Nobody told me this about solar panel shading

[caesarv]

I have 2 strings of 6 panels each (Longi LR5; no optimizers or micro-inverters.) One string is in full sun while the other gets partial shade in the morning due to my neighbor's tree. When all are in full sunlight, each string produces about 3kW. However, when that one string is partially shaded over two or more panels, I get less than 100W from the entire string. I had assumed the bypass diodes would just eliminate the power from the shaded panels, leaving the rest intact. Apparently I was wrong.

In full sunshine, I can cover any ONE panel, and sure enough, the wattage drops by one panel's worth (~500W.) This proves the bypass diodes are working. However, when I partially cover two (or more) panels, the entire string drops to near zero. I was not expecting this. This severely impacts my morning solar production.

I am assuming that the MPPT in my 18Kpv can deal with one panel being bypassed, but when two or more are partially shaded, it's algorithm cannot find a suitable setting to deal with the combination. Does this make sense? Would adding optimizers on the two or three partially shaded panels help? Any other ideas?

Meanwhile, I will be trimming my neighbor's tree to minimize the shading (he is okay with this), but this will not completely eliminate the morning shading.

 

[MisterSandals]

However, when that one string is partially shaded over two or more panels, I get less than 100W from the entire string.

By "string" you mean for the particular 6 panels in series, right?

If this string producing 100W is connected in parallel to an identical string producing 3kW, they should be fine and produce 3.1kW together.

If not, can you explain how you have your array (all of your strings) wired and what you have between the array and MPPT?

 

[Brucey]

Depending on the minimum voltage for the mppt trackers, when you cover two panels your pv voltage may be dropping below minimums.

 

[SomethingAppropriate]

Same happens to me . If 3 of my 9 have 1/8 shading each. I'm down 97%
All in series, no where near mppt cutoff

 

[timselectric]

A partially covered panel isn't impaired enough to trigger the bypass diodes.
A fully covered panel (or internal string) is enough to trigger the bypass diodes (or single diode).

 

[EPicTony]

what's a bypass diode ???? Is it built into the panel ? First I have heard of it. I guess I'm dumb.

 

[timselectric]

Is it built into the panel ?

Yes
Typically there are 3.

 

[SomethingAppropriate]

A partially covered panel isn't impaired enough to trigger the bypass diodes.
A fully covered panel (or internal string) is enough to trigger the bypass diodes (or single diode).

Are you saying that if I have a series string of 9 I would gain if I covered all of the partially shaded panels on the same side? For 1 month out of the year I have 3 panels partially shaded all the time and my output goes to 8% in full sun.

 

[timselectric]

Are you saying that if I have a series string of 9 I would gain if I covered all of the partially shaded panels on the same side? For 1 month out of the year I have 3 panels partially shaded all the time and my output goes to 8% in full sun.

Cover the 3 panels completely, to engage the bypass diodes.
This would increase output to that of approximately 6 panels.
But keep in mind that bypass diodes get hot when in continuous use.
This will shorten their lifespan.

 

[SomethingAppropriate]

Cover the 3 panels completely, to engage the bypass diodes.
This would increase output to that of approximately 6 panels.
But keep in mind that bypass diodes get hot when in continuous use.
This will shorten their lifespan.

Thanks! So still better to just disconnect them and have them catch thier 70 watts on a separate controller.

Are all 3 diodes in the connection box on the solar panel? Easy to find and replace?

 

[timselectric]

Thanks! So still better to just disconnect them and have them catch thier 70 watts on a separate controller.

Probably
Or install a bypass (DC rated) switch or relay, for the 3 panels.

 

[caesarv]

By "string" you mean for the particular 6 panels in series, right?

If this string producing 100W is connected in parallel to an identical string producing 3kW, they should be fine and produce 3.1kW together.

If not, can you explain how you have your array (all of your strings) wired and what you have between the array and MPPT?

Yes, I have two strings of 6 panels each. Each string has all 6 panels wired in series for a total voltage of about 260V (under load.) The two strings are each wired to two different MPPT inputs on my 18Kpv inverter; I ran 4 wires to the inverter. There is nothing between the panels and the MPPT inputs other than a switch; i.e. no optimizers. Each panel puts out about 49V (regardless of sun or shade) with no load. When in full sun, the available current is about 12-14A...measured in early January, so, not optimum. The ground mounted panels are rated at 535W but are bifacial, so I do get some additional wattage from reflection of the white rocks on the ground. As typical, each panel has 3 bypass diodes.

 

[timselectric]

Another reason for the bypass diodes to not kick in, is if they have failed (burned).

 

[caesarv]

Thanks! So still better to just disconnect them and have them catch thier 70 watts on a separate controller.

Are all 3 diodes in the connection box on the solar panel? Easy to find and replace?

Older panels usually have a small box at the top with 3 diodes. One could open that box to test and possibly replace the diodes if needed.
Newer panels usually have 3 smaller boxes on the left, middle, and right sides. They are typically encapsulated with some potting compound and would be very difficult to replace. I did open the 3 boxes on a spare panel to check the diodes and all were okay. It was a real pain to get to the diodes.

 

[MisterSandals]

Each panel puts out about 49V (regardless of sun or shade) with no load. When in full sun, the available current is about 12-14A...measured in early January, so, not optimum. The ground mounted panels are rated at 535W

What are the actual specs on the panels? Your numbers don't add up:

49V x 12A = 588W
49V x 14A = 686W

 

[caesarv]

What are the actual specs on the panels? Your numbers don't add up:

49V x 12A = 588W
49V x 14A = 686W

Yes, I realize that. Much depends on method of testing/measuring, and if taking the bifacial advantage. I did not want to get bogged down with too many numbers. Here are the actual specs...but they do not include any bifacial gain.