|
Post by 55 on Sept 22, 2020 21:16:13 GMT -8
I've been designing and redesigning a heating system for my home, intended to replace an outdoor boiler that is aging and slowly beginning to leak fluid. I have PEX in my floors and a couple of radiators with blowers that has all proven relatively effective over the last many winters. Looking through various threads and videos on the subject, and chatting with a couple of well-known designers of rocket systems, I see that most folks are contemplating using a closed system with a copper tube-in-hot-water, fluid/fluid heat exchanger plus a circulation pump. As my current boiler is an open system with an open breather on the tank, running a send and return loop directly from and to the reservoir, rather than through an immersed exchanger coil, I figured I'd see if I could design my own batch box-fired system based on the same principles. What I have arrived at, is two possibilities. Both involve building a tank that is a "box inside a box" approach, so that the fluid tank has a large passage running right through the inside of the fluid, which would then have the flue gasses from the batch box run through it. By making this interior passage into a very wide and very narrow-height rectangular shape, a controlled amount of flue volume can be given a much greater surface area than a cylinder of the same volume provides, making the heat exchange process faster, especially if the tank is a long and narrow design, allowing a longer flue run through the tank. The tank could be either mounted horizontally, with the core passage offset towards the bottom so the gasses pass horizontally mostly "under" the fluid mass in the tank, or, it could be mounted vertically, so the gasses are fed to the top of the tank from the flue and passed downwards through the interior passage and around the tank. I have seen a system using stainless steel tubes in a similar idea, on Peter van den Berg's (did I spell it correctly, Peter?) site, though I think it was another closed system with a limited fluid volume and no reservoir tank. In my design, ideally, the tank would be mostly enclosed in a series of mass brick stacks to extend the tank's retention against the constant heat loss being circulated to the floors, and all of that, housed inside a bell which is fully lined with ceramic fiber board or ceramic wool, for exceptional heat retention of the mass and tank. The system would be open vented and unpressurized, and would have other routings available for the flue, such as a mass bench or bypass, and could even be operated thermostatically to switch between the two, not unlike the automated damper control on my existing boiler. Ideally, the system would be continuously run for as long as it took to get the tank (200 gallons/800 litres, maybe) up to a sufficient temperature, which with my current boiler, is anything over about 140 degrees F, up to about 175 F or so, measured at the zone distribution manifold in the house.
Having both the heater core plus the fluid tank indoors and close to the circulating system will be incredibly more efficient than the current 40-50 yard run of underground tubing being heated by an inefficient wood hog boiler which exhausts much of its output straight into the winter air, and loses enough fluid heat into the ground that it visibly melts snow along the heater line run in the fringe seasons.
The main drawback I can see in my system, is the water tank exchange potentially killing the draw until it gets a bit heated. I figured running the bypass to light the stove and heat the exhaust, and then adding on the tank flow a bit at a time, keeping just a notch of bypass open to keep the flue warm, would probably work, though it'd take some babysitting. The vertical approach may function better in this regard. The other thought, is to build that same system concept, but with less efficient, slower heat exchange, in trade for a more consistently manageable draw, and just let the flue gasses heat some slower-transfer masonry mass, and have that mass do all or a major part of the water tank heating.
I'm trying to refine my plans enough to build it once and have it actually work... Any thoughts on these ideas would be much appreciated.
|
|
|
Post by peterberg on Sept 23, 2020 7:33:48 GMT -8
Hi 55, welcome to the boards. The best system to transfer heat to water is the crossing direction method. Water up, hot gases down as in a bell system. I don't see any advantage of a passage right through the tank to be honest. The brick bell around the tank and the tank itself upright means the batchrocket can be built beside it, feeding the hot gases out of the riser directly into the bell/tank assembly.
One important point: doing it like this would mean the 90º direction change above the riser should be much wider than the riser itself. This is to compensate for the lot of friction you will get there, say 200% of the riser csa would be good. And all sturdy refractory materials, of course.
|
|
|
Post by 55 on Sept 23, 2020 16:55:40 GMT -8
Hi, Peter, and thanks for your input. We've emailed each other a few months back. The idea of the passage through the tank, was to increase the transfer rate and efficiency by multiplying the surface area available for the gasses to contact the tank and fluid. Yes, the system would be drawing hot fluid off the top of the tank and returning cool fluid at the bottom, so the flue gasses and fluid flow would be opposed.
I felt the increased efficiency of transfer using a passage through the tank would translate into less fuel consumed to get the tank up to temperature, and in a shorter time span. Among other things, I noted the tankless "tube" system on your website, in arriving at this alternative concept.
Because of some other considerations of system flexibility, I have been leaning towards building a riserless core instead of a vertical riser, so I could create the flue output lower down, vertically-speaking. I was then thinking to feed it to a vertical manifold (think, a small, super-insulated CFB bell) that would optionally distribute the gasses to whatever use, via an on/off gate on each output, and the top of the manifold would feed the 200% CSA "corner" into the bell and tank, also with a gate, so I could run other components without heating the tank (I'm thinking to incorporate a small cook top/oven and a passive mass bench as well as a startup bypass). If I use a riserless design and a manifold preceding the tank and bell, even very well-insulated, that's another reason to increase the tank surface area, too, in my opinion.
In the system as (roughly) described, do you feel the tank with just its exterior surface area and 800 litres of circulating fluid in it, will be sufficiently heated to get to temperature fairly quickly and maintain its heat under fire, while the large radiant heat system is stripping it off (I estimate at least that much fluid again, flowing through the PEX, to and from the tank and dispersing the heat into the floors and two radiators equipped with blowers)?
Thoughts on the rest of my palaver? Thanks for your help, Peter, and anyone.
|
|
|
Post by peterberg on Sept 23, 2020 23:54:50 GMT -8
In the system as (roughly) described, do you feel the tank with just its exterior surface area and 800 litres of circulating fluid in it, will be sufficiently heated to get to temperature fairly quickly and maintain its heat under fire, while the large radiant heat system is stripping it off (I estimate at least that much fluid again, flowing through the PEX, to and from the tank and dispersing the heat into the floors and two radiators equipped with blowers)? I don't have the foggiest idea, honestly. I am a front end man, all things concerning the core and combustion quality. Bell technology crept in because I couldn't measure a system outputting 900º C so in order to get the chimney temperature low enough a bell setup seemed the best option. Can't help you with this, sorry.
|
|
serg247
Junior Member
The mountain can not be conquered, it can allow it to ascend...
Posts: 111
|
Post by serg247 on Sept 24, 2020 0:09:37 GMT -8
|
|
|
Post by 55 on Sept 24, 2020 6:14:42 GMT -8
Thank you, but I'm not sure I want to click on the link without any background... what are you offering, Serg247?
|
|
|
Post by 55 on Sept 24, 2020 6:51:02 GMT -8
In the system as (roughly) described, do you feel the tank with just its exterior surface area and 800 litres of circulating fluid in it, will be sufficiently heated to get to temperature fairly quickly and maintain its heat under fire, while the large radiant heat system is stripping it off (I estimate at least that much fluid again, flowing through the PEX, to and from the tank and dispersing the heat into the floors and two radiators equipped with blowers)? I don't have the foggiest idea, honestly. I am a front end man, all things concerning the core and combustion quality. Bell technology crept in because I couldn't measure a system outputting 900º C so in order to get the chimney temperature low enough a bell setup seemed the best option. Can't help you with this, sorry. Funnily, I was thinking last evening, that your major forte seems to be in ultra-refining the combustion process. So... any thoughts on the use of a riserless design (a la Matt Walker) versus a riser, for my application? Specifically, I'm asking about each system's potential for doing the fluid heating job, all other considerations aside. I gather the riserless core is the same for all planning purposes as a riser. Thoughts? Thanks for your help, Peter.
|
|
fuegos
Full Member
not out of the woods yet
Posts: 177
|
Post by fuegos on Sept 24, 2020 10:46:12 GMT -8
Matt has used his riserless core for a water boiler & one of his comments i remember is that not to underestimate the capacity of a water filled core to strip heat out of a system .Take a look at a medium sized car with a 85 KW engine & look at the cooling system ? Years ago we had a Rayburn stove with a back boiler that was behind the brickwork of the firebox .the system held 30 gallons at the most & would take quite a few hours to get up to temps.Maybe reaching out to Matt Walker would be a good idea ?
|
|
|
Post by 55 on Sept 24, 2020 12:51:19 GMT -8
Hi, Fuegos, and thanks for your input. I'm curious about the Rayburn, and how that water tank was heated.... I'm presuming it was just receiving transfer through the brickwork, and that it was a "normal" everyday kinda fire burning beside/below the tank, and not a finely-tuned batch box you could forge with?
I actually email Matt frequently, and we discuss lots of design points, but he says I have to do my own research on the water tank question. He also says the riserless design is the same for "all planning purposes" as an equivalent riser unit. Ultimately, he doesn't have as much experience with water as with heater cores, to comment too deeply, beyond making the same suggestion as Peter, of gasses down and fluid up, flow-wise. I guess we're all facing The Final Frontier.
I have an outdoor boiler here that I want to replace, and I've run it for eight winters so far, and I know what it takes to heat its 580+ litres. That said, the batch rocket will be much more efficient and burn far hotter, and the tank will be able to hold heat much better in my design, not to mention being about 200 litres bigger, and will also be indoors with the mass heater and not out in minus 20-something with cheesy foam insulation under sheet metal and pushing 40m of buried lines that heat the ground impressively well. I think I'm probably down to doing some math on system heat distribution loss versus flue heat energy transferred via [X] unit of tank surface area, and I'll know if I'm in the ballpark.
As of now, using the inner passage tube alone, not including the tank exterior, my hollow-tank design presents about 10 times the surface area to the flue gasses for heat transfer as the boiler does, which is mostly a box of water with its floor and seven 2" cross-tubes above a normally-aspirated, damper-controlled, non-rocket fire. My tank is about 30% larger than the boiler's, so it'll consume some of that extra transfer, for sure. The tank and tube are shaped to maximize transfer through a large surface area into shallow, wide water, with the idea being that the tank will heat very quickly and be easy to maintain at temperature.
Still too much heat transfer in my design? I want to balance efficiency of design with it actually burning and working, too.
This said, the flue gasses will flow downwards and will tend to "push" the exhaust a bit. I also intend to use a bypass to preheat the vertical exhaust tube, and I can easily build this rig with flue flow regulation at the input to the tank passage, to use for a cold start after running the bypass for a bit, so it restricts the gasses to just heating the inside of the tank bell and the exterior tank walls at first, rather than the inner flue passage as well, until the water heats up a bit, so it doesn't trash my draw too completely, and I can then open up the passage and move things along faster once its running properly.
Thanks for reading. Thoughts, criticisms or outright laughter and pointing? : )
|
|
|
Post by gadget on Dec 19, 2020 14:39:17 GMT -8
55, You want to build a heater entirely INSIDE the tank. Its the only way to go if you want to be serious about heating water. If your home is solely heated by radiant then you definitely want a 100% heat the water type of setup. Don't bother wasting time with heated copper tubing or other options that are low output. A heater IN a large open tank is also your safest possible design. I don't think following a modified batch box design passing through the tank will work for this. There would be to much room heating unless you want that also the draft issues mentioned. I'm guessing you have a classic outdoor boiler with out a draft blower? They do ok but they guzzle down the wood by the truck load. The basic classic wood boiler has an uninsulated burn box inside the water tank and a chimney. Simple and reliable but prone to smoking and horrible gas mileage. They require the water temp to stay over 140F so the unburnt volatiles don't form creosote due to an incomplete burn. Since your using radiators your going to need 140F+ anyway. I'm assuming your using a mixing valve for lower water temps to the floor? What type of radiant floor do you have? If your radiators will go lower temp, you could switch to a lower water temp and that will kick your efficiency way up! Problem with this classic design (assuming thats what you have in the old boiler) is the low burn temps and short burn duration(path). This is why you don't want the burn chamber uninsulated next to water. Put the burn box in the water for max heat extraction, but it needs to be insulated to keep temps high in the burn area. Also a decent insulated burn path with some mixing early on as possible. High temp burn, good mixing and long residence time at temp = good clean efficient burn!! They make newer boiler designs that have insulated burn boxes, secondary burn chambers, and typically heat tubes to transfer the heat. They also have to have blowers on the end of the flue for the package to work. They don't smoke out the entire neighbor hood and they use way less wood. I would focus on a modern design similar to this. The other decision is if you want it outdoors. This is much safer but no matter how much insulation, there is going to be losses. If you already have really good insulated buried pipe and don't mind going outside to heat, this is a good safe option. I'm going radiant in the next house we are building, here is what I'm designing; - entire heater built into a very large storage/buffer water tank (1-3K gallons) This large volume will be enough BTU's stored so I don't have to constantly be feeding a fire all day long. This will save the need for a separate tank and heater from the storage tank. Keep it simple. Calculated BTU needs for size. - Tank will sit inside my residence in an attached room or basement. heat losses go towards heating the home. This will eliminate the need for tons of insulation and will reduce my heat loss from an outside boiler and buried lines. I also want to feed the fire from inside the house. - It will be a very hot large volume batch heater (2 hours a day burn time or less)to get the storage tank up to temp quickly. I built a test heater and it put out a ridiculously tremendous amount of heat. It had some design problems but what I learned from it will help me build a powerful wood boiler thats safe and quick. I don't want to spend all day feeding a fire. I might also again make a large box for full size stumps. donkey32.proboards.com/thread/3552/el-tornado-experimental-build-thread-The tank can take other inputs like from direct solar or a heat pump. My radiant will be low temp 80-100F so tank temp will rarely be over 110F. Domestic water will preheat through the tank on its way to the water heater. Wood will be a backup supplement heat. Solar driven geothermal heat pump will be primary heat source. The heat pump will run off panels while the sun is shinning. This works with a buffer tank setup. Flow path should be like the following -Insulated burn box (Im doing a swirl burn box for good early mixing) -extended burn path (box, tube, etc...) -heat exchangers (bells, fire tubes, etc..) -induction blower -flue pipes (plastic if lower temp flue gases) If you do the build, please start a build thread. I would consider going with a blower on your exhaust. There are trade offs but you will get more heat out of your wood. You cant build a high efficiency wood boiler without one. With a blower, you can do what ever design you want. I should add that I'm also planning on putting in a masonry heater in the middle of my house with a classic chimney. We want something for backup and also like to have a nice fire to sit by. I think it is easier to have 2 separate heaters then to try and make one heater do multiple task.
|
|
|
Post by Jura on Dec 29, 2020 12:04:26 GMT -8
acl This will eliminate the need for tons of insulation and will reduce my heat loss from an outside boiler and buried lines. I also want to feed the fire from inside the house. We have 2m 3 buffer tank which will be fed from a 15 cm rokcet (IMHO too small), a cook range, and (in case of emergency)2 x kW electrical coil heaters. () it is ready to be extended for solar (once they are mounted) It was pretty important to project the buffer tank the way water layers do not mix extensively.(it has kinda funnel inside at teh 2/5 of its height) I wouldn't skip insulation. it is really worth of insulating it with perlit or glass foam. as the lost heat dissipated in a house volume will be "lost" (not thermodynamicly) but not directed to heaters will not add to a comfort feeling. And we shall keep the buffer temp as high as possible.
|
|
|
Post by Jura on Dec 29, 2020 12:13:09 GMT -8
In the system as (roughly) described, do you feel the tank with just its exterior surface area and 800 litres of circulating fluid in it, will be sufficiently heated to get to temperature fairly quickly and maintain its heat under fire, while the large radiant heat system is stripping it off (I estimate at least that much fluid again, flowing through the PEX, to and from the tank and dispersing the heat into the floors and two radiators equipped with blowers)? As a safe rule of thumb (AFAIR implemented in a national norm) we say 50l of water in a buffer tank for a 1kW of the stove/kiln
|
|
|
Post by gadget on Dec 29, 2020 16:07:13 GMT -8
acl This will eliminate the need for tons of insulation and will reduce my heat loss from an outside boiler and buried lines. I also want to feed the fire from inside the house. We have 2m 3 buffer tank which will be fed from a 15 cm rokcet (IMHO too small), a cook range, and (in case of emergency)2 x kW electrical coil heaters. () it is ready to be extended for solar (once they are mounted) It was pretty important to project the buffer tank the way water layers do not mix extensively.(it has kinda funnel inside at teh 2/5 of its height) I wouldn't skip insulation. it is really worth of insulating it with perlit or glass foam. as the lost heat dissipated in a house volume will be "lost" (not thermodynamicly) but not directed to heaters will not add to a comfort feeling. And we shall keep the buffer temp as high as possible. Hey Jura, Thats a decent size buffer tank (750 gallons). Maybe not enough for a full house over night but still pretty good size. How are you heating the water? Copper tubing? From my experience, using a rocket heater to warm water does not work very well. Now if you placed the whole thing inside the tank, then maybe you can get some water heating at a decent speed. At that point, why not just build a large burn box wood boiler with insulated burn chamber? If your going to use your 2kw electric element heat it would take you 66 hours to go from 68F to 140F. If your using DC elements, it gets much worse. If you had a 2kw heat pump with 4 COP, you could cut that down to 16 hours. Not enough daylight time for solar but you can get some supplement heat with it. I hope you have a tiny house. I would only consider the "very large hot fire placed inside the tank" route. Its what works, its safe, up to 95% efficient and industry figured this out years ago. Every other option just fall to short. I've tried.
|
|