On Mon, Sep 19, 2005 at 08:28:10PM +1200, Russell McMahon wrote:

Swapping energy sources is a project that I'm hot to work on right now.
With natural gas prices going through the roof here in the U.S. I'm
planning on adding solar to the package a kick NG out as my primary
heating fuel. So I've been doing a ton of research on solar.

> >>For example it's not clear to me at all that the best artificial
> >>way
> >>to collect solar energy over a large field is with solar panels.
> 
> Any method of collection solar input at the earth's surface has a
> maximum possible output of only 4 or 5 times what is achievable now.
> ie 20% versus 100%. Say 4 times as much would be "nice", but that's
> not so vastly much more that it would make all the difference,
> especially so if the cost was say twice as high per area.

I also seems to me that it doesn't make a lot of sense to cover the
ground with extremely expensive panels that can only receive 1 sun's
worth of light. Concentrating solar would be a much better return on
your investment.

Of course with panels the efficiency is flat terrible as outlined
below.

> 
> >>efficiencies of 10-18%, there is a lot of room to do better.  A
> >>bunch
> >>of mirrors for concentrating the radiation and a custom designed
> >>heat
> >>engine for the task should be able to do considerably better.  The
> >>theoretical maximum high temperature for the heat engine is the
> >>temperature at the surface of the sun, and the low is ambient air.
> >>The carnot efficiency should be pretty good.  Even if only half of
> >>that is actually realized, it should still beat solar cells
> >>handily.
> 
> Carnot efficiency is (Thot-TCold) / Thot (= deltaT/Thot).
> You don't need to get anywhere near solar radiation temperature to get
> respectable efficiencies. call Tcold 300k (27Cish) for nice round
> figures. At Thot = 1200 K =~ 930C you get 75% Carnot efficiency. 
> That's
> enough to start with. Even at say 1000C collector temperatures you
> have very nasty material problems.

All true. But there's still the question of if efficiency is the critical
attribute. We're pretty sure that the Sun is going to be around for a few
billion years more. Collection systems don't necessarily need to be ultra
efficient. But they do need to be cost effective. Measurements in system
efficiency should be made in $/MJ or $/kWh.

> 
> >Now you have a few choices for heat engine. Stirlings have NOT
> >caught on
> >well have they? Could it be the pricing? Or is it that to approach
> >high
> >efficiencies they need working fluids that are caustic and expensive
> >to
> >contain?
> 
> Sort of, no, sort of.
> Stirlings are doing very well thank you in niche applications. (cf eg
> www.whispergen.com * ). Costs are higher partially because volumes are
> low and applications tend to be specialist. WHEN someone gets some of
> the more major problems sorted out the Stirling engine will rule. I
> think Philips said something like a $US6 billion development effort
> was needed and a payback period after success of under a year was
> anticipated. Finding anyone who believes this enough to put up the
> $US6B(now more) is the trick.

On the global scale it's chump change. There are energy companies that
could pay that out of pocket. It's not in their interest though.

> Greatest problems with Stirling engines is sealing a moving piston
> with little or no lubrication at high temperatures and very large
> pressures using a has that loves to diffuse through steel. Lower
> temperature lower pressure Stirling engines using air as the working
> fluid work just fine. but energy output per volume is "modest" making
> them highly unsuited to portable/mobile applications. So people start
> to do thinks to increase energy density. And the fun begins. First
> swap the working 'fluid" to Helium or better still Hydrogen. Helium
> will hold 5+ times as much energy as air all things being equal and
> hydrogen about 14 times as much (!). Both also have other advantages
> over air BUT are better escape artists. Hydrogen especially will 
> escape
> through anything given time. At high temperatures and pressures it
> will happily diffuse through steel. Small gaps in rotating seals etc
> are even easier. Diffusion is not only not good for containment, it
> leads to 'hydrogen embrittlement" of metals. Needless to say, high
> temperature, high pressure Hydrogen is an exciting product to work
> with.

I'd trade some of that efficiency for reliability. As for rotating seals
part of the problem can be solved with free pistons which can be completely 
encased in steel with the only outlet being for the electrical wiring.

> Serious Stirling engine'ites try to run at pressures of 2000 - 4000
> psi (15 - 30 MPa) and temperatures as high as materials so far won't
> let them.

What kind of power are we talking about in the 500C-900C range. Experiments
have shown that temp differentials in that range are pretty easily achievable
in using parabolic, Fresnell, or Curnutt style reflectors, even for the
do it yourselfer. Not real expensive either. Check out this parabolic solar
furnace that an be built with little more than an low airtight drum whos
mylar top is stretched by creating a partial vacuum behind it. Build one
with an 8 ft diameter and you can focus up to 50 Suns worth of energy into
a relatively tight spot. The question is there a homebrew Stirling setup 
that can convert such energy into useful work?

> >Is there a LOW cost heat engine? One that can be made from existing
> >hardware
> >for example?
> 
> Absolutely. You can make a Stirling engine that runs on under 1 degree
> C differential out of plastic and cardboard. Carnot efficiency is
> "poor". You can also use eg an old (or new) Briggs and Stratton bottom
> end as the basis for a modest medium temperature engine. Stirling
> engines are not at all hard to make if you don't want a high energy
> density. Something akin to the old Lister  single lunger run for ever
> diesels is eminently doable. *AND*, because it is external combustion,
> it will run on about anything that you can get to burn.

I'd like to see some pointers for a real moderate power engine that a DIY
with some scrap equipment and a bit of tooling could put together. Something
along the lines of what the Phoenix Turbine Builders Club is doing for
Tesla turbine technology: http://phoenixnavigation.com/ptbc/home.htm

The closest that I've seen so far is a demo engine built from a Briggs
and Stratton motor. You can find it here:

http://www.geocities.com/Yosemite/Rapids/2068/?200521

It's become clear to me that building small solar concentrating "cannons" are
in the realm of possibility. But how to capture the energy for useful work
cheaply is still elusive.

[Some snippage on the Madras and other topics]

Any further pointers for DIY collection of concentrated solar to convert
to electricity would be greatly appreciated.

James, I read your water injection into the engine page. I'm unclear on
how such a setup can be used to gain motive force for real work. Care
to add some addition explanation?

BAJ
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist