Introduction: Wood Fired Hot Tub

Picture of Wood Fired Hot Tub

Wood-fired Hot Tub
Perfect for 6-10 friends.  It can be carried in the back of a small pickup truck.  It takes about 3 hours to get up to temperature and cost about $650.

-55 gallon closed steel barrel ($30)
-Barrel stove conversion kit ($50)
-14 2" x 4" x 8" firebricks ($30)
-2 2 foot sections of 6 inch black stove pipe ($13 each)
-50 feet of 1/2 inch flexible copper pipe ($80)
-10 feet of 1/2 inch rigid copper pipe ($15)
-10 1/2 inch straight pipe fittings ($5)
-10 1/2 inch elbow pipe fittings ($5)
-plumbing solder and flux ($5)
-Rubbermaid 300 gallon stock tank ($300)
-1/2 horsepower centrifugal water pump ($40)
-10 foot garden hose ($15)
-1 foot 3/4" CPVC pipe ($5)
-assorted fittings to connect tub to pump and stove with garden hoses ($50)
-ground fault switch to turn on pump ($15)

-jigsaw with metal blade for cutting holes in barrel
-hacksaw for cutting pipe
-slip-jaw wrench
-drill with sharp drill bits
-propane torch for soldering pipes
-shop vacuum with blower attachment (optional)
-socket set

Step 1: Build Barrel Stove

Picture of Build Barrel Stove

Get a 55 gallon steel barrel with a closed top.  I got mine from a guy in Bradford, VT who calls himself the "Barrel Man," but if you don't live in the area, you can usually find them on craigslist.  Make sure you get a barrel that hasn't been holding anything toxic, since you'll have to spend a fair amount of time crawling around the inside during the assembly process.  This barrel formerly held palm oil, and there was a small puddle of it on the inside (it had the consistency of crisco) and every once in a while I accidentally stuck an elbow in it.

I bought a Vogelzang Barrel Stove Kit, Model# BK100E from  Assemble the stove following the directions in the kit.  The kit comes with the door, the chimney collar and the legs.  The stovepipe is not included in the kit.  I bought mine at Home Depot.

When cutting the holes for the chimney and collar, make sure you use a blade designed for cutting sheet metal, and when drilling holes, make sure you have a sharp drill bit, otherwise it will take forever to drill your holes.

After assembling the kit, I added firebricks to prolong the life of the bottom of the barrel.  I used 2" x 4" x 8" bricks there were 3 rows of 4 and a final row of 2 for a total of 14.

Step 2: Coil Copper Tubing for Heat Exchanger

Picture of Coil Copper Tubing for Heat Exchanger

The copper tubing comes from the store as a flat coil.  First straighten it out by hand.  Then crimp one end of the tubing with a pair of pliers.  Fill with fine dry sand.  The tube should be held almost vertically when filling with sand or the tube may be filled unevenly.  Once the tube is full of sand, crimp the other end.  The longest length of tubing I could hold vertically and still reach the top to fill was about 10 feet, so I cut the tubing into 10-foot lengths and filled each one separately.

The tubing can now be bent into a coil.  I found a piece of 4 inch pipe to wrap the tubing around.  There were a few sections where the sand had not completely filled the tube and it kinked.  I cut these sections out.

After bending the tubing into a coil, you have to get rid of the sand.  Cut the crimped ends off with a hacksaw.  Then shake and tap the coil to remove the sand.  Getting the sand out of the coils took a lot of time.  After most of the sand was out, I used compressed air to remove any last remnants.

5 short coils were soldered together to make the large coil.  In order for the coils to fit together nicely, make sure they're all coiled in the same direction.

Step 3: Put the Heat Exchanger on the Stove

Picture of Put the Heat Exchanger on the Stove

The heat exchanger was inserted into the chimney (2 sections of 6 inch stove pipe).  I drilled a hole in the bottom for the tubing to come out.  Two pieces of angle iron were used to reinforce the tubing and the chimney, along with a lot of bolts (probably more than necessary).

The heat exchanger was plumbed to the bottom of the barrel to keep the hoses low to the ground (don't want anyone tripping over them) and also to keep them away from the heat of the stove.  The copper tubing was terminated with a female hose fitting.  In the photos you can see a short section of 3/4 inch iron pipe, but that was not necessary.

Finally, we added handles by connecting short sections of galvinized pipe (electrical conduit, I think) to the barrel with wire rope and bolts.

Step 4: Connect the Heater to the Tub

Picture of Connect the Heater to the Tub

The tub is a 300 gallon Rubbermaid stock tank.  It has a 1 1/4 inch threaded opening at the bottom.  Using a series of adapters, I was able to connect this to 3/4" CPVC pipe.  Next, there's a union fitting.  This allows the pump to be easily connected and disconnected.  Then I added a ball valve.  The ball valve was basically unnecessary.  Then I added another adapter to bring the 3/4" CPVC up to the 1" threaded diameter of the pump intake.

The pump output is also 1".  It's connected with regular PVC (CPVC probably would have been a better choice, but it's a short enough section that heat flexibility isn't a problem) to an elbow.  The elbow is important because if you try to connect the garden hose to a vertical fitting, it will kink.  The elbow has a male hose fitting on the end.

To control the pump I use an inline portable ground-fault circuit interruptor.  The "test" and "reset" buttons can be used to turn the pump on and off.  Side benefits of added safety from electrocution.

Step 5: Transport to Location

Picture of Transport to Location

Pile everything in the truck.  Strap the tub down to keep it from blowing away.

Step 6: Fire It Up and Enjoy

Picture of Fire It Up and Enjoy

When we fired it up, we decided to attach the exhaust of a shop vac to the barrel to get the fire going a bit faster.  It worked great.  In fact, it caused the end of the barrel to glow.  With the stove closed up, we were able to adjust the temperature of the tub by turning the shop vac on to increase the temperature in the tub and turn it off when the tub was warm enough.

Step 7: Analyze Performance

Picture of Analyze Performance

I used a digital logging thermometer to measure the temperature of the tub over the course of the evening.  There are a few missing data points, I think this might have happened when the logger got splashed.

Time 0 was at 4:30 pm when we lit the fire in the barrel and started filling the tub.  The outside temperature was about 65F and the water from the hose was 45-50F (I didn't get the thermometer until 2 hours in.) After 3 hours the tub had reached a comfortable 105F.  The heating was greatly accelerated by using the shop vac blower.  When the blower was turned on you could feel the water in the hose get warmer.  The stove heated it at about 20 degrees F per hour.

The next 5 hours (hour 3 to hour 8) was spent lounging in the tub.  The temperature was controlled by turning the shop vac blower on and off.  During this time the temperature fluctuated between about 100F and 108F.

At hour 7.5 the tub had 10 people.  When 8 of them left we needed to fill it up with more water.  I think the temperature drop from time 7.50 to 7.55 was a result of the "cold hose" being pointed at the thermometer.  Later on the temperature jumps to 130F, which is probably from the "hot hose" coming out of the stove.  I was in the tub at the time and it was quite comfortable despite local variations in temperature.

The stove was turned off after 9 hours and the tub cooled down at a rate of about 7 degrees per hour for the next 5 hours.

Doing a bit of math, we can see that the heater was generating about 13.4 kW, the tub was losing about 3.4 kW and the remaining 10 kW is what was getting used to heat the water.  To put it another way, we lost 1/4 of our energy from cooling, which suggests that the tub might benefit from an insulated cover.

I'd estimate we used about 1/12 of a cord of wood, which works out to about $17 (for wood at $200/cord).

Step 8: Mistakes to Avoid

Picture of Mistakes to Avoid

Always keep water circulating through the copper coil.  Otherwise the solder joints will melt and start to leak.  Started the fire about 10 minutes before turning on the water, and the coil started leaking so much that it put out the fire.  To fix the problem, I had to disassemble the chimney, take the copper coil out, clean off the solder joints (all 5 of them were leaking), resolder and reassemble the stove.  I'd estimate the procedure took about 4 hours.  It occurred to me that brazing might have been a better choice for joining the tubing sections instead of soldering.


Steinzel (author)2017-05-01

You wear clothes in the hottub? LOL

Very nice instructable!

ntense99 (author)2014-10-15

Dude! I am building an outdoor wood fired furnace using the same barrel stove - your creative solution to stabilize the chimney are exactly what i needed :) I plan to add on even more chimney height!

I too am going to put a 3/4 inch copper coil (around 100 ft or so) inside the tank and then send that into my house. Thanks!

dan268 (author)ntense992014-10-15

Wow, looks great. Making the copper coil is a huge pain. How are you planning to do it?

There are lots of suggestions on the internet about filling the tubing with sand. I found that it was very difficult to get the sand in, and very difficult to get it back out. There were lots of gaps where the sand hadn't completely filled the tubing, and this would cause the tubing to kink (which later had to be cut out). Getting the sand out took a few hours of banging and rotating. So my recommendation is to find something other than sand to put in your tubing when you're trying to bend it. Maybe wax?

Let me know how things go.


ichibanshark (author)dan2682017-01-02

Purchase a couple cans of 140 or 210 or 220 melt. Heat, pour in , let cool, maybe a half minute , bend that section, rotate slightly, heat THROUGH the copper letting it flow a short distance, remove heat, bend and repeat. Infinitely reuseable

quadracer (author)dan2682015-02-15

Use water and freeze it ;)

ntense99 (author)dan2682014-12-09

Hey Dan, just finished uploading a bunch of stuff about this - The majority of the content I uploaded is on my YT channel. Here is a high-level instructable I made:

The wood furnace is up and running nicely! Thanks man!

dan268 (author)ntense992014-12-09

Looks great. Do you have any problems with creosote building up on the pipes? The last time I fired up my hot tub, it took a lot longer to heat up, and I think the creosote build-up may be the reason. I'm planning to try to clean things out a bit before next time.

ntense99 (author)dan2682014-12-09

Ohh hell yeah! I have Creosote galore! It builds up fast too! You know, I just leave it. I notice that when I run the appliance in overdrive, it tends to burn off any creosote. That's how I take care of it. However, there are parts of the appliance that "should be" cleaned - like my top-down batch feeder. Gassified wood that does not combust sticks to the inner wall, creating a nice crispity, crackly creosote egg-shell! Well, if your appliance has a thick build up of anything (even installing the fire bricks), it will definitely add to the operating temp timing.

aduraja (author)dan2682014-12-02

Hi Dan,when you first get your copper tube ,it come in a big loops, what you can do is fill the tube with water and put it in a coffin freezer (if you have one)once the the water is frozen , you could use any size OD pipe to shape the copper tube as you wish,my two cents :-)....

dan268 (author)2016-12-13

About 2 feet. Detailed description here.

JulianK17 (author)2016-12-13

How deep is the tub?

DougM43 (author)2015-12-30

Something not yet mentioned, which I'm pretty sure you thought of, is to use a counter current exchanger design. This means you'd want the water to flow through the copper pipe in a downward direction through the chimney, against the flow of the smoke. If you had the water flowing up, the water and the smoke would quickly match temperatures, and the upper portion of the exchanger becomes much less effective. Alternatively, in the counter current design, you use the cooler smoke at the top of the chimney to heat the cooler water, and by the time the water gets to the bottom, you'll have the hotter smoke to heat the hotter water. A temperature differential is maintained through the entire exchanger coil this way, which is important as I'm sure you know. I learned about this in an ecology class. Fish use a counter current design to oxygenate their blood, and I'm sure it's a common engineering principle.

dan268 (author)DougM432015-12-30

That's an interesting idea and one that I thought of, but never systematically tested. I've set up the heat exchanger in both orientations (just depends which hose I connect to which side of the pump), but haven't noticed anything dramatic.

Here are my thoughts on the theory. The temperature difference between the water going in and the water going out was at most 10-20 degrees F. When I really got things cranking on the stove, it would start to glow red, suggesting that the temperature was around 1000-2000 degrees F (the chimney wasn't glowing, so it probably wasn't that hot, but might have been close). Heat exchange is determined by the temperature difference between the pipe and the exhaust gas, as you pointed out. So if we consider an example where the water in is at 90F, the water out is at 100F, and the chimney exhaust gas is at 1000F, the temperature difference changes from 900F to 910F, which is about a 1% difference.

A much more serious problem that I've noticed is creosote buildup. The efficiency of the copper coil heat exchanger has been getting much worse over time. I think the last time I fired up the tub I was only seeing a 5 degree difference between the inlet and outlet water, so it may be time for a re-design.

DougM43 (author)dan2682015-12-30

I agree, the change in water temp is low. However, I'm betting the change in flue gas temp is high, especially with all that water cooling it down. There's not a lot of energy per unit volume in gases, so it looses temp fast. With the fire cranking and the water cold, I bet the output at the top of the chimney isn't hot enough to burn you. All in all, I agree that the difference in design probably isn't much in this case, but I bet the counter current design works at least a little better.

The rapidly cooling flue gas is part of the reason you're getting more creosote than you would in a regular chimney. Creosote forms when the water created by combustion cools to the point that it condenses on the walls of the chimney (or in our case, the copper coils), and then solidifies into creosote. The fixes for this in normal wood stoves are 2:

1) completely combust the fuel to where there is no unburned carbon left to form the deposits. This one is tricky. Stove design is important, where you want to use either secondary combustion or a catalytic converter to get complete combustion. This is a complicated and expensive way to fix the problem. Also important here is using dry wood. Wet wood adds water to the flue gas, and slows combustion to create cooler gases with much higher concentrations of unburned carbon. Wood from firewood dealers is rarely dry, even if they say it is. 20% water by weight is good, 10% is better. Find this by taking a small piece, weight it wet, and then put it in a 210° oven for 1 day and weight it again to get the dry weight. The equation is 100*(wet-dry)/dry. If you're using even somewhat wet wood, this is the bulk of your problem.

2) Keep your flue gas hot. This is done in two ways: burn a hot fire, and insulate your chimney. Contrary to popular opinion, insulated chimney pipe (aside from where it passes through the ceiling/wall), is not a safety feature for the reason most think it is. It's sole purpose, especially on the exterior chimney section, is to keep the interior flue wall temperature high to reduce condensation and creosote buildup. In your case, insulating the chimney might help, but it only solves half the problem. The water in the pipe is doing most of the cooling, and we can't insulate that.

My guess is you're using wet wood if you're needing to use a shop vac to get hot temps.

The flue is a convenient place to put the coil, but difficult to make effective because of the creosote. It's also the coldest part of the stove. You'll always struggle with creosote with this design. I think if you used copper pipes running across the top of the inside of the fire box, it would be more effective and less of a headache. You could use elbow fittings on the outside to send the pipe back and forth without having connections to fail inside the stove. I think this is the way I would do it if I were limited on resources.

I'll make one of these someday. When I do, I plan to use a boiler design with a water jacket. I'll use 3/16" or 1/4" steel plate to weld an internal fire box which will be surrounded by a chamber (water jacket) filled with water. The hot tub water will be circulated through the chamber, and heated directly by the heat of the fire. I'm sure there'll be issues I'm not thinking of though.

dan268 (author)DougM432015-12-31

I agree that the chimney is not the best place for the coil. I wanted the coil to be vertical to allow for easy drainage, so I wouldn't have to worry about the water freezing when I'm not using the stove.

To get rid of the creosote problem, the flue gasses have to be kept above 250 F. I think the only way to do this is with a bit of insulation between the water and the fire. I like your idea of putting the water piping along the top of the stove, but I think I'd just leave the whole thing on the outside.

FYI, if you want to go the water jacket route, a 30-gallon steel drum nests nicely inside a 55-gallon drum.

AaronR6 (author)2015-11-19

If you allow the water to bypass the pump, a natural convention current will occur, mitigating the risk of your low temp solder joints from melting.

dan268 (author)AaronR62015-11-19

Thanks for the suggestion.

The solder joint melting was caused by user error. The tub only had a few inches of water in it when I started the fire.

The pump is a centrifugal pump, so when it is turned off, there is no barrier to water flow. Nevertheless, I haven't found much of a convection current. I think that for a system like that to work, both the inlet and outlet have to be submerged. Also, the highest point in my coil is about 5 ft. above the water level of the tub, and the natural convection systems seem to have the top of the coil at a level that's lower than the surface of the water.

Parmenion (author)2015-05-17

enielson (author)2014-03-19

This is an awesome instructable. I am dying to give this a try.

ArtificerMade (author)2014-01-14

You can braze the copper just fine using something like stay-silv or (way cheaper) a phos-copper brazing filler. In fact, using phos-copper brazing rid means you won't need to use flux.

kwhit190211 (author)2013-10-01

I would try to solder the joints using silver solder for a higher temp. As far as brazing copper that's a impossibility, as the brass won't stick to the copper. I found that out long ago when I was making a grill out of schedule 80 copper pipe..

bmaust (author)2013-05-18

could you use an old hot tub for this instead for the Rubbermaid tub?

dan268 (author)bmaust2013-05-18

Sure. Any kind of tub would work. The Rubbermaid tub is nice because it's easy to transport, and already has a hole in the bottom for connecting the pump. If you're not planning on moving the tub, though, there are a lot of good alternatives.

danzo321 (author)2013-05-02

wonder if rope in the tube could work.

dan268 (author)danzo3212013-05-04

I'd be worried that it would get stuck. If you try it and have success, let me know.

ibarnett52 (author)dan2682013-05-07

wax or sand works great

dan268 (author)ibarnett522013-05-07

I used sand for this project, but getting it in and then getting it out was a huge pain. Wax sounds like a much better idea, though.

iMartijn (author)2013-05-05

When you turn the pump off, how do you keep the water inside the coil from overheating?

dan268 (author)iMartijn2013-05-05

You need to leave the pump running all the time. Otherwise the fire will cause the solder to melt and everything will fall apart. If the heat exchanger was just a single piece of tubing (or perhaps if the solder joints were all outside where the heat is lower) you might be able to just let the copper heat up.

iMartijn (author)dan2682013-05-05


The I don't understand the part of the shop vac. I'm not even sure if I know what a shop vac is. (English is not my default language)

dan268 (author)iMartijn2013-05-05

The shop vac blows air into the stove to make the fire hotter.

mr fat (author)2013-05-04

Thats such a great idea and its something I had thought of doing for a long time!

zawy (author)2013-05-02

If anyone remained confused and in the dark like me even after reading everything: water instead of air is being pumped through the copper coil and the coil is inside the chimney, not in the tub. If you want to geek out on the heat conversion, assume R-factor of 0.1 for air flowing over a surface, 50 ft times pi*1/2 inch / 12 in/ft = 6.5 ft^2 exposed surface, assume average 700 F stove output temp above water temp, and use heat loss equation for the R-factor: Watts = degrees F difference times ft^2 / R-factor / 3.4 which gives Watts = 13,400 which is what you measured. 1/12 cord of wood is about 50 kWh (nearly 4 hours at 13.4 kW) which is about what you got out which would mean 100% conversion efficiency instead of what I would have expected to be 20% so I would have thought nearly 1/3 cord was used.

dan268 (author)zawy2013-05-02

The way I did the calculations was to look at the energy put into the tub (mass of water multiplied by change in temperature). The heat value of wood can be found on the internet. I think I ended up finding about 20% efficiency when I did the calculations that way.

If you're still confused, let me know and I'll post the spreadsheet I used for the calculations.

zawy (author)dan2682013-05-03

I agree with your energy calculation as received by the tub as I showed the design of the coil should provide that much. My confusion was that I made a mistake on converting the energy in a cord of wood. 1/12 cord of wood is about 500 kWh instead of 50 kWh, so my calculations based on your heat exchanger design agree with your measurements.

Dantheplasticman (author)2013-05-02

Could you run tubes straight up, then use a 180 fitting and come straight back down and arrange the tubes in a circle? That way it could be taken in and out to clean if needed.

dan268 (author)Dantheplasticman2013-05-02

One of the reasons I decided to go with a vertical coil was to make the tubing easy to drain in the winter. If you don't have to worry about freezing, vertical tubing might work.

Since there's no air vent in the top, you have to jiggle the stove a bit to get everything to drain out.

There was a lot of creosote build-up on the heat exchanger. I'm going to try cleaning it out with a chimney brush and see how that works.

scristian (author)2013-05-02

this is realy well explained gj :)

jmcgaghey (author)2013-05-02

I haven't personally done it, but I have seen it being done with small diameter tubing and it works.

you could have also gone and got a set of tube bending pliers I have used those for making sharp angles for brake line tubing which is pretty much the same diameter.

WimG (author)2013-05-02

expensive counterpart, but great for stealing ideas:

lordzamiel (author)2013-05-02

For bending the heat exchange coil, you can buy basically a long spring in the plumbing department specifically for coiling copper tubing - no kinks, fast & easy. is an example - Hope it helps somebody =)

jmcgaghey (author)2013-05-01

a better way that would save you a lot of time would be use water instead of sand and then freeze the tubing. Gives even coverage within the tubing and easy clean up after.

dan268 (author)jmcgaghey2013-05-01

Oh man, I wish I had thought of that earlier. Have you actually tried it?

triumphman (author)2013-05-01

Hey who made those bubbles ?

Moerdith (author)2013-04-30

I was talking to a guy who was working on making one, and his design actually got it to percolate, similar to a coffee pot. I really liked the idea because then there really isn't a need to have a power source. Last I spoke to him it was apparently working, and he was just trying to get a more efficient heat transfer rate.

dan268 (author)Moerdith2013-04-30

Cool. I'd love to see some pictures. Is percolation different from thermosiphoning? I like the idea of not having to use electricity. From what I've read, though, thermosiphoning seems to be a bit finnicky. The stove puts out a lot of extra heat, and your comment got me thinking about using it to circulate the water. Maybe some kind of steam engine.

Kozz (author)2013-04-29

"That's so hot..." </p.hilton>
But seriously, that looks like fun.  I can't help but think what you might be able to do with a rocket stove mass heater.  I saw an article on one recently and have been reading more and watching videos obsessively. 

I wonder if you could create the mass ( the "mass" in rocket stove mass heater) out of a body filled with water.  You may get some serious gains in heat retention (and combustion of ... well, combustibles).

Best wishes, and keep 'tubbin.

About This Instructable




More by dan268:Wood Fired Hot Tub
Add instructable to: