Copper cylinder bursts frequently

esquilax

Hi folks

My parents' house contains a gravity fed open vented heating system. Originally it was just a solid fuel stove and a cylinder for hot water. Over the years this was then followed by the installation of radiators, and later followed by the addition of a diesel boiler. The stove can heat hot water and to some extent, by switching on a circulating pump, the radiators. The boiler heats both. It's a single coil cylinder and there are no valves, electric or otherwise.

There are numerous things I don't like about the whole set up and at some point I'll draw a diagram of the system and see what people here have to say about it. In the meantime there's one specific problem I wanted to see if anyone was familiar with: the copper cylinder springing leaks ridiculously frequently.

At first before copper got as expensive as it is now they used to replace the cylinder, these days they remove it, get the leak (generally on a seam) braised and reinstall it. We joke about it being an annual event but this time it's actually happened twice in three months, prompting me to try and get to the bottom of it once and for all.

I know many houses with copper cylinders that have been in place for years without once ever springing a leak. My father reckon the current one is at least his tenth, and I have personally been involved in its removal and repair at least five times.

I can only presume it's something to do with excessive pressure on the cylinder (which has a sticker saying it was tested to 20 bar incidentally) but if anyone has any ideas on what kind of oversight on the installer's part could cause it, please tell me.

If there are any specific aspects of the system you'd like me to report back on please ask.

Many thanks
Gavin

ART6

Are you sure the label says 20 bar? That's 300 psi and a lot of pressure!

When you say it leaks at seams, do you mean seams that join the top and bottom to the cylinder drum, or where the pipes enter? If it's gravity fed there should only be the head pressure of the attic water tank on it -- two or three psi perhaps. Is there a proper vent pipe back to the cold water tank, and can you blow air back down it without much resistance?

I suppose (a wild theory but---) if there was something like a non-return valve in the cold feed from the tank and either a blocked vent pipe or no pipe at all, then as the cylinder contents heated it would attempt to expand, increasing the internal pressure a lot. Way of checking that would be to not run off any hot water until the cylinder has been standing for an hour or two, then open a hot tap. If it squirts hot water at a pressure that drops off almost immediately, then the cylinder is being over presurised. Highly unlikely I would think, but ten cylinders replaced??:eek:

items

Must be your water, correct me if I'm wrong but you prob have a fresh water well? If I'm wrong and your off a mains supply their must be others on same line experiencing same problem.

I've seen a few cases of whats know as pin hole. Some cylinders pin right through copper, others pin at cylinder seem. Most cases pin holes can be fixed by using either so low temp solder or high temp solder but once a cylinder pin holes more tend to follow so best replace cylinder. If your cylinder is regularly springing leaks you should have water analyzed and have a filter fitted to suit, you might already have a water softener, maybe quality of water has changed since softener first installed.

Worst case you could replace cylinder for a stainless steel cylinder, most stainless cylinders are designed for un vented use, I am sure with a bit of imagination and research its possible to fit a stainless steel cylinder into a gravity system, their may well be cylinders available, never fitted one for gravity use myself, first time hearing about a cylinder replaced 10 times so just thinking this up.

Have a read over this, should give you some idea on whats happening,
http://www.bsee.co.uk/news/fullstory.php/aid/2321/Avoiding_corrosion.html

esquilax

ART6 and Items: very belated thanks for your help on this. I got sidetracked by a more serious problem involving a leaky roof and forgot to get back to you.

Firstly, ART6, you're right, the sticker on the cylinder says it was tested to a more plausible 20 PSI, not 20 Bar as stated above. There are no non-return valves in our system and the pipe to the expansion tank in the loft is clear and functional (I didn't have the pleasure of blowing through it but I have heard water spitting out of it occasionally). So it should be physically impossible for that cylinder to be under any pressure beyond normal atmospheric pressure. On this particular cylinder the leaks have all occurred around the seam where the cylinder drum is welded to the top. They start as pinholes and get bigger until at some point we notice a wet hot press.

items: thanks for the link - an interesting read but none of the things they mention stand out as applying to our problem. We did used to have well water in our hot water system, which was very limey and the first cylinders removed were covered in a thick crust of it. Rather than get into water softening, about 15 years back we started capturing rainwater and using that for all water in the house except the kitchen cold tap.

That either one of the well water or rain water could be agressive to copper seems plausible, but to have the problem with both seems like extraordinarily bad luck. That said, I would imagine the rainwater would be slightly acidic and I've never actually tested it.

However, if the water is corroding the copper, why is it always the cylinder that leaks? Shouldn't some piece of copper pipework (none of which has ever been replaced) have sprung a leak in the time we've been through all those cylinders? The article implies that it should.

Anyway I like the stainless steel cylinder idea and will probably go with that as part of a general overhaul of the system. We have another issue with the pipes banging when there is a good fire lit in the stove, and my father always thought this was related in some way to the bursting cylinder. I can't see how it could be and I'll post a separate thread about it to get your thoughts on that too.

Edit: here is that thread:

Thanks again
Gavin

10-10-20

Gavin, do you have green staining from the taps in sinks? How old is the heating system in the house? Is it gun-metal piping throughout?

esquilax

10-10-20 wrote: »

Gavin, do you have green staining from the taps in sinks? How old is the heating system in the house? Is it gun-metal piping throughout?

The only place we get a bit of a green stain is in the bathroom sink. I'm not sure why we don't get it in the bath or toilet, they're all the same water.

The system piping is all copper apart from four 28mm tee pieces which for some reason are gunmetal.

10-10-20

Have a look at this:
http://www.awa.asn.au/AM/Template.cfm?Section=Copper_Blue_Water_in_Pipes

* Water from rainwater tanks has very little of the salts (hardness) in it that mains water often does, so it is actually more likely to corrode copper than mains water is. Once again, though, this is not predictable.

Out of interest, where are your patents located?

esquilax

10-10-20 wrote: »

Have a look at this:
http://www.awa.asn.au/AM/Template.cfm?Section=Copper_Blue_Water_in_Pipes

* Water from rainwater tanks has very little of the salts (hardness) in it that mains water often does, so it is actually more likely to corrode copper than mains water is. Once again, though, this is not predictable.

Out of interest, where are your patents located?

That's an interesting one, thanks. There is a blue stain under the cold tap but not the hot tap so they're right about that. I always figured it was just because the cold tap was left running for longer.

My folks are in Durrow, Co. Offaly. Site of a high cross that was being very badly weathered by acid rain and had to be taken indoors for conservation, FYI...

10-10-20

It could be sulphur.
If Tullamore is in the direction of the prevaling wind from a coal powered generating station, then I could see a link. The prevaling wind in Ireland is between South and West. Tullamore is also between two bogs, so whether there is a link there, I could only guess.

10-10-20

Ok... here's the potential link:
http://www.brdisolutions.com/pdfs/bcota/abstracts/4/z229.pdf

So, I propose that you have sulphurous rain either as a result of the burning of peat in the locality, or even from the abode itself. This rain is being used to provide water for the house (as documented by yourself) and that this is creating copper sulphide when reacting with the piping.

What I don't understand is, why is the cylinder being corroded, but the cold tap shows the signs of the deposition of copper... I'm sure I read the reasons for that in the past though.

And finally... Zoom out to see the location of Tullamore between the two significant areas of bog lands, shown in red/brown.

http://maps.google.com/maps?hl=en&safe=off&client=firefox-a&hs=5r&rls=org.mozilla:en-US:official&oq=&gs_rfai=&um=1&ie=UTF-8&q=tullamore&fb=1&gl=ie&hq=tullamore&hnear=Dundrum&ei=krKmS7_qJpq80gTOwIn8CQ&sa=X&oi=local_group&ct=image&resnum=4&ved=0CCgQtgMwAw

aah yes

Esquilax, see thread below ...

http://www.boards.ie/vbulletin/showthread.php?t=2055852183

I would bet several hundred euros, the repeated condition of the cylinders you describe is that of slightly aggressive water from either rainwater or well sources usually ranging from the 5.5 pH to 6.5 pH range.

This is an area where I specialise and other qualified water treatment engineers, - it can be corrected no problem with a whole house pH correction system. You will never have a compromised cylinder again or green / blue stains again, with a correctly specified pH unit fitted.

A thread in a specialist US water forum gives a handy insight into pH and effects on plumbing ...

http://www.watertechonline.com/detail.asp?Board_ID=7&Parent_ID=139277#139277

This thread was about RO filter water which is also slightly aggressive, similar to rainwater.

aah yes

Just a quick mention to confirm all RO systems should have all-plastic manufacture and work fine at reduced pH values, - the treated RO water can also be raised in pH afterwards if required above 7 pH to 8 pH if required similar to bottled water values.

Perrier bottled water has a pH of 5.9, so pH is not a worry when consumed in many beverages as low as 3 or 4 PH such as Coca Cola and orange juice, it is just a concern for pipes.

esquilax

So it's really looking like acid rain is the culprit then - thanks everyone for the input.

gws: the rainwater system was cobbled together for practically nothing (from 10x 1000L plastic tanks, run up to the house in plastic pipe - so we got that much right by coincidence)

A system to buffer the pH is an interesting idea and good to know such a thing exists. What complicates the decision is that the stainless steel cylinder would solve other problems too and in fact might end up being required in future anyway.

Having the water tested is more expensive than I imagined (e.g. €85 from these people) although there's also this which I think would achieve the same result. Either way, it'd be good to know just how acidic the rainwater actually is. Maybe we're being slowly dissolved just by showering in it

aah yes

For testing I would recommend any HSE - INAB certified lab, although there are a range of good private labs.

If you are testing one parameter such as pH, then ask for testing around 10 euros from a certified lab, or send two samples to two separate ones to confirm consistent, accurate results from both. (Taking into account we are looking to avoid thousands of euros of damage long term.)

For a really accurate investigation into the nature of your water condition regarding agressive water and copper leaching, test for the following ...

1) Copper

2) pH

3) Total Alkalinity

4) Total Hardness

5) Get your own idea of the stored water temperature.


From the results, an accurate Langelier Saturation Index can be calculated to measure true pH potential, plus the copper exceedance will give an idea as to copper levels being lost from plumbing.

The ideal LSI is "0" but not much more than -0.1 or +0.1, also pH is safer at levels above 7 towards 8 pH.

Do not pay 85 euros for a rainwater test for a whole range of parameters that are not applicable to rainwater.

Low cost pH meters will not guarantee any real accuracy. Between 100 to 1000 euros budget the better electronic pH meters still take some care and calibration to get any real accuracy or consistency with pH results.

Google Merck for pH strips sold in the pH 6.5 to 10 pH range, sold by companies like Hach agents in Dublin, "Celtic Water" I think, or "Uisce" in Kilkenny.