MPPT in parallel mode - any good?

We are having a grid-tied system installed, and the first powerup test was done recently. We ended up with a PowerOne inverter that has independent MPPT 2 channels. After the installers left, I was reviewing my notes. It appears that the installers wired the panel in "Parallel" MPPT mode rather than "Independent" mode. From the wiring it looks like the + and - from the panels are just connected in parallel inside the inverter, and then both MPPT inputs are wired in parallel as well.

We have some mid afternoon shading issues where a shadow from the neighbor's peaked roof creeps up our roof and begins covering the panels. The PV company knew about this situation, however our project manager changed midstream so I suspect the info was lost? I've put in a call to them and will make sure they address this.

In any case, my question is more of a theoretical one: assume we have 2 stings of panels, and one string has one panel partially or completely shaded.

What would be the difference in production if these were wired into two independent MPPT inputs, vs if they were paralleled? My (limited) understanding is that if one panel in the string is completely shaded, the whole string would be producing zero, at which point it doesn't matter how it's wired, but that it would matter for the time when one panel is only partly shaded?

NB I'm not sure how they wired the two strings - there are two rows of panels, and the original plan was to do a donut arrangement so the center group of panels was on a different circuit than the outer group.

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: MPPT in parallel mode - any good?

    The reason for the parallel MPPT inputs is to separate the total array into two halves. Why would you do this? In case one gets shaded at some time during the day or some panels having to face in quite different directions.

    Sounds like you should get in touch with the installer and talk to them until you are satisfied the job is done right.

    And no, boys and girls, we are not going to go through the whole "it doesn't make any difference" argument again.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: MPPT in parallel mode - any good?
    The reason for the parallel MPPT inputs is to separate the total array into two halves. Why would you do this? In case one gets shaded at some time during the day or some panels having to face in quite different directions.

    Sounds like you should get in touch with the installer and talk to them until you are satisfied the job is done right.

    And no, boys and girls, we are not going to go through the whole "it doesn't make any difference" argument again.

    I expect (well hope) that no one is going to try to dispute that it does make a difference when differences in shading are involved.

    And, because of the existence of bypass diodes, the partially shaded string will NOT be producing zero, it just will be producing the same current as it would without the shading, but at a significantly lower voltage. That is where the mismatch occurs, and if the MPPT controller does not scan low enough in voltage it will not even realize that the second string exists. And there will be power lost because one MPPT setting cannot match both strings. The size of that power loss will depend on the details of the voltage difference.
    One shaded panel out of a string of 10 will be less of a problem than one shaded panel out of a string of 3!

    Your installer certainly should fix this, since the inverter has the capability to handle it.
    SMA SB 3000, old BP panels.
  • soylentgreen
    soylentgreen Solar Expert Posts: 111 ✭✭
    Re: MPPT in parallel mode - any good?

    Update : the PV installer confirmed that they had wired it in parallel mode, but said that they sometimes do that for initial testing / qualification of the system. They've switched it over to dual MPPT mode and now it clearly shows two different circuits, each generating almost the exact same volts/amps/wattage when there's no shading. So this is good.

    But there's a new mystery: the inverter has a display for "lifetime max wattage" and "daily max wattage". The lifetime max is about 4700watts, which was set back in late december when the system was wired in parallel mode. I've watched over the last few days, and the "daily max wattag" is running about 3300 watts. This is on a 5.2KW DC nameplate system.

    What's weird is this: if I believe the #s, the lifetime max wattage was set in late december (just around the solstice) and is about 50% higher than the daily max wattage now, when the system is running in dual MPPT mode. This seems backwards - the insolation would have been lower (as it was the solstice) and one would think that running in dual MPPT mode it would eek out a few more watts.

    Note: overall i've compared the daily output when running in MPPT dual mode, to PVWatts 1 estimates and it's pretty much spot on, so my hunch is that the 4700W number may be bogs. After all, 4700W in a 5200W system would be 90% of max nameplate production : would this even be possible in San Diego on the solstice?

    I've thought of possible explanations:
    * true value momentarily: the 4700w max value was a momentary spike
    * true value : the panels were brand new and produced a burst of energy but now they have "broken in" and are now acting normally (implausible)
    * the panels are dirty (possible, but unlikely - it's been less than a month)
    * metering error: the meter just doesn't work right
    * some weird scenario where running in parallel mode actually generates more watts than 2 separate MPPT circuits (possible)?
    * air temperature differences - the 4700W record was set on a really cold day


    Thoughts?
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: MPPT in parallel mode - any good?

    Edge of cloud event?

    Tony
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: MPPT in parallel mode - any good?

    Crystalline solar panel may lose about 3% in the first few months/year... Thin film amorphous panels may lose 10-20%.

    Very cold days, and some types of cloud conditions can "focus" a significant amount of extra sun--30% extra? Don't know.

    I think the MPPT charge controllers are more variable than we tend to think they are. I had a new model Xantrex GT 3.3 kW inverter that showed >3 to 3.3 kW many times. It failed and was replaced with a new/old version (5 years older) and that one seems to never go above ~2.5 kW... Don't know--And at this point, I am not going to care too much as this is the third inverter, long out of warranty (may try SMA next time), and my bill is still a net "credit" vs my power usage (oversized array vs my needs--had always planned on some sort of e-car some day).

    Are the differences in hardware/software/calibration/array decaying over time (my first array was replaced after about ~5 years when its output fell to ~1,500 peak)? Don't know and I have no way of figuring it out without lots of monitoring and such.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: MPPT in parallel mode - any good?

    I wouldn't worry about any maximums for just the reasons Tony and Bill mention. It's the consistent average output that matters.
  • soylentgreen
    soylentgreen Solar Expert Posts: 111 ✭✭
    Re: MPPT in parallel mode - any good?

    I didn't know about the cloud edge effects, but I like that theory and it seems to fit the data : On the day that the record 4.7kw was set, it was bright + clear + cold with occasional compact clouds going by.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: MPPT in parallel mode - any good?

    Well cold temps will increase a panels output Voltage which the MPPT function will convert to additional current, and edge-of-cloud events focus more light on panels which increases current output. Either or both will give you a temporary blast of extra power. But of course it's not something you can count on.
  • jschner
    jschner Registered Users Posts: 22 ✭✭
    Re: MPPT in parallel mode - any good?

    When my Aurora PVI-5000 was installed and powered up for the first time on Dec. 31st, it had a maximum Peak around 4750Watts already in it even though the peak for the day was 3600 watts. It was a new unit right out of the box as I helped unpack it. I'm thinking the maximum 4750W peak came from the manufacturer that way, maybe from QC testing? The 4750W maximum was exceeded on Jan 11th due to cloud edge effect at 5055Watts. http://pvoutput.org/intraday.jsp?id=15767&sid=13640&dt=20130111

    So is this SG from TT?
  • soylentgreen
    soylentgreen Solar Expert Posts: 111 ✭✭
    Re: MPPT in parallel mode - any good?

    That's an even better explanation : "it was set at the factory" :-) Mine is 4790W to be exact, though I believe i have an aurora 6000 so you'd think they woul d have tested it closer to 6KW?
  • jschner
    jschner Registered Users Posts: 22 ✭✭
    Re: MPPT in parallel mode - any good?
    That's an even better explanation : "it was set at the factory" :-) Mine is 4790W to be exact, though I believe i have an aurora 6000 so you'd think they woul d have tested it closer to 6KW?

    Just a guess on my part but the number being left in there from pre-testing seems logical. Should be a sticker on the side of the inverter that can tell you the model.