my first attempt at home made panel

[RifRaf]

no vibrating table, just supported between two couches with the cushions removed, so i cannot sit on them for 3 to 4 days but it allows me to get under the panels to see whats going on

the vacuuming sound like a great way to remove bubbles, not thinking i shook the smaller container of sylgard once before adding it and ended up with the most bubbly mixture, i just work the bubbles out by tapping the mixture and rotating the cups, i use clear plastic cups after mixing the two parts together so i can see all the bubbles

for thinning i use xylene, a 1 part xylene to 3 parts sylgard, and stir till well dissolved before adding the second part of sylgard combo. this appears to work from my limited testing

i am not recommending this method though, for one xylene is very flammable, however once cured there is little to none left, its just to thin it out, after 2 months the panels still look ok, i have tiny bubbles in places, like 1 to 2mm, but can live with that, and the odd pocket where an edge of a cell lifted a bit, but will fix those with extra sylgard once i finish the array

[martinjsto]

here is the first panel now completed and wired into the junction box. i am realy pleased with the results and output and I have learned a lot for the next new improved modle
one thing i have found though, is that when i have the diodes connected in the junction box i loose half the voltage. not sure what i am doing wrong i must have them wired wrong, see pics below, worst comes to worst i will not have them in the panel but similar to RifRaf and have them in the array instead.
anyway a lot learned and a new panel in the making. this one is pumping 7 amps at 20v so i have a 140 plus watt panel.
A big thank you for all the assistance i have recieved from this forum, especialy from RifRaf, thanks mate.

[RifRaf]

congrats , and happy to be able to help.

your diodes must be wrongly positioned, ignore the resistor for now. so we can fix the diodes can you please draw a set of 36 cells/boxes in one row, one end is positive and one end is negative of you panel. then draw in your bypass diodes and connect to the long string of boxes where you have them connected.

i found that by looking at your panel on paper as one long string of cells the bypass diodes make far more sense. if you can post that i'll will try edit it if it does not look right

[martinjsto]

ok, here is what I think you mean. pic 1
is my layout in 1 long string pic two three and four is the actual panel wiring
pic 5 is of the actual wired junction box, I have not soldered the joins untill i get the diodes working.
thanks.
edit. the pics look like i have 2 sets of 2 diodes, this is not correct sorry, the diodes undernieth the drawrings of the cells represent the ones in the junction box.

[RifRaf]

all looks ok to me at least. to get 1/2 voltage one diode must be acting as if in shade mode, is one getting alot hotter than the other? maybe the voltage drop is less than across the 18 cells you are trying to bypass making it conduct? try a standard rectifier diode if you have some with a higher voltage drop to see what happens. hopefully someone with more experience in these issues could suggest something.

when you are testing and seeing this reduced voltage with the diodes, what is the voltage across A to C, and C to B, does the current stay near 7A?

[SpaceOddacy]

I'm sure it has been asked but I can't help myself.

How long does this encapsulation material last? (assuming no air pockets and it seals up perfectly)

[martinjsto]

Hi Spaceoddesy, I think I read the dow corning company gives it 30yr plus lifetime, given perfect conditions I suppose, If i can get my first batch to last 10 yrs I will think I have done well. hopefully if done correctly paying attention to details(somethink i need to work on hehe) the panels will get better and last 20 plus years, no reason why not.

RifRaf, do you clean off the residue flux off the cells before encapsulation? if so what do you use? i believe the roslin pen flux we use is not supose to be a problem but from electronic circuit board making i have seen this go bad before, something i should of thought of before encapsulating.
I use to us electronic board cleaner a preasure pack spray cleaner,

found the problem with the diodes. its the old ID10tT error.
I had the diodes backwards. rewired them today and tryed the 10amp standard ones suplyed with the kit. interestingly the standard diodes loose 1 volt even with no shade but the TO220 package didnt show any loss on my cheap multimetre.
so today, with results varying slightly with cloud cover, i am getting 20.5v to 21.2 v at 6.75 to 7.36 amps across a and b, the amps on either side of the diodes, a to c, b to c are the same 6.75 to 7.63 and the volts exactly half.
so looks like its functioning fine,
i will fix the air by vacuming and injecting with fresh silgard once the new batch arives. making a frame to mount the panel/s on top of shed for a season test, I wont put on house due to fire and insurance issues.
It runs to a battery setup im doing in the shed and into a inverter to keep the telly running and the beers cold (with a few more successful panels).
they will eventually end up on our farm as panels for our new house once we get started, unfortunatly a few years away yet but time to make better panels

[SpaceOddacy]

@martinjsto
Thank you for enlightening me, I appreciate it.

[RifRaf]

martin, as the cells i use were factory tabbed and clipped at the cell edges there is no flux over the cells, the few that i had to retab have a bit but am not too concerned about it. if properly sealed from the air should be ok. if you are concerned though would use 100% isopropyl and a soft brush to remove the flux from the cell face, this will get it off and not leave any other residues

[martinjsto]

Thanks RifRaf, there fragile enough to handle so another process adds to the damage risk so thats where the spray type for circuit boards should be better. i will have to see whats the solvent they use first. I keep the use of flux to a minimum, so we will see how it goes.
made the mounting frame and mounted the panel yesterday and connected to a main junction box then into the shed and the cheap charge controler and the batteries for a few months test before taking down to block. left all day on charge and this morning connected up the 300w inverter and then plugged in my soldering iron to solder a few connections with the sun power, a good feeling.
will leave it now with a couple of battery drill chargers for a load.
im not to sure about the battery condition but they seemed to charge up ok and are holding 12.3 v.

[martinjsto]

last couple of pics.

[russ]

Nice looking work you do!

Best of luck Martin!

Russ

[martinjsto]

just a quick update, no fires yet
batteries charging and holding well, amperage has dropped off a little today, still showing good charge to batteries, not sure if its the solar conditions or the panels, do they settle in after a while? and stablize to a relatively stable current (weather conditions accepted). still need to finnish the few air gaps with new sylgard.
installed RCD's for the 240v side of the inverter and permanent volt/amp gages for the panel imput. need to go down to block to retrieve two N200 batteries to give me better storage capacity.
on a different note: I would like to state that the issue i had with the delivery and consequence short postage of my gear, has been settled, i have finaly been recieving deliveries of the correct buss wire and tabbing wire, even some spare and also 20 new exellent cells to replace the dammaged ones. this seller has certainly made good although there communication and logistics has a few issues, but i am now confident you will get the correct good quality gear from them in the end.

[DeltaFox 25]

You can put sylgard down on the glass first then put your cells down and then more sylgard . this way you have no air space there.

[martinjsto]

i am trying 3 different methods to discover which uses the least amount of sylgard rather that just pour and see. ive seen that method of laying cells in a coating on the glass, and also the pour around 3 sides way on you tube by mark, RifRaf pours onto cells and let the sylgard flow around the cells and has excellent results with no backing as using both methods show. i first tryed the cappilary method which is realy good results in some parts except i didnt allow enough sylgard for a 36, 6x6 cell panel so ended up with gaps. my ambition is to achieve a method that is easy to reproduce, use the minimum sylgard and achieve 100% encapsulation with no air or waistage of sylgard for a glass/cell/sylgard panel.
the sylgard is a big part of the expence but a nessasary requirement it would be nice to be able to know exacltly how much is required per panel. i have started to build a vacumme/resin injection system utilising the glass then the cells then a plastic bag type temporary substrate or another thought is another layer of gladwrap coated glass with spacers. the idear is to place the cells on the glass, wire with busses and junction box connectons which is the issue with two glass layers as a mold. covering the back with airproof substrate and vacuume the air out before releasing the resin to be drawn into the voids completly encapsulating the cells. once set remove the bottom glass or temp substrate leaving a airless panel. i will try this for my 2nd panel and post pics and info on the results. if i can achieve a good vacuum and not crush the cells in the process then the system is proven in the carbon fiber aircraft industry to achieve some of the best repeatable controled results.