Solar charging with some shade

[2oldman]

I'm facing west today, so I'm getting a pretty good shadow on one of six series-connected panels. Panels are all around 135w @ 12v.

Shade is on the far-left panel - about half of it:




Not a bad charge for that big of shade- 51.9v x 10.6a



The shaded panel is the last in a series string. It's a Kyocera. Looks like the diodes are working well.

[CA_Traveler]

BFL13 wrote:
If that shaded panel is half by-passed, then it is putting out half its current? (I really am confused about the current amount)

Because you keep returning to the same incorrect assumptions.

Try this: Assume you can magically rewire a shaded bypassed section of a panel so that the section is no longer in the circuit. Would you not agree that the panel is now producing the same current but with less voltage - say 1/3 less voltage with a typical 3 diode panel?

Also we're discussing array voltage and array amps. Of course the controller output amps are now less as it now has less input power due to the reduced input voltage and it has to maintain the output voltage to the battery.

[CA_Traveler]

Salvo wrote:
What happens if one panel is totally shaded? There have been tests where now the string has significant reduced current. To run this test, all you have to do is point all panels but one towards the sun. Rotate the other panel 45 degrees away from the sun to simulate shading.

I was not sure of those test results at the time. This thread shows 1 panel producing power while the other 2 are in full shade. I was interested in how much charging could occur in the storage lot.

I'm not suggesting mine was a good shade test but rather "it is what it is".

[BFL13]

If that shaded panel is half by-passed, then it is putting out half its current? (I really am confused about the current amount)

So now the whole string is down to that same half-current? So why is Imp still 7.3a out of rated 7.6a and not more like 3.8a?

Vmp is way down at 77.8, but Imp is hardly down at all. Input watts is down from the low V, but the current bottleneck with series story is that it should be down from low current too.

In fact the OP photo shows the shade covering more than half the vertical columns (I think a couple un-shaded cells at the ends of the shaded columns don't count for anything--the column is killed?) So you would think all its current is shut off if it is a typical two diode panel 18 cells each. Whatever, the panel's output current should be low?

So is the sting output current the same as Imp or not? (if it is the same why is it not low?)

[CA_Traveler]

BFL13,

Look at the physical layout of the above diagram. At the bottom all of the necessary wiring is very close to the junction box and 2 short additional wires provides all of the connectivity for the bypass diodes in the junction box. This is a very common way to wire bypass diodes.

A 6x10 array will have 3 diodes for the 3x 2x10 cells. A 4x9 array will typically have 2 diodes for the 2x9 cells, etc.

Many other wiring combinations are possible but costs go up.

[Canadian_Rainbi]

smkettner wrote:

Thanks for that. Though at this time of the morning, I too have a headache from reading this entire thread before breakfast and only one cup of tea.

[time2roll]

[BFL13]

pianotuna wrote:
bfl13,

There are usually 3 strings of cells with 3 diodes.

It seems to vary how many diodes and sections there are. ISTR my Sharp 130w had two. My Sun panels had just one. But the big 24 has 3. However the K 135 seems to say it has 8! Here is what they say they do:

11. BYPASS DIODES
Partial shading of an individual module in a source circuit string (i.e. two
or more modules connected in series) can cause a reverse voltage
across the shaded cells within the module. Module output current is
then forced through the shaded area by the remaining illuminated cells
and other PV modules in series with the partially shaded module(s).

The current forced through the shaded cells within the PV module (or
modules) causes additional module heating and severe loss of power.

The purpose of bypass diodes is to provide a low-resistance current
path around the shaded cells, thereby minimizing PV module heating
and array current losses.

The KDxxxSX-UFBS series modules employ bypass diodes that have:
? Rated Average Forward Current [IF(AV)] Above maximum system
current at highest PV module operating temperature.
? Rated Repetitive Peak Reverse Voltage [VRRM] Above maximum
system voltage at lowest PV module operating temperature.

So the shading causes loss of "power" which is from loss of voltage in the VA, since the amps rating does not go down with heat.

Here is more:

http://electrialstandards.blogspot.ca/2015/07/bypass-diode-function-in-solar-panels.html

[pianotuna]

bfl13,

There are usually 3 strings of cells with 3 diodes.

[BFL13]

Almost had another glimmer.

You are saying that the 36 cell 135w panel with Vmp of 17.6 and Imp of 7.6 (say) is 36 cells in series with each cell having a Vmp voltage of 17.6/36= 0.49v and an Imp of 7.6?

So if 1/2 shaded (in the right way say 2 columns of 9= 18 shaded) it will be a panel of 18 cells in series having 8.8 volts Vmp but still with Imp of 7.6?

So that 18 cell panel in a string of 36 cell panels, still has the full Imp but lower voltage to add in to the string's total V?

Edit--except when I cover part of a single panel I still get fewer amps than when uncovered.

[Salvo]

Both claims can be correct. It all depends on bypass diodes. If there are no bypass diodes or the diodes do not conduct then claim A is correct.

[BFL13]

I have/had? a glimmer but need clarification, if available 🙂

A. the claim is the lowest current panel (the shaded one) drags down the current for the string, including the Imp from the string.

B. the other claim is that the shaded panel has lower voltage so that reduces the total voltage of the string (and so its wattage) but the current remains the same.

How can both be correct?

Now in the OP photo of the controller it shows Imp of 7.3 and Vmp of 77.8 with the shading. Imp of a 135w panel is about 7.6, so let's ASSume the insolation at the time is a little less than STC, and allowed for an Imp of 7.3 and that was what it would be even with no shading. (claim B)

How can the Imp be so high at 7.3 if the shaded panel's current is cut back and that has lowered the array current to that? (claim A)

Vmp at 77.8 is way low compared with say 6 x 17.6 Vmp = 106v at STC, so the array wattage is much reduced for sure, but the claim is that is that 77.8 is the Vmp the controller has selected based on whatever.

If the Vmp "knee" on the IV curve for the prevailing insolation is that low, so is the insolation low (lower curve on the graph, knee farther to the left), but then how can the Imp still be way up there at 7.3 (out of 7.6) as on the upper insolation curve near STC?

If I had a string and an MPPT controller that displayed Imp (not all do) I would like to run a test where I shaded one panel somewhat and watched Imp to see what it did.

[2oldman]

BFL13 wrote:
what about the OP's 48v set-up and grid tie inverters?

I don't have grid tie inverters.

[Salvo]

I agree with your description.

What happens if one panel is totally shaded? There have been tests where now the string has significant reduced current. To run this test, all you have to do is point all panels but one towards the sun. Rotate the other panel 45 degrees away from the sun to simulate shading.

CA Traveler wrote:

Simple example: My array is producing 5A at 90V and 5*90/14=32A which the battery is accepting. A shadow actives 2 bypass diodes of the 9 and now the controller finds a new maximum power point at 5A and 90*7/9V and the output is now 5*90*7/9/14=25A to the 14V battery.

[BFL13]

2oldman wrote:

CA Traveler wrote:
I have no idea what the two of you are discussing ....

Neither do I.

Don't try! I don't know either, brain has fried.

So moving right along, nothing to see here folks, what about the OP's 48v set-up and grid tie inverters? It looks like they have stick house inverters that don't even use a solar controller? So why does the OP need his new 48v inverter and his controller, now that he has gone 48v?

[2oldman]

CA Traveler wrote:
I have no idea what the two of you are discussing ....

Neither do I.