Largeish Cabin RMH for WI Event

I have the opportunity to help some friends install an RMH in a cabin they use in winter, and we are looking to do it as a workshop/session during the Wisconsin Permaculture Convergence at the beginning of September. We've had a discussion of the basic options and I could use some feedback on my assessment and suggestions.

The prospective area is large and open with a woodburning stove currently installed near a wall, with an 8" exhaust pipe running through the roof to a proper external height (yay!). The 'clients' have been on a tour of Wheaton labs and liked the RMHs there. They are fine with a J-tube and metal barrel in the middle of the space. I've helped with a couple of J-tube builds at WL events and made my own with a second in my near future.

I think the first decision needs to be cob bench vs. pebble style vs. some sort of brick construction. Cob is their preference at this point and we have some people with experience that will be able to help during the convergence.  The question is about weight. It is a suspended wood floor with an enclosed and insulated crawl space. Underneath, the entry to the crawlspace is on the opposite side of the cabin and it looks questionable whether we will have to clear out some earth to just get over under the spot. The floor structure is of 2x6s (eek) suspended across long beams spanning maybe 30'. The beams look to be embedded in the stone wall perimeter and resting on cinderblocks on the ground. No idea what lies below the surface yet, an inquiry has been trawled out. The beam seemed to be about 10" diameter from my recollection, maybe a bit less.

The owners describe the cabin as 'pretty redneck' and would be happy with a structure that has a good chance of improving things- they are not looking for top notch contracting or superfluous code obedience. We also need to be able to bring this all together for the event schedule, make significant progress on it during that time and have something working to show off to attendees, although finishing aesthetic touches and operational tweaks can come after the event.

My preliminary plan would be to jam as many cinderblocks as possible under the beam where the bench would fall and build a 15-20' ducted cob bench centered along it. The 'clients' feel this placement would be close to ideal for them.

Another option would be pebble style as my most direct experience was with the solarium RMH and have reviewed the Fisher Price house and Red Shed ones also. But this would seem to produce limited heat retention in the mass compared to cob, and this cabin is large enough that max heat retention would be utilized.

So more viable alternatives would be some sort of brick bell or hybrid brick/cob bench. Looking at the large 'box' brick bells I've seen in recent threads, it would not seem to fit the aesthetics of the room. I'm wondering if using brick to make an open channel that would flow gasses similar to the path of the ducting in a similarly dimensioned bench could produce similar heat retention with better weight distribution on the floor. Would a cob finish on the top of a brick bench structure work well? What would other pros/cons of such a structure be?

The idea of placing the bench next to the beam and knocking out the floor to build it directly off of the ground there is on the table, but the carpentry work of reinforcing the surrounding floor and making a mass foundation would require extra time and risk of pioneering techniques in my experience box.

The building is of cordwood construction, built back before Rob Roy and Richard Flatau brought the current levels of refinement to the technique. Full rounds were used along with a cement-heavy mortar, so cracks and air gaps are common and need annual attention. I'm undecided if building a structure against the wall would do more to help seal possible leaks or to interfere with needed maintenance, but as mentioned above, a bench more central to the room was preferred by the clients after discussion.

Apologies if this it too much at once, just wasn't sure where to start but would like to be sure we're on a good path before placing any orders. Anyone care to correct any assumptions I've made or sell us on an alternative to a ducted cob bench coming from a J tube? I did describe the benefits of a batch box as I'm aware of them and J tube was still the mutual preference for this build.

I suggest an 8" J-Tube with an open stratification bench.
Use brick for the walls and cement pavers for the seat.
Use a barrel to get your quick heat.
Use cob to cover all the brick work.

thomas rubino wrote:I suggest an 8" J-Tube with an open stratification bench.
Use brick for the walls and cement pavers for the seat.
Use a barrel to get your quick heat.
Use cob to cover all the brick work.

That is the rough sketch I worked out with my collaborators as a primary option. But I'm looking to quantify the difference between this and the cob style laid out by Ianto's and the Wisner's books before finalizing the plans.

The main point is finding a good weight to heat retention ratio. Are there any books/websites that have formulas and/or examples for weight loads and heat retention capacity around the brick stratification structures that would be comparable to what is in the rocket mass heater Builder's Guide regarding cob mass?

Also, with the general concept of a "stratification bench", assuming we want the exhaust flue 'kissing' the barrel to help start the system, does it help to build a channel into the bench to direct flow to the opposite end before returning to exit vs. one long open bench? My initial design for this was an 18" high bench/table, 30" wide, as long as needed to give a solid ISA figure*, with a wall of brick down the middle to both support the top and with a lone gap at the far end so gasses would need to travel the length and back before exiting...

*It occurs to me ISA figures are founded on single or double wall brick bell construction. Does this change if the brick is covered by cob? This is my main concern- the full cob bench has thoroughly researched examples in these books and my collaborator has built a complete one, I've used Allerton Abbey's and seen video of it's construction, etc. None of us have dealt directly with a brick bell and the resources (threads here or at the donkey32 forums generally lack the comprehensiveness of all that. Cob over brick does sound like the most promising design, but is it a good idea to take it on for the first time when doing a demonstration build?

The isa numbers do not change with wall thickness or composition.
The heat will take longer to work through a thicker mass.
I suggest starting with a thinner layer of cob, and the option to add more later is always available.
Info on weight distribution is easily located, but not specifically aimed at strat chambers.
Bells are certainly lighter than a solid mass.
Exiting the bell should be where it is most convenient to the roof jack.
If it is near the barrel, great. If not, use a bypass instead.
An open bell is better.
Rather than a solid wall to support the "roof" use columns only where needed.

Coydon Wallham wrote:
*It occurs to me ISA figures are founded on single or double wall brick bell construction.

It also occurs to me ISA formulas I know of are for Peter's batch box designs. I'm guessing one shouldn't just use the ISA numbers for an 8" batch box on an 8" J tube system?

Also, does an 18" bench equivocate to a 48" high box bell for ISA figures? Won't stratification be more effective with a taller vertical span?

For 8" J-Tubes, the isa numbers from a six-inch Batchbox are used.
A strat chamber works the same, tall or short.
Heat rises and then stratifies as hotter air replaces it.

thomas rubino wrote:For 8" J-Tubes, the isa numbers from a six-inch Batchbox are used.
A strat chamber works the same, tall or short.
Heat rises and then stratifies as hotter air replaces it.

How was the 8" J-tube = 6" BBR conversion arrived at?

Reading up on the double dead-end bench bell that was built at the 2015 MHA convention, it's mentioned that the benches worked great, but that a baffle over the exit to the flue was needed to ensure the benches encountered the warmer air before it exited, so I'm inclined to think the more 'squat' the bell is, the more precautions must be taken with the exhaust. Or perhaps that was only correlated with the dead end design of the benches, a flow-through bench would have no problems?

I looked through batchrocket.eu and saw that a 6" batchbox would utilize a 57sqft ISA. Even though Peter shows a "hybrid" system there with an oil barrel over the riser feeding into a brick bell, he doesn't offer any details as to calculating the ISA for it. Is 10.5 sqft a good number to use for a 55 gallon drum? Leaving 46.5 sqft for the 'secondary bell' bench?

The first one.
Matt Walker converted his 8" J-Tube to a 6" Batchbox, and it worked fine.
Peter was impressed.
No detailed testing was done, as the hope is that J-Tubes with a Piped mass will be updated to use a strat chamber.
Many conversions have been done since then, with no issues using 6" Batchbox ISA numbers.

With my preliminary calculations, this is looking to be about a 10-12 ft bench (12x~22" ID, encased in 1-3" of cob). Considering I was looking at a 20+ ft bench with ducting, a brick stratification bell would be a significant reduction in ability of the mass to store heat over the 'old style'.

This would also affect the ability to place the barrel where I was thinking, as the shorter bench might not span the distance to the roof jack.

So I should clarify if my thoughts on this are correct. The cabin has a split loft for the second floor. Half of the second floor, which has three bedrooms, is about 8' above the first floor on 2x4s over crossbeams. The roof jack is along the wall furthest from the second floor and goes to a 16' vaulted ceiling on that side.

My thought was that by placing the barrel under the second floor, more radiant and convective heat would work to warm the structure of the second floor rather than going to the vaulted ceiling. I'm assuming 3+ ft clearance over the barrel would not present a problem to the wood up there as that would be the same for most installs. Is this as important a design element as I'm thinking?

Unless I'm missing something about the ISA calculations, this would require a very thin/short bench and/or an angular exit flue. Is there a certain degree of angle that introduces a notable reduction in draw created by the exit pipe?

I'm probing the interwebs for all the arcane secrets behind this dark art of the hybrid RMH and one prominent set of questions is begging to be addressed in my mind:

A) Does the burn tunnel/barrel need to be in a separate area and ducted into the stratification "bell" bench?

B) Can one large bench simply be made to match ISA requirements and the firebox positioned within the footprint such that the bench is level with the top of the feed tube?

C) Can the bottom of the barrel simply empty into the stratification chamber without a manifold collecting and directing the falling gasses?

I'm picturing all of the BBR designs that have the box inserted into the bell, while wondering if the J-tube would struggle to get hot with the cooler (but still much warmer than ambient) air from the bottom of the bell around it...

My 8" J-tube has the whole riser inside the bell cavity, and works great. The combustion core is all insulated well.

My bell has a similar ISA to the 6" BBR spec, and extracts enough but not too much heat so that the exhaust stovepipe is hot to the touch but bearable for a while early in a burn, for a few seconds after an hour or more of burning.

I have found that my 8" J-tube draws strongly with a 6" metal chimney, even at startup.

Glenn Herbert wrote:My 8" J-tube has the whole riser inside the bell cavity, and works great. The combustion core is all insulated well.

My bell has a similar ISA to the 6" BBR spec, and extracts enough but not too much heat so that the exhaust stovepipe is hot to the touch but bearable for a while early in a burn, for a few seconds after an hour or more of burning.

I have found that my 8" J-tube draws strongly with a 6" metal chimney, even at startup.

Are there pictures or diagrams posted on it?

I think I miscalculated the ISA on my theoretical bench above, just on the math, but reviewing what has been written in various places am not even clear on the principle. I think @Peter van den Berg wrote something on his site that any interior walls/columns (except for the firebox) will count against ISA totals, but read elsewhere that they wouldn't...?

I think thick internal columns would count as ISA, but thin isolated brick struts would not because they would quickly get saturated with heat and stop absorbing more.

My RMH project is documented here.

Glenn Herbert wrote:I think thick internal columns would count as ISA, but thin isolated brick struts would not because they would quickly get saturated with heat and stop absorbing more.

My RMH project is documented here.

So it looks like the transition from your riser to the bell is simply the bottom of the cook surface with no other restrictions?

The cook surface is actually lower than the riser top, and not directly above it. So riser exhaust goes straight up to the top of the bell.

Not sure if there's enough context in these to convey the whole concept, but here's the building footprint and the working elements of the RMH with strat chamber:

Various things I'm not sure how to implement:
Floor insulation. For the stratification bell, I'd like to put down insulative cob on the wooden floor. I have horticultural perlite plus a number of wine and beer bottles (a nearby restaurant might supply some liquor bottles if they would be any better). The more bottles that can be used the better, since they would otherwise be headed for recycling and are much easier to deal with than perlite/cob mixture.

Any experience with incorporating bottles to insulate like this? Are they efficient? What kind of spacing would they require between the floor or each other?

When it comes to the brick walls of the strat chamber, could they go on the cob/perlite mixture, or would they need regular cob? How thick?

I built bottles into the cob surrounding the core of my rocket fired oven... can't say how well they work, though one near the riser base exploded in the first firing as the cob was wet and sealed the bottle neck.

I think they would work fine in the floor of a bell. I would bed them in some perlite-clay and space them enough to have perlite-clay struts between them, and a continuous layer above. It would probably be wisest to keep some space open at the base of the bottles so that air could flow from side to side under the bell to prevent heat buildup in a long firing.

I would want the brick walls to be supported on cob or on the floor with some gaps at the base; the base of the bell will probably not get too hot, but better safe than sorry.

A standard method of flooring a bell is to space some cement board up on bricks so air can flow freely underneath. This requires buying cement board, but is fast and easy, and takes little space from the height of the bell cavity.

Glenn Herbert wrote:I think they would work fine in the floor of a bell. I would bed them in some perlite-clay and space them enough to have perlite-clay struts between them, and a continuous layer above. It would probably be wisest to keep some space open at the base of the bottles so that air could flow from side to side under the bell to prevent heat buildup in a long firing.

I would want the brick walls to be supported on cob or on the floor with some gaps at the base; the base of the bell will probably not get too hot, but better safe than sorry.

A standard method of flooring a bell is to space some cement board up on bricks so air can flow freely underneath. This requires buying cement board, but is fast and easy, and takes little space from the height of the bell cavity.

Did your bottle experiment use bottles with the opening out, exposed to the room environement? I'm just planning to use them for the contained space, mouths blocked in.

Is there an expected temperature maximum that the floor would see in a properly ISA sized bell?

I was planning for the base of the brick wall to go brick>bottle>brick>bottle for the whole course of the exterior. The exterior wall would have 3" wine bottles, the interior floor 2" beer bottles. I'm not sure how tricky it would be to work in perlite-clay around those bottles while using regular clay mortar right around the bricks, or if that would make much difference. Would that be useful to prevent any thermal expansion from breaking the glass, or needed at all?

I also have a section of reclaimed pipe (1 1/4"OD) I planned to cut up and embed with the perlite-clay base under the firebox (with both open ends exposed to the room). Any potential problems there?

The bottles were totally embedded in cob. The mouths should be able to breathe a bit to avoid explosions. The floor of a bell is not likely to get hot enough fast enough to be a problem.

The floor of a proper bell would not get near as hot as the ceiling or even walls, but could get too hot for long-term wood contact in some situations. I can't say more than that.

I don't see any benefit to having the bottles exposed at the base of the brick wall; I would just have the gaps in the bricks at the base with the bottles inboard of the bricks. No interference or accommodation needed. If you had one of those bottle cutters, you could cut off the bottoms of all the bottles, put them end to end, and have free air circulation which would reduce heat to the floor much more than enclosed bottles.

I don't see including some 1" iron pipe in the floor as doing enough to be worthwhile.

Out with the old...

In with the new. (Still waiting on more clay to finish the base layer of structural cob in a few spots...)

The first firing went smooth, ran a couple of hours without incident. The following two days, it's been cold plugging severely. Even today with a few pieces of birch bark and some paper under the exit flue before starting, the firebox stalls out after ten minutes of a tepid burn. An hour or so of feeding the firebox a slight crack of air, as well as pointing a space heater directly on the base of the exit flue the whole time, is needed to get a draft going.

I'd say all of the wet cob (removing the cleanout covers after it plugs reveals moisture dripping from the roof of the bench inside) and ambient temperatures in the 60s F are clear causes for this and it is likely to be fine once the heating season starts, but am still not sure about placement of the bottom of the exit flue.

I created a ~2" depression in the stratification chamber bottom around where the flue would land, and set the bottom of the duct a little less than 1" above the rest of the floor of the chamber, so between 2-3" of space all around the opening to the 8" duct exit. Is using the inverse of the barrel top-riser clearance numbers a good guide for exit minimums?

That is a cool design, Coydon. I see how the barrel would help heat the upper floor. While the exhaust is over where the old woodstove was.

Yep, though one of the first additions will be a polished metal shield on the ceiling directly over the barrel intercepting the radiant heat there to reflect it towards the dining/kitchen area instead of the support beam.

Coydon Wallham wrote:The first firing went smooth, ran a couple of hours without incident. The following two days, it's been cold plugging severely. Even today with a few pieces of birch bark and some paper under the exit flue before starting, the firebox stalls out after ten minutes of a tepid burn. An hour or so of feeding the firebox a slight crack of air, as well as pointing a space heater directly on the base of the exit flue the whole time, is needed to get a draft going.

I'd say all of the wet cob (removing the cleanout covers after it plugs reveals moisture dripping from the roof of the bench inside) and ambient temperatures in the 60s F are clear causes for this and it is likely to be fine once the heating season starts, but am still not sure about placement of the bottom of the exit flue.

There's a lot of water in the cob and masonry, that's for sure. On top of that, even bone dry fuel will create water vapor, quite a lot of it actually. As a rule of thumb: half the weight of the fuel to begin with, is chemically converted to water vapor during the course of the burn. Adding all that to the water of the build, resulting in a very recalcitrant heater. During the first burn all seemed to go well, until more water was converted into vapor and condensed on any surface that was cold enough. Keep on going, that's all I can say, eventually the darn thing will dry out and come up to speed.

Coydon Wallham wrote:I created a ~2" depression in the stratification chamber bottom around where the flue would land, and set the bottom of the duct a little less than 1" above the rest of the floor of the chamber, so between 2-3" of space all around the opening to the 8" duct exit. Is using the inverse of the barrel top-riser clearance numbers a good guide for exit minimums?

This depression *could* work, depending on the size of it and where the exit is located.
Is this depression at least as wide as the bench' internals and shaped as a square, it could work. Provided the exhaust pipe is right in the center of the square.
Is the exit located at the very end? In that case, one quarter of the circumference is blocked by the end wall so no gases are able to enter from that side. When that's the case, the pipe should be higher above the floor in order to provide adequate space. I won't recommend this layout, I would regard it as much better when the floor is flat and the pipe ends about 4 inch from the floor.

Regarding your end question: yes it is, provided the calculation is done in a proper way. See the space above the riser as an imaginary ring. Think about it this way: the gases, rising out of the riser need to go through a 180 degrees bend in order to stream down.  When this ring's circumference, multiplied by its height is a figure that is twice as large as the riser's cross section area, then it's about the right size. Just by giving the gases ample space to round the bend without too much friction means a possible pinch point isn't there at all.
The same thing happens at the exit, the distance between the floor and the end of the pipe works the same way, although the change in direction is just about 90 degrees since most of the colder gases stream along the floor. So, in theory, the surface area of that imaginary ring should be at least 1.5 times as large as the exhaust pipe.

Calculating... the 8" pipe's csa is 50.26 sq. in. So, the imaginary exhaust ring's surface area should be at least 75.4 sq inch. Devided by the circumference gives the height of the ring. In your case, that would be 3" as the minimum distance from the floor, provided the exhaust pipe is free all around. As said before, the depression might be a negative factor.

Glenn Herbert wrote:I don't see including some 1" iron pipe in the floor as doing enough to be worthwhile.

I ditched the pipe idea and put a sheet of .5" cement board on top of a grid of cement bricks to form the base of the firebox/barrel. The face of the bricks has been cobbed over, with channels left between them.

The bottom of the stratification chamber has wine bottles between each brick on the first course, completely cobbed in by thermal cob in the floor of the bench. Not sure if there will be any insulative benefit (from the bottles), but it has decorative qualities and saved on some cob requirements.

Peter van den Berg wrote:
Regarding your end question: yes it is, provided the calculation is done in a proper way. See the space above the riser as an imaginary ring. Think about it this way: the gases, rising out of the riser need to go through a 180 degrees bend in order to stream down.  When this ring's circumference, multiplied by its height is a figure that is twice as large as the riser's cross section area, then it's about the right size. Just by giving the gases ample space to round the bend without too much friction means a possible pinch point isn't there at all.
The same thing happens at the exit, the distance between the floor and the end of the pipe works the same way, although the change in direction is just about 90 degrees since most of the colder gases stream along the floor. So, in theory, the surface area of that imaginary ring should be at least 1.5 times as large as the exhaust pipe.

Calculating... the 8" pipe's csa is 50.26 sq. in. So, the imaginary exhaust ring's surface area should be at least 75.4 sq inch. Devided by the circumference gives the height of the ring. In your case, that would be 3" as the minimum distance from the floor, provided the exhaust pipe is free all around. As said before, the depression might be a negative factor.

The exhaust pipe is constrained some on the sides around it. Would cutting a vertical slit in the pipe and bending it back to enlarge the opening on the side toward the firebox be enough to increase the draw ability, or will it be necessary to break the cob, cut and reposition the whole pipe?

Coydon Wallham wrote:The exhaust pipe is constrained some on the sides around it. Would cutting a vertical slit in the pipe and bending it back to enlarge the opening on the side toward the firebox be enough to increase the draw ability, or will it be necessary to break the cob, cut and reposition the whole pipe?

If you are able to reach into the bench, cut the pipe and bend the ends open, it might help. Breaking cob and redoing the exhaust thing I would keep as a plan B.