Posted by Morris Dovey on January 6, 2009, 8:21 pm
I'm not aware of such a panel with a 4x10 form factor.
No again. I will assert that I've seen (and measured) considerable
variation in the shape of that curve - and, to the point here, the
/slopes/ of the rise and fall portions.
It would seem that the area where we don't see eye-to-eye is in the
realm of conversion efficiency - and I suspect that may be closely
related to our interest in very different applications.
I intentionally build panels with a (seasonally) variable efficiency,
optimized to produce both a maximum of heat in winter and a minimum of
heat in summer - and don't much care about either projections (as
opposed to delivered results) or year-round heat production.
I'm not denigrating evacuated tubes for use in concentrating collectors.
I'm working on a Stirling engine (which operates off a temperature
differential) whose hot head is a pipe at the focus of a parabolic
trough - and I've been wishing that I could use an evacuated tube in the
design (non-technical considerations prevent) because the bare pipe is
so incredibly lossy. In that application the ratio of trough width to
the width of the target area is greater than 100:1, and just the
reflected /visible/ light losses are horrendous. An evacuated tube would
seem to offer significant help, if not with the reflective losses, with
at least re-radiated and convective/conductive losses.
For space heating, however, I'm very much more interested in energy than
I am in temperature - and because lossiness is a function of temperature
differential, I work very hard to design and build panels that run at
the lowest temperature I can manage in order to deliver the greatest
amount of energy.
Originally, I tried to prevent/minimize the losses - and then (largely
as a result of discussions here) began to design to make use of the loss
mechanisms to /contribute/ to heat delivery. One example is that the
absorber is highly reflective and absorbs on both surfaces - and yet,
for all its reflectivity, it looks flat black (and it's 'black' in a far
wider frequency range than just visible light!) The question that turned
out to be key was "How much energy does a photon lose (yield) when it's
That led me to examine all of the loss mechanisms to see how they could
be used to /improve/ efficiency. The answers have been surprising - and
[ You can see the flat panels at http://www.iedu.com/DeSoto/solar.html -
the trough at http://www.iedu.com/DeSoto/Projects/Stirling/Heat.html -
and a concept drawing of the high-temperature engine at the bottom of
DeSoto, Iowa USA
Posted by azuredu on January 6, 2009, 8:45 pm
Hi, me again.
Interesting but I don't see how you can achieve this with reasonable
cost and performance.
I imagine. Do you know how 100:1 is difficult? You have to adjust
(align). If you use my method (http://wims.unice.fr/xiao/solar/
index.html) and build very carefully and with good material, 100:1
with sufficient accuracy is possible. However, watchout for the
gravity bending. If the trough is too long, this will put the tube
Have you read my article on how to insulate tubes in such cases? This
is not an easy subject, most people designing small troughs are locked
up by this one.
My recommendation is never try to use glass-to-copper vacuum seals. At
this size, it will give you huge disappointments.
Posted by Morris Dovey on January 6, 2009, 9:04 pm
I had no idea it was difficult. We built the trough and measured the
focal bright line at it widest point (3/8 inch). Then we divided the
width of the trough the width of that line. It didn't /seem/ difficult.
Why be "locked up"?
Ok - we weren't planning to do that, so (hopefully) we'll avoid major
DeSoto, Iowa USA
Posted by azuredu on January 7, 2009, 5:20 am
The bright line on the receiver is not a reliable indication.
I recommend that you follow the method described in my do-it-yourself
instruction documents, last chapter:
Only when you don't see the light escaping that you can say that the
precision is correct.
By the way, I know how to judge a trough from its image. Yours seem to
have important deformations near the edges. This problem is typical
for constructions using ribs. The middle portion should be OK, but the
edge portions are probably all lost.
A simple method: examine the trough from the front, and watch the
image of the receiver. This is a magnified image, which should be
straight and regular from any angle. If you see it become serpentine
at some place, then there are problems.
Posted by Morris Dovey on January 7, 2009, 8:53 am
The project objective is to produce an inexpensive (less than US$00)
engine that performs direct conversion from radiant solar energy to
mechanical energy and delivers at least one full horsepower.
At this stage of the development process, it is sufficient that the
reflector provide the minimum necessary heating of the engine's hot
head. Since that requirement appears to have been met in the "quick and
dirty" prototype, attention has moved on to other aspects of the design.
To more directly address your concerns - since the reflector is a
continuous surface (without discontinuities or isolated singularities) a
deformation would need to significantly widen the bright line before it
would be considered a problem.
The fine-tuning will come after there's something to tune. :)
DeSoto, Iowa USA