Rail1 on left, then Rail2 and Rail3
Two panels hooked to one Enphase M215 via splitters; the west panel is propped up to allow photo.
Small black plastic breaker box that has one breaker for each rail
White stuff around breaker box is Emporia monitoring system. That is supplying the data below.
White 14ga romex will feed AC to the far end of string of drop cable.
19 drops is a little beyond the limit of supplying AC from one end only.
All three rails with the building that supplies some morning shade to Rail1
I intend to post some daily energy harvesting data here on three rails, each with 19E panels and 19W panels. Rail2 and Rail3 each have 38 microinverters, one for each panel. Rail1 has been converted so 19 microinverters each serve two panels. Rail1 gets some morning shade from a nearby building and some evening shade from Rail2. Rail2 gets some morning shade from Rail1 and evening shade from Rail3. Rail3 gets only morning shade from Rail2 and none from the west.
Cost of the 1 panel 1 inverter units is about $110/panel. Cost of the 2 panel 1 inverter unit is about $79/panel. After the conclusion of this test I expect to start converting Rail2 and Rail3 to 19 inverters each. These conversions free up inverters and cabling for new installations.
Saturday Sept 5 was a mostly cloudy day with overall energy production about 60% of a good sunny day.
Rail1 21.88 kwh
Rail2 19.93 kwh
Rail3 24.32 kwh
Sunday Sept 6. More sun than yesterday. "Partly cloudy". Maybe 80% of a good sunny day
Rail1 24.0 kwh
Rail2 22.66 kwh
Rail3 27.37 kwh
Monday Sept 7. Pretty good sun! Maybe 90-95% of an all sun day.
Rail1 30.37 kwh
Rail2 29.27 kwh
Rail3 34.52 kwh
Tuesday Sept 8. Fully overcast all day. About 50% of a good day.
Rail1 19 .92 kwh
Rail2 16.71 kwh
Rail3 22.03 kwh
I'll likely call an end to the test after Tuesday's data. Tuesday looks like cloudy all day. The results are clear: The energy lost to a pair of E/W panels sharing an inverter is not significant; especially when ~40% hardware savings are considered. One second power sampling does reveal that energy is lost when both panels get good sun. For that reason, I believe the scheme will work best when panel slopes are steep.
Well.... one more day Wednesday Sept 9 was cloudy and rainy all day. Maybe 30% of good day.
Rail1 11.18
Rail2 8.45
Rail3 12.55
I don't believe shading is the cause of low production on Rail2; I suspect one or more non functioning panels though I have no easy way to confirm. The production on the converted Rail1 is about 10% lower than that of Rail3; that meets my expectations.
Here is a post after the 3rd rail was put in production about a year ago:
https://wmckemie.blogspot.com/2019/09/3rd-twofaced-rail.html
And another post dealing with the beginning of the 3 rail project:
https://wmckemie.blogspot.com/2019/07/first-new-project-on-across-road-meter.html
I put in the first rail, the one recently converted, not fully intending to add rails 2 and 3.
2-3 years later. I was having good success sharing several varieties of single panel inverters. At the beginning, Enphase M215s were attractively priced at about $35-$40 for the inverters plus around $15 for each segment of the drop cable needed to connect inverters to the 240vac supply.
ReplyDeleteMore recently, I switched to two panel chinese inverters; each inverter serving four panels. I installed 6 rails at my new "hill top" transformer. I am now going back to try to diagnose and correct the problems. I fear the chinese inverters do not work as well sharing as the older single panel inverters. I may have to reconfigure those six rails.