Poor thermosiphon effect -- add pump?

I have a woodburning cookstove with a boiler insert on one side that is plumbed to thermosiphon into a hot water tank. It works but not as well as i'd like. It does pretty well at getting the water tank up to 100-110 degrees, it will eventually get it up to 115-120, but really struggles to get the tank higher than that. The water coming from the boiler to the tank is often around 140 but can be higher (i've had it up to 220!).

One major indicator of our problem is that the TPV release valve will start sputtering because the water is overheating in the boiler, but the water in the tank will still only be 110 degrees or so!

This leads me to believe that the thermosiphon is not performing well and that if i could get the water to circulate more quickly from the tank to the boiler and back, that I'd end up with cooler water (not overheating) in the boiler and hotter water in the tank. What do you all think?

I wonder if installing a pump would do the trick, and if so, what kind of pump, how to install it and how to control (automate?) it and keep everything safe and reliable?

PS - This is not my house, i'm renting it. The house is somewhat experimental and the owner is open to making changes / improvements.

Thanks!
Troy
Dancing Rabbit Ecovillage

Troy, is there an open expansion tank for this system?

Is your pipework all in good condition with no extreme bends and the hot feed always going upwards toward the top hot feed on the tank?

Is the bottom cold outlet of the tank higher than your stove?

You could install a pump easy enough. A circulator pump would work and is an easy 120 volt. You could get fancy with controls, but honestly just an off-on switch should be enough.

Troy, this situation is potentially very dangerous, the hot water in the woodburner isn't going anywhere and there's a risk that the internal boiler could explode. Keep your temperatures down until you have found a solution to the problem.

Travis, I agree that adding a pump will help but it means always relying on having electricity to have hot water. Is Troy off-grid or are there a lot of power cuts in Missouri ?

One of the key phrases in the opening post, "The house is somewhat experimental" leads me to think that there could be a simple explanation for the problem but it would good to know more about the system itself - diagrams and photographs would be helpful too.

I've loads of questions, sorry.

Is there a thermostat for hot water in the woodburner?
Does the system also run radiators and if not, is there a small radiator to dump excess heat?
Has the system worked OK up to now?

Thermo-siphon by it's nature is not a powerful motive force for fluid. It's "free" but not forceful, even when the hot part is very hot. The TP valve is a real heads up, as are the temperatures you mention at the coil. If that coil bursts and dumps a lot of water/steam it could get spectacular in the wrong way.

When trying to make a thermo-siphon effective, the piping needs to be in as straight a line as possible, few and gentle bends. You don't mention what pipe size or material you have. Larger pipes help, and if the run has got several bends in it you might consider going up a pipe size. The coil will still be a restriction but the larger pipe reduces the problem with bends.  Horizontal piping does not help the siphon, just slows it down. The higher the tank is relative to the coil the better. The better insulated the hot pipe is the more effective the system.

I tried numerous times to get thermo-siphons to work with standard domestic hot water heaters, but it was a total crap shoot. Some sort of worked, many just did not . Part of the problem was we did not have a lot of latitude for how, and how far, we ran the pipe - the heater was fixed where it was and the end of the run, usually a bathroom, wasn't going to move much either...

Rufus

Irene, I am not really sure you know what Troy has for a woodstove here. They are very old and predate electricity so they are very safe, in fact if you reread his first post you will see that he actually has a pressure relief valve on the system even if it was made in 1880 or whenever his stove was made. His system is very well protected, he is just having issues getting really hot water from the boiler, which I assume is actually a tank in the firebox, up to the water tank on top of his stove.

My understanding of these stoves is, they are not boilers, nor were they ever meant to be. Again this predates electricity and on demand domestic hot water, so they were designed as water heaters. They took the drudgery out of loading a big tub of water onto the stove top, warming it up, and then using that tub of water for laundry, baths, etc. Just as today we would not set our hot water heaters at 140 or hotter, these water tanks on these old cook stoves were not designed for really hot water either.

But that does not mean the water cannot be cranked up in temperature as you suggest Troy. I think if you moved water a little quicker from the firebox to the tank heater, you would get higher temperature. Back in 1880, this was not really required. I am not sure why you would need to today, but since the pressure inside the system cannot exceed the set presure of the relieve valve (either 12 psi or 30 typically), you should be fine.

If you do it right, your system might actually be a little safer than what you have now, because you are not putting the tank in the firebox on the relief valve as often. By looping the system, you should be getting a higher temperture in the upper tank, but lower in the firebox. In other words you are just averiging the two extremes.

Of course you may also be running your woodstove firebox at too high of a temperature as well. Thermosiphon systems like it low and slow, whereas you might be getting a hot quick fire. in this I am not suggesting you are doing anything wrong, just suggesting another possibility of why you are running on the relief valve.

BTW: Me and my wife get absolutely giddy with old stoves and I am kind of jealous of the gem that you have! We have old stoves, but not ones with side tanks on them! :-)

Thanks Travis,

It's true, I'd no idea what stove he has, he didn't say but I've designed thermosyphon heating systems and installed a lot of woodstoves and back boilers and thought I'd help - I've been worried all evening !

Our latest system, a Rayburn, heats 400 litres of water and six radiators. I installed a pressure release valve in the water tank but the installation manual also recommends an open expansion tank for a thermosyphon system and of course with the output of this particular stove, I'd never run the water heating without having a small radiator open to act as a dump load if the water overheats.

It's difficult to give advice without knowing much about Troy's stove and his pipe runs and so on - but I'm relieved to hear that you consider that his system's safe, so I can go to bed now - it's 1am here !

Irene Kightley wrote:Thanks Travis,

It's true, I'd no idea what stove he has, he didn't say but I've designed thermosyphon heating systems and installed a lot of woodstoves and back boilers and thought I'd help - I've been worried all evening !

Our latest system, a Rayburn, heats 400 litres of water and six radiators. I installed a pressure release valve in the water tank but the installation manual also recommends an open expansion tank for a thermosyphon system and of course with the output of this particular stove, I'd never run the water heating without having a small radiator open to act as a dump load if the water overheats.

It's difficult to give advice without knowing much about Troy's stove and his pipe runs and so on - but I'm relieved to hear that you consider that his system's safe, so I can go to bed now - it's 1am here !

Dear irene, thank you so much for realizing I was in no way being condecending towards you. Sometimes the written word is hard to understand and emotions are not properly conveyed. I kind of figured you did not know exactly what he had (and I do not know exactly, but think it is something very similiar to the stove in the photo). That is fine, you were not entirely off, and neither was Rufus, a relief valve is a back p plan, but maybe it is not best to rely upon it in the first place. I could tell instantly too that you knew what you were talking about, just maybe not with this particular type of stove.


You do bring up a point I did not think of though, MOST circulators are not check valves, so if the power goes out, and the system can thermosiphon, it will. But there are some circulators with intergral check valves that will not.


In any case I hope you are deep into sleep by now. If You are, i am deeply jealous as I have not slept well for years. Even with prescribed medication, sleeping through the night escapes me. i am deeply jealous of those that can get that.

Irene Kightley wrote:Thanks Travis

Actually, after a few moments of thought, I think I need to owe you an apology. I reread what I wrote and could have worded what I said a lot better.

I have noted your many postings on here, and honestly do have a deep respect for your thoughts and insights and could have worded my reply to ensure you knew that. I feel you add so much to the forum, and hope if offered someday the chance to moderate, you do so. I deeply respect moderators.

I could have done this in a private message...a "Moosage", granted, but feel my lack of character in replying to you, was best done on the forum. It is an apology that is truly warranted.

Rufus Laggren wrote:Thermo-siphon by it's nature...

Rufus, I only quoted you so that you would hopefully get a notification and see this as well.

You are absolutely spot on in this reply. Your reply came as I was typing so it was ahead of mine, but I was in no way refuting what you said, though it looked that way.

But I wanted you to know, that I am in no way side-stepping you in my public apology to Irene. I have noted your presence on here as well, and think your insight truly improves this forum. Like her, if you ever get offered to moderate this forum, I hope you consider it. Again I truly respect moderators and feel you would make a great one.

Don't be daft Travis, I'm not bothered - honestly.

I've been around in forumland for a while (Admin for the French permie forum for 9 years) and I'm pretty sure of myself, she said.

I'm still up, I can't sleep either.

...oh, and where is the photo of the stove you mentioned ?

Troy: Does your stove look something similar to this?

Travis, seems to me you say things pretty well. Certainly no problem here.

I in fact don't have a clue as to details of Troy's installation but figured I could justify my existence with a review of the basic basics of thermo-siphons. They are, in my experience, finicky creatures and it may help to allow that this particular bit of plumbing can depend on a bit of luck. A pump is the "white man's way" of just applying brute force and have done and with these contrary buggers that may turn out to be necessary.

Cheers,
Rufus

High everybody, thanks for the feedback so far. I'll try to address all the questions.



The stove is fairly new, it's an Esse Ironheart. The boiler is indeed a tank style (as opposed to coil) on one whole side of the firebox.

The tubing run is copper, 3/4 or 1" i think. It has several bends including basically a 180 and has a very flat total rise/run ratio. The tank is a Superstor Contender 50 gal. It's not very high off the ground and that may be part of the issue as well (that the tank is not much higher than the stove). The tank is only a few feet away from the stove.

There are multiple TP valves and expansion tanks in the system, so it should be reasonably safe. It's still quite disconcerting though when it boils over and we certainly try to avoid it.

We are grid tied and have solar power as well, however, it's still not assured that we'll always have power. Whatever we implement must be safe without electricity.

We also have evacuated solar tubes outside that heat a coil inside the Superstor hot water tank (when they're not pumping cold fluid, but that's for another post).

There is no thermostat or controls.

If we need to dump heat we can run the radiant floor system or just run hot water at the taps, most of the time though, we don't have enough heat.

I'd love to be able to get the tank temperature up in the 160-180 degree range, so that we'd have a greater quantity of usable hot water and also the tank would stay hot longer, also we could use the radiant heat more. Some of the other buildings here have very large water tanks that they heat up to 160-180 with big wood fired boilers.

I think the system has always worked this way, the owner and past tenants have mentioned it happening to them.

I wonder if we have a poor/weak thermosiphon due to low-ish tank height, kinky plumbing, and/or other issues. Is thermosiphon force proportional to the temperature differential? It seems like the water warms up reasonably well from room temperature up to 100 or so, but then it's much slower to get it warmer beyond 100. Perhaps as the water in the tank warms up, the thermosiphon force is reduced because the temperature differential becomes lower. This effect could be something of a downward spiral to the point where the water is barely flowing from the boiler, and since it's spending a lot of time there, it gets overheated and that's how we wind up with boiling water in the hot part of the system even though the storage tank temp is quite low (120 maybe).

Possibly relevant is that in the summertime the water is heated by the solar tubes only, and I'm not sure i recall it ever getting much above 110-120 then either. That system is actively pumped.

Would installing a 120VAC pump on a control that ran when the boiler is hotter than the tank work well and be safe? It would need to not substantially impede the water flow when off, so that the system would be safe during power outages.

Okay, I got you now.

You need what is called a heat dump loop. In your case dumping the extra heat into the floor would be the best bet. To do that you need to have an aquastat that detects the water's temperature, and when it hits a predetermined temperature, opens a zone valve to dump the excess heat into. A few check valves would keep the system from back feeding, and a circulator may or may not be needed. Hopefully not as that requires a bit more controls and set up. That would depend on your exact set up however. (Zone valves work on 12 volt, and circulators are 120 volt, so it requires a transformer and relay to make sure the zone valve and circulator come on together.

Travis, that will prevent that scary booming rattle when the water inside the Esse boils but it won't solve the problem of not having really hot water.

It's a shame to waste the energy heating the floor, especially in the summer, when the solar isn't giving us much and we need loads of hot water for washing blankets or making conserves.

This Esse stove (UK built and really popular) has about a 9k output depending on the fuel, with about 2.5kw for water heating and it should run one or two radiators too, unless you use an awful lot of hot water.

"I wonder if we have a poor/weak thermosiphon due to low-ish tank height, kinky plumbing, and/or other issues. "

I feel that this is what you should address and my approach would be to keep things simple and use natural physics instead of technology to solve problems. Thermosyphon works really well if the hot pipe coming from the stove has an incline upwards and the final hot run to the tank is almost straight up. The inlet to the hot entry on the tank should be a metre or so higher than the stove, with as few joints and turns as is possible. Changing that could be cheaper that buying and installing a pump.

More questions for Troy !

The radiant floor heating running from the stove, is it operated by a pump which you switch on when needed ?  

Is your solar tied into the same pipe system or is it on a completely separate system to the water tank with two serpentines ?

I've _never_ seen a thermo-siphon work through a check valve of any kind. Also, check valves stop working - either open or closed - after 3-5 years. IOW, they need maintenance.

It will help to keep the hot pipes hot, ie. insulate them as much as feasible in your situation. It is not uncommon for 50gal (and smaller) tanks to be raised 4-8 feet; some people put them in a conditioned attic space. Your normal plumber would groan about that type install but, unless he's feeling real rich or he's old and decrepit like me, would take the job, no problem. Over 50gal, you start getting into industrial size muscle. Old beautiful cast iron tubs (well, ugly ones too) still take the cake for most gawd-awful to horse around. Up to and including 50gal HW tanks don't even rank. <g>


Rufus

Oh, and "160 to 180 F...."

That is _seriously_ HOTTT! It is hot enough to be dangerous in its own right all by it's lonesome. Old style hydronic systems can run that type of temperature, but they consist of a totally closed system (since about 1940) with T&P valves specified to handle the BTU's that may need to be dumped if something goes wrong. The materials are made of steel and cast iron with maybe a copper boiler. No civilian ever gets the option to contact that heating fluid at all. Restaurants use equipment that _may_ use water up to 160F. (rarely -  it's almost always 145F.), but it's w/in a totally enclosed appliance engineered for safety and there are serious penalties for anybody messing with the design or installing improperly.

IF you ever fabricate something that hot that supplies domestic water, you MUST install a tempering valve right at the tank which limits to 120F. water delivered to the home fixtures. Without that tempering valve, properly installed and properly set you are civilly and likely criminally negligent and LIABLE. Think the old lady and McDonalds.

At one time I was  a licensed plumber and we get very narrow minded about stuff like this...


Rufus

Moving the tank could be a pretty big job, it would require changing a fair amount of plumbing... i'll look at it and consider that. I'd be afraid that we'd make the effort and still not have satisfactory results. It would be nice to keep it simple and passive though.

There is a mixing valve on the hot tank output that controls output temp to the hot water system (and a separate one controlling output temp to the floor tubes) so that the water in the tank could be very hot but still only 120 coming into the fixtures.

Floor tube pump is operated by a wall switch.

The boiler circulates the actual water in the tank, there is not a coil or heat exchanger, i believe that's called an open-system, right? The solar tubes circulate through a coil in the tank, indirect heat. They use an antifreeze mixture to prevent damage in the outdoors parts of the system.

Thanks again for the feedback

Where are you based Troy?

Is there differential temperature sensor which activates the solar pump when the water in the tank is cooler than the fluid coming from the solar manifold?

It would be ideal if you live closeby John (and you're a plumbing person) it's difficult doing everything in writing!

Is there any reason the system cannot be upsized by larger diameter plumbing? That would move a larger volume of water from the firebox to the tank without adding a circulating pump.

John, at Dancing Rabbit Ecovillage, near Rutledge, MO.

Irene, no sensors currently, but we need some, and a controller. That's a separate issue though, and i don't want to chase that tangent in this thread. I'd be happy to elsewhere though!

Travis, up-sizing everything would be a bugger of a job, i doubt the owner would go for it. Also, i wonder if that would work if there were still a couple of spots that couldn't be upsized, like the fittings at the boiler and the water tank? And again, what if it doesn't work? It would be a lot of effort and expense.

I really lean toward an electric pump because I think it's the thing that is most likely to really work. However, i'm not sure what are best practices to make it as safe and functional as possible in the event of failures or power outages.

I agree with your assessment.

The cheapest way to see if moving water faster would work better, would be to put a circulator pump into the system. A circulator is about $120, and then just wire it to a 120 volt outlet. Let the system heat up, then plug in the pump and see what the temperatures do.

If that worked, then you could add better controls once you proved a pump would work.

Troy

Having a thermostatic tempering valve on a HW tank that's going (maybe) to get really hot is _the_ proper install. Good set up.

As Travis says, a quick and dirty pump install is the best way to get "proof of concept", get a better handle on what's happening and what might work. Depending on how many people the (sort of) working system now serves you might be able to (briefly) get really crude, ugly and creative with the test pump install. But you probably want to be  at least somewhat confident in your plumbing skills. And it wouldn't hurt if the whole system could be out of service for a few days w/out upsetting people. There are few _really_ cheap pumps...

> open system
Means that the fluid has a connection to the open air so that it never gets pressurized. (Except if part of it gets heated so fast so high that it flashes to steam and that steam can't get out fast enough - all bets are off. But if your stove system is open, there shouldn't be too much risk.)

One other item of real concern is what type of antifreeze you're using in the solar HW system that circulates in a coil in the domestic hot water tank. Those coils have been known to leak for any of a number of reasons. If the coil springs a leak and the antifreeze is also anti-human (ie.  poisonous) and it contaminates the tank, somebody drinking that hot water as part of a quicky tea build could suffer. Not to mention getting it their eyes or breathing the mist in a shower.

The attempt to ensure a very high confidence level for domestic water hygiene explains much of the high cost of plumbing and the bureaucracy surrounding it. There is a _serious_ institutional effort to keep domestic water clean in 1st world countries. Different jurisdictions regulate indirect storage tanks differently and there is "controversy" about the risks and costs associated with these systems. But it's definitely something to be aware of. It's been years since I studied solar hot water and I don't recall if there are "food grade" antifreeze options. I _think_ there may be, but just can't recall. If the thermo loop from the stove is closed, not open, meaning the _tank_ is pressurized and that pressure is significantly  _higher_ than the solar HW loop maintains, there is less danger - but it's still not a good system unless the antifreeze is actually safe for humans. Murphy messes with the best laid plans.

Rufus

Good points Rufus,

Our system is not open, it's pressurized. What I thought that open meant was that the fluid going through the boiler is actually the water that will eventually come out of the faucet, as opposed to a contained separate fluid like the solar heating loop.

Great point about the possibility of poisoning from a leak in the solar loop, I hadn't thought of that! I'll have to check with the owner about what exactly is in there. We don't usually consume the hot water due to it being nor difficult to keep clean of bacteria and such (the whole system is fed from rainwater catchment). We do shower in it though.

I'll do some googling on "thermostatic tempering valve", I don't know what that is.

Oh, a temperature mixing valve, yes those are already installed in multiple places in the system.