| 
 
 
 
                  
  
 
  
  
   
   Handy Link..... Sun Tracker.A tool by SunEarthTools.com to
 calculate the sun angles for your
 location at different times of the
 day and season.
 | 
                      		
                        		| Make your own Solar Panels! | Page 1 | 2 | 3 | 4 | 5 |  By Oztules  It's  something I said I would never do..... but now I  have. 
 
 If you have no value on your  time, and want to do your own DIY solar cells, then it has always  been possible to cobble some cells together, stick them to a surface,  and place plastic or glass over them, measure the output..... and  generally feel good.
 
 A fancy frame helps with the karma...  but you know in the end it is only temporary.
 
 As the water  vapour creeps into the sealed wooden framed ones, or the window glass  you scrounged up...breaks because of the heat/cold and localized  heating in particular cells, and it all falls in a heap. It can get  pretty sad for a lot of devotees that have tried.
 
 There are  plenty of youtube stories and how too's...... but all (but the ones  using sylgard (or whatever it's called)) are doomed to failure)......  except those who have tried to use EVA (ethylene vinyl actetate).
 
 I  viewed a few treatments on the youtube sites, and none of them  inspired a great deal of comfort. They all revolved around using a  heat gun and a heap of good luck to melt the EVA around the cells,  and onto the glass.........
 
 Why I thought, did they make solar  cells 30 years or more ago, and still basically the same today...and  yet the amature panel maker comes up with lots of bad ways to do  it??
 
 I pondered this for a while, and in the meantime, a  plumber friend turned up with a few solar hot water panels..... also  25 or more years old..... they were stuffed..... but they did use  tempered glass, and it was iron free glass.... and it was free glass  ( this was the clincher).
 
 
  So I sat and looked at a few 2 sq  meter panels (1x2m).......... yep, had to have a go too.... but it  cant be cut!
 This  is where I decided to look around the world at the current state of  play. This was sobering, no real effort at making proper panels like  the big boys built without the big boys price tags... why not?
 It  appears that folks thought you needed a laminator style of machine,  which were very expensive, and a barrier to making your own. But if  you look at what needs to be achieved, it is really only a few simple  steps.... so lets look at the how and why.
 
 Traditionally they  use a sheet of glass (iron free tempered if you can get it), a sheet  of EVA, the cells themselves, another sheet of EVA, and whatever your  having yourself. After the last EVA sheet, you only really need to  put something there that will not wick, and will give you something  for the very very sticky EVA to bond to that is not your "jig"  It can be painted later if you want more protection. Usually they use  tedlar, but really any inert permanently waterproof material will do  for protection.
 
 The simple theory of what we are trying to  achieve is this:
 We need to get the cells fully encapsulated in a  UV stable envelope, that resists water and air from denaturing the  cells. It must allow for differential expansion rates to occur  between the silicon wafers, the glass and the medium itself.......  and no bubbles.
 
 Ethelyene vinyl acetate is a plastic sheet  .5mm thick. It is soft plastic in texture and seems innocuous  enough....... however, when it gets to only 65 degres C, it melts.  This means that we need not get to very high temperatures to get  something to happen.
 
 I'll bet there are plenty of hopefuls (  like me for instance) that grabbed some EVA, some glass, and the  wifes oven, and made a prototype cell......but it is filled with  bubbles!..... It does stick to the glass and, it envelops the cells  well, but it is useless because of the bubble population.
 
 It  is at this point that you start to get a bit anxious about where to  get a laminator...... but you don't really need one. We just need to  emulate what it does.
 
 Next thing we look at is what happens  when it melts. To get the EVA to be useful, we need to get enough  energy into the stuff to cross link the molecules. It changes the way  the material behaves, and increases the melt point considerably as  well. At low temp (65C-80C) it melts to itself very well, but is only  slowly converted into the final product, so we have to keep it at  these temperatures for long periods..... but as we increase temp,  things happen a whole lot faster, and 5 mins may be enough at 145C to  get it to cross link, and stick very strongly to the glass and the  cell itself.
 
 It also needs some encouragement from  pressure..... or a vacuum perhaps, and use the air pressure to do the  job of supplying the pressing force.
 
 Now we know it works  perfectly well without vacuum, but you will have myriads of fine  bubbles.... blocking out the sunlight.... well, not so perfect after  all.
 
 So we need a vacuum pump. Ideally one that can draw a  near perfect vacuum. The vacuum you can achieve sort of directly  relates to the bubbles you will get left with. High vacuum.... no  bubbles. It's that simple. The vacuum will also press the cells very  firmly (200 plus pounds per cell for the 6x3 ones) against the glass,  and this will help the glass bond, and make for a very flat cell,  with no voids in the plastic envelope.............as always there is  a but:... you must have NO leaks at all, or you will get uneven  vacuum in the sheets, and you will get bubbles forming.... probably  in the inter cell space, but bubbles all the same.
 
 The heating  profile I will use is like this:
 
 15 mins at around 50C with  full vacuum
 15 mins at 65C with full vacuum
 70-80 mins at  100C-120C (or hopefully more) with full vacuum
 This last bit is  dependent on the EVA you get. Low temp EVA is used for window  lamination, where electronic components may be incorporated into the  glass matrix....
 For “normal” EVA if we can get the temp  to rise some more.... then good. Even if the  containment  bag (ok  garbage bag) fails at this point, it's work is done for all practical  purposes... better if it does'nt.......but we get what we can.
 So all in all.......it  looks doable.
 First step for me was a BIG oven.
 
 
                          
                            | An oven  that would handle the 2 meter by one meter panes of glass that came  from out of the solar water panels. (it turns out that for some quirk  of history, there are plenty of these old solarhart type panels  laying around over here). |  |  
                            | We know that the  temperatures involved are not terribly high in order to achieve the  bonding and cross linking that is required to get the cells to be  protected, and pressed against the glass without any bubbles..... so  I started with a wooden construction..... yes for an oven.
 I  had recently pulled down a wool shed for a fellow, and I was given  the wood.... lots of old 4x2 hardwood.... so this was to be the frame
 It looks like this >>
 |  |  
                            | I was also given some  packing material in 8x4 foot sheets of low grade 3/4"  pineboard.(someones kitchen apparently turned up in them). These  would serve for the bottom and lid of the oven. For the sides I just  used some of the old floor boards from the shearing stand..... 
 The  whole thing is lined with ceiling roof bats insulation and then all  held in the walls with a sheet of TYVEC building wrap material....  could be anything really.... even paper.
 |  |  
                            | Ok, that gets us the oven  walls and floor, and the lid was just a sheet of the 8x4 pineboard  with insulation laying on the top.... Now we have an insulated  container that will take our panel. |  
                          
                            |  |  |                            What this means to the  60watt panel builder.....is that even an insulated cardboard box will  do for an oven (I used one with a fan heater blowing from in front  for the initial trial, with blankets and dooners for the  insulation...... perfectly good for a 60 watt panel)
 Next step  is a heating element.
 
 This was simply solved by destroying a  fan heater/radiator. These things draw about 2000 watts and  incorporate a fan and a nichrome heating coil. They spill out a fair  stream of not too hot air, so are unlikely to melt our plastic bags  (comes later) when we bring the oven up to temperature.
 
 I  built the heater body out of a "think safety" sign (ironic  really), which housed the heating element at one end, and the fan  motor at the other. The fan blades were plastic, and would melt in  this environment (proximity to the glowing element), so I stole the  aluminium blades from my big MIG welder, and put the plastic ones  back in that.
 
 This gives us a 2000w heater that blows warm  air..... but not too hot,and will allow us to gently heat the oven up  to about 100-120 degrees, without any cunning trickery required to  temp control it. Because the oven is so well insulated...... we just  keep adding watts from the heater until we get to where we  want.
 
 Finished it looks like this:
 
                          
                            |  |  |  
                            | Perhaps finished was too  strong a word to use...... but it does work. |  |  
 Page 2  |