Successful 3-1/2" portable low-mass rocket

[ringoism]

Some more (and less) eminent forum members had lent me very useful advice in late 2011 when I was first considering the feasibility of such a project, and gratefully, having finally started on it, here's what it's come down to after a week of fabrication and testing:

Success (I think) re: 3-1/2” portable, non-mass rocket:

“But Why?” you ask…

We’ll get to that further down. And still plenty of developments forthcoming, but for now:

1. First, my definition of “success” (totally unscientific and subjective, mind; And note total burn-time as yet of under 20hrs):

a) 6-hr plus continuous run capability (i.e., it didn’t stall, just wanted to go to bed - but then got to writing!). This on hardwood / slightly green lighter wood.
b) no smoke out the stack (once hot), even with green wood as 50% mix with seasoned hardwood
c) decent ram’s horn/vortex flame pattern (at varying heights depending on wood size / composition)
d) burn tunnel / heat riser red-orange hot (" " " " ")
e) barrel too hot to touch anywhere (well, possibly for a half second at the extreme bottom)
f) flue stack merely warm - can be held by hand indefinitely, even with riser at its hottest
g) enough heat output to keep me warm out on the porch at night (with unzipped thin fleece jacket and wet uncovered hair) at near-freezing temps (now if that’s not subjective, what is?!!)
h) no smoke-back at the inlet, even mounted outdoors with some wind present
i) design would seem easily duplicable at around $50-100, including local skilled labor cost (Himalayan India)
j) not measured, but could venture a wildly guessed consumption rate of around 1-1.5kg/hr of small-dia. hardwood. Whatever, it seems quite little compared to what we're used to.

2. Technical specs / dimensions / details:

a) Fuel magazine: insulated vertical "J"-type, 3-1/2” dia [now 5" - see below], 14”H (from tunnel floor) (capped).
b) Burn tunnel: 4.5”W x 3”H (ceramic tile-lined with superwool backing).
c) Heat riser: 36” (from tunnel floor) tall; 3-7/8” dia lower end, 3-1/2 at top; 404 stainless; wrapped with 2-1/2” of superwool (actually Isolloyd Supercera) 1260C-rated blanket secured by (surrounded with) steel insect window screening around the diameter.
d) 7” tunnel length between magazine / riser.
e) Primary air supply: side-draft, adjustable via sliding drawer/burn tunnel opening (so yeah, a capped J-magazine with L-drafting)
f) small secondary air flow from below grated area under magazine.
g) Chassis: MS with insulative / scale-resistant linings in burn areas.
h) Barrel: 11" diameter steel with generous flow clearances at top (3”) and sides (approx. 1-1/2”).
i) Flue: approx. 9ft. high, thin uninsulated galvanized, 3-1/2” dia.
j) Ramp: ½”H, 2”L (possible) “trip-wire” made of brick on tunnel floor between magazine and tunnel entry. I credit and thank Chrisburge here: donkey32.proboards.com/thread/739/mini-rocket-mass-heater?page=1&scrollTo=9036
k) At present, no “P”-channel.

3. Run methodology:

a) Starting: pull out main tray (a.k.a. burn tunnel/primary air adjustment/ash collector), fill with very small kindling, sprinkle a little kerosene, push drawer in, add several foot-long pieces of ½” dia. wood to the vertical magazine, drop in a match (or use my butane mini-torch), cap the magazine, and viola, flame goes horizontal as heat is forced to the riser, burn starts immediately.
b) Re-load: when the “rocket sound” eases off a bit, it means that long pieces have burned down short – i.e., just higher than the burn tunnel roof; at this point, reduce primary air (slide drawer inward) to around 20% opening, uncap the magazine, drop in 4-5 more pieces of 12-14” long, approx. ¾” – 1-1/2” dia. pieces of wood (better when uniform, so they burn down at similar rates), re-cap the magazine, and finally, re-open the primary air supply (slide drawer outward) till optimal burn sound is achieved (around an inch usually). So long as long sticks have burned down to around 3-4” lengths and the primary air is mostly closed down before uncapping the magazine, very little if any smoke escapes into the atmosphere during re-loads. I did not observe any smoking out the stack during loading, either, once the tunnel/riser were hot and some coals were present at the bottom of the magazine – in which case ignition of new fuel is near-immediate. With the wood I’m using, it’ll take around 20 minutes to burn down, with the present length of vertical magazine (thinking of extending this).

4. Future improvements (?):

a) full stainless sheet chassis for longevity (4-6 years minimum lifespan desirable).
b) burn-chamber lined with thicker insulating firebrick (compare burn vs. Indian soapstone?).
c) cast riser, if the stainless (as most suggest) should fail to hold up.
d) some cooking-oriented features, that being a major parallel use for woodstoves here; current 11” barrel top good only for one pot, and really need space for at least another small one.
e) higher magazine (feed tube) to increase time between reloads.
f) broader market appeal would be proportional to any possible height reductions (which I realize is tough with a rocket), since local custom is primarily to sit / eat / cook while seated on the floor.

5. The “Why?”:

a) Low cost / portability / easy duplication and universal fitment a must for any improved woodstove in the local context (large-scale “buy-in” impossible re: heavy, built-in, customized cob systems).
b) 3-1/2” the standard flue pipe here – existing holes in stone/brick walls / concrete roof slabs, etc. cannot accommodate larger.
c) considerable daytime mass storage possible via passive solar during the mostly sunny winters (and locals mostly sit / work / socialize outside anyway, till it’s cloudy / snowing).
d) local population (including ourselves) have no issues sleeping in near-zero temp (C) rooms (lots of warm quilts - with whole families under them!).
e) local habit is to split up fuelwood to very slim profiles, so no issues there re: loading such a small magazine.

So long story short, there didn’t seem much point for us (or other locals) to have a system that would be emitting heat continuously for 24 or even 12 hours, which of course is what mass-systems are so good at. Homes are non-insulated here, so the benefits of mass are a bit questionable along those lines, too. The fuel one would spend heating up that mass is not going to be well-spent if the rooms that mass will be radiating heat into mainly unoccupied except for mornings and evenings, and moreover represent high rates of heat-loss between occupied periods.

Obviously the conditions / socio-economic context / personal preferences represented here are not going to be applicable in a lot of other places, so please take this for what it is: a first prototype / attempt towards creating something that could help in reducing ecological degradation in this or similar contexts: a) heavy deforestation (primarily for fuelwood), and b) diminishing air quality due to the local habit of burning unseasoned wood in very crude, smoke-generating box stoves. Any improved stove must moreover be offered at a cost that mostly-ecologically-unconcerned locals will feel is “worth it” (i.e., it has to have a short payback period in terms of fuel savings); The only “improved” stove currently available is too large and too bulky for typical domestic use, besides costing upwards of $500 – far too pricey for creating mass-appeal here.

Anyway, your thoughts / suggestions / critiques most welcome. I’ve spent sufficient time researching on this and other forums / blogs / etc as to have become slightly mind-boggled on the whole subject, so I’ll be grateful for any kind of comments more specifically directed towards this project.

Mind, I don’t own a Testo or even a temperature probe, so I’m not making wild claims here; But hope that at the very least this will prove encouraging for anyone else who’s thought of building a sub-4”, low-mass system, but weren’t sure if one could even run.

Thanks to all the helpful and creative people out there,
Eric

PS: I’m a mechanical engineering grad who hated (most) math but nonetheless worked a bit under a decade in the auto industry (finally in test equipment design); I’ve experienced just a little change of life since moving to a small village in the Indian Himalayas, getting married, having a kid (almost a second now), and “retiring” to a life of making fruit jams and fixing villagers’ kitchen blenders (etc). So less of an engineer now, and bit more of a tinkerer, pack-rat and cheapskate.

And please forgive the lack of photo/video – internet connection pretty dreadful, and anyway the unit is looking like a bit of a patch-job at the moment, pending some less makeshift, more permanent fabrication work (bit embarrassing really).

[peterberg]

Eric, welcome back!
You are using superwool or equivalent, stainless steel and ceramic tiles for the construction. If I remember correctly, materials were a problem but apparently this is sorted out one way or the other. Congrats!

[josjor]

Pics or it didn't happen. : )

Really. We want to see pics.

[ringoism]

Dec 12, 2013 18:22:11 GMT -8 josjor said:

Pics or it didn't happen. : )

Really. We want to see pics.

Ahhh, yes... me, too. Patience...patience... thought of a still photo of the riser glowing orange, but don't have a camera bad enough that I wanted to subject it to that heat (thought of using my wife's ...but thought again...). Drum is at the metalsmith's today getting fitted with a properly welded ash cleanout / flue pipe mount (I had improvised that part yesterday, since he was too busy to get to it). And I need to get some legs on the thing (sitting on bricks at the moment) and get a more appropriate stone cut for the barrel top (green marble cracked). And my old glue barrels (two on end) still have their ugly original insignias... ETC... I DO have standards, you know . It'll do the word "presentable" some justice soon, I hope; at which point I can post something in "finished stoves". This was just kind of a "heads up".

Thanks,
Eric

[ringoism]

Dec 12, 2013 13:39:06 GMT -8 peterberg said:

Eric, welcome back!
You are using superwool or equivalent, stainless steel and ceramic tiles for the construction. If I remember correctly, materials were a problem but apparently this is sorted out one way or the other. Congrats!

Peter, I owe you some gratitude, as I worked off the basic dimensions you shared for a 4" system pretty closely; you had also suggested a stainless core, and so far I don't see any scaling there, as I do (lightly) on even the unexposed MS bits.

Yes, materials can be challenging, but like most things, they're easy to get (if a 15-hr. mountain bus ride there and back is "easy") if you know where to find them in the "big city". The search is difficult, as these small-time Delhi traders can be rather curiously specialized / narrow in their product offerings; The vendor I bought the Supercera wool and stainless sheet from (it seems they specialize in products that come in rolls!) had never even heard of "refractory products" (brick, castables, etc). Then I found out a few days back (via internet) that I'd walked right past a seller of firebrick last time without even knowing it - they're in the electrical products market (a vendor of electrical insulations, too!). But still no clue as to where to look for perlite. Or 309/310/321/etc stainless. Let's see.

Fortunately, between the tourist installations and the tunnel / hydro-electric projects up here, there are some good pickings at the scrap dealers' if I keep checking. Found a nice heavier-gauge stainless (mystery grade, of course) utility box today that I'll re-form as my chassis for stove #2. And talked with a friend who was contracted to build a brick pizza oven last year, who said there's some leftovers (brick, not pizza) at the job site; so should have those by tomorrow and will try lining my burn tunnel with half-widths of those instead of the crack-prone tiles. Also discovered (online) the Hindi word for soapstone (most stone here is referred to by its place of origin, rather than technical names), so will see if the granite/marble dealer up here (who had no idea what I was referring to) can now order me a sample.

Very grateful to you, Peter, for your efforts in this field, and for the personal inputs / encouragements that played a significant role in getting this thing up and going.

-Eric

[ringoism]

Stove is installed indoors as of today. Heats our uninsulated, rather leaky brick-walled & RCC slab-roof 2 rooms (275sqft total) very decently with outdoor temp around zero (C), emits no smoke out the flue (five min. from start) or feed (ever), and has consumed perhaps 7kg of dry 1-1/2"dia. hardwood in a full run of 6hrs. Starting immediate, loading easy, auto-feeds without a glitch (i.e., no fiddling). moderate, fairly even "rocketing" on this size wood (though tunnel/riser get hotter with smaller/lighter wood). Presently burning down/out as we prepare for bed. Once we get curtains over our third set of windows, some weatherstripping around the front door & windows and get the third (attached room) closed off with a door, I'm expecting very effective heating from this small, seemingly quite efficient unit.

Compared with the outdoor test runs, the flue is running a lot warmer, I guess because the temperature differential is a lot greater now and it's drawing more strongly. I could keep my hand on it indefinitely earlier, but now not more than a few seconds. Still can hold my hand over the exhaust at the top comfortably, so not too bad, but know it can be better (bit cooler). I'm going to decrease my riser/drum gap to the minimum by rolling some more superwool blanket around the riser, and see how much effect that has towards better heat transfer. If the flue's still hot after that, I'll consider adding a bell - probably just a parallel steel drum or whatever else I can find - or otherwise something to channel the outflow along our massive divider half-wall (make a bell out of it), and see how that works.

Did shoot some video today, but will have to set up a Utube acct. & upload when I get time; Being that our second son was born (ten days early) on Monday (!), time is not something I've got a lot of to spare at the moment... but when I DO get it up and you can see the video comparing our stack to those of all our neighbors, I think you'll understand just how positive a step this is in the right direction, in at least one respect.

I also did discover (online) that 310-grade stainless is indeed available in various cities in India - both manufactured and scrap (though, with India being a major world scrap-processor, the minimum order quantity in some cases was shown as 100 tonnes!). Ceramic-fiber refractory tubes, etc, also made here someplace or the other, and on top of that, another ex-pat neighbor has sourced vermiculite (he builds brick pizza ovens), and shown me the location of some local clay that holds up well to heat - all this to say that stove #2 should turn out a bit different than this first trial one, though the sizing seems about right and will probably remain the same.

Regards,
Eric

[josjor]

Congrats on the baby!!! If you hurry up with the pictures of the stove, we'll even let you post a pic or two of the baby. : ) You're guaranteed to get a "Good Job!!" from everyone.......on both projects. : )

[ringoism]

2013-12-22:
Here's some attached photos of the (rather ugly but bound to improve) stove. 

Very happy with the stove after our first snowy day of the year. Ran it continuously for about nine hours – 8:30AM till 5:30PM, with ash / coals buildup in the tunnel still not enough to prevent the stove from running fine. I emptied it with only ash and a few very small hot coals still inside, and re-started it at around 7PM, having removed the insulating sleeve in the feed to open it up to the full five inch diameter, that it might accommodate some of the larger, chunkier, knotty pieces of wood that the sleeved, insulated 3-1/2” feed wouldn’t allow; also takes more wood at a time, to keep burn temperatures up. The effect was very noticeable - the drafting was more robust / consistent and the barrel was throwing a lot more heat.

No proper sun the past two days, so going into evening as much as 50hrs had passed with virtually no natural heat input; thus a lot of cold, hard (uninsulated brick/concrete) mass, the worst being the snow-covered slab roof.  But with all this and being sub-zero (C) and still snowing late evening, we were warm inside with just sweaters - occasionally too warm when sitting less than several feet from the stove. The divider curtain between the rooms is in place, so heating is concentrated in these mere 120 or so square feet.  

I burned hardwood as large as 2” diameter, some of it very dense, and so long as it was dry and joined with a couple other pieces, it kept burning at a good rate. With the feed being of extended height now, I can burn up to 16” lengths - I guess this is as long as a batch box generally accommodates. But of course, the wood can last quite a while in this J-system between loads, since it’s still only the lower tips (3” worth) that are burning - the last two pieces I put in this evening must’ve burned for nearly an hour. The auto-feed, incidentally, really works beautifully. Just about the time the sound dies down a bit and you start thinking you might want to check it, a burned-out tip will break off, fresh wood drops down, and the roar starts up again. Lovely. Practically never requires any fiddling.

The draw was so good after opening up the feed / increasing the wood load, that I really think it’d be able to push the outflow through another radiating bell or a bit of mass. At the simplest, I could re-position the stove and run a length of sloped horizontal pipe, in order to let it radiate into the room before going vertical – though since a bare pipe at that level would interfere with our floor-seating area, I do think a bell-bench or backrest on the half-wall would work out best and be just about the right height, too. Easiest thing I’ve thought of so far would be to simply lean some large (2.5x2.5ft) floor tiles up along the length of our divider wall, close the ends with brick, run my inlet/(bit lower) flue pipes to it on the same end, and seal everything up carefully with clay. That’d give a nice, fast-heating backrest for anyone sitting on the floor, and would also heat up the divider wall mass, too, maybe allowing us to shut down a little earlier in the evening and just let that radiate.

The mild steel chassis is (predictably) scaling some, but the stainless on the top of the burn tunnel and in the riser seem to be doing okay so far. Anyway, upgraded materials will be there in version 2. For now, I’m feeling blessed to have a heat source – just in time for the baby’s birth and the first snowfall - that uses less fuel and pollutes far less than the very low local standard provides.

Regards, Eric

Attachments:

[ringoism]

(Evidence that we do live in the Himalayas, and that it did snow last night!!!)

Attachments:

[ringoism]

Ahhh... caroling with some dozens in the market earlier, drove back home past snow-covered fields and mountains, then an absolutely delicious meal (wife's such a great cook) with a bit of home-made wine (my own specialty), and now sitting around in the warmth, glow, and peaceful rumble of the rocket stove with the new baby, listening to Handel's masterpiece. I'm sure I've never had a better Christmas Eve.

Best wishes and Joy to all,

Eric

[matthewwalker]

Eric, I've loved following along on this build. Thanks for the photos and updates, it looks like a great little heater. Your last post made me warm!

[2tranceform]

The snow covered mountains hills are sure beautiful. I plan to experience the Himalayas one day. I hope that Christmas day is just as special for you and your family!

[ringoism]

Dec 24, 2013 9:44:28 GMT -8 matthewwalker said:

Eric, I've loved following along on this build. Thanks for the photos and updates, it looks like a great little heater. Your last post made me warm!

Matthew, I sincerely count it an honor for someone as yourself to have even noticed this very humble attempt.

Really appreciate your very many inputs to the forum, besides. I suppose there are quite a few other professionals out there who'd rather keep their "trade secrets" than share with all with the world!

-Eric

[ringoism]

Re: J-feed, L-airflow rationale (with a final question for the experts):

No great understanding of the science after one single, crude build (nor without gas analyzer numbers), so just surmising here and piecing together bits of what I've read / thought through, in light of this experience. The experts may feel free to correct me wherever necessary:

Typical "J" Problems:

1. In a traditional "J" built by the book, small systems would be constrained by equally small feeds ("system size" rule). My system is 3-1/2", but a 3-1/2" feed would really be too small to be of much use, unless there's an awful lot of nice straight 1/2" diameter scraps available to burn (preferably long ones, since they'll burn fast, too). And unless you've got plenty of wood burning, the small riser doesn't really create enough draw to keep things lively, i.e., to create temps hot enough for the desired levels of efficiency. Member Chrisburge seems to have correctly identified this as one of the inherent problems / challenges of small systems.

2. "System size" is a sort of questionable thing when it comes to "J" feeds anyway, since depending on the shapes / sizes / load quantity of wood, widely varying amounts of your "system" CSA are going to be blocked with wood, (Actual CSA = system CSA minus total CSA of loaded wood) and airflow down through and around that wood is going to be inconsistent; Problematically, when you have more wood packed in there/burning is when you need correspondingly more oxygen to get your air/fuel ratios right - BUT unfortunately what you get is exactly the opposite - more restricted airflow / less oxygen since more wood is blocking more of your inlet. The P-channel as originally conceived seemed partly to prevent pieces in the feed from blocking the leading edge of the tunnel, thereby providing an (uniformly) unrestricted portion of the inlet for free airflow. But it doesn't seem to be getting oxygen any nearer to the fuel (wood) itself, and it's not adjustable to compensate for load quantities / wood types (fast-burning vs. slower burning), etc, so might not always be idealized at 6-7% or whatever the figure was (obviously 6-7% of a 1/2-full (as %CSA) feed is not going to represent the same actual percentage of airflow compared to a feed that's 3/4 full).

3. The downward draft airflow in a "J" is not necessarily going to be directed towards the tips of all the wood pieces equally; in fact, the "path of least resistance" is probably going to be towards the leading (drum-side) edge, since it's easier for air to travel through a few pieces of wood there than through ALL of the wood. Thus, pieces at the leading edge receive more airflow / oxygen, and pieces at the back (further from the drum) receive less. This might cause inconsistencies of temps in the lower feed area, and more coals to accumulate in these lower-airflow areas, and cause unequal burn rates of equal-length pieces, as well, complicating re-load timings.

4. A taller feed-tube helps wood auto-feed better (i.e., not get jammed up / stuck) since pieces won't be splayed out at different angles as they can be in a lower tube; It also prevents wood from being loaded on any angle that puts the burning tips end up in weird places - like further into the burn tunnel or too close to the sides, where temperatures would tend to be cooler and airflow less. BUT the taller the feed, the more draft of its own it creates, counteracting the draw created by the heat riser. I think this is why it's been recommended that only an equal-size feed tube of no more than 1/3rd the height of the riser be used. If the feed tube were of larger diameter than system size, it could create even more riser-counteracting draw at an even lower height.

How this design MIGHT be avoiding these problems:

1. My feed is a little taller than 1/3rd riser height; but as no airflow is required (or in fact allowed, since it's capped) to pass through it, it creates no reverse-draft effects AT ALL. And no corresponding tendency towards smokeback. I've observed that system runs much more "rockety" when pulling through the end of the burn tunnel than when pulling down through the feed, in support of this assertion.

2. My feed, as it flows no air, can drop the "system size" rule and be opened up to 5" diameter, allowing adequate amounts of wood to be loaded / burned with plenty of air gaps around them; combustion thus improves, keeping temps & efficiency up. Both the extended height and the taper to a smaller diameter at the top help keep pieces "centered" and parallel (not splayed), i.e., more uniformly accessible to oxygen inflow; the taper also keeps (downward) air velocities up (at the feed opening) during uncapped reloads, so that even if longer wood pieces are still present / burning a bit in the feed, the smoke won't escape.

3. Another benefit of the feed-tube cap would seem to be that since there's no cooler intake airflow passing by it, wood is very well pre-heated and closer to combustion temperature by the time tips break off and pieces auto-feed down. This would seem to contribute to an overall cleaner burn, since fresh wood ignites immediately and burns at closer to ideal temp from the very start. And yet, since there's no air/oxygen flow present in the capped feed, wood doesn't typically start burning upward into the feed, as could happen in some cases with well-insulated (hot) feed tubes and very dry / fissured / bark-clad wood (like I am using).

4. Since primary air enters through the end of my burn tunnel, it must pass through ALL of the burning wood tips quite equally - so the temperatures / combustion characteristics in the bottom of the feed area should be relatively uniform from front to back and side to side and moreover from top to bottom, since additional secondary air enters through the grate below the wood tips. So it seems everything burning is exposed to roughly similar amounts of direct airflow. Thus efficiency improves and coals buildup is probably also reduced on account (the reason I can burn 9hrs. continuously through a small tunnel without it getting blocked).

5. Peter had written somewhere that excess airflow / oxygen can be actually be counter-productive, and that when the system is really "roaring" it actually can be running less efficiently overall. With my burn tunnel being in the form of a sliding drawer, primary airflow is completely adjustable to compensate and idealize oxygen levels for changes in wood type / size / quantity; Changes in the quality / volume of the "roar" are readily apparent as the drawer is slid in either direction (an added benefit is that ash removal is extremely easy / mess-free - simply remove the drawer entirely and dump it in the dustbin - that's it!).


All this has got me wondering now about the technical function(s) of the P-channel (I have not yet utilized one) - so if someone can refer me to an exact explanation for its efficacy, I'd appreciate it (been through Peter's early thread at a couple points); I'm assuming that it's partly providing a guaranteed, unrestricted portion of fresh intake air (oxygen) in a system where varying wood loads can variously block the burn tunnel - thus it helps prevent over-rich burn conditions (I used to get telltale black smoke just after re-loading, when I ran primary air through the (3-1/2") feed only); It also provides that fresh air in a pre-heated condition; So is this simply promoting elevated temps in the tunnel, thereby burning off measurable pollutants caused by incomplete burning / inequal temps / inconsistent burn characteristics of individual pieces of wood further back in the feed area?

I realize I'm going out on a limb here and kind of comparing apples to oranges; but I'm thinking back, a little analogously, to the 70's/80's when pollution controls were being introduced on U.S.-market cars: Most of the American automakers struggled with cumbersome and performance-reducing add-on devices like EGR (hot exhaust gases reintroduced into the intake) and spark-retarding and lean-burn systems (promoting hotter combustion temps) and catalytic convertors (hotter temps in the exhaust stream): essentially systems designed to cope with the reality of rather inefficient combustion characteristics of the mostly antiquated engines themselves. A totally different approach was represented by Lotus (the Elite, around '82-4, maybe?), which managed to pass then-current EPA emissions standards without such systems, on account of having a much better-designed cylinder head that burned everything uniformly/completely enough in the first place (not to mention being a lot more efficient in terms of fuel-consumption).

Just thinking that in a rocket stove (like in maybe any other combustion system), if you can find ways to promote some consistency / uniformity in 1) the amount of fuel actually burning, and 2) the oxygen flow around / mixing with every part of that fuel; then temperatures will be more uniform in every portion of the combustion area, and thus efficiency (in terms of both heat output and the burning off of pollutants) will be enhanced.

As it stands with the generic rocket system, turbulence (via trip-wire, etc) in the burn tunnel (as in a swirl-type automotive combustion chamber) certainly helps with the mixing, and the extreme heat of the riser (like catalytic convertors / Mazda's "thermal reactor" exhaust manifold) burns off pollutants; and maybe it could be argued that it doesn't matter so much where the burn actually occurs, so long as it happens somewhere along the way; but strictly speaking, aren't these (tunnel/riser) kind of the second/third stages of combustion by these points? That is to say, aren't significant improvements also likely to be found more towards the beginning of the combustion process (i.e., in the bottom of the feed, where oxygen is actually first meeting fuel - (analogous in cars to fine atomization of gasoline / strong spark and the velocity / direction of intake charge, maybe)? Conceivably, if combustion was more uniform / complete in the bottom of the feed, maybe such high temps might not be required further on in the system in order to attain full efficiency / low emissions. And maybe we wouldn't have to subject ourselves to as many problems presented by scaling metals, cracking refractories and the necessity of more expensive specialized materials, etc.

I believe it was (again) Chrisburge who suggested that everything that needed to burn in a woodfire for maximum efficiency could be burned within considerably lower temperatures than what rockets are often producing. Are we making things too unnecessarily difficult for ourselves from a materials perspective, by in a way relying excessively on secondary or tertiary stages of combustion in our rocket stoves?

Just brainstorming here...

-Eric

[Cramer]

Well thought out treatise and certainly food for thought.