Hummingbird House: Design for Everyone Part II
Leslie Jackson: Talk about your home’s heating system.Larry Weingarten: On the inside, the heating system is using thin copper tube, the kind that is manufactured for solar panels. Northstar Gardenia in Los Angeles was the place I got it from. They supply solar-panel manufacturers with finned copper tube. I just figured out what lengths I would need so that I could use them cost-effectively. So that if I needed a 6 foot and an 8 foot over there, okay, there’s 14 feet. And so I figured out in even footages all the lengths I needed. They shipped me a 20-foot wooden crate, and I got all my heating fin from them. And I had the chart I made up, and I cut it all to lengths and put it in. Very little waste.
LJ: A little extra thought.
LW: A little extra thought. And I did order some spare tubes, just in case I stepped on something.
LJ: And you could afford to, because you had saved so much material by thinking about it in advance.
LW: Probably because I did much of the work with my own hands, the cost of construction wound up being about $100 per square foot. There is nearly 1,000 linear feet of finned tube in the house, but it was $1.00 per foot—not expensive material compared to other heating systems. Still, the distribution is oversized for a traditional 20 degree delta T.
You look at the heat loss calcs per room and that ultimately tells you how many linear feet of tubing you need per room, because that’s how much emitter you’re going to have. But I put a lot more in than I probably needed to, because I wanted to be able to use very low-temperature water, which would mean more surface area.
LJ: Saving on that end, but is there a trade-off? Using more material to save fossil fuel?
LW: I did build in a way to heat the house using heat from the gas water heater, but I’ve only tested the system to see that it works; I’ve never used it beyond that. Other heat is from a wood stove. Were I grid connected, I might be saving more fossil energy. As it is, I may only be saving solar energy. The plus there is that I run the house on rather small solar-thermal and electric systems. I have six 4 x 8 solar collectors, and the electric system is 630 watts.
If I can use 80°F water to keep the house at 70°F, there are almost no days when I don’t get some solar. There have been times when it’s just cold, but that’s very rare. There’s generally some heat. Even if it’s not fully 80°F in the solar tank—well, it’s 70°F. The house isn’t going to get too cold. There’s 1,000 gallons of mass there. It actually steals the heat out of the water well enough that the pipe coming back from the house can feel quite cold to the touch, because it’s cooling it down to whatever the temperature of the baseboard on the outside walls is. It’s starting off warm up at the top, but it’s cooling all the way down to the coolest air temperature in the house at the floor.
LJ: And this is all done by gravity. So explain the cycle.
LW: Let me start by pointing out two separate gravity systems. One is for water supply and the other is for heat distribution. Domestic hot (and cold) water is supplied from a well, solar pumped, with a ¼ hp pump, up to a storage tank. Gravity feeds down from there.
There are two different sides of the solar set-up, space heating and domestic hot water. It’s easier to think of them separately. Space heating has the heat collection system, heat distribution, and heat storage. Start with the storage. It’s a 1,000-gallon tank, fairly well insulated. To put heat into it, I have solar panels on the roof, filled with water and vodka, and a pair of PV-powered pumps in parallel. Sadly, nobody made a pump that could provide the flow for six collectors. So the PV gets sun, causes the pumps to spin; the more sun, the faster they spin. And that moves the water-alcohol mix through collectors to a coil in the bottom of the tank, which is in the basement (see Figures 1-3). The tank is about a 7½-foot-tall, 5 feet in diameter and made of PEX (cross-linked polyethylene). That’s supposed to be good for hot water. When I got it, the instructions said it could only stand room temperature water in it.
LJ: And what is the vodka doing in it? How many gallons of what proof does it take?
LW: I chose vodka as the right heat exchange fluid. System volume is about ten gallons, half of that is 80-proof vodka. Usually glycol is used, but it’s considered toxic, can turn acidic with overheating, requires more power to pump, holds less heat than alcohol, costs more and needs periodic replacement. Vodka is readily available and even goes on sale sometimes! You don’t ever want the water in the solar panels to freeze. Because if it does, it’ll expand and burst the tubes. That’s what every solar system is based on; how do you keep the panels from freezing. There are a lot of different ways. I could have used a drain-back system, or a freeze recirculation system, but both of those have higher electrical energy needs and more complexity.
So that’s what’s putting heat in the storage tank.
Heat distribution, or taking heat out of the storage, consists of two parts. One is space heating, and the other one is domestic water heating.
Domestic water heating is just a 60-foot coil of 1-inch copper tube. I just bought a standard box of soft copper that has a fair volume in it and a lot of surface area. I didn’t need to use the whole thing, but I thought why not? There won’t be much difference in temperature between the hottest water in the solar tank and what comes out of that pipe. So I’m getting everything I can get. And it’s all single-wall heat exchanger rather than double wall, which is needed when using toxic fluids like glycol or heat transfer oils. Should some food-grade alcohol somehow make its way into the water supply it wouldn’t create a health risk. I’ll add that double wall exchangers cost more and don’t transfer heat as well as single wall does. But anyway, there’s that big coil that feeds the water heater—the backup gas water heater.
LJ: What are you using for backup, and how are losses minimized?
LW: I’m using a direct-vent, 40-gallon propane heater. It’s insulated to R-16 and lives in the warm room with the solar tank. The distribution piping is 3/8-inch PEX, hooked up to a copper manifold mounted directly on top of the heater and well insulated. All hot piping is insulated and inside the shell of the house. My thinking was to reduce the volume of water in the line, for faster hot delivery with reduced waste. I didn’t want an electric recirculation pump, not even a demand system. When water is drawn, only the volume of the 3/8-inch tube is wasted, because the manifold is already full of hot water.
The space-heating part is another large coil. It’s a shorter coil of 1¼-inch pipe in the top of the tank. And the reason it’s a larger diameter is so that there’s much less flow restriction inside it, because that’s a gravity circulation system, unlike the gravity-pressure-driven domestic hot water (DHW). When I turn on the tap, there’s pressure pushing water through the DHW coil. Flow through the 1¼-inch space heating coil is driven only by the difference in weight between hot and cold water, which may amount to—in three stories—½ inch, maybe 1 inch. One column, if it’s warm, might be 1 inch taller than the cold water. To clarify, if you start with two vertical pipes, same temperature, same height, you can make one cold and the other hot. The hot water will expand. Cold water pushes hotter water up. So in this 25-foot-tall column, you may get another inch in height. That’s the pump. That’s what drives it, the weight of that inch of water. So it’s a very minor thing!
LJ: How is the heating system controlled?
LW: Modified wax-filled greenhouse window operators, (one on each floor) close spring-loaded ball valves to allow flow as needed. Springs pull the valves open. Flow is proportional. As the house warms, flow is reduced.
And this is all detailed in great meticulous mathematics in old textbooks and in old engineering books. And it’s very clear, really.
LJ: Handed to you on a silver platter!
LW: Well …You gotta dig, you gotta dig.
LJ: You may be one of the first people to implement this system in this way.
LW: I think it’s the first gravity-driven radiant system done. Gravity hot water has been done forever. Gravity-driven space heating was done, but with the old cast-iron radiators. They used huge steel pipe. All this math has already been done. But no one could use a 1¼-inch pipe to serve their house, because there was no house that used that little heat.
But I did the calcs, and my main line is a 1¼-inch pipe. There (in typical older, leakier houses) you would have 3- or 4-inch lines. You understand the difference in volume—it’s huge. By an order of magnitude. And the water they used was often a lot hotter. By having a really good, snug house envelope, the amount of heat I need to move through this thing is almost inconsequential. Heat loss calcs put it at 27,000 Btu per hour. So that’s equal to a small water heater.
The system in the walls is, again, no pump, so it had to be designed so that there was no way a bubble of air could form anywhere in it. A bubble of air would kill it. Just totally block flow.
So there are loops running at the floor. Loops that are flat, they’re horizontal. They are hidden in that baseboard. Loops are also running higher up in the walls hidden behind the book shelves where the wall bumps out. But you will—if you look around, you will see that the bookshelves are not all exactly the same height. Because the pipes need to drop slowly, so any air will collect at the high point. I have a bleeder on it, which is just a pipe at the very highest point on the system. I’ve run a ¼-inch tube from that all the way down to the basement. So there I have a water feeder at the lowest point on the system. I can hook up a garden hose and fill it with water. And then at the same time as I open it up to fill it, I open the valve to let air out.
LJ: Yeah. Kind of burp it.
LW: Burp it! It’s my kid! And once it’s running a solid stream of water down that pipe, I’m done. There’s an expansion tank on it for thermal expansion, but that takes care of it.
OK, so now I’ve got it full; there’s no air in it. There’s no automatic valve of any sort. There’s no air bleeder that gets stuck with hard water, like they do on every solar system. It’s sort of like a Ferris wheel. There’s a guy Dan Holohan (author of The Lost Art of Steam Heating) who likes to explain gravity flow this way: Any time it moves here, it must move there. If it’s going up here, it’s going down there. And the hot water is rising because this cold water over here is falling. It’s heavier than the hot water. It’s more dense, it’s going to fall. Hot water floats on cold water, if you will. So your Ferris wheel moves. It moves as fast as it can, until friction starts to interfere. The theoretical speed of hot water flow is dramatically faster than the actual speed ever really is, because friction gets in the way. You know—friction on the pipe walls, pipe size. Fittings. So I did whatever I could to use long-sweep (or long turn) fittings. Always reamed the copper ends of the pipes, so that there’s no little lip sticking out, getting in the way. All these little details, trying to make it as smooth and easy for the water to move as I can, because if you have a bunch of 90 degree ells over here, it’s going to affect flow over there, because it’s a Ferris wheel.
LJ: So the only pump that you have is in the solar collector.
LW: Right. Think of it as the solar collection part. There are two little 19W pumps. They’re magnetic drive, so there’s no shaft seal friction to have to overcome.
And I have those hooked up to a 75W panel, which is a little big, but it seems to work out fine. I put a linear current booster in it. It’s of Canadian manufacture; made by Solar Converters, Incorporated. It tempers, if you will, the electricity going to the pumps to get the voltage high enough to get them to kick on, and that works as it should. It all seems to be sized right. So it does its own proportional pumping. The more sun, the faster it pumps. When a cloud passes over, it slows down. You can hear it. You know it’s working. It’s a feedback thing for the people as well. Which is just luck. I didn’t know I’d be able to hear those pumps.
LJ: It’s testimony to how quiet the house’s other systems are that that’s about all you hear. Now the water is moving through your walls through these pipes …
LW: Vertically. The heat distribution system has headers top and bottom, the top one with a slight slope, and then it’s dropping straight down through the finned copper tube. You know it’s designed as a solar-panel absorber. Absorbers work backward just as well as they work forward. I’m emitting heat instead of absorbing it. There are many things that I’ve used in this house in a manner different than was intended by the manufacturer. In this case, I’m using solar collector finned tube to emit rather than absorb. That is, I put heat into the room from water in the tube, rather than putting the sun’s heat into water in the tube. In conventional radiant heating, heat transfer plates are often used under a wood subfloor to conduct heat from the tubing to the floor in a way very similar to what I’m doing. Also, they’re mounted on the surface of the SIPs, with Sheetrock between them just holding the edge of the copper down. They’re 3/8-inch tubes wrapped in copper fin sheet that’s crimped on. So with the Sheetrock up to the edges of the fins and then mudded over, the copper is maybe 1/16-inch inch below the surface. So if you need to start from scratch, if you need to get your radiant system up to speed quickly, it can do it. It doesn’t have to heat a bunch of anything to get the heat into the room. If you’re buried in a 4-inch concrete slab, you’ve got to heat the whole slab up.
LJ: Is the tank at atmospheric pressure? If so, how long does it take for the alcohol in the vodka to evaporate to the point where you need to add more?
LW: The tank is at atmospheric pressure, but the loop with the vodka is a closed loop, using a large soft-copper heat exchanger in the bottom of the tank. Plain water from the tank does evaporate, and it would be nice if I checked it yearly. It’s gone two years, though.
LJ: Does it smell like booze?
LW: I made automatic venting for the solar collectors, so that if they get too hot, that hot air is vented out of the box, preventing the fluid from overheating. Before I did that, the oak tree by the house twice got sprayed with a nice hot vodka mix. Otherwise, (sadly?) it is a sealed system, with no odor.
LJ: And so that board … that Sheetrock is your thermal mass—is that the case? Or rather, the water is your thermal mass.
LW: I like to think of the water as the thermal mass. The Sheetrock is just standard Sheetrock, 5/8-inch in some places, but actually not anywhere there’s fins. I didn’t build mass. I kept it a lightweight house. I figured the mass is in the tank: 1,000 gallons and the mass moves to wherever it’s needed to give heat.
Oh, sizingwise, too, I have 3 gallons of storage per square foot of collector area. That’s way too much. You’re supposed to have 1½ gallons. Maybe even 1. So I’ve got 3 gallons because I wanted the low- temperature Btu. Because I designed it to use low-grade heat to keep the house warm. That’s backward from what most people would do: They use a smaller volume of much hotter water. But by having the greater volume, it makes the solar panels work better, because they’re operating cooler. So there are synergies. Linda Wigington of Affordable Comfort likes to talk about “house as a system.” Everything affects everything else. It does.
LJ: It’s all coexisting. The nature around your house, as well as you and Suzanne, the wind and the sun.
LW: Well, I try to get in the sun’s way. It’s a north-facing solar house, speaking of that.
…and speaking of that, we will continue with Larry’s findings about passive solar design—and other things—in our next issue. Stay tuned.
For more information:
Holohan, Dan. The Lost Art of Steam Heating. Bethpage New York. 1992. Dan Holohan Associates Incorporated.
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)