Can two microinverters be fed from the same solar panel

I have successfully interfaced a single 327W 96 cell panel (Benq Sunforte) with a commercial micro GTI using a custom mosfet source follower design interposed between the panel and the GTI dc input. The panel Voc max is 65v and panel max power occurs at 55v. The problem I have is that the the inverter rated at 250W will only output 226W because of the way it is designed. Can anyone suggest what might happen if two such micro GTI had their dc inputs wired together in parallel so that the full 327W 230v output or somewhere near might be realized? The challenge was to avoid using power optimizers feeding into a constant voltage string GTI something that requires two or more solar panels.

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    The problem is that GT Inverters (and solar charge controllers) that use MPPT (maximum power point tracking) behave as if they are the only device attached to the solar panel/array.

    And an MPPT controller "tests" the solar panel looking to satisfy the equation Power = Voltage * Current or specifically Pmp = Vmp * Imp (mp=maximum power).

    One way of doing this is for the MPPT controller to "sweep" the array. Start at I=0 amps and then I=0.1 amps, 0.2 amps, etc. while measuring the Array Voltage.

    The controller creates a table then does the P=V*I for each entry. And looks for the maximum value of P. At that point, the controller uses Vmp that is finds, and uses that for the next 10 minutes (it varies current to "hold" Vmp-array). When you have two MPPT controllers attached to a single solar array, the two controllers can confuse each other while doing the solar array characterization.

    I am not sure, but I don't believe that modern MPPT devices do the "sweep" method to find Vmp--But I am not sure what they do exactly. But, I believe whatever method they use, the two controllers would still "confuse" each other.

    You could try connecting two MPPT controllers to one solar panel and see what happens.

    However, some of the new GT inverters these days are "transformerless"---And I have no clue if you can even "safely" parallel two GT inverters together on one panel. It is possible that could end up frying everything. Or they will not work because of array grounding sense circuits would confuse each other.

    It just does not sound like a good idea.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • SolarPowered
    SolarPowered Solar Expert Posts: 626 ✭✭✭
    BB. wrote: »
    The problem is that GT Inverters (and solar charge controllers) that use MPPT (maximum power point tracking) behave as if they are the only device attached to the solar panel/array.

    And an MPPT controller "tests" the solar panel looking to satisfy the equation Power = Voltage * Current or specifically Pmp = Vmp * Imp (mp=maximum power).

    One way of doing this is for the MPPT controller to "sweep" the array. Start at I=0 amps and then I=0.1 amps, 0.2 amps, etc. while measuring the Array Voltage.

    The controller creates a table then does the P=V*I for each entry. And looks for the maximum value of P. At that point, the controller uses Vmp that is finds, and uses that for the next 10 minutes (it varies current to "hold" Vmp-array). When you have two MPPT controllers attached to a single solar array, the two controllers can confuse each other while doing the solar array characterization.

    I am not sure, but I don't believe that modern MPPT devices do the "sweep" method to find Vmp--But I am not sure what they do exactly. But, I believe whatever method they use, the two controllers would still "confuse" each other.

    You could try connecting two MPPT controllers to one solar panel and see what happens.

    However, some of the new GT inverters these days are "transformerless"---And I have no clue if you can even "safely" parallel two GT inverters together on one panel. It is possible that could end up frying everything. Or they will not work because of array grounding sense circuits would confuse each other.

    It just does not sound like a good idea.

    -Bill



    In Laymens terms-
    IMP splits in half with 2 micro inverters comined in 1 circuit, the circumstance is its counter opposite, than when to combine 2 panels to double IMP in that circuit.
    Unless its a micro inverter that can sample as low as .0045 IMP, lower light/angle of incidence levels make it harder to harvest power.
    Applying a second inverter will also lessen efficiency. If one inverter (hypothetically) is efficiencent by 94%, and the other inverter is 94%, but niether are meeting the the algorithm (max wattage need for MPPT),
    The efficiency will be worse as an addition problem. If one inverter lost 6% in conversion, and the other lost 6% in conversion, it isn't a 6% loss, its a 12% loss in total watts if it doesn't meet the the requirements for the algorithm. The inverters will only operate at maximum efficiency if the maximum wattage for both inverters are met.
  • SolarPowered
    SolarPowered Solar Expert Posts: 626 ✭✭✭
    I would also like to add, inverters are built and designed on capacitors.
    Capacitors have what is called voltage leakage.
    If maximum wattage in mppt isn't met the capacitors cycle less in a longer duration of time, causing more leakage and causing more losses, also causing capacitors to store that power until it reaches the appropriate discharge rate, the less a capacitor cycles, the hotter the capacitor become. This is the majority where efficiencies are lost. This is also how inverters do not last long, as the capaciotr (elctrolytic, thin film) become the greatest point of failure.
  • Solstice
    Solstice Registered Users Posts: 2
    This is also how inverters do not last long, as the capacitor (electrolytic, thin film) become the greatest point of failure.

    I agree and also agree with BB that connecting two GTI MPPT to the same panel might not be such a good idea. Having looked at the specs for micro GTI, without exception all have their max safe dc operating voltage set at or near 60v. Never understood why this was until I blew one and took the lid off to have a look inside. The power electrolytic capacitors were rated at 63v limiting the dc input voltage. The next voltage up is 100v and caps are physically larger. The inverter dc inputs went short circuit because the MPPT max tracking voltage was 50v and it was forever in hill climbing mode? The inverter was in use for about a year before it failed.

    The source follower black box clips the voltage to 58v and has a yellow LED that comes on when in clipping mode. This was happening quite a lot and the manufacturer suggested that the inverters own protection mechanisms were not catching the over voltage spikes. They also mentioned because the inverter could never reach max MPPT it would never output its rated power.

    The replacement micro GTI has its max MPPT voltage design at 55v and in use things are much more stable the yellow light never comes on. I'm assuming that the panels max MPPT power voltage drops slightly as insolation power levels fall working to favor the GTI. The way I saw things was that if the two inverters were of the same make and model they both would operate to find the max MPPT point in a race to the top.

    Another issue would be that whilst all of the 327W might be put to use, the use of a second 250W inverter would under utilize one or both of them.

    These panels are meant to be used with constant voltage power optimizers such as the Solaredge P500. Because of electrolytic voltage limitations these optimizers cannot use capacitors especially the P500 because its dc MPPT range is 8 to 80v compared with the replacement inverter MPPT range of 22 to 55v. The power optimizer is generating power at 8v compared with 22v making it more efficient at energy production. The panel has a lot going for it: its 20% efficient and PID (potential induced degradation) free. Its lighter and smaller compared with say a 300W Seraphim 72 cell panel.

    The P500 says its max dc output is 60V but would that voltage fall to say 58v under load allowing the use of a micro GTI to generate power from a one panel system?