Designing Back-boiler fed central heating system

[Chevetter]

Hi Chaps and chapesses,

 

Long time since I've been on the forum I know, but I've come back to ask for your advice+opinions.

 

Basically I've been doing a liveaboard while re-fitting type thing for just over a year now (which is somewhat of a challenge, if you've ever tried it) and with winter coming I've saved my last week of time off from work for building my central heating system.

 

My boat is 40ft with a 25.5ft cabin containg 3 rooms - a 9ft lounge/bedroom, a 4ft bathroom and a 12.5ft kitchen/diner.

 

I already have a Boatman stove with backboiler installed and it's 'just' a matter of installing the central heating system. Here's a pic of my design - comments, suggestions, criticism etc welcomed

 

(apologies for the shoddy drawing)

 

So far I have the following questions:

 

Reading other topics on the forum, I think that a Bolin pump will be the best choice as it needs to be quiet and be able to go at the lowest point of the system. Is that sensible?

Do I need to care about matching the number of BTUs of the rads to the number of BTUs of the back boiler?

Is there much benefit to fitting a thermostat to control the pump? I was thinking of just having a switch.

Header tank: where do I get one? Presumably it needs some kind of pressure release valve and dump pipe out of the side of the boat? I was thinking of going to the local automotive scrapyard and hunting under the bonnets of various cars until I found something appropriate?

 

None of this is bought or built yet apart from the boatman so the whole design can be changed if necessary.

 

Thanks in advance

[Chris Pink]

A couple of observations;

 

wisdom says go continuously uphill from the stove to the furthest point of the system, with the header tank higher still (preferrably at the furthest point) then come downhill on the way back. if you use 28mm copper tube then you can do away with the bolin (noisy, very expensive - what's the point of getting dirty in the scrappie to save £15 and then spending over £150 on a crappy little pump).

 

The copper will also dissipate some of the heat and make calculation unnecessary.

 

There was a very heated argument (excuse the pun) on here a while ago about the virtue or otherwise of a pressurised system and therefore the desirability of a PRV.

 

You will need a temperature operated switch for the bolin if you don't want the water to boil locally. This can be fitted anywhere and can be high temperature operated on the high side or low temperature operated on the low side (by the stove). Personally i think a pump is an admission of failure in a well designed central heating system.

[Robbo]

You'll need a gravity feed radiator from the back boiler.

[Smelly]

Your header tank needs to be higher than the system. It's effectively the reservoir for the system.

 

It doesn't need to be pressurised and will be less complicated/prone to fail if it isn't. A Stop cock in the feed with a big "danger" notice on it to remind you not to close it completely when the fire's lit will help prevent too much pump pressure simply being vented.

 

A gravity fed rad is a very good idea, if not essential, to dump the heat should the pump fail. 28mm pipe is essential for gravity fed systems to minimise friction slowing or preventing the thermosyphoning developing.

 

Don't forget, it's only about an amp the pump will be drawing but over a day that's an hour's engine running in itself, gravity fed is better, if not so aesthetically pleasing with all those big pipes.

 

The copper might dissipate some heat, but we found that; as we had cut the "safety features" off the fire to maximise output the fire gave out a lot more than it might. I've got 3.5kw of rads and a calorifier on a supposedly 2kw backboiler. i added an extra 500 watt rad to stop it boiling when the fire was at full tilt.

Edited September 15, 2009 by Smelly

[Chevetter]

Thanks for the comments.

 

I'm afraid thermocycling just is not going to happen. Like I said the stove's mounted on a hearth which is a good 300mm higher than the floor, plus the tilt of the boat puts it a long way above any potential radiators in the kitchen. I can't start chopping the physical features of the boat about so I guess by Chris Pink's definition I have already failed Anyways I need a pump, in which case is the Bolin the best one?

 

Re: the header tank: how big do they typically need to be? Is it actually possible to buy one? Like I said, I was just going to use one from a car.

 

I'm also not 100% sure what you mean by a 'gravity fed rad'? How would that fit into my system?

 

Things that I can move: radiators, pipes, pump, header tank.

Things that I can't move: walls, the stove.

[Machpoint005]

Won't it still thermosyphon even if the stove is not the lowest point on the system? Surely it works because the hot water goes out of the top pipe - and eventually it will come back to the bottom pipe when it cools, because it can't go anywhere else on a closed system.

 

In other words, if the boatman's stove is heating the water, it will try to get out upwards - job done?

 

I would think that the circulation depends on the temperature differential between the top and bottom of the back boiler - which will be the same whatever the height of the stove above the bottom of the system. The stove is the only heat source, and therefore the motive power for the water circulation. It doesn't matter how slowly the syphon operates so long as there is still some heat in the water at the far end of the system. As someone else has said, if you use large bore copper pipes, they heat the boat as well as the 'radiator' (which doesn't radiate - it's not hot enough - and it doesn't have to be 'hot' like a house heating system anyway).

 

All you are trying to do is distribute the heat from the stove more evenly through the boat. You don't need to be able to fry eggs on the radiator.

 

By the way, with a thermosyphon system you need a header tank as the highest point, and the radiator(s) must have hot water going into the top and out of the bottom.

 

Waiting for someone to tell me why I'm wrong, now...

[Chevetter]

Won't it still thermosyphon even if the stove is not the lowest point on the system? Surely it works because the hot water goes out of the top pipe - and eventually it will come back to the bottom pipe when it cools, because it can't go anywhere else on a closed system.

 

In other words, if the boatman's stove is heating the water, it will try to get out upwards - job done?

 

I would think that the circulation depends on the temperature differential between the top and bottom of the back boiler - which will be the same whatever the height of the stove above the bottom of the system. The stove is the only heat source, and therefore the motive power for the water circulation. It doesn't matter how slowly the syphon operates so long as there is still some heat in the water at the far end of the system. As someone else has said, if you use large bore copper pipes, they heat the boat as well as the 'radiator' (which doesn't radiate - it's not hot enough - and it doesn't have to be 'hot' like a house heating system anyway).

 

All you are trying to do is distribute the heat from the stove more evenly through the boat. You don't need to be able to fry eggs on the radiator.

 

By the way, with a thermosyphon system you need a header tank as the highest point, and the radiator(s) must have hot water going into the top and out of the bottom.

 

Waiting for someone to tell me why I'm wrong, now...

 

Ah right, ok. For some reason I was under the impression that the return from the rads had to be downhill. I'll work out how high the top of the radiator in the kitchen has to be to be uphill from the stove and hence if its feasible.

[Chevetter]

How about this revised design?

 

[WotEver]

How about this revised design?

 

I'm no expert, so could well be wrong, but shouldn't the return pipe between the two rads also be 25mm to keep a 25mm circuit?

 

Tony

[Robbo]

I'm no expert, so could well be wrong, but shouldn't the return pipe between the two rads also be 25mm to keep a 25mm circuit?

 

It is, look again!

[WotEver]

It is, look again!

Well, in that diagram it's red and the diagram says red is 15mm

 

T.

[Robbo]

Well, in that diagram it's red and the diagram says red is 15mm

 

It's blue till where the two radiators split, where then it's red to the separate rads.

[Machpoint005]

How about this revised design?

 

 

Just a suggestion - do away with the towel rail radiator altogether, just make sure you can hang towels over the radiator supply pipe. That way you'll get a nice warm radiator in the saloon, and the towels will still dry (they probably won't take much longer, and anyway, so long as the shower towels are warm and dry the next day when you want them, that's all you need).

 

The warm pipe passing right through the boat will air it off a treat, not forgetting that the boatman's stove will give warm air heat right through as well. Think of the volume of a narrowboat cabin compared with that of a living room. As a first guess, one is no more than a quarter of the other?

 

Edited to say: we hang tea towels and hand towels over the hot water pipe (on hooks, not on the pipe itself) and it works a treat.

Edited September 18, 2009 by Machpoint005

[Chevetter]

It's blue till where the two radiators split, where then it's red to the separate rads.

 

My thinking was:

 

Make the feed 28mm all the way down the boat in order for good conductivity of heat.

 

In the return there's no need for this (28mm pipe is expensive)..... although I could be wrong. I figured if its got to go uphill anyway then a bit of 15mm pipe isn't a big deal.

 

Can anyone see a problem withmy design? Anything that I could do better etc?

 

If not I'm actually going to build it

 

Just a suggestion - do away with the towel rail radiator altogether, just make sure you can hang towels over the radiator supply pipe. That way you'll get a nice warm radiator in the saloon, and the towels will still dry (they probably won't take much longer, and anyway, so long as the shower towels are warm and dry the next day when you want them, that's all you need).

 

The warm pipe passing right through the boat will air it off a treat, not forgetting that the boatman's stove will give warm air heat right through as well. Think of the volume of a narrowboat cabin compared with that of a living room. As a first guess, one is no more than a quarter of the other?

 

Edited to say: we hang tea towels and hand towels over the hot water pipe (on hooks, not on the pipe itself) and it works a treat.

 

Hmm, well it certainly saves me having to buy a towel tail thats for sure plus simplifies the plumbing and ensures a nice flow of hot water up the feed pipe. I suppose it will get rather hot and there's only one of me anyway....

 

My boat is actually fully spit in terms of air flow (although I do keep the doors open) so thats a bit of an issue, which is the problem that the radiators are meant to solve.

[WotEver]

It's blue till where the two radiators split, where then it's red to the separate rads.

 

Exactly. That is why I wrote "shouldn't the return pipe between the two rads also be 25mm to keep a 25mm circuit?"

 

In the return there's no need for this (28mm pipe is expensive)..... although I could be wrong. I figured if its got to go uphill anyway then a bit of 15mm pipe isn't a big deal.

 

As I said in my earlier response, I'm no expert. However, the idea of the 28mm pipe (my bad for saying 25mm earlier) is to reduce the flow resistance. I'd have thought that reducing the bore to 15mm, even for a short run, would increase the flow resistance for the entire circuit.

 

But hey, I could be wrong, which is why I'm waiting for someone who knows to jump in here.

 

Tony

[blackrose]

You'll need a gravity feed radiator from the back boiler.

Not if you have a pump!

[Machpoint005]

Not if you have a pump!

...which is why it's better to use no towel rail, and large bore piping throughout - better 'thermosyphon' effect because you get more effective convection.

I imagine an increased copper tube size is cheaper than a pump and wiring - and it costs nothing extra to run it!

[Robbo]

Not if you have a pump!

 

Even if you have a pump, you need something to take away the heat from the back boiler if something happens to the pump not to work.

[ChrisPy]

As I said in my earlier response, I'm no expert. However, the idea of the 28mm pipe (my bad for saying 25mm earlier) is to reduce the flow resistance. I'd have thought that reducing the bore to 15mm, even for a short run, would increase the flow resistance for the entire circuit.

you need 28mm for the entire loop.

The resistance to flow is basically proportional to the square of the flow velocity, so the resistance of a metre of 15mm is the same as that of 4metres of 28mm.

 

another aspect is a 'steam safety release valve'.

beware your stove boiling the water and ejecting lumps of steam.

I have a 15mm pipe that runs straight up through the ceiling (deckhead) and ends in a U-bend discharging onto the roof.

 

On reflection, I 'spose i should have a large sign adjacent to the u-bend saying 'DANGER - RISK OF SUDDEN RELEASE OF HOT FLUIDS - POSSIBLE SCALDING HAZARD' (I've worked in this HSE-ridden industry far too long).

Edited September 19, 2009 by ChrisPy

[Smelly]

I get a feeling you'll regret not having a 28mm return. Admittedly by the return you've got gravity on your side but it'll potentially slow the syphon down... If I had the wonga I think I'd go 28mm to prevent kicking myself later (and I did so no kicks for me!).

 

To save pipe and space, mine has the rads in series on the hot line, then they drop into a common return through balance valves.

 

I'd still put a stop cock between the header tank and leave it turned right down, otherwise, heat going up as it does, you might end up venting the hot pipe as opposed to forcing water through the rads. I suppose it might be worth trying 22mm with pipe bending springs to form the bends but if it's too much of a restriction you'll just end up with a volcano out of your expansion tank which, take it from me, isn't funny...

 

If the return angle up as opposed to the right angles that might be better... For 28mm I've used 1 1/4 inch engine hose to make the bends and that's worked. I've got a gradual rise of about 300mm on one side of the return and the rads on that side work nicely.

 

Otherwise it looks good to me.

 

edit... Chris has a point... I've used an exhaust water scoop from MC out of the side to vent steam as opposed to drilling holes in the roof but too give Chris the credit he inspired it...

Edited September 18, 2009 by Smelly

[Chevetter]

Hi chaps,

 

Again comments much appreciated, please see heating system design Mk3. Am I getting there?

 

On the subject of the vent pipe: do I really need this to be down the stove end? Can I not just have a vent out from the header tank into the bilge? It would make the whole thing so much easier.

 

Also, you will notice that my 28mm pipe runs past the top of the rad. Is there any need to do this or can I just stick the feed into one side and hang the header tank off the other side hence saving nearly 1m of 28mm?

 

[ChrisPy]

the vent is a safety device to prevent lumps of steam banging around in the pipes.

it should be directly above the stove.

[Chevetter]

the vent is a safety device to prevent lumps of steam banging around in the pipes.

it should be directly above the stove.

 

Does it have to be literally directly above? That will create a heck of a lot of work for me!

 

A better idea from my point of view would be to take-off directly above the stove but actually vent the steam somewhere else.

[Chevetter]

Right, been having a bit of a think, conversing with a workmate who has a lot of experience of installing central heating in houses and we've come up with the following:

 

 

Should the boiler boil up or produce steam in this design it gets dumped into the bilge. That way no messing about cutting holes above the stove etc.

 

I assume you're all about to tell me that this is a terrible idea though....

[Chris Pink]

I assume you're all about to tell me that this is a terrible idea though....

 

 

no.

 

Looks good

 

why the valves? having them shut by mistake could be interesting.

 

is it also possible that the overflow, once it gets some water through it will start syphoning?

 

not clear from the diagram, the return from the radiator should run downhill if possible