Home-made solar panels

Saw this link on hackaday. It sounds completely impractical, but still, a cool experiment and in the spirit of RepRap.

[scitoys.com]

[www.scribd.com] heres some ebook with lots of ideas about solar cells,some of it sounds bogus but not all
should you somehow get a watt or twou out of your cell you can try cpv tech, that is using large parabolic mirror to consentrate light on the cell, 2 suns make 2X more power that 1, you can easily consentrate 100+ suns on one point, the question remains how you are going to cool the cell, with copper cell it might not be too hard to put water cooling on the back of it. if you normally egt 1w out of the cell with 100 suns you get 100W, some imporvement huh?

Edited 1 time(s). Last edit at 06/05/2008 05:00AM by r2kordmaa.

... sometimes i was thinking of a cheap'n'dirty home-brew solar-cell, which didn't work photovoltaic, but instead with heat from concentrating solar-mirrors.

Simply connect two different metall-wires in respect of the electro-negativity (e.g. konstantan with iron) to make many small Seebeck-elments (=inverse peltier-elements).

Then mold-make an array of small solar-fokussing mirrors (like an egg-holder) with smal holes in the centre of the cavity.

Then insert a bundle of serially connected Seebeck-elements in every cavity, so that the elements are in the focus (best make a blob of black absorbing material around the bundle-tips).

Then place the same amount of elements behind the solar-array (in the 'cold' shade), then connect them so that you have the cold and hot elements communicating and connect this macro-elements again serially.

When setting this module in the sun, then the elements in the focus could heat until 300

VDX Wrote:
> Simply connect two different metall-wires in
> respect of the electro-negativity (e.g. konstantan
> with iron) to make many small Seebeck-elments
> (=inverse peltier-elements).

Seebeck elements aren't very efficient. What you really want in this design is a Stirling Engine.

One design I saw a while back used an old TV satellite dish (pre-digital type). Mirrors were placed on the surface of the dish, with a Stirling setup at the focal point.

There have been articles around for a while now about making transistor junctions with an inkjet-like printer. Perhaps this could be used to create a series of photodiodes?

... a Stirling engine is much more complex than a couple of fused wires and would wear off by time, so a solid-state system with no need to further support and maintenance should be better.

The poor efficiency of the Seebeck-elements should be a factor of the amount of them - combine 10x2 elements (10 hot + 10 cold) per solar mini-dish, arrange 20x20 of the dish-cavities in a panel of say 1x1 meter and then calculate if an 1x1m big photovoltaic element could compete with 4000 Seebeck-pairs with an estimated temperature-difference of 300

Viktor, what kind of overall efficiency do you think you could get out of the Seebeck elements? IE, with 1 kW/m^2 solar radiation, how much power could you reasonably put out? Even more importantly, how much is it going to cost to build up per watt?

I had a quick look at the wiki page on Seebeck effect, but I didn't see an easy way to calculate power output. In general, it looks like solid state thermogenerators generally run at about 5% to 10% efficiency, but I'm not sure if that's relative to the Carnot cycle or overall efficiency. With mirrors and whatnot, we're probably only looking a couple percent power generation. If it's cheap though, that might work for now.

Google is talking about building 1 GW of generation using solar heated stirling engines and some wind power; they figure they can get the cost down to near $1/W, using a lot of up front research and economies of scale.

I would love to see a reprappable solar array though - that would really bring the cost of solar down into the realm of economically feasible.

Wade Wrote:
-------------------------------------------------------
> effect, but I didn't see an easy way to calculate
> power output. In general, it looks like solid
> state thermogenerators generally run at about 5%
> to 10% efficiency, but I'm not sure if that's
> relative to the Carnot cycle or overall
> efficiency.

Those sorts of efficiency ratings are generally taken as a percentage of the Carnot efficiency given the delta-T.

... i didn't calculate for the efficiency ...

My idea was to arrange as many handmade two-wire-elements in a solar focus as possible, the same count of corresponding elements behind the mirrors in the shadow and then let it go ...

For me this would be the simplest way for a solid-statem, longlasting and completelly DIY solar-element, regardless of efficiency or whatever.

You need only some glass-sheets (old windows) as cover, the moulded 'egg-holder'-array with mirror-coating and the two coils of wire for the S-elements.

Then build it and forget it ...

***
Some years ago i (nonsmoker ) found a solar cigarette-lighter made from a 4"-plastik-solardish and a steel-spring in the centre for fixing the cigarette - worked very well for lighting twisted paper for firing our barbeque too

And now i'm thinking to reuse an old satellite-dish with 1,2m diameter, cover it with aluminium and set some dozens S-elements in the focus with the corresponding 'cold'-partners in a waterbath nearby and experiment with efficiency.

Here i have to track the sun, but i think it should be possible to receive nearly 1000

I like the idea, but I'm wondering if I want to run a reprap, how many of those hand soldered joints do I need to solder? 4000? Or 4,000,000? Cause if it's 4,000,000, I think I'll stick with $7/W silicon cells. Which may soon be $1/W anyway, check this out:

[www.colostate.edu]

Also, [www.nanosolar.com] is working on printing solar panels. Probably a bit beyond Reprap's scope at the moment, but a few years down the road, who knows?

Quote
r2kordmaa

[www.scribd.com] heres some ebook with lots of ideas about solar kit,some of it sounds bogus but not all
should you somehow get a watt or twou out of your cell you can try cpv tech, that is using large parabolic mirror to consentrate light on the cell, 2 suns make 2X more power that 1, you can easily consentrate 100+ suns on one point, the question remains how you are going to cool the cell, with copper cell it might not be too hard to put water cooling on the back of it. if you normally egt 1w out of the cell with 100 suns you get 100W, some imporvement huh?

yes there are plenty of good ideas.. I do have plan of installing these panels on my home roof and I do need these fresh ideas..

Edited 2 time(s). Last edit at 11/01/2014 04:10AM by DarrellBurrough.