"Designer" for solar reflectors
brian white
sun across the sky, the best shape for getting maximum concentration of
light over one hour, 2 hours, 3, etc is NOT going to be parabolic. In
fact, nobody knows what shape it will be.
Check 5 minutes into this video from Bill Gross for an idea. (It will
not be THAT complicated) http://www.youtube.com/watch?v=TSMzKg6fwJ8
So I thought some way to guesstimate it would be useful.
The idea is pretty simple. In my case, I want to design a reflector that
sends all the light to the target for 2 hours. So I use a laser to
simulate the sun at the start of the 2 hours and another one to simulate
the sun at the end of the 2 hours. As long as the light from both lasers
bounce off the reflector and both hit the target, the reflector is
correctly shaped. Basically, you set up the shape of the reflector one
strip at a time until you get the whole thing done.
I think very little work has been done on this (because the math is too
complicated) so if a few people make these little "designers" and play
around with them, I think very useful shapes and concepts will show up.
There is no "best" shape but perhaps close approximations will show up
fairly quickly.
The designer idea is at the link below and I will add to it over the
next few days Brian >
http://www.instructables.com/id/Method-for-designing-reflectors-for-unattended-sol/
Josepi
However, there are lots of lies in that presentation.
- After all the bullshit about the poor parabolic reflector design they end
up with a parabolic reflector, anyway.
- The other main lie is after $250,000 of research the end result is a
parabolic reflector with twelve tracking "pettles" for $3 each, with
tracking motors and microprocessors. Sign me up! I will even give them $72
for 100% profit. Not likely!
- Then the whole video ends up with a PV promotion for the worst inefficient
design layout I have ever seen. Too much blank space in a compact module.
Entertaining though.
"brian white" <br...@shaw.ca> wrote in message
news:B1LMm.12294$Sw5....@newsfe16.iad...
Robert Scott
avoids the problems of off-angle cosine loss as experienced by static flat PV
panels. However the off-angle cosine loss can eliminated by breaking the PV
panel up into smaller PV panels, about the size of one of the inventor's "petal"
reflectors, and motorizing each little PV panel, just like he motorized the
petal reflectors. It is still questionable whether the gain in power by
tracking is worth it. If a tracking collector costs twice what a static
collector of the same size costs, then I think I will just buy twice as many
static collectors instead. The overall off-angle cosine loss is not so great
that it justifies a 2:1 price difference. The real advantage in concentrating
collectors occurs when ultra-high temperatures are generated, which allows more
efficient thermal engines. But then you are comparing two different
technologies: PV and thermal engines. And that comparison is likely to depend
on the scale.
In the invention cited there is no great breakthrough in reflector shape design.
When a reflector is broken up into individual little pieces, then it does not
matter much what exact shape the pieces are. The fact that each piece is
individually motorized makes them all reflect light to the same place. And in
fact, as Josepi pointed out, they are still using parabolas.
Robert Scott
Ypsilanti, Michigan
dow
> The idea here is that if you line up a reflector with the path of the
> sun across the sky, the best shape for getting maximum concentration of
> light over one hour, 2 hours, 3, etc is NOT going to be parabolic. In
> fact, nobody knows what shape it will be.
> Check 5 minutes into this video from Bill Gross for an idea. (It will
>
> So I thought some way to guesstimate it would be useful.
> The idea is pretty simple. In my case, I want to design a reflector that
> sends all the light to the target for 2 hours. So I use a laser to
> simulate the sun at the start of the 2 hours and another one to simulate
> the sun at the end of the 2 hours. As long as the light from both lasers
> bounce off the reflector and both hit the target, the reflector is
> correctly shaped. Basically, you set up the shape of the reflector one
> strip at a time until you get the whole thing done.
> I think very little work has been done on this (because the math is too
> complicated) so if a few people make these little "designers" and play
> around with them, I think very useful shapes and concepts will show up.
> There is no "best" shape but perhaps close approximations will show up
> fairly quickly.
> The designer idea is at the link below and I will add to it over the
If you think about a little bit of the surface of a reflector, it must
act like a plane mirror. As the sun moves across the sky, the
drection in which sunlight is reflected by this little mirror *must*
change. So it basically impossible to design a reflector that will
reflect all the sunlight that strikes it to a stationary target over
an extended period of time.
What is possible, however, is to design a reflector such that *part*
of its surface will reflect sunlight to the target at some specific
time, then another part will do so at a later time, and so on. There's
a shape, which I know you have seen, called a "compound parabola" that
can be used to do this. Basically, this consists of two or more
parabolas side by side and fused together. The parabolas share a
common focal point, but are aimed in slightly different directions. At
some time, one of the parabolas exactly reflects sunlight to the
focus. A bit later, the sun has moved, and another of the parabolas
reflects sunlight to the focus. If the objective is to send sunlight
to an extended object such as a cooking pot, the compound parabola
will do this fairly well continuously for a considerable time. Light
does not have to be reflected exactly to the focal point in order to
hit the pot, so one parabola reflects sunlight to the pot for a few
minutes, then another parabola does it for the next few minutes, and
so on. However, at any given time, only part of the total area of the
reflector is reflecting light to the pot.
There is no reason, apart from simplicity, why the parabolas should
be side by side. They could be cut into many pieces and put together
in any pattern. However, the basic fact that only part of the total
area is reflecting light to the pot at any given time will always be
true.
dow
do...@27.usenet.us.com
> What is possible, however, is to design a reflector such that *part*
> of its surface will reflect sunlight to the target at some specific
> time, then another part will do so at a later time, and so on. There's
> a shape, which I know you have seen, called a "compound parabola" that
> can be used to do this.
The video shows something resembling an old Victrola audio horn that
focuses varying light angles to a focal point, some with multiple bounces.
--
Clarence A Dold - Hidden Valley Lake, CA, USA GPS: 38.8,-122.5
Robert Scott
>...So it basically impossible to design a reflector that will
>reflect all the sunlight that strikes it to a stationary target over
>an extended period of time...
That's right. Another way to see this is to think of the situation in reverse.
Suppose the collector was the source of the radiation and the whole sky was the
target. Where in the sky would the radiation from the collector go? By
symmetry, if most of the radiation from the sun that hit the reflector went to
the collector, then most of the radiation from the collector that hit the
reflector would have to point right back to the sun. In the case of a pure
parabola, this is exactly what happens. But of course it can't happen with a
moving sun. So if you are sitting at the collector and looking at the
reflectors, they could not all look sun-bright.
Robert Scott
Ypsilanti, Michigan
brian white
Decision made, game over!
And I am also dissapointed that people did not bother to go to the
bottom and check the link.
There are cusp reflectors that can take nearly 180 degrees of light and
get it to hit the target. (The target has to be a pipe in line with the
cusp). There are also winston w troughs that have large acceptance
angles.
All I am trying to do is revolve shapes like that into "dishes" or
"flowers" that work. You are not going to revolve in a circle (like
every parabolic dish in the world because we know a circular revolving
will not work.
The objections would be correct if the cooking vessel or target is a
point. Nowhere in the real world is the target a point.
It has area, it has volume. and this means we have leeway up and down
and left and right. Also note that I am not doing the same thing as Gross.
I am lining up the long axis of the "dish" or "flower" with the path
of the sun. That is crucial. There might be one flower or 2, I am
pretty sure there will not be more.
Brian
Morris Dovey
> I am a little disapointed that people have refused to look into it.
> Decision made, game over!
Eh? What were your expectations - that people would immediately inundate
you with 3-D renderings (or actual photographs) of the compound
reflectors they've built since reading your post?
That's not much of a rational basis for your snap judgement.
> And I am also dissapointed that people did not bother to go to the
> bottom and check the link.
Ok. I watched the youtube video and didn't find anything that I can use
in the context of the work I'm doing. Having reached that conclusion, I
didn't, in fact, choose to see more of the same. We have disappointment
in common, but for differing causes.
> There are cusp reflectors that can take nearly 180 degrees of light and
> get it to hit the target. (The target has to be a pipe in line with the
> cusp). There are also winston w troughs that have large acceptance
> angles.
Sure. Now please show me one that costs less other methods and that will
give me an adequate supply of air heated to 1000�F for 6 hours/day.
> All I am trying to do is revolve shapes like that into "dishes" or
> "flowers" that work. You are not going to revolve in a circle (like
> every parabolic dish in the world because we know a circular revolving
> will not work.
> The objections would be correct if the cooking vessel or target is a
> point. Nowhere in the real world is the target a point.
Agreed - the target in my application is a full inch wide.
> It has area, it has volume. and this means we have leeway up and down
> and left and right.
You might have that leeway - but I do not. I'm working with a 1"
diameter tubular target and any "leeway" results in lost energy.
> Also note that I am not doing the same thing as Gross.
> I am lining up the long axis of the "dish" or "flower" with the path of
> the sun. That is crucial. There might be one flower or 2, I am pretty
> sure there will not be more.
And so (why not)? Is there a cost tradeoff? A performance tradeoff? Is
there a point of diminishing returns such that two petals are good and a
third is wasted?
--
Morris Dovey
DeSoto Solar
DeSoto, Iowa USA
http://www.iedu.com/DeSoto/
Josepi
efficiency?
You have created a very expensive tracking system. Can't you point it?
Was this whole post only spam to promote your product?
Trough solar system with one very simple tracking mechanism will beat this
hands down for cost and probably output.
"brian white" <br...@shaw.ca> wrote in message
news:x5lNm.34212$6c2....@newsfe03.iad...
brian white
Best of luck with your projects.
If this does not apply to yours, that is fine.
You did not comment in the first round of comments
so I have no problem.
And I see now that a lot of people confused me with the video makers.
I took issue with much of the video too.
The important part for me was "The video shows something resembling an
point, some with multiple bounces.
-- Clarence A Dold - Hidden Valley Lake, CA, USA "
I just hope it is useful to somebody.
Ok, for the rest of you, it is NOT my video.
I only linked to the video so that people could get the idea that
dishes and troughs are not the ONLY option. People do not hear me
because I am nobody. Gross is the opposite and people listen.
His 6 flower thing probably cost about $50,000 to design. A 2 flower
thing designed with my method might cost less than $50 to design.
The trumpet flower thing with 6 horns was what I was trying to show
people. (I think it is very overbuilt, by the way)
I knew most people would not be interested. (My post is solar cooking
related and very few here are interested in solar cooking).
Currently parabolic solar cookers are been paid for by carbon offsets
and given away in villages in India. People even get paid to use them.
But there are problems because they have to be adjusted a lot and people
complain about getting flashed by the light in the eyes.
I was just hoping that 1 or 2 people might take a look, and try out the
designer to see what they came up with.
(I did not expect replies for a few days).
A 2 trumpet flower system with a 2 hour acceptance angle would need
adjustment only at the start of the cooking period and there would be
minimal flash risk.
Gross used different parameters than I use for the trumpet flowers.
Designs made with the set up that I show must be alligned with the path
of the sun so most likely 2 trumpet flowers would be the outcome.
Brian
brian white
It might be possible to apply the horn flower for solar cooking
applications.
And I am not selling anything. I have made a simple method for
designing the horn flowers. (Different flowers for different acceptance
angles).
And if you want to sell solar cookers, who wants to have a trough in the
yard?
Wouldn't people prefer the pretty flowers?
Brian
brian white
And if the reflector was collecting light from the whole sky. But it
isn't.
In my case, the reflector is going to catch sunlight from a 30 degree
wide segment of sky above the reflector and reflect that light on to a
target that will be at least 6 inches diameter. No other piece of sky
is involved.
I believe that can be done.
I have not got proof yet because i have only started to make my first model.
Brian
Josepi
larger target area in the zone and let the reflector get misalligned.
The horns look quite interesting but the multiple pettles are just too
expensive to make with so many trackers.
You gave the impression you were involved in the subject company in the
video. When people picked it apart you appeared defensive. The first part of
the movie and basis of your post was the parabolic shape was not the best
and then link to a movie that clearly defends the parabolic shape after a
bunch of nonsense.
"brian white" <br...@shaw.ca> wrote in message
news:K6oNm.39751$%j4.3...@newsfe18.iad...
brian white
silly. I guess I did not give enough info in my post.
Sorry for that.
For practical solar cooking, parabolic dishes give lots of power but
people are having trouble because they have to adjust them constantly
and they are getting flashed in the eyes way too often too.
I think that maybe the horns can eliminate both of those problems.
I believe that you only need 2 horns (not the 6 that gross shows). I
could be wrong of course.
Also, You could do the horns "dumbed down" with just 4 strips of
material to make a squareish horn. The key is to get reliable and
pretty constant heat delivery without moving the device for 2,3 or 4
hours. Parabolic dishes simply do not deliver this in a non tracking
environment.
It is just an option. Perhaps it is not good enough.
I was just opening it up to the general public.
Brian
Morris Dovey
> Morris,
> Best of luck with your projects.
Thanks. I like what you're doing - but felt that your expectations of
response were a bit much.
> If this does not apply to yours, that is fine.
:)
> You did not comment in the first round of comments so I have no problem.
Good. My personal focus is fairly narrow - and I didn't feel that I
really had anything worthwhile to contribute. I read, I looked, and I
kept my mouth shut, which is how I try to conceal my ignorance and/or
lack of inspiration. (The vastness of my ignorance amazes even me. :) )
> And I see now that a lot of people confused me with the video makers.
> I took issue with much of the video too.
I read it a little differently. I think it was Gross who was being
faulted for claiming that a (mathematically) interesting project is more
significant than it really is as a practical solution to a real problem.
> The important part for me was "The video shows something resembling an
> old Victrola audio horn that focuses varying light angles to a focal
> point, some with multiple bounces.
Yeah. That multiple-bounce part might an efficiency killer. Maintaining
the optical precision of a surface that complex - and providing
adequate reflectivity might present serious production cost problems;
and that will (I think) keep it out of reach for the people who most
need solar technology. :(
I kinda liked the independently-controlled petal concept (but not the
control system). Gross talked up what could be done with tiny motors and
micro-controllers - but I wonder how that system might behave in
breezy/gusty conditions, and how a user in (for example) Rwanda might go
about diagnosing and repairing a problem with their collector/generator
system.
> I knew most people would not be interested. (My post is solar cooking
> related and very few here are interested in solar cooking).
You might be surprised - but being interested doesn't necessarily mean
being involved. People here are working on all kinds of different
projects - but we all learn from one another. It's entirely possible
that you may arrive at some insight from which we all benefit.
I think what you're working on could be important to a great many more
than read this newsgroup - but that's really unknown until you produce
some improved result (true for all of us, not just for you).
> I was just hoping that 1 or 2 people might take a look, and try out the
> designer to see what they came up with.
> (I did not expect replies for a few days).
At least one person took a look and rough-drafted a program to create
and model a reflector (that may or may not end up looking like what
you've done). The math involved is mostly trig and, as Gross pointed
out, computers can do a lot of number crunching once there's software to
crunch /with/. I have enough on my plate to be able to guarantee that
it'll take me a _lot_ longer than a few days to know if I'm even on a
right track.
FWIW, I make a point of _not_ registering at sites that require me to
remember yet another user ID and password. Which means that I was
looking at your work through a figurative peephole - which (for me) made
understanding as difficult as it could be made. I don't blame this on
you, but it's something you might take into consideration.
do...@27.usenet.us.com
> FWIW, I make a point of _not_ registering at sites that require me to
> remember yet another user ID and password. Which means that I was
> looking at your work through a figurative peephole - which (for me) made
> understanding as difficult as it could be made. I don't blame this on
> you, but it's something you might take into consideration.
Those miniature pictures were a turn off for me.
If they were big enough to be able to see something, I might have been
interested enough to sign up for free, with a throw away account I would
never use again. But nonsense like that annoys me.
--
Clarence A Dold - Hidden Valley Lake, CA, USA GPS: 38.8,-122.5
Josepi
properly.
It's like bait and switch by the user himself.
"brian white" <br...@shaw.ca> wrote in message
news:lrpNm.45130$rE5....@newsfe08.iad...
dow
.
> The important part for me was "The video shows something resembling an
> old Victrola audio horn that focuses varying light angles to a focal
> point, some with multiple bounces.
> -- Clarence A Dold - Hidden Valley Lake, CA, USA "
The problem with "horn" or "funnel" solar collectors is that only a
small fraction of the light that falls on the wide end finds its way
out of the narrow end. If you trace the paths of light rays going unto
a funnel, you'll see that most of them end up being bounced out of the
wide end. Also, of course, multiple reflections cause a lot of light
to be absorbed by inefficiencies of the mirror.
If you don't like parabolic reflectors, I suggest you try a completely
different approach. Blacken your cooking pot, and put an inverted
glass bowl over it to act as a greenhouse. Try to use a bowl that is
only slightly bigger than the pot. Even without any reflectors to
concentrate sunlight, the pot in the greenhouse can get hot enough to
boil water or cook food.
dow
gaiatechnician
I did not know they required people to register now. You click on the
miniature pictures to enlarge them.
It used to be that anyone could see what was there. That was why I
joined.
gaiatechnician
I agree on the multible bounces. I have not finished my model yet but
so far everything is bouncing to hit the target in the 2 hour time
period. I have not been tracing the light path, I have used a laser
pointer to make the path.
The horn shape may be difficult or impossible for an amateur like me
to make but I am sure they could be stamped out very easily in a
factory.
Brian
gaiatechnician
I think the picture there might help.
For solar, maybe as extra input to the sides of a box cooker, I think
it has potential.
Especially if it is inputed near the bottom.
Also, note that in a solar collector situation, the curve will be
different because it might be designed to catch 30 degrees or so of
sun movement. You can constrain the design just to catch light from a
30 or 50 degree segment or whatever degrees of the sky if you wish. (I
think that is a sensible thing to do).
I know by now that most are not interested so I shall let it be.
dow
>
> - Show quoted text -
Being easy to make is less important than being easy to maintain. The
device might be stamped out in a factory, but the end user will have
to keep the reflective surface clean. Cleaning the inside of the
narrow tube that carries the light toward the target from each "horn"
would be a serious challenge in a third-world environment. Grease from
the food being cooked will get into these tubes, and will ruin their
optical performance.
Human-powered tracking systems, using simple parabolic mirrors, are
quite practical. There are always plenty of children around in third-
world villages. They could easily keep the mirrors aligned properly.
This was done in ancient Egypt, and could be done again today. (There
are a few places in Egypt where it is still done, for the benefit of
tourists.) No motors, computers or electronics are needed.
dow
dow
Actually, the Holmdel Horn Antenna has precious little to do with what
you want. It is designed to bring radiation (radio waves) that is
arriving from one particular direction to a focus near the narrow end
of the horn. The wide end of the horn is closed by a wall that is
paraboloidal in shape. The radiation enters through the opening in the
side of the horn, is reflected off the paraboloid, and is focused as
it travels down the length of the horn. It is *not* reflected off the
side-walls of the horn, which are there only to give mechanical
support. Similar antennas have used open lattice-work in place of the
solid walls. Many similar horns were used in the early years of
satellite communications, to send signals to the satellites and
receive signals from them.
dow
Frogwatch
Am happy to see that people have advanced far beyond parabolic
surfaces and into optimum concentrators and illumination optics.
For fabrication of such optics, you might consider "optical
replication" to make them cheaply (at least for reasonably small
optics). You start with a "mandrel" that is a male version of the
shape you want and then the optic is molded on the outside. The
mandrel can be re-used many times making the technique kinda cheap.
The mandrel can even be coated with any desired coating that ends up
being on the inside of the reflector surface.
For example, you can start with mandrels made by computer controlled
diamond turning. These mandrels can be coated with plastic to make
really cheap optics (quantities of dozens)or electroformed with
various metals to make nice metal shells anywhere from .002" up to .
040" thick. We make optics this way anywhere from 3" diameter
(maximum) and 6" long down to nearly microscopic ones. These are fun
to make and I am happy to make some for very little for Infra-red and
visible light.
Our money comes from making such optics for x-rays, not exactly common
and very expensive.
Morris: If you are getting into designing non-imaging optics, you
might consider investing in "Opticad" which is a great ray tracing
program that already has compound elliptical concentrators and
compound elliptical concentrators coded as examples. It costs a few
thou but is really powerful.
BTW, I see that the new Edmund catalogue has some glass CPC optics.
brian white
I put 2 videos on youtube to further explain what I am trying to do
here. The second one shows the basic shape of the reflectors. Just a
mock up because I have not yet figured how to hold everything in place
as I "design" the reflector. I think the real reflectors will be more
"scallop" shaped going to tubular towards the middle.
The videos are
http://www.youtube.com/watch?v=NFjuoDVU4c8
and
http://www.youtube.com/watch?v=xbiVo12kuKA
It helps to watch the first one but perhaps all you need is in the first
5 minutes of video no 2
Brian
brian white
I put 2 videos on youtube to further explain what I am trying to do
here. The second one shows the basic shape of the reflectors. Just a
mock up because I have not yet figured how to hold everything in place
as I "design" the reflector. I think the real reflectors will be more
"scallop" shaped going to tubular towards the middle.
The videos are
http://www.youtube.com/watch?v=NFjuoDVU4c8
and
http://www.youtube.com/watch?v=xbiVo12kuKA
It helps to watch the first one but perhaps all you need is in the first
5 minutes of video no 2
Brian
brian white
brian white
Anyway, I have been continuing slowly with the project.
I had some success making the shape with little square mirrors.
There are 3 links for anyone wanting to follow it.
http://solarcooking.wikia.com/wiki/Solar_design_T-Square
with basic explanation and older pics (and changable by anybody)
Please check out the animated gif for a procedure.
http://solardesign.ning.com/photo/albums/solar-design-tsquare (more up
to date but less text to explain what is going on)
and http://www.youtube.com/watch?v=W6N2IGcVHrg
(video of the little models). You can probably see the shapes best in
the video. (It was hard to see depth when I photographed the mirrors).
Anyway, it does seem to work.
If anyone tries it, be aware that with the cheap little laser pointers
the laser is not in line with the barrel! If you make a hole for it in
a piece of wood or metal like I did, you just have to twist it round in
the hole until it is pointing straight down before making any other
adjustments. It might be worthwhile using small laser levels instead to
avoid this hassle.
brian white wrote:
> The idea here is that if you line up a reflector with the path of the
> sun across the sky, the best shape for getting maximum concentration of
> light over one hour, 2 hours, 3, etc is NOT going to be parabolic. In
> fact, nobody knows what shape it will be.
> Check 5 minutes into this video from Bill Gross for an idea. (It will
> not be THAT complicated) http://www.youtube.com/watch?v=TSMzKg6fwJ8
>
> So I thought some way to guesstimate it would be useful.
> The idea is pretty simple. In my case, I want to design a reflector that
dow
> Most people will not be interested because this is solar cooking related.
> Anyway, I have been continuing slowly with the project.
> I had some success making the shape with little square mirrors.
> with basic explanation and older pics (and changable by anybody)
> to date but less text to explain what is going on)
> (video of the little models). You can probably see the shapes best in
> the video. (It was hard to see depth when I photographed the mirrors).
> Anyway, it does seem to work.
> If anyone tries it, be aware that with the cheap little laser pointers
> the laser is not in line with the barrel! If you make a hole for it in
> a piece of wood or metal like I did, you just have to twist it round in
> the hole until it is pointing straight down before making any other
> adjustments. It might be worthwhile using small laser levels instead to
> avoid this hassle.
I like the idea of using a lot of small plane mirrors instead of a
large mirror with some complicated, curved shape. I think you should
plan to make this a permanent feature of your device, rather than just
a development tool. In "third world" situations, plane glass mirrors
are fairly easily obtainable. Materials for making curved reflectors
that will be easy to keep clean and will not corrode are not usually
obtainable.
I am worried that you are looking at just two directions of incoming
light, simulating the directions of sunlight at the beginning and end
of a two-hour period, and are (I think) assuming that if the reflector
works for these two directions it will also work for all directions in
between. I suggest you should check if it does work adequately well
for intermediate directions. If it doesn't, you may have to make some
modifications.
Good luck....
dow
brian white
hard to make mylar or alu foil to a curve but relatively easy to put it
on flat material. Same as plane mirrors.
I made a full size clam shaped reflector (Just by scaling up a little
model) for a good sunny day back in Februry. It did work, about 3 hours,
and just one move of the reflector to bake some potatoes in a 7 liter
pot. If I had ribs on the back of the reflector to make it keep its
shape properly, it would have worked a lot better. (I still have not
done the rib thing). I checked that it would also work for the
intermediate directions when I made the model
I did make one model where the curves of the primary reflector and the
secondary reflector met too soon and it didn't reflect all the
intermediate directions to the target.
The good thing about clay is that you can easily readjust the model!
I have some videos of the cooking, etc up on youtube. I was lucky to get
some sun in February, usually here it is dull till late March.
Brian