|
|
Post by pyrophile on Dec 22, 2017 3:03:55 GMT -8
About bells cooling :
It is important to remember that thoses "bells" including a very high heat riser don't act totally as bells because the gases leave the riser at almost the same height as the roof of the bell, at the top of it! Then any cold gases entering the firebox go up to the top of the bell (almost) and go down, cooling all the bell. It is different when there is a second bell where gases enter from the bottom and where cold gases stay in the bottom, not cooling all the bell...This a the "true" bell effect!
Bneoit
|
|
|
|
Post by independentenergy on Dec 22, 2017 3:37:37 GMT -8
About bells cooling : It is important to remember that thoses "bells" including a very high heat riser don't act totally as bells because the gases leave the riser at almost the same height as the roof of the bell, at the top of it! Then any cold gases entering the firebox go up to the top of the bell (almost) and go down, cooling all the bell. It is different when there is a second bell where gases enter from the bottom and where cold gases stay in the bottom, not cooling all the bell...This a the "true" bell effect! Bneoit thanks for this clarification, also for some time I'm wondering if a riser near the roof of the bell leads to a sort of chaos in the upper part of the bell "disturbing" the bell effect, the high-speed gases could disturb the stratification? What if you think? |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 22, 2017 3:54:33 GMT -8
During testing I tried 6 different lengths and two different widths for the vertical part. Also with slits in the sides, one slit in the far corner, open top or not, closer to the port, threee different tubes for the horizontal part, etcetera. The configuration I am using now turned out to be best, the threshold was added later on (5 iterations) to damp down the first violent part of the burn and feeding the floor channel adequately at all times. So your statement above do sound far too simple, I would say. OK Peter, thanks for the clarification. I'll first shorten the threshold to 60 mm to be in a similar configuration as your redbell, and then we'll see. Regards, |
|
|
|
Post by pyrophile on Dec 22, 2017 4:48:17 GMT -8
Independentenergy, imagine a jet of water going up, the speed and gravity make the water fall in an umbrella pattern. The lightest gases will stay in the upper part, the cold heavy gases will go down. But the problem is more when your fire is finished, if you have leaks in the door, ashdoor or anywhere : then fresh air enters the stove and cools all the empty volume by going downfrom the top of the riser to the bottom and escaping.
|
|
|
|
Post by esbjornaneer on Dec 22, 2017 7:58:03 GMT -8
It would mean something that needs to be poked at... but how about a threashold that is hinged and can be flicked down from upright position for the burn, to horizontal (or covering part of the 2ndary inlet) in the coaling phase? And reset for the next burn. Would this do anything for your embers?
|
|
|
|
Post by independentenergy on Dec 22, 2017 8:22:42 GMT -8
Independentenergy, imagine a jet of water going up, the speed and gravity make the water fall in an umbrella pattern. The lightest gases will stay in the upper part, the cold heavy gases will go down. But the problem is more when your fire is finished, if you have leaks in the door, ashdoor or anywhere : then fresh air enters the stove and cools all the empty volume by going downfrom the top of the riser to the bottom and escaping. thank you! really! ! I did some tests and even just by closing the secondary air intake the temperature of the brick surface goes down more slowly, to an extent that I did not expect. I have always wondered if leaving the secondary air open could cool the bell, and in fact it is! ! therefore it is very useful to close primary and secondary air when the stove is in the coal combustion phase! But I think that leaving it open at least a little is safer. my door closes very well, when I close it I notice that the coals go out |
|
|
|
Post by rakettimuurari on Dec 22, 2017 11:50:27 GMT -8
Very nice testing rakettimuurari ! Were the temperatures of the heater equal at the initial time, right before the two firings ? regards, You have excellent point there concerning the transparency of the test procedure  Should have measured & reported those too. Nevertheless, both tests were conducted exactly 24hrs after previous full batch burn in very similar circumstances. Most telling part for damper open case for me is the 1) asymmetrical cooling of the heater and 2) some certain spots closer the fire bath, which give almost identical numbers with previous test; first resulting from the cooling air flow and second indicating that initial temps must have been somewhat the same. I am prone to think that asymmetrical cooling most likely also indicates that bell effect here might not be fully in operation due the cramped nature of the bell i made ('cause oven and its supports). On the other hand the stratification should be starting already in the oven, but the height is only 25cm. The exhaust slit is some 2,5 x CSA to help in the slowing the gases but then again the space between oven external ceiling and bell ceiling is some 17 cm. However, asymmetrical placement of the firebox and closing the damper seem to result in very uniform distribution of the heat in the end. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 23, 2017 2:24:37 GMT -8
Hi everyone, Yesterday night and this morning were the first trials with a DSR-like configuration ! I put a half brick (3cm thick) to close one third of the top of the riser.  Visually the double vortex is still there but it happens more on the last two thirds of the riser. It's more "condensed". The plancha is clearly getting hotter right above the riser and there is a very clear hot spot at the height of the port, on the opposite side of the riser. The bricks were glowing red ! CO-wise, I've only performed an analysis this morning and the results seems not as good overall (with the best being 150 ppm) and the heater is LESS stable -- the tipping point is around 9.1% O2. The tipping point happened 3 times (9.0, 9.0, 9.2% O2) clearly and even a fourth time at 9.0% but it's less clear. Combustion analysis :  Regards, |
|
|
|
Post by matthewwalker on Dec 23, 2017 6:48:38 GMT -8
I suspect a centered port, such that there is no exit straight up the wall of the riser on any side, would perform better.
|
|
manU
New Member
Posts: 41 
|
Post by manU on Dec 24, 2017 6:16:58 GMT -8
Hello. Nice tests! Thanks again for sharing.
Your comment about a more condensed burning zone seems like a good sign to me.
In my previous comment when I said "a flat deflector plate right over the riser that creates a "second port like" opening to the side, maybe 5 to 7 cm high." I meant the deflector covering the whole riser which is raised 5 or 7 cm creating the port horizontally open to the opposite side (left, away form the chimeney) and sends the flames horizontally, away from the plancha`s hot spot. I can`t open your sketch up file to draw it because I have an older version. Wanted to clarify this, thinking maybe you would like a different idea to try and my english wasn't good enough to describe it.
Also some questions:
When you measure for example a 9% tipping point, what amount of that exess air do you think has to do with the leaking plancha? 1%, 2%, 3%? Maybe the real tipping point of the core is for example 6% and there is a fixed 2 or 3% added by the plancha? I don`t have much experience sealing this kind of setup but these are things I would try: aluminum foil, hi temp silicone, rock wool, ceramic fiber wool, mud.
About double vortex: maybe a wedge opposite to the port to split the flame in two would force the double vortex?
Last question: Isn´t DRAFT an important thing to measure and compare? In my experience wind outside is a major factor in my stove`s behaviour, I frequently have to use the damper to control this.
Your approach with this stove has caught my interest
Regards
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 27, 2017 23:13:43 GMT -8
Hi manUThanks for your comment, I'm not really sure with a horizontal 90° turn like this.. seems to me that it would slow the gaz velocity too much. For the vortex now I'm not sure neither that it's a good shape. For sure it works with a long riser but for a small riser like this it seems not turbulent enough in my opinion. Here is a V8 sketchup file of the cookstove : uzume-asso.org/assets/docs/experiences/cuisiniere_donkey/batch165_sidewinder_cuisiniere_maison_sketchupV8.skpI think the added excess air is around 1% at the top of the burn or something like that. Maybe a little bit more. That's the drift I've seen from the first measures to the last ones. I have flat sealing joints like this one that I will use to seal the plancha. Yes you're right I also think draft is important to measure. I mainly want to see the minimal initial draft to start a good fire in the stoves to see how small we can make the chimney flue for each size of batch. At the begining I thought for example a 230 mm batch would need a 230 mm chimney flue.. but now I know it's not necessary. About the effect on combustion it's too hard to measure.. I can only measure draft separately so it's not possible to measure it during the burn. matthewwalker I think that it's worth tring but I won't do this in my actual setup by fear of bending the plancha too much ! But I will have the occasion to try this in one month on another heater. esbjornaneer I'm afraid that a hinged threshold would not last very long here. I now have a 60 mm threshold and the amount of coals is lower now. When you start a new fire the old coals burn with the rest of the wood so that's not a problem. When I empty the firebox I have to take off a lot of coals mixed with ashes but then I put them back. That's a little bit annoying but not a huge problem neither. Regards, |
|
manU
New Member
Posts: 41
|
Post by manU on Dec 28, 2017 17:40:16 GMT -8
Thanks for the v8 file and the answers.
As for slowing down the gases, yes that was part of the idea. And using the heat that was too much for the plancha, to heat up a radiant surface (the deflector plate) that gives back a lot of radiant heat back to the burn zone. At last a second port or throat for more turbulence. Just trying to increase the "trhee T"s... But I understand what you said, it may be too much of a "brake" for your system.
I use a deflector in the stoves I‘be built which have a short, 5 or 6 inch diameter riser, deflector covering maybe 2/3 of riser, and have no problem with draft. And solved the problem of a glowing red spot in the metal oven‘s floor. But I do aim for a safe margin on stack temps, because I don't do by pass and prefer to avoid using the primer, keeping it simple for the users. So draft may be stronger.... but that's my explanation of why I had that idea.
My interest is to learn what simple features can improve combustion, and as a "Non Testo owner" user/builder following your tests is of great help. I am glad there are new threads on the Testo topic to follow too, and have seen your answers and agree in many points. Very valuable to have people like you, MatthewW, Peter and many others sharing your experience and knowledge.
Regards
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 29, 2017 0:48:55 GMT -8
Hi everyone, A new trial with a threshold of 60 mm instead of 80 mm. The threshold is 75 mm away from the inside part of the door. In the Peter's redbell the threshold was 55 mm away from the inside of the door, so it's slightly different. The heater was cold (27°C on the front lintel), there was a layer of black coals coming up to the top level of the threshold. I used 5-6cm in diameter logs. Slightly bigger than what I normally use for "hard testing". Lightning from the bottom. I didn't use the glass to see the vortex from the top. There is a round metal sheet instead inserted in the plancha. Mean values : - global efficiency = 87.5%- combustion efficiency (calculated) = 98,0% - TF = 132°C - CO = 451 ppm - COnd = 1848 ppm - CO13% = 457 ppm- O2 = 13,1% Combustion analysis :  I'm still trying to push the heater to its limits but as I used bigger logs it was hard to get is as low a 9.0% O2. I succeeded to push it to the tipping point two times only — 9.0 and 9.2 O2% which is approx 0.5 O2% higher than the previous configuration with an 80 mm threshold. So it seems to be less stable. At best the CO was at 80 ppm but during the most part of the burn it was between 120 and 80 ppm. End draft of 20 Pa with TF = 140°C I measured the max. temperatures on the plancha during the burn : at t=15 min, T=350°C at t=24 min, T=450°C at t=32 min, T=480°C Some comments : - It's very clean burn I'm impressed ! The start and the end of the firing were better than usual. I need more testing to confirm this behaviour. - The temperature on the plancha is particularly high, on the next iteration there will be a better system to protect the plancha from bending. - The tipping points happened when the flames where going OUT of the small bell. Now the exit of the small bell is 1.5 times the CSA. I'll try to make it lower to match the CSA and to force the flames to stay longer in the bell. Regards, |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 29, 2017 1:11:19 GMT -8
manU thanks for your comment, I'm happy it's useful to you. Where your builds rockets no ? Because there is a big difference with rocket stoves -- they can burn well with a very very small draft. Last time I analysed a rocket ( uzume-asso.org/experiences.html#chauffe_eau) the draft at the end and during most of the burn was 1.3 Pa only ! In the first batchrockets I've built I didn't add no bypass and in my opinion that was a mistake because the chimney flues where not good enough and because the batch needs a decent draft to burn cleanly. With the bypass and still a poor chimney it's much better. I'm not done with the draft measurements but I think I can say that the draft during most of the burn on this cookstove, when the chimney flue temperature is around 120°C one meter above the plancha, is a depression of 13-15 Pa. That's why I'm not conviced about the flat deflector plate you proposed.. but maybe I'm wrong ! On the next version of this cookstove there will be a thermometer in the chimney flue and the instruction will be to keep the bypass open until the temperature is 150°C or something like that, depending of the chimney draft. Regards,
|
|
|
|
Post by peterberg on Dec 29, 2017 4:16:37 GMT -8
A new trial with a threshold of 60 mm instead of 80 mm. The threshold is 75 mm away from the inside part of the door. In the Peter's redbell the threshold was 55 mm away from the inside of the door, so it's slightly different. The heater was cold (27°C on the front lintel), there was a layer of black coals coming up to the top level of the threshold. I used 5-6cm in diameter logs. Slightly bigger than what I normally use for "hard testing". Lightning from the bottom. There's another difference regarding the threshold: mine is about 55 mm from the door, yes. But the threshold is made of a steel L-profile while the horizontal flange is positioned ever so slightly below the door's T-profile, directly above the floor channel's opening. This leads to a reduction of over fire air, the distance from L to T profiles is a mere 25 mm when the door is in closed position. Opening the door a crack will enlarge the primary air opening about as much as 50%. Since I don't have a bypass, I always start the heater with air inlet fully open and the door a crack. When the end temperature rises over 80ºC I close the door. The end temp will drop 5 to 10 degrees as a result and rises again slowly up to 100ºC. Firing two full loads a day will let the temp rise to 120ºC by the way, not necessary now because the temperature levels are still above freezing. Edit: I start a fire by top lighting it, no exception. Refilling on hot coals is another matter, I tend to use the biggest pieces I have in store, about 8 by 8 cm. |
|
Should have measured & reported those too. Nevertheless, both tests were conducted exactly 24hrs after previous full batch burn in very similar circumstances. 
