Solar panels

A solar panel is, as the name says, a panel which collects solar energy and converts it to electricity. At least that's my simple view of it. A solar panel consists of a background on which solar cells are "stuck". These cells are interconnected and ultimately there is an electric wire which can be linked to an inverter. In a system panels can be put in series and or parallel. Among others, depending on the inverter.

What a professional panel is made of I do not know. I've read somewhere that the rear side is of aluminum and the solar cells are placed on top of that and on the top there is (special) glass over it. I must be honest that I would like to know more about this but do not have time for it. As mentioned earlier, I came to the site of Chris van der Zwaal and he described a way of building a panel which sounded like a good method to me.

Do it yourself solar panel

The aim is to build solar panels. Not solar cells, but panels and connect them via an inverter to the electricity grid. These panels are as follows: The back is made of glass. Then the cells are glued on top of that. The solar cells are inter connected and ultimately linked to an electrical wire. On top of that a glass plate. The two glass plates are sealed by sealants. Ultimately,  an aluminum profile if put aound it's done. Quite simple :-)

The cells

Only a few parts, however, many questions. First but the most important and expensive part are the solar cells. I bought the cells through ebay in America (www.ebay.com). There are a number of suppliers, but only a few large ones. One of them did not standardly shipped to the Netherlands and another I did not think he had really interesting cells. So finally I ordered with Ever Bright Solar Inc. It takes about two weeks before your belongings arrive. This is mainly because it is very long at the TNT / customs. See the solar cells section. I think if you live in America the cells will be there in a couple of days.

The glass

There are many types and sizes of glass. Glass will let light through, but not all. It is therefore important to find suitable glass. Fortunately, there is a sector that has the same requirement: greenhouse. So for my panel, I went looking for horticultural glass for a good price with the right size. I found used glass at www.marktplaats.nl which was former gardeners 4mm thick glass with size of 165 x 73. Which is a perfect size (I hope) for a panel. The cells measure 8.1 x 15.0 cm, and in total there will be 18 x 9 cells on a panel. That would be nice to have some space for the sealant and bus wire. I chose this because the glass is cheap. The only problem: the glass can break which would destroy my entire panel. During my quest I spoke with a gardener who told me that if you use the glass on a sloping roof all hail to the size of pinpong balls will not break the panel, only hail the size of tennis balls will create a problem. This has given me a little reassurance and I only hope that the panels never get hit by serious hail. There seems to be also some special glass for solar panels to be available. Search for solar glass on the net and you will see, but I would not know where you could buy this and what it costs, but I expect that it is expensive. For now I will just go for the gardeners glass.

The sealant.

To glue the solar cells on to the glass and to create an airtight construction a sealnt is to be used kit. This sealant, glues very well on glass, but also other materials like aluminum. For solar panels there is a "special" sealant made. This sealant I could only order in America and that I was a little expensive and would take a long time to get it due to transport. So I searched for a sealant which is available for sale locally. Eventually I came to zwalum 1001-U because according to the manufacturer it was used for solar panel construction. A problem, however, this kit was not for sale. Eventually I found an internet shop that sold it and ordered it there. But then I got another sealant. Instead of 1001-U I received NO (yes it is called NO). When I looked at the specifications of these two sealants side by side I found that they were the same, so I think it is ok. Anyway, the sealant glues very well on glass and metal and the use is of -50 to 120 degrees celcius, which I believe should be sufficient. Please note; I have read that the panels can easily reach 100 degrees celcius in the sun on the roof, so you do want material that can handle high temperatures.

Electric wire.

Also for this there is "special" material available in the solar world, and all for a friendly price :-). So I searched for "solar cables" alternatives. There are of course plenty of cables for this sort of thing. First, I looked for "ship" cables. Used between ship and shore and must be able to handle something and they can according the specs. The ordering of these cables is difficult because they are usually in great lengths and only delivered to companies. Eventually I came to a cable from conrad that had good specs. Up to 180 degrees. Good weather resistant, so that will hopefully last.

Silica gel.

The idea for the panel is to be airtight and weather resistant so that the cells will last. To do this a sealant is used. At the time of closing there is a certain amount of moisture between the panels. This should be removed to ensure that the panel doesnt condensate. To remove moisture in the panel I will put silica gel between the plates just before I close them. This silica gel can be ordered (sealed), but you can also use "old" silica gel. Silica gel is often in packs in white bags. The grains in the plastic bag consume moisture and keep it. As I have read silica gel does not degenerate. "Old" silica gel can be dried in the oven or microwave. On the Internet you can find much more information. I do not have old bags so I have ordered some new, but if I would have old one I do not know what I had done.

Diode

A diode in a solar panel can be used in two ways. Either as a blocking diode which is placed in series with the paneeel to block returning power. Or a diode may placed over one or more cells (part of a panel) as a bypass diode. The latter is made to protect for shadow effects. The shadow effect is created when one (or more) cells are in the shade, and the other cells are not. The cell in the shade is not a power generator but instead becomes a power consumer. This cell can consume quite much. This has two adverse consequences. First, it is at the expense of the generated power of the other cells. Secondly, the cell gets hot and wears out over time. The latter is often referred to as hotspots.

The blocking diode is relatively straight forward. It is in series on the plus side of the panel and as soon as the panel does not produce but consumes than the diode will block the current. The disadvantage of this diode is a voltage drop and thereby uses power. It is therefore very common to use Schottky diode for this type of diode. I am not very familiar with diodes but I understand that the diode has to handle as a minimum, the VA of the panel. So a Schottky 45V 4A should comply. I probably choose 50V 8A. but that will depend a bit on where I can buy them. I read that it is important to have them be able to handle a bit more amps.

The bypass diode is another case. There are many websites which write about them and show when to use them:

http://www.daviddarling.info/encyclopedia/B/AE_bypass_diode.html

In the above schematic it shows how a blocking diode to be placed and how to place the bypass diode. The following link shows a page (shockwave required) how and when the diodes work:

http://pvcdrom.pveducation.org/MODULE/Bypass.htm

And a text on bypass diodes to solar cells in space. The costs are of course irrelevant in this case. To be honest I can not really pick out what would be a good diode, but I have not read the article very well:

http://www.emcore.com/assets/photovoltaics/3O_D9_03.pdf

A bypass diode is also called a shunt diode. Hmm, google books:

http://books.google.nl/books?id=U6ZQ3gbD8YUC&pg=PA81&lpg=PA81&dq=bypass+diode&source=bl&ots=bx68Bf0Dhe&sig=h_0cAYY1Cy5ShJAR9FItI6krI1g&hl=nl&ei=wbouSuDSMYXK-Abmu-CICg&sa=X&oi=book_result&ct=result&resnum=10#PPA81,M1

Normally, on every 18 cells a diode is placed. But that is physically quite difficult. Below the schematic structure of a panel:

Zonnepaneel 72 cell schema wiring
 
For a string of 18 cells to protect with a bypass diode, you need a large bridge. This is not really useful in this format. If I would use 4 diodes to place on any 18 cell I get an enormous amount of buswires. One possible solution is the following:

Zonnepaneel schema diodes (5 stuks)

In the above diagram I have 8 groups of 9 cells used. My idea was that group 2/3, 4/5, and 6/7 each are protected by a diode. And that a possible solution is that group 1 and group 8 are protected by a diode over 32 cells. The latter is sub-optimal, but in any case it is something, and when two or more cells of this group (32cells) are in the shade than that would (if I understand correctly) not create any hotspots. The problem I have is with the central diode. This would (at least that was the intention) protect group 4/5. But I suspect that if a cell in group 4 or group 5 is in the shade than one of lower diodes will become the bypass diode. Unfortunately I do not have enough knowledge of electronics and the behavior of cells in order to determine.

After posting the problem on a forum and not receiving any response, I have decided to use 5 diodes to work with, but there will be buswire to bridge 9 cells. So I come to the following schematic:

Paneel diode final

The diagram also indicates what is plus and minus of the cell. Again each blue block suggests a group of 9 cells in series. This solution will cost around 1.5 meters buswire.

Also I found a Schottky diode. Initially I wanted to use the SBR10U45SP5 or SBR1045SP5 of diodes inc, choosen mainly on the basis of the advertisement they have circulated about this diode. The greatest advantage of this diode is that it is only 1.2 mm thick, but I have 3mm space between the glass so it can be slightly thicker. The diode of diodes inc I could purchase in volumes of 5000 or more, so that was not feasible. Then I looked and I searched and found the ON Semiconductor diode MBRD1045. This diode is 2.4 mm thick but has slightly better specs. Lower voltage drop, lower leakage current and higher amp max. Also, this diode works up to 175 degrees celcius so that should be no problem. This is the diode that I've ordered and will use for my solar panels.