Feed tube way too hot! Suggestions?

[bryan]

I have built up a RMH using light weight insulating firebricks (they insulate about 5x more than regular ones). Early trials all looked really good -- super fast heat up, clean burn, good draft. So I went ahead and started mortaring it all up.

Got it all nicely built with a few definite hiccups on the way, but now that it is mostly complete it is not working nearly as well. First problem is that the draft is not as strong as the prototype.

Dimensions are pretty standard for an 8" system except it is reduced a bit to about 7.25" equivalent. Burn tunnel is 12", heat riser is 40", feed tube I've tried from 5" to 15" (measured from the top inside surface of burn tunnel).

Playing around a bit with barrel clearances have helped, but not quite good enough. I have only 5ft of horizontal duct with 5ft of vertical. This was a rough guess because with a short horizontal the exit gas is a lot hotter than it will be so I didn't want a big vertical outlet section because that would be an unrealistic amount of 'chimney'.

The problem is 2 fold. First is not quite enough draft, the second is that the bottom of the feed area gets too hot. What then happens is that the whole mass of wood down in chamber ends up catching fire and it even sometimes climbs up the wood. The result is that a lot of the wood is pyrolizing because there is not nearly enough oxygen. It will smoke a good bit and will even sometimes make very noxious yellow gases. There is tons of heat output when this is happening, really its too much. Center of barrel top gets to 1100 degF and outside edge is 500-600. This all occurs after about 30 minutes of burn.

So I need more air but less fire, which is an odd combination. Seems like if I significantly reduce the feed tube opening (like cut in half) it gets a lot better, but that then cuts in half how much wood I can load in (so doubles how often I have to tend it). If I try putting a lid partially on top of the feed tube I get less airflow, more smoke and after say 15min the whole feed tube area is too hot and the whole length of the wood is smoking.

Ideas?

[careinke]

I had a similar problem, Check out my "Another Greenhouse RMH" thread. I ended up putting a piece of re-bar towards the back of the feed tube. Worked great.

[bryan]

Actually I had read that post and tried something similar. It is definitely an improvement, except that it also cuts in half how much wood you can load into it at one time and thus means you will be feeding it twice as often. That actually is not that big of a deal, but was hoping to find a way to have my cake and eat it too

[Donkey]

Err..
The downdraft feed box you find in RMH's tend to be small, fiddly things.
I've tried to make bigger ones with similar (kinda) results to yours.. Gotta keep the cross section the same all the way through and can't cheat the feed box bigger (sadly), at least not that I've seen yet (with the downdraft feed boxes)..

The #1 cause of poor draft is around the barrel, either at the top or the bottom of it, usually the bottom where it makes a seal with the bench, etc. That's the easiest place to screw up, and no matter how many times I've built a rocket stove, I STILL almost always need to pull the barrel up and carve out the lower chamber a little and when I set the barrel down again, I make it just a tad bit raised.

I'll need more data if that doesn't just fix it.

[bryan]

Ok,

I had been fiddling with the barrel a fair bit already. The bottom transition into the tube is actually a welded piece. A simple description is that looking down from the top of it you have the 8" tube mated to the barrel and on the one side is a triangle that is 10" on a side that is on the top and bottom. Sort of like a big chamfer blending the barrel to the tube. Actual part is a bit more complicated than that. But it made a noticeable difference over simply connecting the 8" right to the barrel. Could still use a bit of refining, but think it should be decent at this point.

The height of the barrel top above the stack started off at 1.75". It drafted noticeably better at a little over 2" and as high as 3" also seems to work well but is much more fussy about how hot the fire is. So thinking 2.25 is about right.

The one dimension that is a good bit off of the base Ianto book design is the clearance between the heat riser and the barrel. Due to the materials I had on hand (use water heater tank, ventilation ducting) I wound up with about 2.25" clearance all the way around the barrel. So the cross section going down the barrel is about 75% larger than the burn tunnel and heat riser cross section. Thought that having it bigger would not be an issue, but maybe it is. Can try stuffing something in there to take up more space just to see if it changes anything.

At this point the level of draft seems to match with online videos that I have seen. Whether already hot or if just started, when a fresh load of wood is put in it burns well when only the bottom of the wood is burning. But then it typically overheats and more and more of the wood mass starts to smoke and catch fire.

Also, to clarify, my feed tube area is NOT larger than normal. It actually started off as 6" x 7" which is consistent with the cross section of the burn tunnel. I have reduced it down to 4.5x7 and still had trouble. Smaller than that and it noticeably starts to hurt the air flow. Careinke's method of using the standard size feed tube but putting a piece of rebar in the middle seems like a better approach because then there is no air restriction from cutting the whole feed tube in half. But it is less than ideal since you can then only load 1/2 as much wood and also seems like plenty of times the sticks could get hung up on the rebar.

Considering trying a sort of square metal feed tube that goes all the way down in with the hopes that having 3 metal walls down in the burn area will draw out some of the heat and reduce the tendency of the whole load of wood to catch fire. The problem I suspect would be that having a heated metal feed tube could create some reverse chimney effect by preheating the air.

So maybe the solution lies in still greater draft?

[Donkey]

Hmm.. Are you burning a lot of softwoods?? They tend to "ladder up" and out..
I wouldn't reduce the feed size any, system size (in cross section area) is good, though I like 'em (just a little) longer and narrower (though the same volume) so I can get my arm into the burn tunnel for cleaning.

There shouldn't be a problem with LARGER gaps in the barrel.. I've built stoves with huge gaping spaces in the barrel area with good results. I'm not really picturing your bottom end barrel to tube junction..
I've found that the bottom junction is difficult to envision, difficult to describe, and easy to get wrong. The exhaust path down the barrel is basically a torus shape and getting a clean transition, all the way around, into a single pipe without pinching flow is harder then it first seems. You actually need to grow in volume, starting from farthest away from the pipe, growing towards the pipe, like and exhaust manifold in a car.

I recently had trouble with a little six inch stove in a cob hut here at my place.. The small space made for tight tolerances around the barrel and I had to pull it off 4 or more times to carve out underneath and get just the right seating, etc..

[bryan]

Been out testing again tonight. Found some things out and some new questions too:

1) Feed tube sizing. I significantly reduced the feed tube size down to 6.5" wide by 4" front to back. The system size is about 7.25" dia so that is a major reduction from a cross sectional area of 26" in the feed tube to 41" after that. It drafted a bit better and majorly reduced the smoking. Still smokes a little though. Also it stopped the fire from climbing out of the brick portion of the feed tube. Still gets a bit too much burning at one time though. Still has large swings in barrel top temps, but not so extreme. Had been going all the way up to 1100degF at the barrel center, now 'only' hits 800.
2) To allow me to adjust the height of the barrel and test the gap between the barrel top and the heat rise top, I had left the bottom of the barrel unsealed. This left about a 1/8" gap for air to go in. I realized it was drawing cold air in there and was stealing draft from the feed tube. Noticeably better after sealing it, but now figure I have to go back and retest different barrel heights with each time resealing the base.
3) Wood type is definitely a factor. Had been using scrap 2x4's for testing. Tried some oak and it behaves much better. Unfortunately when it is finished out at the yurt I mostly have doug fir. Probably won't be as fast burning as kiln dried lumber though.
4) Final tuning still left with some smoke. It smells lightly of a wood fire (before it stank like poison) and in poor lighting you can't barely see it. But if I used a shop light and aimed it at it then I was quite visible. How clean are you guys getting these things to run??

Now that I've changed more than one thing at the same time, I probably need to go back and retry a more standard size feed tube. I suspect that because of the ultra-insulation behavior of the refractory bricks vs clay that bottom area is a lot hotter than normal. So it may not be able to be as large as a regular one, but think I may have gone too far in the direction of reducing it.

Donkey in terms of the barrel into the bench pipe, what do you mean by the gases move down in a torus shape? A torus is a doughnut. Do you mean the bottom of the barrel should have a bulge as if it had a torus stuck in it? Maybe offset towards the pipe side? What I made was basically if you took the pipe right into the side of the barrel and then stuck a pie piece on one side. I kept it fairly simple since it was a weld fab in my case. But sounds like I need to take another look and blend it over a much bigger area.

Bryan

[canyon]

It can be tricky to discern steam from smoke. Depending on outside temperatures and moisture level in the wood etc. steam will look like smoke except it dissapates quicker in a ghostly fashion. Unless I load up with some knotty spruce or something with less than ideal seasoning I only see smoke when lighting and coming up to temp.

[Donkey]

Feb 9, 2011 22:03:41 GMT -8 bryan said:

Donkey in terms of the barrel into the bench pipe, what do you mean by the gases move down in a torus shape? A torus is a doughnut. Do you mean the bottom of the barrel should have a bulge as if it had a torus stuck in it? Maybe offset towards the pipe side? What I made was basically if you took the pipe right into the side of the barrel and then stuck a pie piece on one side. I kept it fairly simple since it was a weld fab in my case. But sounds like I need to take another look and blend it over a much bigger area.

The gap between the barrel and heat riser forms a torus through which the exhaust flows. Depending on how big the gap is, your necessary flow area is spread out throughout that torus. You've got to gather the exhaust from that torus, all the way around the barrel, without pinching flow, to where the pipe takes it away, off to the side someplace. I've been surprised many times in the past at HOW QUICKLY the needed volume (of the flow-path) can grow.

It's a true-ism that where the shapes change will be tricky volume adjustments. For instance, I like to use tubular steel for heat risers and brick for the feed/burn tunnel.. For an 8 inch system, the void in the brick part is roughly a 7" square, the pipe though is an 8" circle.. When you set 8" pipe on a 7" square, you just made a narrow spot. To repair, you gotta cheat the bricks out a little bit right where the pipe will set, the shapes still mis-match somewhat, but it doesn't seem to harm anything, probably creates more turbulence which is good.

I've found that the transition at the bottom of the barrel is trickier still. The volume changes are difficult to visualize and in curvy cob, crazy hard to calculate.. You have to move from a torus, around the guts of the stove (which are always in the way) to a circle (the pipe) while maintaining enough support for the barrel and everything else.

Usually, in cob, it's as you say, a larger torus surrounding the barrel, bulging towards the pipe.
You might indeed need to blend it over with a larger area.

[bryan]

Well after quite a few hours of fab time I welded in some very large chamfers blending from the barrel to the outlet duct. Noticed some improvement in the drafting, though it was not dramatic. Rechecking the temps though showed that the gas flow was much more even going into the piping (before I had a very clear hot spot in one section).

Testing with different types of wood and different diameters was very enlightening though. Not suprisingly scrap 2x4's burned the fastest, followed by doug fir from the yard, then madrone then oak. Basically found (as others probably already knew) that the softer it is the more likely the fire is going to climb up the feed tube. So splitting soft stuff into larger chunks was a big help.

Also played more with different sized openings for the feed tube and also something similar to Careinke's suggestion to use rebar to leave an air space behind the wood. Similar results-- soft wood it worked better with a smaller feed tube opening of 4.5x7 and hard wood behaved just fine with a full 7x7 opening. So going to finish building it with a 7x7 opening but then make a sort of wedge shaped stainless piece that will drop down into the opening to act as a size reducer for when I burn soft woods. Sort of a plate with a slot opening the bottom.

If it works particularly well I'll post details.

[nedreck]

The shape of the entrance to the flue can have significant impact even if the cross sectional area remains the same.

Bryan, take a hard look at it and consider the distances involved, in other words how far does the air travel on the top of the entrance to the chimney vs the bottom's distance.

I mention this because we have a very low pressure exhaust and if that design is creating eddie currents and builds excessive turbulence, you may have more than buoyancy fighting you on exit. When you are working within the 1/10ths of an inch of water column pressures, very minor issues such as eddie’s can certainly rob you of up to 50% of your draft capabilities.