mixed panel question

I've been running two arrays into two charge controllers at 24 volts into one battery bank.
These are have been located on two different buildings.

One has Matrix and Evergreens at Vp 17.4, Voc 21.7, totaling about 1000 watts, going into an Outback mx60.

The other is a "democracy rack" mostly of old SMUDs, quads and tris at Vp 17, Voc 20, totaling about 500 watts.
I did have 8 Siemans M75's, Vp 15.9, Voc 19.8, 400 watts hooked up with these SMUD's. Total of this array was 900 watts
This array goes into a Solar Boost 50 MPPT.

I am changing my 2nd location and moving these near the first array. I can hook the Sieman's up to either array.

Which would be the most efficient.

Thanks in advance. What a wonderful forum.

Charles

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: mixed panel question

    Roughly, the maximum panel wattage your MX60 can econnomically put into the 24 volt battery bank:
    • 60 amps * 29 volts battery charging * 0.77 system derating = 2,260 watts or so maximum
    Next, looking at the Outback array, what is the series/parallel combination of solar panels are you currently running?

    Roughly, you can connect in parallel any panel (or panel string) that is within ~10% Vmp of each other (i.e., 20 volts and 22 volts--just made up numbers to illistrate the rule). And for series connections, again Imp within ~10% of each other.

    It is not to say that you cannot connection panels with specs. >>10% missmatch--but it can cause loss of array output and/or confuse the MPPT charge controller (i.e., putting Vmp=24 volt panels in parallel with Vmp=16 volt panels).

    In some cases, assuming you are in a hot climate, your 15.9 volt panels, for example, may not properly equalize a 12 volt battery bank on a hot day (equalizing a 12 volt bank requires ~15 volts, and Vmp can fall on hot days below 15 volts).

    For the Outback, you could put two x 15.9/17.x panels in series and just meet the nominal voltage requirement for charging your 24 volt battery bank... Or you could put 3 to around 5 or so panels in series and operate the cabling at less current, and give lots of head room for battery charging and wiring drops.

    The other thing for safety, is each of these parallel connected strings should have a series protection fuse/breaker/combiner box to prevent any short in any single panel/wiring from getting fed current from the rest of the panels and possibly starting a fire.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: mixed panel question

    "Roughly, the maximum panel wattage your MX60 can econnomically put into the 24 volt battery bank:
    • 60 amps * 29 volts battery charging * 0.77 system derating = 2,260 watts or so maximum"
    that's not how to rate how much pv you can feed into an mx60 that i've ever seen. 24v x 60a = 1440w. nec likes it at 80% of that for 24v x 48a = 1152w. derating for ptc is ok in theory, but it is only about 10% and you should have some headroom anyway to allow for the mppt to do its thing. nec does not derate to ptc or account for wire losses in their calculations to the controller. luckily, the mx60 has a software mod for 70a. 70a x 24v = 1680w and at 80% for nec (which can allow for mppt gains too) 56a x 24v = 1344w.

    the m75s have a low vmp and may go too low under hot conditions for good mppt action and should probably be handled differently, like possibly series/parallel.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: mixed panel question

    I thought the 1/1.25 derating of the 60 amps was the solar panel Vinput current to 48 amps (which I disagree with anyway)... I have not understood that they do an NEC derating on the 60 amp output (which Outback lets the owner program to ~72 amps on the MX 60 models--as I recall).

    The 0.77 derating I suggested above was the rough "optimum" amount of solar panels (in watts) to put on the controller that would, for the most part, result in ~60 amp output in the middle of the day (average sun, average warm panels, average dust in air, average dust on panels, average losses in charge controller, etc.).

    It was not intended to be a discussion of wire sizes and NEC current derating thereof.

    The average MPPT controller will control its output current (and voltage) to meet its specifications (and Listing requirements--if any). So, there should be no additional input (or output) deratings required. The controller is designed for 60 amp input--which means that the input wiring (and fusing, if any) would be designed for 60a*1.25=75amps (or round up to 80 amps). To now reduce the input current to STC rating of 48 amps seems to be an excessive safety factor--But can be "worked around" by running Vmp-array>>Vbatt-charging voltage--Then Iarray current can never exceed 48 amps, but you can still have your full wattage of panels (MPPT controller doing the voltage down converting).

    The double derating on the Solar Array Input Current by NEC would, sort of, make sense for PWM controllers where the current does go above STC ratings in a few situations (very hot weather, lots of sun, and reflections). But the change in current due to hot weather is not very much, so I think the extra 1.25 derating seems to be excessive to me.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset