| 
 
 
 
                  
  
 
  
  
   
 Buliding a 200 watt dump load
    
     
A dump load is used to drain off  
excess power. This design uses a  
few 50watt light bulbs and with  
some parts from the junk pile,  
shouldn't cost more  
than $20 to build | 
                      		
                        		| Make your own Solar Panels! | Page 1 | 2 | 3 | 4 | 5 |  
  Back to Page 1.
 Sober reflections:
 
 
 Eva  onto toughened glass has a track record, and most folks know of  panels that are at least 1/4 of a century old and still running fine.  The EVA of today is better than the EVA of yesteryear, and will be  less likely to change colour than the older material. I have some BP  panels 15 years old, which have clearly changed to a brownish tinge  over the cells... from the heat, not the UV alone. The inter-cell  spaces are still clear, as are the last 1/2 inch of the outer  perimeters of the cells.... but even then, the performance is  probably only 10% less than stated on the stickers.
 
 If you  build the panels for 50c per watt, and they only last a few years....  then it would have been cheaper to buy them made.... as they should  have lasted at least 12 times longer.... so to keep replacing is  $6.00/watt......??
 
 To reiterate the preliminary vacuum stage.
 
 With large cells, the  metal tab on the front will try to take the pressure of the air when  the vacuum is pulled on. It can then try to push the tab wire into  the eva, to spread the pressure load. When the air temp is cold, the  EVA won't give easily enough, and the silicon cell will try to wrap  around the tab wire so it can press against the eva as well.......not  being flexable, it will instead split along the tab.
 This is  easily mitigated by raising the temp up to 50C or more (less than 65C  though) for the prevacuum, rather than room temp.
 
 I found room  temp fine on a hot day, but too cold at night...... so now use the  oven to get 50C. This softens the eva, and makes it easy for the tab  wire to sink into the eva, so the silicon presses evenly against the  "glass".
 
 Moving right along, we  now have an oven to cook the panel in.... we have bought the vacuum  pump, and are now prepared to make a test panel/cell..... just to see  if it is true.
 I decided that it would be best to do a show  and tell for a single cell at this point, in a step by step  approach.
 
 So here we go:
 
 
 
                          
                            | First get a piece of glass  that will fit your cell on. Wash/dry the glass with alcohol  (Methylated spirits).
 Grab the bit of your EVA your going to use,  I have chosen to build a single 6x3 cell, with some other bits thrown  in just to give an interspace between cell and cell bits..... like  this:
 |  |  
                            | Next throw on the top EVA  piece: |  |  
                            | Now a piece of tedlar, or  even just sheet plastic ( I used white building wrap this time)  anything you want the rear of the cells to wear. It must be  non-porous, rot proof, water proof and everything else proof. It only  performs the role of keeping the EVA away from anything that may want  to harm it.... and to allow you to get the panel away from the vacuum  bag. 
 The EVA cross links into something a bit like silicon. It  is  rubbery, and not especially strong..... although try getting it  off the glass once cross linked and bonded properly and you may  decide that it is very strong.
 
 Here we are with a piece of  building wrap  on the back. (The wrap is porous but the EVA will soak  into it and fill the pores.... it has been found to be better to just  use plastic sheet... the EVA when heated initially will be a runny  liquid.)
 |  |  
                            | Now we have all the  ingredients to make it work, we just need something to allow the  vacuum to suck the air through..... an air wick if you will. This can  be any porous material you can get hold of. I find for small 60watt  panels, that a few thick teatowels covering the back do ok, for  bigger panels, it is best to find something a bit better.... beach  towels? Glass matting would be good if you have it, I tried shade  cloth on the 2mx1m panel, but found that it shrinks a little when it  gets heated. Not enough to damage anything, but enough to make the  cells not line up as I had placed them. For this show and tell I just  folded up a rag that was lying around near me... and it looks like  this:
 |  |  
                            | 
 Next we need a plastic bag  big enough to place our panel in. Unless your into building big ones  like I have, most likely a big garbage bag will be the best for the  job... cheap, plentiful, and plenty strong enough for the job. Here I  used a transparent one for demonstration of seeing the stuff in the  bag
 |  |  
                            | Next we poke a hose into the  bag and seal it up I use plastic wire ties. This seems to hold the  bag opening onto the vac hose pretty well.... except for this time.  The plastic bag must have a hole in it, as I could not get down to  max vacuum...... |  
                            | Do NOT place the rubber suction hose over a  cell. It will crack that cell, so just place it between your airwick  material and the glass... or inside the folds of the material, but  away from the cells themselves, and let it suck from there. Provided  there are no leaks,  all the air will all eventually seep through the  wick. If there is a tiny leak, your wick will need to be very very  porous to still work. If it is a bit difficult for the air to move  through, a leak will kill this project.... if it is a very free wick  material, a leak may be tolerated.... best without though.
 |  |  
                            | So I have replaced the clear  leaky bag with a trusty black garbage bag....
 
 |  |  
                            | Note the wire ties: |  |  
                            | Now for this exercise I  won't use the big oven, but instead will show just how simple this  can be....... 
 Here is the latest you beaut high tech  laminating machine...... or the build it in 1 minute oven..... and it  will work every bit as well as an expensive  laminator
 |  |  Of course we know there is  more to it than this ......like the heater.
 For this size ... and  up to at least a 60W panel, a fan heater pinched straight from the  loungeroom would do to get it up to 80-100C..... or if you wanted to  prove a point, perhaps broil it in boiling water.....ie half a 44  gallon drum parked over the camp fire........simmer till cooked.
 Well I guess that would  work too, but I'm trying to make clear that this does not need to be  high tech, to get a perfect result.... just temp time and vacuum .If you are going  with very low temp, then it will take a very long time... ie 80C for  80 minutes, or 140c for 5-10mins
 
 I have toyed with the idea  of using the sun... ie use the some of my spare solar glass from the  hot water heaters, and make a solar oven... very similar to the solar  panel itself... as the oven.... I haven't tried it as yet, but if you  can boil water on a hot day with a sheet of glass and the black  backing..... then , it can crosslink the EVA easily too.
 
 
 Note the solar cooking  has now been done in Africa to do it similar to this.... but without  the vacuum...... they were still happy.
 
 
 Your trial one will  give you an idea of how long to cook it with what equipment you have  to get your brand of EVA to crosslink properly and stick to the glass  VERY well.  
                          
                            | If it has not been left  for long enough, you will be able to peel it off the glass,  we don't  want that..... try just putting it back in and cook some more...  noting how many minutes gets the required bonding... if it is too  poor this extra cooking may not be enough, but has worked  before. 
 Our trial one looks like this
 |  |  
                            | and the whole sophisticated  arrangement looks like this: |  |  
                            | And yes, it got to .....and  stabilized at about 112c... so I cooked it for 45mins anyway. Note  the improved insulation (brown towel) pushed the temp up another 20  degrees from the picture first showing the box oven without this high  tech improvement. |  |  
                            | I don't think you can  overdo the time on these low temp crosslinking exercises, although it  is probably worthwhile exercise to use a normal cooking oven with a  good thermostat to get an accurate picture of how the stuff bonds to  the glass... how long for how well. That is the only critical figure  in all of this.... and theoretically  then you can always put it back  in the oven without any problems at all..... but I haven't had to as  yet 
 And here is what we have got  ourselves.:
 |  |  
                            | So it does work, and it is  dead simple, and very easy to get right.... just test your EVA until  you get to know how it works for you. 
 The only difference to  doing this size
 |  |  
                            | or this size:
 |  |  It                          is how much time you have, how  many cells you have, and how much glass you can scrounge up
 
 Page 3  |