Permaculturing in Portugal

One family's attempts to live in a more planet-friendly way

On rocket stoves …

It seems rocket stoves are as much part of the natural building vernacular as glass bottles in cob walls: de rigeur for any self-respecting stomper-of-mud, stacker-of-straw and fashioner-of-eccentric-curves. Being innately somewhat contrary and suspicious of fads and fashions, even ones I’m participating in, this fact alone would usually send me running in the opposite direction. But reading about rocket stoves, I was attracted by their low tech simplicity, their apparent ease of construction, how they lend themselves to self-build projects, how they can be made from junk and be fueled with the small branches and sticks that are no more than kindling for more conventional wood-burning stoves, and how efficient a burn they can achieve. So they were penciled in firmly for the buildings here – for cooking and heating water – pretty much from the start.

But theory is one thing: practice another. With a big push on the main building planned for this year, it was time to start experimenting – constructing different configurations of firebricks and clay and stuff and firing it all up to see what works and what doesn’t.

Rocket stove core

I started with ideas for a cookstove, based on the L-profile cookstoves developed primarily by Larry Winiarski of Aprovecho Research Center for use in developing countries. I wanted to see just how practical these stoves would be to use on a daily basis for cooking, so my initial aim was to put together the core of a rocket cookstove – the burn chamber and heat riser – and see how easy it was to light and to run. For this, the only critical dimension is that the heat riser above the combustion chamber is at least 2-3 times its diameter, so I built a 200mm diameter heat riser from the ends of stacked firebricks rising 600mm above the combustion chamber. Air gaps between the firebricks were sealed with a clay plaster mix.

Rocket stove core

Lighting the stove was simple and, allowing for the temporary nature of the construction, it drew and ran well, but after a weekend of playing with it, I kept coming back to this. A well-stuffed feed hole seems to allow the perfect amount of air through the gaps between the sticks and under the grate to support a good burn, but the rate of consumption means that the stove requires almost continual attention to keep it that way if the burn is to be kept efficient, especially if burning softwood. (And if it’s not kept efficient, then no less attention is going to have to be given to cleaning out flue ways on a regular basis …)

Rocket stove core

It seems like an awful lot of constant fiddling and effort when there are tried and tested stove configurations with a long history (kachelofens, masonry stoves, Finnish contraflow stoves, etc) which can heat mass and burn just as efficiently and a lot more controllably with a minimum of attention. The heat riser works flawlessly and well, but I kept finding myself wishing for a larger combustion chamber to fill with larger pieces of wood, and a door to better control the air flow.

The more I thought about it, the more it became clear that while it’s useful to have a stove that burns small pieces of wood efficiently, wood is wood is wood. In other words, within any one tree species grown under similar conditions, a given mass of wood has a reasonably constant calorific value. The bigger the chunk of it, the greater the amount of heat that can potentially be produced. A burn chamber that’s limited to a certain cross-sectional area and which requires a certain amount of air to be drawn through gaps between pieces of wood to maintain the efficiency of the burn, can only run within a limited temperature range. A burn chamber that’s not limited in this way, which can hold a larger mass of wood, and which is not so dependent on how the wood is loaded (and kept loaded) to maintain the efficiency of the burn has to be more flexible, more controllable and less trouble to run, as well as capable of running within a larger temperature range for potentially no loss of efficiency, temperature being controlled through both fuel mass and air flow relatively independently rather than interdependently.

Having closely watched how the wood-burning water heater, or bailarina, in our cob bathroom performs – which is essentially a rocket stove heat riser with a less limited combustion chamber – it seems to me that this is the case.

Then I came across this post from Alex Chernov written to the Masonry Heater Association of North America’s members, and Max Edleson’s reply on his own website and thought YES!! Exactly! Running these thoughts by a bunch of natural builders on a Facebook group served only to confirm these suspicions.

My thoughts now are following the lines of using a rocket stove-type insulated heat riser to deliver heat from a more conventional firebox with door to a cooking surface, but I’ve yet to investigate Finnish contraflow heaters …

Next Post

Previous Post


  1. Joe April 16, 2013

    The mistake a lot of people make with rocket stoves is that they follow the design but use the wrong materials. The most important aspect of a rocket stove is a super insulated combustion chamber and chimney to keep the burn temperature as high as possible–normally fire bricks made with a perlite and clay. It looks like you’ve used normal clay bricks on your stove and they will absorb a lot of the heat through conduction into the mass, keeping the burn temperature way down. This sort of defeats the object of a rockets stove, the object being achieving an efficient burn and therefore using less wood. If you can find some vermiculite, you can make some bricks to line the combustion chamber and chimney. You should then the stove burns much more efficiently.

  2. Quinta do Vale April 16, 2013

    Joe, this was just a test configuration to see what a rocket stove is like to run. Had I built one, it wouldn’t have been using these materials. My comments about the nature of the stove still stand. Personally, I reckon masonry stoves have the edge. The critical point is the combustion chamber and the separation of air flow control and fuel status. A firebox with a separate draft regulator does this very well. I want to be able to light the fire, load the combustion chamber, set the air flow for an optimally efficient burn, and walk away. With a masonry stove I can do this. With a rocket stove I can’t.

    Rocket-type heat risers can be found in plans for masonry heaters going way back and in such things as Badeofens/bailarinas – wood-fired water heaters. Such heat risers aren’t super-insulated, admittedly, but the important point to me is that firebox design has evolved a long way through constant experimentation over centuries of use in Northern Europe/Russia to arrive at something which is every bit as efficient to run (or more so), stores heat every bit as effectively (and over a smaller footprint) and is a whole lot easier to use.

  3. Joe April 22, 2013

    OK. In regards to the constant attention to keep it fuelled, maybe a vertical feed to the combustion chamber, using longer pieces of wood will solve that problem. I’ve always thought, though I’ve not tried it yet, that seasoned, coppiced wood would be idea for a rocket stove. A few lengths put into a vertical feed should be good for a fairly long, unsupervised burn. Just a thought.

  4. Quinta do Vale April 22, 2013

    That’s a possibility, though vertical feed tubes apparently do tend to smoke when the fire is first lit. This is a cookstove design, so a vertical feed is less than ideal also, since it requires space which could otherwise be occupied by cooking surfaces. But the main point is this: I see absolutely no point in persevering with a rocket stove design when excellent time-tested designs exist for building low-tech cookstoves that perform every bit as well and are much easier to operate! It seems like a no-brainer to me :-)

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

© 2024 Permaculturing in Portugal

Theme by Anders Norén