Making and Using a "compound" Parabolic Solar Cooker (Large Project)




Introduction: Making and Using a "compound" Parabolic Solar Cooker (Large Project)

About: 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 couple of years. In mother earth news in January, i read that their compost expert does t…

Best shape for solar reflectors?
Parabolic dishes to a point ONLY when they are pointed EXACTLY at the sun. But the sun moves and that point quickly spreads all over the place unless your are watching over your solar cooker all the time.
My conclusion after a couple of months of working with a prototype is that a parabolic dish is the wrong shape!
This instructable is about the shape I found to replace the parabola and how you can use it to make a simple reliable "Compound" parabolic solar cooker. Mine will focus the heat on your cooking pot for up to 3 hours without moving it! Here is a video link to help too.
You can skip to the 7th page if you just want to see a template for the cooker.
Use the template to make "petals" from any sheet material.
Then tape the template together with masking tape or duct tape and stick your reflective coating on it.
Finally a use for corrugated election signs when the election is over!

Step 1: Problems With Parabolas!

Why I needed a new shape
The tracking solar accumulator
It was a solar cooker that consisted of a pot of material and a parabolic reflector that turned slowly to track the sun on an equatorial mount. This parabolic reflector was not the bottom of a parabolic bowl as most reflectors are but rather, a section from the side of the bowl.
I did this daily tracking with a low tech tracker that I made.
It was just an adjustable waterclock (also on instructables)
The sun moves higher and lower in the sky with the seasons and I intended to do this adjustment twice a week to keep my dish pointed exactly at the sun and the light exactly on my pot of food. I had to make a "mechanical mathematician" for making my section of a parabolic dish, and I had to make a low tech tracker too.
One problem was that every time I adjusted for the sun's change for the season, the center of gravity changed and I had to readjust the balance of the dish. Time consuming!
Another is that parabolic dishes are precision devices.
If you aim slightly wrong at the sun or if your tracking mechanism is slightly inaccurate, or even if you bump your dish a few times (as I did) you will not get the hoped for results.
This are some of the reason that I needed a more forgiving shape than the parabola!

Step 2: Searching for a Solution

I searched through the literature about shapes for passive reflectors and found Winston concentrators and cusp reflectors.
But the cusps did not concentrate much while the winston concentrators might be suitable for a box oven but not for what I had in mind. They concentrate in an area behind the compound parabola.
Eventually I decided that if I rotated a parabola round its focus by a certain number of degrees and joined the 2 inside curves It might work for me.
Using that shape focuses the light in an area at or below the focal point of the combined parabolas.
I have added image notes to the 2 pictures below to explain it a bit. Shape 1 is the parabola for a parabolic dish.
Shape 2 shows how to turn it to make a 45 degree compound parabolic dish (suitable for up to 3 hours of solar cooking without moving the dish!)

Step 3: Making the Dome

Deciding shape and size
I chose a pot, painted it black and used it in a model to decide on a good compound parabolic dish.
I saw that I could make a 3 hour 45 degree compound parabolic dish to shine the light on this pot.
so i set about making a mold for the dish.
First I marked out the piece of a parabola on a piece of wood and cut the wood to that shape.
Then I placed the wooden curve on a nail on a central axis.
I added rubble to under it to bring it close to the wooden form.
Finally I cobbed out the the wooden form all round.
Cob is a mixture of sand clay straw and water.
I had to let it dry for about a week.

Step 4: Making the Dish

Making the dish on the mold.
I had some old sign corugated plastic and I decided to use this for my first dish.
First I cut it so that it would fold down somewhat over the mold.
I held it down with rocks and the sledgehammer.
Look near the sledgehammer. See how the cuts overlap when it gets held close to the dome?
Wherever they overlap, you cut along again and remove the undercut.
Then you hold it all down with rocks or whatever and tape the sections together.

Step 5: Adding the Mylar

Completed dish must be made reflective
I did not have enough material to cover the entire dome but who cares?
I cut pieces of mylar to cover the inside of the dish. I glued them on with potato paste.
I used big or small pieces depending on the curve. Smaller in the more curved sections to avoid wrinkles.
I let it dry and then tested it.
I hung a 7 liter pot over it and adjusted the dish until I thought it was pointing at the sun.
Later I put a turkey bag over the pot to conserve heat.

Step 6: Cooking!

I heated water and steamed soil water in the big 7 liter pot.
Steamed soil is better for seedlings because it has no bugs and weed seeds are killed by the steam.
For cooking i used a little brown ceramic pot.
For a test I stewed some apple.
I forgot about it and it boiled over in the cooker.
The thermometer goes to HI when it is over 100C

Step 7: Conclusions, Template and Pdf of My Results

So, I have not made the new tracking solar accumulator yet.
What I have made is a much more usable stand alone solar cooker!
The compound parabolic shape means that it does not need to be pointed exactly at the sun. The dish has to be in an area from the focus to the back of the dish. In my case the focus is at 10 or 11 inches from the back so my pot is almost perfect.
Here is a template so that you can make your own.
I think if you make the template from plywood, then use the plywood to cut the shapes in mylar, you can easily make 12 segments. Then just tape them together at the back and you should have a umbrella cooker the same shape as mine.
Use thin wire to keep your umbrella open (Up to you to figure out how!) and start cooking.
Graphs of my results in 2 tests heating 7 liters of water in a pot are shown in the other photos.

I also include a draft pdf file so you can check out my results.
Please note that I have a new instructable that is devoted to making your own template using basic technical drawing skills or a computer design program. If you can read a house plan and elevation, you can probably make a template!

Step 8: First Attempt With the Template. Cheap Emergency Solar Cooker!

Simple cooker with the template
I decided to make an umbrella cooker even though it is almost October and results are unlikely to be useful. First I cut out the 12 segments, then masking taped them together. I cheated a little on the next stage, the wiring. First i duct taped along the rim so that the wire would have something substantial to stick to.
I placed the umbrella over the cob dome and made the rim of wire right there round it and then ducttaped over the wire. Then with the umbrella still on the dome, I cut 2 cross wires to go over it and bent them to hold onto the rim wire.
In hindsight I should have had 4 wires to hold it to shape!
I also redid my first compound parabolic with kitchen foil to see how that would work.
I will just put in some pictures to show the results (next page). I only got to about 85 C but as you can check, that is sufficient to boil potatoes and to bake them too!
Probably not many people thought of baking potatoes in a black pot! It works, even on Sept 29th.
My umbrella cooker was a first attempt and I did not in enough brace wires to keep the umbrella up. I just had the rim wire, and 2 wires criss crossing across the top. I think it needs 4 criss cross wires to be effective. Also, I had that tiny glass pot where I needed the big pot to grab all the light.
It still worked though!
Brian White 1st October 2008

Step 9: Cooking and Baking

Boiled and baked potatoes!
I boiled potatoes in the emergency umbrella cooker and I baked in the big black pot in the (now coated with kitchen foil) solar cooker I made on the cob mold.
I was not happy with the performance of the emergency cooker but it clearly worked.
Partly the small pot was a problem but also there were lots of creases that probably bounced light in wrong directions.
As you can see both worked on the 29th of September. The temperature in either solar cooker did not go much over 80C.
I thought it was innovative of me to raise up the potatoes for baking!
I imagine in future designs that the potatoes can be put on little metal skewers on the insides of "baking" pots so that the heat can go straight into the middle of the potatoes for excellent baking!

Step 10: The Future of the Compound Parabolic Solar Cooker?

Some little visions.
The compound parabolic solar cooker is part of my seeking after my holy grail.
A low tech, tracking solar cooker. I think it brings it way way closer!
But it is a lot more than that. It is a pretty good reliable and easy to make stand alone solar cooker.
Even the little one I made puts about 200 watts into the food. On an engineering forum, a guy calculated that it could pay for itself in electricity costs in 50 meals. (assuming the material cost 30 dollars).
He calculated that my 15 watt solar panel would pay for itself in "wait for it" Wait, WAIT ______
Wait 70 YEARS!!!!!!
If anyone can make the umbrella cooker a bit more close to the proper shape (without wrinkles like I had), the cost comes right down to under 10 dollars!
That could be a life saver all over the world when people go hungry.
The compound parabolic shape means that you can use pretty low tech tracking and still get good performance. The seasonal change in the suns altitude in the sky is 46 degrees.
That means that my cooker (if on a single axis tracking system), could cover almost the entire YEAR!
That is incredibly valuable. Accurately Tracking the sun is criminally hard to do.
With this little thing you do not have to be really accurate.
I also see this compound parabolic dish (in a half dish version) as being super useful for upside down solar ovens on tables. You have your tracking half dish under the table focusing the light through a hole in the table into the underside of the highly insulated solar oven.
So you can get higher heat and a greater transfer of energy too.
Some people think me weird to want to use it for soil steaming. I spoke on an online forum and some gardeners are quite liking the idea.
Including one with a tv show! Expanding the uses of solar cooking is good in my opinion!
Brian (Diagrams to come soon)

Step 11: Scaling Up

What happens if you make things bigger?
For instance, If you took all the lengths in the dish and the pot and just made them twice as big?
What happens?
I have checked some of it. And here are some guidelines.
The surface area of the pot increases at the same rate as the area of the dish.
But if you double the surface area of the pot you increase the volume by about 2.8 times!
This means that twice the size of device will put exactly the same solar flux on the surface but there will be a lot more capacity inside.
I think that this means that compound parabolic solar devices will have more or less ideal sizes for specific uses.
Also this aspect, the volume increasing faster than the collecting area, might mean that 15 degree and 30 degree compound parabolic dishes will become the norm in larger units.

Step 12:

Using software to compare solar cooker performance
Solar cookers come in a huge variety of shapes and sizes and lots of them are great when pointed directly at the sun but performance becomes poor very quickly as the sun moves. I have been on the lookout for software to help with design since august 2008 or earlier. I finally found something useable by me. Art of illusion. With it, I compared a parabolic dish, various compound parabolic dishes and a hemisphere as the sun moves in 5 degree increments. (20 minute incriments). It is only preliminary results of course but the hemisphere wins as the best stationary dish!
Who would have thought! Parabolic dish came --- LAST!
In fairness my test was pretty crude. I just set up 7 cameras to render the scenes from directly overhead to 30 degrees off. (2 hours later). I used a red ball to represent the cooking vessel and I reason that the size of its reflection in the dish will approximate the cooking power.
Nothing was optimised. I put the ball at 000 at the xyz cooordinates and moved the dishes until the reflection image filled the dish with the camera overhead. Then I made all the other dishes invisible and rendered the scene with each camera in succession. Finally I put the 35 images in one jpg file for people to do their comparison. I have also uploaded a slightly better version of my art of illusion scene file for anyone else who wants to try this. I got a lot of help in the art of illusion forums with this and I thank them for it. The help thread is
If you use my scene file, you need to reduce the magnification from the default 100 to 1 or 2 to see the dishes The scene file should be .aoi when you download it. I have been told that it can download as .temp (Just change to .aoi and it should be fine)
Explanation of picture. With the sun overhead, all the dishes show a full red image, meaning that the sunlight will be all reflected to the red ball at the focus! So they are all equally good when pointed at the sun! By 10 degrees off (40 minutes later) a black cresent appears in the parabolic dish reflection meaning that this dish is much poorer in those conditions. It reflects conciderably less light to the red ball than do the other dishes! By 30 degrees off target, pretty much all the dishes are useless, the only red that we can see is the red ball itself. The dishes just reflect the sunlight back into space. (The people at art of illusion are not too impressed with my poor choice of magnification) etc. I should have had the red discs much bigger. Sorry
16th feb, I got help with a new scene file. If you use it, you can render a stop motion movie of jpg files of each dish very quickly, you can also move the movie timeline and your camera moves quickly.( I have video on youtube to help you get started with this. is part one of software is part 2 It is poor quality but if you get through it, it is worth it.)

The scene file for making movies is pretty neat. It is called dishs all same size nullworks.aoi I used one frame per second and 48 frames in total
i left the 7 unused cameras so you can use them as markers as your move camera 1 in an arc over the scene
20th feb I added the 5 .mov files here for download so you can inspect how the dishes reflect light over time.
Depending on the dish, the reflection pattern is very Different! I urge people to download and inspect these quicktime movie files. They show a dish from ovehead to 45 degrees off the side.

Hungry Scientist Contest

Second Prize in the
Hungry Scientist Contest

Be the First to Share


    • Mason Jar Speed Challenge

      Mason Jar Speed Challenge
    • Bikes Challenge

      Bikes Challenge
    • Remix Contest

      Remix Contest

    21 Discussions


    4 years ago

    Caan you plz inform the reflector sheet which u have used is of which material ??

    moreover plz provide me with the information that which materials can be used as a best solar reflectors


    10 years ago on Introduction

    ihave idea you put inside the box or card board


    Reply 10 years ago on Introduction

    Yeah, you could have a box cooker up where my oven bag and pot are.
    This was the first test of the idea and the second test is beginning soon with completely redesigned equipment.


    Reply 8 years ago on Introduction

    Hi Brain,
    I would like to know the type of leaser level that you used for the that interesting compound parabolic experiment? can you also tell me what is the aproxx cost of that type?


    Reply 8 years ago on Introduction

    Is that the experiment with strips of mylar? It was a mastercraft laser level.  Laser levels are much harder to find these days in Canada.  The el chepo ones are gone. Basically any laser level that does a line will work. You just have the line pointing straight up and down.  Mostly they have adaptors to do the line and you just plug that in.  If you buy a laser level (for general use), it is far more important to have the dot than the line. The dot can be seen much further away.  This is probably my most useless laser level.  It only does a line and even though it is pretty powerful, it is pretty much useless outdoors, in daylight because a line fades very quickly further than 10 or 12 ft from any laser level. My best laser level was a jobmate 9 inch one that I got for 15 bux.  It has a better brighter laser than laser levels costing 4 times as much. (And just as accurate). By the way, laser levels can be checked with a water level for accuracy.  Some are not so accurate when you check them!

    Reply 8 years ago on Introduction

    Hi Brain..
    Thanx for these interesting tips.
    I am actually working on compound parabolas, but could not able to get it right...the compound parabola that you have shown is continuous.(above video).
    but i have found few site in net that shows some different picture for it..
    I am totally confused can you plz help regarding this.


    Reply 8 years ago on Introduction

    Hello, Krishnashaw, The fossil freedom thing is a compound parabolic TROUGH based on a winston patent that has expired but the picture is scaled WRONG. Do not use it! I scaled it up on paper and played with it a while before realizing that it is crap and the angles in it are useless!
    My thing was the first part of my attempt to turn that trough into a dish. (Because parabolic dishes have such a short time before you have to move them and because dishes have greater concentration.
    So basically the bad news is that my stuff here is just part of the process to making a better dish. It is probably a bit better for solar cooking than a parabolic dish but it is not as good as I had hoped. My latest version is the clam shaped solar cooker. That should be best of all but I still have a LOT of work to do on it. So search "solar design t-square and clam shaped solar cookers" for my most up to date stuff. (And please be prepared to work on my foundations, and also be prepared for mistakes by me) I am not perfect and I do not have the math for this but the trial I made of the clam shaped solar cooker seemed to work pretty well.
    If you want to go into the theory of troughs and compound parabolic troughs, send me a message and I can send you pdf's with the correct dimensions and with several other trough designs too. (the pdf files are not my work but were obscure yet very interesting stuff and helpful for the hobby trough and dish maker). You might never find them on your own. Hope this helps
    Brian White


    Reply 8 years ago on Introduction

    Hi, Brain.. sorry for the late reply... i was actually busy with my exams..
    it would be good if u can send me the materials you mentioned above..
    the problem is that i have already done experiment on compound parabolas and don't want to leave this work with out seeing its end.. so i will do it and lets see if i can make it a little efficient..
    i just need your materials to study..
    my email address:


    11 years ago on Introduction

    Good work, and good intentioned. The earth needs more people as you. I am making a similar design, equatorial mounting included. I tried many methods to make the parabolic collector, and my conclusion was that the metalized plastics are not convenient because each wrinkle, large or small, behaves as a hole in the reflective surface. And there are thousands of them. Then, I am using glass mirrors. The problem with they is that it is necessary to calculate the form of each mirror with certain precision, for which I had to disinter my old know-how of trigonometry, and then to orient them carefully one to one so that they aim all the sunlight to the same place. They are 85, for an circular array of one square meter. These difficulties are real and despicable nothing, but once surpassed, the result is acceptable. You live as you say in a very high latitude (mine is 33º south), which obliges you to maximize the cares. In full summer the abundance of sunlight permits to cook an egg putting it inside the car with the glasses closed, but in winter one must have a parable or an arrange of mirrors well designed and well built. Otherwise, one wastes time trying to cause to boil a small pot of water. Pardon my English, I live in Argentina. I speak Spanish.


    Reply 11 years ago on Introduction

    I accept your argument about the wrinkles. I found kitchen foil produced more wrinkles than the metalized plastic. A bigger problem is that some of the metalized plastic lost its metal. I do not know if they are coated front and back or just at one side. I suspect that I laid some sheets upside down and the metal reacted with the wheat paste. But I have not looked into this. I simply cannot tell if they are coated both sides or just one side. But I also think that this time I produced acceptable results. i have them in the pdf file at the end. has results from a sk14 solar cooker in India heating water 10 liters in a 12 liter pot. So basically the solar cooker is about twice the size of mine heating 10 liters in a 12 liter pot. It reliably took just under 2 hours to bring the water to near the boil. Mine heated 7 liters in an 8 liter pot. My best results was just over 3 hours to get to a similar temperature. But remember it is September in Canada! So compared to one of the best that there is (The SK14) it did not do too bad at all. I love your version of English. The meaning is clear and it is almost poetic. Thanks for the encouragement. We try to do an acceptable job in the time available! Brian


    Reply 9 years ago on Introduction

    i think there is a parabolic cooker potentially with a same unattended time


    Reply 9 years ago on Introduction

    Thanks, I retired my compound parabolic and made a clam shaped solar cooker for unattended use. I think that the clam shaped will easily beat the parabolic because it collects more sunlight. You do get a double bounce of light sometimes but the extra collection area more than makes up for any losses from the extra reflector.


    Reply 11 years ago on Introduction

    The plastic sheet is metalized on one side only. You can detect it damaging the surface with the tip of a knife or something similarly sharp.

    Another problem with the glass mirrors is the weight: one square meter of 2 mm thickness (the thinnest I can obtain and cut) it weighs 6 Kg, which obliges me to put a counterweight to balance the parable, and that at the same time obliges me to reinforce all the structure.

    There are also light plastic mirror, but is very expensive for me. And besides the salesperson told me that it does not resist the elements a lot of time, it goes losing the shine quickly.

    We continue insisting, that some day our efforts will be valued by who do not share the generalized attitude of spending and to waste the fuels.

    if you want, you can see two brief videos of my first successful test of my arrangement of mirrors, winter 2006, in or, if you can't enter there, in and

    wilson, te
    wilson, te

    11 years ago on Step 7

    The template is difficult read. Reflective light makes it difficult. Is there another source for template? tkstaco


    11 years ago on Introduction

    This is really good, I was hopping to make a parabolic solar cooker a while ago, but found it too hard to make...


    Reply 11 years ago on Introduction

    I am trying to do one just with mylar and my template today. I got 12 segments cut out of the mylar and I taped them together with masking tape. I had the mark the mylar first round the template with a marker and then slice along the lines with a scissors. It did not work just slicing round the template. You get a jagged cut that way. It is not great, probably need more segments to do a better job. I still have to figure out how to hold it in umbrella shape! But I bet it will work. If it works half way well, the umbrella one would be awesome for emergencys or camping because it stays cooking for 3 hours unattended. I think the key to getting these things working really well is NOT to use a parabloic shape. I think a hibred between a compound parabola and a cusp will end up being the right shape. Do my optics experiment and decide the best shape yourself! Brian