J * tube to Batch Box Conversion

Over the Christmas holidays I began the conversion of my J tube Rocket Mass Heater into a Batch Box. There were many reasons for the change but the biggest two were:
The workshop where its located is only visited now and then which becomes a hassle to keep up with the constant relighting and filling of the stove. Secondly, I really wanted to get some experience with a batch box that I have been reading and dreaming about for years now.
It wasn't like I had all the materials ready to go, it was just a feeling of now is the time and lets begin! I'll make due with what I have and improvise the rest.
So the adventure begins......
First off, here are some pictures of my J tube just before and during tear down:

Sweet Gerry!  I'm proud of you!  Took the plunge before I did !
This is great, I get to watch your build and learn what to do ... and maybe, what not to do :)
What I really want to see, is its completion and how well it works in your shop!!!
I think I would convert my shop stove first. Saves lots of grief... then after that is successful move on to Liz's studio... and then... after she likes that one, I could maybe have one in the house!
So carry on my friend , your floating a boat, for more than yourself!

The batch box construction begins. I mostly followed Peter van den bergs dimensions but made some modifications knowingly to see what would work to keep cutting to a minimum and to make it as simple as possible.
I had a bunch of split firebricks salvaged from my J tube core and decided to insulate with rock wool. Not the best insulation choice perhaps but on my J tube, the rock wool I used to surround the core for about 2 months held up quite well so I went with it. I did notice that it was more dry and crumbly (perhaps the binder that holds it together disintegrated) but its integrity was still good. I liked it because it was fast to insulate and if dismantled again, can be removed quite easily. One day when everything is running the way I like it, I may switch to ceramic fibre blanket or board. Same goes for the heat riser.

I used furnace cement to glue all the parts together instead of clay slip only because I was getting tired of my J tube bricks shifting on me (particularly around the feed tube). It may have been a good choice for short term strength but probably going to be a bear if I do any modifications in the future.

Over a few months I was experimenting with my own refractory bricks and had the best results with perlite, clay and furnace cement. I decided to use these as the base for the heat riser. After installation, I put a thin coat of clay slip and furnace cement over the entire inside surface to see if it would help with longevity of the bricks. Time will tell.
The angle didn't turn out quite right so it ended up being a little pear shaped. Whoops.

Thanks Thomas! Already learning so much just by doing.
I know its not going to be a role model 'by the book' perfect creation, but it does get me to try all those things I've been wondering about with no remorse.

Gerry is that the same riser you were using before?
And is that an 8" system?
How long is your bench?

thomas rubino wrote:Gerry is that the same riser you were using before?
And is that an 8" system?
How long is your bench?

My old riser is still (temporarily) being used until I can make another one. It just sits on the new base I made. The top gap turned out to be 2".
I know Peter says at minimum to go at least system csa (so that would mean 6" for me) but before shortening it, I thought I'd give it a try. I ran it for about a week and had mixed results, particularly during startups - slow pulsing, fast pulsing, flames out of the primary air hole, needing to open the door to get enough draft etc..... Once running though, it cranked out the heat and had no problems with draft.
A few days ago, I cut 7" off the old riser thereby making it a 9" gap. The total riser height (measured from the floor) is now 46". The chart for a 6" system says 43 3/16".
Initial results indicate no change in performance. Everything appears to function the same. I thought for sure it would have affected at least the draft but nope. I may not get the glowing red top as fast but I never use it for cooking anyways.

I took a picture of the cut off section of the old perlite/clay heat riser after about 2 seasons of use. The inside perlite is blackish and heavily pitted about 1/2" in, particularly at the bottom and less so near the top. It looks very similar to all the bricks I've seen that have had spalling. The rest is in good shape. I'm sure it could survive longer but its not the type of riser that was meant for long term use or handling anyways. Each time I moved it bits of it would fall off in crumbles.

My bench is approximately 5' long, 18" wide and 18" tall with an insulated plunger tube taking up some of the space. Not necessarily suited to the 57sq ft allotted to a 6" BB system but I'm working within a limited space.

Here is a picture of my first small fire to dry things out and heat set the furnace cement.
Primary air is just guessed at and at this point there is no secondary air.
Because the stove and mass had gotten cold, it was a little hesitant to start the draft. Pulsing was very slow (about once every 3 seconds) and smoke often puffed out the incomplete door seal. Pulling the glass away from the brick face helped a lot with controlling the pulsing. Perhaps my primary air size was too small. Much to experiment with. I've read in the forums that there is a downstream restriction if there is pulsing so I'll look into it.
BTW, the glass is arched on the right side so there is a small gap at the top and bottom so this is also adding air into the fire.
There also was smoke for about half the burn coming out the chimney, maybe because of no secondary air.
Once it did get going, it no longer pulsed and seemed to flow correctly.
Even with a small fire, it really put out more quick heat than my 6" J ever did. Top of barrel was about 800F.
Also, there is no 45 degree bricks on the bottom yet. Soon to be cut though.

Hi Gerry,
The batch box rocket system need secondary air, it won't run clean without. There's one exeption though: when it is run open secondary air will come over the fuel to the port anyway. But overall efficiency is quite a bit lower then.
Pulsing is indeed a signal of a restriction down the line. It might be that your J-tube never pulsed but due to the higher heat and velocity of a batch box there might be a spot that is retrictive now. You have a bench with a stove pipe in it that doubles back? In that case the 180º bend might be acting as a restriction, depending on the temperature of the gas stream at the spot.

The above is something that often plays up when the conversion is from 6" J to 6" batch. Never happened when conversion was from 8" J to 6" batch. Better to check every change of flow direction and solve any pinched spot first before hacking into your bench and widening the 180º bend. Provided there's lots of room everywhere, a batch box rocket shouldn't pulse. Even the 8" thingy at Paul Wheaton's never did despite the 5 times change of direction including one of 180º.

Peter van den Berg wrote:Hi Gerry,
The batch box rocket system need secondary air, it won't run clean without. There's one exeption though: when it is run open secondary air will come over the fuel to the port anyway. But overall efficiency is quite a bit lower then.
Pulsing is indeed a signal of a restriction down the line. It might be that your J-tube never pulsed but due to the higher heat and velocity of a batch box there might be a spot that is retrictive now. You have a bench with a stove pipe in it that doubles back? In that case the 180º bend might be acting as a restriction, depending on the temperature of the gas stream at the spot.

The above is something that often plays up when the conversion is from 6" J to 6" batch. Never happened when conversion was from 8" J to 6" batch. Better to check every change of flow direction and solve any pinched spot first before hacking into your bench and widening the 180º bend. Provided there's lots of room everywhere, a batch box rocket shouldn't pulse. Even the 8" thingy at Paul Wheaton's never did despite the 5 times change of direction including one of 180º.

Hi Peter,   Running the stove without secondary air was just a one time deal to heat cure the furnace cement. I have since made a floor channel similar to Matt Walkers description (without a slit in the pipe facing the port). I have seen your square version with angled deflector at the top and would like to try that too.  See pictures. Question: Do you ever see any flame coming from the top of the pipe where the hot oxygen it being delivered to the top of the port? I'm recalling a regular wood stove with secondary air tubes along the top that have small holes and sometimes you see then all shooting out flames similar to a propane cooking stove element.
My bench is actually a bell - Pictured below. My top gap is 9" and you can see the manifold transition is fairly large and open (but maybe not for a batch box?). The drop tube (insulated with rock wool) is however right close to the manifold and barrel which could be a restriction - bottom left of the first photo and last photo. I have no other place to put it due to all our shelving in the shop. For an experiment, I could remove the insulation to give it more room, I just know that it will increase the stack temperature dramatically which is why I insulated it long ago. Sorry, I don't have and csa measurements for any of this at the moment.

Gerry Parent wrote:Question: Do you ever see any flame coming from the top of the pipe where the hot oxygen it being delivered to the top of the port?

As you may know, my floor channel arrangement reaches up to about half of the port's height. It is done this way because from my own and some experiments from others the most effective part to add fresh heated air is in the top half of the port. During each and every run flames that seems to come out of the channel can be seen in my heater, continiously for about half of the burn.

Gerry Parent wrote:My top gap is 9" and you can see the manifold transition is fairly large and open (but maybe not for a batch box?).

A manifold that is open like that is close to ideal I'd say.
It might be that the plunger tube is too close to the exit into the bench.

As I see it, the horizontal feed part of Matt's channel is open at the riser end. It should be closed off and maybe it is, but in case it isn't it would be advisable to mount it in such a way that air can't stream out of that end.

Peter van den Berg wrote:As you may know, my floor channel arrangement reaches up to about half of the port's height. It is done this way because from my own and some experiments from others the most effective part to add fresh heated air is in the top half of the port. During each and every run flames that seems to come out of the channel can be seen in my heater, continiously for about half of the burn.

For clarity Peter, are you saying that it would be better to cut my vertical tube in half so its half the height of the port? As you know, this was based on Matt's design and on his website he states: "A section of 2" I.D. tubing is attached vertically to the end of the square tubing, and should be approximately 9" long, or as tall as the port." Perhaps this was an old design or description but I would be willing to try it if you think it would be better.
Only once (for only about 30 seconds) have I seen what appeared to be some flame igniting the air coming out the end of the pipe. Nice to have the confirmation that this is something to look for.  

Peter van den Berg wrote:It might be that the plunger tube is too close to the exit into the bench.

Not quite sure I follow. See an older picture for the overview of my system and how the exhaust pipe right up against all those shelves. Perhaps I could add an elbow to the end of the plunger tube and extend the piping on the floor of the bell to the opposite end further away from the manifold.

Peter van den Berg wrote:As I see it, the horizontal feed part of Matt's channel is open at the riser end. It should be closed off and maybe it is, but in case it isn't it would be advisable to mount it in such a way that air can't stream out of that end.

Yes, its open as a stand alone unit but once it gets fit into the floor channel, it gets blocked by the brick it butts up against thereby sealing the end. Do you think it should be properly sealed with a welded end or will this suffice?

@Gerry,
Matt's pre-port channel differs from my floor channel on a number of aspects. Both deliver air at the top half of the port.

Matt's is about the same csa throughout, and the round vertical part is indeed as high as the port. Air is coming out of the top and flows down into the port. As it is, it delivers the air at about the same point as my p-channel arrangement but due to the hot vertical part more air is likely to be delivered. As I understand from Matt, the top end of the vertical tube tend to corrode away so during the season it's slowly getting shorter. Matt tend to run his heater all day long as he lives in a mostly uninsulated cabin according to my latest information.

My floor channel (latest amended design on the site) is different in that the horizontal feed is twice as large as the vertical stub, csa-wise. The stub ends just over halfway up the height of the port and acts, due to it being smaller than the feed, as a venturi in itself. That means there are two quite separate effects: air speed in the stub is significantly higher than in the feed which means cooling of this steel part is quite good. The other effect is air velocity delivered into the port itself is quite high as well, helped by the 45º cap on top of the stub. Presumably, lengthening this stub can't be done without deminishing the venturi effect. Corroding of the stub in our red bell heater is minimal, it's in its 4th full season now without holes. We are living in a passive house by the way, fuel usage during one season is about half a cord of softwood species, mostly construction scraps and pallets.

The horizontal part of both channels expand during the burn so they tend to "creep" away from the port. With that in mind, it would be better to have a welded end  on the feed part so leaks won't occur.

Regarding the plunger tube: I was mistaken about that, it is in the manifold, not in the barrel itself.

Good Day Peter;
I was cruising your batch site and Matts site last evening.  
I see Matt has started using a piece of stainless exhaust tubing as the vertical portion of his port, bolted to the horizontal run. It is 2.5" in dia.! And it gets larger?
His site was updated in Oct of 2017 stating that he is now using this arrangement on all his stoves.After 3 years he is seeing no deterioration of the pipe.  
Unfortunately his link for that pipe was no longer valid. I have written him asking for more info.

While cruising your awesome site, I did not notice your improvement that you mentioned to Gerry.
About your latest floor port design getting smaller as it goes vertical.  I can see how it would increase air flow.
I will be going back to your site this evening and attempting to read all about it.

EDIT)  I went back this morning.
I Found your new design right away. There is so much information I somehow missed it.
I do have a problem.  I can't seem to down load sketchup file.
I did join them but I am hopeless about computers.  
Could you please tell me the dimensions on your 6" B.B. floor port ?  That would be much easier for me.  

That is a lot of excellent information Peter. Thank you!
I will weld an end cap on the end of the horizontal tube when I get a moment.
I made it easier to remove the floor channel (for replacement or modifications) since I made a detour in my decision of how to do the door. I put aside the Robax ceramic glass (makes for a really nice LARGE view of the fire!) and cobbed up an enclosure to mount a casserole lid door. See pictures.
I had to extend the length of the floor channel to exit the front of the stove by making a form out of waxed cardboard then cobbed over it to form the channel (1 1/4" x 1/1/4"). I also made a similar form for the primary air (2" x 2"). Air controls for both are waxed cardboard squares with a small bolt through the center as handles.
To keep the glass from getting overly hot, I backed the front of the stove away from the fire - maybe about 4" at the bottom and 2" or so at the top (since the glass sits on an angle).
No idea how this will all work out in the long run but I wanted to try the casserole door thing while I'm still in full experiment mode. In the meantime, I'm going to research the easiest way to mount my big and clear ceramic glass door.

Thomas, I emailed Matt about a week ago about the broken amazon link to the exhaust pipe and he responded with a thanks for the heads up....I guess he hasn't gotten around to sourcing out another. I did do a quick search for exhaust pipes similar to the one he was talking about and did find a few. Here's one: muffler pipe

Some of the photos

some more:

Excellent Gerry!  A second casserole door stove! Looking Good!
Thank You very much, for the link to the stove pipe. I would say it is exactly what Matt was working with.
Now I need to decide if I want to go that way or with Peters modified design....

I must tell you that you have inspired me to jump off the cliff this spring / summer and build my first batch! (Hopefully my work schedule lets me)
Might run up to Columbia falls and buy some of those same Insulated fire bricks that Paul snagged.
Still have plenty of Superwool as well.

I'm going to give the casserole door a good fair run for its money (the whole $2 I paid for it at Vallue Village). Since there are both high praises and equally scary stories about its use, I wanted to see for myself what will come of it.
The ceramic glass will always be there if all goes to pot!
I'm glad your inspired too! Superwool and insulated bricks would sure be the cadillac way to go. A little pricey for me to experiment with, but when I really want to sink into a long term robust system and do everything 'correctly', I'll be sure to invest into these more long term and efficient materials.
In the meantime, I'm ever so greatful for the patience and skills of Peter, you and anyone else that wants to join in the experiments with me. :)

thomas rubino wrote:Could you please tell me the dimensions on your 6" B.B. floor port ?  That would be much easier for me.

Ehrmm... I presume you mean floor channel. What I used was 35x35x2 mm (1.38" square and 0.08 thickness) for the vertical stub, 60x40x2 mm (2.36"x 1.57" same thickness as the stub) for the horizontal feed.  You could look for dimensions that are close to the decimals, you could double up the stub duct for the feed so that saves you a search for the right measurments.
I have somewhere pictures of how that can be done... Just cut it out and weld it together, stub on top.







Length of the stub would be just over half of the port height, measured from feed to top of highest side rim and the end should be welded close. I prefer this method because just a single length of one duct is required to fabricate the channel.
Hope this is clear.

Hi Peter;
Thank you very much for the quick reply.
The closest sizes readily available here are, 1.5 inch square X 1/8" thick for the stub.
And for the channel I can get 2.5" x 1.5" X 1/8" thick.
Are these acceptable size substitutes ?  

From your photo's)  I see what looks to be maybe a  4" - 5"  cut out on the thin sides apx a 1/4" - 1/2 "  from the end of the channel , with the diamond shape for the stub cut into both pieces.
Are those measurements suitable ?  Or are there more precise measurements?

On the stub)  It appears that it starts out as a square cut and then a 45 degree triangle is removed on the angle facing the port and another on the feed side.
Then a deflector is cut to direct the air?   The deflector looks to be a diagonal cut flat piece, that stops mid point of the stub?

Here is the link to Peters site which explains the floor channel Thomas is referring to (sketchup file is found here also):  Floor channel

thomas rubino wrote:The closest sizes readily available here are, 1.5 inch square X 1/8" thick for the stub.
And for the channel I can get 2.5" x 1.5" X 1/8" thick.
Are these acceptable size substitutes ?

The stub is quite close, so that would be yes. The one for the channel isn't large enough on its own and overdone when used double. Just use the 1.5" square double side by side and it's exactly twice the stub's dimensions, as specified. By the way, the pictures are showing a rectangle duct from another experiment, I provided those just to illustrate how it has been done.

thomas rubino wrote:From your photo's)  I see what looks to be maybe a  4" - 5"  cut out on the thin sides apx a 1/4" - 1/2 "  from the end of the channel , with the diamond shape for the stub cut into both pieces.
Are those measurements suitable ?  Or are there more precise measurements?

The cutout in my example is 4" long, this is perfectly adequate. Starting 1/2" from the end to keep things stable. Keep in mind the cutout for the stub is just as wide as the inside of the stub. I'd weld it just one spot on all four sides and it remained perfectly stable. Stub distance to the end of the feed duct is 10 mm or 0.394". I'd say 0.4" is close enough.

thomas rubino wrote:On the stub)  It appears that it starts out as a square cut and then a 45 degree triangle is removed on the angle facing the port and another on the feed side.
Then a deflector is cut to direct the air?   The deflector looks to be a diagonal cut flat piece, that stops mid point of the stub?

The deflector or stub cap is there to direct the air, yes. I welded just a flat piece on the 45 degree upstream side and cut it to shape afterwards. It doesn't stop mid point but go further over to the point where the stub ends. The first one like that took me half a day figuring things out and all, the result looked awful. The second was done in just over an hour and looked much better.

So here it is all together and fired up. I'm keeping it unfinished until I get more experimenting with it done.
I found that it would start to pulse after the fire was going pretty good. As soon as I cut back the primary air about half, the pulse stopped. See video.



The space where the gasses come out of the manifold and into the bench between the plunger tube and wall of the bench/bell is about 6 1/2" (165mm) X 16 1/2" tall (419mm).
This works out to a csa of about 107 sq inches. The plunger tube is off the floor of the bench/bell by about 4 1/2" (114mm). Any of these areas too tight that may be causing the pulsing?

Gerry Parent wrote:The space where the gasses come out of the manifold and into the bench between the plunger tube and wall of the bench/bell is about 6 1/2" (165mm) X 16 1/2" tall (419mm).
This works out to a csa of about 107 sq inches. The plunger tube is off the floor of the bench/bell by about 4 1/2" (114mm). Any of these areas too tight that may be causing the pulsing?

According to my calculation the plunger tube could do with a minimum of 75 mm (3") so 114 mm is plenty of space, provided it's free all around for the gasses to stream up.
I am unsure about the opening between manifold and bench, I can't picture it. But a narrow and high opening isn't the same aerodynamically speaking, as one that's wider than high. It all depends where that opening is situated and from what angle the gas stream is coming.

Most of the time, a slow pulse can be deminished by opening the primary air further. What's the reaction when the door is opened while pulsing is going on?
And the opening between barrel and manifold, how's that sized and arranged?

Peter van den Berg wrote:According to my calculation the plunger tube could do with a minimum of 75 mm (3") so 114 mm is plenty of space, provided it's free all around for the gasses to stream up.

The plunger tube is unfortunately not free all around. It is situated right up against the bench/bell wall.
Picture #6 of my earlier post shows the plunger tube positioning against the bench wall : insulated-tube-in-place-ready-for-cob.JPG
As mentioned, it was an old photo so if in your mind you remove the barrel and manifold setup on the right of this photo and replace it with the first photo in this post: core-wrapped-with-rock-wool-and-some-loose-vermiculite.JPG that's what I'm working with now.
Tonight, I'll remove the barrel and take a 'fly through' video.

Most of the time, a slow pulse can be deminished by opening the primary air further. What's the reaction when the door is opened while pulsing is going on?
And the opening between barrel and manifold, how's that sized and arranged?

I'll light it up again tonight and see what happens when the door is opened up when its pulsing.

Here is what the pulsing sounds like before, during and after the door is open. Its not always exactly the same every time but for the most part the pulses start about 15 minutes into the burn (when things are really cookin'), then stop and smooths out when the door is open, then starts back up again (often right away) when the door is replaced.
You will notice that my primary air form made out of waxed cardboard decided at that moment to ignite into a flaming death. Gonna have to come up with another slightly more permanent air control.



This video takes us into the bowels of the dragons belly. Timeline (in seconds) is as follows:

13 - bottom of plunger tube
22 - looking up towards the manifold and space between plunger tube and wall of bench where the gasses travel through to get into the bench. You can also see the bottom of the heat riser.
38 - A different view of the gap in question and the flow the gasses would take as they enter the bench.
41 - A light being with a lightsaber.

@Gerry, I am in the dark what is causing that pulse. All dimensions seems to be OK, no pinched areas that I am aware of. When the restriction cannot be found in the heater itself it should be somewhere downsream, like in the chimney. In case the pipe is a single diameter over its entire length, it might be a chimney cap that's too restrictive.

Just a thought: during testing of the diverse incarnations of the floor channel I found out that a flat piece of steel on top of the channel's feed would stop pulsing. It is mounted at right angles on the upstream end of the feed, as high as the sloped sides of the firebox. The name I came up with for this part is the threshold since it restricts the air in the lowest portion of the firebox. Starting is slower this way and there's a tendency of leaving some charcoal behind.

That last one isn't a disadvantage per se, normally I close the air inlet before glowing phase is over. The next run will burn easier through under the lowest logs of the stack due to the fact that self-ignition temperature of charcoal is quite a bit lower than that of the wood itself.

Don't know if this helps, I am out of ideas for now.

Peter,  Here is a picture of the chimney cap. I had cleaned all the chimney pipe in the late summer so I know their good. The gap is 2 1/2" (63.5mm) with no screening .

Is this something like what your talking about the metal plate or "threshold"? how close to the "upstream end of the feed"?   see picture below

On his webpage, Matt Walker says he puts his primary air port about 1/3 of the way up from the floor. The bottom of mine is about 1.5" (38mm). Wondering if raising it up this much would have the same effect as your threshold?

In the thread bb pulsing Byron Campbell talks about solving his pulsing by creating a velocity stack. Any comments?

In regards to the pulsing. I take it as a sign that something isn't perfect with the system. Does it mean though that it causes a dirtier burn, or just more of an annoyance of having to listen to a locomotive making laps inside the house? From some other videos I've seen where pulsing is really prominent, in comparison it doesn't seem like mine is really bad but still would be nice to find the source. I'm sure it'll be found as I continue to tinker. I'll be sure to let you know.

Gerry Parent wrote:Peter,  Here is a picture of the chimney cap. I had cleaned all the chimney pipe in the late summer so I know their good. The gap is 2 1/2" (63.5mm) with no screening .

Might be a bit tight, although the difference is fairly small. I'm inclined to say 3" is minimum.

Gerry Parent wrote:Is this something like what your talking about the metal plate or "threshold"? how close to the "upstream end of the feed"?   see picture below

Yes, that is what I meant. At least 2 1/2" from the primary air opening. If at all possible use a straight piece, in front of the sloped floor bricks. This threshold tend to keep the coals and ashes in as well.

Gerry Parent wrote:On his webpage, Matt Walker says he puts his primary air port about 1/3 of the way up from the floor. The bottom of mine is about 1.5" (38mm). Wondering if raising it up this much would have the same effect as your threshold?

You could follow Matt's method, of course. Keep in mind there's a marked difference as compared to my solution. Mine runs with a single air inlet at the level of the floor channel, so quite low. The incoming air, being icy cold as compared to the inside of the firebox tend to form a "puddle" of cold air in front of the channel's feed so it will be served at all times. During the burn the under pressure in the port is getting stronger, pulling harder at the floor channel at the expense of primary air. This is exactly what one would like to happen, during the burn emphasis shifting to secondary right at the time it is most needed and back to primary during glowing phase. Threshold and floorchannel should be both at least 2 1/2" away from the air inlet for this to work properly.

Gerry Parent wrote:In the thread bb pulsing Byron Campbell talks about solving his pulsing by creating a velocity stack. Any comments?

At some point I tried this too, but the Testo wasn't happy with it, lots of CO. He's right about having the fuel away from the main air intake though.

Gerry Parent wrote:In regards to the pulsing. I take it as a sign that something isn't perfect with the system. Does it mean though that it causes a dirtier burn, or just more of an annoyance of having to listen to a locomotive making laps inside the house? From some other videos I've seen where pulsing is really prominent, in comparison it doesn't seem like mine is really bad but still would be nice to find the source. I'm sure it'll be found as I continue to tinker. I'll be sure to let you know.

Pulsing is an imperfection in the sense that more of the produced wood gas is combusted in the firebox instead of in the riser. In general, results weren't excellent during pulse activity.

Peter,
Interesting and informative facts Peter. Thank you.

So the first thing I tried tonight was the threshold. It was easier to make one out of a chunk of brick and fill in the gaps with cob than fabricate one from steel. Positioned the face of it 2.5" away from the primary air opening and slightly higher than the angled bricks. It definitely stopped the pulsing with the door closed. See video for details.

The one thing that I really noticed though is that the startup was very sluggish. The fire kept wanting to go out until I started to blow at the base where the primary air would normally be feeding air without the threshold in place. Even once it got going, it took about 20 minutes before the fire was full under way (which was when I took the video). It felt like air flow was being restricted too much with it there. Perhaps lowering it a bit, reducing its thickness or a steel plate would be better? I also noticed a bit more smoke in the firebox particularly near the roof area. Opening the door got rid of the smoke.  

Not sure about what you meant about the cold air being "served at all times".

You said "Threshold and floorchannel should be both at least 2 1/2" away from the air inlet for this to work properly." The floorchannel is buried under the primary air and both of their intakes are seen at the front of the stove so I'm not sure how distance for the floorchannel plays any part in how I have it configured.

So the primary and secondary air are self regulating due to the influence of the threshold? If this is true, why isn't it used on all batch boxes?

Gerry Parent wrote:It felt like air flow was being restricted too much with it there. Perhaps lowering it a bit, reducing its thickness or a steel plate would be better? I also noticed a bit more smoke in the firebox particularly near the roof area. Opening the door got rid of the smoke.

Reduding height and thickness would help to tune the heater, yes. As I said, starting up will be slower. What I didn't say is that I leave the door open a generous crack (I believe another expression is: keeping the door ajar) until the burn is well under way. Smoke in the firebox doesn't mean it also comes out of the chimney as long as the afterburner flame is going on.

Gerry Parent wrote:Not sure about what you meant about the cold air being "served at all times".
You said "Threshold and floorchannel should be both at least 2 1/2" away from the air inlet for this to work properly." The floorchannel is buried under the primary air and both of their intakes are seen at the front of the stove so I'm not sure how distance for the floorchannel plays any part in how I have it configured.
So the primary and secondary air are self regulating due to the influence of the threshold? If this is true, why isn't it used on all batch boxes?

A puddle of cold air is formed in front of the threshold so there's always plenty of fresh air available for the floor channel. Is this refrase better?

Your configuration is different from mine, as a consequence it won't work the same. In my solution primary air and secondary air are fed through the same air inlet, they aren't separated like yours. The inlet being low in front of and separated by a space of 2 1/2" from the floor channel means this will be fed plenty of fresh air at all times. The rising draw in the floor channel will automatically tip the balance to more secondary air at the height of the burn. As long as you have both inlets separated it won't work this way.

Primary air and secondary air are up to a certain extent self regulating in my air inlet configuration but not by the threshold on its own. And why this inlet system isn't used in all batch boxes, let me guess: this self regulating mechanism isn't readily visable, people want control, fiddle with the inlets themselves and mostly it's understand as too simple and elegant, that can't be true. And it seems difficult to comprehend how it works. It is based on the notion of colder air stays low and stronger pull draws in more air.

Eureka! Peter. It just clicked in what you've been trying to say to me. One air inlet it shall be then. I have some time today so I'll go and make those corrections and report back.

Peter van den Berg wrote:"That last one isn't a disadvantage per se, normally I close the air inlet before glowing phase is over. The next run will burn easier through under the lowest logs of the stack due to the fact that self-ignition temperature of charcoal is quite a bit lower than that of the wood itself."

So from my understanding, one of the functions of the floor channel is to heat the incoming air so that when it exits the top of the stub it doesn't cool the fire and keeps combustion more clean. With that said, each time I start the stove, I've been cleaning out the ash and any charcoal so that the top of the floor channel is exposed directly to the heat of the fire. With what you said though, leaving the charcoal behind is actually a good thing which makes total sense. I can see the ash being an insulator if too much is left behind so I guess its a moderation thing.

Gerry Parent wrote:So from my understanding, one of the functions of the floor channel is to heat the incoming air so that when it exits the top of the stub it doesn't cool the fire and keeps combustion more clean. With that said, each time I start the stove, I've been cleaning out the ash and any charcoal so that the top of the floor channel is exposed directly to the heat of the fire. With what you said though, leaving the charcoal behind is actually a good thing which makes total sense. I can see the ash being an insulator if too much is left behind so I guess its a moderation thing.

What I was describing is a system to serve a lazy person, that's what I am deep inside hidden by all that perseverance. No problem to do a lot of work to get it to a state of self-regulating so that I am able to lean back. I never had the ambition to build heaters myself as a way of living. I hate production without any creative flow, tinkering is much more satisfying.

That said, during development I noticed lots of charcoal were left behind, presumably caused by the cooling effect of the horizontal floor channel part under the fire. So I left ash in there since the stub was the main source for heating air anyway. Blocking the floor channel during full burn changed the colour of the stub in no time to bright cherry red.

Before loading the heater I rake a bit back and forth to get the charcoal on top and that's it. I use the up side down method, bigger logs first, smaller ones on top of that and kindling last, lit with a single barbecue lighter. Once the fire is in glowing phase I let it go until a generous layer of coals is left and close the air inlet. Next evening I rake the coals up and start the process over again. I don't use much fuel, about 12 lbs in one run, scooping ashes out is done once in two weeks or so and even then I don't empty it completely. Perfect for a lazy guy like me.

Oh yes, since I use construction scraps and pallets for fuel there are lots of nails in the ashes. More nails means less charcoal, probably those steel pins transfer heat down in the ashes to burn more coals.

"Lazy" is not how I would have ever described you Peter!  Its funny how we view ourselves vs how others see us   :)

Thank you for the tips on the operation of a batch box. I'm sure I'll be asking more questions about your techniques but would like to get some more of my own experience first and then compare notes.

So onto the modifications I made today:
As you can see in the video, the threshold is made from a piece of thin metal spanning the front of the stove and held in place with some cob. It is about 5" away from the combined primary/secondary air opening. The floor channel entrance is about 3" away from this same combined opening. Is this what you meant?

As you mentioned earlier, I also like the fact that it keeps the ash and charcoal contained and away from the air feeds.
Also, keeping the door open while lighting helps quite a bit in getting it going without it initially starving for air as you mentioned too.

Since they were already there and before cobbing one of them permanently shut, I left both entries available to see what the difference between the two would be on the pulsing (also shown in video). Closing one, opening the other and vise versa. It wasn't really a fair trial as the bottom opening is 2" x 2" and the top is 2.75" x 2.75" but it did have a definite influence:
Looking through the larger opening (on top), only half of it is blocked by a straight shot of incoming air and this caused it to pulse somewhat - Not as bad as before but was definitely occurring.
The bottom opening is totally blocked by the incoming air no pulsing occurred.  I can now also see your description of the air having a chance to "pool" and acting as a buffer to keep supply and demand in check. Very fascinating!

In regards to an earlier comment you made about the secondary air shooting flames out of it for half the burn, I still have only noticed this phenomena sporadically. So far the only way I can get this to happen (temporarily) is to dampen down the main air entrance a lot perhaps so there is less of the gasses being disturbed by the fast incoming air and can ignite showing themselves? Maybe an analogy something like a candle not being able to stay lit when there is a breeze. Is this a sign that my floor channel is not operating correctly?

Gerry Parent wrote:In regards to an earlier comment you made about the secondary air shooting flames out of it for half the burn, I still have only noticed this phenomena sporadically. So far the only way I can get this to happen (temporarily) is to dampen down the main air entrance a lot perhaps so there is less of the gasses being disturbed by the fast incoming air and can ignite showing themselves? Maybe an analogy something like a candle not being able to stay lit when there is a breeze. Is this a sign that my floor channel is not operating correctly?

You are getting closer, although the inlet opening is a bit too tight for a combined inlet. Mine is 150x30 mm (5.91"x1.18"), adequate for the majority of circumstances. That said, when the heater core is hot in the situation that I run the heater twice a day I bring that back to 2/3 to about half at the height of the burn, depending on wind and so on. Stronger draft means smaller opening to keep the air volume within certain boundaries.

This not seeing any flame shooting out of the floor channel isn't a sign of malfunction per se. Remember your floor channel is like Matt's pre-port tube, in my configuration the stub is smaller and square, reaching to about half of the port height and, important, the feed is double the size of the stub. Floor channel distance to air inlet is OK.

Peter & Gerry;    
Thank you!  I'm really enjoying watching you guys get Gerry's dragon roaring!
It's giving me a much better understanding of the principles involved and their reaction to changes.
Peter, I understand better now, why you are fascinated with batch box's and rather bored with the simple J tube design's.
Keep posting guys, first thing I check for in the morning is if there is a new post from one of you!

Peter van den Berg wrote:You are getting closer, although the inlet opening is a bit too tight for a combined inlet. Mine is 150x30 mm (5.91"x1.18"), adequate for the majority of circumstances

OK, so if I make the (smaller) lower opening the same size as the upper opening (which would then get plugged) which has approximately the same csa as yours, then I'll be good to go. Any benefit of its shape (yours being rectangular) or is square OK?
While we're on the subject, what kind of air restricting device do you recommend? I've been using a piece of rock wool stuffed into the air inlet which works but is rather crude! I was thinking more in the line of a typical stove damper or sliding plate.

Peter van den Berg wrote:This not seeing any flame shooting out of the floor channel isn't a sign of malfunction per se. Remember your floor channel is like Matt's pre-port tube, in my configuration the stub is smaller and square, reaching to about half of the port height and, important, the feed is double the size of the stub. Floor channel distance to air inlet is OK.

Noted. When I go into town next, I'll pick up some metal to fabricate your floor channel and compare results.

Thomas, I love your "flame pattern"! Who needs to fabricate anything when you've already got the perfect system right out of the dragons mouth. No wonder why your better half won't allow you to have one in the house! OH NO...not the drapes on fire again!

thomas rubino wrote:Peter, I understand better now, why you are fascinated with batch box's and rather bored with the simple J tube design's.

Actually, I became fascinated by the J-tube design during spring of 2008. Doing all sorts of experiments to find out what the underlying principles were and how they could be optimized. In 2009 I worked a couple of months on optimizing the J-tube itself, the results of that were among others published in the Rocket Mass Heaters book, third edition. There happened to be a strong conservative opposition not to amend the J-tube design so I refrained from doing further development.

When Lasse Holmes came along with his idea of what later became the batch box rocket I jumped on it because it was a new concept so I wasn't disrupting anything. I've always said the J-tube rocket is serving another niche as the batch box rocket. There's no need for deviding into two camps in my honest opinion but it feels like that, there have been rumours in rocket heater community and on permies in particular.

It would be very interesting what would come out of a merger between J-tube and batchrocket technologies, given what is known about those two at this point of time. It could lead to a whole new breed of rocket heaters but although it's tickling my imagination, it won't be me to pick up development.

Gerry Parent wrote:OK, so if I make the (smaller) lower opening the same size as the upper opening (which would then get plugged) which has approximately the same csa as yours, then I'll be good to go. Any benefit of its shape (yours being rectangular) or is square OK?

Square would be OK, rectanglular has the advantage of being low and spread out so the air streams up over the threshold more easily without an extra change of direction.

Gerry Parent wrote:While we're on the subject, what kind of air restricting device do you recommend? I've been using a piece of rock wool stuffed into the air inlet which works but is rather crude! I was thinking more in the line of a typical stove damper or sliding plate.

A sliding plate is probably best, although  builders in France find it better to have a valve that could only be open or closed and nothing in between. They have a point there, people tend to fiddle with it and try to slow combustion down which is not what we want.

Peter van den Berg wrote:Square would be OK, rectanglular has the advantage of being low and spread out so the air streams up over the threshold more easily without an extra change of direction.

I've noticed with the square shape that the air is quite turbulent in this 'pool' or depression area by observing the smoke and little bits of ash whipping around. I have no attachment to the square shape, so low and rectangular it will be then with a slider (as I am the only one running the stove and can't really tinker with it to see what happens)

In regards to your comments to Thomas, all I can say is that I would never want to have my open minded tinkering nature stifled by others opinions.
I would like to say a lot more but probably would end up being a 'cider press' conversation at that point so.... long story short, you have been nothing but an inspiration to myself and I'm sure many others out there and would dread the idea that you were holding yourself back because of some limiting ideas..... therefore, you have my vote on helping you to keep your imagination tickled for as long as possible! GO Peter GO!  


Picture sourced from: https://solarsystem.nasa.gov/resources/889/inspiration/