under floor heating leak

Oh dear! The pipes went down and were tested as not leaking. The screed went down dry and no leaks showed up. The floorsheets went down and then the ply and ceramic tiles. And then the plumber came to connect the manifold to the boiler.

and we discovered we have a leak somewhere under the floor which wasn't there when the screed went down.

Aaaaaargh!

The conservatory floor is downstairs and is a suspended wooden floor with kingspan between the joists supported by slats between the joists with the pipes and screed laid on top of that but between the joists. The floor sheets rest on the joists and have been screwed in. The ply has been screwed to the floor sheets. The tiler asked how thick the floor sheets were to ensure that the screws he used didn't go right through. The only access to the under floor void is through the air bricks.

Is there any way we can : a) plug the hole in the pipe without disturbing the tiles? Radweld springs to mind but I have a feeling it might break the boiler. I'm assuming that the b) if a) is not possible can we find out where the pipe is leaking without ripping up the whole floor (is there some detector that works through the tiles (eg by measuring microwave absorbtion or something). Is it possible to remove just a portion of the tiled floor without disturbing the surrounding tiles??

Any suggesstions welcome as we really don't want to start all over again!

Despontantly

Chris J

Reply to
ChrisJ
Loading thread data ...

Myself, I'd be tempted to try some of that CH leak sealant liquid you put in the system then allow to circulate.

It's sorted me out on one occasion. I bet a screw has gone through a pipe!

Is pressure dropping rapidly of is it over the course of rays rather than minutes?

Reply to
PeTe33

One of the pitfalls of laying piping under floor tiles/laminate ect ect. :-( Glad my piping is behind skirting. :-)

Reply to
The3rd Earl Of Derby

My thoughts too. I hope the floorer and tile have good insurance or are willing to redo it all for a four pack of beer!

The pressure drops within about 30 minutes. We've isolated the manifold from the rest of the CH and the leak is definately there or under the floor.

ChrisJ

Reply to
ChrisJ

Bad luck, that must be so annoying.

Have you considered abandoning the UFH and installing conventional rads?

Reply to
Steve Walker

Just been reading up and have found that fernox do a raditor leak fixer for fixing inaccessible leaks. Would this work in the plastic UFH pipe? Are these fixes generally permanent or is it a bodge that will need repeated fernox applications? I'm presuming we have a screw through the pipe or at least into it.

Its a presurised combi boiler. How would I get the fernox in?

ChrisJ

Reply to
ChrisJ

How much space is there under the floor? Could you remove a few bricks and crawl in? Another long shot is a stethoscope. Put as much pressure on the system as possible and follow the route listening. you may get lucky. Personally I wouldn't trust leak sealer on plastic pipe. The sealer is unlikelt to stick to the plastic and as the plastic flexes it any deposit is likely to crack off.

Good luck

John

Reply to
John

Having fitted UFH .. you let a tradesman screw to the floor ? ... lesson learned the hard way.

The fix is to expose the portion with the leak ...cut out the damaged portion and fix with a repair fitting. Probably not what you wanted to know ..

Reply to
Osprey

Perhaps adding a flourescent dye as used by aircon installers may help. If you can shine some UV light into the void afterwards you may see where it was leaking.

Yes, How you do it will depend on the type of tiles. A Fein multimaster can take out individual tiles neatly if required (although not withoy damaging the removed tiles). A trend routabout cutter can be used to make a neat access hole to the floor void if required after.

Reply to
John Rumm

The stethescope idea sounds good, maybe compressed air (used with great care) would make better noise for detection.

Something I'd try...

Suck/blow a small bit of rag on fine fishing line all the way through using a vac.

IF this works then pull stronger line and then a ball that is a fairly snug fit through (keeping the line at both ends).

If it jams at some point, that's where a screw is. Try it from the other end to double check.

Maybe then pull some fine wire through attached to the line from where the ball is, connect to isolated 240v (shaver transformer) and use a cable detector above to trace where it stops.

Good luck!!!

cheers, Pete.

Reply to
Pete C

One other idea if a line can't be pulled through, blow a close fitting steel ball though with water, if it gets stuck use a metal detector to find where...

cheers, Pete.

Reply to
Pete C

The only technique I know of for detecting leaks in UFH uses helium. I've never had to use it (touch wood) and so don't know the details of how it works. There used to be a technical paper available on the US Radiant Panel Association (heating contractors') website called "Leak Detection with Helium", but it's now only accessible to RPA members. If you get details, I think you'll find it's a specialist process. Helium is used as a tracer gas for leak detection 'cos the molecules are small & will pass through small perforations.

It's used to find leaks in some duct/pipe systems in conjunction with a forward light-scattering photometer, so it may be the same method that's used for refrigerant systems. There's probably a BS on this(?). Advise me if this looks promising & I'll have a rummage in the BS catalogue on Monday.

The only other system I've used utilised compressed air; the air leaks generate ultrasonic noise which are normally inaudible but could be detected with a microphone/transducer/ headphones/amplifier type device. Although effective with exposed pipes, I doubt this would work through a screed because HF sounds are easily attenuated.

I think anything other than finding and fixing the leak(s) will be a bodge. Problems will be huge if you find you have to start excavating after occupation.

I think you're going to have to grasp nettle/bull by horns, identify the leaking circuit(s) and start digging. :-(

Reply to
Aidan

the floor sheets were to ensure that the screws he used didn't go right

through.

Re-reading your post, I think I'd break up the tiles where the tiler used screws, remove screws, pressurize loop with air or water and look for leaks around the screw holes. Keep screws as evidence.

Reply to
Aidan

Not if there is a nail or screw through it. Even if it works temporarily the nail/screw will corrode quite quickly both making the hole in the pipe bigger due to the expansion of the rust and eventually corroding through leaving a hole too big for leak sealant to work.

If ply was laid before tiles it is common to screw it down at about

12-24 in centres. I'd guess one or more (possibly all) of the screws were too long and pierced the piping.
Reply to
Peter Parry

The message from Peter Parry contains these words:

Seems odd to put ply down on top of UHF. Surely it'll insulate the floor.

Reply to
Guy King

It does a bit - but not significantly as it is a high mass/low temperature system. It's no worse than carpet.

Reply to
Peter Parry

The message from Peter Parry contains these words:

But he says "floor sheets, then ply then tiles". Why not just tile straight onto the concrete?

Reply to
Guy King

The floor is suspended wooden floor on joists over an under floor void. 90mm Kingspan was suspended on battens between the joists with its top

30mm below the top of the joist, the pipes laid on the insulation and the remaining gaps filled with screed to the level of the top of the joists. Chipboard sheets are then laid followed by 3mm ply and then tiles. Not sure why the 3mm ply was needed but I'm glad it was as it means we can cut it out and start again without having to clean the old tile cement off the floor sheets.

The solutions with blowing the rags trailing fine wire and applying mains to them then using a wire detector sound great as did the using compressed air. The former I think might not work due to the fact that there is 50m of pipe in that half of the run with tight bends every 2 m. I think the drag on the wire would mean the rag never got as far as it should. Using a metal detector to detect the ball bearing I suspect would fail due to the grid of screws holding down the ply at 8" centres.

Carpet is much more insulating than 3mm of ply. The ply is screwed to the chipboard floor sheets (no concrete in this floor) with really small screws that don't go through the chipboard sheets.

Action so far:

The manifold splits the floor into two loops of approx 5m pipe each. We presurised the loops individually to find which half the leak was in. Guessing the leak was going to be where the pipes crossed the joists then I started at the door working along the wall and removed tiles, ply and floorsheet until I found a wet patch, fortunatly after only having removed about a dozen tiles.

Next job is to remove enough floor cover to replace the holed pipe with a new section. Then get it retested to make sure there was just one hole and then recover the floor.

The Underfloor heating company based in Sheffield came out next day (Saturday) at no charge to help locate the leak and were really helpful even supplying the compression fittings and a length of pipe to patch it once I'd found it. I can't recommend them strongly enough but also I don't think I'm allowed to mention them by name but try googling for "under floor heating sheffield uk s8 9eh" if you want to find out more.

Thanks to all those who suggested ways of dinding the leak.

ChrisJ

Reply to
ChrisJ

Boy I am glad you found it.

Now for hindsight..what CAUSED it???

..I was so effin careful when I did mine about not routing it near where screws were to go, not constricting it, and using little bits of slip on corrugated tubing where it crossed screed boundaries..

Like most things that were very carefully planned and thought out FIRST..its been faultless.

Reply to
The Natural Philosopher

The message from ChrisJ contains these words:

Don't see why not - good service deserves an airing.

Reply to
Guy King

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.