Diodes and bypass diodes

[longwolf]

Yeah, the pictorial is not good, the gray wire shorts all the cells out. Tab wire goes from topside, to bottom side, not across the whole row.

Well, that was a given. I guess I could try to shade the tabbing wires where they go under the next cell.

[longwolf]

I hope I got the polarities right. I'm assuming that the electrons are collecting on the top of the cells, making the tops the negative side.


EDIT: Oups, grabbed the wrong image for editing, I just replaced it with the right image.

[Mike90250]

diodes have to bypass cells:

build a stack, and try (before you epoxy it all together)

[longwolf]

Maybe these will clarify things.

[longwolf]

I guess the last pics didn't do it.
Maybe these will. If they're wrong, please show me where because I can't see it.

[Mike90250]

Try a couple stacks, and shade it. See how it works before you "seal" it all . Sorry about all the torture.

[mpan1]

I guess the last pics didn't do it.
Maybe these will. If they're wrong, please show me where because I can't see it.

Hey there!

Images very nice!

What was the result?

Worked for you?

[longwolf]

Sorry for the delay, but I'm working on this as I can.
I finally got to do the test on Sunday.
And I'm not going to pretend to understand the outcomes.

I took 4 strings with 4 cells each and connected them in series, the same way they'll be layed out in the panel.
I then took 3 diodes with jumpers and used them for the tests.
The diodes are rated at 15 amps/ 30 volts.

The first image shows the 1st test arrangement, the one that looked like it should work.
Well, it didn't. The voltage dropped from 8 to a bit over 4 volts and the diodes were getting pretty hot.
Just to be sure, I tried reversing the diode polarities, it got even worse.

I almost chucked it in. I was doing this test in a frizby golf park, with the cells laying on a piece of cardboard and I didn't know when the discs might start whizzing by.
But I decided to try another arrangement, the one shown in the second image.
It worked!

1st test was without diodes
Rows Covered ; Volts ; Amps ;
------------------------------
0 ; 8.02 ; 7.35 ;
1 ; 7.73 ; 0.149 ;
2 ; 7 ; 0.03 ;
3 ; 6.09 ; 0.023 ;
------------------------------

Test with diodes
Rows Covered ; Volts ; Amps ;
------------------------------
0 ; 8.0 ; 7.26 ;
1 ; 7.75 ; 7.26 ;
2 ; 6.88 ; 7.22 ;
3 ; 6.51 ; 0.156 ;
------------------------------

Now, I'd really like to understand why my preferred arrangement failed.
I'd also like to know if there's a better arrangement than in the 2nd pic, preferably with images.

Right now I'm sitting in my van, it's near 100 degrees in here and I can't run my fans for fear of damaging my batteries, so PLEASE, enough of the Socratic method.

[mpan1]

longwolf!

At the momen, i am wiring the cells and trying to check the space for the diodes connections!
I ll do the effort (i have some elecrtic wiring diagrams) and as soon as i finish, i ll let you know!

[mpan1]

Longwolf!

Check this out!

http://www.ibselectronics.com/pdf/ac/Diotec/applications/solardiodes.pdf

[Mike90250]

Longwolf!

Check this out!
http://www.ibselectronics.com/pdf/ac...olardiodes.pdf

Very nice link. the schematics there should aid in getting the diodes & panels to work together.

Keep in mind, that diodes can be easily damaged by heat while soldering.

[longwolf]

The more I look at the following pic, the more it looks right.
It's just the full panel done as the test that worked.

[longwolf]

Sorry it's taken me so long to do an update.

I was going to do step by step on how I made the panel.
I took pics of most of the steps I used, but my cheap chinamart camera's battery died and lost all the pics

My panel is made with 6x6 cells rated at 0.5V and 8A per cell.
You take 36 cells and wire them in series for 18V at 8A.
Most panels made like this use 4 strings with 9 cells per string.
That way they use less bus wire.
I set my panel up with 9 strings with 4 cells per string.
That allowed me to use eight bypass diodes.
The diodes are rated for 30V at 15A.

I took a base reading of the finished panel, then I started covering rows, one at a time, and took more readings.
Here they are:

Rows Covered | Volts | Amps
0 | 20.3 | 7.6
1 | 18.63 | 7.35
2 | 17.77 | 7.39
3 | 17.32 | 7.41
4 | 17.08 | 7.44
5 | 16.75 | 7.38
6 | 15.2 | 7.31
7 | 13.3 | 3.8
8 | 12.3 | 3.3
9 | 11.3 | 0.063

I tested the diodes ahead of time and they average a drop of 0.125 volts.
So I would have expected a drop of at least 2.125 volts for each row covered.
I don't know why I was getting such great readings, but I'll take them
Oh, the diodes only become active when their section of the panel becomes shaded.

You can see the final wiring setup in post #27 of this thread.

These are the square type diodes with a hole that allows you to mount them.
I was going to attach them directly to the frame and let it act as a heat sink, but it turns out that the mounting bracket has electrical continuity with the anode. Which means they would have all been shorted together.
So I made 2 inch long L brackets and insulated them from the frame. It worked out well because they also help support the glass and hold the wiring in place.

Right now I'm getting by by doing handyman work.
This design will let me carry ladders or lumber on the ladder rack, when I need to, and still get a good charge!

[RifRaf]

am using the same 6x6 cells and trying to work out the best way to incorporate the bypass diodes. is the diagram linked to below correct? the idea is to have 2 strings of cells, each going into its own 600W regulator. maybe the blocking diodes are not required with this arrangement of bypass diodes? http://rifraf.dyndns.org/solar/solarlayout3.png

any tips or comments on the diodes, diode values etc would be appreciated

[longwolf]

am using the same 6x6 cells and trying to work out the best way to incorporate the bypass diodes. is the diagram linked to below correct? the idea is to have 2 strings of cells, each going into its own 600W regulator. maybe the blocking diodes are not required with this arrangement of bypass diodes? http://rifraf.dyndns.org/solar/solarlayout3.png

any tips or comments on the diodes, diode values etc would be appreciated

Hi RifRaf,
I'm no expert and was hoping someone else would help you.

If your panels are going to be stationary you should try to place them so that there won't be any shading on them. In that case the panels won't need diodes built into them.

If you do a search you should find diagrams of how to place the diodes between panels in case of a single panel failure.

good luck!