Alternatives to Solar Cells

[gizmo]

Hi All

I've moved house recently, onto a 6 acre block on the side of a hill. Unfortunately the hill faces north, and winds are very turbulent, mostly coming from the south east over the hill top, and the small HAWT I put up is jumping around like crazy trying to find the wind direction.

So I have a couple of options, VAWT or Solar. I do like windmills, and a VAWT would make better use of the erratic winds. But I also want to look into solar power, but would like to look into the feasibility of using methods other than solar cells to collect solar energy. Solar cells are just too expensive.

The roof of the house faces north ( I live in Australia, so north is good, plus I'll use metric units ), and has about 100 square meters of coverage. In Australia the solar power per square meter is approx 1000watts. So to get 2kw of power using 15% efficient solar cells I would only need 14 square meters of cells.

But what about other ways of converting the solar energy to electrical power. I'm thinking along the lines of a water heater, like used for solar hot water systems. These collectors can heat water to over 90 degrees C, almost boiling point ( 100 C ). They are relatively cheap and can be built in the average home workshop. Its not uncommon here to see a length of black poly pipe on someones roof, a cheap home made solar collector.  

Would it be feasible to use this heated water to power a large slow LTD ( Low Temp Difference ) engine, or stirling engine, driving an alternator. Yeah I know it would not be efficient, but if I were to use my whole roof area of 100 square meters, and want 2000 watts of power, the whole set up would only need to be 2% efficient.

The system would need a stirling engine, water pump and controller to kick over the engine once the water temp is high enough to use.

Or are there other ways of converting heat directly into electricity. Peltier (spelling?) effect devices are pretty expensive, how about a home made thermocouple?

Open for discussion.

Glenn

[fuzz]

Maybe this:

http://www.redrok.com/main.htm

[fuzz]

Or these:

http://muextension.missouri.edu/xplor/agguides/agengin/g01971.htm

http://phoenixnavigation.com/ptbc/articles/ptbc48.htm

[wooferhound]

What about using solar heating for heating ?

[gizmo]

I live in norther Australia, we dont use and never need heating. Thats like asking an eskimo if he wants a air-conditioner.

Its the middle of winter here and I'm walking around in short pants and a singlet.

Glenn

[commanda]

Thats like asking an eskimo if he wants a air-conditioner.

Only if it's reverse cycle :-)

Sorry, couldn't resist.

Amanda

[commanda]

I'm kicking around some crazy ideas involving trough reflectors and thermocouples. I've spent half the day googling & can't seem to find this bit of information.

I seem to recall from studies some 30 or 40 years ago that the potential difference generated between two dissimilar metals could be derived from the periodic table of elements.

Does somebody have this information bookmarked, or know what the magic terms are to feed google?

I did find somebody who makes a gas-fired generator in 2 & 8 watts, about the size of a shoe-box. Expand this out to several square metres and heat it with a solar concentrator is what I had in mind.

Copper pipe filled with plumbers solder, or pure lead, looks promising. Cu being 29, and Pb being 82 on the periodic table.

Amanda

[BeenzMeenzWind]

The links make interesting reading. One question though. Does antifreeze damage plastic tubes?

I've thought about making one as a supplemental water heater. It probably won't be able to provide all the heat on it's own, as we don't have a south facing roof. The one on the front is sort of south west though, so should work a bit.

[BeenzMeenzWind]

I tried Googling 'galvanic potential' and got a lot of the kind of thing you're looking for.

There is a table here: http://www.leadertechinc.com/EMIgalvanic.shtml

Hope that helps.

[BeenzMeenzWind]

I'd have thought the galvanic potential would be related to the material's utility in a thermocouple. Electro-chemistry is electro-chemistry isn't it?

I seem to remember all the really good thermocouples are made of like platinum, palladium and the like. I assume this is for corrosion resistance though, as most of the ones I've seen are exposed to gas turbine exhaust and the like.

[kenputer]

Glygo will not damage plastic tube,all the infloor heating system I install are filled with it they make it for that purpose.

[Arno]

Once upon a time, I read if you take a copper wire and a iron wire, twist the ends together and heat them, a current will flow.

Sounds cheap enough, so what if you made up a thousand or more, at least as long as the thickness of the insulated wall behind your wood stove, pushed them through non conductive tubes in the wall, wired them in series with half the joints out side at 32 degrees or below, and half inside at 80 or 90 degrees inside?

Is there anything usefull? What about doing all your walls? Isn't it great to dream?

Maybe when I wake up I'll check back to see what you all think!

arno

[bob golding]

think the biggest problem with peltiers and other low voltage junctions is getting the voltage up to something useful efficiently.

[bparks]

Nice table!  I've done a lot of searching about thermocouples too, the main question I haven't found an answer for is what are the variable that affect the amount of voltage & current produced at each junction.  ie what's the effect of larger or smaller gauge wire?  My though would be to use cheap foils, ie aluminum & copper foil, then build a stack, alternating foils.  Would probably need an insulating layer too.  I don't know if there's any benefit to the long junctions in this design, or does the size of the junction not matter?  Seems like this would result in a fairly cheap, simple & compact unit.

[BT Humble]

Its the middle of winter here and I'm walking around in short pants and a singlet.

Looxury!  It was -3C this morning inside my caravan.  

Still, you get used to it - I only troubled to put on a pair of ugg boots for my naked 5am stroll to the outhouse... ;-)

BTH

[elvin1949]

amanda

look for thermopile

later

elvin

[inode buddha]

Off the top of my head (no real research), I would divide the difference. Instead of only a big VAWT or a big solar, I would design something that uses both, but not as large. In other words the solar system would be smaller, and so would the VAWT; but they would both work together. The over-all "duty cycle" of energy fed into the system might increase, and that would be a good thing.

[bparks]

Ok, I found this

http://www.ferrotec.com/usa/thermoelectric/ref/3ref13.htm

that seems to be a better explanation of designing thermopiles to generate power.  I admit, I don't follow the whole thing though.  I think I need to get some materials & play around with the variables.  I guess the main thing I'm curious about is how the surface area of each junction affects output (if at all).

[whatsnext]

I think the problem with galvanic generation is that it is not RE. One metal galvanicly corrodes, which is an issue I would think. OT but, am I the only one who read about the Sterling/solar project that's showing something like 60% eff? Who needs PV or wind if that is true?

John............

[alcul8r]

Amanda,

try Seebeck with Effect, voltage or coefficient

Rex

[Ungrounded Lightning Rod]

No problem getting all the voltage you want:  Just hook 'em in series.  (They already are a bunch in series internally - you just need to hook several ASSEMBLIES in series to get up to the voltage you want.)

But getting enough power out of 'em is another matter.  I'll post about that separately.

[Ungrounded Lightning Rod]

Going after the heat - especially at a low temperature like the boiling point of water or lower - means you've got other efficiency issues.

You're using a heat engine - which is carnot cycle.  At high temperatures (like superheated steam or fire) you're going to be around 30% efficient if you're PERFECT, far under 20% if not.  At lower temperatures (like 120 farenheit or lower) you're going to be far lower.

Thermopiles are an option, especially if you use a concentrator.  A one-dimensional concentrator (a trough) gets things up to process steam temperatures, good for both heat engines and thermopiles (which are also heat engines, really).  Can't get much higher than that with a peltier or you'll destroy it.  (Two-dimensional concentrators, i.e. sun-tracking dishes, can get up to metal-melting temperatures and get a lot more of the energy out of a heat engine.  But then you're adjusting the aim by the minute rather than the month - which means moving-parts automation - and using something like a steam engine or a gas turbine - which means high temparatures and pressures, and thus mechanical hazards - even explosions - if you screw up.)

Peltier thermopiles have an additional efficiency problem:  They conduct heat from the hot side to the cold side, not just in the generation charge-carrier transport, but by good old thermal conduction of microscopic vibrations.  Not good:  It steals the bulk of your energy.  You can get it to work.  But you'll be far ahead to use solar panels instead.  (For starters you'll need a LOT of 'em to collect any significant power.  That make 'em more expensive than solar panels (which is why you still see solar panels rather than peltier panels with built-in trough concentrators.)

Homebrew non-semiconductor thermocouples and thermopiles have MUCH worse thermoconduction losses - which is why they're mostly used for instrumentation rather than generation (though I once saw a bunch of them converting heat from a waterpipe to enough power to run a very low-drain radio...).  They're SO bad that the resistive heating overcomes the thermopile cooling, which is why you didn't see thermopile refrigeration until the semiconductor version was developed.

There's a new technology coming down the pike:  Vacuum thermopiles, using nanoscopic "cold-emission" whiskers ("cold" in this case meaning "not glowing orange") to launch electrons across a gap.  This makes a vacuum tube version of the peltier cell, but with very little thermal leakage.  These are good enough that you could mount them on the front surface of your water heating system and pick off about as much electric power as, or more than, a solar panel.  Unfortunately they're still in the lab, and won't be available for some years, if at all.

So unless you want to do a concentrator and play with engines I'd say stick to photovoltaic panels until the new vaccum thermopiles, or some other breakthrough, comes out.

[commanda]

This article is about using Peltiers "in reverse" (the Seebeck effect) for generating power.

The worked example is particularly good. It shows that it requires 8 of the 6amp peltiers to generate 12 watts (1.5 watts output each). Solar panels, in Australia, have a best buy price of about AU$8.00 per watt. The 6amp Peltiers I can get from Oatley electronics locally for AU$13.00. This equates to $8.67 per watt, and doesn't include the cost of heatsinks, reflective parabola, or cooling. And cooling is not a minor engineering problem. 0.031 Watts/degC is a serious heatsink. I suspect the cooling problem is probably why there hasn't been much progress using these devices to generate power.

So, as a viable project, it's in the ballpark; it's all a matter of degrees. Getting the Peltiers at the right price is the trick.

Amanda

[Ungrounded Lightning Rod]

(though I once saw a bunch of them converting heat from a waterpipe to enough power to run a very low-drain radio...)

In this case by "waterpipe" I mean the kind with burning veggie matter whose smoke is inhaled after being filtered through water, not the kind in your walls feeding your sink and bathtub.  They got a fraction of a watt by using a crown of thermocouples on the pipe's bowl, where the red-glowing stuff was.

A big communal pipe made enough power for one of the smokers to listen to the radio on earphones.

[veewee77]

Perhaps the smoker just hallucinated the radio thing.. .. .. .. .

DS