The Sundogger




Introduction: The Sundogger

About: I work at Middle Tennessee State University as a Professor of Physics and Astronomy and Direct the Computational Science Ph.D. Program. I've been a programming nerd, a woodworking geek, an astronomy dork, an…

The Sundogger is a solar hot dog cooker I built using my X-Carve CNC machine. I decided to make this project to learn a bit more about carving larger pieces of furniture on my CNC machine. However I was also motivated by a lifelong interest in both solar energy and in hotdogs. When I was a high school student, a mentor of mine built a solar hot dog cooker for the Earth Day Celebration at the local community college. I grew up being fascinated with solar energy, so I wanted to build something like this for my backyard for decades. Around the same time, I remember seeing a similar device for sale in the Edmund Scientific Catalog. My version of this cooker is an homage to this now ancient device.

It is important to note that this while this project is named the Sundogger, it is vegan friendly. It will work with any kind of tubular food that can be put on a skewer or can be made into a cylindrical shape. If you end up using other kinds of foods, make sure they are cooked to a safe temperature before eating them.

Originally I tried to build this project out of foam-core instead of plywood. After some initial tests, I found that I couldn't make smooth cuts with this material. The foam core board fragmented too easily even when using the sharpest knife. I considered moving to a hot wire cutter, but opted to move toward a plywood-based project instead.

In my explorations of the local crafts store, I found a great material to use for the reflector - an aluminized piece of poster paper. The reflectivity seemed high enough to make the project work without having to suffer through the inevitable crinkling of aluminum foil that occurs when you try to smooth over a large surface area. If you can't find this material, aluminum foil mounted on cardboard should work well enough. I also considered purchasing a acrylic mirror sheet to improve the efficiency. In the end, I opted for the simple and cheap poster paper solution for this prototype.

For the frame, I decided to try and make the entire project out of a single 1/2 inch "handy panel" of plywood a bought at the local big box hardware store. These 2 ft x 4 ft pieces is just the right size for the parts needed for this project.

The total costs of supplies was about $35, including the plywood, the reflective poster paper, the finish, and the bolts.


  • 1/2 inch plywood panel - interior grade - 2 ft x 4 ft
  • 1 piece of reflective poster paper (available at the craft stores)
  • 6 - 1/4 inch x 1 1/4 bolts #20 bolts, washers, and butterfly nuts
  • 4 furniture bolts and cross dowel nuts (I used the Hillman 5/8 plain steel barrel bolt + 1/4 inch x 50mm bolts)
  • Set of two small cabinet hinges
  • Bamboo or metal skewers - at least 24 inches long (sharpened dowels will work)
  • Wood glue, finishing nails
  • Wood finish - if you like


  • CNC machine with usable working area of at least 24 x 28 inches
  • Files, sandpaper, knife and saw for trimming and cleaning the wood after the carve
  • Clamps and perhaps a pneumatic nailer to join the wood together during gluing
  • Drill - to install cross bolts, install the hinges, and clean up the CNC cut holes

Step 1: Solar Hot Dog Physics

To understand how the Sundogger works, it helps to step back at talk about the physics of hot dogs.

At Earth the total energy flux (flux density) from the Sun is called the the Solar Constant. The value of the Solar Constant is approximately 1360 Watts per square meter or 1.951 calories per square cm as measured on a surface that is perpendicular to the incoming sunlight. This number doesn't change since the distance between the Earth and Sun is approximately constant over the yearly orbit.

The Sundogger we are building has a width of approximately 24 inches or about 60cm. If our Sundogger was a full meter long, we would have an area of 0.6 x 1 = 0.6 square meters. However really understand hot dogs cooking in the sun, it turns out it is easer to think about the energy per centimeter of length of the cooker rather than the total energy of the cooker. The parabolic shape of the collector concentrates the energy on a skewer, so the energy for every cm of length is going to be the energy that is concentrated from the local 1 cm of width in the collector. In our case, we have 1.951 calories per square cm per minute x 60 cm (width) = 117 calories per minute of solar energy for every cm in length along the skewer.

Detailed scientific measurements have show that a typical hot dog has a diameter of about 1 inch of about 2.5 cm. This gives us a hot dog radius of about 1.25 cm. (Precision measurements of hot dogs put this number at 2.726 cm in diameter - but your actual hot dog diameter may vary.) The volume of a hot dog - or anything - is its length times its cross sectional area. The cross sectional area is going to be A = Pi times the radius squared. This means that every linear centimeter of the hot dog has a volume of (1.25 x 1.25 x 3.14) = 5ish cubic cm.

The mass of any object is the density times the volume. According to the manufacturer of the hot dogs I used, each dog has a mass of 57 grams. With the length of the hot dog measured at about 12 cm, this gives us a volume of about 4.8 grams per hot dog cm. This estimate puts the typical hot dog density just below 1 grams per cubic cm.

Combining these energy input per centimeter and the mass per centimeter, we find that we have 117 / 4.8 = 24 calories of energy per gram being added to the hot dog every minute. Thus in every second, we get enough energy to raise the hot dog temperature about 24 degrees Celsius every minute when it's internal temperature is about 20C.

Of course, this estimate is doesn't include several important factors:

  1. The 1360 Watts per square meter figure is actually the amount of sunlight hitting the outer atmosphere of Earth. The actual amount that reaches the surface is typically about 1000 Watts. The actual number depends how far the Sun is from overhead. (Thanks to the user redrok for pointing this out.)
  2. The energy from the collector to the hot dog is not transferred at 100% efficiency. The reflective paper we are using probably reflects less about 50% the light directly to the focal point.
  3. The hot dog itself also reflects some of this light rather than absorbing it. The hot dogs albedo (the net reflectivity) is perhaps about 0.2 to 0.4, suggesting that 20% to 40% of the visible light reflects off its surface.
  4. Much of the remaining energy is converted into thermal radiation that escapes from the hot dog's surface rather than cooking it. The exterior heats up, and then Planck's law causes much of the energy to escape back into the air rather than thermally diffusing into the interior. This thermal radiation effect probably changes as the hot dog''s internal and surface temperature rise.
  5. The energy need to raise the hot dog's internal temperature depends the hot dog's current internal temperature, surface temperature, and the thermal conductivity it has.

I briefly considered doing a simulation of the hot dog thermodynamics to understand these effects a bit better, but decided it was outside the scope of this Instructable. Instead, we are going to just kind of eyeball these effects to make a prediction about the cooking time. Assuming the actual net efficiency of the cooker is probably about 20%, the temperature increase of the hot dog should be about 5 degrees Celsius per minute in bright sunlight. To get the temperature up to a safe temperature of 80C from a starting temperature of 20C, we should probably let them cook for about 15 minutes if our efficiency estimates are accurate.

I should note that the hot dogs used in this experiment were pre-cooked, so they technically do not need to be cooked at all. However, the USDA recommends a cooking temperature of 145F for pork. It seems best to use as a guideline for this project.

Finally - a note about terminology. The flux density is defined as actually a measure of power per square meter per second. Watts is a measurement of power - the amount of energy change per unit time. Calories or Joules are a measurements of energy. Calories is almost always related to the thermal energy or heat in a system. Temperature, on the other hand, is a measure of the average thermal motion of molecules inside an object. Temperature is related to the heat, but is a bit different. The properties (specific heat, mass, etc) of a body determine how temperature and heat are related to each other. I was a little loose on this terminology when I first wrote this up.

Step 2: Carving the Wooden Frame

Design and Layout

Carving the wooden frame is pretty easy if you have access to an X-Carve or another CNC machine. However, there are a couple tricks that I found helpful.

I did the layout of the project using the Inventable's Easel software. Easel is a free on-line CAD/CAM package that works with all their X-Carve and Carvey machines. However, it also now works with any GRBL CNC machine. The Easel file is easy to use. It already has the CNC tabs included (to keep the pieces you are cutting from breaking loose during the carve) and the depths are already set for the carve. However, if you have another kind of CNC machine, I have included the SVG layout file for your use as well. You can access the Easel project at:

The project layout is laid out on a 2ft x 4ft piece of 1/2 inch plywood. You could use MDF or another material as well, but the 1/2 inch plywood yields a sturdy and attractive final piece. Since the usable area of the standard XCarve is 30.5 inches by 30.5 inches, you have to cut the top and the bottom halves separately. Basically, you carve the reflector frame and cross pieces, and then rotate the plywood and cave the other support pieces on the other half of the sheet.

When I made the design for this project, I used a Python code to generate the SVG file with the parabola on it. Using a Python code to generate designs is a pretty nice way to make mathematically complex designs. This layout could have been done directly in any drawing program that supports making parabolas, but Python was the best tool for me given my background. I've included my code and the resulting SVG file if you want to play with the parameters and create your own files. You will need to edit the code to change the parameters of the piece, but I think there is enough comments in the code to make playing with it fairly easy. Using Python to generate the SVG file is pretty easy when you use the Matplotlib package. However, the resulting file didn't import cleanly into the Easel. It included some additional unsupported SVG commands including the clip paths. It also had a free redundant points in the file. I manually just edited those lines out of the template. I included both the edited and original SVG files generated by the Python code so you can reproduce this.

Once I got the basic SVG outline imported into Easel, I created the basic cooker frame. The parabolic pieces include feet on each end so it will sit flat on the support frame. The idea is to use two of these parabolic pieces and hold them together with horizontal cross pieces. To connect the cross pieces and the parabolic pieces, I created a set of dog-biscuit joints using Easel. The width was designed to fit the width of the reflective poster paper I found at the craft store. I also added a 0.1inch rectangular indent to on each of the parabolic pieces to help align the vertical struts that hold the skewer. I also added a nice friendly project name on each side of the cooker. Holes are added to the cooker so we can connect tracking brackets to the support bed.

Once you have machined the cooker frame, you rotate and reattach the rotated plywood sheet back to the waste-board. I used the same zero location for the piece between my carves. I also broke up the carve for the support pieces into to several stages. This isn't strictly necessary, but it made it easier for me to monitor the work in the shop.

The longest two and three shorter pieces in the design form the support bed of the cooker. The two short pieces with a single hole in them will support the skewer above the parabolic cooker. The three cross pieces and the triangle pieces will be glued and nailed to together to form a sturdy frame. The holes in the long pieces are used to connect the tracking brackets to the support bed. I added a 0.1inch dado in the frames to make assembly go a little bit easier in the support bed.

The two longer pieces with the curved ends and the long slots are used to tilt the cooker toward the Sun. These tracking supports will go on either side of the cooker and attach the support bed to the parabolic frame.

The remaining two vertical pieces hold are connected to the parabolic elements of the cooking frame. They hold the skewer at 12 inches from the parabolic - the focal point of this cooker.

Carving the Sundogger

For this project, I used a 1/8 inch bit on the CNC machine for this cut. I used the default settings for plywood - a free rate of 40 inches per second, plunge rate of 12 inches per minute, and cut depth of 0.05 inches.

The carving time for the project is about 4 hours and 30 minutes. It takes about 2 hours and 45 minutes to carve the support bed, the skewer holders, and the tracking brackets. It takes a little less than 2 hours to carve the cooking frame and supports.

The CNC machine really shines when you are creating the cooker frame. Cutting the dog-bone connectors and the parabolic by hand would have been difficult. However using the CNC machine to cut out the support bed was not really necessary. It probably would have been faster to do these cuts using my table saw - or to just use 1x3 wood for these parts. It is easy and fun to use the machine, but making long vertical pieces doesn't require a CNC.

The only critical element of the project is the parabolic shape that defines the shape of the reflector. However, it turns out that even this shape isn't even that critical. For this design, the maximum deviation between a parabolic and a simple circle was only 0.7 inches. It would be pretty easy to just use a router, band saw, or jigsaw to follow the template well enough to get a good enough focus. Even just cutting a circle with a radius of 24 inches would work well enough to focus most of the light on the hot dog. Don't let the desire for perfection prevent you from doing science! There is another great Instructable about how you can Build a solar hot dog cooker without using a CNC machine at all.

Step 3: Cleaning Up the Carved Parts

Once you have finished with your carve, you need to free the parts the plywood sheet. I normally do this using a small flush-cut saw and a box cutter. As you can see in the pictures above, the support tabs that remain in the pieces need to be cut through to remove the pieces from the remaining plywood. These tools work great for doing this tasks.

In some cases, you will have a tiny amount of warping in the wood during the machining. This is particularly a problem when you do large pieces on your machine that are clamped down only at the edges. Even a tiny amount of warping means the CNC machine won't completely cut through the wood in some places. You will have to cut through this thin layer of wood (usually 0.05 inches or less) with a knife or saw. Unfortunately, the resulting edges on the piece can end up looking a bit rough.

After the pieces are freed, I use a combination of wood files, sandpaper, and even an oscillating drum sander to clean up the edges. For this project, freeing the parts and cleaning up the edges took about 90 minutes. I went slowly and systematically through the parts to make sure everything was flush and fit together well.

For the dog-bone joints in the cooker, I test fitted these together and used my wood file to clean up the edges. It only took a few minutes to get everything to fit tightly together.

Step 4: Assembling the Cooking and Support Frames

Assembling the Cooking Frame

Assembling the cooking frame is easy. After the wood has been cleaned up, just insert the tabs from the cross pieces into the dog-bone holes on the parabolic pieces. Make sure the connection is completely flush. Hold the assembly together with a clamp, and then drill a hole through the parabolic pieces into the cross pieces. Once the holes for the bolts have been cut, locate and drill the locations for the dowel bolts. Using the dowel bolts made it easy to get a nice tight connection that could easily be disassemble if needed. Of course, you could just use glue, screws, and nails if you wished. This assembly of the cooking frame took about 30 minutes. It too me a bit longer than expect to clamp the frame and align carefully align the drill when making the holes.

Assembling the support frame

To assemble the support frame, you need to use the two longest pieces, the three shortest pieces and the six triangular pieces. The triangles act as angle brackets to provide strength for the assembly. The inset grooves are designed to hold these triangles in place. To hold this all together, I used a combination of wood glue and my brad nailer. This step also took about 30 minutes. Most of the time was spent setting up my brad nailer and making sure the alignment of the cross pieces were square and well aligned.

Finishing the Wood on the Sundogger
If you wish, you can stain, paint, poly, varnish, lacquer, shellac, or seal the Sundogger. I hate the finishing stages for projects, so I opted for a simple spray coat of shellac for the project. With shellac you can put thin spray coats on every 30 minutes, and only need to do a light sanding and tack cloth between the coats. It involves minimal clean up, and takes basically no effort. I put on three coats one evening, feel very happy with the results.

Step 5: Connecting the Support Frame to the Cooking Frame

The Sundogger needs to be able be adjustable so it can track the Sun. To allow these adjustments, we will attach the cooking frame to the support frame with hinges and the allow the opening angle to be set with the tracking supports.

Adding the hinge

Carefully align the end of the cooking frame t(oppose that of the support holes) with the base of the support frame. You want to have the end of the cooking frame completely flush with the support frame. Once everything is aligned, clamp the pieces together. Using the cabinet hinge, attach the two pieces along the seam. I put the outer edges of the two hinges about 3.5 inches from the outer edges of the cooker. When the hinges are installed, you should be able rotate the cooking frame while the base remains flat on the ground.

Attaching the reflector
Attaching the reflecting material to the Sundogger is easy. Measure and trim the reflective paper so it fits on the parabolic back of the device. To attach it to the parabolic surface. I just used a staple gun for this step. If you used an acrylic mirror or a piece of sheet metal, screws would be a better choice.

Step 6: Final Assembly Before Cooking

Once you have added the hinges and finished the project, you can do the final assembly of the parts for cooking. I take these parts off when I store the cooker. This only takes a few minutes to do.

Adding the tracking supports

The tracking supports are attached on the support frame with 1/4 bolts, washers and wing nuts. You can put this through any of the holes on the bottom of the frame. The slots in the tracking supports are attached to the parabolic pieces with 1/4 bolts, washers, and wing nuts as well. But picking the holes you use on the support frame and the parabolic side pieces, you can adjust the vertical angle that the tracker faces to point directly at the Sun. Attaching the skewer holders Next connect the skewer holders to the cooking frame. Align them along the outside of the parabolic pieces. They are design to place the hole that holds the skewer 12-inches from the parabola if the vertical piece is touching the bottom of the pocket hole found above the words "Sundogger" on the side pieces. I opted to attach the skewer piece to the sides with a 1 1/2 bolt x 1/4in bolt, some washers, and wing nuts. I clamped the pieces together and drilled a hole about an inch up from the bottom of the skewer holder through the side piece. By using a wing nut and the 1/4 bolt, this piece and be removed or folded flat for transportation.

Daa.... Whoops

When I first cut the CNC skewer holders, I managed to used the radius of curvature of the parabola rather than the actual focal distance of the parabolic to set their length. I ended up using a hand drill to add secondary holes in the the skewer holders at the correct distance and eventually trimming these pieces to length on my band saw. This was an easy fix - and the problem has been fixed in the Easel and SVG files.

Step 7: Cooking Hot Dogs!

Cooking the hot dogs - or any skewer-friendly food - is easy.

  1. Wait for a sunny day.
  2. Turn the cooker away from the Sun.
  3. Put the food on the skewer, and then thread it through the two holes at the focus of the device.
  4. Turn the cooker so it is facing directly toward the Sun.
  5. Once it is facing the right direction, loosen and adjust the two tracking bolts and to adjust the tilt.
  6. You can look at the shadow of the skewer holders on the ground to see if the device is aligned. Make sure you can see their shadow on both sides of the cooker and that the tilt is adjusted so they make as small of a shadow as possible.
  7. If everything is aligned, the bottom of your hotdog should be brightly reflecting the sunlight.

To test the Sundogger out, I ran a two backyard science experiments. I used a digital cooking thermometer to record the temperature of the cooking hot dog during the afternoon on July 27 at 3pm and July 28 at noon. Exterior air temperatures on both days was 88F. The graph of the temperature on both days is show in the graph above.

During experiment 1, I had a few small clouds passing in front of the Sun during the experiment. The temperature of the hot dog warmed up mostly uniformly constantly through the experiment, although you can see the times when clouds are briefly blocking the energy coming from the Sundogger. I discontinued the experiment because the interior hot dog temperature reached 150F, and because my wife was hungry. It took about 35 minutes to bring the hot dog temperature to its final temperature. At the end of the cooking time, the temperature was rising about 2F per minute. The temperature rise seemed a bit slower than expected, but it certainly gave a satisfactory result.

During experiment 2, there was a bit of high overcast during the first part of the experiment. Since the interior temperature of the hot dog was at 32F when the experiment began, there was a rapid warming phase in the hot dog for the first 10 minutes that rapidly brought the interior temperature to the ambient outdoor temperature. After that, you can see that the temperature increased more slowly than in the first experiment because of the atmospheric haze. About 32 minutes into the experiment, large clouds blocked the sun. The temperature of the hot dog began to decrease slowly. These clouds remained in the area for the next hour, so I was never able to get the hot dog's temperature to get up to my target of 150F. Experiment 2 was also interrupted when a small yellow furry quadruped named Fred began his own investigation into the interesting smells in his backyard.

I can draw a two important conclusions from these experiments:

  1. Clouds are bad.
  2. This temperature decrease was missing from the initial estimate of the calculation. Although we are adding thermal energy to the hot dog from the Sun, the hot dog is also radiating heat away because the air temperature is less than it's surface temperature. It turns out the rate this occurs is approximately proportional to the temperature of the hot dog minus the air temperature. The air helps initially warm up the hot dog, but then slows down the heating toward the end. This means we only get about 1 to 2 degrees Celsius per minute.

It turns out there is some interesting physics involved in cooking hot dogs. Newtonian cooling approximations combined with a linear input of thermal energy might do well for modeling the temperature profile. I might end up giving this problem to my graduate students in the fall so they can calculate the thermal coefficients involved.

Of course, students could do many other science experiments with this project. You could try to improve the efficiency of the device by using a better reflector, a bigger frame, changing the reflectivity of the hot dog (by blackening aluminum foil?), or reducing the thermal losses through some kind of insulation. There are so many avenues for future work! However for now, I am just going to enjoy a nice lunch in the backyard with my wife.


If you have a CNC machine, reproducing this project should be pretty easy. If you don't have a CNC machine, you still can built a version of this that works with just some basic hand tools. Take a look at the other similar Instructables for inspiration. Don't get stuck just because you don't have exactly the right tool!

It was challenging trying to figure out how to machine this out of a piece of wood that didn't fully fit on my CNC machine, and to incorporate a python generated SVG file into the design. I think it's a great project to make for Science Fairs or just to have in your backyard. I enjoyed doing the design, the construction, the theoretical hot dog physics, and the experimental hot dog physics. I hope you enjoyed as well!


Thanks to Ben Becker, Neal McClain, and the gang at the MTSU Makerspace for their input and useful discussions during this project. Also, thanks to my old college Physics Professor, Don Penn for inspiring this project and for all the mentorship he gave me as a student.

Backyard Contest

Participated in the
Backyard Contest

Be the First to Share


    • Plastic Contest

      Plastic Contest
    • The 1000th Contest

      The 1000th Contest
    • Battery Powered Contest

      Battery Powered Contest

    15 Discussions

    spark master
    spark master

    1 year ago

    I like the idea, I like it better with out tools most people do not own. Also please refer to things in Fahrenheit and inches. I know you are a scientist, I know the excuse is "Well I think Metric, as does the world" , but reality is, if you want average people to follow, you need to use the vernacular, even if it hurts. This site is global, and yet the readership is mostly American, the average American does not know a mm from a tube of toothpaste.

    Now I will await as flames rise with all the people who like yourself, use metrics daily take umbrage with this post. I took 3 years of real science, Bio/Chem/Physics, a million years ago in HS and we, like you, computed everything into metric first step. Then again I am an olden farte and we did have slide rules. In Japan they used to have an abacus race a calculator. Calculator always lost, why? no enter key, the answer is just there.

    For me, the brain cells that once housed the metric system are long past dead. Although I did use it for a hile to make 2 Meter Antenni.

    Now the nice part. I did love the piece, It is totally overdone for what it is, but heck. iff'n I had a cnc machine and skills to use it, this looks fun.


    1 year ago

    Especially love the Venn diagram.


    Reply 1 year ago

    This project has a niche market. :)


    1 year ago

    Did you experience drippings from the hot dogs falling on the reflector? I did on mine. :(

    Note the nail on the center of the top right edge. This is used to align the panel. When the shadow of the head is concentric with the bottom of the nail, you are at the best alignment with the sun.


    Reply 1 year ago

    Yes! It did get a bit drippy. I used some 409 and a paper towel to clean it up. I also used the skewers for alignment. I think using cup hook might be helpful to set it up.


    1 year ago

    Hilarious! Loved your write up - entertaining and really informative. Nice work.


    Reply 1 year ago

    Thanks. I had fun working on the project and the writeup. I wasn't sure about including the hot dog physics, but felt like it would be fun to do some back of the envelope calculations to provide some context. I also enjoyed doing the experiments, although it was heartbreaking to watch the hot dogs cool down when the clouds came.


    1 year ago

    This is incredible.

    v2 should include a solar power rotisserie.


    Reply 1 year ago

    The problem then becomes the thermometer cord for the hot dog core temperature. I guess going with servos would work.... hmmm.


    1 year ago

    No JohnW539 you don't get 1360 Watts of Energy.
    1. Firstly, it's not energy, it's Power.
    2. You only get about 1360 Watts per square meter if you are outside the Earth atmosphere and near Earths orbit.
    3. The rule of thumb for us on Earth is it's about 1000 W/m^2 for full sky radiation. The other 360 W/m^2 is lost to the atmosphere as heat.
    4. Lastly, the rule of thumb for solar concentrators is you only get about 800 W/m^2 for radiation directly from the Sun. The other 200 W/m^2 is wasted because it wouldn't hit the hotdogs.


    Reply 1 year ago

    Technically we are talking about the flux density of the incoming radiation, but sure - energy per unit time is power. The other effects you mentioned are lumped into the category of efficiency in the writeup. Atmospheric losses, reflectivity of the collector, the reflectivity of the hot dog, and thermal losses all drop the amount of energy that actually goes to cooking. I did simplify this discussion a lot intentionally. For a short while, I thought about doing full simulation on hot dog cooking but decided it was going into way way too much detail for a CNC project.

    Alex in NZ
    Alex in NZ

    1 year ago

    Beautiful piece of work. Thank you for sharing it :-)


    1 year ago on Step 7

    I am going to share this in many solar cooker Facebook groups. I hope you join us in solar cooking discussions around the world. I have a CNC and will try this.


    Reply 1 year ago

    Thanks! Let me know how it work out.