Detecting to within a foot or so of where a mains pipe leak is?

I've traced the pipe run by injecting a signal into the pipework and tracing with a small pocket size short wave receiver. Easy access, under liftable floor boards, to the pipe run in the middle of the house, with pavement stopcock open I can hear a leak via a piece of broom handle and with company stop c*ck off there is no leak hiss noise. With kitchen isolator off and pavement one on there is a hiss. No obvious damp in that area of the house. Assuming I successfully freeze the pipe and water in the middle I can then determine which side of that mid point. One end is in garden soil and the other under concrete of the kitchen floor , under there somewhere is a conversion from lead pipe to copper pipe, likely problem point ?.

Any ideas?. Would a microphone and silicon grease to the concrete floor or metal plate rammed in the soil pick up enough sound when put through an amplifier be able to zero in ?

Reply to
N Cook
Loading thread data ...

You want to watch that, I hear the water board listen in on shortwave radio for any would be DIY'ers messing about with their pipes.

Reply to
ben

It would be a very foolish thing to wipe a lead to copper joint then lay a concrete floor over it.

Reply to
Dave Plowman (News)

| I've traced the pipe run by injecting a signal into the pipework | and tracing with a small pocket size short wave receiver. | Easy access, under liftable floor boards, to the pipe run in the middle of | the house, with | pavement stopcock open I can hear a leak via a piece of broom handle | and with company stop c*ck off there is no leak hiss noise. | With kitchen isolator off and pavement one on there is a hiss. | No obvious damp in that area of the house. | Assuming I successfully freeze the pipe and water in the middle I | can then determine which side of that mid point. One end is in garden soil | and the other under concrete of the kitchen floor , under there somewhere | is a conversion from lead pipe to copper pipe, likely problem point ?.

Yes worth looking at. | Any ideas?. Would a microphone and silicon grease to the concrete floor | or metal plate rammed in the soil pick up enough sound when put through | an amplifier be able to zero in ?

All tool hire places have pipe finders, which are better than metal detectors, because they will detect electric cables as different from water pipes. A bit pricey.

It is also worthwhile considering replacing the lead with copper, because lead in drinking water is a *bad* thing.

Reply to
Dave Fawthrop

Water companies do something a bit like this.

Using a mic attached to each end of the pipe (i.e. the two stopcocks in your case), connect each to the input of a dual trace scope. Now if you tap the pipe at one end, you'll see the pulse instantly at that mic, and delayed at the other mic. (You might also see other reflections from where the pipe joins the mains, and the like.) Similarly, if you tap the pipe at the other end, you'll see the same but the other way around. Ideally, adjust the time base and trigger so these two pulses are separated by about

2/3rds of a single sweep of the timebase, i.e. the timebase represents just a bit more than the time taken for the sound to travel the length of the pipe between the stopcocks.

If you turn up the gain, you should see the pink noise generated by the leak on the two traces. Now if the leak is at one of the stopcocks (very common in the case of the outdoor one), you will see the pink noise is shifted between the two traces, with the gap being the same as when you tapped on the pipe before and with the mic on the outside stopcock showing the data earlier. Conversely, if the leak is in the middle of the pipe, you will see the leak (pink noise) is completely lined up between the two traces. For other leak positions between the middle and end of the pipe, the shift will be proportional, and the mic showing it first will indicate which side of the middle of the pipe it is.

This is all well and good in theory, but matching up two pink noise traces may not be easy. I rather suspect the water company equipment does this by computer. This also assumes constant sound velocity in the pipe, which might not be the case if there's a join with different types of pipe either side of it. And no, I haven't tried doing this for real...

Reply to
Andrew Gabriel

Just as well I'm an electronic engineer and understand what you say. I would probably take the T/B trig-out signal, buffer and repeatedly trip a small solenoid to tap the pipe. I now know the pipe run goes copper from company stop c*ck to just inside the house, perfectly good sweated joint there. Then the original and good lead run up to somewhere under concrete where there must be another such conversion. I've bunged some water in the freezer and with that broken up and packed around the pipe along with maybe some freezer spray I know I can freeze the water (as I've done this before to repair a pipe where it was not possible to fully turn off the mains supply ) to narrow down which end is the problem. Then unusually try the mike and 2 channel scope thing - very rare that I use both channels of a scope. Listening on the broom handle the noise seems louder at the garden end rathe rthan concrete covered end but that may be due to unconstrained pipe there so can resonate freer.

Should it be a leak in the lead run I know the following technique works and I even placed on my otherwise electronic repair/tips files. I had to repair a lead pipe at the bottomost part of its run where it had chafed on the edge of a brick so the sweating was on the underside of the pipe so worst possible situation.

"repairing holes in old lead water pipe - use 2 hot air guns,one to dry out enclosed water and generally heat the pipe and the other one to locally heat to sweat in lead solder. Even a novice can do a functional repair like this as there is much less chance of the pipe totally melting away as with blow-lamp heating." on

formatting link

Reply to
N Cook

More likely the copper hasn't been isolated from the concrete and it's corroded, in which case you want to rip the whole lot out anyway.

Reply to
Rob Morley

One of the odd bits of test equipment I have is a stethoscope for trying to locate the source of squeaks in mechanical equipment. I tried going over the concrete floor every foot of run listening for any noise but nothing heard, also tried with a metal rod with a wooden cupboard knob on the end instead of broom handle as a listening stick , better response actually on the pipe but nothing heard thru' concrete.

I won't be doing the same with stethoscope, in daylight anyway, over the garden unless freezing halfway shows it is definitely at that end.

Reply to
N Cook

Hmm. Me, I suspect I'd try using a computer with a couple of mics inputting to left and right channels. Then do a trivial app to take the wav file, and spit out a text file for plotting with gnuplot.

That is a handy technique.

Reply to
Ian Stirling

The low tech approach which has worked for me both underground, under-floor in concrete leaks, and dmapness in general:

Find the boundary of approximately equal wetness (or dampness) go for the middle. This works because the soil or the concrete or whatever are reasonably consistent in the way water travels.

Reply to
Ed Sirett

This afternoon i looked out a protimeter masonry probe, minus the meter but I can use just a DVM instead for relative measurements. Will try tomorrow after the mid point freezing

Reply to
N Cook

With buried unprotected copper and lead, I'd just abandon the whole lot and replace the entire run with 25mm+ MDPE from stopcock to stopcock.

Christian.

Reply to
Christian McArdle

The leak was 0.00026 L / min berfore icing up at inlet to house and 0.00026 after icing up so the leak (or maybe the major leak) is in the garden somewhere.

-- Diverse Devices, Southampton, England electronic hints and repair briefs , schematics/manuals list on

formatting link

Reply to
N Cook

The leak was 0.00026 L / min berfore icing up at inlet to house and 0.00026 after icing up so the leak (or major leak) is in the garden.

Reply to
N Cook

Hmmm. That's about the size of leak I had when they kindly installed my meter with the wrong size fitting. Did this leak come with the meter, or has it just started?

Chris

Reply to
chris_doran

| The leak was 0.00026 L / min berfore icing up at inlet to house | and 0.00026 after icing up | so the leak (or maybe the major leak) is in the garden somewhere.

Is there somewhere on the line where the plants grow *very* well and everything is very green in a drought.?

Reply to
Dave Fawthrop

About 9 gallons a quarter, hardly a major leak, one flush of the bog.

Reply to
<me9

Listening for potentially non-existent whistles, is about 90% less reliable than looking for smoke. Plumbers often put peppermint essence in smoke, so that if it can't be seen, there is a very good chance it can be smelled.

Plumbers have been using this method for hundreds of years, and have yet to change it, for the good reason that it works. And it's cheap.

Why re-invent the wheel? Your local plumbing supplier should have smoke bombs for this very purpose.

Reply to
N Cook

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.