Club 80-90 Forums

Having joked with CovKid that making a solar panel to fit a space on top of his tintop was the sort of thing that he'd like to do, I realised that actually it's the sort of thing that I'd like to do. So I ordered some cells from www.solar-deals.uk, and I've started making a template for the substrate.

My plan is to make a flexible panel that will sit in the windscreen and work as a combined blackout screen/sunscreen/solar panel. If I don't happen to be parked facing the sun, or I'm parked in the shade, I could remove it and prop it up facing the sun nearby. I already have a 100W panel fixed to the roof of my poptop, but I don't let it dictate where I park or which way I face, so it doesn't always face the sun.

My first stab at templating the screen made me realise how unwieldy a single panel would be - it's virtually impossible to place it inside the screen due to the rear view mirror, the grab handles, the dash binnacle etc. So I'm going to make two half-panels (driver's side and passenger side), with a slight overlap, and a plug and socket to join them together. I'll probably add suckers to the corners to hold them in place, but I think the visors will do that sufficiently anyway.



Each half panel will have room for three rows of five of the cells (they're 125mm square), allowing for my stick-on sun visor across the top of my screen and the wipers at the bottom. A solar panel works by summing the voltage generated by each cell - the current through every cell will be the same since they're wired in series. The overall power, given by volts x amps, is limited by the current produced by the 'worst' cell. So if one cell is shadowed, it will produce a lower current and that will limit the output of the entire panel. So it's important not to place any of the cells behind wiper blades or sun visors.

My cells arrived today. They're rated at 3.4W each. 30 of them in series will make just about 100W. They're gorgeous things to look at - they have a velvety appearance and reflect virtually nothing. I guess that's what you want in a cell designed to convert sunlight to electricity - you don't want it reflecting the light it's supposed to be using.



I daren't touch the surface, tempting as it is, as apparently they mark very easily. They'll need encapsulating once the panel is assembled, with something like this, although that seems rather expensive for what it is. I'll look out for a cheaper supplier. I plan to use 1mm thick PVC sheet as my substrate - I may change that to 2mm if the 1mm feels too flexible.

Looks very interesting.

will be interesting to see how this goes, I didn't even know you could make your own panels!

Tony

in series ?
mm

I like the fact it doubles up as a sunshade.

[quote="CJH"] A solar panel works by summing the voltage generated by each cell - the current through every cell will be the same since they're wired in series. The overall power, given by volts x amps, is limited by the current produced by the 'worst' cell.
[quote]
Is there not a way around this?

Definitely fascinated by this, but I knew as soon as you posted joking about it, that this was coming... Suspect I wasn't the only one haha!

metalmick8y wrote:in series ?
mm

Smosh wrote:

CJH wrote: A solar panel works by summing the voltage generated by each cell - the current through every cell will be the same since they're wired in series. The overall power, given by volts x amps, is limited by the current produced by the 'worst' cell.

Is there not a way around this?

Yes, definitely in series. Each cell makes about 0.6V, so 30 of them in series makes 18V, which is quite typical of a solar panel. I must admit I don't understand it fully. I've heard before that a panel on a boat, say, can suffer because there's often a lot of rigging and masts about that can cast a shadow on at least one cell, so the entire panel suffers. But what I don't understand is why, if all the other cells are in full sun, does the one shaded cell drag the others down - why can't the ones making full power drag the shaded one up? I guess they're like a series of taps - if one is partially closed the overall flow rate drops.

I'm also not sure how you'd wire up more of these same cells to make a bigger panel, say 120W or 150W. Solar charge controllers, at least the one I have, don't seem to be rated above about 23V, and at some point too many of these 0.6V cells in series would take the total over 23V. I guess at that point you have to start adding strings of them in parallel, but I haven't looked into what the constraints are. To make a 150W panel could I, for instance, add a string of 15 cells in parallel with my 30 cells? The current would be the same, but the voltage in the 30-string would be twice that of the 15-string, so I don't see that working. I'm sure it's all explained out there on t'interweb, but I don't need to know it for my panel so I haven't looked.

yes hook together your strings of 30 in parallel and the voltage will stay the same so no issues, for some reason i thought you were linking 100+ in series
mm

metalmick8y wrote:yes hook together your strings of 30 in parallel and the voltage will stay the same so no issues, for some reason i thought you were linking 100+ in series
mm

Ha, no, just a modest panel to start with. But you're right - voltages have to be matched when joining strings in parallel. I found loads of useful information at www.alternative-energy-tutorials.com. There are two rules for joining cells (or panels).

1) The series rule I knew - joining cells in series sums their voltages, and the overall current is limited by the current provided by the worst cell. This image is from that tutorial site above:


2) The parallel rule is, quite logically, the opposite - joining cells in parallel sums their current, and the overall voltage is limited by the voltage of the worst cell. Again, from the site above:


So in my earlier example, trying to join a 15-string to a 30-string in parallel would be a total waste. The overall current would be the sum of the two strings, but the voltage would be limited to the voltage of the 15-string, so half the power of the 30-string would be wasted. In fact the total output of the 30+15 array would be no better than the 30-string on its own.

So that's important if you want to have more than one panel on one controller in parallel - match the voltages or you're wasting power from the higher voltage panels.

It also means that I need to tweak my design a bit to push the voltage up a bit to match my existing 100W panel on my roof. If I don't do that then I'll be wasting power from that fixed panel if I connect them to my controller in parallel. I'm going to try and squeeze 16 on each side, giving 32 in total with a voltage very similar to my fixed panel.

The voltage between your 2 panels will never be exactly the same in varying everyday operating conditions, due to orientation, cell type etc, so I doubt you really need an extra couple of cells. A series diode in your diy panel would a good addition though, so the one panel doesn't backfeed the other if one gets shaded (your bought panel likely already has one).

Thanks Mark, that's a good point. It's true that the two panels aren't likely to be facing the sun in exactly the same way. But what is it that varies with the amount of sun? I assumed it was mostly the current that varied. I did do a quick test with one of the cells pointed roughly at the low sun, and it gave almost the rated voltage but very little current.

Another thing to think about is the type of controller. For experimenting, and to avoid damaging my expensive MPPT controller, I've bought a cheap PWM controller off ebay. But ultimtely I'm planning to add this second panel in parallel on my MPPT controller. My understanding of MPPT is that it drives the panel at the voltage that gives the maximum power (volts x amps) in the current conditions. I don't know how it 'drives' the panel, but I would think that two panels in parallel will look like one big, higher current, panel, and it will have to drive them both at the same voltage. I checked my pack of 30 cells, and in fact there are 33, so provided I can fit them in it still seems like a good idea to try and match the voltages.

My existing panel does have a diode inside the wiring box, and it's on my list of bits still to buy for the DIY panel, along with encapsulation film, a flux pen, suitable connectors. It's not going to work out any cheaper than a ready made panel I think, but that's not the point