mass catering

[satamax]

Jono, there's, in fluid mechanics, something called the boundary layer.

Basicaly the molecules of air or other gases stick to the walls of the rocket J or L or whatever kind of aparatus. The first layer is completely stuck, then the second layer is slowed down by the friction onto thoses stuck molecules. Then the third layer goes a smidge faster etc. The full speed can't be reached before at least half of an inch. (well in roofing we consider anything in a space of less than 3cm, stalled air, in which no convection movement of any significance can occur)

Plus the change of direction at the elbow of the feed tube to the burn tunnel, and from the burn tunnel to heat riser; makes a hump in the gas flow, creating even more drag.

Here's why a four incher is hard to get right.

Frankly, i think you could go 6 inch J, and be sorted. Tho, insulate your feed tube correctly, otherwise, if you want a small fire for simmering, with just a few sticks, you might end up with blown off flames, and a crap control of the fire.

[jono]

Ah yes, i was wondering about how the corrigations on my 3" corrigated steel duct (shown above) were affecting the boundary layer. Good to know those numbers. Its easy to see how the combination of boundary layer and square corners could really mess up a small j tube.

My experience of my own 6" RMH is that it can burn a smaller fire pretty well so long as i restrict the feed tube with something, and gases would circulate within the feed tube but still ultimately get drawn into the burn tunnel. The problem with a standard 6" j is just that, used properly, it will be way to much heat, which might damage equipment, burn food, boil like crazy, overwhelm the heat extractor (pan) and waste loads of energy up the chimney and, if used on a summers day, turn my kitchen into an unbearable sauna full of stressed and hard working people (especially if i have three rocket stoves for up simultaneously, as in the above picture).

I wonder what happens if i put a 4" square feed tube on a 6" system?

[jono]

on a different note, but still within the theme of the thread; i have been ordered to do giant paella. making a giant steel pan on a tripod i can do, but how the hell can i heat this with even moderate efficiency?!

Like this?



by the way satamax, i really appreciate your advice and am sorry if this is anoying

[jono]

you know, I drew that because i thought it was a hilarious plan, but i actually quite like the idea. I have a massive pile of messy old bricks that i need to do something with anyway. a key part of the design is that there is a slab on pilars directly over the exit of the heat riser to stop the intense heat from burning a hole in the pan, bit that slab can have an inch gap above it so that it doesn't cause a cold spot either. That thing is going to take a lot of power to get up to temp too, so an 8" rocket is definitely on the cards. one thing is for sure, that is never going to be an extremely fuel efficient cooking method. hasn't even got a lid!

[DCish]

Jan 18, 2015 16:37:16 GMT -8 jono said:

).
I wonder what happens if i put a 4" square feed tube on a 6" system?

If all you ate looking to do is keep people from over-firing the 6", I can't imagine how that would be a problem. Heck, if 4" is the lower limit of reliable operation, how about a 5" with an optional feed restrictor (say a split brick slid into the feed) to keep operator error to a minimum?

[jono]

Good point

[jono]

This 4" feed tube on a 6" system cannot be a good idea. Like why not put a 10" tunnel and riser on a 6" free tube? Because it will compromise draw, turbulence and the every area of the performance of a stove. No. The point of custom building the j for the pot is that it can be designed to be exactly the right size to burn at optimal efficiency when producing exactly the right amount of heat for the size of the pot.

[jono]

Jan 15, 2015 10:21:15 GMT -8 peterberg said:

Don't hold your breath, a 4" J-tube would be probably all you need for this application. And those are notoriously difficult to get right.

I just found your thread "small scale development". It seems you came to the same conclusion as i did, that small systems can work fine but they need a longer heat riser. Am i wrong? Taking your assumption that the shape on the back wall of the burn tunnel was inconsequential, It seems like the design you settled on is almost a normal j, with more heat riser and the two standard Peter upgrades, trip wire and Peter plate (sort of). Presumably, that is just about the ideal design for my institutional stove j? Or, maybe not the ideal design, but as good as i can get without buying a testo and start a long development process.

Maybe, or maybe I'm better with as small a batch box as i can make work. I read somewhere that if loaded up with half the fuel the batch box will burn with half the power for the same 45 minites, so that could make the difference. What is your guess as to how small the batch box can go and still work well? 4" with an extra long heat riser? Would you guess that a 4" batch would be about the right size for this application?

[peterberg]

The development 4 incher is in daily use now. And yes, the riser is proportionally longer than you would expect.
The back sweep at the bottom of the riser isn't inconsequential though, in particular combined with that little feature called the kick tail. The common sharp corner at the back act as a drag factor and because the corner is wider as compared to system size ash will inevitably accumulate there. The back sweep eliminated both the problems, and the kick tail will force the flame to the center of the riser. With less heat stress at the back of the riser as the result and hotter exhaust gases at the top of the riser. Combined with trip wire and p-channel will make this small J-tube much more effective.

And yes, a 4" batch box is probably all you need for your application. I know of some builds that small and they work well. The risers are not overly long by the way, just 8 to 10 times the base figure. Batch boxes won't have the same limitations as J-tubes, they need a large top gap for instance, at the very least 150% of system size. When running a full batch the flames will come out of the riser, good for heating up a large pot quickly. Adhere to the spreadsheet for the proportions and build it really well, that's all I can say.

[jono]

Hello everyone,
Life got very demanding for about a year there, but I did build a quick 4" batch box, bought a 77 liter aluminium pan, and 20kgs of insulating refractory concrete, and learnt to use sketchup for a different project, now I have 3 weeks until an event at which I would like to be serving food from my apro-batch. so this is what I am puzzling about now. 1. if I have a 4" batch box with a 6" gap between the top of my heat riser and the bottom of my pan in an aprovecho style cooker, what shape should the structure be that connects the top of the heat riser to the bottom of pot skirt? Im thinking of having an inverted cone like so

2. this thing is about 1.4m tall. I think I'm going to need a stirring step, but that is probably the only way around this problem. I don't think it would be any shorter with a 4" j,

3. last question, probably for peter. Aprovecho give a system for calculating the gap between the pot and skirt which is based on the diameter of the pot and the volume of the combustion chamber of their L tube rocket. Is this idea going to work? what worries me is that as I do not own a flue gas analyser and cannot afford a testo, I'll never know if this kind of heat exchanger compromises the batch box's efficiency or not. is their a minimum cross sectional area that I should be going for. there is a lot of surface to create drag so I'm guessing I want more than the csa of the riser in that skirt gap. maybe you could give me one of your best guesses?

I'm hoping to have this thing as a bike trailer for a stall called "food that doesn't cost the earth" at a local climate awareness event on the 22nd of October. If I cant solve these design problems I'll just have to do an L combustion chamber, but either a batch or a pimped J would make a much more satisfying stove

[satamax]

Jono, on my rocket range, i have a 6cm top gap. For a 12cm heat riser, but a box and port more suited to a 10cm. Heat riser is only 62cm long iirc.

Tho, with a 11cm chimney at the back, about 2m high, it wasn't performing too well. Now with 11m of chimney behind, it turns fly ash into glass balls (just joking)


If you build your thing, use a chimney, like aprovecho,, but put another tube around and insulate in between. That would help the draft.

[peterberg]

This summer, during a memorable workshop in Warsaw, I've tried a 4" batch box in a 55 gallon barrel. The top was closed so no sunken pot in there, only a 120 mm chimney pipe came through the lid close to the rim. This pipe was reaching down to the bottom of the barrel just leaving a space of 80 mm underneath. The top gap was cross section area of a 100 mm riser devided by the circumference of that same riser , multiplied by 1.5. Boils down to 37.5 mm, rounded off to 40 mm. I catered for the expected dimpling in of the lid because of the heat but that isn't applicable to you. Running at full tilt, the temperature of the lid got up to 600 C and the top quarter of the barrel side got also very hot (sorry, no numbers). The lowest quarter of the side remained surprisingly cool. In this construction, the bare chimney pipe got hot but not overly so, it still could be touched briefly. Due to part of the exhaust pipe being inside the barrel this pipe remained constantly hotter than the lower parts of the barrel. Because of smoke coming out of the firebox opening we had to arrange a make-shift door of a simple down hanging plate with a piece of it bent at right angles to rest on the top of the firebox. Instead of a p-channel, it featured a floor channel. Just a straight piece of rectangle steel tubing resting on the floor of the firebox, ending right in front of the port.
This contraption proved to be running surprisingly well.

Based on this experience, may I suggest some alterations?
The riser seems overly long in your drawing. Just keep to the 8 times base figure and you are fine.
When building a system using a square riser, the sides of the square should be the same as the diameter where all the measurements are derived from. In other words, the riser of a 4" system should be 4" by 4".
You probably can get away with 1.5 barrel, the pot hanging in there like in the institutional stove just fitting tightly in the lid.
No skirt or anything, just the batch box core underneath it and the exhaust somewhere halfway the side or maybe a bit lower. Chimney by means of an elbow to vertical plus a stretch of 3 meter stove pipe.
Top gap not larger than 50 mm, leave the rest to physic's law.
This construction will eat almost certainly more fuel as compared to a real InStove, but on the other hand it is a very straightforward one. And very powerful at that.

[jono]

Thanks guys, this forum has such a supportive community.

Hmm, sounds like what I will end up with is a stove that at best might have a higher combustion efficiency, but will definitely have a much worse efficiency of heat transfer into the pot, so will ultimately be not nearly as good as the classic instove. Is that because the L tube combustion chamber is able to work with the restricted draw of the pot and skirt in a way that neither a refined J tube or a batch box will? If that is the case, it seems like I am not going to make a better instove by using either a 4" j or batch, because even if the combustion is a bit more efficient, it needs more wood and could still have higher emissions for the same amount of cooking, and it will be more of a hazard in a public environment (because the whole thing will be super hot) than the classic instove. sounds like my enthusiasm for the peter designs is only going to make the instove worse.

perhaps I could run a trial to refine the shape of the L. I could make a bunch of test shapes for the combustion chamber, and measure the temperature rise in 60L of water for a given weight of wood burnt. Ill ultimately still be making my combustion chamber from insulating castable refractory, though I might use sand cement and vermiculite for temporary test pieces as they only need to hold together long enough for a test burn. so this is the first shape I have to make (not to scale): the classic instove L design

Now im thinking would a trip wire help? could I bring the secondary air in differently? could I have a sloped fuel magazine to make it a gravity feed system if I have the primary and secondary air separated from the fuel feed tube? Can I redesign this so that it creates better mixing to maintain combustion efficiency with less air in the system and thereby further increase the efficiency of the heat exchanger?

I know that the way to make progress here is to try a bunch of shapes of combustion chamber with consistent fuel batches and take careful measurements of the temperature change in a given volume of water, but I'm just trying to get some initial idea of what shaped I should be experimenting with. Maybe I dont stand a chance here, and I would be best to just go with the tried and tested shape.

[jono]

I have built a 4" batch box to peters exact dimensions before and it worked great. However, I want to mess with the formula to see if I can make it more tolerant of a restrictive heat exchanger, because if I can half the amount of wood I need to burn to boil a litre of water, even if I double the emissions per gram of wood burnt, this is still the same amount of pollution for the amount of work done.

My first thought is that the port in a batch box probably has about twice the flow resistance of the riser, having a smaller cross sectional area and a lot more edge. My second is that I don't really want to load it up with a big batch of wood, I would rather have to refuel more often but have more control over the fire. So I built this disposable experiment, which has a narrow box and an extra high port. 10cm diameter riser, 10cm wide box, 3.5cm x 20cm port.



The secondary air enters the bottom, through that hole, and the tin has a plate that guides it into the very bottom of the riser. Joints have been sealed with clay.

I wrapped this in a piece of ceramic quilt, built a fire in it, filled it with bits of wood and put a temporary door to restrict primary air to the bottom third of the door, also sealed with clay. The rams horn pattern was extreamely robust, starting imediately. I held a large sheet of steel over the top of the riser and gradually lowered it to work out how small i could make the top gap before noticably stalling it. At 1cm the fire rises up out of the top of the riser to fill the gap with yellow flames. I assume this is because there is insufficient oxygen to complete combustion within the riser. a 2cm gap had little noticable effect on the burn, though I imagine a testo would notice the difference. I also did this experiment, though much quicker and less precisely, with a thick sheet of glass. the rams horn pattern persists even when the combustion is very restricted. Not sure what I learnt, other than that the top gap seems to be able to be quite small before it causes serious problems. I'll buy some barrels at the weekend and build the rest of the stove so I can do some water boiling experiments

[jono]

intuatively, it seems to me that part of the reason that the standard instove L tube is more resistant to stalling than a J or batch would be that the fire is right under the riser. Looking at the rams horn combustion pattern being drawn up the riser, it occured to me that not much was happening in the bottom corner of the riser opposite the port. which made me think that maybe you could squeeze a double venturi rams horn port into an L, with the combustion beneath it by having the port disect the riser at an angle like so.