Calorifier sprung a leak

[alan_fincher]

26 minutes ago, ivan&alice said:

300 GBP was about right to replace this calorifier as-is ? oh well, at least it is the kick up the pants I need to fix my central heating.

A 70 litre single coil with  large bore coil specifically for gravity cost me around £520 2 years ago, (although that price did include an immersion heater and thicker than usual insulation.

I used this company

 

https://www.coppercylinder.co.uk/

[jetzi]

In this picture you can see the hot water outlet. It runs out of the top of the tank, left and left again, vertically down and then to where im holding the pipe and the rest of the boat.

[jetzi]

 

16 minutes ago, stegra said:

That's a prv. It should have a rating on the cap, if you can get in to see it. 

Ah I see, it is rated 3 bar.

 

I assumed that a PRV worked like a gas regulator, regulating the pressure of the hot water system. That's why I thought it needed to be on the hot outlet. This seems like more of a safety device to prevent overpressurisation of the cylinder.

 

Edited November 16, 2019 by ivan&alice

[Chewbacka]

34 minutes ago, ivan&alice said:

Yeah, I'm going to try, but I don't have high hopes. Otherwise I just have to shell out for a new calorifier. My big concern is whether there is something with the system layout that could have caused this other than just age. I don't have any PRVs or NRVs or expansion tanks, the only thing I have is a accumulator tank on the cold water side.

One of the calorifier manufactures specifically states there is no warranty if you do not have an expansion tank because of the increased cyclic stress.  My pump is a 2 bar pump, so with an expansion vessel and a non return valve the calorifier will not go much above 2bar even when the hot  water expands.  So the tank cycles between about 1 bar (when the pump turns itself on) and 2 bar.  Without the expansion vessel it will occasionally cycle between 1bar and 3.5bar (the extra pressure coming from water expansion as it heats up with nowhere to go) which is the pressure release valve release pressure. So much more stress and this will increase the chance of fatigue cracks.  Of course if during the water heating time you draw off hot water from time to time, it will release the excess pressure.  In your case as the tank has a max rating of 2.5bar I would not want a pump pressure much over 1.5bar  and definitely would fit an expansion vessel.  So your tank sounds like it is nearing the end of it’s life (fatigue) so have a go at repairing it, but don’t be surprised if it leaks again, though it may take days or it may be years, difficult to predict.

Edited November 16, 2019 by Chewbacka

[jetzi]

21 minutes ago, alan_fincher said:

A 70 litre single coil with  large bore coil specifically for gravity cost me around £520 2 years ago, (although that price did include an immersion heater and thicker than usual insulation.

I used this company

 

https://www.coppercylinder.co.uk/

Hmm, then that's only an extra 100 GBP or so for the large bore coil - perhaps that is worth it.

 

Do you have your calorifier mounted high off the floor to make the gravity feed work?

 

And is your single coil driving both the rads and the heat source? (engine, eberspaecher, webasto?)

[Alan de Enfield]

2 minutes ago, ivan&alice said:

 

Ah I see, it is rated 3 bar.

 

I assumed that a PRV worked like a gas regulator, regulating the pressure of the hot water system. That's why I thought it needed to be on the hot outlet. This seems like more of a safety device to prevent overpressurisation of the cylinder.

 

 

 

 

 

 

It would appear that a PRV that opens at 3 bar (42 psi) is protecting a calorifier that 'bursts' at 2.5 bar (35 psi)

[jetzi]

1 minute ago, Chewbacka said:

One of the calorifier manufactures specifically states there is no warranty if you do not have an expansion tank because of the increased cyclic stress.  My pump is a 2 bar pump, so with an expansion vessel and a non return valve the calorifier will not go much above 2bar even when the hot  water expands.  So the tank cycles between about 1 bar (when the pump turns itself on) and 2 bar.  Without the expansion vessel it will cycle between 1bar and 3.5bar which is the pressure release valve release pressure. So much more stress and this will increase the chance of fatigue cracks.  In your case as the tank has a max rating of 2.5bar I would not want a pump pressure much over 1.5bar  and definitely would fit an expansion vessel.  So your tank sounds like it is nearing the end of it’s life (fatigue) so have a go at repairing it, but don’t be surprised if it leaks again, though it may take days or it may be years, difficult to predict.

I see - thanks for that. I'll stand the cauliflower in a big bucket until I replace it.

 

It sounds like an expansion vessel is a good idea to add regardless. Just had a quick look on Screwfix and they can be had for 30 odd squids and range in size from 8 to 25 litres.

4 minutes ago, Alan de Enfield said:

It would appear that a PRV that opens at 3 bar (42 psi) is protecting a calorifier that 'bursts' at 2.5 bar (35 psi)

Right - so perhaps this leak was caused by a rise in pressure more than the cauliflower could handle but less than the PRV would vent? I'm wondering why it happened now though - we've had the boat over a year without any issues and presumably the previous owners did too.

[mrsmelly]

Just now, Alan de Enfield said:

 

 

 

 

 

It would appear that a PRV that opens at 3 bar (42 psi) is protecting a calorifier that 'bursts' at 2.5 bar (35 psi)

Yes, quite. I recall that when we first became liveaboards when we had a leaking cauliflower actualy second boat as first didnt have one, or a leaking water pump I used to do various repairs with kits and or solder,  etc etc. Whilst it can extend life I find life is too short and I was always expecting something to fail hence for the last several years we just replace with new. When I bought the present boat it had a slight weep and I started messing about and then thought better of it and replaced the cauliflower,  living aboard aint cheap and never has been.

[Chewbacka]

2 minutes ago, Alan de Enfield said:

 

 

 

 

 

It would appear that a PRV that opens at 3 bar (42 psi) is protecting a calorifier that 'bursts' at 2.5 bar (35 psi)

It doesn’t ‘burst’ at 2.5bar, just that it is designed to have a reasonable life if used at less than 2.5bar.  As you exceed that pressure the fatigue life is progressively reduced.  There will be a safety margin between the safe working pressure and bursting.

3 minutes ago, ivan&alice said:

I see - thanks for that. I'll stand the cauliflower in a big bucket until I replace it.

 

It sounds like an expansion vessel is a good idea to add regardless. Just had a quick look on Screwfix and they can be had for 30 odd squids and range in size from 8 to 25 litres.

Right - so perhaps this leak was caused by a rise in pressure more than the cauliflower could handle but less than the PRV would vent? I'm wondering why it happened now though - we've had the boat over a year without any issues and presumably the previous owners did too.

It is called fatigue cracking, and your tank has been going over pressure from time to time for 20 years, so considering you don’t have an expansion tank it has done well, though it would have lasted longer with the expansion tank.

[Alan de Enfield]

4 minutes ago, ivan&alice said:

Right - so perhaps this leak was caused by a rise in pressure more than the cauliflower could handle but less than the PRV would vent? I'm wondering why it happened now though - we've had the boat over a year without any issues and presumably the previous owners did too.

The seam has been gradually 'ageing' and eventually reached it bursting pressure. ?

 

SWMBO would suggest 'muck in the carburettor'  - she's normally right (or so she tells me).

[stegra]

You could try using epoxy putty instead of soldering. I've used it successfully a couple times where getting to the leak would have involved removing the hot water cylinder. Never had any complaints and once was at my sister's house and she isn't one to hold back. It comes in a sausage that you cut to length and mix with your fingers then smudge on until you've built it up to a decent depth (having thoroughly cleaned the surface with wire wool first). 

 

Search for 'pipe repair putty' or "epoxy putty'. Some of the reviews are pretty poor but that's mostly people trying to use it without draining the system, or even releasing the pressure. Here's an example:

 

https://www.diy.com/departments/evo-stik-putty-50g/191081_BQ.prd?gclsrc=aw.ds&&gclid=Cj0KCQiA2b7uBRDsARIsAEE9XpHyug4n65yhWY_JgP2lJlJNDmtstxUTg4WmK85cC7VZw3IgblaATKcaAlOpEALw_wcB

[Tony Brooks]

4 hours ago, Nightwatch said:

We have recently replaced our cauliflower and installed an extension tank on the hot side. When the hot water is run we gethigh water pressure until the pump kicks in. Then after urging tap off he domestic water pump runs for approximately and up to 20 seconds.

 

Is this because there's too much pressure in the hot side of the system? Can I let out some pressure to regulate?

 

No one seems to have replied to this.

 

No, its the pump compressing the air in the accumulator while building up pressure. It is possible that you have air in the top of the calorifier doing the same thing.

 

The high initial pressure is probably just the pressure stored in the accumulator but if you do not have one then it could also be caused by a higher pressure caused by expansion compressing the air in the calorifier.

 

Basically the pressure in the system is initially dependant on the pump cut out pressure. Once a tap is running and any accumulator has given up its stored water the pressure will be dependant on the difference between the water flow from the pump and that from the tap.

 

Altering the pressure and an accumulator or expansion vessel can only vary the volume of water it can store.

[alan_fincher]

1 hour ago, ivan&alice said:

Hmm, then that's only an extra 100 GBP or so for the large bore coil - perhaps that is worth it.

 

Do you have your calorifier mounted high off the floor to make the gravity feed work?

 

And is your single coil driving both the rads and the heat source? (engine, eberspaecher, webasto?)

 

No, I don't have the tank high up, and in fact I have yet to get around to making it so the existing pump can be bypassed and it runs on gravity.

However the calorifier is towards the front of the boat and the stove towards the back, so everything is right, and gravity should work.

I only have one heat source for it - the solid fuel stove.  I don't have a diesel boiler, and our engine is air cooled, so can't heat the water.

If I had wanted a second heat source, I'd have specified a second coil, (no doubt at extra cost!)

SureJust/SureCal also said they could provide a "special" suited to gravity, but they were far less responsive than "Copper Cylinder" who ended up getting the business.

I was staggered that a custom made cylinder literally arrived with me the day after I had placed the order - most impressive!

A picture, if it is of interest, though now surrounded by a cupboard... (Note use of large expansion vessel)

Edited November 16, 2019 by alan_fincher

[jetzi]

9 minutes ago, alan_fincher said:

the calorifier is towards t  of the boat and the stove towards the back, so everything is right, and gravity should work.

I don't know much about this, but I was thinking of mounting the stove and cauliflower together, with the heat from the stove causing the water to rise into the tank and then circulate down throughout the system.

 

Missing a word but I assume in your system you have put the stove at the front? I can see from just the first rad there that there is a complete circuit that doesn't seem to go anywhere near the stove?

[cuthound]

1 hour ago, ivan&alice said:

I see - thanks for that. I'll stand the cauliflower in a big bucket until I replace it.

 

It sounds like an expansion vessel is a good idea to add regardless. Just had a quick look on Screwfix and they can be had for 30 odd squids and range in size from 8 to 25 litres.

Right - so perhaps this leak was caused by a rise in pressure more than the cauliflower could handle but less than the PRV would vent? I'm wondering why it happened now though - we've had the boat over a year without any issues and presumably the previous owners did too.

 

Make sure you buy an expansion vessel suitable for potable water. They are made of stainless steel rather than mild steel.

 

The mild steel ones are cheaper but are intended for heating systems which should contain a corrosion inhibitor.

[alan_fincher]

30 minutes ago, ivan&alice said:

I don't know much about this, but I was thinking of mounting the stove and cauliflower together, with the heat from the stove causing the water to rise into the tank and then circulate down throughout the system.

 

Missing a word but I assume in your system you have put the stove at the front? I can see from just the first rad there that there is a complete circuit that doesn't seem to go anywhere near the stove?

No, the stove is at the back end of the main cabin, (connected by the pipes running off to the right of the picture).

The calorifier is near the front of the boat, (so "uphill from the stove), though not at the very front.

The pipes running forward through the blue bulkhead supply just one further radiator in the bedroom at the very front of the cabin.  At least 30 feet between stove and calorifier, actually.

I did contemplate trying to locate the calorifier very close to the stove, so the gravity circuit would have been very short, but in the end I decided I just didn't have the space to do this without some major upheaval.  Probably worth doing if you can though.

You are correct that one circuit is supplying both the radiators and the calorifier, though either can be turned off - you will notice some gate valves on the supply to the calorifier coil - shut one of these, and only the rads are heated.   They are also useful if the stove isn't being run, but the calorifier is being heated by immersion, as they can be closed to stop the calorifier trying to heat the radiators, (which does happen to some degree if they are both left open).

 

[jetzi]

1 hour ago, alan_fincher said:

At least 30 feet between stove and calorifier, actually.

I did contemplate trying to locate the calorifier very close to the stove, so the gravity circuit would have been very short, but in the end I decided I just didn't have the space to do this without some major upheaval.  Probably worth doing if you can though.

Ok, interesting. Really I need a third coil for what I want to do! Perhaps if I put the calorifier on one circuit, and the backboiler, Eberspaecher and rads on the other, with taps to turn the Eber off when the backboiler is in operation. Actually I probably wouldn't even need taps since the Eber's pump would prevent hot water from flowing that way.

[Jen-in-Wellies]

10 hours ago, ivan&alice said:

Ok, interesting. Really I need a third coil for what I want to do! Perhaps if I put the calorifier on one circuit, and the backboiler, Eberspaecher and rads on the other, with taps to turn the Eber off when the backboiler is in operation. Actually I probably wouldn't even need taps since the Eber's pump would prevent hot water from flowing that way.

If you get a custom made cauliflower, then a third coil is easily done for a few more £.

 

If/when you do replace the existing cauliflower, then you need to match the pressures of each of the components.

Water pump shut off pressure needs to be less than the opening pressure of the PRV.

PRV opening pressure should be less than, or at most equal to the max operating pressure of the cauliflower.

There are recommendations for the pressures you pump up the accumulator on the cold water side and the expansion vessel on the hot water side, but I can't remember what they are!

 

As an aside, if copper calorifiers are soldered togehter, then solder suffers from something called low cycle fatigue. If a joint is stressed, then the solder creeps to accommodate the stress over time. When the stress is reversed, then it creeps the other way. Eventually cracks in the solder form. The stresses in a calorifier heating up and cooling down and being overstressed from the high pressures from a missing expansion vessel are perfect for causing this. My first cauliflower didn't have an expansion vessel and failed at the top seam like yours.

 

Jen

Edited November 17, 2019 by Jen-in-Wellies

[jetzi]

To report back, I have temporarily fixed the leak.

I couldn't get hold of any copper nails, and I soon realised the reason why they are needed when I started soldering. Despite a lot of prep and flux, there was no way I was getting solder to adhere to that surface, and there was no way the solder would be drawn into such a fine crack. The cylinder acted like a large heatsink and it was very, very difficult to get the area to the right temperature to melt the solder. After a fiver's worth of solder was piled up on the floor I switched to Plan B. This was a tube of "external leak sealer" which was had from Screwfix for the princely sum of 3.15 GBP. I then put a bicycle inner tube around the whole seam and tightened up a ratchet strap. So far, so good.

 

 

Once I cleaned the surface, it became clear that it was not a pinhole, but actually a small crack about an inch long. And it was not on the seam, but rather close to it. So unfortunately I think this means a new hot water cylinder is needed. At least this means I can re-design the system and get one custom made to suit my needs. Hopefully my fix will last until then.

Also, on a point of order, people often talk about the cylinder as the "calorifier". Is this correct? Surely the "calorifier" is the circuit that runs in a coil in the hot water tank, through to the engine and back?

 

One last question - how does the system "know" when the water is up to 85 degrees and to start using the skin tanks for cooling?

[rusty69]

12 minutes ago, ivan&alice said:

One last question - how does the system "know" when the water is up to 85 degrees and to start using the skin tanks for cooling?

Hey, well done.Good bodge (I hope). I was gonna suggest some sort of epoxy (milliput) yesterday, but thought you probably wouldn't wanna bodge it thaaat much,so didn't suggest it.

 

The calorifier circuit is probably installed before the engine thermostat, so it will heat the coil before the skin tank.

Edited November 17, 2019 by rusty69

[jetzi]

Just now, rusty69 said:

Hey, well donr.Good bodge (I hope). I was gonna suggest some sort of epoxy (milliput) yesterday, but thought you probably wouldn't wanna bodge it thaaat much.

 

The calorifier circuit is probably installed before the engine thermostat, so it will heat the coil before the skin tank.

I certainly didn't want to bodge it Messing around with the solder was so unbelievably frustrating that I was very glad I knew myself well enough to have bought some of the gunk as a backup plan. Good bodge so far.

[Tony Brooks]

26 minutes ago, rusty69 said:

Hey, well done.Good bodge (I hope). I was gonna suggest some sort of epoxy (milliput) yesterday, but thought you probably wouldn't wanna bodge it thaaat much,so didn't suggest it.

 

The calorifier circuit is probably installed before the engine thermostat, so it will heat the coil before the skin tank.

 

On most boats the calorifier circuit is in parallel with the thermostat-skin tank circuit but will still get hot while the thermostat is closed.

25 minutes ago, ivan&alice said:

I certainly didn't want to bodge it Messing around with the solder was so unbelievably frustrating that I was very glad I knew myself well enough to have bought some of the gunk as a backup plan. Good bodge so far.

Did you drain the calorifier to at least an inch or so below the crack, if not it was the water acting as a heat sink.

 

Soldering things like copper tanks need a very large and very hot iron, toy electronic type irons are no good at all.

 

I would use active plumbers flux for this type of job as used with Yorkshire/end feed pipe fittings.

 

 

[jetzi]

1 hour ago, Tony Brooks said:

On most boats the calorifier circuit is in parallel with the thermostat-skin tank circuit but will still get hot while the thermostat is closed.

Did you drain the calorifier to at least an inch or so below the crack, if not it was the water acting as a heat sink.

 

Soldering things like copper tanks need a very large and very hot iron, toy electronic type irons are no good at all.

 

I would use active plumbers flux for this type of job as used with Yorkshire/end feed pipe fittings.

Thanks Tony. Yes I think I did drain the tank, at least I tried to - I took all the pressure off the hot water system and then I released the drain valve at the botto of the tank. I was surprised that only a couple of litres came out before the drain slowed to a dribble. That didn't make a lot of sense to me. Of course I couldn't really see how much water was in the tank being that it is made of copper

 

I used a propane blowtorch to do the soldering along with flux. The tank did get hot enough to melt the solder but it cooled off so quickly that the solder simply wouldn't adhere. If it had just been a pinpoint I could have managed to drop a blob on it but a whole inch was impossible with my level of (lack of) skill. If I was going to try again I would have got a brazing torch and tried to seal it that way, but perhaps it's for the best I didn't because with the current bodge job I can have a hot shower tonight.

 

Thanks for all the advice. Now I'll get on designing the gravity central heating system. When I have a plan for that I'll be sure to start a thread to get some input

23 hours ago, alan_fincher said:

A picture, if it is of interest, though now surrounded by a cupboard... (Note use of large expansion vessel)

One last question about this - I see you're using copper pipes here @alan_fincher. And my tank is also copper. My pipes are all plastic though. Don't we tend to avoid using copper on boats because of the vibration and work hardening? Couldn't that be the cause of the split in my tank?

[Tony Brooks]

How did you let air into the calorifier when you tried to drain it? For future reference lodge the PRV into a mid position (held open) between clicks. I don't think you drained it enough. Even a blow lamp should have heated the copper enough.

[alan_fincher]

11 minutes ago, ivan&alice said:

One last question about this - I see you're using copper pipes here @alan_fincher. And my tank is also copper. My pipes are all plastic though. Don't we tend to avoid using copper on boats because of the vibration and work hardening? Couldn't that be the cause of the split in my tank?

I don't think vibration and work hardening are a significant issue with using well secured copper pipes and fittings on canal boats.

To me the bigger issue with copper, if it is on a boat that is left unattended in winter, are to do with temperatures and freezing.  Copper pipes and fittings left with water in, (assuming no anti-freeze), seem far more susceptible to problems than the same thing done in push-fit plastic.

So my now strong preference is to plumb in plastic for most of the boat.  However what you are seeing in my pictures in copper is all part of the heating circuit connected from boiler to rads and calorifier coil  This is al protected permanently by anti-freeze, so there are no worries about it freezing up.

Because I want to switch to gravity, and with long runs involved, the main pipe s all 28mm.  I've ever investigated if this is available as push-fit plastic, but I assume the use of more than 22mm would be fairly unusual.  As the copper pipes are visible through the whole length of the boat, I think it will look far more presentable than trying to make plastic plumbing look so.  I understand plastic plumbing used on central heating circuits does tend to droop between the supports as it gets hot.  Also in the main cabin connectors will all be capillary or end fed - far eater than any plastic offerings I've come across.  They might even get polished to make a bit of a feature of them.