not enough power over night

    It is difficult to tell how well your system is performing... What are the size of your solar panels (watts).

    300 VA maximum... VA is not power--You should be talking about Watts (VA is either equal to or greater than Watts)... So, are you talking about 300 Watt maximum load on the AC side of your system?

    Yet you guess that you use 4kVA*hours per day--Let's call it 4kWHr per day (or 4,000 Watt*Hours).

    Assuming your weather is similar to Asheville NC--using 0.52 derating, a 1kW solar panel system will generate around 72kWH per month, or 2.4 kWhrs per day (average for month of July).

    Your system, just to break even, would need around:

    [4kWH/2.4kWH]* 1kW of panels = 1.667 kW or 1,667 watts of Solar Panels (PTC).

    If you have more than ~1,667 watts of solar panels--then somewhere you are having problems (more power than you think you are using, failing batteries, etc.). If you have less than 1,667 watts of panels, then you may just have too few Watts of solar panels (lots of assumptions here--numbers are very approximate).

    Regarding battery size... 4kWatt*hours at 24 volts = 0.167kAH or 167 Amp*hours per day

    Your batteries are S-530 6V, 400Ah (20Hr)... And you have eight of them.

    400AH*8 batteries * 6 volt / 24 = 800 AmpHour at 24 volts battery bank...

    Estimated battery bank should be 6 day storage*167Amp*hours per day=1,000 Amp*hours at 24 volts...

    Assuming I got all of the numbers correct--you battery bank is slightly undersized for the load--but well within functional limits (you are not damaging the batteries with daily usage).

    Regarding a good battery monitor--The Xantrex XBM has been recommended by others here... And the new Xantrex LinkLITE/PRO series looks very interesting.

    With a battery monitor--you will very quickly see if you have enough solar panels for the daily load you are using from the batteries.

    Understand you cannot shut down the loads for 3 hours and measure the resting voltage--but you should be using a hydrometer with temperature correction to check the electrolyte levels--very accurate. Just a pain to do this several times per day (plus you make a mess, can contaminate the electrolyte, and loose a little bit each time). This will tell you the state of change of the batteries at the end of the charging session and at the beginning the next day.

    You will probably need to check all the cells a couple times (make sure all are matched--if not, you have other issues)... Then you can just take a couple samples every so often to determine charge state (and switch cells every month or so).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Not enough power over night

    Consider what is running at night that could run during the day instead? Pump water during the peak sun, let the tank pressure drop over night. Phantom loads that could be shut off at night? Internet modems, routers unplugged at night?, satellite receivers, TV sets fully unplugged? Cell phone chargers, radios, anything with a brick power supply?

    Tony
  • James Wilson
    James Wilson Registered Users Posts: 28 ✭✭
    Re: Not enough power over night

    We have four Sharp 167 panels, then two strings of four of Sharp 162 watt panels. This is a rated 1,964 watts. These feed an Outback MX80 which is configured at a nominal 24 volts for two parallel strings of four Rolls S-530 batteries. We don't have great location for the panels, but still manage 9 KWh on good days and average about 6 KWh as viewed on the MX80 .

    I assume that bts is a temperature sensor, yes we have one stuck right in the middle of the batteries.

    The MX80 is set to 31 volts for two hours for equalize. I just haven't seen the batteries get above 28 or 29 volts. I did turn off the inverter one day for half an hour to try to let them equalize, but this irritated the pump controller on the solar water heater and I spent a couple of hours trying to get the loop cooled down enough to run again. Good lesson there - need an alternate power source for the pump.

    At any rate, the batteries do bubble, just not as vigorously as when at 31 volts. Leaving the controller in EQ does let them bubble for longer periods of time, but it's not like the batteries are constantly in EQ because they are never really EQing. Does that make sense? :)

    Electrolyte levels are fine, but I do need to find the hydrometer (recently moved) and check the specific gravities - could be a bad cell taking everything down.

    While we are here - when I installed this battery bank I just hooked up two strings in series the paralleled them at the ends. Should the middle batteries be paralleled also? Would it be a bad thing to parallel the middle batteries now that they are a year and a half old?

    I do have a nominal 6 KW diesel generator that I can use via the Trace 4024. I wanted to try and use it for EQing, but it's currently a little far away fro the 4024 and the 4024's charge drops out when I put it into EQ mode from the generator. Good idea about letting the generator do the bulk and let the MX80 do the EQ, I'll try that.

    I've been measuring VA on the AC side instead of watts because of some of the reactive loads on the AC side. Watt hours (real) used will be somewhat less than the VA hours (apparent) used.

    Four KWh per day would be a safe maximum number to use as measured by a killowatt meter - 9:00 am to 9:00am.

    I thought that battery banks should be sized for three days, not six! Ok, five is good! :)

    I figured this differently - say 400Ah per 6 volt battery = 2400 watt hours per battery. I think that works out correctly.

    Then 2400 Wh per battery time 8 batteries = 19,200 Wh. That seems to be right...

    Then multiply 19,200 KWh by .5 for 50% depth of discharge for a capacity of 9,600 Wh. Am I making sense? :)

    Then a usage of 4,000 KWh per day gives me 2.4 days of capacity. Please point out any flaws to this reasoning!

    I don't think that I'm damaging the batteries - I trust the MX80 to tell me what's going into the batteries and loads. Of course the killowatt meter is on the far side of the inverter from the batteries so it isn't telling me the full story. I was looking at the Pentametric meter to get the full story and to let me know how much I put in from the generator and hydro (when not in a severe drought). The Pentametric is expensive and installing all of the shunts will be a bit of a pain, but it really does look like something to monitor in and out of the battery is very necessary for an off grid system! I'll look at the Xantrex also.

    We have looked at our overnight loads, but they still don't explain where we're losing 50% of our power...

    The refrigerator will decide to defrost any time that it feels the need and I haven't been able to find a schematic so that I can install a nighttime override switch.

    Smoke alarms use about 17 watts constantly, then there's another 20 watts that I haven't found that are probably a combination of LED night lights, GFCI breakers, solid state timers, and cell phone chargers. None of these really register by themselves, but may add up to 20 watts.

    I found that the solar water heater pump was using about 60 watts at least past midnight - the controller claimed that there was enough temperature differential to be of some use. I turned it up to full throttle where it uses 100 watts and put it on a solid state timer - much better.

    Tivo went on a timer also. Cell Internet access, WiFi router, everything else goes off after about 9:00pm and don't go on until the panels are making power the next day.

    The clamp-on AC/DC ammeter is on its way! At any rate, I sized the system with some parasites like this in mind.

    We are in sort of a valley so we only get direct sun from about 10:00am to 4:00pm. :( One thing to do to alleviate this will be to put four of the panels on the house which faces West and gets full sun until about 7:00pm at this time of year. I'm sure that a lot of the losses we have now are because we just have too many hours with all of the power coming from the batteries.

    Does anyone know if we can parallel these Sharp panels if some are in full sun and the others are fully shaded?

    Thanks a lot for all of the answers and ideas!
  • BB.
    BB. Super Moderators, Administrators Posts: 33,613 admin
    Re: Not enough power over night

    You have the batteries figured out... Watt*Hours vs Amp*Hours*Volts is the same thing...

    Just watch out for those small loads... Add them up to a 100 watts * 24 hours per day--and you got 2.4kWhrs--or over 1/2 your load.

    Add up some 80% for battery efficiency and 85% for inverter efficiency... 6kWH in:

    6kWhrs * 1/.8 * 1/.85 = 4.08 kWhrs usable at the inverter output...

    So--using very rough numbers, you are just about balancing power in vs power about for your system...

    Given the morning/evening shading--it sounds like your system is producing what it should be from the solar panels/charge controller... Don't see any indication that something is broken.

    As you know, check for phantom loads, turn off anything not needed, perhaps run the generator in the morning to bump up the battery by 1-2kWhrs, and look at more solar panels/more power from somewhere.

    -Bill

    PS: I should add you should look at your power for the other ~6 months of the year when you will be getting 25% less power (or even less with shading issues) from your system... More generator+fuel, more panels, and/or less load will be needed for the "darker" seasons.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Not enough power over night

    I know that smoke detectors are required to be wired in series so that when one goes they all go. Used to drive me crazy that I had to run a 3 wire all over the house (when I was building) just to power the smokies! Somebody somewhere MUST make a low voltage series smokies that you can run off either a battery bank, or a large rechargeble portable battery.

    As Bill says, watch out for the little phantoms out there, 5-10 watts over the course of a day is huge. Consider putting your gfi's on switches.

    Good luck,

    Tony
  • James Wilson
    James Wilson Registered Users Posts: 28 ✭✭
    Re: Not enough power over night

    There are GFI breakers in the actual breaker box these days. I can turn them off, but with all of the other maintenance to watch out for the system ends up owning me!

    All of these little loads are included in what the killowatt meter is telling me - it's between the inverter and house breaker box right now. Everything else is fairly efficient so we shouldn't be having problems.

    The killowatt meter says that we have used 2.3 KW hours in the last 12 hours. Adjusted up 20% for battery losses that is 2.76 KWh then another 15% for the inverter losses then we've already used 3.174 KWh!

    I still need to measure the specific gravity in all of the cells to see if I need to do an equalize cycle yet. I usually do one once a month, but Surrette recommends once every six months or as needed...

    Based on all of the recommendations and my observations here is what I am going to do:

    First of all a, a second refrigerator and washer that I've been hosting on my system is going away soon, that should help with the load!

    I'm going to take four of the solar panels and put them on the roof facing west. They will get direct sun a little later in the day, but they will still be getting sun about three to four hours later. This probably won't change the total amount of power made each day, but it will stretch the hours that power is actually being made. The shape of this valley makes normal solar panel siting more complex.

    I'm going to set up the Water Baby hydro so that the water flow will turn on and off automatically so that when enough water has built up in the little dam it will start running. It will be set up so that it's valve only opens up at night and when there is enough water for it to make power. Depending on the nozzles I put on it it can make between 36 and 175 watts. I'm also going to move it so that it is about 5 feet lower Now if it would only rain! At night. :)

    We had another 8 KW hour day! :) (5.2 KWh after losses)


    Thanks for all of the help!
  • Brock
    Brock Solar Expert Posts: 639 ✭✭✭✭
    Re: Not enough power over night

    For the heck of it can you check each 6 volt battery and take a reading. It sounds like you're doing SG, but as a backup checking each battery might tell you something as well. Since they are in series and in parallel at the ends they all might be just a tiny bit off, but if one is way off it will stick out. When our last bank went it took just one battery to mess the whole thing up.
    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