6″ Dragon Burner masonry heater using chimney flues – Part 3

The re-design of the flue solved the thermal stress issues and improved both the thermal curve and exit temperatures.  The temperature profiles and Testo numbers are included in the charts below, but showed low stove emissions and high efficiency.

Testo Gas Analysis

Here are some charts from the Testo 330 Gas Analyzer. We only ran it for 90 minutes of the 3 1/2 hour test. The analyzer is not really designed for long test periods. The filters and readings drift and go wonky after a while. We start all charts when the O2 number drops below 20.


The 1st chart here graphs the increase in negative pressure vs the stack temperature. Starting at ambient, (around 94°F) through 155°F or so towards the end, you can see what a big difference the temperature differential makes to the drafting.

Dragon Heater ( Rocket Heater shippable core ) Efficiency and Emissions for 6 in. Flue Build

Dragon Heater Flue Build #2 Efficiency and Emissions (Click To Enlarge)


CO was impressively low. Its carbon monoxide levels were low from the very start.  Cooking on a gas stove is an exposure rate of 100 ppm, and smoking a cigarette is 400-500.

Efficiency - Once again even the efficiency numbers were very good from the very start, over 92% efficient and never dropping to even 85%.

Dragon Heaters are exempt from EPA requirements. It is considered a constant burn stove because it has no option to reduce the levels of oxygen. Many cast iron stoves attempt to control the heat by controlling the amount of oxygen allowed. However, it’s performance would clearly exceed even the most stringent efficiency requirement of 68% efficiency in catalytic converter stoves.  Masonry heaters require 58% efficiency. It is not unusual for plain old cast iron stoves to only operate at 25% -30% efficiency.

Dragon Heater ( Rocket Heater shippable core) Oxygen and CO2 data, Flue Build

Dragon Heater Flue Build #2 Oxygen and Carbon Dioxide (Click to Enlarge)



As you can see in the chart of all temperatures, the fire was allowed to die down at about the 9000 second mark, about 2.5 hours into the burn. We left the temperature probes on until the 4 hour mark. So the last portion of any of these charts is after the main wood has been consumed.

 Outside surface temperature of both bells

In the chart below you can see that both bells have a similar surface temperature even though the 1st bell has an additional column of flues that the heat must migrate through before reaching the surface.  After the fire is gone, the 1st bell outer surface continues to increase in temperature as the heat from the inner flue migrates to the outer flue surface.

Rocket Heater Masonry Heater built with Dragon Burner and chimney flues

Surface Temperature of the top of the Bells

Temperatures inside the 2 bells

Temperatures inside the 2 bells show that the bottom readings of both bells is fairly close to each other and the exit temperatures.  The 2nd bell continues to collect heat from the slower draft of the dying fire.

Temperature chart of a double bell masonry heater built with 6" dragon burner

All Temperatures

This final chart shows the temperatures of the heat riser vs all the other probes. You can see how much the heat is dispersed across the 2 bell resulting relatively low temperatures everywhere except the outside of the heat riser.

Chart of double bell masonry heater built with a dragon burner and chimney flues

Shows the temperature as it leaves the heat riser versus various locations in the stove. (click to enlarge)



We think the exit temperatures could be a bit higher, but we were running this at over 93°F ambient, not exactly cool weather. Draft should increase during more realistic winter conditions.

The temperatures outside the heat riser column rose to over 350°F even with the firebrick lining. It was not logged well because the adhesive on the tape holding the thermocouple failed and it would not stay on the flue.

In order to lower the temperature on the flue surrounding the heat riser, we will do an additional revision. We will change most of the interior liner from fire clay brick to insulation. Making this area cooler will  to allow for skinning the stove in tile or stone.  It will also force a bit more heat into the bells, which have plenty of capacity for additional heat.  With the change to insulation around the heat riser exhaust, we are hoping to push the exit temperatures a tad bit higher.

In general, we were pleased with its performance and think it is a winner. It

  • Small footprint 30″x36″
  • Drafts well even in summer
  • Low CO emissions
  • High efficiency
  • Captures all of the heat
  • Can be constructed in a day
  • Inexpensive to build

 Click here to continue to part 4

Copyright 2013 Dragon Heaters

If you enjoyed this article, Get email updates (It’s Free)


  1. Allen Lumley says:

    I find the part #3 4” dragon but not a part #1 or part #2, In short I find it nearly impossible to navigate your site,
    It makes it difficult to regard this site as a potential source to send people to who are looking for a Cast core model this is unfortunate as you have what looks like a good working 4” A. L.

  2. Allen Lumley says:

    - My Mac does not play well with others, BUT ! the only thing at the bottom of my page is the ‘get updates section’ and the ‘post comments section’ ! A.L.

  3. This is awesome. The model seems to be a rocket built using the design of the European masonry bell stoves. If so, then we can finally get away from those god awful barrels. Since the flues are clay brick, my mind wanders to the ability of making an oven or hot plate addition. What type of tile will you be using for the outside?

    • SMathieu says:

      Yes, this was our goal to eliminate the barrel, as well as reduce the footprint and offer some aesthetic choices other than cob.

      I think a hot plate or oven would work as the cap on the heat riser portion, but keep in mind it will reduce the heat into the bells and it’s a bit high. From the floor to the top of the heat riser is about 40″ If you left 6″ of space above the riser for exhaust path your oven or hot plate base would be around 46″.

      • What about using a clay bell similar to the pizza oven design parallel to the riser? The heat would flow down and then up to the side cooking bell. I’m asking because there are a couple of us playing around with outdoor rocket oven/grill designs.

        As far as a heating stove, I congratulate you again on this wonderful bell design. You didn’t answer but I was wondering if you would be using the soapstone tiles as wood stoves and Russian stoves have been adopting. I look forward to future developments.

        • Actually, an outdoor cook top and pizza oven, flue design are next up in the build cycle. I will posting the detailed design probably this week and hopefully a build will happen the next week. We have gotten a lot of request for this. We will be using a 4″ burner on the cook top model to address the height issue. Cooking surface can be 36-38″ high with a 4″ and since its primary purpose is cooking, it should be enough heat, we will see.

  4. Doug Himes says:

    I was looking over this really neat design and was wondering how to heat a two story home with something like this? I have duct work in place from current heating system and have not seen how this system’s heat could be channeled though the ducts. How many square feet do you think this system can heat?

    • Without seeing your layout its a bit hard to comment, but the warmer air will rise, often the heat is too much at higher levels so small voltage fans can be used to move warmer air to lower levels. But it depends entirely on your house layout.

      Of course it depends on your insulation, where you live etc. But generally I would say 1500-2000 sq ft. If you have more square footage, you can consider 2 located in different parts of the house or an 8″ version, which has a slightly larger footprint, and produces more heat.

  5. Interested in kit and more info. Good work