Follow Us:

Browse Flowering Elbow

Latest Blog Entries

Outside FE

Making a wood stove from scrap

Wood Stove From Scrap – 1 Design Notes

We have got to the point where there is no longer a gale blowing through the workshop (yay!): doors are shutting, and fiddly insulation and air tightness detailing is well under way. The obvious next step is to start considering a heat source. Passive solar gain is already in effect – the inside is staying about 5 degrees above the external temp, which is about zero degrees at the moment (yes working out there is cold). Although that’s quite encouraging, and it will be interesting to see how much better things get when the porous straw walls are properly rendered, we are thinking that another heat source is going to be needed. Especially when doing less energetic and more technical hackery.

Having a reasonable supply of wood on the Golden Hill site, a wood burning device of some sort seems like a good, carbon neutral plan. Bongo has been wanting to have a go at making a wood burning stove from scrap materials for some time anyway so… First step: research, think about materials, and dream up some possibilities. Making the simple shape of a wood burner shouldn’t be that difficult – oh yeah, we now have a nice MIG welder, but that deserves a future blog post of its own! The thing is that nice flat plates of scrap steel are not so easy to come by, but what does come up from time to time, are discarded steel containers – from old barrels, to scrap steel safes.

A quick search on the net for ‘DIY wood burner’ shows that using a scrap metal container like this certainly isn’t a new idea- it’s a nice short cut to achieving a general stove shape. A lot of the implementations are pretty crude though. Most are only suitable for outdoor use, with smoke belching from every poorly sealed orifice. OK, I exaggerate a bit, there are some exceptions out there, and the others are good for what they are: quick and dirty wood eaters, like so:

The typical implementation of the gas bottle stove has a door without seals and ill placed air intakes. Better ones have a baffle to encourage heat transfer.

We want something a bit more elegant, and a lot more efficient to go inside. So let’s start with some universals of wood stoves to get us thinking about possible designs:

10 General principals of wood burning stoves[1]:

i) Wood burns most efficiently when it is HOT. Therefore insulating the burn chamber is a basic requirement.

ii) The more efficient a burn, the less polluting smoke comes out the flue, the less ash is produced, and the more heat you produce for every bit of wood you burn up.

iii) The creosote and other particulates in the smoke from wood fires will burn if given the right conditions, namely, a spark, oxygen, and enough heat [2]. This is usually referred to as ‘secondary burn’ [3]. As well as needing enough heat, achieving secondary burn usually requires a secondary air supply (to provide more oxygen in the right place). This ‘secondary air’ needs careful metering – too much will lower the temperature, not enough will result in an incomplete burn.

iv) Even if the wood burns completely and efficiently, this is all for naught, if we can’t extract that heat out into the room – in many ‘super efficient’ compact commercial stoves and diy designs most of the heat goes quite literally up in smoke. In the worst cases, there is no heat exchanger and the internal flue pipes are insulated so the heat goes straight up and out. In some designs, good heat exchangers are built into the stove itself and in some cases these work well, but they are usually very expensive and can weigh a good few tons.

v) Not really a stove point as such and we have touched on it already, but let’s say it again: sealing up and controlling airflow in the space to be heated is essential. If the air in the room changes regularly it will be very hard to keep it warm.

vi) During combustion fresh air is pulled in and all the exhaust gasses should go outside. If you aim for this, and all stoves do / should, you will have to replace the air that is sucked into the stove from somewhere. Thus the cross section of the stove’s air intakes, should equal that of the exhaust flue. Ideally this air inlet should be piped directly from the outside. If it is not the stove will be drawing in air from the room, which in turn sucks cold air into the room from outside through all the tiny gaps in the building envelope. If the room doesn’t have enough gaps, you get smoking and back-draught problems.

vii) To get really nice hot burn and secondary burn the air should be preheated before combustion. So we are going to want to devise some clever way to pipe the air through the fire before releasing it in there for combustion.

viii) Making sure exhaust gases don’t leak is important. Carbon monoxide is odourless and is the primary danger to address with wood stoves. A carbon monoxide monitor/alarm seems like a good idea.

ix) To maintain enough ‘natural’ up-draught to draw exhaust gas up the flue and provide fresh air for combustion, the air temperature when leaving the flue to go outside should be about 120º C  (250 ºF). Fans can be used to reduce this, but more on that later…

x) Accurate monitoring and feedback of a wood burner’s performance is the key to adjusting the running conditions correctly. There are a few obvious signs of a good efficient burn: nice clear to light grey smoke (white smoke clouds signals polluting inefficiency) but other readings can be useful. A thermometer in the flue, reading the temperature at the point gasses leave the room is very useful (see previous point). Also good, an O2 sensor, giving an Oxygen reading in a similar location, will tell us a lot about the efficiency of combustion and allow us to optimise the amount of air going in for primary and secondary burn.


Some possible stove design ideas


Standard canister style stove with air preheating (aka a posh pig stove):

Many have been built in this style, but few have preheated air to allow for efficient primary and secondary combustion. We are not going to be burning coal so there is no need for a grate (according to most sources wood burns more efficiently on a layer of insulative ash). In this horizontal design we could line the walls with lightweight fire brick for added insulation and pipe in the preheated air in two heavy gauge metal tubes: one large bore (say 3 to 4″) for primary air supply, and a smaller one for secondary. Something like this:



 The Rocket-mass stove style: sideways burning down draught:

As well as having the air ‘naturally’ preheated as it is pulled over the hot coals, side and downwards burning stoves have another significant advantage: they can metre out the fuel because it is fed by gravity into the burn chamber. Please excuse the drawings here, but hopefully you get the general idea:

There is an interesting video comparing the classic wood stove and a ‘rocket mass heater’, and a load of stuff on the rocket mass heater here.  The semi-self-feeding mechanism is good, because to burn efficiently (ie hot), things burn quickly, and you don’t want to have to be putting new wood on the fire every two minutes. It also means that you can feasibly burn smaller sticks, which are typically wasted wood. What’s more, if dry, this kind of wood burns hotter because it presents a much higher surface area at which combustion can occur than say a big log (big logs that last ages but smoulder and cool the fire = bad) . This is why pellet and wood chip boilers can achieve such high efficiencies.

However the rocket-mass heaters all have a very large thermal mass, usually a clay or cob bench. That’s great for the home, but really I think we want an air-to-air heat exchanger so we can quickly warm up the workshop. So our possible design will leave out thermal mass, and exhaust to a heat exchanger (possibly constructed from another cylindrical container or two) instead. I think we would want to have preheated primary and secondary air supply too:) It might look something like this:

One cylindrical scrap container could make the main body of the stove, while 3/4 of another can act as the combustion and fuel feed chamber. Here the heat would be controlled by metering the wood fuel, rather than restricting the airflow, which can lead to inefficient burn. It should be possible to burn a certain number of sticks, to give various different heat outputs. Both secondary and primary air are preheated and blown into the combustion chamber from outside with small fans. The whole stove is insulated with a separate heat exchanger mounted above.

Becoming a fan of fans in stoves

Fans seem like the solution to so many problems in wood burning stove design. OK so they add complexity, make a little noise and require electricity but… They can boost efficiency of the stove so that, in effect, they produce way beyond the amount of energy they use. Here’s how:

1) They can be used to force the room air through the heat exchanger which vastly improves the effectiveness of the  heat exchanger and/or the need for such a big one.

2) They can be used to force air into the combustion chamber as in the design above. Because the air is blown, and we are no longer relying on the natural up-draught of the flue, we can do two things we couldn’t before: First, we can heat all the air (primary and secondary) fuelling combustion. Second, we can extract ALL the heat from the flue gasses, so that the exhaust air leaving the flue is the same or similar temperature to the room air [4]. And lastly, we can move the air into primary and secondary burn zones, right where we want it, in high-velocity, low-volume jets that help to mix up fuel, air and fire further improving the burn.


This is all great, and I think the heat exchanger will certainly feature some peltier driven fans to improve performance, but pumping in preheated air for combustion does raise some practical concerns. If it works as hoped we will be getting up to melting-metal-temperatures in the combustion chamber, and only refractory cement or similar materials will endure. I guess we will want to make the fans’ speed easily controllable, so that we can carefully adjust for optimum burn, without going OTT and melting/cracking the whole thing.

So yeah, that’s the thoughts on design so far – sort of settling on the kind of thing in my last three diagrams. I am torn between having a viewing window vs not. Having one would be nice, but doesn’t seem like it would work as well with the sideways burning design. Having a glass front to the combustion chamber is always going to be a weak spot in term of insulation, but then again, you can tell so much about how a burn is working just by direct observation – it may well be worth having glass.

Now if we can get some other stove builders, designers, and users to chime in with ideas, experiences or advice, we might soon have a ‘part 2’ of the stove mission :)

  1. [1]Compiled from various sources – some good ones are: David Hood 2008 “Biomass: Sources and Technologies”.  Mark Bryden, Dean Still, Damon Ogle & Nordica MacCarty “Designing Improved Wood Burning Heating Stoves”, see  Chris Laughton 2006 “Home Heating with Wood”
  2. [2]Ignition Temp of wood gasses: Hydrogen 400ºC (750ºF), Carbon Monoxide  607ºC (1125ºF), Methane 538ºC (1000ºF)  – according to Borman & Ragland (1998) “Combustion Engineering”
  3. [3]Secondary burn is the reason downwards and sideways burning stoves can be so efficient. The smoke produced at the initial burn point of the wood is then dragged down (or across) through the hot embers, which further heats and ignites the smoke.
  4. [4]This can still only be done if the combustion itself is working efficiently, and all the creosote is being burnt, otherwise you will get deposits building up and blocking the heat exchanger. When the fire is first started this will always be a problem – but because fans are so controllable, if you use one on the heat exchanger you can have that off or set on slow until the burn is up to temperature. In fact a small fan powered by a peltier would self regulate its own speed in relation to the heat – could be ideal?

Responses to Making a Wood Burning Stove -1 Design

  1. Dave


    Good conceptual plan so far, now I have seen the bulked out diagrams i see how the self feed will work, previously I was confused about how it was going to seal or whether it was going to seal at all. I like this self feeding fuel aspect very much, currently burning thin(1/2″ to 3 “) stuff in my conventional stove and getting good results in comparison to the open fire we had last year, but the reloading frequency is quite high. I wonder if all goes well you will be able to produce fuel ‘fagots’ held together with twisted split lenghts of stick ready for quick loading?

    I’m liking the idea of the viewing windows from a control and learning point of view alone.

    You will see when you pick up the stuff we ordered from the internet the other day that I have included the CO2 detector we looked at, not that I doubt the design will work but better safe than sorry. Control and correct function of the fans will be the tricky issue, as there is going to need to be some cut off system on the fans when loading fuel or the workshop is going to get smokey quickly. I can however voucher for the effectivness and consequent loss of effectivity of a conventional wood stove pulling cold air into a room while running, since the new windows have gone in at masion dave the draft under the door has become comical. And our stove is rated at below 5kw the point at which a air vent is legally required.

    In regard to the heat up of the base issue, what hearth material are you considering as some of the heat build up could be disapated into the floor if configured right, obvisously not directly sat on the floor as the concete will prob crack up? I’ll do some learning on this…

    All in all looking forward to warm workshopping, wouldn’t a stirling engine heat exchange fan look good, if potentially noisy.

  2. Bongo

    Hay Dave,
    Agreed, there will need to be some fan control mechanism to shut the fans off when the door is opened. Shouldn’t be too difficult, though temperature resistance of the electronic components is always going to be the issue…

    In many ways a fan mounted near the end of the flue, which pulls the air, would be superior. But you then need a pretty robust and heatproof blower. Not sure how easy that might be to get hold of / hack together??
    Fans at the inlets that push air in could possibly be mounted outside, at the entrance to the inlet pipes, so you wouldn’t be able to hear them inside. I can think of a few potential sources for suitable scavenged fans that should work for that…

    Like your sterling engine idea, would be a nice project, but think the noise may cause mild insanity after a while. A peltier one would be good.

    As for the stove base, i was planning to have it insulated to protect the metal (which probably wouldn’t deal well with the intense heat in the combustion chamber), and lined with a hard refractory cement, and the actual stove up on little legs, to hold it off the floor. I guess the idea of putting heat into the floor is what the rocket-mass stoves are all about.. I just think the time delay between lighting and getting heat out of that kind of system might make that less convenient for the workshop?

    Agree about the glass – would be very useful!

  3. Dave

    interesting that the rocket mass stoves are very similar in some ways to the design of stoves I have seen in slovina where large portions of the main living room are given over to a large ‘heat mass’ cover in tiles, often with benches incorperated into the desgin. In use they seemed quite effective with a relatively short burn resulting in the stove being warm to the touch many hours later.

    I suspose the desgin for the work shop is influenced by the fact that in many ways the workshop is more insulated that the average house, certainly more than ours, and a short efficient burn should warm up the space quickly and then remain so due to the insulation.

    what diameter flue are you considering using? as this can influence the soot build up if there are dramatic changes where the transit through the flue drops in a sort of eddy effect.

  4. Bongo

    Well we have a old stainless flue from a retired oil boiler lined up for it. I think the ID of that is about 6″. I looked the other day and thought it would be sufficient, didn’t actually measure it though – should be enough.

    You are right about the soot build up, that will be one of the design challenges of the heat exchanger. Really to get the best efficiency the air wants to be running through very narrow flat passages, where as much it ‘scrubs’ the metal walls as possible. Now the trick will be in having such an arrangement without adding too much air resistance, eddies and all that. I guess we will need one or more removable hatches so it can be cleaned – but we don’t want to be doing that too often!

    Again the fans will help here – hopefully they will keep the air moving enough that this wont be a problem.

  5. Travis Skinner

    Hey bongo,

    Sounds like a great project and with your climate, quite necessary. I’m no expert on stove design, but I did just build a propane forge for blacksmithing. I’ve found that ceramic insulation covered with refractory cement and coated with ITC-100 works wonders for insulative material that can with stand very high temperatures (over 2000 degrees. The ITC-100 is a high temp coating that significantly increases the life of the refractory material and increases temp by reflecting “radiant heat back into the furnace or forge to achieve high temperatures quickly and efficiently.”

    You can get the ITC-100 here

    I have read there is another product called Plistix 900F that is cheaper and works as well as ITC-100, but I do not have any experience with this product.

    I thought that this might be helpful if you are worried about melting your metal, as the ceramic insulation with with refractory cement and ITC-100 has worked well for blacksmiths working at very high temps. Just a thought.

  6. Bongo

    Hay Travis,
    Thanks for the tip. Had a little look at your forge build on your blog – very nice. The ITC-100 link seems to be dead, but I have done a little searching about on google for it. Sounds like good stuff but super expensive.

    I also wonder if it would be a bit waisted in this application: if you are using propane burners and placing metal in the forge, that is quite a clean environment compared to a woodstove combustion chamber. I wonder if its ability to reflect the heat back into the stove would be limited/neutralised by the ash and soot and general blackness of a wood burner?
    Also not sure how long it would last with bits of burning wood scraping on it? Any ideas as to how abrasion resistant it is?

    Thinking about it, it might make a good lining for the secondary burn chamber though…

  7. Eric Rousseau

    do you know SEDORE WOOD STOVE? you use the same principle. but your is more axe for recup heat. i like your stove

  8. Bongo

    Hi Eric. Thanks for the comment. I just watched a you tube vid of a sedore wood stove. They look good! Sounds like the secondary air needs some work though (or maybe that was just an idiosyncrasy of that particular owner?

  9. Bax2theMax

    Hey if you could post some pics of the completed, that’d be awesome. Currently designing a stove myself but have very little experience with wood stoves. The tank I am planning to use is a 250 gal propane tank, 3/8″ thick walls. Not planning by any means on using the entire tank as it’s 38″ diameter x 7′ long. However I do plan on using a decent portion of it as a fire box.

Add a Comment, Question or Musing

What is 12 + 3 ?
Please leave these two fields as-is:
To comment, answer the following, so we know that you are a human :-)