How to wire a single long row of panels that are progressively exposed then shaded?

I'm planning to mount 16 panels in portrait orientation in a long single row under the ridge of my roof. In the morning because of trees to the east, the panels will be exposed to sun progressively from west to east. In the afternoon trees to the west will shade the panels one-by-one from west to east. I'm planning to bring 8 panels to each of two MPPT's, but I'm not experienced enough to know how to connect them so as to maximize my output.

My first thought was to connect each 8-panel group to an MPPT in a single 8S1P string. But because the eight panels will be at first exposed to the sun one at a time and then later blocked by shade one at a time, I wondered if connecting each group of eight as a 4S2P string might be better. I just can't visualize what the power output would do in the face of progressive uncovering and covering of the panels during the day. (Note that while a diagram for this parallel arrangement would normally show the panels in two rows the physical layout would still be a single row of eight for each MPPT.)

I'd appreciate any suggestions from those who have a better idea than I do about how the output would respond in these two configurations as the panels are progressively exposed and shaded.

I’ve done this before along my fence line. As the shade comes across the first set of 4, you’ll still be able to produce power off the second set of 4.

Would a 4S2P still supply enough voltage to your SCC? If it does, and doesn’t exceed the max amperage of the SCC, it should be more beneficial than 8S because the entire string would be affected by the shading.

Keep this in mind. If you have 8 panels in series and one is shaded, you’ll lose most of the potential power of the total.

Skypower said:

Keep this in mind. If you have 8 panels in series and one is shaded, you’ll lose most of the potential power of the total.

Click to expand...

With bypass diodes it is not quite that bad.

For the OP, I would have two strings with each end of the stack grouped together. Not alternated.

You haven't mentioned anything about what the voltage output of each panel is, or what the voltage limit of your controller is? Are these large 60-cell or 72-cell panels? Eight in series might go over 240V or 300V respectively, with the Voc being even higher.

Before reaching the wiring phase, you need to carefully outline what your production limits are. How far away from the panels will the controller be? Even with only 4 panels in series, that might have high enough voltage to negate any effects of voltage drop.

Yes, I didn't mention those other details because I've assumed those aspects would be under control and only wanted to ask about the wiring-configuration issue. I'd probably be going to a Sol-Ark 12K, which has an MPPT range of 150-425V. They'd be 400W+ panels, probably 66 or 72 cells unless half-cut.

My thought is that in the 8S1P arrangement the progressive activation of the bypass diodes as the panels are shaded would keep the string voltage as high as possible and minimize the power loss as panels become shaded compared to a 4S2P arrangement of the same 8 panels. Am I on the right track here?

time2roll said:

With bypass diodes it is not quite that bad.

For the OP, I would have two strings with each end of the stack grouped together. Not alternated.

Click to expand...

The bypass panels are my hope too. See my reply just above.

I don't exactly understand what you meant in your last sentence. Would you mind expanding on that a little?

Sprucebeach said:

My thought is that in the 8S1P arrangement the progressive activation of the bypass diodes as the panels are shaded would keep the string voltage as high as possible and minimize the power loss as panels become shaded compared to a 4S2P arrangement of the same 8 panels. Am I on the right track here?

Click to expand...

Are you planning on using only 8 of your 16 panels now?
8S and 4S2P are both only a panels.

For 16 panels:
8S2P seems reasonable (depending on the Voc of your secret panels)
4S4P also very reasonable but would require high voltage panels (Vmp over 40, the more the better).

MisterSandals said:

Are you planning on using only 8 of your 16 panels now?
8S and 4S2P are both only a panels.

For 16 panels:
8S2P seems reasonable (depending on the Voc of your secret panels)
4S4P also very reasonable but would require high voltage panels (Vmp over 40, the more the better).

Click to expand...

Secret panels. Good one! I don't have any panels in my possession or on order. I'm just thinking about plans and wanted to explore this configuration issue.

Apparently my original post wasn't very clear. I'll have 16 panels and two MPPT's and will connect eight panels to each MPPT. The question is the same for each string of eight so I've been discussing it in eight-panel terms.

Sprucebeach said:

I don't exactly understand what you meant in your last sentence. Would you mind expanding on that a little?

Click to expand...

The original post said two controllers. I assume that meant two separate strings of panels electrically but physically all in a single row with progressive sun or shade sweeping across during the day.

I would have 8 on the left connected to controller 1 and 8 on the right connected to controller 2.

Panels like this: C1,C1,C1,C1,C1,C1,C1,C1,C2,C2,C2,C2,C2,C2,C2,C2

Not alternating: C1,C2,C1,C2.......

Sprucebeach said:

Apparently my original post wasn't very clear. I'll have 16 panels and two MPPT's and will connect eight panels to each MPPT. The question is the same for each string of eight so I've been discussing it in eight-panel terms.

Click to expand...

Okay, that makes sense.
Without knowing more about your shading, it sounds like the more "zones" you can make, without other compromises, the better your array will function as a whole in regards to shading.

With a 150V min voltage, your balance act will revolve around meeting this requirement too.

If you got 45Vmp panels for example, you'd be able to make 4S2P arrays for each MPPT. This would give 2 "zones" per MPPT.
If you are able to keep 3 of the 4 zones for example in sun, you will always be at 75% or better.

There's an opportunity to get 20 panels as well for 5S4P if you have the room. The 5S (5x panel Vmp to meet your MPPT minimum) will perform better in low light conditions too (early morning, evening and cloudy days).

MisterSandals said:

Okay, that makes sense.
Without knowing more about your shading, it sounds like the more "zones" you can make, without other compromises, the better your array will function as a whole in regards to shading.

With a 150V min voltage, your balance act will revolve around meeting this requirement too.

If you got 45Vmp panels for example, you'd be able to make 4S2P arrays for each MPPT. This would give 2 "zones" per MPPT.
If you are able to keep 3 of the 4 zones for example in sun, you will always be at 75% or better.

There's an opportunity to get 20 panels as well for 5S4P if you have the room. The 5S (5x panel Vmp to meet your MPPT minimum) will perform better in low light conditions too (early morning, evening and cloudy days).

Click to expand...

I'm struggling with what happens when shading starts with a 4S2P array. When two of the four panels in one parallel substring become shaded and the bypass diodes have kicked in the voltage produced by that side of the parallel layout will be halved. The other four-panel substring in the parallel layout, being located farther along the roof, will still be producing full voltage. What exactly happens then?

Sprucebeach said:

I'm struggling with what happens when shading starts with a 4S2P array. When two of the four panels in one parallel substring become shaded and the bypass diodes have kicked in the voltage produced by that side of the parallel layout will be halved. The other four-panel substring in the parallel layout, being located farther along the roof, will still be producing full voltage. What exactly happens then?

Click to expand...

The parallel group producing the lower output simply drops out.

Again I would connect them in sequence to keep a set at full power as long as possible.

This: C1A,C1A,C1A,C1A,C1B,C1B.......

group A in series paralleled with group B that is in series.

Sprucebeach said:

I'm struggling with what happens when shading starts with a 4S2P array. When two of the four panels in one parallel substring become shaded and the bypass diodes have kicked in the voltage produced by that side of the parallel layout will be halved. The other four-panel substring in the parallel layout, being located farther along the roof, will still be producing full voltage. What exactly happens then?

Click to expand...

4S2P is 4 in series and 4 in series to create 2 4S strings. These 2 strings are connected in parallel, thus the 2P.

So if 2 panels in a 4S string are shaded, the current will drop out significantly but the voltage will remain near Vmp.
The 2 4S strings will combine in parallel with roughly the same voltage but the sum of amps between the 2.

Perhaps the all the panels that I’ve used have had marginal diodes, but I’ve been amazed at what a few fat sticky leaves on one panel can do to a string of 440W halve cell panels.

Thanks for these responses. I take the points that MisterSandals and time2roll have made about using 4S2P arrays. As I understand it, in that configuration one shaded panel in a four-panel substring would essentially kill the output from that substring and the array output would be cut in half.

If I look at using an 8S1P array instead, my mental picture of what happens is clouded by my not understanding when a panel's bypass diodes might be activated and what effect that would have on the series array. For example:

Eight panels in series. One panel starts to get shaded by leaves and branches and soon is producing virtually nothing. Do that panel's bypass diodes ever activate so that the other seven panels can continue to produce at 7/8 of the former voltage and power? If that were true it would be better than losing half the power with the 4S2P array. But since knowledgeable people are recommending 4S2P I suspect that I've got that all wrong.

My understanding of the diode function is that the diodes will allow the rest of the panels push power through the shaded panel. This feature is primarily in the design to protect the panel. If current is forced through the actual cells they will heat like resistors and wreck the panel. So the diodes are included and all is safe.

However there is a penalty to push power backward through the diodes of about .5 volts each and there are usually three. Maybe more on the larger panels as I am not familiar with them. Somebody will correct me if I am off a bit here.

So in addition to producing no power that shaded series panel will trim some additional voltage from the others for a bit more power reduction than just the lost panel.

Gets to be 6 of 1 half dozen of the other. Also I would assume the shade is early and late in the day when production is lower so it is not a prime loss of power.

Sprucebeach said:

The bypass panels are my hope too. See my reply just above.

Click to expand...

OK when sun is off-angle, but if some panels are shaded when others are aimed straight at the sun, producing full Imp, diodes may not be selected and heatsinked for that much current. Resulting possibly in shorted diodes, possibly in catastrophic damage to the panel.

time2roll said:

View attachment 136454

Click to expand...

Therefore, orient panel so shade hits just one diode-bypassed section, not all three at once.


As was mentioned, panel voltage vs. MPPT operating range and max voltage. Battery voltage is related to minimum MPPT voltage.
You might have a situation where anything from 1 panel through 8 panels in series can be used, so progressive shade seems OK.
But price of SCC vs. current and operating range may indicate some other arrangement would be more cost effective. (high voltage SCC stepping down to same low voltage cost more and handle fewer watts.)

Just my rooie 2cents here, but our 24 (100w) panels are in two rows of 12, 3 strings per for. 4S6P. Per WP's advice for partial shading I bought this PV Combiner with reversing diodes. Hope this helps. It's one of the few on AMZ that have them.