> The 17.2 value is when the panels are at 25C ... on the the roof they are
> much hotter and the 17.2 falls 0.45% with every "C" above 25C the panels
> get

Cool. Did not know that. Now these puppies are roof mounted, about fifteen degrees above pitch, natual "silver" colored aluminum roofing.
I do remember they were too hot to touch when I was bolting down the clips for the uni-rack; had to wear gloves. That would put them in the 45-50 degrees C range, I'd guess.

> Where is this installation located ?

St.Lawrence Island Provincial Park, in Thousand Islands, Ontario, a ParksCanada research station. Right on the St.Lawrence Seaway, across from upstate NY, I think, at that point.

> I'm not sure what "running fine" means ... most folks wouldn't know if
> the system is sub-par performance ... just go to th site mid-afternoon on
> a hot day and see what the array voltage reported by the inverter is , if
> it with a few volts of 150 , theres performance being lost

Interesting. By "running fine" I mean the engineer flipped the breakers, and apparently the appropriate amount of amerage was moving through the "donut" (I called it an ampmeter, the engineer called it something else).

> If I was the "engineering outfit" I wouldn't go lower than 12 panels in a
> string with the sharps ...

Well, like I said, I was just there to be lead monkey . True enough, though, I also like to run equipment in the mid-range, and trade standard is at least 10% over minimum and less than 90% of max.

I will look into this, though, if for no other reason than education.

DJ

That explains it ... They are more concerned about the effects of cold , vs hot ... But Runnning the numbers , even for -20C ...

VOC 22 volts * (45C (delta of +25C to -20C) * 0.45%) = @26.5 volts , * 11 panels = 292 volts ... still well below the 450 VDC input

Sorry , I would have to say they didn't run the numbers on this one

> That explains it ... They are more concerned about the effects of cold , vs > hot ... But Runnning the numbers , even for -20C ...

> VOC 22 volts * (45C (delta of +25C to -20C) * 0.45%) = @26.5 volts , *
> 11 panels = 292 volts ... still well below the 450 VDC input

Ok, so they couldn't have used just one Fronius IG, then. At least they got that right . And actually, hitting -35 C wouldn't be impossible in that location, right on the river, for a few weeks in the dead of winter.

> Sorry , I would have to say they didn't run the numbers on this one

All evidence does indeed point to that conclusion, and I do recognize the logic .

DJ

Even at -35C panel temps , the string would only be at 308 VDC ...

Xantrex and SMA both have online string sizing programs that make this sizing easy ...

But , I think they wll be fine how it is , it would have been 100% at the edges with 12 panels , but since I doubt the have allot of hot weather 11 in series should be OK

> Even at -35C panel temps , the string would only be at 308 VDC ...

> Xantrex and SMA both have online string sizing programs that make this
> sizing easy ...

I'll have to keep this in mind if I ever design something that big myself.

> But , I think they wll be fine how it is , it would have been 100% at the
> edges with 12 panels , but since I doubt the have allot of hot weather 11 in
> series should be OK

Oh, Gup, don't believe all the stories about the Frozen North .

They'll see a month or more of mid-nineties this summer, during, I might add, the main supposed production time for these panels .

DJ

Fronius also has a sizing tool on line.

http://www.fronius.com/worldwide/usa.so ... /tool.html

> Fronius also has a sizing tool on line.

>http://www.fronius.com/worldwide/usa.solarelectronics/service_support/tool.html

Well! Fronius liked that scenario of eleven panels, but only if we ran two strings in parallel for the 15ish amps, which we did. But it was definitely a low end system, and as Gup suggested, 12 in a string would have been a better choice.

Cool.

DJ