Thermosiphon Direction of travel

As my woodstove will be lit probably starting this week and will remain lit and loaded likely until late April, every year I contemplate a hot water loop to take some of the demand off my propane hot water heater. My stove sits about 9 Feet from the hot water tank which is in a closet opposite the stove off the living room. The hieght of the coil in the stove would end up ~32" high and the inlet on the top of the water heater is ~58" high. Seems ideal for a thermosiphon set up, but I have a question as to the logistics of that. For a Thermosiphon to operate properly does the rise need to be continuous and singular. By this I mean do I need to go in a straight line from the stove outlet to waterheater inlet. This poses a serious problem as it would put a diagonal pipe across the center of my living room space from roughly thigh to shoulder height. Obviously this is not an option. However a hewn beams runs this direction across the cieling. Can I go from the stove, up to cieling height along and across the beam, then drop down to the top inlet of the tank (again ~58")??

The return would go from the bottom drain of the tank to basement and across then back up to the stove coil which could be anywhere from maybe 20"-32" high on the stove depending how I set up the coil.

So will a thermosiphon operate set up like this? I could sketch a picture if my ramblings are too hard to follow let me know.

I would rather not go with a pump system as the woodstove runs continuously regardless of if we have power it also complicates things to go this route.

Bump.  I can't answer your question definitively, but the thermosiphon works on the principle of the hot and cold cycling, so as long as your height differences are not that extreme, I think you should be fine.  I hope someone else with more experience pipes up (pun intended) for a better answer for you; there likely are some restrictions on height.  The way I see it, the hot water is going to rise out of your heating system and be replaced by your colder water from the bottom of your hot water tank; there's your thermosiphon, no matter what; but you need efficiency too.  The biggest issue that I can see from your description, if you have a continuous hot fire going, is that at some point you may overheat your system.  So be sure whatever pressure release valves you have are working well.    

I subscribed to this thread to see the answer an expert gives...  My belief is that you can pipe it up to the ceiling and then over.  I might try to have it continue a bit of an upward slope as it goes along the ceiling beam.  

I always get confused with siphons though when you connect up to the water heater tank.  Since the hot water in the tank is at the top of the tank, will the hot water from the wood stove want to push that hot water down or should you connect up to the bottom of the water heater?  My real hunch is that the hot water tank should be one story above the heat source.  Then the hot water can rise and go in the top or middle of the tank and then the coldest water in the tank can sink down to the heat source.

I have a similar situation except that my water heater is 2' away from my wood stove

Be sure to be safe with how you do the heat coil.  Don't create a boom-squish type of situation with steam.  When/if I try this, I'll want to rig up a water heating system that doesn't change the UL listing of my stove.

Generally thermosiphons are overstated by nature. They are simple enough in theory, but seldom really work well. In your case, even as simple as it is, you would have to put a high point vent in and a check valve or two. Even then, consider this, the first combustion engines had thermosipon cooling systems and quickly abandoned them for pumps.


It might just be easier to connect a circulator pump with a relay that comes on when the hot water has a call for heat. Put another overtemp relay on going to a heat dump like a cast iron steam radiator and you would be much more safe from a squish-bang situation. All would be easy to do do with a relay system with priority, 24 volt zone valves, and a 120/24 volt transformer.

As a wine-maker, I do a lot of siphoning... I'm pretty sure that the proposed system would work poorly if at all. You'd have hot water trying to rise out of the water heater, and hot water flowing in the opposite direction trying to leave the stove. They would create a hot-pocket near the ceiling that would tend to prevent flow in either direction. Then, when the inevitable air bubbles formed in the system, they would likewise form a plug, near the ceiling, that would stop the flow. (The force exerted by a bubble is much greater than that exerted by thermo-siphoning). I highly commend Travis's suggestion of a pumped arrangement.

I like Travis' idea of the pump. I tend to think though, that the pump might need to run continuously, even if it's at a very slow rate of flow, to prevent the coil of pipe at the wood stove from becoming a boiler.

Yet another - not an expert.

Earlier today, I appended a reply to a root cellar in Montana thread here on Permies.  In looking around since then, I stumbled across a too brief description of a kind of thermosyphon heating/cooling/ventillating comment in some other website also on a root cellar type idea.  So, I went looking some more.

Most thermosyphons seem to be hydraulic (all liquid) or evaporative (mixed liquid/gas).  There are thermsyphones which work on a loop principle, and if so, the loop systems can derive greater heat transport.  I believe in any loop system, you need to have (almost) monotonic elevation changes.  And one side of the loop needs to be on average hotter than the other side.  If the pipe is going up on the hot side and then goes down (far enough), that can be enough to stop or severely slow the circulation in the system.

For all the reasons stated above I worked on another solution to this to make mine a stand alone solution.  I tried copying the old John Deere A.  It has huge water piping and a fan to remove the heat so it will cycle.  It was to impractical for me for that size piping so I decided to go with a pump.  It cost me a couple hundred buck to do it though.  I used a couple TEG that were rated 15 watts at 12 volts at 300 deg f temperature drop.  I had to use 4 of them to get the pump to work because I only run my water to 200 deg.  The pump comes on and runs when there is enough heat.  The hotter it gets the faster it pump. It self regulates.  I did put a backup system with a 50w taco pump to start the system and if it fails that I can easily run with a battery while fixing any problems with the TEG.  So far so good, but only one year in testing.

You will need to place the tank higher than the heat excanger at the heater, give it large enough waterways and it will flow just fine. A coil will not thermosyphon for you its a bank of traps like a drain trap uses to keep the pipe sealed.

A heat exchanger that is a parallel manifolded riser set or a column will work. For single columns they need to be large, 3" is getting there. A tank with its lower third in contact with the stove will do the same. This allows a conveyor belt within the pipe (or tank) from heat source to a storage in the reverse of say, a double pane window, where the spacing of the panes (1/2 or less) stops gases from thermosyphon/convection heat loss.

At 3", liquids will have room for a circuit or conveyor belt for heat, warmed water rising and cooler water falling within one pipe. A pot of water does this on the stove!

For a heat exchanger on a wood heat plant without a pump or circulator, there would likely be several to dozens of vertical risers or pipes that have a common manifold head and foot horizontal, connecting them. A ladder on its side is a similar concept visually.

If you feed the diagonal opposed corner and place the bottom of the tank any higher than the outlet or at least theas high as the inlet, heat will drive a thermosyphon until you stop feeding the heater wood and it cools. This detail, parallel manifold, diagonal opposed inlet and outlet, causes fluid to flow evenly through all risers.

Even a small tank could be the HE, imagine a briefcase shaped tank pressed to the back of the stove or heater and two 1" pipes, (big is good but need not be) one feeding the bottom and one coming from the top and the storage tank in line and higher than the HE outlet. That machine will work too, possibly not exactly as laid out in that you may have the HE inside the heater or other details which i cant know.

I am absolutely not a physics professor, and not a fluid dynamics expert. I have though, installed quite a few thermosyphon solar heating systems for domestic hot water and space heat, air and hydronic.

My heater has a coil pressed into contact with a heat shield on the back of the outside of the stove and a circulator to run the heat to an unpressurized storage tank and dhw from an immersed coil. Great system, ivan labs circulator and an Art-Tec pv or battery powered differential thermostat.

The point of that is...... its coming out and i will arrange it as a thermosyphon system for the same reasons you have..... never going to have an overheat (produce steam pressure) caused by a failed controller or circulator, sensor, etc. !

I have plenty of experience with thermosiphoning systems and all work real good.
Pipe size is important, and difference in elevation between the heat source and the entry point to the hot water tank.
I use 15M lengths of 1/2 inch copper pipe around a 6 inch flue pipe, with the cold water supply from the tank bottom, entering the coil at the bottom.
At times this pipe will start boiling and puffing steam into the vented storage tank at times.
The exit point of the coil [ hot water ] is about 4 feet below the bottom of the hot water storage tank.
The tank is 3 ft tall and the hot water enters about 9 inches down from the top.
The pipe between the coil and the tank is 3/4 inch.

I also heat the same tank in summer with a flat solar collector.
It sits on the ground, is angled at about 45 degrees
The cold inlet goes to the base and the hot water exit is about 5 feet off the ground and about 7 feet below the tank base.
I use 1 inch copper pipe between the tank and the collectors.
In summer the panels are 100 degrees C.

Any pipe run must be continuously rising, allowing no air bubbles to form inside the pipe,