diy solar

diy solar

Bypass diode sources and how to spec them...

gfmucci

New Member
Joined
May 21, 2021
Messages
74
What is a good source for bypass diodes? And what is the best way to spec them for a specific panel?

For example, if a panel is ~21 volts, 3 in series, each 11 amps, rated at 200 watts?

I've searched several electronics suppliers and I don't notice anything that seems appropriate.

Can anyone tell me the value of these diodes from the photo below? Can they be tested in circuit (without removing)?
20210719_153849.jpg
 

Attachments

  • 20210719_154039.jpg
    20210719_154039.jpg
    152.8 KB · Views: 7
Last edited:
So you would need a 65 volt 11 amp Schottky diode. Add a safety margin, and up that to 100 volt and 15 amp.
Mouser has over 1,400 that meet or exceed those specs.
Thanks. I went to that link. None of the photos, no matter which of the 1,000 combinations of Schottky variables I select, look like my diodes.

This was my result from doing a search on the Mouser site:
 
If I were to ask about a blocking diode, would this change at all?
 
Last edited:
Voltage rating should exceed the voltage of the panel or that panels sub-sections. In the sun there is no current thru the diode, just (reverse) voltage across it.
When a panel or subsection is shaded current from other panels/subsections will pass thru it. What do you expect that to be? 250W/48V = 5.2A
 
How do you size a blocking diode?

The OP has three panels in series at about 63 volts each pushing 11 amps. Would you stick to a 15 amp diode and up the voltage to handle the sting?

I am installing a string of panels on a charge controller that will be nowhere near the same angle and azimuth to the other two panels in the 3S3P series. This is when a blocking diode is supposed to come in handy to prevent the strings with different outputs from back feeding.
 
If I were to ask about a blocking diode, would this change at all?
The voltage rating would need to be higher than the panels output, and all series panels. 2 48v panels in series = 96v, so 150v rating or higher.
 
The blocking diode of the shaded panel/or string will be reverse bias by the Voltage source from other panel/string.
For example yo have 4 panels in parallel, each panel has Voc of 50V, that means if one of the panel is in the shade, the blocking diode for that panel will have to be able to handle the reverse bias of around 50V, so you should get the blcoking diode of 100V and the current rating that is double the current rating of the panel, the diode will also get quite hot when it is in forward bias so heatsink will be needed. I try one of those inline MC4 blocking diode rating at 15A and they are junk, the MC4 was so hot to touch when only 5A was flowing. You can use the heavy duty blacking diode module like MD40A1400V that is typical used in the combiner box that has blocking built-in.
 
Voltage rating should exceed the voltage of the panel or that panels sub-sections. In the sun there is no current thru the diode, just (reverse) voltage across it.
When a panel or subsection is shaded current from other panels/subsections will pass thru it. What do you expect that to be? 250W/48V = 5.2A
Normally a subsection is always ~13V max and the diode will never see the full Isc. The allrounder 15SQ045 Schottky diode should match 99% of all cases.
 
Last edited:
How do you size a blocking diode?

The OP has three panels in series at about 63 volts each pushing 11 amps. Would you stick to a 15 amp diode and up the voltage to handle the sting?

I am installing a string of panels on a charge controller that will be nowhere near the same angle and azimuth to the other two panels in the 3S3P series. This is when a blocking diode is supposed to come in handy to prevent the strings with different outputs from back feeding.
You should be using fuses with more than 2 parallel strings. Per resources posted here.
https://diysolarforum.com/resources/fusing-guidelines-for-solar-panels.144/
MC4 Fuse Holders
A friend had many panels, all in parallel. One panel failed and drag down the output. It had a shorted bypass diode.
 
Last edited:
The blocking diode of the shaded panel/or string will be reverse bias by the Voltage source from other panel/string.
For example yo have 4 panels in parallel, each panel has Voc of 50V, that means if one of the panel is in the shade, the blocking diode for that panel will have to be able to handle the reverse bias of around 50V, so you should get the blcoking diode of 100V and the current rating that is double the current rating of the panel, the diode will also get quite hot when it is in forward bias so heatsink will be needed. I try one of those inline MC4 blocking diode rating at 15A and they are junk, the MC4 was so hot to touch when only 5A was flowing. You can use the heavy duty blacking diode module like MD40A1400V that is typical used in the combiner box that has blocking built-in.
OP is speaking about bypass diodes.
If a string is shaded the diode will conduct and must handle the current of the remaining cells which is always < Isc.
In normal operation the diode will be reverse operated and will never see more than ~25V
 
Last edited:
No, in parallel connection of the panels, if Voc of the shaded panel/string is less than the Voc of unshaded panel/string the blocking diode of the shaded panel/string will be reverse bias.
Also if one of the parallel panel shorted out, the blocking diode will also stop the dumping current from flowing into the shorted panel so that blocking must have high enough blocking Voltage rating.
 
Thanks. I went to that link. None of the photos, no matter which of the 1,000 combinations of Schottky variables I select, look like my diodes.

This was my result from doing a search on the Mouser site:
The one in your panel looks to be the typical 15SQ045 Schottky diode, 15A 45V.
So NMNeils suggestion of 100 volt diodes is inappropriate? I also thought that I would need at least 65 volts. I looked at a bunch of sources for 65 volt or greater Schottky diodes. I couldn't find anything more than 45 volts at 10 amps. I wonder if anyone even makes them.
 
Can diodes be tested in line or do the have to be removed from the circuit?
 
No, in parallel connection of the panels, if Voc of the shaded panel/string is less than the Voc of unshaded panel/string the blocking diode of the shaded panel/string will be reverse bias.
That is also normally the case: the bypass diode is normally reverse biased by the forward voltage of a string of 18 cells. It will not see very much more excepted when a string is cut.
 
Last edited:
What DC Voltage do you get when panel is in full sun and you put Red probe on the Cathode and the Black probe on the Anode of each bypass diode?
 
So NMNeils suggestion of 100 volt diodes is inappropriate? I also thought that I would need at least 65 volts. I looked at a bunch of sources for 65 volt or greater Schottky diodes. I couldn't find anything more than 45 volts at 10 amps. I wonder if anyone even makes them.
100V diodes is not wrong, but useless.
Bypass diodes are normally reverse polarized by the forward voltage of one string of 18 cells: never more than 13V, excepted if the string is completely defective and open circuit. but then the panel is dead anyway.
 
Last edited:
Back
Top