The Second First Rocket Stove

It was on my mind for many months … rebuilding the first rocket stove. It worked good – though not great. There were some design errors and compromises and during the last winter there were more smokeback events that I could not explain. I wasn’t keen on taking apart something that (kinda) worked and embarking on another build project. So I played around with it in my mind for a long time. Eventually I had a design I felt comfortable pursuing, I did a simple cob test to confirm the materials I had … counted and purchased bricks … and decided to go for it.

At the last minute I decided to give it a chance to become a workshop build and so I published an invitation and sent out word to people I knew and thought may know other people who would be interested … this was 3 or 4 days before the planned weekend build … and once person did sign up … making him the first participant in the first workshop I have ever offered at Bhudeva. I had two pairs of helping hands – Annelieke and Horatiu.

The project was born when I did created this layout:


I was able to take real measurements, finalize brick counts … and get confident enough about my vision to move forward. The first thing we had to do was to take apart the existing stove … which was magical … the knowledge that most of the materials can be reused … that the rest are non-toxic and can simply be tossed out anywhere on the land where they will be reassimilated by nature … its one thing to know this and another to experience it:




I was surprised to find the metal heat riser mostly in tact … though it was dry and chipping. Most of the clay-perlite insulation was used in thew new build …  which … began by recreating the layout in place to find the exact position it would be in relation to the existing chimney.




With the position fixed we were able to get to work on building a raised floor:


And then, layer by layer, building up the core of the stove:




… and when we brought in the barrel for a first fitting it started to feel like it just might become a real life rocket stove:


In the following image you can see the experimental part of this build. I discovered these honeycomb bricks and decided to use them to easily create heat channels and storage mass. There are two air passages (barely visible in the image) that allow the hot gasses to flow from the barrel into the two-brick chamber on the left hand side of the image – where they flow up. Then (as can be seen in later images) there is a top chamber that allows the gasses to flow across and down the two-brick chamber at the top of the image (right up against the wall) – where they flow down and then out through the chimney. There were three experiment going on: 1) using honeycomb bricks; 2) introducing a vertical flow both with both bottom-up and then top-down flows; 3) and gaining improved heat storage by having mass outside (the shell of the bricks) and inside (the honeycomb pattern).


This is as far as we got in the two days of work we had available. Horatiu and I agreed that he would come back for another day of work during which we will complete the build and fire it up for the first time. So during the next few days Annelieke and I continued doing some preparatory tasks. The most prominent task was the heat riser. Annelieke started doing a perfect and wrong job. Can you guess what is wrong in this image:


Annelieke is doing fantastic work getting the bricks aligned and leveled … but she is laying them without overlaps … creating a beautifully symmetric and unstable structure. This is something I take so much for granted that I did not spot until a few more layers were built and it became very prominent. So It had to be taken down and rebuilt properly:


While she did that I built some insulation chambers around the core (to extend the insulation that would be placed around the heat riser) and started filling them with the clay-perlite mix from the old core … and as you can see in the bottom-left corner I started playing around with cob … hoping for a better experience (I’ve had very poor experiences in the past):


On the day Horatiu came back we finished building up the honeycomb brick chambers and the top chamber in which gasses could pass from the up-flow chamber to the down-flow chamber:



The top chamber was closed with bricks and we then added on the sheet-metal container for the insulation:


… filled it up with clay-perlite insulation:



… and sealed it:


… and suddenly that was it … everything was ready for a barrel:


… and lighting a fire … the smoothest lighting of a new rocket stove I’ve ever experienced … excellent draft (probably helped by the fact that the core had a few good days to dry):


I was then left on my own to slowly transport cob-worthy material, to mix it up in reasonable one-person batches … and slowly build it and transform the stove from something very mechanical and engineered to something organic and mysterious:




There were a few places where seeds apparently got into the cob mix … and given that there was a lot of moisture inside this happened, in a few places:


It now, though still slowly drying, looks like this:


The second lighting of the stove, in contrast to the first, went very poorly. I am guessing it had to do with the loads of cool moist cob. This is where the experimental part may have also kicked in … the gasses may have had a hard time establishing a complete and continuous flow throughout the stove, resulting in serious backsmoke. During the third lighting I was careful to preheat at both cleanouts, to start very gradually and only when the stove was flowing well to put in a full load … and … to my great relief … it ran perfectly again. The fourth lighting was not so good … I wasn’t as patient. Since then I’ve lit it a few more times and it has been going fine.

I estimate that, aside from the bricks, I put on over half a ton (maybe up to three quarters) of cob. Thats a lot of moisture. At the end of the first lighting (before cob went on) when the full load finished burning the bricks at the back were noticeable warm. Not so during the next few lightings. There are many liters of water in there that need to dry. This is something that should be taken into consideration in a construction schedule. I started the construction early so there would be time to experiment and make corrections. I did not take into considerations how long this would take to dry … it still is drying.

Cob was much more friendly this time … finally. I played around with different finishing techniques … I still am. It is a subtle thing finishing and there seem to be numerous paths to go about it. It is very pleasant work (when it works) to be able to mold shapes, smooth corners, add colors. It felt like a complementary and balancing process to the more structured, measured, aligned process of building the core. It felt free, open, secure,  … embracing and welcoming. It is a pleasant way to finish a build and a much more pleasant result 🙂

The stove has already worked for a few cold nights. It’s still hard to say how good it works because: its not that cold yet, there is still humidity in the mass, the barrel itself is partly wrapped in cob … so a few things still shifting and changing. I am looking forward to experiencing how it works … both the immediate heating and the heat storage for the night. I have a feeling that it is going to be more efficient in terms of wood consumption (then its predecessor) … I am curious how it will compare in terms of heat storage (the previous stove was all storage, slow to heat up but then radiated plenty of warmth throughout the night – sometimes even overheating the room).


  1. klemen
    Posted October 14, 2014 at 7:30 pm | Permalink

    Nice! I am preparing to make a finish layer of clay plaster on my stove… Did you use anything to fix the plaster not to be dusty/abradeable? I am thinking of using casein….

  2. Posted May 3, 2015 at 4:36 pm | Permalink

    For a easy start on a rocket stove you can light a paper in chimney. The heat from that paper will establish the draft and the main fire will start easier.

Leave a Reply