Pinpointing a very slow leak associated with a calorifier.

[alan_fincher]

Some months back I had a custom made vertical calorifier supplied for Flamingo, which I installed, and works well, (massively better than what it replaced).

 

However for a while I have been aware of a damp patch on the plywood floor it is standing on.  Not large, maybe 6" across, but definitely damp and never drying up.  This patch was at the "downhill" end of the boat, but is also immediately below three different connections to the calorifier.  Working from bottom up these are cold feed in, return from coil & feed to coil. There is only one coil, (our engine is air cooled, so can't heat water), but the coil is large bore designed in the hope it will work with gravity from the solid fuel stove, if I modify the system to make the pump optional, (still to try!).  The foam lagging is extra thick at 75mm, so it is very hard to inspect what is going on near the copper itself.

 

Until today I could find no trace of any leak at any of these 3 connections, and assumed the damp patch might relate instead to a leak somewhere underneath.  However I can now find the occasional drop of water either gathering on the lagging near the lower 2 connections, or having run down below any connection over the insulation to floor level.  Whatever the leak is it is very slow - sitting waiting for a drip never seems to produce one, but go back half an hour later and there may be a drop or two.

 

At first I thought it was the bottom connection, (cold feed in) but as I have spotted drips at a level above this I think it must be from a calorifier coil connection. However I have only seen water below the lower calorifier coil connection, but it is the upper one that maybe shows some suggestion of a leak, (green staining if you look past the lagging to where the coil passes through the cylinder, and is secured to it by a large nut).

 

At the moment I have no idea where the leak is coming from, whether it is a manufacturing fault, whether it is on my plumbing, or whether I have damaged anything.  Obviously I don't wan't to start cutting away lagging on an expensive cylinder that's under warranty, but I don't have the evidence to prove a manufacturing defect.

 

Does anybody have any suggestions for pinpointing such a slow leak please, (and I'm confident it is very slow). 

If it were leaking from the highest coil connection, I would expect to see the leakage running down the outside of the foam insulation from that point, and there is zero evidence of that.  I would have though the foam was well stuck to the tank, and to have closed cells, so would not expect a leak at the top end of the coil to be running out at the bottom.  Am I wrong though?  Is that a possibility, please?

[Graham Davis]

Get some cheap talcum powder and shake that over the area.

[Alan de Enfield]

Put a wrap of kitchen roll around each of the joints (tape it down if necessary), wait for hours / days - if it is a joint leaking the highest one that is damp will be the one leaking. (unless more than one is leaking)

[Bee]

I think you may just have to cut away some foam, it won't make any difference to the tanks efficiency, just stick some more foam back later. I had to cut foam away from a blanked off immersion heater opening, took many attempts and different gaskets to fix it. Can't think there's any other way really.

[Boater Sam]

The water will run behind the foam. It could even be from the top hot connection on the dome. If you find that it is the securing nut for the coil, I would invite the maker to inspect rather than trying to cure the drip by tightening the nut, the gasket is inside the calorifier and if you twist the connection even slightly it may well get worse rather than better.

Talc is a good way of seeing where it is coming from.

[Stilllearning]

I suspect both Bee and Sam have it about right, unfortunately. Is it possible to send a brief video of the leakette to the manufacturer?

[bizzard]

A bit like an internal hemorrhage, lets hope it gets better soon.

Edited October 1, 2018 by bizzard

[alan_fincher]

13 hours ago, Graham Davis said:

Get some cheap talcum powder and shake that over the area.

 

Yes, worth try I guess - I don't have anything like that here at the moment, so next trip I guess.

 

4 hours ago, Alan de Enfield said:

Put a wrap of kitchen roll around each of the joints (tape it down if necessary), wait for hours / days - if it is a joint leaking the highest one that is damp will be the one leaking. (unless more than one is leaking)

 

Yes, I'm trying that.  I ought to be conclusive, but results seem inconsistent.  I'm starting to wonder if there are at least 2 very small dribbles.

 

4 hours ago, Bee said:

I think you may just have to cut away some foam, it won't make any difference to the tanks efficiency, just stick some more foam back later. I had to cut foam away from a blanked off immersion heater opening, took many attempts and different gaskets to fix it. Can't think there's any other way really.

 

Cutting lagging away certainly might help, but having started a conversation with the supplier and manufacturer as it's under warranty I'll not "modify" anytthing unless thet agree.

2 hours ago, Boater Sam said:

The water will run behind the foam. It could even be from the top hot connection on the dome. If you find that it is the securing nut for the coil, I would invite the maker to inspect rather than trying to cure the drip by tightening the nut, the gasket is inside the calorifier and if you twist the connection even slightly it may well get worse rather than better.

Talc is a good way of seeing where it is coming from.

 

The only way I can see to attempt to tighten the nuts securing the coil would be with some kind of massive box spanner or long reach socket that I don't have.  It's 1" BSP, I think so fairly huge.  I agree trying to shift them might make things worse not better.  I'm surprised water can run between the foam and the copper, I would have thought it pretty well adhered all over, but I agree the evidence is that it may well be.

 

2 hours ago, Stilllearning said:

I suspect both Bee and Sam have it about right, unfortunately. Is it possible to send a brief video of the leakette to the manufacturer?

 

It would be a more boring video than watching paint dry.  After several hours there is a good chance that any single drop of water will not finally appear at the same point as the last one did.  It really is a very slow leak at the moment.

 

I was supposed to be reverting to the supplier, but can't conclusively nail this as a manufacturing defect rather than less than perfect plumbing by me.  I guess I'll talk to them again anyway, but I see a real prospect that if I uninstall it and somehow courier it to them, (and I have no original packing), that they will declare it as no fault found.

 

Thankyou to everybody who has given their helpful thoughts so far.

Edited October 1, 2018 by alan_fincher

[Phil.]

1 hour ago, alan_fincher said:

 

Yes, worth try I guess - I don't have anything like that here at the moment, so next trip I guess.

 

 

Yes, I'm trying that.  I ought to be conclusive, but results seem inconsistent.  I'm starting to wonder if there are at least 2 very small dribbles.

 

 

Cutting lagging away certainly might help, but having started a conversation with the supplier and manufacturer as it's under warranty I'll not "modify" anytthing unless thet agree.

 

The only way I can see to attempt to tighten the nuts securing the coil would be with some kind of massive box spanner or long reach socket that I don't have.  It's 1" BSP, I think so fairly huge.  I agree trying to shift them might make things worse not better.  I'm surprised water can run between the foam and the copper, I would have thought it pretty well adhered all over, but I agree the evidence is that it may well be.

 

 

It would be a more boring video than watching paint dry.  After several hours there is a good chance that any single drop of water will not finally appear at the same point as the last one did.  It really is a very slow leak at the moment.

 

I was supposed to be reverting to the supplier, but can't conclusively nail this as a manufacturing defect rather than less than perfect plumbing by me.  I guess I'll talk to them again anyway, but I see a real prospect that if I uninstall it and somehow courier it to them, (and I have no original packing), that they will declare it as no fault found.

 

Thankyou to everybody who has given their helpful thoughts so far.

You could always try adding bright dye or uv leak detection dye. Some can't be injested so not all suitable for potable water, but there are some on the market that are. Just don't go to a disco if you have showered in the treated water, you'll glow.

[dor]

2 hours ago, alan_fincher said:

 

The only way I can see to attempt to tighten the nuts securing the coil would be with some kind of massive box spanner or long reach socket that I don't have.  It's 1" BSP, I think so fairly huge.  I agree trying to shift them might make things worse not better.  I'm surprised water can run between the foam and the copper, I would have thought it pretty well adhered all over, but I agree the evidence is that it may well be.

 

My hot water tank at home had a slight weep and it took me a long time to track it down.  Eventually found a pinhole under the insulation and it had been running down the inside of the insulation foam.

[MtB]

It might be helpful to narrow down the leak as originating from the domestic hot water or from the heating circuit water. Try turning the pump OFF and opening a hot tap and see if the drip stops forming. 

[Tony Brooks]

AS the heating water should have antifreeze in it, as would the cooling water, a very small dab on a figure and touch it ion your tonge. If it tastes   bitter or sweet it has antifreeze in it.

[OldGoat]

I had a similar situation some years ago  - little dribbles - then a catastrophic leak while we had friends on board. Hacking away at the insulation discovered that the top dome had corroded through the tank body and the join was paper thin.

My guests (city folk) were amazed at my 'knowledge' and attempts to solder the two parts (pointless) and attempts to source fittings locally to fix it. (My thought s about a so called 'marine / canal quality' so called heavy duty calorifier were less polite). It came from a very nice, friendly chandlery on the T&M, but that's another story.

 

I wonder if yours may have the same affliction? Calorifiers do not always take kindly to constant flexing - especially if they're really not much more than a domestic type.   

[MtB]

4 hours ago, OldGoat said:

I wonder if yours may have the same affliction? Calorifiers do not always take kindly to constant flexing - especially if they're really not much more than a domestic type. 

 

Or more specifically, they work harden and crack if the pump is fitted without an accumulator. Regardless of whether the pump manufacturer says their pump needs one. 

[koukouvagia]

19 hours ago, alan_fincher said:

 

 I'm surprised water can run between the foam and the copper.

 

When I removed a leaking calorifier I took off the foam lagging.  Only about half the lagging had adhered to the copper, leaving a gap between the insulation and the metal.  I don't know whether the foam had shrunk over the years or whether it was manufactured like that.

[alan_fincher]

12 hours ago, OldGoat said:

I wonder if yours may have the same affliction? Calorifiers do not always take kindly to constant flexing - especially if they're really not much more than a domestic type.   

 

8 hours ago, Mike the Boilerman said:

Or more specifically, they work harden and crack if the pump is fitted without an accumulator. Regardless of whether the pump manufacturer says their pump needs one. 

 

This was a made to order marine calorifier from a supplier well recommended on here.  It is still less than a year old, but because we don't run the solid fuel stove except when it is cold, has spent most of its installed life isolated while we are actually using a Morco.

 

It is pressure tested to 5 bar by the manufacturer, has an 8 litre expansion vessel attached as well as there being an accumulator at the pump end, and the PRV is only 3 bar.  A gauge on the PRV indicates the pressure seldom at anything more than 2 bar.

 

I would be disappointed if the calorifier, (which I have paid for a 5 year warranty on), is in physical distress after so little use.

I still haven't nailed the problem.  I made a perhaps unwise decision to further tighten any joints I have made in copper with compression fittings somewhat tighter than I normally would.  That seemed to provoke a couple into a slow leak, so I assumed I had overtightened and crushed olives, but when I took apart, it all looked pristine.  I put it back together again, and I'm in a "no apparent leak" situation again - but I thought I had been there before.

 

I'm trying to avoid resealing any olive based joints with Fernox LS-X in them, as it always seems a right bodge to me.  However past experience suggests to me it might be worth a try, as it generally stops a dribbling leak dead.  Particularly as most of the joints involved are on the pipework that is not being pressurised - i.e. on the central heating loop from the multi-fuel stove.

2 minutes ago, koukouvagia said:

When I removed a leaking calorifier I took off the foam lagging.  Only about half the lagging had adhered to the copper, leaving a gap between the insulation and the metal.  I don't know whether the foam had shrunk over the years or whether it was manufactured like that.

 

Yes, There is every evidence that some water may have leaved at the upper coil connection, but is actually emerging where the lower coil connection passes through the lagging.  I think my assumption that they come with it well stuck to the copper all over is probably a duff one.

 

 

[Tacet]

2 hours ago, alan_fincher said:

 

This was a made to order marine calorifier from a supplier well recommended on here.  It is still less than a year old, but because we don't run the solid fuel stove except when it is cold, has spent most of its installed life isolated while we are actually using a Morco.

 

 

It's not a repair as such, but very small weeps on the hot side sometimes seal when the chalk and crystals form through evaporation of the water.   Your limited use of the cylinder to store hot water might mean you can't wait long enough to test your luck. 

 

 

[blackrose]

If you cut away a small amount of foam and it's saturated then you have your evidence - assuming you've checked the plumbing joints and it's not them. If you can get a bit of kitchen roll around the joints and they're dry then you know it's the tank and I then wouldn't waste any more of your time pinpointing the leak. Just get the calorifier out and return it from whence it came complete with saturated foam. It's up to them to locate the leak and either get the manufacturer to fix it or replace the tank. 

Edited October 3, 2018 by blackrose

[MtB]

8 minutes ago, blackrose said:

If you cut away a small amount of foam and it's saturated then you have your evidence - assuming you've checked the plumbing joints and it's not them. If you can get a bit of kitchen roll around the joints and they're dry then you know it's the tank and I then wouldn't waste any more of your time pinpointing the leak. Just get the calorifier out and return it from whence it came complete with saturated foam. It's up to them to locate the leak and either get the manufacturer to fix it or replace the tank. 

 

Being ever the pragmatist myself I take the opposite view. It is FAR more likely to be 'installer error' even if Alan can't find the leak, and I bet they just return the same calorifier saying they factory-tested it and its fine.

 

In addition, this leak is microscopic and really doesn't actually matter in the big scheme of things. Most really tiny leaks will (and, I bet, including this one) 'self heal' once the calorifier is actually in use and hot. As said earlier in the thread, the calcium in the escaping and evaporating water will bung the hole up. 

 

 

[blackrose]

Having assisted a neighbour remove a leaking tank that had been leaking for a while, that's not true in my experience Mike. It was a manufacturing defect. Leaking seam.

 

I accept that's only one example, but the calorifiers installed on most canal boats aren't really manufactured to marine standards and the quality can be poor. Go to a boat show and look at some proper marine calorifiers and the difference is clear. 

 

I also accept that installer error is far more common, which is why I suggested checking all the joints first. 

Edited October 3, 2018 by blackrose

[alan_fincher]

6 hours ago, blackrose said:

I also accept that installer error is far more common, which is why I suggested checking all the joints first. 

 

But take the case of my two coil connections......

 

The two threaded connections on either end of the coil pass though the calorifier wall, and are secured by flanged nuts on the outside, (presumably with some kid of fibrous washed providing the seal).

 

To connect to the projecting male threads, I have had to use a pair of  1" BSP threaded female fittings that reduce to 22mm compression.  The joint between the adaptor and the thread on the coil is hopefully sealed by ptfe tape - the connection to the 22mm plumbing relies on compression oloives.

The problem is that both the nuts securing the coil and the screw threads on my added adaptors are virtually next to each other, and down a hole, surrounded by 3"on foam insulation.  If either of these joints s leaking, it is virtually impossible to determine which.  Any tissue paper inserted down the gap that comes out damp could be being made wet by either source, or even a leak higher up on the tank if water can run down any gap that may exist between the foam and the copper.

 

I agree wkith MTB that there is a reasonable chance this is my error, and there may not be any manufacturing error.  it will be an enormous effort to uninstall this and ship it without damage back to the supplier.  That's why I'm doing everything I can, (including seeking advice here) to rule out something I have got wrong, before embarking on that course.

[springy]

Is it possible to drain the calorifier & then pressurise with air (being careful not to over pressurise) - you may then be able to use a liquid leak detector by bubbles to locate the actual leak ?

 

springy  

[MtB]

45 minutes ago, alan_fincher said:

To connect to the projecting male threads, I have had to use a pair of  1" BSP threaded female fittings that reduce to 22mm compression. 

 

Ah, that's the wrong fitting. The right thing to use is called a "cylinder union" and it like a giant tap connector. One of these (or the straight version):

 

 

http://www.teslauk.com/product/7581/end-feed-cylinder-unions

 

Winding on a female copper-to-iron adaptor seriously risks twisting the copper coil inside which apart from damaging it, can loosen the backnuts holding it in place through the cylinder wall, which have fibre washers on the inside of the tank. I wonder if this IS your leak....

[alan_fincher]

23 minutes ago, Mike the Boilerman said:

 

Ah, that's the wrong fitting. The right thing to use is called a "cylinder union" and it like a giant tap connector. One of these (or the straight version):

 

 

http://www.teslauk.com/product/7581/end-feed-cylinder-unions

 

Winding on a female copper-to-iron adaptor seriously risks twisting the copper coil inside which apart from damaging it, can loosen the backnuts holding it in place through the cylinder wall, which have fibre washers on the inside of the tank. I wonder if this IS your leak....

 

Damn - so do I, now!

[MtB]

29 minutes ago, alan_fincher said:

 

Damn - so do I, now!

 

If it IS your leak, tightening the exterior backnuts on the coil connections will probably fix it. 

 

To do this, you need a "Monument wrench"...