Protecting the solar cells

[Ogou]

Hi everyone,

I'm new on this site.  I just bought a 5lbs of solar cells

I want to make a solar panel but I'm very concern about protecting the cells. I can see it's also a main concern on this website.  A folk here even use mineral oil to protect his cells...

In my research on the internet I found that many PV manufacturers use protective silicone-elastomer called « sylgard 184 ».  Did anybody use this sylgard 184 as a protective sealant for their cells?

I also found another product I think can be use as an encapsulent for the cells.  The sillicone-elastomer is UV resistant, with negligeable shrinkage and it is not brittle.  It is call clear crystal 202 made by www.smooth-on.com.  This product is normally use for casting, molding in sculpture but I wonder if it is good idea to use it as an encapsulent if it has all this properties?

Thanks

Ogou

[DanG]

I have a bunch of loose PV cells too & am waiting till I can get 12 or 15 year life w/o surprises...

Having the cells lying on the reinforcement plate and covering with 1/8" or 3/16" of encapsulant is an early and usefull cell protection scheme. It imparts higher light transmission losses and a shorter (10-15 yrs) usefull life expectancy but for utility and harsh conditions it can't be beat.

Many marine navigation aide and meteorological station buoys use something along those lines - Illinois DOT use some thing similar too I've noticed. I've owned bouy arrays mounted on 3/8" STIFF fiberglass board, wired underneath and coated with 2-part silicone liquid that failed QA from cells being too close together that I got for $2 a watt back in mid-eighties.

An engineer aquaintance spoke about thin-as-water 2-part self setting encapsulant with the disadvantage that after production started overseas it was discovered seagulls and other waterfowl would spend all day trying to eat it. The found a harder formulation.

The only problem is there is no known solvent for set silicone so no cell replacements - but with backplane wiring only having a thing coat on it open cells can be jumpered out.

An evergreen type 24V PV array is 25"x61" inches for 1525 square inches surface area. There are 231 cubic inches to one gallon so that would cover the 24V array to about .15 inch deep - or a 12V panel to .3 inches; or 1/4 inch with the backplane given a light coating. Five pounds of cells after sorting may produce six 12V panels so you'd need at least 6 and maybe 8 gallons - but I have yet to price materials more then just in passing.

Do you have prices on the products you mentioned?

[Ogou]

Well I want to make a 50W solar panel 20 in x 40 in.  I will put an acrylic edge (1/8 inch thick) around the glass, to contain the encapsulant. My maximum volume will be 20*40*1/8= 100in3. So a gallon will give at least  two  50W panels.  But I will need less than 100 in3 because of the volume occupied by the 36 cells.

 36*6* 3* 1/64 = 10.125 in3

For the price :

For the "Clear Crytal 202", you can find it on sculpt.com follow the path :

Online catalog - casting product - polyurethane - Rigid Polyurethane Casting Compounds

Pint kit - Crystal Clear 202 (1 pt. part A, 1 pt. part B) - 1.9 lbs - 25.99$

Gallon kit - Crystal Clear 202 ((1 gal. part A, 1 gal. part B)- 15.2 lbs - 147.99$

5 Gallon - Crystal Clear 202 (5 gal. part A, 5 gal. part B) - 76 lbs - 599,99$

The "Sylgard 184" is made by Dow Corning, if you go on their website, you can "find a distributor" on your state.  I called and the price for 8,8 lbs (unfortunately it is not by volume) is 253$ + shipping. If 1 gallon of "Clear Crystal 202" is 15,2 lbs.  I can assume 8,8 lbs will be close to half gallon.

JRN

[willib]

as i undestand it ,HEAT buildup in an array is a problem because as the cells heat up their efficiency drops, so if you could vent some of the built up heat , it might be better.

willi

[DanG]

If you have Evergreen style cells the peak power NOCT (normal operating circut temp) is 44°C (111°F) with -0.49%/ °C derating above NOCT - so peak power would be reduced 6.5% at 135°F.

With backplane connections well encapsulated the reverse area could be allowed to breath but 5 to 7% losses are expected however fashion the array is assembled. Heat will be mostly irradiated away as IR energy anyway - just keep it from being reflected back at itself or other panels...

[DanG]

I would calculate the thickness formula as 'needs extra' for gaps around and under cells then for 'less needed' above cells - hail storms, twigs and sticks hitting end on like an arrow, etc. is alot to ask from 1/10th of an inch of encapsulent.

Those ridiculously high prices are retail for small portions - I am looking for a "B" grade coating that may need surface coat refreshing every year vs. spending nearly as much as a store bought panel that still requires recoating periodically.

I looked hard at epoxy mixes and a slow cure rate (days) - epoxy surfaces will 'bloom' with opaque blemishes during cure but that can be dressed and coated w/ a polyurathane. Pure epoxies have minimal shrinkage during cure but it may be too much - many of the cull cells we have are laced w/ microscopic flaws just waiting to enlarge. Many of the cells I have want to shatter more then they want to stay together.

Here's a picture of cells I am trying to hatch currently (thanks TomW)

-

-

[DanG]

For surface mounting... The ClearFlex 95 looks promising. Water-clear, UV resistant and .0028 in. per inch shrinkage with a 95 shore A hardness (similar to polypropylene, just under nylon) but requires oven cure for thin castings seems about the best I've seen so far.

Nine 3x6 cells are 162 sq inches, I'm thinking 28 inch by 7 inch strips for 196 sq inches surface with four strips being 12V panel. Strips would also make oven curing somewhat easier for me. Tighter tolerances could be used, example: 170 sq inch but little error margin.

Spec sheet says 26.8 cu inches per pound - 3 'gallon' kit lists at 21.3 pounds weight (may include containers) so 20.7 pound product = 555 cu inches volume. If sold by true volume it'd be 693 sq inches but it looks like 'gallon' in name only.

555. cu in yields anywhere from 9 to 14 strips w/ thicknesses from .33 to .2 inches
     
12. strips equal three 12 volt panels for $60 each and a modular construction to allow for some repairs. That price is approaching tempered glass in bulk order (one sheet approx $150, 25 sheets approx $70 each was one quote I got)
    
    

The next larger quantity would reduce that to $54 a panel, and the $4000 quantity lowers per panel price to $47.

Thinking two part casting process also - one action to bond cells to back board and allign cells perfectly flat then cover with the spendy optical clear poly.

Seems doable so far IF one can find a stable (nill expansion when heated & no flexing) back board for cheap, and warp free framework for cheap also...

http://tinyurl.com/7v5ke ClearFlex data

http://tinyurl.com/759qu Shore Hardness definition

-

-