Bypass Diodes

[Mike90250]

If the diodes are not properly heat sinked at their heat sink terminal, they will overheat. Tab wire is too thin to carry much heat, maybe 2 or 3 layers will give enough copper to carry heat away. The diodes only get used if a portion (not always where the diodes are located) of the panel gets shaded

[Zardiw]

Which is why the bypass diodes are most commonly located in the junction box rather than sandwiched into the panel at some other location. Even in the junction box the manufacturer typically relies on the wiring to conduct heat away rather than take the cost and space hit of a heat sink.

Well these are specifically designed to be sandwiched......hence their thinness. I've already started leading them with bus wire....pretty sure that will adequately transfer any heat away from them very well. There IS a center square on them that may be designed for a heat sink.....but I'm not planning on using it. Also they're rated to 200C.......fwtw........I doubt the panels will get that hot........will see......in the Summertime it can get up to 120F here in Palm Springs.......which is only 49C...But in the sun I suppose they will get a lot hotter.....I had some copper heating coils in a glass box and it got up to 188F ....which is only 87C.

Hoping there won't be any really hot spots. This first panel is made with factory reject cells.....I've tried to use the highest output ones....Guess we'll see how it turns out...........z

[inetdog]

... you should heatsink them - don't forget their case is electrically live.. With .63V @ 7.5A, you will be dissipating about 5 watts per diode, so chose a heatsink accordingly, strive to keep junction temp below 90C, sure their spec is 150C, but they won't last long that hot.

Which is why the bypass diodes are most commonly located in the junction box rather than sandwiched into the panel at some other location. Even in the junction box the manufacturer typically relies on the wiring to conduct heat away rather than take the cost and space hit of a heat sink.

[Zardiw]

Got my ByPass Diodes!.. (SBR12U45LH -13)......They are awesome. Will add them to my second panel.

Here's the spec sheet on them from Mouser: http://www.mouser.com/ProductDetail/...2bjUGdJbh3w%3d

They are really thin and will work great between the glass sheets.

Am also in contact with MicroSemi who just came out with a 'Smart' surface mount type bypass diode that is only .7mm thick. and designed to be incorporated inside panels.

z

ya glass underneth the cells would work wonders. im sure I was thinking that to.
Anyway im off to next town where they have dollar store and grabbin 60 more of these tiny cells. will be back in 2 hrs

[Zardiw]

More $0.02: While wood is a fine material for many applications, my experience is such that it's particular characteristics, one of which, BTW, is relatively good thermal insulating qualities compared to most metals, make wood less than ideal for solar panel applications, either for module framing or panel/array support, pretty much regardless of surface coating or other treatment.

I have to agree.......The problem with wood is if you get a Hot Cell......it will Burn the wood possibly starting a fire.....

He did say he was going to use layers of foil and heat resistant paint......but it will still trap heat and possibly char/catch fire.

And all the trouble and expense........I think it would be way better to just use glass.............z

[Mike90250]

Wood for signs is one thing.
Wood frames with glass and sealed in PV cells is a solar oven, and will cause the problems I listed earlier.

[J.P.M.]

More $0.02: While wood is a fine material for many applications, my experience is such that it's particular characteristics, one of which, BTW, is relatively good thermal insulating qualities compared to most metals, make wood less than ideal for solar panel applications, either for module framing or panel/array support, pretty much regardless of surface coating or other treatment.

[Zardiw]

I have to completely disagree with you there friend. Ive done plenty of Paint work for signs for business hand painted. And for over 10 years The wood remains as good as new.
No colorizing no warping even for the thinnest of wood my friend. It defiantly depends on the wood you use and how well you treat it.
But infact I use various techniques on my wood.
When your a artist who has done commercial signs by hand. You learn how to treat the wood for preservation for long periods of time.
And I do believe that my signs will go well over 25 years looking like new with no warp at all.

See what most people do when they use wood for solar, They dont do any sort of treatment to that wood. Let alone even paint it or seal it or varnish it nothing.
And Most of all.. Aluminum heats faster than wood. So in that regards you are only cooking your cells alot faster being encapsulated than you would with wood.

But if you dont agree that is fine. You wont always be correct. Im just saying this from experience with commercial sign work.
I know many frown upon it. But not many know how to treat wood at all.

Good conductors of heat: Copper, gold, Aluminium
poor conductors: Wood, plastic, glass, china, cork

How are you going to seal the cells......will you use glass on the front?.....I think you should go glass/glass and seal it with silicone. That's what I'm gonna do. Found some old patio doors that are tempered glass.....will build it on that and then silicone the back with sheets of single strength glass.....should last 50 years.....lol......That's what the top solar panel mfgr is doing now. Getting away from encapsulating with EOV or whatever it is..........z

I have to completely disagree with you there friend. Ive done plenty of Paint work for signs for business hand painted. And for over 10 years The wood remains as good as new.
No colorizing no warping even for the thinnest of wood my friend. It defiantly depends on the wood you use and how well you treat it.
But infact I use various techniques on my wood.
When your a artist who has done commercial signs by hand. You learn how to treat the wood for preservation for long periods of time.
And I do believe that my signs will go well over 25 years looking like new with no warp at all.

See what most people do when they use wood for solar, They dont do any sort of treatment to that wood. Let alone even paint it or seal it or varnish it nothing.
And Most of all.. Aluminum heats faster than wood. So in that regards you are only cooking your cells alot faster being encapsulated than you would with wood.

But if you dont agree that is fine. You wont always be correct. Im just saying this from experience with commercial sign work.
I know many frown upon it. But not many know how to treat wood at all.

Good conductors of heat: Copper, gold, Aluminium
poor conductors: Wood, plastic, glass, china, cork

[Mike90250]

Wood is barely OK for a toy panel you use for a year. Over time, and repeated heating in the summer sun, it will carbonize, warp and do all sorts of strange things. Year 3 it's likely to burn up for you, most likely in the middle of the day when you are gone.

I think wood would be okay for solar panels if done right. if you use, heat resistant, non flammable paint on the wood. and also couple of layers of Foil. Just my two cents.

[Zardiw]

WOW.....look at these.......they are thin enough to incorporate into the panel itself.......where the cells are.....not in the J Box.....only have .38V forward voltage drop........they are not 'smart' diodes though:

http://www.digikey.com/product-detai...ITR-ND/2791834


Edit: Those are the ones I'm gonna use.......only 1.42 at Mouser
z

[Zardiw]

I'm going to be building some 5 x 12 panels and due to the size of the glass I have available, I'll be putting them between every 2 strings of 5......should protect the panel really well against shading........z

[bcroe]

So where can you buy them.......do they make thin ones?

z

Thats a cool part. Its based on the idea that an FET tuned off looks like open in normal direction,
a diode in the reverse. Or a short (in either direction) when turned on. I used that concept for a
180A solid state power control a while back. The problem is, with only 2 terminals and only
26 mv voltage, where do they get enough voltage for gate drive? The video shows, they hickup
every so often back to a diode, to get enough voltage to recharge. Might generate some electrical
noise, not nearly as much as interrupting the current. Bruce Roe