|
Post by hollandlef on May 11, 2019 12:38:56 GMT -8
Hi All,
I live in southern Spain and we're renovating our entire cortijo here. A part of this were are putting a new heating system. I have previously posted about my thoughts here: donkey32.proboards.com/thread/2008/heat-exchanger-batch-rocket-boiler but our requirements have changed somewhat since then - we're now really just looking for an efficient water-heating solution, because our space-heating will be hydronic
Our primary heat source will be the sun, heating a 1m3 thermal store using solar thermal panels. This heat will be used to warm DHW via coils immersed in the store, and the store water will also be drawn off directly to pump around a radiant floor. The solar system is an open-vented "drain-back" design, so the store is unpressurised. I'm proposing to provide back-up heating for winter months using a batchbox rocket that will heat the store water, but I need to work out a way to do this safely and without excessively complex fabrication requirements.
There are some practical design constraints:
- The solar panels are going on the roof of our house (3m high)
- The top of the thermal store must be below the lowest point of the panels in order for the drain-back system to work correctly
- The boiler will be installed at ground level
I've seen the batch-box boiler on Peter's site, but this is complex, requires stainless-steel welding and frightens me somewhat because the direct heat is applied to relatively small volumes of water in the heat exchangers and there's a potential for steam flashes if something goes wrong. I am not confident to build something like this safely.
Instead I'm wondering if maybe we could heat a small intermediate copper tank - perhaps 120l - directly from the exhaust gases of the rocket stove and have this thermosiphon to the main thermal store - which may need to be lifted up slightly relative to the height of the intermediate tank in order for this to work.
Some questions: - Is this remotely practical? Are we likely to be able to extract a meaningful amount of heat from the exhaust gases by just passing them over the outside of copper cylinder full of water?
- Are there still safety concerns with this? The whole thing would be open-vented, and the volume of water being directly heated would be large enough that I'd hope there was minimal risk....but it would be good to get some opinions...
- Could the copper tank sit inside an insulated chamber to one side of the main burn chamber/heat riser? So effectively the exhaust gases are immediately directed sideways after leaving the heat riser, into an insulated bell where they fall down around the outside of the copper tank, and then escape at the bottom into flue that leads to a vertical chimney outside the house? My concern is that this might stall as there is nothing playing the role of the "barrel" in the typical rocket stove design - an immediate heat sink that absorbs and radiates some of the energy of the exhaust gases as soon as they leave the heat riser, encouraging them to fall and continue on their journey through the system. I have attempted a very crude sketch:
Thanks in advance for any tips/advice
|
|
|
Post by peterberg on May 12, 2019 1:42:45 GMT -8
Some questions: Is this remotely practical? Are we likely to be able to extract a meaningful amount of heat from the exhaust gases by just passing them over the outside of copper cylinder full of water? Yes, I think it will work this way. Heavily depending on the space around the intermediate tank, the larger the space the slower the hot gases will move and the more heat will be extracted. Are there still safety concerns with this? The whole thing would be open-vented, and the volume of water being directly heated would be large enough that I'd hope there was minimal risk....but it would be good to get some opinions... In my opinion it would be safe although in your sketch a vent from the buffer tank directly to the outside is missing. Could the copper tank sit inside an insulated chamber to one side of the main burn chamber/heat riser? So effectively the exhaust gases are immediately directed sideways after leaving the heat riser, into an insulated bell where they fall down around the outside of the copper tank, and then escape at the bottom into flue that leads to a vertical chimney outside the house? My concern is that this might stall as there is nothing playing the role of the "barrel" in the typical rocket stove design - an immediate heat sink that absorbs and radiates some of the energy of the exhaust gases as soon as they leave the heat riser, encouraging them to fall and continue on their journey through the system. I'm inclined to think it would be best to house the tank in the same bell as the combustion core. This bell insulated from the inside so all the heat produced would go into the tank. Enabling the gases to cool and fall down, the tank is acting as the primary heat extractor this way like the barrel in a Ianto Evans' J-tube RMH. There are as many variables as you can think of: space around the tank as mentioned above, surface area of the tank, level of insulation inside the bell, efficiency of the thermosyphon, diameter of the pipes leading to the larger buffer, temperature of the water to start with, and so on. To alleviate some of these concerns I'd recommend to install a bypass from the bell to the chimney in order to get the system up to operating temperature before extracting lots of heat. There will be some soot accumulated on the surface of the tank but in my experience this won't be a big problem. Rather the contrary: I could imagine having a black tank would extract even more heat. The hot pipe from tank to buffer need to be insulated to avoid excessively heating up and flashing water to steam. Just random thoughts.
|
|
|
Post by hollandlef on May 15, 2019 21:46:00 GMT -8
Hi Peter,
Thanks for your quick reply - sorry not to have responded before - our building project started this week and that has been predictably disruptive!
Building the intermediate tank into the same bell as the core sounds like a good plan - it would simplify the design in many respects. I was intending to insulate from the inside, yes, to minimise the amount of "redundant" mass that needed to be heated up. The bell will be in a room where we e.g. dry laundry so any heat that "leaks" will not be wasted but obviously the primary goal is to get the water hot. Given that the core is going to vent directly into the bell, I'm guessing we're going to need some pretty tough insulation, especially directly above the core. Is there anything in particular that people recommend for this?
Are there any principles to follow in working out the volume of the bell relative to the size of the rocket core or is this just pure experiment? I'm probably looking at using a cylinder something like 675 high x 450 diameter - actually probably smaller because I think that's before I strip the insulation off it; internal volume ~85l. I was thinking about a 6" core - maybe even 8"? I'm guessing I want a decent amount of space for the gas to circulate rather than just rushing over the surface of the cylinder and out of the exhaust, especially since little of the heat will be being absorbed by the bell itself. I guess one option would be construct a sort of meandering air path that guides the gases by having bricks that project from the walls of the bell almost to the edge of the cylinder inside it, forming a crude downwards spiral... but I guess that might wind up being counter-productive. I'm not sure where the balance lies between extracting sufficient useful heat and stalling the thing - I've only ever built a very simple "barrel" rocket so far.
How would a bypass work if the core is directly inside the bell? Have a closeable exit from the bell immediately above the core that joins directly to the chimney? I guess that should work ok...
|
|
|
Post by peterberg on May 16, 2019 0:18:00 GMT -8
Given that the core is going to vent directly into the bell, I'm guessing we're going to need some pretty tough insulation, especially directly above the core. Is there anything in particular that people recommend for this? Superwool, made by Morgan Thermal Ceramics. Thickness 25 mm would be adequate. Are there any principles to follow in working out the volume of the bell relative to the size of the rocket core or is this just pure experiment? I'm afraid it's just pure experiment. But allow me an educated guess: core plus riser do need quite some space, front to back. Gases need to go around the core, left and right. In practice, this would be a square(ish) footprint of the bell. Beside the riser, well above the firebox, there would be plenty of room to house the tank. The recommended space above the riser is the diameter of the riser as a bare minimum, twice of that is plenty. So build your core, build a square bell around it, insulate it from the inside and place the tank as high as is practical in the space that's most suited for it. Of course there need to be some space under the core to allow the gases to stream to the exhaust. So lift the core 40 cm or more from floor level on a steel frame, also convenient to load the firebox. I'm probably looking at using a cylinder something like 675 high x 450 diameter - actually probably smaller because I think that's before I strip the insulation off it; internal volume ~85l. I was thinking about a 6" core - maybe even 8"? For such a small tank a 6" core would be sufficient, I'd say. I'm guessing I want a decent amount of space for the gas to circulate rather than just rushing over the surface of the cylinder and out of the exhaust, especially since little of the heat will be being absorbed by the bell itself. I guess one option would be construct a sort of meandering air path that guides the gases by having bricks that project from the walls of the bell almost to the edge of the cylinder inside it, forming a crude downwards spiral... but I guess that might wind up being counter-productive. What you are looking at is heat extraction by lots of length, i.e. a serpentine. It works, but you'll add also a lot of friction and thereby running the risk of stalling the draft quite easily. On the other hand, in a downdraft situation (gases need to go down to the exhaust) extraction is merely done by giving lots of space to slow the gas velocity down. The longer stay (of the gases) around the tank will extract lots more heat this way. How would a bypass work if the core is directly inside the bell? Have a closeable exit from the bell immediately above the core that joins directly to the chimney? I guess that should work ok... Having a bypass directly above the riser means thermal stress to the valve could destroy it in time. It's much better to place the bypass in the side of the bell somewhere halfway the height of the tank, out of the region of highest heat. By doing it this way heat extraction level is significantly lower with bypass open so the combustion core and the chimney stack will be given the chance to heat up properly.
|
|