|
|
Post by hollandlef on May 1, 2023 1:06:22 GMT -8
Hi all,
After several years of not quite getting around to it, I have finally had a go at building a batchbox rocket based on the info/designs on Peter's website. It's a 200mm system with a floor channel made from a 80x50 horizontal box section feeding into a 45x45 vertical box section for the "beak" that sits in front of the slot into the heat riser. The firebox is made of firebricks with a cast refractory top; and the heat riser is made of 5cm thick insulating refractory. Firebox and riser are both insulated with a layer of ceramic-fibre blanket.
The heat riser vents into what will be a pizza oven, which is in fact a kind of "mezzanine floor" that takes up about 2/3 the width 3/4 of the way up a brick bell. There's a small bench leading out of the bell and there's also a heat exchanger for heating water, but I don't think any of that is particularly relevant for the problem I'm having. I'll attach a sketchup file and some pictures to give you a clearer picture.
The basic issue is that I can't get an effective secondary burn in the heat riser - and especially it doesn't seem that my floor channel is really doing anything. If I get it running with no door on, it draws fine and pulls the flames through the slot into the heat riser, and I do get a satisfying roaring sound, but the exit temperature is nowhere near as high as I'd expect (pizza oven thermometer doesn't go over 300), and more worryingly, if I get it really going with lots of small, dry branches, it burns really, really sooty. I don't think this is a problem with the exit path choking the thing, because I get exactly the same result with the pizza oven door open, which provides a large, unobstructed exit path 50cm away from the heat riser; and it's not like it's backing up - it's still roaring away into the heat riser. If I close off the front of the firebox to limit primary air down to the kind of CSA it's supposed to have ( ~25% of my riser CSA), I don't get all the soot/black smoke, but it stops roaring, the burn slows right down and the temperature of the gases at the top of the heat riser drops dramatically. Interestingly, blocking the entrance to the floor channel doesn't make a discernable difference - it's as if this isn't really doing anything at all. The other concerning thing is that the stove isn't burning all the fuel effectively; I quite quickly end up with a large mound of glowing embers in the firebox that isn't burning hard enough to send much flame/heat into the riser, and this then burns slowly for many hours, leaving behind a lot of ash and charcoal pieces.
In summary: the thing burns hard and dirty (you can see flames coming out of the top of the heat riser + a lot of smoke in the pictures) when it first starts, if I have the front wide open. But if I close it down or just let it burn down for a bit, it stops "rocketing" and just smoulders away slowly. Any suggestions?
|
|
|
|
Post by martyn on May 1, 2023 3:28:09 GMT -8
How long ago did you build the stove and how many fires have you lit?
The most common reason (if build to spec) is moisture in the bricks and any other refectory material, any moisture will hold back the heat Considerably!
I cant download your file so I cant see how you have designed the oven but I am not sure how well your oven will work as a pizza oven as the floor really needs to be as hot as the dome.
Have you cast the riser from insulating refractory? That might take an age to dry out too….
|
|
|
|
Post by hollandlef on May 1, 2023 4:46:11 GMT -8
How long ago did you build the stove and how many fires have you lit? The most common reason (if build to spec) is moisture in the bricks and any other refectory material, any moisture will hold back the heat Considerably!
The core of the stove was complete over a month ago now, and the chimney was cast nearly 2 months ago. We've lit 2 fires in the core before the rest of the stove was constructed - one very small to help drying out, and one slightly bigger. The rest of the stove is pretty new - finished over the weekend, and we only lit a couple of fires so far. The moisture in the rest of it might well explain cooler temperatures in the oven but I'm still puzzled why we're getting so much incomplete combustion with lots of soot given that the core really ought to be pretty dry by now. I _think_ it's built to spec; I followed the spreadsheet on Peter's site for the dimensions of the firebox + heat riser. I'm using the May 2019 revision of the floor channel on the page batchrocket.eu/en/designs. The CSA of the horizontal feed tube is a little bit less than a faithful scaling of that design (3744mm2 rather than 3918mm2) because of the constraints of the steel tubing available. The stub is 45 x 45 section, which is exactly the same proportion as the original design, and the stub height is 160mm to the middle point of the beak - which again, is the same proportion as the original AFAICT, i.e. pretty much exactly half-way up the port. I find it hard to believe that the 4% reduction in the CSA of the floor port is completely killing its effectiveness to the point where you can't really tell the difference in the burn when it's blocked vs open
Does the download link not work at all? It's a sketchup file and should work ok in the web app. Anyway, yeah, that's a risk but there wasn't really a good way round that without the pizza oven being insanely high off the ground. The floor of the oven does take a lot longer to heat up, obviously, but it does get reasonably hot eventually and it retrains heat pretty well (pre-cast heat accumulation slabs). The pizza oven application was somewhat secondary to the water heating side of it anyway.
|
|
|
|
Post by hollandlef on May 1, 2023 9:02:00 GMT -8
Interestingly, I just noticed this comment about DSR1 on Peter's site:
This is actually pretty close to the behaviour I'm seeing with no door on the firebox: a hard burn with lots of sooty smoke. Is that just a normal symptom of too much primary airflow? If so, I guess my real problem is not so much that this behaviour - which should be fixed by throttling primary air down to the size specified in the designs - as the fact that I'm not going my secondary burn when I do this.
|
|
|
|
Post by peterberg on May 1, 2023 12:02:20 GMT -8
Hi Lucian,
Running a bare core will work, in a well-built bell system there won't be much difference. Running a bare core of the DSR1 is radically different as compared to one built into a bell. It's like comparing apples and oranges.
There's a marked difference between a core equipped with a door and a p-channel or floor channel and one without, according to my trusted Testo analyser. Most of the time the difference isn't visible to the eye, that's for sure.
Checking whether or not the floor channel is drawing can be done with holding a match or lighter in front of the floor channel's feed. The flame will get sucked into the feed every time.
The feed of the floor channel being somewhat smaller than the recommended size won't pose an important burden on the workings of this heater I'd think.
Running a straight core without a door means the gases are sucked more freely through the system, cooling the front of the fire at the same time. That said, my heater need to start with the door open a generous crack on top of the normal recommended air inlet, the door can be closed in two steps once the chimney temperature is above 100 ºC.
In general: a fresh build won't work as well, to say the least. In order to get the right idea how it works the thing need to be dry first. In a domestic situation this will cost at least a couple of weeks to a full month, provided the heater is fired daily twice or more with half loads of very dry fuel. Once it's completely dry, one can be sure what the definitive behaviour of the system would be.
Now to the actual heater: I studied the pictures and the drawing. It looks like a full scale 200 mm system, judged by the size of it. Doing a rough calculation regarding the Internal Surface Area reveals that it's overspecced by a fair margin. My calculation shows it's ISA is 11.87 m², although it looks like the outside of the boiler part is insulated so this figure might be smaller by 1.2 m². Still larger than the recommended 9.4 m² so problems could be expected.
The boiler part is added on top of that ISA, even when empty. Filled with water I fully expect it would stall the thing completely. The exchanger consists of 36 pipes, 50 mm diameter and 1.2 m long, representing an area of 6.73 m², on top of the 11.87 or 10.67 m² depending on the insulation around the exchanger. The exhaust pipe is situated at the end of the bench, there's no bypass as far as I am aware of. Flames out of the riser usually are a sign of incomplete combustion, so smoke in that situation is not uncommon.
So, on top of the usual drying period, the heat extraction area of this heater is grossly overspecced in my opinion. Hard to say what to do about it, sorry for that. Taking out the boiler part and the bench could alleviate all this, leaving everything in place would make a bypass really, really necessary.
|
|
|
|
Post by martyn on May 1, 2023 12:13:10 GMT -8
Deleted post due to Peters response posted at the same time.
|
|
|
|
Post by hollandlef on May 1, 2023 13:07:14 GMT -8
Hi Peter,
Thanks so much for you swift and detailed response. Good to know that there's no relationship between the behaviour I'm seeing and the behaviour you described of the DSR1.
The issues you have identified with the bell sizing may not be as bad as they first appear; the diagram and the pictures omit some quite important details. First off, there is a bypass! The heat exchanger sits inside a vertical tube of vermiculite/ceramic fibre boards; there are steel plates that be swivelled to direct the air either through this tube/the exchanger or through the gap in the oven floor beside it directly into the main portion of the bell. You can actually see these plates in the SKP file behind the main design, but they're not shown in their correct place inside the stove. Secondly, the water will be pumped by a speed-controllable pump driven by a Raspberry Pi, and there is an anti-condensation valve to ensure that the water coming into the exchanger will not be excessively cold once running; taken together, this should allow me to avoid pulling too much energy out of the system at once - no doubt this will take some tuning. Thirdly, there's a lot more insulation than is obvious at first sight. As you can see in the photos, the rear of the bell/bench is built against an existing stone wall. There's 50mm of ceramic fibre blanket between the brickwork of the bell/bench and the wall, which should cut heat loss from that portion significantly. The oven section at the top is also completely insulated with 25mm of ceramic fibre blanket between the firebrick skin and the outer brickwork + a 50mm ceramic fibre board on the top of the oven. I'm not sure how to calculate the effect this will have on the heat loss precisely, but it's very noticeable that the top course of bricks in the bottom part of the bell is _significantly_ hotter than any part of the oven section. If anything I think I've probably erred on the side of overly reducing the heat output of the bell here.
At any rate, with the heat exchanger bypassed, I've not noticed the thing stalling - whatever the core is pumping out seems to make it through the bell along the bench and out of the chimney, whether that's relatively clean vapour or the dark sooty smoke from when the thing is burning really hard with a heavy draught. As mentioned in my original post, I can also test the core somewhat in isolation by opening the oven door, at which point the thing exhausts directly through there for preference, and I haven't noticed a particularly obvious difference in burn with the oven door open or shut; if it's smoking it smokes that way as well.
We will continue to test it over the next few weeks in the hope that drying and running it in will improve things dramatically. If you have any other suggestions about experiments I can do to tune the core behaviour as I go I'd be very interested to hear (I will check the draw of the floor channel with a flame when we next light it) I will update the SKP to reflect what we actually built a bit more accurately when I get a moment + maybe make a longer video to show a bit more about the construction in case anyone is interested.
|
|
|
|
Post by peterberg on May 2, 2023 6:54:46 GMT -8
It that case, only teething problems could remain as an issue.
The 25 mm superwool between the firebrick skin and the outer brickwork might be too much.
Some more tips what to expect and do's and don'ts.
Please avoid lots of small branches burning all at once in the firebox, the resulting rapid off-gassing could be too much for the system to cope with. Using somewhat larger fuel will result in better combustion.
Please don't forget there shouldn't be any piece of fuel inside the port or riser, however small. People seeing the brighter and faster moving flames in the riser and tend to feed that hotter fire directly. Try it yourself: while the fire is going on with lots of flames in the riser, stick a long piece into the port. The result will always be lots of smoke, even in a bare core. Drawing that piece back into the firebox will stop the smoking within about 30 seconds, depending on the size of your system.
Please refrain from throwing big pieces into the firebox, this will dislocate the firebox' back wall, use a welding clove for reloading. Running the thing with a door means the fire is different than without. Best to close a door, in case you are going to install one, in steps so the fire is able to adapt to the different circumstances.
Happy stoking!
|
|
|
|
Post by hollandlef on May 2, 2023 12:38:06 GMT -8
Ok, that's all very useful information. I was probably using too small branches; and on the most problematic burn I think stuff was probably ending up too far into the port. I'm going to have another go at a burn tomorrow and I'll let you know how i get on.
|
|
|
|
Post by fiedia on May 3, 2023 5:50:10 GMT -8
Dark fumes: my outdoor open prototype systematically did so when I put too much wood at once. On a former 5” batch rocket, I had to reduce the air inlet to avoid thermal runaway (with Peter's help). Besides waiting for your HR to dry out, you can check if your post combustion gets better by increasing your air inlet CSA (door closed). I could not open your skp file (my version is 17, yours 25) and I did not see your door nor the primary air inlet on the pictures. I would like to have a closer look. My 200mm batch did not heat enough to get a nice post combustion (door closed). I had to widen air inlets to get much better results (thanks to Peter again). See post donkey32.proboards.com/thread/3825/air-inlet-optimization-batch-box. Regarding oven temperatures, my oven sitting right on top of the firebox gives floor temp above 250°C but the oven walls and ceiling are not hot enough to cook a nice pizza.   The temperature inside the bell decreases very fast vs height. Up to 600°C at the ceiling, less than 300°C just 22cm under the ceiling. The temperature decreases also very fast after the burn peak. The oven sitting on the firebox remains hot much longer. I guess the best pizza oven might be inside the DSR2 afterburner. I wonder if anyone has tried yet. My floor channel is also 80x50 but ending with a 50x50 “neck”, both are 3mm thick. |
|
|
|
Post by hollandlef on May 3, 2023 13:53:44 GMT -8
That's interesting info. Right now, I have no door. I'm experimenting with a piece of vermiculite board balanced on firebricks pushed up against the front of the firebox to control the airflow. So far, it seems like the best behaviour is actually with no door at all (firebox front completely open). I could certainly live with that if necessary - this is an outdoor installation so it wouldn't matter.
What's still a little upsetting is that I'm still not getting anything like the kinds of temperatures i was hoping for at the top of the heat riser. I was expecting to see the 5-600 degrees that you've been recording, but unless I overcharge the thing with small sticks and generate a whole load of black smoke and soot, I'm struggling to get more than 300. From what I can see there is (at least initially) some post combustion going on (i can see the curls of flame in the base of the heat riser through the port), but it just doesn't seem to be generating that much heat.
The other thing is that it only really burns properly for a relatively short time; after about 30 minutes I get a lot of embers but minimal flame and no kind of secondary burn. The thing smoulders away slowly for a long time (overnight) leaving a lot of ash and charcoal.
I'm not using the _most_ high quality wood - it's mainly small trunks (3-4"), unsplit and with the bark on, but well seasoned and dry (I'm in southern Spain and we're in the middle of a drought, so _everything_ is dry right now). A mixture of stuff - some willow, some olive, some chestnut. It's not split ash, that's for sure, but we always used to burn this stuff in our old stove very successfully so it didn't think it should cause too much trouble.
We'll keep burning and tweaking and see if I can gather more info. I'll post a few more pics/videos soon.
|
|
|
|
Post by martyn on May 3, 2023 21:53:08 GMT -8
A video might help us see what the issue is…..
What about your chimney, is it a straight vertical run, is it insulated?
Fresh builds are very often slow to dry out, there have been lots of similar post on the forum but the systems do eventually dry out.
|
|
|
|
Post by fiedia on May 4, 2023 10:41:18 GMT -8
What's still a little upsetting is that I'm still not getting anything like the kinds of temperatures i was hoping for at the top of the heat riser. I was expecting to see the 5-600 degrees that you've been recording, but unless I overcharge the thing with small sticks and generate a whole load of black smoke and soot, I'm struggling to get more than 300. Now that my air entries are tuned, I get higher temperatures by closing the door. There is a balance between bringing enough fresh air to get a clean burn and avoiding too much fresh air which cools down your fire.
A batch run lats around 45mn. The idea as for most mass heaters is too get a short clean burn and store the heat inside a big mass around the fire. The ember phase lasts longer. The firebox remains hot longer than the HR. That is why I put my oven right on top of the firebox.
|
|
|
|
Post by hollandlef on May 4, 2023 14:41:01 GMT -8
So today we did our first pizza cook. Despite the oven not getting _especially_ hot it was extremely successful. After reading the comments here and looking at the design of the oven, I supported a piece of refractory concrete directly over the heat riser to provide a really hot base. The gases hit this, come forward, and then are forced over the top of the pizza before exiting into the main body of the bell below. This way we get both the hot floor to cook the crust and the hot air to cook the topping. Once it was going we could cook a really good pizza in 4 minutes. I know it's not really how you're "supposed" to fire a batchbox rocket, but we found that for the purposes of pizza cookery, feeding it with a carefully judged amount of relatively small pieces to keep a solid secondary burn going in the riser was actually very effective and quite convenient - we just topped it up a minute or so before cooking each pizza. Video of the modification (small amount of profanity when I put my hand to close to the oven): share.icloud.com/photos/02fnIqHD9ULGeblfkAS8_XvegSelection of not enormously helpful videos of it burning: share.icloud.com/photos/00e97u-Cnh6J3386vlPbTxuuQ You can see various levels of air intake from completely open to somewhat shut off. I'm not actually quite clear what the ideal CSA for the air intake is with this style floor channel, given that the primary and secondary air is going to be shared. When I next run it I might try really closing it down again. What has tended to happen so far is that it just slows right down and the temperature in the oven falls substantially - so it would seem like I'm starving it. But maybe I'm just shutting it down to early, or maybe it's just not drawing through hard enough. I'm also going to try and put an extra section on the (rather short) final chimney to help pull through the admittedly somewhat ambitious amount of bell + bench. |
|
