Charcoal Retort




I have had a couple of attempts making charcoal using 44 gallon drums and they have worked okay, but needing a new project after buying a mig welder, I decided to make a retort.

The theory of a retort is that you heat the wood to be made into charcoal up to a point where it starts expelling all of the wood gas and impurities, which is then fed back into the fire to keep the process going.

After reading up and watching a number of videos on rocket stoves and retorts, I decided to make one with the idea of combining the two into one rocket stove powered charcoal retort.

Teacher Notes

Teachers! Did you use this instructable in your classroom?
Add a Teacher Note to share how you incorporated it into your lesson.

Step 1: Obtain and Flush the Cylinder

I started out with an expired 90kg propane tank purchased from a scrap dealer and a length of 100mm box section. The box section was the dearest part as the scrap dealer did not have this in stock, therefore I had to buy it new.

The first thing and most important thing to do was to remove the tap valve and fill the cylinder with water to ensure all the gas was removed before I started cutting and welding it. I filled the cylinder up until it overflowed to ensure all gas was gone.

Once the water had been emptied, I cut the top off the cylinder with an angle grinder.

Step 2: Make the Rocket Stove

To make the rocket stove I cut the piece of box section off on a 45 degree angle and welded it back together to form an L shape. I made the small section 100mm shorter so I could physically fit it into the cylinder and later I would weld the 100mm back on, once the box section is in place. If not done like this you cannot fit the stove into the cylinder and have enough poking out the front.

The longer part of the stove was made to length so it sits lower than the top of the cylinder.

Step 3: Weld the Rocket Stove Into the Cylinder and Add the Chimney

Next I cut out two 100x100mm holes in the cylinder. One in the top for the chimney and one in the bottom (front) for the rocket stove to sit. I cut these with an angle grinder and welded the rocket stove in place. I then welded 100mm more box section onto the front, so I had a decent amount out the front to hold the wood being burnt.

I welded a 50mm flat bar around the chimney top (lid) box section to seal off the chimney as the top sits on cylinder and internal box section to ensure a snug and aligned fit of the chimney section.

I also welded a piece of 50mm flat around the top of the cylinder to create a seal when the top fits on.

Step 4: Add the Retort Vent and Paint - Ready for Trial

I cut a hole in the top front of the cylinder and one in the top of the rocket stove and then welded some 25mm tubing and angle between the two points. This enables the wood gas to feed back into the fire once the charcoal wood heats up and produces this gas.

I drilled a small hole and mounted a BBQ temperature gauge into the cylinder also so I could monitor the internal temperature.

At this point I painted the whole thing with heat paint.

Step 5: Firing It All Up for the First Time

Once all painted, I loaded it half full with Tea Tree (manuka) to do a trial run. I fitted the top on with some force (hammer) as it was a tight fit. I needed to line up the chimney box section to get the top to go down and lock in place.

There was a small gap around the edges to the lid, so I used mud to fill this, as I had seen this done on other charcoal making videos and this seemed to work well.

I lit the fire and sat down to watch as I would need to keep feeding wood into the rocket stove.

The temperature increased quite fast and the rocket stove roared.

Step 6: Once It Got to Temperature It Started to Self Feed

Once the unit heated up to over 200 degrees Celsius, it started to put out visible gas which was sucked back into the fire and I assume burned. It also put out some sort of tar which was also burnt.

Once it reached 300 degrees Celsius the gas coming from retort was clearly flammable and the fire started to roar.

I stopped feeding wood in at this point as it had become self feeding with the wood gas.

I kept watching it for about an hour after this as every now and again it would do what i could only describe as "cough" and blow itself out, so I would start it again with a gas torch and it would be off again.

I decided to keep a few smaller bits of wood burning after this happened a few times and that seemed to fix the problem as it would relight if it coughed itself out.

Step 7: The End Result

The next day once it had cooled down, I removed the top from the retort with a hammer as it was locked on tight and emptied the charcoal out.

I tried all different sizes of wood in this first run and the larger pieces had not fully turned into charcoal but all of the smaller pieces had converted well.

I hope this experiment gives others some ideas and I would like to hear any recommendations to enhance this design.

I really wanted to prove the theory worked... which it definitely did.

Be the First to Share


    • Made with Math Contest

      Made with Math Contest
    • Multi-Discipline Contest

      Multi-Discipline Contest
    • Skateboard Contest

      Skateboard Contest

    23 Discussions


    5 months ago on Step 6

    I think you did a good job on this project, but I did notice your amperage on your welding unit was probably set a bit too low based on the welds.

    2 replies

    Reply 5 months ago

    Thanks for the tip, the main problem was i has a disposable argoshield bottle that had run out and i didnt realise so most of the welds were with no gas.


    Reply 5 months ago

    Your stick was melting okay, but the workpiece wasn't quite hot enough. Sticks with flux don't require a gas shield as they make their own.


    5 months ago

    Great idea! I think you are really onto something with this concept.
    Have you thought about insulation with stone wool or something similar?

    3 replies

    Reply 5 months ago

    I saw one that you sat on a fire that was a 44 gallon drum and they wrapped that to keep the heat in. This is why i chose a vessel with a smaller diameter as i thought this would not be needed. Grteat sugestion though may try.

    PaPa RaMsKuzmanic

    Reply 5 months ago

    Hey buddy, good work!
    I know this design.
    I have built 3 variations of it so far.
    It can make perfect charcoal.
    It can also break your heart!

    To stop the extinguishing of the flame you need a "turbolator", like the one in the video, which is basically a disc almost cut through from either side then twisted to offset the 2 halfs. It could be anything really, It disrupts the airflow and somehow magically stops the puffing and chuffing.
    Insulation, never mind glass wool or anything else of that sort, it will eventually go up in flames and you get to breath the nasty, which is plain nasty.
    You need ceramic wool or the affordable way, which I did, was to make a box around it and fill with loose vermiculite.

    Your timber feed stock for charcoal and fuel must be super dry. I live in Scotland, its damp here a lot, this is one of my biggest challenges.

    Secondary air. Figure out a way to introduce secondary air, this I believe is a major stumbling block of mine so far.

    Taz above said it.... Bigger diameter burn chamber.
    James Hookway, the inventor of this design states no smaller than 6" or 150mm diameter burn chamber for the rocket stove. Mine is 125mm as that is what I had, but it bugs me to this day if this is the root of my charcoal woes..

    Mind you, best charcoal I ever got out of this was with a cookie flue setup, I had 2 x 90 degree elbows as I was redirecting it out of a window in a stone building, maybe that acted as a baffle, but omg, see when this thing really, really gets going, you will bow your head to forces far greater than yourself....

    Thats all I can think of just now, but please feel free to hit me up if you want to talk turkey some more..

    Good luck!!!

    Really like the temp guage btw, Nice!

    KuzmanicPaPa RaMs

    Reply 5 months ago

    Thanks for your feedback i will look at making a few changes.


    5 months ago on Step 7

    Wood that has dried one year works best. This gives enough time for most of the moisture to evaporate but still retain the volatile oils. The moisture left in the wood has to be heated up to create a gas, which is not flammable. Consequently, the moisture is robbing heat which can be better utilized to volatilize the oils which you want to burn to create a hotter fire. A hotter fire makes better charcoal.

    Cut your wood into smaller lengths, that will enable the oils to get out of the wood easier. I generally cut their length to a few inches.

    You may want to consider an angled feed ramp so your heating wood will be self feeding. Still provide a horizontal tube for clean out and fire breathing. Use the angle ramp for retort access then the tar can drip closer to the heart of the fire.

    Nice job!


    5 months ago

    Great Project & helpful comments too, “green” wood will produce the black stuff & it is flammable. Not sure if it built up in your burn chamber but could cause unwanted results. In regular wood burning stoves it can cause chimney fires


    Tip 5 months ago on Step 7

    You are adding heat to the charcoal only around the central tube. You need to insulate the outside surface of the tank to prevent heat loss. You could also speed the process by lighting up the lower pieces of charcoal wood, loading up the rest, and feeding some air into the charcoal chamber while you are firing up the rocket stove. Then cut off the air to the charcoal chamber and finish.
    Another technique is to load up the charcoal chamber, and light up the top layer just before you put on and seal the lid. Feed air into the bottom of the charcoal chamber. As the fire burns down inside, the upper layers are protected from oxygen by the burning layers below. As soon as you see fire at the bottom, you cut off all air and allow to cool. The bottom layer will not be fully converted, so use this for the next run.


    5 months ago on Step 5

    Pretty impressive, but I'm sorry you had to buy the box section. Don't know where you live but here in NYC and I presume other dense urban areas one of the commonest--and most ignored--street-find materials is the mattress frame. They're made of steel angle iron and run 6 ft long and more. Should be easily welded into box sections.Think it would work?

    1 reply

    Reply 5 months ago

    That's a lot of welding. Maybe a piece of pipe would work, and you can get 90 degree sections too.


    5 months ago

    nice job done on the welding and construction of this piece too.


    5 months ago on Step 7

    there are a few things about this design that make it less efficient for producing charcoal. the most critical one is that the rocket stove is designed to produce maximum heat out of the top of the stack pipe not the side of the pipe.this is why you got an incomplete burn for biochar. the fire that is cooking/pyrolysing the wood must have a long residency time to transfer heat to the compartment that is pyrolysing the material being turned into biochar. perhaps you could lay the machine on its side there by slowing the exhaust gases out the chimney, connect the wood gas tubes to collect the gas from the top of the horizontal tank to several locations spread along the bottom of the outside tank spreading the secondary combustion heat more evenly along the tank. the unit could still be upright to be loaded with wood then laid down perhaps in a cradle to be burned. you will still need to experiment with the size and dryness! of the pieces loaded on the outside /primary burn container and the biochar pyrolysing container. good luck in your development process.


    5 months ago

    I suspect the fire coughed out due to insufficient O2 for combustion. An orifice at the end of the return tube will reduce fuel flow and should help. Perhaps weld a short coupling to the end, and then add a pipe plug with a small hole drilled in it; this way you can change the pipe cap and can experiment with different hole sizes to find the right orifice diameter. Definitely use a lot of anti-seize on those threads if you ever want to remove it. Also, adding a few holes in the side of the square tube near the end will increase the open area allowing for more fresh air draw; if you do this with say one 1/4" and one 1/2" hole it will also whistle letting you know when combustion is dying down. Adding mineral wool insulation will help with the conversion of larger pieces, but realistically I think the only way to get those larger pieces to complete the process will be a second burn. I would recommend just starting another fire in the rocket tube with more raw material once the gas discharge rate gets too low to sustain the fire. A second heat cycle should not damage the completed charcoal but should help get the larger pieces through the conversion process. Alternatively, once you unload the unit, toss the incomplete pieces back end at the bottom with fresh smaller wood on top and run it again; that should force the larger pieces from the previous burn to complete their conversion.


    5 months ago

    Wonderful proof-of-concept; WELL DONE! I'm admittedly ignorant, so I wonder what increasing the diameter of the return tube would do. Also, would there be any advantage to making the return tube removable/replaceable to be able to clean out tar and experiment with its diameter? Two things I don't understand: 1.) Do you have to feed wood through the air intake? Could there be a separate "door" through which you could introduce larger pieces of wood? 2.) Given the high temperatures you achieved, can you speculate as to why the larger pieces weren't completely carbonized? Would it just be a matter of letting the stove run longer? ANYWAY, please forgive me if my questions are stupid or obvious. You've done a fantastic job with this Instructable, and I thank you very much for sharing it. I'm in awe of your knowledge and abilities. With kind regards, Taz


    5 months ago

    Well done! Perhaps the gas gets a bit too rich when the heat spikes. May be a blow off valve and control mech to regulate the gas flow. Awesome build though. That welder probably just paid for itself.