Introduction: "Compound" Parabolic Solar Cooker, Make a Template With Technical Drawing

Compound parabolic dishes have a super advantage over parabolic dishes. You rarely have to reposition them as the sun moves. So you can leave them unattended for long periods!
There are many possible compound parabolic dish configurations and rather than me try to satisfy everyone, I though it would be better to give you a method to make your own petal templates.
Then you can make a 1 hour, 2 hour 3 hour or 4 hour compound parabolic solar cooker.
Whatever you want is within your grasp! You can easily make petal templates for parabolic dishes with this method too.
For 12 petal templates the error in the method (because you are using flat material to make a surface curved in 3 dimensions) is about 1%. I believe that is acceptable.
If you do not want to read the whole instructable, just read the image notes on the first image.
It should suffice for many people.

Step 1: Compound Parabolic Concentrators, What Are They?

The typical compound parabolic concentrator is a fairly old device, They are shaped like a bicycle light or a tulip flower with a hole at the bottom and as long as the sunlight enters within the acceptance angle it gets reflected all the way down the concentrator and gets trapped there at the bottom. They are made by putting the focus for 2 identical parabolas on the curve of the other one. To do this you have to twist each parabola (which gives you the acceptance angle) and the resultant 2 dimensional shape is spun on its axis to get the flower shape of the compound parabola.

Many times the reflectors for box ovens are based loosely on this type of concentration. They are good because if you face the box cooker the right way it stays collecting sunlight well for a long time.
But I wanted something different. I set out to apply the same principle to parabolic dishes.
Parabolic dishes are great because they can focus a lot of energy but they quickly go out of focus and have to be readjusted. There is virtually nothing out there about applying the same principles to parabolic dishes so I had to start from scratch. My first compound parabolic dish was a great success but not many people were interested in copying my construction method. They prefer a template.

Step 2: How to Make the Compound Parabolic Shape

easy way to make a compound parabolic shape. You have a parabolic shape already with its focus.
All you need to do is rotate it as shown and you have the compound parabolic shape!
45 degrees acceptance angle gives a 3 hour solar cooking dish. The axis of the dish will be through the 22.5 degree line. Half of the acceptance angle.

Step 3: 45 Degree Compound Parabolic Petal Template

The first 2 images show the figures to make your own and the shape of the particular template.
Scale up until you get the right size.
Figures that are important for this template scaling are:
6.5 cm from top of the dish to the focus
The height of the dish is 11.7 cm and
the radius of the dish at the bottom is 12.3 cm
The cooking pot should fit in the scaled up 6.5 cm space between the focus and the top of the dish when the solar cooker is in operation.

Step 4: A Little About Compound Parabolic Dishes and Parabolic Dishes

For now I am just leaving an animated gif here to show the differences between parabolas and compound parabolas for dishes. Please note that this is very much a shot in the dark for me. There are likely to be errors (but not substantial ones) in my explanation. Eventually others may confirm or improve on what I explain but I wouldn't hold my breath. Please go with this until something better comes along.
Note. Parapolas are like circles in that there are large and small ones but the fundimental shape is exactly the same. Unlike circles, nobody has ever seen a complete parabola. (the ends go on infinitely).
Compound parabolic dishes can have their curves come in and meet up. I show the MAXIMUM EXTENT for useful curve for a 30 degree (green), 45 degree(blue) and 60 degree(yellow) compound parabolic dish in the animated gif. Note of course that making the maximum extent would be costly and you can just use a cost effective portion nearer the focal point just as people do with parabolic dishes!
For my compound parabolic dishes all I did was twist one half of the parabola round its focal point.
This concentrates the light in the area between the back of the dish and the focal point as long as the incoming light is within the acceptance angle of the dish.

Step 5: How Do We Test a Compound Parabolic Shape for Effectivness?

This is part of my continuing attempts to understand what I am doing!
Below are pictures of a 60 degree compound parabolic model that I made around January 18th 2009
As you can see, depending on the angle at which the picture was taken, the reflection of the apple or little ball was very different. (The apple happened to be closest to the size of the distance between the focus and the bottom of the dish). I think the photos are a clue to how effective the dish can be.
I also used a lazer spot to test its effectiveness.
I have not done anything scientific just yet but it is a start! As you get closer to directly over the dish, you get an increase in the size of the reflection that probably coresponds to how effective the dish will be when the sun is in that posititon.

Step 6: Known Issues

I. When I rotate the parabola around its focus for say a 45 degree compound parabolic dish, I am left with a 45 degree"gap" which I just fill with and arc of a circle. But an arc is probably NOT the best way to fill this space to reflect the light towards the cooking vessel. A line or perhaps a dimple pointing at the focus might be better and will allow for greater concentration of light and a smaller cooking vessel.
A circle is good enough for me for the moment. I am sure someone will find a more accurate solution in the future. This will allow for more concentration of energy and larger reflector to cooking pot ratios while maintaining the long unattended cooktimes.
2. I do not know exactly how the light reflects off the dish as the sun moves across the sky.
Some light may miss the cooking pot when the sun is not directly over the focus.
3. A large (2 square meter) stationary dish (perhaps 60 degree compound parabolic) may be suitable for same time every day cooking all year long if aimed at the equinox sun at its highest point.
but I do not know how it would behave when the sun traverses 23 degrees above the aimed direction (high summer) and 23 degrees below, (mid winter) and the tropical equivalents of these times.
This awaits software modeling or small physical reflector models and laser tests.
Thank you Brian White

Step 7: Other People's Work

Other people have traveled the compound parabolic road.
Anyone who designs a flashlight reflector or lampshade reflector to spread out the light from the bulb is doing something similar, just doing it in reverse!
A guy called Roland Winston had patents for compound parabolic troughs. The patents have now expired and are online. He is a well respected scientist. He did things probably the perfectly mathematically correct way so I will put one of his patent drawings here. I found it online on about the 14th of January 2009. (Pity I did not see it earlier)
Here is a picture of one of them from the patent application itself! As you can see, he has a bump or dimple under the receiver. Once again, I do not know how this will work when we make a dish.
He seems to use 2 parabolas without a combined focus. Please take a look.
I do not know if I am allowed to show the entire patent (I have it in pdf format on my computer) it is easy to search for it, if you just put in the patent number and winston in your search.
As you can see, it is fairly similar to what I came up with but it is a trough not a dish.


curtlezumi (author)2015-03-13

Regarding the cone by Ronald Winston: The dimple at the bottom
allows the heated item to be inside the cone and heated from all sides.
If the target was at the bottom of the cone, some heat can conduct out
the bottom.

Another key part of the design is to have the light
reach the target with just one reflection or bounce. With each
reflection, a small amount of light is lost, this adds up relatively
quickly so ensuring only one bounce is required over the acceptance
angle provides the best heating. If you look closely at the angles,
you'll be able to see that a ray just grazing the upper edge of the cone
that hits near the dimple will be reflected to the target with a single

Finally, I've heard that using a string and pencil will
also allow you to trace parabolic cones like the ones in the patent. One
end is fixed to your target, and the other end the other end represents
the light source. It may require some playing around to get curves that
look right and you may need to reposition the 'light source' end to
keep the incoming 'rays' parallel (Just keep the light source end the
same 'hight' from the target otherwise the curve will be wrong). You can
use this technique to create unique collectors, I saw one that was
almost like a nautilus seashell.


Hi, I have gone further along since this to the solar design T-square and clam shaped collectors. You are not correct about 2 bounces. Sure you lose a bit of power (to the dish metal that will re radiate it) but the gain in acceptance angle can be worth it. It is a trade off. One bounce, move the dish lots, multiple bounces, move it less. The only way to have a dish collector with a greater acceptance angle than a parabolic dish and a decent heating time between moves is to have more than one bounce. I wanted a dish for solar cooking. The main trap that every researchers falls into is that they make dishes with a wide acceptance angle in all directions. But for solar cooking you only need a collector that you aim at the sun and it accepts for about an hour. In an hour the sun moves (like a spotlight tilting) in a very specific way so a dish should have its acceptance angle in line with that tilting. So all the designers should be taking this into account. Not a single one does, except perhaps the designers of the butterfly and papiollon reflectors. I pointed this out 3 or 4 years ago and nobody "got" it. One of the emperor has no clothes type statements. So flipping obvious that only a child would get it. I made 2 models with a "claymation technique" and scaled one up to a full size reflector (something like 1 meter primary collector. I cooked potatoes in a big pot with it in february. Aimed it at the sun just once. I have pics of the cooked potatoes but they are not on this computer. After that I gave up. Nobody is really interested. Everyone comes back to parabolic dishes and anybody who solar cooks will tell you that parabolic dishes are not good for solar cooking because you have to attend to them all the time. And it is dangerous to be out in the heat like that.

pdrodrig (author)2012-04-18

Hi all,

I read carefully all information posted by gaiatechnician.

I found this webpage that might be interesting for those of you reading this post:


static (author)2011-07-24

I was wondering if you experiment with the compound parabola. Where I live if there is sunshine, on any given day there is enough of it, there's no need to go with extra ordinary measures. A blessing in Winter, a curse in the summer.

gaiatechnician (author)static2011-07-24

I have done hardly any experimenting this year due to workload. I gave a better and far more useful reply on your other comment, check it out.
Thanks Brian.

mscir (author)2011-06-01

Great article, thanks!
Please post the patent number for the work you mention by Roland Winston, I'd like to see what he did.
I wrote a freeware program that calculates parabolas of any depth and diameter, you can see or download it here . I'm thinking of writing a calculator to help people make compound parabolic collectors as you've described in this article, they are extremely useful because they don't have to be aimed as parabolic reflectors do, looking at his patent might help me get a few ideas about how to approach it.
Thanks, Mike

gaiatechnician (author)mscir2011-06-01

Hi, Mike, glad to hear from you. I gave a talk in February and in the research for it I found a very good pdf about Winston and other curves. (All seem to be for troughs). Anyway, several might be good, winstons with a 47 degree acceptance angle, (horizontal and aimed at the equinox sun would work), and there are 2 "seashell" troughs, and the trombe minell cusp. I don't remember the patent no. but if you send me your email address, I can send you the "interesting curves" PDF, etc.
It contains some of the math that you need.
Anyway, the seashell troughs are completely new to me but they help show that I was on the right track with my "clam shaped" solar cookers. (Even if I only ever made one).
Have you seen the solar design T-square? Thats what I used to make the clam shaped design. Maybe you could do a better design and more repeatable design with your grasp of mathematics? Brian

User113 (author)2011-03-31

Patents themselves are public domain, you can always copy and publish them (the documents themselves), even if they have not expired. The idea behind patents is that by telling others how to make and use the invention, they (the others) will be able to use that knowledge to improve other things. There is an incredible amount of information residing in the U.S. patent database, just waiting to be rediscovered and/or reapplied.

However, if the patents have not expired, you just have to be careful about making, using, offering to sell, selling or importing anything that is covered by the patent's claims. All modern U.S. patents have claims. See 35 U.S.C. 271.

nrevelis (author)2011-02-28

would like to make one to accept a 9.5 in. pot .Just scaling up approx 49 in petal. can scale up be used or do i have to try to start from scratch.

nrevelis (author)nrevelis2011-03-02

I misspoke I intend to take half of template and cut a curve as in the template using 3/8 ply. then duplicate this curve 6 timesto create a half circle. On half circle i will place formica to make a crved half circle,affixing formica as needed with fir strips for backing to hold curve and half circle, repeat for other half making of course a ltittle overlap to connect the two half circles . My computations using your template shows me that this will create shape that is approx 31 1/2 inches in diameter at top and approx 48 inches in height along centerline(focus line ). will line surfaces with mylar. Will also affix a swivel platform for pot and add a side door opening to allow access to food pot . Mount side brackets to allign parabolic to sun and put on a swivel base for any other adjustments to alignment. It seems that this shape is very close to what you would make with wet clay. If you take this concept further you could use these curves to make your cob dome. My real concern is have i done the scaling right. . all i did was measure and expand by 10x. Surly do arrpeciate ypu taking the time to respond. I am a almost total novice . Have built large solar oven that gets to 400 on a nice day . This should have about 800 sq in surface and be suitable for cooking . Question "is this doable?

gaiatechnician (author)nrevelis2011-03-02

I think you should check out and see if it suits your needs.
If you make "petals" scale them up to be similar size to the ones I used there and there should be no problem. I collected a lot of details about that solar cooker and it worked pretty good. The pot was a similar size to yours. (Fits in an oven turkey bag). I forget exactly what size it turned out but it was more than 800 square inches. The second version (just with mylar without a backing (and an attempt to hold it open like an umbrella) did not work so good. The mylar must be stuck to a rigid surface for best results.
I hope that helps  
I still think you should try the cone solar route, it is newer, less traveled but has great potential.

nrevelis (author)gaiatechnician2011-03-08

took your advice googled winston cone came up with a drawing of cone had it blown up 10x and am in process of making one not being a techie i would not even try to understand the math but if a guy like winston says it works i will take his word for it. if i am able to get it done right i will let you all know how i did it. nick

gaiatechnician (author)nrevelis2011-03-08

Just be careful that the drawing is correct. Some of the winston drawings I saw on the net were scanned in and one of the dimensions got changed somewhere along the way.
Yeah, the math is scary hard. I avoided it too.

gaiatechnician (author)nrevelis2011-02-28

Hi, Thanks for the comment. In my world, I get sidetracked a lot and I have been off this path for a while. From this I went to clam shaped reflectors. However I think these compound parabolic dishes are probably a bit better than parabolic.
I suggest that you do not use the petal method.
A guy called Magnar suggested I look into something that I have been calling cone solar. My cone solar project got stalled for the winter but there are several videos on youtube about  it.   I certainly do not claim to have invented cone solar but currently videos about it are very thin on the ground.
So, I suggest you make the elevation and plan of your proposed dish,  with a circle or ball representing the size of your pot and work from there to design your cones for it.
I made this playlist of videos  just for you!  I hope it helps. I will add videos from other people as they become available and probably remove mine.   Best of luck   Brian

Kwena Moxadi (author)2011-02-15

Hello, I'm a first year student for engineering and we've been given a project to design and construct a solar cooker. I'm finding it hard to follow these steps because I'm not familiar with all the technical steps. please summarise these steps in simpler language for me to follow, starting even with potential material to be used. I'd appreciate it so much if you do. pictures not necessary.

Hostlerguy (author)2010-05-14

Nothing in your instructions showed how you made the base for the reflective strips..... There's a huge leap from drawings to finished bowl to glue reflective strips to


Yeah, sorry. I made a template and rotated it in the middle of the bowl filled with wet clay. The reflective strips were stuck on by squirting a plant mister on the clay and sticking the mylar directly on to the wet clay.
It was really more to show myself and other people what happens with off focus paraboloids and other shapes.  YOU are the SUN and the reflection you see is the same size (pretty much) as the area that reflects onto the target.
I spent ages trying to do this in software. It ended up being the basis for the solar design T-square idea. 

robbtoberfest (author)2009-01-15

Is the energy less intense on the focus than a standard parabolic reflector because your sacrificing direct aim at the sun?


Yes, the energy ends up in an area behind the focus. So you put your cooking pot in that area to catch the heat. Ideally you make the cooking pot the same size as that area. The important points are that although it is less intense, it is (almost) the same amount of energy. And in my view, "sacrificing direct aim at the sun" is no sacrifice at all! I discovered last summer that keeping a parabolic dish automatically pointed exactly at the sun is really really hard! No matter what you do, the seasons can screw up your parabolic tracking device or at least add substantially to the cost. Having a device with virtually the same power ( without the need for exact pointing at the sun and constantly readjusting its aim is a super advantage). It means that cheap tracking (the clock based tracker or even the water clock tracker) might be good enough to keep food cooking or water pasturising all day. Brian

About This Instructable




Bio: I am a stone mason. My hobby is making new solar cooking and gardening stuff. I have used solar heat to cook soil for a ... More »
More by gaiatechnician:Trompe and airlift pump model for schoolsDIY Constriction Airlift Pumpswater flea water filter for fishpond!
Add instructable to: