Demystifying (and Cutting to Size) Flexible Thin-film Solar Panels (Fuji F-Wave)

For the last 2 - 3 years you can find flexible thin-film solar panels quite cheaply (around 50-80euros per 100w) on the internet. The bulk of them are a solar panel called Fuji F-Wave 92W. The flexibility of this panel is amazing in such ways, that you can make a roll out it and treat it like a carpet (its nearly 3,5 meters long)

But there is a catch. Those 92W panels as they are sold, are not exactly usable by hobbists. The reason is the way they are packaged. Every solar carpet, has an open circuit voltage of 430V, and an operating voltage of 320V.

Even if we take into account the much more plausible Nominal Operating Cell Temperature -NOCT- (most of the soler panels you buy state only the ideal STC which you never reach, this panels produce "only" 300V when used (around 72W in real conditions).

There few electronics that can handle this voltages, and all of them are prohibitively expensive.

The reason these panels are packaged and marketed like that, is that they are very convenient for big installations. Installing 20 of this in parallel will give you "only" 4,7A of power for a total 1,4KWatts and that means thinner cables. In such condition buying an expensive inverter that can handle 400V becomes more plausible.

I had one of this panels for a year now but never found the time to play with it. Because in the back of my mind, my idea was to somehow tame the beast and make something hobby usable (or wearable) out of it. I can proudly state that this aim is now reached :)

take the plunge with me.

Trying to find data about this panels is not an easy feat. The only technical specs, are here, and here in a pdf in japanese (but you can make some things from the numbers or the diagrams).

So the first thing one notices is that this solar panel is consisted from 4 seperate ones. Each one is 83cm long and 42cm wide. You can easily see the connecting tabbing wire. The company also states that these parts are easily cuttable. And yes thats the case if you use a sharp knife cutting through the plastic insulation. So we can seperate the panel into 4 parts and bring down the voltage to more manageable levels. Lets look what we can expect from one panel of (17.5Watt real conditions I dont state the standard one as they are non achievable)

Open circuit Voltage of 99,5V

Peak Operating Voltage of 74,5V

Peak Operating Current of 0,235A

things are looking much brighter indeed. But even these values are not exactly easy to manage in hobby level to make a USB or a car battery charger. There are ICs that can regulate down to 5V (USB) or 14V (car battery) (like the LM5008 ) , but thats not an easy feat, so I wanted to go further.

If you look on the technical data on their site, it states that each one of those 4 17,5Watt panels consists of 68different cells in series times 2 (in parallel), and that these cells are cuttable. Thats of course easier said than done. The whole panel is 0.2mm thick and most of this thickness comes from the plastic insulation. But dont dispair, after some experimentation I found an "easy" solution. But first lets get a better grasp how these cells are constructed

I use the japanese diagram because it gives you a much better understanding.

So the blue part on the top is the positive pole but you cant access it as it is in the micrometer range. But the wholes in the middle of the blue part

Every cell has holes on it. These actually serve the purpose of connecting the very thin film (micrometers thin) to the connecting tab "wire" in the bottom. The holes in the center are for the positive, the ones in the edges for the negative.

Those amber lines you see is the film seperating the two cells, Thats where you cut. But be carefull. You can see amber lines from both sides of the panel, you have to chose the ones from the top. The amber lines on the bottom are seperating the "tabbing wire" from each cell and are not in the same position as the amber lines from the top. Every tabbing wire on the bottom rests between two cells and actually connects the middle holes from the one (positive) to the negative holes from the next cell (negative), thus in series connection.

So after you cut it carefully with a sharp knife, you have to seperate carefully the layers on both sides in order to get access to the negative and positive poles of your panel. You dont have to seperate the whole length, just a tiny bit is enough. Use a plastic knife in order not to scratch the amber film or the tabbing wire. The tabbing wire is so thin, that if you apply too much heat with the soldering gun, it will evaporate. I suggest using some tabbing wire for solar panels to connect the ends and not more than one-two seconds of heating :)

After that seal it back with silicone or hot glue and you are ready to go.

By cutting 16 cells together we can achieve . 17-18V x .10A
to .12A = around 2Watts (real life conditions) power. Now thats what I call manageable! By using an LM2596 board from aliexpress or ebay you can bring it down to 5V (USB power). You can also go higher and use 20cells (28V open circuit voltage, 20-22V peak operating voltage) and that means over 2Watts of power which will give you (after the conversion loses) around 2Watts (0,4A) USB power. Its not much but we are now talking about a very flexible panel 21x25cm :)

If you use both parallel panels (so 20cells x 2) you can get close to 1A of USB power for more hungry devices by having a foldable 44x25cm tube.

And if youd like to go small, you can use one of those boards based around the MP1584 IC.

During the next days I hope I will have some pictures from the first application :)

So I humbly declare this panel tamed :)

If you liked my instructable, consider voting me for the solar contest, I wouldnt mind having one of the solar bbqs at all :)