Plumbing/Boiler SOS

[Gaffar]

Calling on any plumbers for assistance/advice

Apologies for the long post in advance

I've just had a Biasi Riva Conpact HE (28kw) boiler installed along with

8 new rads, pipework used was HEPA2 10mm runs around the house with either 15mm copper or microbore ends.

All done as part of extension work and new pipework was connected to old pipework feeding 4 existing radiators using 15mm copper pipes.

Since boiler commisioned, the pressure on the boiler kept dropping to zero overnight, even though nothing in use (i,e, HW or CH), I found a possible leak on a join from new to old pipework and that has just been resoldered by the plumbers, but they did not stick around to check the rest of the system, since they have gone the pressure is still dropping dramatically and is visibly going from 1.5 bar to zero witin 30-60mins.

Any pointers on where to start looking, all new pipework is concealed in walls and joists. Could the boiler itself be faulty?, I have BG coming to do there annual service check on Wed, will they be able to diagnose the cause?

Any comments/advice will be greatly appreciated

Thanks Worried homeowner

[Aidan]

The pressure relief/safety valve on the boiler sometimes drips. The discharge pipe (15mm copper) will be piped through the exterior wall at the boiler. You could fix a plastic food bag over it with a rubber band to see if it fills up or gets blown off. It's a long shot though, with a new boiler. Don't cap it or seal it in any way.

Other than that, you've got to play hunt the leak. You could get the old pipewortk disconnected from the new. This won't help find the leak, but if it's on the new system it's the Installer's liability & problem. They won't deal with it whilst they can point a finger at the old, suspect pipework.

You could also use the valves on the boiler (see Users' Manual) to isolate the heating if it's a combi. Rad valves often drip around the glands.

[Gaffar]

I don't think it is this as I have watched this whilst the pressure has been dropping and nothing comes through

If the pressure is dropping to zero rapidly, does that mean all the water in the system is being drained somewhere?, if so should I not see a wet patch somewhere as an indication as I have filled the system several times and it kepps dropping

Any tools/techniquies that can be used to chase the leak?

It is a combi, if the rad valves were leaking would I not see some evidence of this on the floor (there is no carpet yet)

Thanks for the reply

[Andrew Gabriel]

No. The volume of water expelled (assuming no significant temperature change in the water) will be something like half the volume of the expansion vessel (depending on the air pressure in the vessel). Probably around 4 litres, so you'll probably going to notice that somewhere unless it's running into the ground under the house. Does that seem like roughly the amount of water it takes to get it up to pressure again? If not, then another possibility is that the pressure vessel has failed, which would result in a very much smaller volume of water leaking out and being required to refill. This would also cause rapid pressure rise and pressure relief valve discharge if water is heated up.

A quick look on the web implies this is a condensing boiler, in which case another possibility is a leaking heat exchanger with the water running out through the condensate drain, which might not be obvious unless you have some way of checking it.

When I installed my system, I fitted full-bore isolation valves in the main branches, partly so I could isolate them to help locate a leak if one happened (but mainly because I installed half the system one year, and the other half the following year;-)

Don't isolate the radiator circuit unless you switch off the boiler. It may require it to dump excess heat when you switch off a hot tap.

Yes.

[<me9]

If the expansion vessel is not pre-loaded to the correct pressure the amount of water required to leak out will be small. Is a small amount of water required to re-pressurise the system? If so, resetting the static pressure in the expansion vessel may reduce the problem. A small leak can be hard to find on a hot system due to evaporation of the evidence. Pumping it up to maximum (usually about 3 bar) and looking for dampness around joints/valves, whilst the system is cold, may find the leak.

[Mr Fizzion]

Somewhere near the lowest point in the system you should have a drain c*ck. You could drain the water into a bucket (well several buckets!) and measure how much water is in the system after it has been filled with the filling loop, then refill and wait 60 minutes and measure again.

You could disconnect heating flow and return from the boiler and drain down the system then pressure test with air. Use screwfix 10179

and some pushfit end caps: 15121 or 17170 or there's a plastic range too - you don't seem to be able to get 10mm stop ends in cuprofit: is also a reliable supplier: you need to do the pressure test is a car foot pump - you connect it to the test guage and pump like hell! :-) (After putting the stop end on the other pipe)

Some leaks can be heard as a hiss, very small leaks can be detected with leak detection spray:

't put your head near the end cap - apparently they shoot off sometimes (not happened to me yet!)

Mr F.

[Mr Fizzion]

Or as others have pointed out you could just see how much water it takes to repressurise...not sure there is any point in measuring how much water comes out!

Mr F.

[Aidan]

2) Pressure testing with air is inherently hazardous. If bits break or come loose they fly around at potentially lethal velocities. Do not "pump like hell". The system has been proved to be safe with water up to a certain pressure, so I wouldn't exceed that. 5 -6 psi should be adequate to hear leaks. The noise is high frequency, so young ears will hear it better that oldies. In fact a lot of it is in the ultra-sonic spectrum. 3) The OP isn't a plumbing expert; I would suggest he should not be doing such testing. The installers may claim his work has invalidated the warranty (e.g., you've overpressurized the boiler & punctured the heat exchanger). I'd still suggest disconnecting the new work from the old so that the soundness of the new work can be proved. 4) A newly filled system might need topping-up as air bubbles make their way to the automatic air vent & are vented. Presumably, it's more than this could account for. 5) If it's part of an old system, what happened to the open vent, cold feed and feed & expansion tank? Have they been disconnected & removed? If valved off, could the F&E tank be filling up & the water going out the overflow? This would account for the lack of any wet patch.

[Mr Fizzion]

OK, perhaps not like hell, but filling 10 radiators with air even to

1.5 bar takes quite a lot of pumps - what I meant was go at it fast not go at it hard or else you'll be doing it all day.

I pressure tested my new system to 4 bar. Nothing came flying off. I had one small leak in a failed Hep2o joint and one major failed cuprofit joint (the O-ring seemed to be badly damaged) plus some small leaks on a couple of radiator valves which were cured by adding a bit more ptfe on the threads.

Mr F.

[Aidan]

Mr Fizzion wrote: - what I meant was go at it fast

4 bar must have taken much pumping. A lot of effort went into compressing the air in the system. If anything HAD come flying off it would have had a lot of energy behind it. Things flying off are quite likely since the purpose of pressure test is not only to find leaks, as you did, but also to cause the failure of any weak joints under controllable conditions.

The classic case of failure under air pressure was the Comet 1 crashes. The pressurized fuselage was failing, but they didn't know why or where. One of the bits of wreckage recovered from the Mediterranean crash had the imprint of a small Italian coil stamped onto it, as clear as if it had been done in the Italian mint. The coin had obviously been chucked around at great velocity by the explosive failure. This was at (I think) something like 5 or 7 psi, but a large volume of air was involved.

When the RAE were investigating the causes, they wanted to repeatedly pressurize/depressurize a fuselage (and cycle the loads with hydraulic jacks). If they had used air, the bits would have been chucked all over Farnborough when it did fail. They built a sectional steel tank around the fuselage with the wings poking through rubber sleeves. They filled the tank and fuselage with water & pressurized/depressurized the water in the fuselage. When it eventually failed, it the damage was limited to a crack at the corner of a window.

You should only pressure test with air as a last resort and with the area evacuated. How did I learn this?

"You're not testing it with air, are you?" "Yes, mate. We always do it like that" "The spec says you have to test it with water. You'd better turn it off." "Why? I've been doing this for years and I've never had any....." Whoompf. Thud. Hisssssssssssssssss.

Or words to that effect.

[Gaffar]

Thanks for all the replies guys, just to give you an update, I found a leak on one of the radiators upstairs, albeit small, the ceiling board below the joint was visibly damp, called plumbers back and they repaired the joint and went, pressure again dropped to zero within a couple of hours after they had gone, called dthem back and they suspected that the joints they had made from new piping to old system could be the only other obvious suspect, having hacked into the wall (luckily hidden now by kitchen units) they found a big leak with water disappearing under the house. Since they have fixed this, touch wood, no pressure drops in the last week (although the system is still not in use at the mo, other than to test the heating once for 4 hours)

Thanks once again for all the advice

Regards Gaffar