diy solar

diy solar

Reverse current protection diodes

Logic28

New Member
Joined
Nov 9, 2022
Messages
64
Location
Holbeach Lincolnshire - UK
I am building my first PV system and I have a question in my mind. I have noticed that most Combiner Boxes have the reverse protection diodes positioned after the individual fuses before the isolator and merging point (Bus Bar) however since I will have a smaller consumer box next to the combiner with one isolator for each string, it would make it for a tidier arrangement to have the diodes in the smaller box and therefore before the fuses which are in the much larger combiner box. I cannot think of a reason why that would be wrong however I'm sure some of you might have a better understanding and prove me wrong.
Thank you
 
I have softened my stance on blocking diodes with the prevalence of used panels these days. I've had arrays hit by lightning and leakage can be higher than with new panels. You also never know when a panel or two may go bad. I haven't installed any diodes yet, but they are far more useful than fuses.
 
Reverse current protection diodes are rarely (basically never) needed.
There are mixed opinions on this.

I think the reality is that there can be small to moderate benefits to blocking diodes in some situations.

Let's do a thought experiment with the two extreme conditions:

Case 1: Best Case) All the panels are in full sun and at the same angle to the sun.​
In this case, the different voltages between the strings would be minimal and there would be no backflow from one string to another. In this case blocking diodes are of no real value. (In fact, they can be counterproductive. See below)​
Case 2: Worst Case) One string is in full sun with no shade but the the other strings are in pitch black.​
In this case, the blacked-out strings will have no voltage and the other string will drive reverse current into it. This would waste a lot of energy. In this case, blocking diodes would prevent the energy loss​

In reality case 2 would never happen. All of the strings would have some amount of light so the real-life worst case would be something between the two cases. If one string is in the shade, the current may be low, but what little sun it is getting will drive it along the I-V curve to a relatively high voltage, helping prevent back-current. If the shading is not very bad, there would probably be little or no back current. However, I suspect that if you have a situation where some of the strings get into deep shade there may be a small or moderate advantage, but I have never tried to test this situation enough to quantify it.

What complicates the equation even more is that blocking diodes will have a voltage drop.....and this represents an energy loss even if all the panels are fully illuminated. Consequently, the gain of the blocking diodes would have to be enough to counter the loss of energy going through the blocking diodes. For case 1 above, diodes would be counterproductive. For a situation closer to case 2 they may be useful.

As I said, I can not quantify it, but I suspect that blocking diodes in most cases are not particularly useful but may be useful in certain extreme shading cases.

BTW: In the early days of solar, a '12V panel' would be hooked directly to the battery without an intervening charge controller. In this case, a blocking diode was an absolute must because at night the battery would drive reverse current through the panel. With a modern charge controller, this can't happen.
 

Attachments

  • 1669937571661.png
    1669937571661.png
    80.5 KB · Views: 2
Last edited:
In reality case 2 would never happen. All of the strings would have some amount of light so the real-life worst case would be something between the two cases. If one string is in the shade, the current may be low, but what little sun it is getting will drive it along the I-V curve to a relatively high voltage, helping prevent back-current. If the shading is not very bad, there would probably be little or no back current. However, I suspect that if you have a situation where some of the strings get into deep shade there may be a small or moderate advantage, but I have never tried to test this situation enough to quantify it.

When you disregard thermal drift, Vmp stays fairly stable over a large range of solar power. At 5 times less power you still have virtually the same Vmp. I know a person who claims he gets solar power just before sunrise. Unless you throw a blanket over one panel, I don't know if you'll ever get a diode to block. At full panel power, each blocking diode could be consuming 9W!!!

7884-pv-voltage-curve_125777129-660x1024.jpg
 
In reality case 2 would never happen. All of the strings would have some amount of light so the real-life worst case would be something between the two cases. If one string is in the shade, the current may be low, but what little sun it is getting will drive it along the I-V curve to a relatively high voltage, helping prevent back-current. If the shading is not very bad, there would probably be little or no back current. However, I suspect that if you have a situation where some of the strings get into deep shade there may be a small or moderate advantage, but I have never tried to test this situation enough to quantify it.

I have watched my panels produce > Vmp before sunrise with indirect ambient light. This was this morning 3 minutes before sunrise:

1670010455179.png

This is the basis if my "almost never". Something would need to physically obstruct the panel blocking all light from enough cells to drive the Voc of that string below the Vmp of the parallel strings. I just don't' see that happening, and it seems to me that very few array designs use blocking diodes.
 
There are mixed opinions on this.

I think the reality is that there can be small to moderate benefits to blocking diodes in some situations.
. . . . . . .

. . . . . . As I said, I can not quantify it, but I suspect that blocking diodes in most cases are not particularly useful but may be useful in certain extreme shading cases.

BTW: In the early days of solar, a '12V panel' would be hooked directly to the battery without an intervening charge controller. In this case, a blocking diode was an absolute must because at night the battery would drive reverse current through the panel. With a modern charge controller, this can't happen.

After reading all posts I can only think of one potential solution that would eliminate the need for blocking diodes while still preserving full protection against any reverse/abnormal current and that would be one solenoid disconnecting relay per string controlled by a rather simple voltage comparator/minimum voltage, programmable for any specified application's need.
 
Back
Top