Another back boiler system design question

[Jambo]

Hi everyone

Another back boiler question! I've had a good look through the forum to glean what I can about back boilers but we're now getting to the stage of specifying the work needed and I have a few questions outstanding so apologies for the rambling length of this post.

This weekend we had a plumber friend over to stay to discuss our proposed back boiler system. He's not an expert on boats or back boilers but he's a good plumber so I'm looking for some feedback from CWF on our proposed approach to check the functionality and safety. I've added system diagrams below and posed a bunch of questions. If you have a moment to answer any of them I'd be very grateful!

It looks like our boat had a back boiler fitted in the past as we have a blanked off 28mm pipe running form just behind the stove to the first rad (and on down the length of the boat). At some point the stove with back boiler was removed and a Mikuni MX60 was installed. We don't like the Mikuni (we've had problems with it, it smokes and I'm keen to burn as little diesel as possible) so we want to revert to the back boiler system. We're looking at a pumped system with a dump rad in the living room as a failsafe.

Our system currently looks like this:

The flow pipe which runs along the boat is 28mm which possibly suggests a thermosyphoning system originally but the gradient must be pretty slight between the front and back of the boat.

Q1: What kind of angle is needed for thermosyphoning with 28mm pipe?

Our calorifier is only connected to the Mikuni. The second coil is not in use and I won't be connecting it to the engine as we only cruise every two weeks so we wouldn't get much benefit from it. We would therefore like to move the calorifier to the bathroom so that any standing losses go to heating the boat rather than the engine room. A long term plan is to add some solar water heating and this would be more feasible with the calorifier further forward (as our PV takes up much of the rear half of the boat's roof).

Q2: Apart from the potential to take heat from the engine is there any particular reason to have the calorifier in the engine room? It gets bloomin cold in there in winter which must suck the heat out of our hot water quite quickly.

The big rad in the living room is completely unnecessary what with the stove being right next door. So we were discussing the potential for removing that big rad and replacing it with something smaller (physically, not necessarily in terms of output) that can act as a heat dump in case of pump failure. What we are thinking is that it would still be connected to the main heating circuit but would have a valve controlled by a flow switch so that if the pump stops the valve opens and water circulates through the dump rad.

Q3: Would this work and be safe?

The proposed system would look like this:

I guess if it turns out that there is enough gradient for the system to thermosyphon then having this dump rad would be a redundant step but as we don't know for sure that the system will thermosyphon naturally, we're going to need a backup. Another option would be to give this rad its own circuit but I feel that the above approach would be neater and would involve less work..

Some other questions:

Q4: I understand that the circulation pump should be located on the cold return part of the circuit but can it be located on any part of the return circuit? My preference is to have it in the bathroom which would be good from a noise and voltage drop point of view.

Q5: Does anyone have any recommendations for high efficiency, low noise DC circulation pumps?

Q6: We have a Morso Squirrel (I think it's a 1410) which has a rated output of 4.5kW. Our boat is a 65ft x 11ft widebeam with so-so mineral wool and polystyrene insulation (no spray foam). If we're diverting heat to the back of the boat and hot water, will 4.5kW cut it? I understand that back boilers can be retrofitted to the 1410 but is it better to go for a new stove with an OEM fitted back boiler?

Q7: How big a dump rad do we need? Is there a rule of thumb of XkW dump per YkW of back boiler rated power?

Q8: It would sometimes be nice to be able to divert heat to the hot water before it goes to the heating system. As all of the rads (with the exception of the dump rad) would have TRVs I imagine this would be feasible by turning down them down? Likewise would it be possible to add a valve to the hot water tank flow so that heat can go to the rads first?

Thanks for taking the time to read this. If you have any other comments / opinions as to the merits or lack thereof of the above suggestion then I'm all ears!

Edited August 6, 2014 by Jambo

[Jambo]

Looks like my editing has generated a double post. Mods can you delete the other thread please?

[Jambo]

Q5...

 

During the course of my research I came across these pumps which look very good:

 

El SID DC Circulating Pumps by Ivan Labs Inc.

Bell & Gossett ecocirc e3 circulating pump

 

http://solarhomestead.com/off-grid-circulating-pumps

 

Both seem to have impressive performance but it looks like sourcing them in the UK might be challenging.

[Stilllearning]

If you move the hot water tank to the bathroom, I would still be tempted to run piping from the engine to the other heating coil, as "free" hot water is always nice to have, and gives almost unlimited hot water for crew/management/friends to shower while you are steering.

On boats I have had with thermosyphon heating, it has always surprised me how little gradient is actually needed for it to function OK.

Others on the forum are far more skilled and experienced than me though, so over to them.

[Bee]

The only thing that jumps out to me is the header tank, I have always tee`d it off from the highest point in the circuit, in this case in bedroom 2 where the hot pipe turns down to begin the route back. the problem with taking it from the cold water return is that the hot water finds it easier to push against the low pipe and it tries to go round backwards, you will not get a really hot header tank as it cannot circulate through it, it just rises and falls with expansion.

[Higgs]

I would be inclined to have two circuits. The first would be the basic circuit from the backboiler, as though nothing else, not even a pump was needed; with only the rads dropped in.

 

If you wished to then have added use, via a valve, taking a breakout circuit, with pump, to the other utlities and Mikuni. At this point, it could also be possible to isolate the second circuit for summer use, where the back boiler and rads are not needed.

 

The biggest benefit of having the back boiler and thermosyphon is that it is silent running and simple. Adding complication and bends will require a pump. This is why I would suggest keeping the first circuit basic, with no pump needed.

[Pete & Helen]

Q5...

 

During the course of my research I came across these pumps which look very good:

 

El SID DC Circulating Pumps by Ivan Labs Inc.

Bell & Gossett ecocirc e3 circulating pump

 

http://solarhomestead.com/off-grid-circulating-pumps

 

Both seem to have impressive performance but it looks like sourcing them in the UK might be challenging.

 

I have a standard 240 volt pump but I am changing it for one of these

http://shop.solarproject.co.uk/newshop/catalog/product_info.php?cPath=2&products_id=10&osCsid=2ussg9jgl2a05g6saac6kmqoh7

I discussed its use with the company and they are happy with its use in a boat CH system. As yet I have no experience with this type of pump but I know that it has been used by one or two others on here.

[Jambo]

Thanks for the swift replies!

 

Stilllearning - I'm a big fan of making use of virtually any and every source of free energy that's available but given that our cruising is relatively limited ( a few hours every two weeks plus occasional longer trips) I'm doubtful as to the benefits of plumbing the engine in, just for the odd tank every other week. If it looks like it'll be relatively trivial to plumb it in then we might look at it for now and then if the solar water heating idea comes together (and it won't be for another year or two at least) then we can always plumb that in as well.

 

Bee - sounds eminently reasonable. I think that quirk to the diagram is more about my ignorance of the finer points of plumbing! 8¬) Will make sure the plumber connects that at the right point in the circuit.

 

Higgs - having a thermosyphoning system as the core circuit sounds like the safest approach. The first emitter is 8ish metere aft of the stove and if we can get a thermosyphon circuit working that far back then maybe we can get it going all the way back? I just wonder if the reason why it was abandoned in the first place was because of poor thermosyphoning? That's pure speculation on my part though. I suppose we can always try it without the pump and see if it thermosyphons. If it doesn't we can add the pump. I really need to get back behind the plasterboard to have a look at what sort of incline the 28mm pipe is on. We're effectively going to ditch the Mikuni as it's been too problematic (I'll probably keep it though). Summer hot water will be provided through soaking up excess PV or, potentially, solar water heating.

 

Pete & Helen - thanks for the link. I get the impression that conventional circulation pumps just aren't designed to be efficient. That one looks both efficient and good value and it's good to hear the manufacturer is happy for it to be used in this kind of application.

[David Mack]

The only thing that jumps out to me is the header tank, I have always tee`d it off from the highest point in the circuit, in this case in bedroom 2 where the hot pipe turns down to begin the route back. the problem with taking it from the cold water return is that the hot water finds it easier to push against the low pipe and it tries to go round backwards, you will not get a really hot header tank as it cannot circulate through it, it just rises and falls with expansion.

 

With a back boiler and thermosyphon system (and ideally with any boiler) you MUST have a pipe which rises continuously from the top connection of the back boiler and ends in a inverted U-shape over the header tank. That is so that if the back boiler should boil, then any steam can escape safely without the risk of blowing boiling water anywhere. The cold feed from the header tank can then be connected to the lowest point in the system (or thereabouts), which will minimise the chances of getting an airlock when filling.

[Higgs]

There shouldn't be any steam. The reduced size tube for the header rises from the end for expansion. Unless the system is not sealed properly, thus allowing air in, a properly bled system should not boil. The only problem arises from a failure of other things like the pump. It is another reason I prefer no obstruction in the thermosyphon system. The expansion tank or bottle needs to have sufficient coolant in not to be fully drained into the system on cooling, and a sufficiently large expansion tank to contain the expansion as the coolant needs on heating up.

Edited August 4, 2014 by Higgs

[Jambo]

We have a pretty decent sized expansion tank already (I'm not good at judging volume but it must be a good 15 to 20 litres or so).

 

I was wondering if in the (admittedly very unlikely) situation where the TRVs are all left closed, we have a full tank of hot water and the stove cranking away, could there be a risk of boiling even when being pumped? The heat would be moved around the circuit and would not get much of a chance to escape so it would return to the stove still pretty hot. The water would expand into the expansion tank but the heat would still have nowhere to go. This scenario isn't going to happen as we wouldn't leave the stove blasting away with no way to dump the heat.

 

You say that "it is another reason I prefer no obstruction in the thermosyphon system". In the proposed system I sketched out up there ^^^ I have a basic, single emitter thermosyphon system as back up with the rest of the system being pumped. Are you concerned about the reliance of the thermosyphon back up on a mechanical valve controlled by a switch? i.e. if the pump and flow switch were to fail at the same time and there's no natural thermosyphoning around the rest of the system then it would lead to problems?

 

In that case would a better solution be to have a pumped system pushing heat to the back of the boat with a separate, thermosyphoning circuit connecting the living room rad to the back boiler? In which case how can that be connected so that in the event of a pump failure the heat is definitely dumped? Wouldn't it still require some sort of mechanical valve (triggered by flow? temperature?) to switch circuits in the event of pump failure?

[David Mack]

There shouldn't be any steam. The reduced size tube for the header rises from the end for expansion. Unless the system is not sealed properly, thus allowing air in, a properly bled system should not boil. The only problem arises from a failure of other things like the pump. It is another reason I prefer no obstruction in the thermosyphon system. The expansion tank or bottle needs to have sufficient coolant in not to be fully drained into the system on cooling, and a sufficiently large expansion tank to contain the expansion as the coolant needs on heating up.

 

Agreed there shouldn't be any steam. But if a pump fails, or a valve gets blocked, or or the fire runs away it can happen. And its entirely automatic, not relying on sensors or valves etc. Better to be safe than sorry.

 

Another advantage of this layout is that it is pretty well self bleeding, so will cope with dissolved air coming out of the water in the days after first firing up the system.

Edited August 5, 2014 by David Mack

[Jambo]

David, I've had a go at interpreting your suggestion as relates to our case:

 

A pure thermosyphon system would allow us to remove the rad from the living room. I still haven't measured the flow pipe to see what sort of inclination it's on but removing that rad would allow us to extend the wall panels down further, potentially allowing us to increase the incline. The rearmost rad and relief pipe could then come off the highest point in the bathroom.

 

I'm still interested in exploring the potential for a safe pumped system so if anyone has a view on a system which would work I'd be interested to hear it.

[b0atman]

Temperature control valves to me are a no no as where does heat go if room temperature shuts off radiators ideal is correctly sizing rads for areas and having more kw rads than boiler output. Other more knowledgable people will know figures.

One suggestion is have valves before and after back boiler to isolate and a drain off point between these and boiler enabling you to drain off boiler for use of fire only in summer on those chilly nights or to warm you up after a cold dip.

[Higgs]

We have a pretty decent sized expansion tank already (I'm not good at judging volume but it must be a good 15 to 20 litres or so).

 

I was wondering if in the (admittedly very unlikely) situation where the TRVs are all left closed, we have a full tank of hot water and the stove cranking away, could there be a risk of boiling even when being pumped? The heat would be moved around the circuit and would not get much of a chance to escape so it would return to the stove still pretty hot. The water would expand into the expansion tank but the heat would still have nowhere to go. This scenario isn't going to happen as we wouldn't leave the stove blasting away with no way to dump the heat.

 

You say that "it is another reason I prefer no obstruction in the thermosyphon system". In the proposed system I sketched out up there ^^^ I have a basic, single emitter thermosyphon system as back up with the rest of the system being pumped. Are you concerned about the reliance of the thermosyphon back up on a mechanical valve controlled by a switch? i.e. if the pump and flow switch were to fail at the same time and there's no natural thermosyphoning around the rest of the system then it would lead to problems?

 

In that case would a better solution be to have a pumped system pushing heat to the back of the boat with a separate, thermosyphoning circuit connecting the living room rad to the back boiler? In which case how can that be connected so that in the event of a pump failure the heat is definitely dumped? Wouldn't it still require some sort of mechanical valve (triggered by flow? temperature?) to switch circuits in the event of pump failure?

 

 

If you start with a basic circuit - One loop of 28mm pipe, from and back to the backboiler with reduced pipe to header. No obstructions and the system runs itself. Then from that you drop in your rads using 28mm 15mm 28mm T joints. The 28mm loop is still there and unobstructed even with rads fully shut off.

 

If you wish to heat a calorifier, it is going to need a pump, but not in the circuit above. Taking off at a junction, and valve controlled for isolating and opening that secondary circuit. If you keep the 28mm loop of the backboiler system open at all times to flow, the only real danger you have is if your boiler is allowed to overheat by uncontrolled stove heating and coolant allowed to heat off. I also would not put a pressure cap on the expansion tank.

Edited August 6, 2014 by Higgs

[smileypete]

Some more thoughts and ideas to ponder over , for a mixed gravity and pumped system I'd do something along these lines:

 

There's also a couple more layouts here:

 

http://www.canalworld.net/forums/index.php?app=galleryℑ=6507

 

Bear in mind they're just layouts, not a complete turnkey design though... but notice there's NO valve whatsoever between backboiler top outlet and header tank vent, this is a very important safety feature. Plus any escaping steam will help and not hinder the thermosyphon.

 

With the thermosyphon, what's important is the increase in height from the backboiler top outlet to the top of the radiator(s), this can make it quite tricky to have a stove at the bow and and gravity rads at the stern.

 

I don't think the angle of incline is tooo important, subject to the above, BUT provided any trapped air can 'self bleed' by rising toward the header tank vent or an auto air vent, and not get trapped in a section of pipe.

 

With the gravity rads, I'd ideally make them 80-100% of backboiler output, but IF a spare pump is carried and/or a large extra firebrick to wall off the back boiler, then 60% of backboiler output may be OK.

 

If there's a decent rad in the saloon near or next to the fire, it does gives the choice of having all the heat from the fire kept in the saloon, if that's what is preferred at times. Without that, any heat from the back boiler has to go to rads elsewhere in the boat or to the calorifier til it's heated.

 

With any pump the outlet should point up so any trapped air can self bleed out. These days there are pumps for £20ish on Ebay or from sellers such as solar project, they seem to get good reports and are cheap enough to carry a spare:

 

http://www.solarproject.co.uk/page2.html

 

With stove size it can be a bit of a personal thing, depends partly on how much window area and whether single or double glazed, how toasty it should be in deepest winter, and whether the stove will run 24/7 or need to heat up boat and calorifier from cold.

 

Sounds like there's the bones of a (failed?) gravity system already in place, so best be a little careful. Having a fully gravity or fully pumped system can be a little tricksy or risky, so why not combine the best of both? I think after looking at the problem from many angles including cost, the best approach may endup looking like what's in the diagram above.

 

cheers, Pete.

~smpt~

Edited August 8, 2014 by smileypete

[ditchcrawler]

 

I have a standard 240 volt pump but I am changing it for one of these

http://shop.solarproject.co.uk/newshop/catalog/product_info.php?cPath=2&products_id=10&osCsid=2ussg9jgl2a05g6saac6kmqoh7

I discussed its use with the company and they are happy with its use in a boat CH system. As yet I have no experience with this type of pump but I know that it has been used by one or two others on here.

Some handy bits on that site.

[DanTheMan]

Hi folks. Im about to update my own ch system and get rid of the noisy wasteful pump once and for all.

 

The diagram above makes perfect sense and should work perfectly. However.. every radiator I've looked at has the inlet/outlet at each side of the bottom of the unit. Not top to bottom as in smileypete's drawing above. So, does this system only work with some specific radiator models?

 

Top to bottom makes sense because as the water cools it sinks downwards, thus drawing more hot in from the top and increasing the circulation.

 

Do these radiators exist and is there a specific name for them? Or, could someone clarify the diagram to show the system with conventional rads.

 

Any help greatly appreciated!

[smileypete]

Pretty much all rads come with 1/2 BSP female tappings, usually 4 in total with 2 at top and bottom on both ends:

 

http://s7g3.scene7.com/is/image/ae235/77236_P

 

Easiest way to connect is with a 1/2" BSP taper male adapter, with a dozen turns of PTFE tape or some Fernox LSX on the male thread.

 

That said, what sort of pump is it? If it's just a 12V pump being too noisy, maybe it could be slowed down using a 12v bulb in series, with thermostat to bypass the bulb and make it run at full power when the stove gets hot enough.

 

Could save redoing the existing system, as finding the best fittings to use can take some working out. Maybe post some photos of what's already there for ideas, or go into a friendly local plumbers merchants when it's quiet for advice.

 

cheers, Pete.

~smpt~

Edited August 13, 2014 by smileypete

[David Mack]

Pretty much all rads come with 1/2 BSP female tappings, usually 4 in total with 2 at top and bottom on both ends:

 

In a conventional pumped system one of the two top tappings has an air bleed fitted and the other a blanking plug. For a gravity system either use diagonally opposite connections and blank off the other two or use all 4 - both top connections to the upper pipe, both bottom connections to the lower pipe. That will increase the flow capacity in and out of the radiators.

 

Don't fit standard radiator valves. Even fully open they have a pretty small orifice, and with a gravity system you need all the flow capacity you can get. If you want to be able to turn off individual radiators better to use a full-flow ball valve.

[DanTheMan]

Thanks Pete. I must just have some cheap radiators without the top inlet.

 

I dont think my pump is overly noisy or faulty. I just can't relax when I hear my battery amps humming away in the night.

I want to overhaul the system in the next month or so. It's badly done, uses ugly white plastic tube that sags down from under the gunnels and the current thermostats need constant adjustment.. especially from cold. Not something I'd trust with a friend if I go away for a weekend. I think you could call the system, at best.. amateur.

 

So thermosyphon is the way forward for me. It also eliminates the risk of draining batteries or a blown fuse boiling the system dry.

 

Thanks again I'm going to have to have a closer look at some other rads.

[smileypete]

Fair do's

 

Could be well worth getting a copy of the BES catalogue or trawling their website at least, to see what bits are available:

 

http://www.bes.co.uk/index.asp

 

Pipe can bent a reasonable amount eg. with an internal bending spring, especially if it's annealed with heat first.

 

Usually rads have a bleed screw at the top which screws into a 1/2 BSP plug, which in turn is screwed in somehow (and often painted over), the rads have got to be bled somehow :

 

http://www.screwfix.com/p/radiator-vent-blanking-plug-chrome-bsp/40972

 

If more rads are needed Screwfix have a small double rad on clearance, not the ideal size but a couple should handle most of the output of a small backboiler. As with stuff like this, if buying check it carefully, then if it's damaged ask for a further discount.

 

http://www.screwfix.com/p/flomasta-type-21-double-panel-single-convector-radiator-white-500-x-500mm/77236

 

Some circulation pumps are cheap enough to be worth buying a spare, only use 6 watts, would best be controlled by a thermostat anyway:

 

http://www.ebay.co.uk/itm/Solar-DC-12V-24V-Hot-Water-Circulation-Pump-Brushless-Motor-Water-Pump-3M-5M-/171368602480

http://www.ebay.co.uk/itm/Solar-12V-24V-DC-Hot-Water-Circulation-Pump-Brushless-Motor-Water-Pump-5M-3M-UK-/271539097190

 

Keep us posted as to how it's going if you want.

 

cheers, Pete.

~smpt~

Edited August 13, 2014 by smileypete

[Jambo]

Apologies for the delay in responding to these last comments (work has been sapping my time and I've not had a chance to think about this until now).

 

I'm going to get my head together with my friend and we'll discuss what might work best in our situation. I think what smileypete's suggested is along the lines of what we were originally trying to come up with but he's filled in some important gaps (he has a knack for doing that!).

 

Thank you one and all. Most helpful as ever.

[Jambo]

So after quite a few delays we've been installing this week. Broadly speaking it's looking like smileypete's suggestion. We've had our fair share of issues but mostly it's gone pretty well and last night we filled the system and lit the fire (with some trepidation).

 

The thermosyphon circuit to the dump rad works really well and the pumped part seems to be working ok back to the bathroom but the plumbing gets a little funky beyond that as the heating system has gone from back boiler to Mikuni (and now back to back boiler) and some of the resulting plumbing could be charitably described as 'bodgey' and will take some attention to get it working as it should.

 

On the whole we're very happy with how it's gone though and these teething troubles will be sorted today or in the near future.

 

But the problem we're having is that that little solar pump, while effective, is pretty noisy. It has quite a whine to it and that whine is transmitted quite effectively by the pipes. We can try using flexibile connections but I think I'd prefer to buy a much quieter pump in general. I guess ideally it would be 12V but we are a mostly 230V boat with an always on inverter so perhaps a normal domestic circulation pump makes sense? Efficiency is always an aim in everything I do so low power consumption is very desirable too. Does anyone have any suggestions?

Edited November 28, 2014 by Jambo

[rupertbear]

 

 

If you start with a basic circuit - One loop of 28mm pipe, from and back to the backboiler with reduced pipe to header. No obstructions and the system runs itself. Then from that you drop in your rads using 28mm 15mm 28mm T joints. The 28mm loop is still there and unobstructed even with rads fully shut off.

 

I have been reading this resurrected thread with interest. A great deal of useful information.

 

The quote above, which suggests a complete loop of supply pipe to and from the back boiler on a gravity system, makes a great deal of sense. When you add radiators, though, won't the majority of the flow take the route through the large diameter pipes rather than through the radiators? I'm thinking of the analogy of balancing parallel radiators.