Home » Blog » Personal » A project to add to your “not to do” list

A project to add to your “not to do” list

I like building stuff.

I spend much of my life doing “work” that’s decidedly cerebral – writing articles, papers, blog posts, computer code or emails; talking with folks face to face or over the phone. When I get the chance to take some time off, I like to build things, preferably out of wood, preferably things that have a practical purpose. It’s not a surprise then that my December vacation included a fireplace mantel for my sister and her fiancée and a new, improved ger.

My favorite person to build with is my friend Nate. Like me, he spends too much time staring at computer screens. (In his case, he’s trying to make computers do things they’ve never done before, whereas most of my computer wrangling is, at root, interpersonal…) Like me, he’s good with his hands and likes making stuff that works. Like me, he’s a little crazy.

To accompany our cosy, warm ger, Nate thought we needed an outdoor hot tub. Specifically, Nate was interested in building a hot tub that could be constructed almost anywhere, with minimal materials investment, so that a hot tub could be part of any rural outdoor gathering in the future. It’s a worthy goal, and both easier and harder than it looks.

(To be very clear – I had basically no design input on this project. My role was to prevent Nate from burning down my house or woodshop, and to document his success and failure. I claim no credit or blame for the solution he pursued, and I strongly encourage you NOT to try repeating our experiment without some major modifications, for reasons that will become quite clear.)

The first step in building a hot tub is building a vessel that can hold a few thousand liters of water. Nate did this by using a length of stone retaining wall in my back yard, and building three other walls from plywood. The sheets of plywood are held in place with cheap and massive stuff – in our case, bales of hay and stacked cordwood, which is what we had handy. Used tires would also be great, as would auto parts.


The box was lined with a large (4m x 4m) green tarp (for those of you not familiar with tarp, the building material of choice here in New England, green tarp is the “intermediate” grade of tarp between the cheap blue tarp and the luxurious and strong silver tarp.) The tarp held water surprisingly well, but not well enough, so Nate relined the tarp with 4mil vapor barrier. A 6m x 7m sheet of plastic vapor barrier cost $20, the only money we actually spent on the project.

Filling the enclosure with water, we now had a very large cold tub – the water comes from my well, and comes out of the ground at about 5C, a temperature not very conducive to relaxation. The goal was to heat the water in the tub with firewood, hoping to bring the temperature close to a comfortable 40C, perfect for a relaxing soak in the snow.

Nate commandeered an old ash can, attached two thick blocks of 4×4 to the bottom, borrowed about 70kg of weight plates from my gym and used them to sink the can to the bottom of the tub. Without wood in the “stove”, it had neutral buoyancy. A few kilos of cordwood and it sunk securely to the bottom. A little lighter fluid later, and we had a (dysfunctional, wholly inadequate) pool heater.

Starting a large wood fire in an ash can and leaving it unattended in a tub of water seems like a dumb thing to do, on initial consideration. On the other hand, if the pool walls gave way or the can tipped, the fire would likely be doused by the 2000 liters of water in the pool or by the several cm of wet snow on the ground. At the very least, I figured the house – 20m away – was probably safe.

It also seems like a poor idea to climb into a pool of water that’s also otherwise occupied by a hot metal barrel filled with flames. While that’s true, it’s not true for the reasons you think it is. If you brush a limb against the can under water, it doesn’t feel warm at all – it’s transferring heat to the water very quickly, so the metal isn’t warm enough to burn you, unless you jam your hand against it and hold it there, preventing the metal from transferring heat to the water. (The can above the water line is another matter entirely. You don’t want to touch that.)

No, the reasons it’s dumb to get into a pool of water heated by an open woodfire is smoke. Wood fires release a lot of wood smoke, and that smoke gravitates directly to your eyes as you’re attempting to relax in the water, making said relaxation harder than you’d think. Future designs of a stove will require us to add a chimney that pulls the smoke away, and captures more of the heat to warm the water, instead of warming the air above the fire.


Ah yes, the heat. After four hours of a roaring wood fire, the temperature in our tub had warmed up to a balmy 11C. This isn’t quite the failure that it sounds like – after all, the water began at 5C, the air temperature was -5C and it was snowing. Somewhere in the heating process, Nate raided my shop for excess extruded foam insulation, hoping to hold more of the heat in the water. It didn’t work, but the pink insulation adds some nice color to the photos, I think.

(That’s Nate pictured above, enjoying the balmy water. Yes, photos of me in the tub exist. No, I’m not posting them. Consider yourself lucky.)

A future outdoor hot tub would have some major design changes. It would be much, much smaller, giving us less water to heat. It would have a thorough insulation layer on top. (A thick layer of ping-pong balls would be especially fun, as you could get into the tub, displacing the balls, and keep it insulated even as you soaked in it.) It would have a radically redesigned stove, likely an ash can with two chimneys, one for air intake, another for smoke output.

And it would probably be built some warm summer evening, not during a freaking snow storm. As should be clear from this post, we’re idiots.

It’s been helpfully pointed out by one of the readers of BoingBoing that this hot tub plan is “A Bad Idea”. That was part of the point of this post – sorry if that wasn’t obvious from phrases like “we’re idiots” and “I strongly encourage you NOT to try this experiment”. Let me be very explicit: Don’t do this.

50 thoughts on “A project to add to your “not to do” list”

  1. The New York City Math Teacher

    Suggestion – form, uh, twenty feet of copper or galvanized tubing in a helix, dimensions restricted to those of the barrel, soldering one end out a hole at the bottom of the heater barrel connected to a plastic feeder tube. Attach the other end of the heater coil to a flexible hose connected to an aquarium pump or pool pump to more flexible hose and then finally to an l-shaped siphon pipe fed into the bottom of the tub.

    With proper drafting of the heater barrel and a nice, not-too-hot wood fire, you should be able to heat your makeshift hot tub much more efficiently.

  2. Ah, yes a wonderful idea poorly exicuted the sum of many of my indevers as well. Things i would have done diffrent, first put the pink foam on all sides underthe tarp for insulation and extra softness, maby having an old cast iron stove altered for water proof and an ambilical corn for air intake maby with a fan at the far end for extra heating power, maybe a depression in the bottom middle for feet maby maby the stove if big enough, making a small frame to put the roof of the ger on top, an idea for doing with out the stove is build a fire outside of it all together and putting rocks in the fire to heat then scooping them out into the barrel and putting that in the water but bewarned if you use river rocks they may burst and even explode inside the fire and some will crack with u put them into the water i would suggest experimentation on a smaller scale. i have a friend who loves this idea he and i will try it in the spring with some other ideas, will send pictures

  3. Thanks for the suggestions, folks. Just to be clear – part of the design constraints for this hot tub was spending as little money as possible. With that in mind, NYC Math Teacher’s solution would be allowable, while purchasing a ready-made wood heater would not be. An added complication – we ordered a cheap arc welder over ebay and, well, it hasn’t shown up yet. A little welding and we would have, at minimum, produced a sealed ashcan with two chimneys. Who knows – if it shows up in the next 48 hours, we still may… :-)

    Keep the good ideas coming…

  4. You write:

    “Without wood in the ‘stove’, it had neutral buoyancy. A few kilos of cordwood and it sunk securely to the bottom.”

    A friend of mine was badly burned a few years ago, when the “stove” in her friends’ similar DIY hot tub tipped over onto her. What started as a fun night on the beach with friends turned into months of painful physical therapy while her skin healed. She was probably lucky it wasn’t worse.

    I question how “secure” it is to keep a flaming barrel within arm’s reach that is only a little weight away from floating up and tipping onto you. Water is heavier than one might guess, and waves pushing on largish objects exert more lateral pressure than one might guess.

    Just eyeballing the photo of Nate in the tub, I’m going to guess that the water displacement when he got in stole around 10 pounds of downward force, perhaps making the barrel a lot closer to neutral buoyancy than you’d hoped. In a smaller tub with more than one person, that effect would be greater. If that barrel is 50 cm in diameter with 10 cm sticking above the waterline, that’s a potential of over 40 pounds of additional buoyancy it could suddenly experience when people get in (before water pours over the lip).

    While I appreciate the need for scientific experimentation, I would strongly encourage people not to try this themselves.

  5. Maybe a double wall of plywood with a few cans of foam insulation squirted in between the layers… you’d probably have much heat loss through the walls. Oh Nate, one more thing: bathing suit. Please.

  6. wow you guys are like my heros now…

    ok, ok i have an idea. instead of a bucket why not cut out a side of the pool like so

  7. “Then figure out a way to pump water past a metal case with that flame heating it, and back into the tub.”
    this is a good idea aswell, but the way your thinking (and this is a guess) costs a lot. instead i would use the powers of coduction to move the water for me.

  8. augh noo my pool of text
    damn well anyway
    cut out a small side and build a frame to support a small flame pit to give the wood more surface to heat.
    also use old ducts to get the smoke out of your eyes.

    crappy text dealies…

    i figured it out… the greater or equal sign gets rid of anything you say after it because its html… what the fuck…

  9. Pingback: the Tao of David » Blog Archive » Build your own hot tub

  10. Jamie – your concerns are totally legitimate ones. We never got even close to a place where we were having multiple people in the hot tub at the same time. The water rise from one person semi-submerged in the tub was not enough to make the barrel float, and we’d worked out a system to anchor the barrel should we ever have reached the point where we actually planned on using the tub. Idiocy aside, we’re (marginally) smart enough not to get into a tub with a hot floating mass. I posted these photos as a failed experiment, not as a plan to build your own hot tub.

  11. An old Hippie standard “hot tub”:
    find a discarded cast iron tub.
    Put up on blocks.
    Fill with water.
    Build fire underneath it.
    Put something in bottom, up on bricks to keep your butt off the hot iron, like a window screen.

  12. Random thought of a physicist/boingboing reader:

    You’re relying on conduction for heat transport through your tub. That is very inefficient. Also, most of your thermal energy is heating the sky due to the conduction current you have set up in the air above the can. This is bad design. Conduction should be your friend, not your enemy. You should run pipes in at the bottom of the can and out at the top (weld carefully to avoid leaks). Much of the heat that is currently escaping into the sky will instead heat the small amount of water in those pipes, setting up a convection current that will draw cold water in at the bottom and spit hot water out at the top. Note that both ends of the pipes need to be underwater unless one of you plans to donate his lips to start a siphon. I’m guessing that would be Nate.

    If he’s up to it, another possibility would be to switch from wood to charcoal and build a primitive blast furnace on the ground outside the tub (http://www.davistownmuseum.org/toolPreBlastFurnace.html). Instead of iron ore, you have water pipes running through your furnace. Someone has to start the siphon but once running it should go on its own since your entrance and exit pipes will be at the same level and convection will be pushing it along.

    Sounds like a good plan. Let me know how it turns out.

  13. “I spend much of my life doing “work” that’s decidedly cerebral”

    Heh, you could have fooled me! I guess you never took a physics classs.

  14. Pingback: imagesafari clip blog » Blog Archive » …My heart’s in Accra » A project to add to your “not to do” list

  15. Pingback: …My heart’s in Accra » Sorry about that.

  16. I really liked the simplicity of this project. The tub itself is, IMHO, a nice achievement, before even considering the heating scheme.
    I think the heating was also a good attempt, considering you didn’t use a pump. I wonder if the pump-less convection plan outlined in comment 23 would work.

  17. Thanks for the comments! Many of the suggestions about convection, conduction and air flow were things we intended to do, but as the stick welder had not (and still hasn’t) arrived, we were pretty limited in our ability to build anything leakproof. And we were consciously avoiding anything requiring an electrically powered circulation pump, both to avoid any risk of electrocution as well as to try to come up with a design that could be used in a location without electrical power.

    Ethan summarized the lessons learned pretty well. The plywood boards backed by strawbales worked very well as a container. The green tarp probably would have worked but with a slow leak, but the 4-mil plastic liner was completely waterproof and probably sturdy enough. We would have gone with 6-mil, but didn’t find a sheet large at Home Depot. We were expecting the low heat efficiency of the open barrel, but I was surprised a bit by the heat stratification of the water. Lacking a circulation pump (and essentially heating from near the top of the water), the hot water ended up on the surface and lost heat to the air very quickly. Better mixing, a smaller top surface, or at least a better cover would have helped a lot. My guess is that even uninsulated, losses to the sides and bottom were a lot less than losses to open top. A stove design (like the Snorkel) that has the burning chamber fully underwater would probably give an order of magnitude increase in heat transfer efficiency, and might still be tried if the welder arrives.

    I don’t think the safety of this design was quite as bad as it might appear. Because the iron weights were at the bottom of the can, even if floating it would have been very difficult to tip it. I tested this a bit before putting the extra weight of the wood in. And as Ethan mentioned, the underwater portion of the can was cool to the touch, even to the point that I easily grabbed it with both hands to check how heavy in the water it was sitting. And even if tipped, I don’t think it would have been that dangerous. I think the level of danger was pretty close to that of a bonfire at a party: obviously something to be treated with respect, not a good mix for small children or stumbling drunks, but a fairly manageable risk. As this was a prototype used only by sober adults in a pretty well controlled situation, it felt pretty safe. Still, details of the accident described above would be appreciated if there are risks that are harder to see.


  18. Oh, go ahead and spend the $200 on a Japanese wood-fired tub heater. The heater sits outside the tub, greatly reducing the risk of horribly maiming your tubbers, and relies on convection to circulate water through the heating coils. Good science, aesthetically implemented, is always worth spending cash on.

  19. i just thought of something. why dont you use rubber or that bathroom undercovering that you use when you put down tile?
    its really sturdy and water proof so it would be perfect ofr this application. the only thing you would need is silicon for the seams and a shell around it so the water doesnt break those seams. no water leakage!

  20. Everyone who has been taking this post seriously, seriously needs to stop. Idiots! Someone trys to do so much with so little for a bit of winter fun and the “geniuses” all attack with “you should have used a point-three-oh insulation barrier and a thermal circulating heat pump to raise your efficiency about seven-hundred-fifty-nine percent. Then you would be heating to about 55C, with room for 15C of loss to air in that weather, snork!” get over it. Now!

  21. I like what pjcamp was saying. But I think you should start with the can, and make it a tall can with a removable chimmey. Without the chimmey I think the can should be maybe 3 feet (about 1 meter), above the water line and about 1 to 2 meters below the bottom of the tub. So yes, the tub would have to be supported up off the ground with the can going right through it. Then the can should also have vents at the bottom, I’d say as many as you could have with it still being able to support the can. Also it should have some type of a fire resistant enough series of grates to gradually let the coals fall to about where the bottom of the tub is but then when they get too small they would fall to the ground and keep the fire shifting which will keep air moving through it. Then if you could get something that would move the water around. You want as much water circulating to the can and away from the can as possible. more? no.

  22. A small but vehement radioactive source would allow you to stay in the tub without any need to gather more wood, and it would be smokeless. I suggest you steal one from one of those automated lighthouses they have in Finland. Also, if you use old engine oil instead of water, you can get the tub much, much hotter before it starts boiling.

  23. I like pjcamp’s idea of passing metal water-carrying tubes over the flame and using the convection currents that would result to draw up cold water from the bottom of the tub. This is the way certain backpacker stoves draw fuel from a bottle. At its simplest, this system would only require bending some thin metal pipes and securing them to the top of the can while keeping their ends underwater, with the added advantage of decreasing the can’s bouyancy.

  24. Pingback: …My heart’s in Accra » Winter in Western Massachusetts

  25. What about securing the metal tub to the bottom of the spa, and welding an exhaust pipe sufficiently high to keep the smoke away? Essentially it would be like dropping a wood burning stove into the spa, stovepipe and all…

  26. I think yous guys got it backwerd. Round here we try to get the water colder…wait till January…..swim in moving rivers….Recently i came up with a real gud one. Just fill a galvy tub wit half RV antifreeze and half water, doesn’t freez as quick.. Enter at yer own rdisk…

  27. Your attempt at a no-cost wood-fired hot tub gave me a good chuckle because I’ve been there, done that, and succeeded (eventually) in getting a good soak. You could do it right straight off the bat for less than $200 like I did (eventually), but that wouldn’t be as interesting as all this would it. Skip to the end for the short how-to story.

    Since getting a free cedar hot tub from a friend (it leaked, lacked a heater and he was going to make a planter out of it) I spent a lot of money trying to build a no-cost wood-fired hot tub. If my wife asks how much I don’t remember. My current and very nice version can be replicated for less than $200, but getting to this point cost a lot more. I tried everything from the trash can heater to copper coils in my house woodstove. The best wood-fired heating solution was a simplified snorkel stove design I paid the local metal shop $400 to put together for me out of 1/8″ aluminum. I had them include tabs so I could screw it to the wall of the tub and a 4″ pipe going through the firebox to capture extra heat. I made an internal baffle and an adjustable air intake/lid out of old sheet metal. It worked wonderfully and we enjoyed it for 4 years, but ultimately it took up too much room in the tub for us and 4 kids so I sold it for about what I had spent on it. Because the tub leak kept getting worse, I finaly bought a vinyl liner for the tub and installed it, first lining the inside and bottom of the tub with about 12 layers of the 1/4 inch pink foam sheeting usually installed under vinyl siding. My new and favorite heater is a 300 watt water bed heater ($39 with thermostat) installed between the liner and foam. With a good insulating cover my hot tub costs about $8 a month to keep heated at about 102 degrees. (100 degrees is the top of the thermostat, but insulating the sensor slightly bought me a little extra heat I know that somebody with any electrical sense could tweak the thermostat to get the 102 to 108 degrees most hot tubbers like). There’s no plumbing to freeze, no wood to carry, plenty of room, and it works. Naturally, the extension cord is plugged into a GFI circuit, and the 1000 watt stock tank heater we use for a boost when we want the extra heat is ALWAYS unplugged before we even open the tub.

    If you want to do this on the cheap tub and all, 20′ of 4′ stock fencing tied in a circle gives you a 6′ tub (a 4′ tub takes a lot less water and heat, and the cover is easier to make, but the tub holds fewer friends). Lined with several layers of 1/4″ foam and a 6 mil plastic drop cloth you have a tub, though pond liner will last longer. One or two waterbed heaters on the bottom between the liner and the insulation will provide a nice even heat. Change the water occasionally or shock it with pool chemicals or you’ll get sick. Bacteria like 100 degree water.
    If $200 is too much, try the book by Becky Bee that describes how to make a wood-fired hot tub out of an old bathtub and mud. I’m not kidding. It works.

  28. Great stuff everyone, plenty of ideas put into the mixer for me to ponder over! im thinking of using an old domestic water convection radiator to try and heat a tub! placed at an angle above the fire it will allow venting, but also allow for the convection of the water (in through bottom out through top). Im thinking standard radiators may be steel (judging by the gunk that comes out from draining them) but maybe an old cast iron one would be more effective. I have even thought of putting an enclosed system of 2 radiators (oil filled), one in the fire and one in the pool! Have fond memorys of playing stupid games of who could hold their hand longest on the radiators in school that were belting out heat!
    Any thoughts? Dont know if you use simular systems in the US or not for house heating.
    Havent done any of those calculations tho, so gonna be another suck it and see, trial and error efforts.

  29. Chris, any chance of your sharing (or selling) a set of your plans for the Snorkel type heater? I’m in Austria, and I’m dumbfounded by the price of snorkel type heaters over here. (On the bright side, I found a 7 foot oak wine barrel that serves as the perfect tub.)

  30. I applaud your ingenuity and your willingness to see no task too daunting! What else are you to do with your long winters but find uses for previously unrelated materials that are lying around? Bravo.

    My husband wants a ger. Seems worthless in Texas where the winters never reach under 30 degrees. We would end up with a sauna.

  31. Thanks, Katherine. Actually, a ger would be hugely useful in Texas – they ventilate very well, and some Mongolians use them very close to the Gobi desert. With a smokehole at the top and rolling up the walls at the bottom edges, you get a natural thermal current… and you’d get great shade.

    There was a good article in the New York Times on Friday about a family who built a ger in central New Mexico. Might be an inspiration… :-)

  32. Pingback: …My heart’s in Accra » How we amuse ourselves in Western Massachusetts

  33. Help with our Soft Tub hot tub.
    It has a heavy type foam inside.
    We need a new liner and they are about $400.00.
    Is there a way we can put in a vapor barrier type material and use the jets and heat.
    We have the pump and motor, it all works.
    Just need to find a way to have a liner in it that does not cost too much.
    Sue and Terry

  34. Pingback: …My heart’s in Accra » It’s my party, and I’ll fry if I want to

  35. Really well written article about a really dumb idea, and to cap it all Nate was the guinea pig. Do tell if you actually climbed in as well.

  36. Wow! What an interesting idea, however like ‘baz’ said, a little dumb. I actually stumbled across this post after google suggested it to me when I was searching for my own blog. I actually wrote a post about a project that i would not add to your list, which was a post about backyard designs go figure, as this was taking place in your backyard so it seems.

  37. Cudos for originality and courage. You mention a future stove with two chimneys. You could slightly modify your existing ash can to be essentially my all-time favorite wood stove (from a house-boat in Kashmir): Securely attach a lid that fits the ash can (after making modifications:). Add three openings, with flanges: 1. the outgoing smoke chimney, 2. the wood input hole with lid, 3. the air intake tube that goes straight down to near the bottom of the ash-can, with a top adjustable cover. This third one is really important. It heats the incoming air as it is drawn down the tube by the heat rising in the longer chimney (it has a slightly smaller diameter than the chimney), and provides draft oxygen to the bottom of the fire. It roars when the fire is really going. Narrow sticks can also be dropped down it, adding heat to logs that are getting started. When made of sheet metal it burns out after a season, so I’d use a thick pipe screwed into a flange bolted to the lid. (The rest of the parts in contact with the water don’t get hot enough to burn out.) With this minor addition, I think your heater will be 4x faster, with no smoke. If the lid is galvanized, burn off the zinc before sitting near it (not good to inhale the zinc).

Comments are closed.