Cracked PV panes

[domeguy]

I've got 7 24V 25W Golden Photon panels on the way to the land fill, unless a miracle can happen.  Here's the story:

I built this array to hold 9 of my panels last summer.  At first, it worked as expected, but I noticed some cracks - due to my poor design, I think.  I covered the cracks with RTV to keep out water, but more have appeared since.  This array, which at one time delivered 6A now only puts out about 2A.  I've since built another array for another 9 panels that works much better, and is not cracking the panels.  Here is a picture of the array in question:

This is a picture of the panel label:

This is a panel with severe crackage:

And this is a closeup of one of the panels in the array with less severe cracks:

Is there anything to be done with these?

My thought is that they're toast - but the cost of replacing these makes me shudder, and I've seen very good ideas on this board.

Any ideas?

[pyrocasto]

I honestly dont know about fixing them with some type of glue. My only thought would be to take the glass out and replace it.

BTW, your glass probably broke because you're using wood. Unless you can garrantee that the wood wont soak up moisuture then it will and it will warp. I believe this is what cracked my glass on my panel as well.

Where do you live, because I'm sure someone wouldnt mind taking them off your hands for a few bucks intead of tossing them out.

[domeguy]

I'm in Northern Colorado, not too far from the Dans.  The panel on the ground looks like the cracks go all the way through, but most of the cracks don't appear to have broken the thin-film material inside.

I think they cracked for two reasons: The array was initially built in a way that allowed the weight of the array to actually curve the horizontal members.  The allthread that you can see on one of the pictures was added later, to correct this.  The other reason had to do with my rather lame method of retaining the panels to the wood - screws w/ fender washers.  Many of them were too tight.

My other array is made with dadoed slots in four horizontal members, to make three rows of three.  It has been great so far.

-Lee

[DanG]

Wow  nice form in the design of the tripod panel mount! Is it active tracker? Is is an original, or a borrowed design? It sure earned a place in my bookmarks - thanks for sharing, even if its bad news!

The pack-rat in me votes to burn a few days, weeks, in trying to save them.

Even baking the panels as is for 24 hours will still leave alot of moisture behind, and accelerate corrosion already started - It looks like you'll have to operate. I'd sure try and seperate the cover from the cells on the most corroded panel to see what is user-serviceable - cutting one corner of frame along the seam-weld and prying both legs to ease the grip on the glass; use a heat gun to relax the sealant-glue on glass/frame, etc - If the PV strings are modular you can stitch together nearly-unblemished panels back into the original frame w/ new glass.

The only NEVER DO I've seen & learned so far with PV cells is...  Never use wood of any type in the construction or mounting of arrays. It moves - I've seen that even with 3 coats of automotive primer and 3 coats of dulex 'cadilac' enamel it heaves and settles with time. Even most engineered wood, whose binder resins are otherwise stable, sags from gravity alone over time.

Realistically, coefficients of expansion of an array of available metals & alloys should be examined to get a near match on PV tempered glass or other cover material & the PV strings isolated from external axis of movement.

Thusly so, in saying that, here is a picture of 2200+ solar cells in cold storage waiting untill I get the vital 840 linear feet of extruded unobtanium frame rails & 60 sheets of translucent duraluminum. I keep thinking of all the years my panels couldn't make power if I cheat assembling them so there they sit chilling in the root cellar!

[domeguy]

Thanks for the encouragement, 99y.  I gleefully stole the design for this array from a picture I saw of a BP array on a gas station canopy.  It's supposed to allow tracking without any corners way off the ground and I was worried about high winds catching a high corner.  The picture I didn't include shows a 3x3 array with the panels trapped in groves between 2x4s.  I made the grooves deep enough and assembled them with a fair amount of slop so that there is some play between the panels and the wood.

You're the second person to suggest never building an array out of wood.  Obviously, Im missing something.  It seems to me that the situation would be worse with a metal-type array, since I think the difference in thermal expansion is greater between glass and metal than between glass and wood.

I'm not sure I understand the failure mode.  One of the panels, clearly is broken all the way through - it's a goner for sure.  The others show some discoloration inside.  This is a thin-film type of panel, not a single crystal type.  I'm not sure of the chemistry, but I've heard that Tellurium is involved.  I don't know if the presence of moisture causes a chemical reaction that is irreversible, or if this is simple oxidation and possibly some dendritic growth shorting things out.

BTW, I did design both of my arrays for a tracker, maybe one of the RedRock boards, but I manually adjust them for now.

[whatsnext]

Thermal expansion and cotraction of the wood is probably not the problem. Wood shrinks and expands a lot when moisture levels change. And it warps as it ages. I would never tie wood to gas ever. Each panel should be allowed to float away from the wood. Look at how they build windows and you'll see haw the glass can move a bit and window frames are smaller(less power in the warp), painted and protected from rain and sun. You're asking for trouble by staking the panels to wood.

John..........

[whatsnext]

Wow, that's a lot of spelling errors but you get the point. Don't tie the panels down directly to the wood. It moves way more than the glass.

John.........

[inode buddha]

Well, I'm not a PV guy really. I'm not a wood guy either. I do iron work for a living. I think that if you use wood or some kind of metal for a frame, then you should let the panels "float" so that they are not actually screwed down tight onto anything. Just make a groove and slide them in. As for saving the cracked ones, I dunno; maybe there is a way to saw and re-solder them? Anybody knows more than I do, please feel free to correct me. Heck, maybe someday I could afford to buy big panels like that!

[Tom in NH]

Wood is a fine material as long as you work within its limits. As others have said, it moves, swells, and shrinks, and you need to design things so they can "float". If you loosen the screws holding your panels down, maybe that would prevent the glass breaking. I have a lot of wood in my panels, which you can see at http://altenergy.blog-city.com There is plywood backing behind the cells, and of course the mount is made of 2 x 4's I was getting a lot of bulging of my plexiglas cover due to heat expansion of the plexiglas. On my newest panels that I'm building now, I think I've reduced it considerably by drilling oversized holes for the screws that hold the plexiglas to the plywood. The holes are about 50% larger than the screws. It seems to have helped a lot. It isn't fair to say never use wood. You can use it, but if you do, let things move. --Tom

[ghurd]

I think you will find the thin-film is on 1 layer of glass like paint.

The best chance I can think of would be to see IF the glass layers can be seperated.

Then it may be possible to solder the last good thin-film stripe to the return conductor.

That would make it a lower voltage, but if one was good for 18V and another good for 10V, in series it would still reach 24V.

Looks to me like there should still be more than 2 amps being made.

There could be some current going backwards right now.

The PVs with the cracked corners should still have good volts, but lower amps.

The PVs with cracks running along the stripes will have a lower voltage.

Consider testing each panel seperatly to see which way the current is flowing.

Or add a blocking diode to each panel to stop any reverse current.

Good Luck

G-

[DanG]

Domeguy - any chance on a clearer view or rough blueprint of the tripod, pleeease?

The pivot axle attachment & base details...  umnn, could you do a post just on the tri-pod? Or have a similar project link handy?  I think I see a bunch of those in my future.

Tom - Gotta admit there is a difference between attended and unattended installations - Fabricating something that will last 2, 5, 10, 20, 30 years unattended up on at second story roof heights or eighty yards upslope from dwelling, or all-the-above at a weekend place that gets visited 30 times a year demands metal construction. Yes, yes, IF its for two weeks camping each year use whatever, but if its going to be doing work homesteading I want metal.

I'm not anti-wood, personally I wonder if all the salvaged redwood I got squirreled away would be stable enough to use on PV frames w/ a two-storm-window PV capsule insert for a quick setup; for a bank of PV arrays to require the same attention & maintenance as a wooden boat seems counter-productive.

[domeguy]

99y- I'll try to get more pix next week.  I don't know if I still have any of the drawings for the array that didn't break any panels, but I think the details will be clear enough with pictures.

Others against wood for PV arrays - These are made from redwood.  Since I put them up last Summer, it's about time to treat the good one this year.  My other array (not pictured, yet) seems to be working well.  Time will tell...

[whatsnext]

Domeguy, Trust me, I'm not anti wood in any way. You just can't tie things directly to it unless it's flexible. Those tabs you have appear to be fixing the glass panes in a way that prevents them from moving around.

John.........

[domeguy]

John- Thanks for your feedback.  I agree with you.  That's why my second version of array didn't use that same technique.

To all respondents: I really appreciate the opinions and suggestions.  I'll repost later with some results, if any.

-Lee

[richhagen]

I have a few thin film panels of a somewhat similar design, although a different brand.  The glass seems to be pretty tough stuff, none of mine have broken, although I chipped a corner on one.  I did have problems with mine, the contacts on the back corroded (it was a different design than yours and used a metal contact held with pressure against the exposed metal film on the back.  The other problem I had was that the films on the back of the panel weathered and flaked in a few places.  I took them down and have only one panel of similar design in use right now, an ICP.  

Since the working materials of the panel appear to be layered on the back, it may be possible to salvage some of your panels if you can get at the right metal oxide layers on opposit sides of a crack and connect them via a jumper wire bonded to the panel.  It would be a funny looking fix, but it might save them from the scrap heap.  

Like some of the other posts, I would suspect that the frame and the glass were fixed tightly and did not move together, probably due to a combination of moisture change in the wood, and temperature change in all parts of the assembly.  I would bet that on a hot summer day those panels would get pretty warm, and on rainy days, the wood would get pretty damp.  I havn't seen photos of panels broken in the way that yours are before, but what I see leads me to suspect the way that they are mounted.  Rich Hagen

[N9WOS]

I hate to bring bad news, but any time the glass breaks, the thin film layer also breaks. That thin layer you see still connected across the cracks on the back is just a thick anti corrosion coating. The active thin film silicon, and conductors on the panel is less than a mill thick. Basically the same as the metal coating on the back of a mirror.

The thin film coating is on the bottom of the top glass panel. If it breaks parallel to the cell plating lines then you have basically broken the circuit. The bottom glass panel is just to seal out the moisture. That can break without and loss in output

The best thing you can do with them is get a glass cutter and try to find the largest square sections that you can salvage.

Take the metal frame off then take the back cover glass section of, then score the glass front and snap it into useable square pieces. Scratch the protective coating off and use conductive metal tape to atach to the exposed thin film conductors.

Some of them look like you can save about a full half panel, then cut the rest of the panel up into smaller panels.

They will be usable for other small projects.

[hiker]

did someone mention spelling---thought i might toss this in here :}