Editorial to the Industry

Getting Our Act Together

(Editorial to the Woodburning Manufacturers Industry)
(published in the National Woodstove Journal-August 1981)
(about two newspaper-sized pages!)

My recent visits to the wood heating industry trade shows caused me great concern in the area of safety in new product designs. in addition, I noticed that people are generally becoming more concerned about the side-effect hazards of air-tight wood-burning devices. I was quite dismayed however, to find that few of the representatives of the manufacturers present had even the slightest inkling of the causes of the problems, much less what to do about them. I have decided to present a VERY brief exposition here on the definition of the problems, their sources and the approach that should be used in correcting these situations.

Back starting in 1973, I probed deeply into the problem of design of an improved woodstove for my own personal application. (This evolved into the development of the JUCA line of wood-furnaces and heaters). At that time the concept of air-tight woodburning was relatively new and was rapidly gaining popularity over previously available products. I seriously considered using an air-tight design at that time, but after considerable research and testing, I decided against it on safety grounds. As it turned out, it was possible to make a very highly efficient woodburner without resorting to air-tight operation and its problems.

When an air-tight or air starvation heater operates, the fire necessarily burns rather poorly because of a lack of oxygen. Of course, that is because the control of the fire is by a partial suffocation of the fire. This lack of sufficient oxygen allows only part of the wood's fuel value to be recovered. The lost fuel leaves the fire primarily as unburnt hydrocarbons in the smoke. The largest portion of these are generally carbon monoxide and chemicals called creosotes. Under the conditions of a severely held-back fire, fully 1/3 of the wood's chemical energy is wasted in this way.

Of course, there's a concern here for the tremendous amounts of heat wasted in this way, which tend to cancel out some of the gains in output gained by the air-tight operation. More seriously, though, is the area of side-effects. The actual quantity of carbon monoxide and of creosote is dependent on many variables such as moisture in the wood and wind conditions, but the actual gross amount produced in air-tight operation is really astounding under some conditions. It looks like owners are going to be instructed to check their chimney system WEEKLY in some situations. If the chimney did accumulate a lot of creosote, then it could either catch fire (called a thermite) with a ferocious fire that could cause safety dangers in inadequate chimney installations; or if it blocked off the smoke path (in the chimney cap, for example) then the carbon monoxide produced would be forced back into the house, with potentially grim results.

Text Font Face
Text Size
(for printing)
Clearly, a solution should be found for this situation, and one does exist. It is called "secondary combustion". The idea is that since an air-tight stove did not allow complete combustion due to lack of oxygen then a second burning could be produced using the unburnt fuel gases in the smoke and an alternate supply of combustion air just for this purpose. It sounds ideal because it would: improve the overall efficiency of the stove; reduce the amount of creosote that was present in the smoke so less would likely condense in the chimney; reduce the amount of carbon monoxide in the smoke. There's only one hitch. The primary burning reaction is that of carbon monoxide plus oxygen gives carbon dioxide (the natural, safe product of complete combustion). Unfortunately, this reaction can only occur at temperatures over about 1200°F.

Under some burning conditions, the smoke is at a temperature higher than that. For argument's sake, let's say it is at 1400°F. Now if we mix that smoke in equal amounts with room air, the resulting temperature of the mixture will be less than 800°F, far less than the temperature necessary to initiate secondary combustion. You can probably see that you are going to need the secondary air supply to be at about 1000°F or higher under these conditions. Unfortunately, there are some conditions of low fire (severely held back) where the smoke itself is under 1200°F within inches of the logs. In that case, secondary combustion is almost out of the question as the fresh air would have to be so hot as to heat up the smoke to cause burning. It's ironic that in the situation of a severely starved fire that most needs the effect of secondary combustion, it is most difficult to obtain. When the fire burns relatively freely with less wastage and creosote, that is when it would be easier to initiate secondary combustion.

In the past two years several independent researchers have established that in currently available products secondary combustion is virtually totally ineffective. This seems to have caused some manufacturers to abandon the concept. This is a mistake. I truly believe that the only way that an air-tight woodburner can be considered safe to use, is if it has a highly effective secondary combustion system. This thought also applies to coal-burners, too. The reason for the ineffectiveness of currently available systems is that the pre-heating of the secondary air supply is not adequately accomplished. The air that should be in the 1000°F range is often actually only 300°F or 400°F. That actually worsens the situation because it cools the smoke so that more of the creosote in the smoke will be likely to condense in the chimney, and also just wastes a little more house air. I present here a way to calculate the area of heat exchanger necessary to pre-heat the secondary combustion air to a desired level:

For the application of these calculations to a particular model stove, a few pieces of info are needed. Most manufacturers have measured the smoke temperature as it leaves their product. Let's say it is 700°F. And it's not difficult to determine the air flow rate in the primary draft inlet of the stove. Thirty-five cubic feet per minute is not unusual. Finally, the skin temperatures at many points of the body of the stove should be available.

It is necessary to make certain assumptions about the method of operation of the stove and some other variable factors. We will assume here that for thorough secondary combustion to occur, at least as much secondary air will be necessary as primary air, i.e. at least 35 CFM in this example. In the example here, this brings us to the point of having to use heat exchangers to heat 35 CFM of air to about 1000°F in order to gain the useful heat of the secondary combustion. This is no easy task. The amount of heat that must be transferred to the secondary combustion air is given (approximately) by (CFM of air to be heated) times (Temperature rise needed in this air) times (Specific heat of air at the mean temperature)/(# of cu. ft. of air=1 lb at the mean temperature)/ (# of minutes in an hour)

or (35)(1000-70)(.24)/(28)/(60) or approximately

17,000 Btu/hr.

When we're talking about a unit that is only going to develop 20,000 or 30,000 Btu/hr of useful heat, it seems unreasonable to go to considerable effort to transfer 17,000 Btu/hr TO the smoke that's leaving the house. In some cases it is a losing proposition as to the energy balance equations: more heat is used to pre-heat the secondary combustion air supply than is developed by having the carbon monoxide combust, although this is a very energy rich chemical. Each pound of carbon that is carried away as carbon monoxide leaves with over 10,000 Btu of energy that would have been released if total combustion had occurred. For energy balance considerations, it would be necessary to have present approximately 3 1/2 lb of carbon monoxide in an hour's smoke to be able to recover enough to "break even". By the way, that amount of carbon monoxide would take up about 110 cu. ft. of space in the smoke (quite a lot).

The energy balance of the situation should not be the primary consideration here though. The safety implications must be paramount. Therefore, even when the existence of such an extensive pre-heater does not prove itself in improved NET heating efficiency, it MUST be designed in for safety considerations IF THE PRODUCT IS TO BE AIR-TIGHT.

Now to calculate the approximate size of heat exchanger necessary to make the secondary air system effective most of the time. Again using the example above, let us estimate the smoke temperatures between the fire itself and the measurable smokepipe temperature as being directly proportional with the distance along this smoke path. (That's actually not quite true, but the accurate values are complicated to obtain, and the resulting errors are not too great) Making another assumption that the fire's immediate vicinity (within 1/8", say) is at 1900°F during a moderate fire, we can find a location of, say, 1300°F smoke. Now most of the preheating will probably have to occur around the sides or back of the stove where the effective temperature is near our 700°F outside surface temperature. As soon as the pre-heated air gets near 700°F, it won't get any hotter unless you then pass it through the 1300°F area (in our example). Let's say that we will try to heat to 600°F along the sides, and then get to 1000°F in the 1300°F area. To calculate the areas needed is a standard problem in heat transfer with the variables as presented above:

Radiation source temp = 1900°F
Radiation- heater tube mean temp = (700-288) = 412°F,

Using simplified versions of the standard heat-exchange equations:,
RADIATION IN= 1.7/1,000,000,000((absolute source temp) to the fourth power - (absolute temp of tube) to the fourth power),

OR 1.7/1,000,000,000((2360)4 -(872)4)

OR about 5200 Btu/hr/sq ft of exchanger.

Since about half of the total of 17,000 Btu/hr will be needed to get the air to 600°F, we will need about (17,000/2)/(5200) or

1.6 sq ft of exchanger.

If it is a 2" diameter tube for example,
it should be 9.6 feet long (wrapped around, of course).

SECOND EXCHANGER: For radiation (as above) the tube's mean temperature is 1300 - (700-300)/(LN(700/300)) or


The radiation calculates to be

4800 Btu/hr/sq ft.

For heat picked up from the hot smoke passing it:, The convective heat is given by .19(mean temp diff between smoke and tube ) to the 1.33 power OR .19(472)1.33

OR 690 Btu/hr/sq ft.

The total gain for this second exchanger is about 5500 Btu/hr/sq ft. The remaining 8500 Btu/hr must be put into the air supply in this exchanger, so it's area must be (8500)/(5500) OR 1.5 sq ft. This second exchanger would be slightly smaller than the first one is.

I am not aware of any stove on the market that has even a quarter of a square foot of heat exchange area for pre-heating the secondary air supply, while they may need around 12 times that area. It is true that many extreme approximations were used here to try to keep this understandable to many people, but the overall premise is reasonably accurate. Some shading effects, sooting effects, and geometrical shape effects, if included would tend to require even more exchanger area.

I found these things out back in 1973-74 and chose to not design around such a difficult problem. Instead, by allowing relatively free burning, a reasonably complete combustion occurs that produces far less creosote and carbon monoxide in the smoke. It turns out that by doing so, it becomes possible to put extensive USEFUL heat exchangers that remove a large portion of the smoke's heat, even below the creosote deposition temperatures. The little creosote does tend to condense out but it is a surprisingly small amount. Very little normally accumulates in the chimney system. We abandoned air-tight operation because we felt we obtained clean, complete combustion; very effective heat exchange; comparative safety as to side-effects; and a constant output by using large diameter logs.

The JUCA Home Page is at: juca