tiny batchbox rocket bell cookstove

[colin saengdara]

These are pictures of an experimental cookstove that I'm working on. I wanted to build a brick cookstove for our kitchen that would heat up the cooking surface quickly. I want clean combustion, and I wanted a batchbox design so that keeping the fire going would be easier than with a J tube. A Peterberg batchbox is based off of very specific ratios that all begin with the diameter of the heat riser, also known as the system size. In order to keep the cooking surface height low enough to be comfortable for my wife, I backed the system size down all the way to 3 and 1/8 inches. This is rediculously small by normal rocket stove standards, but I like experimenting so i figured I would see if the ratios worked or if this was below some universal constant.

I originally thought that I would cast the core and could change the heat riser diameter and height as needed, but the original build fired like a dream. Pictured is the cast core and heat riser. These are the only photos I've found so far, but will keep looking.

[matt walker]

Colin! That's awesome man! Great little experiment my friend, very cool. I'd love to hear about your materials, the riser looks like the formed ceramic fiber ones I'm using, yet you say you cast it yourself? Interesting. That is going to be a great little cook stove, I may just have to copy it for my kitchen. Love the small footprint.

My weekend experiments went well, I'll have more to share on cook stove height rockets in a day or two after I do a little more vetting.

[colin saengdara]

Hi Matt,
Thanks for your kind words. I saw you guys discussing the height issue and figured I'd better post this stove so you all could benefit from the experience. It's one of those cases where if a tree falls and nobody hears, it didn't fall. hehe

Anyway, this little thing puts out plenty of heat, with a high velocity flame. It's just a matter of how you want to utilize it. I did a homemade castable bell top/ flue channel. But I had difficulty molding the cook surface with my homemade mix. So i said to heck with it, and covered the top in dense fire brick. The top of the 2nd bell is also dense fire brick. The purpose was really just to dry out my bells, but my mind has been elsewhere, and it functions as a rocket mass heater in it's temporary form anyway, so it stays as is for now.

[colin saengdara]

Matt,
Regarding materials. The core is Kastolite 22 insulative castable refractory with stainless steel needles. It was cast in 4 parts so that different aspects could be replaced if needed. Finished density of .88g/cc.

The heat riser was a homemade mix. I DO NOT RECOMMEND THIS MIX AS IS. The performance of it is fantastic, but it needs an addition of sodium silicate to make it less fragile.

The recipe is as follows: BY VOLUME 4 parts fireclay, 1part castable refractory (dry mix), 16 parts perlite, 4 parts diatomaceous earth. I measured by volume and by weight. Then added half the total weight in clean water to end up with the right consistency mix. In the end, the volume was reduced to 62.5% of the seperately measured volumes because the perlite fines and the diatomaceous earth take up the small spaces in between all the different sized aggregate. Individually calculated volumes were 1250 cu in, final volume was 781 cu in. 37.5% reduction in volume. My calculations were that the final fired density was about .725g/cc or 72.5% of the density of water. This is a VERY insulative mix!
By weight: 10lb 11oz fireclay, 3lb dry castable refractory cement, 2lb 10.6oz diatomaceous earth, 4lb 2oz perlite, 10lb 3.8oz water (added 1 lb more to get proper consistency).

During the dryout phase, the piece steamed more or less evenly from all surfaces beginning at the top and working it's way down to the bottom. A total of about 3 hrs for a 3.125" inner diameter, 8" external diameter - so the wall thickness is just under 2.5". There was absolutely no additional cracking, but the piece is crumbly. I have moved it on and off a total of 3 times, during the build of the brick bells. But each time it felt as though it might suddenly burst into a billion pieces!
Therefore, the next time I use this mix I will add sodium silicate as a hardener or replace the dry mix castable refractory with a sodium silicate based furnace cement. I might also up the % of binder from 20% to 25%, or do two seperately to compare. I began with 20% to try to push the limit for a cheap, lightweight, homemade castable refractory. Performance wise, I can't even fathom anything topping this, I'm just going to have to give up a little bit of performance for more strength.

Edited to add weight and correct error in density information from my notes. - Colin

[matt walker]

Fantastic Colin, thanks so much for sharing that. Seems like a great mix. What a cool project, please keep us updated as you progress. I'm fascinated by the tiny batch burner! I've recently built a 6" J for my experiments, and one of the things that strikes me after playing with the batch design over the last year is how much hotter the output is of a batch for the same system size. In a 6", a J burns about 2-3 lbs of fuel at a time, where the batch goes through 3 times that in the same time frame. Heat output follows the same ratio, so it seems to me that a tiny batch like you have there is a very good size, actually. On par with a 6" J as far as output goes. Cool.

[colin saengdara]

Hi everybody. I still can't find my build photos so I took some new ones of the externals of the stove and the hookup to the chimney. As I said before, this one is a mock up to figure out answers to lots of questions I had about how small a system size was practical for a batchbox. It's ugly because it still has all the clay mortar all over it, but it's gong to be taken apart soon enough to reuse the bricks upstairs. This is the view from the front:



It's a two bell stove. The heat riser exits right below the clay "burner". I added the burner in an attempt to dissipate the highly localized hot spot above the heat riser exit and as an attempt to give more surface contact with the skillet. In the real stove, I plan to have the skillet in direct contact with the flame. Then I will be able to close the stove by means of 1 the skillet, 2 a cast refractory insert the same size and shape as the skillet , or 3 a heat riser extension and small barrel.
Pictured is the cob "burner"



So, one of the things I learned was that there was a tremendous amount of heat coming out of this little guy. I temporarily fire bricked the top so that I could test fire it with plenty of mass to absorb the heat output. This current setup absorbs all the heat readily and because the fire brick is exposed, it radiates pretty well too. But the final design will be a cast refractory top with the round skillet and a rectangular skillet as the primary absorptive loads. I plan to use a low density mix for the flue gas path and high density for the exposed top. Then my wife wants a rack, like oven rack, to cover the non-cooking surface of the 2nd bell so she can dry wood, and keep things warm there.

Pictured is the combustion chamber opening. It consists of my original casting lengthened 4.5" by dense fire brick to make it more convenient sized for fuel. Because of the draw from the chimney, there is always negative pressure even when the stove is cold. So lighting is as simple as loading it and burning 1 small piece of newspaper over the face of the fuel. Then putting the burning bricks in front of it to give it the correct amount of primary air. From there it takes off on it's own.



Because it's a test of the bell, the stove's secondary air is not hooked up. I regret this lazy decision, because it's not burning as cleanly as it had on it's own in the test bed. If I had hooked up the P channel, I wouldn't be stuck wondering how much of that is because of the still unburned fuel condensing on the cold fire brick, and how much is because of the lack of P channel.

Unfortunately, I think that the heat riser needs to be extended. Even with the P channel in the test setup flames were regularly reaching 2.5 to 3 ft into the chimney flue. Ideally I would like to have the flames crown just below the cooking surface. That is why I plan to have a gas path that is still lightweight refractory, to give the gases more time to burn. I'm hoping that a red hot skillet bottom might assist in the final combustion stage, but if not, the cast refractory insert will effectively extend the heat riser a great deal more. With that in place, the gases will not hit an absorptive surface until they reach the front cooking surface. This is about double the length of the heat riser suggested by the formula Peterberg developed. (.72 x 3.125 x 10= 22.5" But this is not the application it was developed for. That's why I need to have a good plan for providing high enough temperatures even in the cooking zones to achieve complete combustion. This is where I kick myself for not just hooking up my P-channel, it might make a huge difference here. The last thing I really need right now is an extra variable!

This last photo is the bricked top. the gas exits below the burner and proceeds under the top layer counterclockwise to waterfall down into the bell at the lower right brick.



Happy Holidays! - Colin

[michaelegan]

Looks great Colin. What's a P channel? Keep us updated and take more pictures! also, a much info on how the bells work is really good too. thank you