Back Boilers and Wood Burners

[a twig]

Resolved a small and previously unknown issue with the back boiler on the wood burner today.

For years we have just sat in front of it, wait for the noises of the boiler getting up to temperature then get up and put the pump on. Today I finally got around to fitting a pipe thermostat to control the pump. Evening came around, get a fire going, the pressure release valve starts whistling at the back of the stove, yet the pump hasn't come on...

"Strange" thinks I, so I get up and and investigate why the thermostat hasn't triggered the pump. Flow from back boiler is stone cold... How can this be if the pressure release valve was doing its thing? Anyway, turn the thermostat right down, pump comes on, circulates the water, release valve is then happy again. Can't be the new wiring then. So...

If the flow and return pipes were completely cold, then the water isn't moving, implying that the gravity feed isn't doing anything. Hmmm. Up into the loft, header tank full of water (cold), all good. Follow the pipes, and note that the shut off valve for the down feed from the header tank is in fact shut off. :lol: :lol:

God knows how long it has been like that. Anyway, in my head that should have solved the problem - we shall see when I get the stove going again tomorrow evening if I am right. Failing that it's an airlock somewhere...

 

[Flyfisher]

Depending on the rest of your hot water system, a pipe thermostat may not be the way to go.

Our wood burner is a 'boiler stove' and is connected into our hot water cylinder. Both are on the ground floor so a pump is needed to circulate hot water from the stove to the tank - no problem. When first installed it was fitted with a pipe thermostat to control the pump, so we light the boiler, water heats up, pump comes on and circulates the hot water through a coil in the hot water tank. So far so good.

But, when the stove died down the pump continued to circulate water through the hot water tank, thus keeping the pipe hot, the thermostat closed and the pump running. The boiler stove was then effectively acting as a 'radiator' and the pipe thermostat only switched off when all the heat in the hot water tank had been dissipated!

We solved this glitch by replacing the pipe thermostat with a flue thermostat. The sensor is fixed to the stove flue and so the thermostat only switches on the pump when the stove is operating and providing heat. When the fire dies out the flue cools, the thermostat switches off the pump and no water is circulated through the tank.

As I said, this may not be a problem depending on the rest of the system but it's something to be aware of.

 

[a twig]

Sounds fun! Good you got it sorted out in the end.

Fortunately with our system though, the three radiators that run off the wood burner are a completely separate system to the central heating and hot water in the rest of the house, so that wouldn't be an issue.

 

[fernicarry]

Depending on the rest of your hot water system, a pipe thermostat may not be the way to go.

I put both a pipe stat and a flue stat on ours. The flue stat to run the pump whilst the stove is firing well and the pipe stat for when it dies down but the embers are still warm enough to eventually boil the water. The pipe stat would normally be set pretty high.

That's the theory anyway, the flue stat has never worked properly so we run only on the pipe stat. I've set it lower than it should be otherwise it hardly ever comes on but not so low that I get into perpetual circulation as Flyfisher describes. Its really not ideal in that when it clicks on the stove gets hit with a big glug of cold water from the pipework and usually the pump clicks off again fairly quickly. This repeats throughout the evening. Not good for the boiler and exactly what the loading valve is designed to avoid. I also find that the air damper tends to close because the heat from the boiler isn't being taken away quickly enough and the fire dies down earlier than I would like.

I really should fix the flue stat. Only reason I can see for it not working is that the capillary got kinked during fitting.

-AL

 

[Flyfisher]

Interesting - it just goes to show how different heating systems can be.

I set our flue thermostat quite low so that the pump comes on as soon as the stove is providing heat. The 'glug of cold water' is not really an issue for our system as the pipe work to the hot water tank is mostly within the kitchen floor, which has under floor heating, so the pipe work is usually already warm.

Fair point about the stove still generating significant heat from embers even after the flue has cooled, but overheating the boiler is not a problem in our case because we also have gravity-fed bedroom radiators. While this is obviously convenient, it is also necessary for our system because the hot water tank is on the ground floor and thus relies on a pump to circulate the water from the boiler stove. This is fine until there is a power cut at which point the stove could potentially overheat, but the bedroom radiators prevent this - and we've tested this!

We also have a gas boiler that feeds heat into the hot water tank via a separate coil so the usual system operation is for the gas boiler to come on in the early morning to heat up the kitchen floor (which is open to the sitting room containing the boiler stove) and take the chill off our main living area, then when we light the boiler stove it provides enough heat for the gas boiler thermostat to switch off even with the kitchen UFH and bedroom radiators still on. Thus, the more wood we burn the more gas we save.

 

[a twig]

Well tonight's experiment was interesting - the pipe temperature definitely got warmer, seeming to show that there is now some sort of gravity flow. However, the pump still didn't kick in, as the pipe stat only got to about 25'C before the pressure release valve started whistling away.

If I move the pipe state much closer to the stove, the ambient temperature will melt it! The pipe stat datasheet states that the accurate operating range is only between 40'C & 70'C so despite it being set at 20'C it isn't coming on...

So, either I go down the flue stat route, or get a wide range pipe stat. More worrying though is that the gravity feed doesn't seem to do the job on its own, which doesn't bode well for a power cut. I don't want to be entirely reliant on the pressure release valve if we lose power when the stove is on full chat!

I think a quick trip round the rads with a bleed key is in order. Failing that, the gravity system must not be set up correctly somehow - which is a worry! :shock:

 

[Flyfisher]

The chap who designed our heating system said that a boiler stove should have gravity-fed 'heat leak' radiators totalling 50% of the rated stov e capacity. Our stove has nominal ratings of 7kW to the room and 14kW to the boiler, so we have 7kW of radiators connected.

Actually, 7kW was too much for the bedrooms so we also had a radiator installed in one of the attic bedrooms. These are never used (the floor is currently too dangerous) so we had a powered valve installed in the feed to the attic radiator that is only closed when power is applied. This power is the same power to the stove pump.

So, the flue 'stat switches on the stove pump to divert hot water into the HW tank and closes the valve to the attic radiator, limiting the 'heat leak' to only the bedrooms where it is wanted. There is also a tank 'stat that cuts in when the tank temperature reaches about 85C and switches OFF the boiler pump (because the HW tank can take no more heat), which also switches off the attic radiator valve, which then opens under spring pressure, thus diverting all surplus heat into all the heat leak rads, thus preventing the boiler from, erm, boiling the water.

In practice, we've only been able to reach this limit with all the CH radiators switched off (we have three zones), but it does mean the system fail safe in case of a power cut.

 

[fernicarry]

twig, how far away is your stat? The pipe should heat up by conduction for a good distance so you'd expect to see more than 25' at a distance sufficient to avoid melting the stat.

I was also worried about the pipe stat body melting if it was in close proximity to the stove so its mounted remotely in the cupboard to the side of the stove. The probe itself is right behind the boiler tapping and the capillary is run behind the plaster in a bit of 15mm copper. Maybe something you could consider? If nothing else it keeps most of the gubbins nicely hidden away.

Our pipework has to go down and under the floor so this limited our choice of stove to ones with an overheat coil. This has been tested inadvertently! I also have a UPS that I intend to wire into the heating circuit (yet another WiP). You can get reconditioned ones with decent capacity for not very much on that well-known online auction site. Worth it for the extra peace of mind IMO.

 

[paulc]

I also have a UPS that I intend to wire into the heating circuit (yet another WiP). You can get reconditioned ones with decent capacity for not very much on that well-known online auction site.

Or if you don't mind sourcing a replacement battery, I have a stack of 500VA units - You're welcome to one of them as long as you collect (they are heavy little b**gers).

 

[a twig]

twig, how far away is your stat? The pipe should heat up by conduction for a good distance so you'd expect to see more than 25' at a distance sufficient to avoid melting the stat.

I was also worried about the pipe stat body melting if it was in close proximity to the stove so its mounted remotely in the cupboard to the side of the stove. The probe itself is right behind the boiler tapping and the capillary is run behind the plaster in a bit of 15mm copper. Maybe something you could consider? If nothing else it keeps most of the gubbins nicely hidden away.

Our pipework has to go down and under the floor so this limited our choice of stove to ones with an overheat coil. This has been tested inadvertently! I also have a UPS that I intend to wire into the heating circuit (yet another WiP). You can get reconditioned ones with decent capacity for not very much on that well-known online auction site. Worth it for the extra peace of mind IMO.

Piccie is worth a thousand...

Stat is about 6" away from the back of the stove on the flow pipe. The actual stat housing is getting to around 40 or 50'C from the air temp before the pipe is getting hot enough to switch it on (with the stat set at ~20'C, the pipe seems to have to get to almost 30'C before it switches on, but then 20'C is outside the spec range of the stat so to be expected).

UPS is a great idea, all the wiring and pipework goes through the wall into the kitchen - a small cupboard in there could comfortably hide a UPS.

The pump is around 30 years old and most likely needs replacing at some point. I also need to put in a closure plate that isn't supported by timber battens and isn't made of a******s sheet, so I think a general refurb will be in order this summer. Alternatively I may just ditch the whole lot, and buy a bog standard stove without a back boiler, would certainly make my life easier!

 

[a twig]

Hmmm, looking at that pic has made me think. As far as I can tell, the gravity return from the header tank is only 15mm pipe, and the rest of the system is 22mm and 28mm.

That probably isn't helping things...

To be honest, the plumbing in this entire house is a mess - I've just been putting off dealing with it.

 

[fernicarry]

That looks like a strap-on thermostat. I tried some strap-on thermometer gauges on my store once and they pretty much just read air temperature.... Changed them out to ones with pockets and they read the flow and return nicely now.

My practical knowledge of gravity (plumbing) is limited, but 15mm sounds like it would be too restrictive, you'd probably want 28 all the way. Our Aga boiler used to go to a tank in the bathroom upstairs. That was all 28 with machine bends and a gradual rise on the horizontal bit for a length of about 5m. That's probably the reason we have a false ceiling in the kitchen which has been very handy for other services. There seems to have been a similar arrangement from the fireplace in the dining room, probably for a coal back-boiler. The Aga now just goes through the wall on a very short run but still done in 28 with nice smooth bends.

I guess if a refurb is due you can rethink the whole thing. If you have other sources of heating then losing the boiler might be in order. I like the idea of having redundant sources, gas boilers do break down, power goes off and so on.

 

[88v8]

The chap who designed our heating system said that a boiler stove should have gravity-fed 'heat leak' radiators totalling 50% of the rated stov e capacity.

Yes.
The heat inertia of the solid fuel stove must have somewhere to go.

As far as I can tell, the gravity return from the header tank is only 15mm pipe, and the rest of the system is 22mm and 28mm.

The primary pipes flow & return should be at least 28mm, and the thermo-syphon element of the system must not go through the pump or if it does there must be a bypass. With a flap valve to prevent the pump short-circuiting through the bypass loop.
It took me quite a while to figure out that I needed a flap valve.
During this process I also learned that it's hard to reuse solder fittings, and it makes a lot of mess when they leak overnight.

Indeed, for a thermo-syphon circuit, the more the pipe is big-bore, the better.

I used a pipe stat on the ch system I installed in our previous house. It was a fully pumped system, and to avoid kettling, the pump had to keep running after the boiler switched off. This was accomplished by a pipe stat on the return ppe, and a (electrical) relay to feed the pump. It worked fine.
The stat must be in intimate contact with the pipe, not just perched on it, so the curve of the baseplate must match the radius of the pipe. I vaguely recall having problems with this before I got it to work, but this was ummm, 34 years ago.

Ivor

 

[a twig]

That looks like a strap-on thermostat. I tried some strap-on thermometer gauges on my store once and they pretty much just read air temperature.... Changed them out to ones with pockets and they read the flow and return nicely now.

If it measured the air temp I'd be laughing, as the air temp is about 50'C at the stat when the stove is going, so it would be well and truly switched on! Interestingly the last two nights, I've used better quality wood (seasoned hardwood) - the stove has come up to temperature a bit more slowly, and everything thermostat-wise has worked as expected, with no whistling pressure valve. That also does seem to point to a lack of gravity flow - i.e. heat transferring along the pipes, not through the movement of hot water.

So, thanks for all your advice, I think the plan is for some serious re-plumbing over the summer, in the meantime, I think I might invest in a flue stat to do the switching, and a UPS to make sure the pump doesn't suddenly stop! Although since I fixed the dodgy wiring upstairs we haven't had a power outage so maybe we are a bit safer on that score these days

 

[88v8]

For a gravity system, as I've mentioned, the pipes need to be as large as possible.

Also, the radiators should be plumbed top and bottom opposite ends (TBOE). That is to say, the 'flow' and its associated hand valve at the top of the radiator, and the return and the lockshield valve at the bottom.
On a gravity system this significantly affects the output.
Hopefully you have proper cast iron radiators with top tappings and nice chunky brass valves.

For durability and appearance, the flow pipe should be iron. At the foot of the pipe, under the floor, one can go back into copper.

In our fully pumped system, I only fitted one rad TBOE, and that just for appearance. Used 1" pipe and valves. Alas, no pic.
However, I did all the valve tails in iron, for 'period' appearance. In conjunction with brass valves, looks much better than copper.

Ivor

 

[skier-hughes]

The chap who designed our heating system said that a boiler stove should have gravity-fed 'heat leak' radiators totalling 50% of the rated stov e capacity.

Yes.
The heat inertia of the solid fuel stove must have somewhere to go.

As far as I can tell, the gravity return from the header tank is only 15mm pipe, and the rest of the system is 22mm and 28mm.

The primary pipes flow & return should be at least 28mm, and the thermo-syphon element of the system must not go through the pump or if it does there must be a bypass. With a flap valve to prevent the pump short-circuiting through the bypass loop.
It took me quite a while to figure out that I needed a flap valve.
During this process I also learned that it's hard to reuse solder fittings, and it makes a lot of mess when they leak overnight.

Indeed, for a thermo-syphon circuit, the more the pipe is big-bore, the better.

I used a pipe stat on the ch system I installed in our previous house. It was a fully pumped system, and to avoid kettling, the pump had to keep running after the boiler switched off. This was accomplished by a pipe stat on the return ppe, and a (electrical) relay to feed the pump. It worked fine.
The stat must be in intimate contact with the pipe, not just perched on it, so the curve of the baseplate must match the radius of the pipe. I vaguely recall having problems with this before I got it to work, but this was ummm, 34 years ago.

Ivor

Whereas we have a cold water pipe into the boiler, so if it reaches a certain temp the valve opens and cools the fire. Output to the outside. We have an internal water thermometer which works on temps we set, so when I let the fire start dying down, I just turn the temp on this to 70' and once the boiler water temp has dropped below this it turn off the pump.

I'm just about to add a gas tank and gas boiler to supplement the system, along with Genius home control system............... wish me luck

 

[Flyfisher]

We have an internal water thermometer which works on temps we set, so when I let the fire start dying down, I just turn the temp on this to 70' and once the boiler water temp has dropped below this it turn off the pump.

Yes, that will preserve the temperature of the hot water tank at 70C. However, if the stove has heated the water to, say, 80C and the fire starts to die down, the pump will keep circulating the water until it drops to 70C. This may take some time if the hot tank is large. Effectively, the hot tank water will 'drive' the fire boiler, keeping it at the tank temperature. Of course, this will gradually drop (because the fire boiler will now be acting as a heat sink, or radiator) until the pump switches off at 70C (or whatever else you set it to).

The advantage of a flue 'stat is that it allows control of the system based on whether the fire itself is generating heat, so that when the fire dies down the pump is switched off regardless of the hot water temperature.

I'm just about to add a gas tank and gas boiler to supplement the system, along with Genius home control system............... wish me luck

I'm not familiar with that control system, though I often think about having a more programmable system rather than just various 'stats.

I assume you'll be connecting the gas boiler into a separate coil in the hot water tank? This works perfectly well in our system. The issue is how to control the two heat sources.

In our case we have a large area of underfloor heating and this works very well at lower water temperatures than typical radiators. So, we control the gas boiler with a hot water tank stat set to about 50C, which is fine for the UFH and DHW. When we light the boiler stove, it heats the water to well over 50C thus switching off the gas boiler.

The disadvantage of this simple control system is that the CH radiators are only at 50C in the morning and therefore not fully effective. They take off the chill in the bedrooms but it's not exactly 'toasty' - which is fine for us but may not suit everyone.

Ideally, we would run the gas boiler to heat the water to around 80C but with our simple system control lighting the fire would not switch off the gas boiler until the hot water was even hotter. Ideally, I should re-wire the system so that the gas boiler is switched off whenever the boiler stove is generating heat . . . which is on the 'to-do' list but you probably know how those work

What I'd really like is for all the stats and pumps to be controlled by a fully programmable system - perhaps your proposed 'Genius' system will allow this? I sometimes have delusions of building a Raspberry Pi - based control system . . . . but then the sun comes out and I get on with my life

Good luck with it though and let us know how it works out.

 

[fernicarry]

What I'd really like is for all the stats and pumps to be controlled by a fully programmable system - perhaps your proposed 'Genius' system will allow this? I sometimes have delusions of building a Raspberry Pi - based control system . . . . but then the sun comes out and I get on with my life

I actually bought a Siemens Logo! PLC a couple of years ago now with the intention of moving all the controls for our thermal store over to that. So far its still in the cardboard box....

The plan for zoning the house is coming closer to completion (using heatmiser's 12V stats and hard-wired actuators on all the radiators) so I've been mulling over once more what I'd like to do with it. I'm going to have upper and lower limits on the store that will dynamically vary according to demand and whether or not another heat source is available (or expected to be available in the case of solar). For example the flow sensor on the new shower will immediately kick the whole thing into high gear by increasing the high limit to give a sort of combi effect. On the other hand if the stove is firing then I'll make the gas boiler back off by turning down the high limit so that there is somewhere for the heat from the stove to go. I'm also thinking about an outdoor temperature sensor so that it can tune itself through the year. In the summer you generally only have DHW demand so the store doesn't need to sit at a high temperature and its wasteful for it to do so, sort of like weather compensation.

These PLC's aren't terribly expensive and have the advantage of being able to control mains loads directly and talk to standard industrial sensors (20ma). Of course you could do it all with home-brew even cheaper but I want to end up with something that's bullet-proof. Quite important if I am to control the stove pump with it.

-AL

 

[88v8]

I hope you boffins write instruction manuals while this is all fresh in your minds, so that when it goes wrong in a few year's time you'll be able to fathom how it was supposed to work, and for the benefit of the poor b***er who buys your house X decades hence and hasn't a clue

Ivor

 

[a twig]

Oooo - PLC control - now we are talking!

So actuated valves on all the radiators, and a sensor in each room? That sounds like it could be a lot cheaper than having my heating zoned the traditional way, and should fail more gracefully (i.e. valves open in a failure state, so you always get heating). You could prioritise rooms as well, with more weighting given to your own bedroom and less to the guest ones, so incase every room is demanding heat, you could ensure that yours is ok first!

Hmmmm, got me thinking now. The only stipulation I have is that if I'm away from home and there's an issue, I need to be able to walk the wife through it on the phone. She's a sharp one, but I think she might struggle at troubleshooting PLCs!

Its funny, at work I constantly push automation and progress, yet at home it never even occurs to me. Maybe its a subconscious reaction to working at the edge of technology a lot of the time in my work life, that the Luddite in me emerges in my home life!