Tweet
I read an article recently discussing optimizer failure rates, testing of HD Wave efficiencty, etc. Near the end, the author spoke of a flaw with the current SolarEdge system design, and I'm seeing this flaw myself.
For any one string, SolarEdge's claim is that shading on one or a few panels will have no effect on output of other panels. However, this is not completely true. Consider a single string of 8 panels, 4 facing east, 4 facing west. Let's say it's morning, the 4 eastern panels *should* be producing 200 watts each, and the 4 western panels are only producing 15 watts each. (200w and 15w is based on the actual irradiance of the panel)
So, our string *should* producing 200 * 4 + 15 * 4 = 860w, right? Unfortunately, this string won't be producing that. The eastern panels will not be able to produce the 200w each they should be due to the pull-down effect of the 4 shaded panels. The problem is that the optimizers (most of them) max out at 60v output each.
Each of the 4 eastern panel optimizers will attempt to output the full 200w they are receiving, meaning they will boost their voltage to the max 60v. Those four optimizers will then generate 60v * 4 = 240v of the required 390v string. That means the shaded panel optimizers must makeup that difference of (390-240) = 150v, or 37.5v each.
However, at 37.5v output, and with only 15w of input irradiance, they can only supply 0.4 amps (assuming 100% efficiency).
That means the whole string is operating at 390v, 0.4 amps = 156 watts! Only 18% of what the string should be capable of at the moment.
The author provided one suggestion, use the P505 optimizers, as they can output up to 85v max each. My second suggestion is to keep strings together, so they are basically all illuminated at the same time. Use an eastern string, and a western string. Don't split a string across these faces.
In my case, all my panels face south, but I have large trees that shade early morning and late evening. Three of my four strings go straight across the roof, left to right, with one string at the end which is a 3x3 + 1 "mostly" square. I noticed that as the sun is sweeping from left to right across my roof, the leftmost panels on my strings are maxing at 60v output per optimizer. They were doing about 30w or so, whereas my squared string was doing 150w+. I'm losing about 1kw of early morning production for maybe an hour or less.
Had I known this, I would have made all my strings into square grids instead of in long 10 panel rows. I'm not losing enough production for me to want to rewire the strings in this manner, but I'm adding this to my experience box for next time!
For any one string, SolarEdge's claim is that shading on one or a few panels will have no effect on output of other panels. However, this is not completely true. Consider a single string of 8 panels, 4 facing east, 4 facing west. Let's say it's morning, the 4 eastern panels *should* be producing 200 watts each, and the 4 western panels are only producing 15 watts each. (200w and 15w is based on the actual irradiance of the panel)
So, our string *should* producing 200 * 4 + 15 * 4 = 860w, right? Unfortunately, this string won't be producing that. The eastern panels will not be able to produce the 200w each they should be due to the pull-down effect of the 4 shaded panels. The problem is that the optimizers (most of them) max out at 60v output each.
Each of the 4 eastern panel optimizers will attempt to output the full 200w they are receiving, meaning they will boost their voltage to the max 60v. Those four optimizers will then generate 60v * 4 = 240v of the required 390v string. That means the shaded panel optimizers must makeup that difference of (390-240) = 150v, or 37.5v each.
However, at 37.5v output, and with only 15w of input irradiance, they can only supply 0.4 amps (assuming 100% efficiency).
That means the whole string is operating at 390v, 0.4 amps = 156 watts! Only 18% of what the string should be capable of at the moment.
The author provided one suggestion, use the P505 optimizers, as they can output up to 85v max each. My second suggestion is to keep strings together, so they are basically all illuminated at the same time. Use an eastern string, and a western string. Don't split a string across these faces.
In my case, all my panels face south, but I have large trees that shade early morning and late evening. Three of my four strings go straight across the roof, left to right, with one string at the end which is a 3x3 + 1 "mostly" square. I noticed that as the sun is sweeping from left to right across my roof, the leftmost panels on my strings are maxing at 60v output per optimizer. They were doing about 30w or so, whereas my squared string was doing 150w+. I'm losing about 1kw of early morning production for maybe an hour or less.
Had I known this, I would have made all my strings into square grids instead of in long 10 panel rows. I'm not losing enough production for me to want to rewire the strings in this manner, but I'm adding this to my experience box for next time!
Comment