System Design

moorsb
moorsb Solar Expert Posts: 38
I am putting in a system with 26 GE 66 watt solar panels. This will give me 1716 watts of 18 volt power. I am trying to figure out how big my battery bank should be. I have 14 Trojan T 105 battery The 1716 watt from solar will give me 95 amps to charge the battery bank. T105 12 volt bank using 5 pairs will give the system 1125 a/hr of reserve capacity. Is this bank too large given the 95 amps to charge them?

My system should be able to power 1 kw of load during the daylight and charge the battery, if I only let the battery bank goto 70% I should be able to run 400 watts for 5 hours at night.

I am I on target with my design?

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: System Design

    We really need to know your loads in terms of 1,000 watts * X hours per day + 400 watts * 5 hours per day, etc...

    Roughly, for a battery bank that is used for daily off grid power--Look at 3 days of no sun + 50% maximum discharge... Basically, this is 6x your daily average load.

    If this was an emergency system--you could go to 1x daily load and 50% maximum discharge or ~2x your daily load for the smallest battery bank.

    We like to size the Solar panels for both your load and your battery bank (i.e, small loads and huge battery bank will need more solar panels for proper charging).

    Also, if you have special needs (little power most of the day, and a huge well pump--that forces the battery bank and inverter to be larger to support the surge loads). If much of your load is during sunlight hours (say irrigation pumping)--You could get away with a smaller battery bank (solar panels support the loads while sun is shining).

    For measuring typical appliance loads, the Kill-a-Watt meter is a great start. You plug your appliance into the KaW for 24 hours and get the kWatt*Hours per day use (some appliances, like refrigerators/freezers will use a lot more power in hot weather)...

    To measure 240 VAC and well pump loads is a bit more difficult/expensive.

    -Bill

    PS: Forgot to add--Conservation, conservation, conservation... It is almost always less expensive to conserve a kWH than to generate a kWH (new energy star appliances, smaller laptop computer vs big desk top computer, turning off loads, CFL vs filament lighting, etc.).
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: System Design
    moorsb wrote: »
    I am putting in a system with 26 GE 66 watt solar panels. This will give me 1716 watts of 18 volt power. I am trying to figure out how big my battery bank should be. I have 14 Trojan T 105 battery The 1716 watt from solar will give me 95 amps to charge the battery bank. T105 12 volt bank using 5 pairs will give the system 1125 a/hr of reserve capacity. Is this bank too large given the 95 amps to charge them?

    My system should be able to power 1 kw of load during the daylight and charge the battery, if I only let the battery bank goto 70% I should be able to run 400 watts for 5 hours at night.

    I am I on target with my design?

    *grumble* There must be some site out there telling people to do things backwards, because it sure comes up often!

    First off: what Bill said. Get a handle on your loads. off-grid solar needs to be designed around that.
    Next, I hope you got those panels cheaply. Small (under 100 Watt) panels tend to be a poor value in terms of $ per Watt. That said, what you'll actually get is more like this:
    1716 Watts @ 80% average = 1372. Daily 'harvest' about 5.5 kW hours (best case).
    Now here's where you run into trouble: charging a 12 Volt battery bank. That array would be able to supply 96 Amps of current peak @ 14.2 Volts charging. There is no single charge controller that can handle that. On the bright side, it could charge a huge battery bank: 1000 Amp hours wouldn't be out of the question. But you definitely do not want to parallel up 5 pairs of batteries; you run into trouble keeping the current even going in and out of all pairs.

    Since you have some existing equipment (including batteries), consider going to a higher system Voltage instead. You could go 24, using three strings of 4 batteries (12 total) for example. Although ideally you'd use two strings of four: 225 Amp hours * 2 = 450 (giving you 5 kW maximum usable power - about the same as the panels could harvest) and the panels would be able to recharge these at Trojan recommended 10% (C/10) rate.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: System Design
    moorsb wrote: »
    I am putting in a system with 26 GE 66 watt solar panels. This will give me 1716 watts of 18 volt power. I am trying to figure out how big my battery bank should be.


    Backwards !

    Size the battery bank for the load, no more than 20% discharge at night, then size PV array to recharge in 1 sunny winter day. (3 hours or less, depending on your location.)
    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 ,

  • moorsb
    moorsb Solar Expert Posts: 38
    Re: System Design

    I know I am doing this backward but If I do it the other way I will never get it done as it is too expensive. I got the GE panels for $122 each or $1.84/ watt I got the smaller ones thinking if it hails there might be less damage. The battery I got for $55.00 each as these are factory 2nd. I am just trying to figure out the best configuration. What is realistic given those 2 items are now fixed. I was going to put 2 60amp mppt chargers on the bank. We get nearly 6.5 hour of sunlight per day according to the maps I've seen.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: System Design

    are the batteries 6V or 12V ?

    I'd wire them in series, and go for a 48V system, with a pure sine inverter.
    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 ,

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: System Design
    moorsb wrote: »
    I know I am doing this backward but If I do it the other way I will never get it done as it is too expensive. I got the GE panels for $122 each or $1.84/ watt I got the smaller ones thinking if it hails there might be less damage. The battery I got for $55.00 each as these are factory 2nd. I am just trying to figure out the best configuration. What is realistic given those 2 items are now fixed. I was going to put 2 60amp mppt chargers on the bank. We get nearly 6.5 hour of sunlight per day according to the maps I've seen.

    If you up the system Voltage you may only need one MPPT controller. You can also wire some of the panels in series and then the series strings in parallel to increase the array Voltage. This will reduce line loss on the down lead to the controller and lower the input current (as opposed to all panels in parallel, which would require one fuse per panel as opposed to one fuse per string).

    Don't count on 6.5 hours of sunlight either. If you want accurate data, run the PV Watts program with input for your location and you will see what you can expect.

    It would definitely be advantageous for you to go up to either 24 or even 48 Volts.
  • azrc
    azrc Solar Expert Posts: 43
    Re: System Design

    New at this, but how is this?

    3 parallel strings of PV -> 54v
    Then buy 2 more batteries and have 2-8 series of 6v batteries for 48v.
    2 panels unused.


    Panels will generate ~3hr*24*66w ~= 5kw energy/day. Run 3 parallel strings @54v

    Buy 2 more batteries to allow for 2 parallel strings at 48v.
    Each series battery string is 48v*225Ah ~= 10Kwhr x 2 (parallel strings) = 20Kwhr x 20% (discharge usage) ~=4kwhr storage

    tldr; buy 2 more batteries and keep 2 panels as spares.
  • azrc
    azrc Solar Expert Posts: 43
    Re: System Design
    moorsb wrote: »
    I was going to put 2 60amp mppt chargers on the bank.

    Speaking from inexperience, questioning actually, but if each string in a parallel battery configuration is charged at C/15, and C=225, and you had 2 strings, wouldn't the maximum that you can charge at be 2x(225/15) = 30A, so only 1 mppt @ 40A be sufficient if you had only 2 strings?
  • dwh
    dwh Solar Expert Posts: 1,341 ✭✭✭
    Re: System Design
    azrc wrote: »
    New at this, but how is this?

    3 parallel strings of PV -> 54v
    Then buy 2 more batteries and have 2-8 series of 6v batteries for 48v.
    2 panels unused.


    Panels will generate ~3hr*24*66w ~= 5kw energy/day. Run 3 parallel strings @54v

    Buy 2 more batteries to allow for 2 parallel strings at 48v.
    Each series battery string is 48v*225Ah ~= 10Kwhr x 2 (parallel strings) = 20Kwhr x 20% (discharge usage) ~=4kwhr storage

    tldr; buy 2 more batteries and keep 2 panels as spares.


    Isn't 54v too low to charge a 48v bank? I thought bulk for a 48v bank was up around 58v or 59v.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: System Design

    A 60 Amp MPPT charge controller can handle up to 60 Amps output at 12, 24, or 48 Volts (actually at the charging Voltage levels). In terms of Watts that's roughly:

    60 * 14.2 ("12V") = 852 Watts (about 1000 Watt array at typical efficiency)
    60 * 28.4 ("24V") = 1704 Watts (about 2000 Watt array)
    60 * 56.8 ("48V") = 3408 Watts (about 4000 Watt array)

    Usual disclaimer: calculations for demonstration purposes only; to show the relationship between rising system Voltage and array size utilizing the same charge controller.

    Also with an MPPT controller you do not have to limit array Voltage to nominal system Voltage. The Morningstar will take up to 150 VDC and down-convert for any system. As long as there is no danger of running over-Voltage (as in very cold temps causing super-conductivity in the PV's) there's no problem.
  • azrc
    azrc Solar Expert Posts: 43
    Re: System Design
    dwh wrote: »
    Isn't 54v too low to charge a 48v bank? I thought bulk for a 48v bank was up around 58v or 59v.

    No idea. Just go to 6 strings of 4 maybe.

    edit: how do typical MPPTs work? Are they purely a PWM current transfer, or is there some sort of buck/level conversion going on also? (To allow charging of a high voltage battery from a low voltage panel)
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: System Design
    azrc wrote: »
    No idea. Just go to 6 strings of 4 maybe.

    edit: how do typical MPPTs work? Are they purely a PWM current transfer, or is there some sort of buck/level conversion going on also? (To allow charging of a high voltage battery from a low voltage panel)

    See previous post re: relative battery charging Voltages. :D

    MPPT are buck converters; they have little microcomputers in them that follow a programmed algorithm to adjust output according to battery need and available input power. Far removed from the PWM controllers which simply put all they can to the battery (basically a straight-through connection) until it reaches a predetermined Voltage level and then pulse on/off to maintain that level.
  • azrc
    azrc Solar Expert Posts: 43
    Re: System Design
    See previous post re: relative battery charging Voltages. :D

    MPPT are buck converters...

    Thanks.

    Do they go both ways, allowing charging from low to high (in which case topology isn't that big a deal other than worrying about max efficiency points and construction/wiring cost considerations)?
  • dwh
    dwh Solar Expert Posts: 1,341 ✭✭✭
    Re: System Design
    azrc wrote: »
    Thanks.

    Do they go both ways, allowing charging from low to high (in which case topology isn't that big a deal other than worrying about max efficiency points and construction/wiring cost considerations)?

    No, they only down convert. So to bulk charge a battery with 58v, you'd need somewhat more than 60v on the input side.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: System Design
    azrc wrote: »
    Thanks.

    Do they go both ways, allowing charging from low to high (in which case topology isn't that big a deal other than worrying about max efficiency points and construction/wiring cost considerations)?

    No. They only "down convert". There wouldn't be much point in "up converting", as you might gain Voltage but there wouldn't be enough power to effectively put any current into a battery.
    Also remember than PV's are current sources; they shoot up to Voc with the first sign of light, but don't get down to producing serious power until they get some direct sun. Couple that with the actual Vmp of a "12V panel" (about 17.5) and you see there is no need if you plan it right.
    They will enable charging under lower light conditions than a PWM will. In basic form, if you have 2 12V panels in parallel but there's not enough light to push the Vmp above 12V it won't work. Wire them in series and the MPPT can take "10 Volts * 2" and make some current with it, whereas in parallel you'd get zero, even though there is power there.
    Okay, that explanation sucked! It must be nap time! :p

    Roughly: 10 V @ 2 Amps puts zero into a 12V battery but 20 V @ 1 Amp can put 1 Amp in, and one is better than zero.

    On the down side, MPPT controllers consume more energy than PWM - because there's more circuitry to use power.
  • azrc
    azrc Solar Expert Posts: 43
    Re: System Design
    Roughly: 10 V @ 2 Amps puts zero into a 12V battery but 20 V @ 1 Amp can put 1 Amp in, and one is better than zero.

    You can always suck power out by raising the voltage. 10v@2A=20w available power, so you'd end up only putting 1.67A in. In an ideal world though (no efficiency loss) :) Probably not worth it, but possible.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: System Design
    azrc wrote: »
    You can always suck power out by raising the voltage. 10v@2A=20w available power, so you'd end up only putting 1.67A in. In an ideal world though (no efficiency loss) :) Probably not worth it, but possible.

    Strangely, up-converting DC Voltage tends to use more power than down-converting.

    Somewhere on the forum there's a thread about "panel equalizers" or some such name that supposedly raise the Vmp of shaded panels to match that of unshaded ones in an array. The notion is that you get more over-all power if all the Vmp's are alike, rather than if one is slightly lower. The "raised Voltage" panel drops in current, of course, but has roughly equivalent Wattage it would have at the lower or "natural" Voltage. In essence, it becomes a "lower-Wattage-at-the-same-Vmp" panel in the array.

    You may not find a lot of people using such technology. This is probably because the cost of buying one each for any given array doesn't realize as much over-all benefit as simply spending the money on more panel.

    For any more detailed explanation of the phenomenon we'll have to go to the engineers!:D
  • azrc
    azrc Solar Expert Posts: 43
    Re: System Design
    For any more detailed explanation of the phenomenon we'll have to go to the engineers!:D


    I used to be one of those (for power systems in process control not for solar) so I think I understand the tech behind everything (yeah, right), just not what is available out there on the market. ;)
  • Kamala
    Kamala Solar Expert Posts: 452 ✭✭
    Re: System Design

    Maybe think of energy in terms of potential (volts) and kinetic (amperes.)

    Also, consider the 2nd Law of Thermodynamics. I always have trouble wrapping my mind around it's meaning. But, as I understand, physical systems degrade from order to chaos by nature of the universe.

    Things do not spontaneously heat up. They do cool down so.

    Consider a cut log of wood, which represents potential. It can be burned to release some kinetics (heat and light.) But it cannot be rebuilt to hold more potential.

    All energy transfers/changes are inefficient. The inefficiencies are less going down the potential. Going up is probably a net loss.

    :cry: I must stop. Hail storms come....
  • azrc
    azrc Solar Expert Posts: 43
    Re: System Design
    Kamala wrote: »
    Maybe think of energy in terms of potential (volts) and kinetic (amperes.)

    Also, consider the 2nd Law of Thermodynamics. I always have trouble wrapping my mind around it's meaning. But, as I understand, physical systems degrade from order to chaos by nature of the universe.

    Things do not spontaneously heat up. They do cool down so.

    Consider a cut log of wood, which represents potential. It can be burned to release some kinetics (heat and light.) But it cannot be rebuilt to hold more potential.

    All energy transfers/changes are inefficient. The inefficiencies are less going down the potential. Going up is probably a net loss.

    Inefficiency in going up (or down) is due to inefficiencies in the parts and design. Eout always has to = Ein, so VinIin = VouIout, Vout = Vin*(Iin/Iout).

    Losses are incurred largely in the magnetics involved (design and quality). You can get up to around 95% without too much difficulty. uP is negligible. Just remember Energy_in = Energy_out + heating. Inverters boost voltage levels through the same process (voltage = delta I / delta t)...you can make voltage as arbitrarily as large as you want (limited by the physics of your parts) by maximizing current and minimizing time.