OT: X-band doppler water leak detection


Ed Breya
 

We just acquired a property that has a house with radiant hydronic heating, and there is a leak in the PEX tubing somewhere in one of the upper floor heating zones. The water ultimately drips out through an opening in the ceiling of the lower floor. I don't want to tear up too much drywall or flooring to locate the actual leak point, which could be a few feet away in any direction. Before resorting to that, I figured I'd try an experimental setup with an old Solfan microwave doppler motion detector module.

 

I have collected a number of these over the years, so that's covered, and all I need is to set up power and an audio amp to listen in on the signals from the detector. I had planned on making a rig like this years ago, for locating a yellow jacket nest somwhere in a wall, but they wisely departed, causing the project to be mothballed. In that case, the sound would have been from body motion and wings buzzing, and easily audible, but for a water leak, it will depend on whether there's gurgling, spraying, or dripping action to get an audible signal.

 

Before I get started, I figured I'd check here to see if anyone has experimented with this stuff for this sort of pupose, or knows of commercial equipment that I could get ideas from. The Solfans run around 10.5 GHz, with Po maybe 10-40 mW, and normally rely on water absorption and motion of bodies up to a few meters away to provide an AC doppler signal. The detectors are DC coupled, so can in fact reach way down in frequency and even indicate position/phase of an object close to the horn - but not with high stability.

 

In this case the desired target is small streams or dripping water at maybe 0.1 to 0.5 GPM, from a distance of up to around a couple of feet, through layers of material depending on the view. From below, looking through the ceiling, there's 1/2" hydrated gypsum drywall, then about a foot (2x12" floor joists) of hollow space and possibly some fiberglas insulation, then 3/4" plywood subfloor, then the hollow PEX cavities formed with 3/4" x (3-6" W?) lathe strips. From above, there's 3/4" hardwood flooring, mounted on a 3/4" plywood substrate that rests on the lathe. The water loop is 1/2" PEX in a serpentine pattern, laid between the interdigitated lathe strips. Normal system pressure is about 15 PSIG. The objective is to locate the point of the leak in the tubing, as close as possible, and quickly, to minimize excess water damage. I think the water will be easiest to detect from below, and more accurately located from above, once the approximate location is narrowed down.

 

So, does anyone have any info, knowledge, or experience in this, or is there a better way? I know that there are ultrasonic leak detectors for gases, so I could rig up the loop to be tested with air instead, but then the ultrasonic emissions would have to be detected through layers of wood either way.

 

Ed


KeepIt SimpleStupid
 

Ed:

All I can offer is the existance of GPR or Ground Pentrating Radar.  See: http://en.wikipedia.org/wiki/Ground-penetrating_radar  I would suppose that those services could be purchased.  Aparently, from what I read the harder part is interpretation of the data collected.

Ferrofluid - http://en.wikipedia.org/wiki/Ferrofluid  I have no idea.

Thermal imager and hot or cold water.

Just ideas.

 
So, does anyone have any info, knowledge, or experience in this, or is there a better way? I know that there are ultrasonic leak detectors for gases, so I could rig up the loop to be tested with air instead, but then the ultrasonic emissions would have to be detected through layers of wood either way.
Ed



Dave Daniel
 

In my experience, the PEX pipe used in in-floor hydronic heating systems is intimately close to the special concrete in which the pipe is embedded. I would guess that there is little or no turbulence associated with the leak, which means that detection by acoustic means won't work.

The thermal imager idea is a good one. One possible source of a good thermal imager is your local fire district - if you can convince them to lend you a firefighter and their imager, you may be able to locate the leak. The floor covering would probably have to be something that conducts heat well, though - if the floor is carpeted, thermal imaging probably wouldn't work.

DaveD

On 2/21/2014 6:49 PM, KeepIt SimpleStupid wrote:
 
Ed:

All I can offer is the existance of GPR or Ground Pentrating Radar.  See: http://en.wikipedia.org/wiki/Ground-penetrating_radar  I would suppose that those services could be purchased.  Aparently, from what I read the harder part is interpretation of the data collected.

Ferrofluid - http://en.wikipedia.org/wiki/Ferrofluid  I have no idea.

Thermal imager and hot or cold water.

Just ideas.

 
So, does anyone have any info, knowledge, or experience in this, or is there a better way? I know that there are ultrasonic leak detectors for gases, so I could rig up the loop to be tested with air instead, but then the ultrasonic emissions would have to be detected through layers of wood either way.
Ed




Don Black <donald_black@...>
 

How about listening to the ceiling with a stethoscope. Since the object is to locate the leak you might hear the water dripping.

Don Black.

On 22-Feb-14 12:49 PM, KeepIt SimpleStupid wrote:
 
Ed:

All I can offer is the existance of GPR or Ground Pentrating Radar.  See: http://en.wikipedia.org/wiki/Ground-penetrating_radar  I would suppose that those services could be purchased.  Aparently, from what I read the harder part is interpretation of the data collected.

Ferrofluid - http://en.wikipedia.org/wiki/Ferrofluid  I have no idea.

Thermal imager and hot or cold water.

Just ideas.

 
So, does anyone have any info, knowledge, or experience in this, or is there a better way? I know that there are ultrasonic leak detectors for gases, so I could rig up the loop to be tested with air instead, but then the ultrasonic emissions would have to be detected through layers of wood either way.
Ed






This email is free from viruses and malware because avast! Antivirus protection is active.



Ralph Hartwell <ralph_yg@...>
 

Thermal imager and hot or cold water.
Not bad! That might just work.

True story: Many years ago I had a water leak from a hot water pipe that passed under a concrete floor - but where was it? Not having a thermal imager, I resorted to taking a wet mop and quickly swabbed the floor with warm water. Then I stood back and waited for the floor to dry. As it began to dry up, I marked the center of the first spot that became dry. Several iterations of mopping and waiting gave me a smaller diameter marker point. Cutting through the concrete showed that I was right on target, and the pipe was repaired without extensive digging and expensive concrete work.

As for the microwave detector, that might just work. I would use a DC coupled oscilloscope so I could watch for very small signals.

I have a similar X-band system that is easily able to measure reflected signals from raindrops as they fall outside of my workshop window. I have also used both C-band and KU-band satellite LNB's for detection of microwave noise signals from ionized gasses in discharge tubes and fluorescent lamps from as far away as 15 feet. Nothing more sensitive than a standard installers signal meter was required. I usually use my Tek scope to look at the low frequency signals.

Best DX & 73,

Ralph W5JGV - WD2XSH/7


Peter Gottlieb <hpnpilot@...>
 

Why not drill small holes and use one of those inspection cameras on a flexible gooseneck to look in the ceiling? Start where the water is showing and work from there.


snapdiode
 

Unless you get one of those coherent fiber cameras, the image quality is abysmal.


Albert LaFrance
 

Here’s what you *really* need – technology from Atomic Age:

http://coldwar-c4i.net/EE0952/812.html

 

Albert

 

 

From: TekScopes@... [mailto:TekScopes@...] On Behalf Of edbreya@...
Sent: Friday, February 21, 2014 7:29 PM
To: TekScopes@...
Subject: [TekScopes] OT: X-band doppler water leak detection

 

 

We just acquired a property that has a house with radiant hydronic heating, and there is a leak in the PEX tubing somewhere in one of the upper floor heating zones. The water ultimately drips out through an opening in the ceiling of the lower floor. I don't want to tear up too much drywall or flooring to locate the actual leak point, which could be a few feet away in any direction. Before resorting to that, I figured I'd try an experimental setup with an old Solfan microwave doppler motion detector module.

 

I have collected a number of these over the years, so that's covered, and all I need is to set up power and an audio amp to listen in on the signals from the detector. I had planned on making a rig like this years ago, for locating a yellow jacket nest somwhere in a wall, but they wisely departed, causing the project to be mothballed. In that case, the sound would have been from body motion and wings buzzing, and easily audible, but for a water leak, it will depend on whether there's gurgling, spraying, or dripping action to get an audible signal.

 

Before I get started, I figured I'd check here to see if anyone has experimented with this stuff for this sort of pupose, or knows of commercial equipment that I could get ideas from. The Solfans run around 10.5 GHz, with Po maybe 10-40 mW, and normally rely on water absorption and motion of bodies up to a few meters away to provide an AC doppler signal. The detectors are DC coupled, so can in fact reach way down in frequency and even indicate position/phase of an object close to the horn - but not with high stability.

 

In this case the desired target is small streams or dripping water at maybe 0.1 to 0.5 GPM, from a distance of up to around a couple of feet, through layers of material depending on the view. From below, looking through the ceiling, there's 1/2" hydrated gypsum drywall, then about a foot (2x12" floor joists) of hollow space and possibly some fiberglas insulation, then 3/4" plywood subfloor, then the hollow PEX cavities formed with 3/4" x (3-6" W?) lathe strips. From above, there's 3/4" hardwood flooring, mounted on a 3/4" plywood substrate that rests on the lathe. The water loop is 1/2" PEX in a serpentine pattern, laid between the interdigitated lathe strips. Normal system pressure is about 15 PSIG. The objective is to locate the point of the leak in the tubing, as close as possible, and quickly, to minimize excess water damage. I think the water will be easiest to detect from below, and more accurately located from above, once the approximate location is narrowed down.

 

So, does anyone have any info, knowledge, or experience in this, or is there a better way? I know that there are ultrasonic leak detectors for gases, so I could rig up the loop to be tested with air instead, but then the ultrasonic emissions would have to be detected through layers of wood either way.

 

Ed


stefan_trethan
 

I think the compressed air / leak detector method is most promising, and least damaging so definitely worth a try in any case. You might be able to hear the leak without a detector if you are very lucky, and it costs nothing.


Personally I doubt the thermal camera method will work unless the leak is massive. A few drops percolating through building materials will quickly take on the surrounding temperature and not show up on camera.


Most likely you will have to rip out a larger section anyway, if the materials are soaked, to repair the damage and prevent any mould issues.


ST


On Sat, Feb 22, 2014 at 5:12 PM, Albert LaFrance <albert.lafrance@...> wrote:


Here’s what you *really* need – technology from Atomic Age:

http://coldwar-c4i.net/EE0952/812.html

 

Albert

 



preilley_454@...
 

I have read about finding leaks with dye.   There is some dye that they

use where that a small amount of dye produces a massive amount of

brightly colored water.   I can't remember if the color was yellow or blue.

You put the dye in the water and watch for the first place the color

comes through.


Pete.


Ed Breya
 

Thanks everybody, for the ideas on this project. I do have one of those snake inspection cameras (from harbor freight), and an old technology thermal imaging camera, but not ready access to making Na-24.

 

The thermal camera is an oldie with nice germanium optics, but I don't recall what its wavelength range is, or if it can resolve well enough to see what I need (it's non-cryogenic, so not that sensitive) - I'll have to dig it out and give it a try. I figured on using the snake camera through whatever light fixture openings etc I can access to narrow it down a bit, then try the doppler method.

 

If that doesn't work, I'll try the ultrasonic/compressed air method, but I was preferring to not have to cut the original lines to splice in test fittings - there's not much room around the zone valves to tubing connections area. Another option would be to purge the entire system and pressurize it with air, but there's about a dozen zones, and thousands of feet of tubing to deal with. On the other hand, I was planning to flush the whole system anyway, so maybe it wouldn't be too bad to include an air test too.

 

Regards,

Ed


 

One thing that might work is to try to freeze some point in the center of the loop and see if the leak stops. Then you keep doing successive binary tree sections until you narrow it down. Maybe dry ice could be used to create ice blocks in the line.
 
How do the pros do it?
 
Tom
 
 

----- Original Message -----
Sent: Saturday, February 22, 2014 12:45 PM
Subject: RE: [TekScopes] OT: X-band doppler water leak detection

 

Thanks everybody, for the ideas on this project. I do have one of those snake inspection cameras (from harbor freight), and an old technology thermal imaging camera, but not ready access to making Na-24.

 

The thermal camera is an oldie with nice germanium optics, but I don't recall what its wavelength range is, or if it can resolve well enough to see what I need (it's non-cryogenic, so not that sensitive) - I'll have to dig it out and give it a try. I figured on using the snake camera through whatever light fixture openings etc I can access to narrow it down a bit, then try the doppler method.

 

If that doesn't work, I'll try the ultrasonic/compressed air method, but I was preferring to not have to cut the original lines to splice in test fittings - there's not much room around the zone valves to tubing connections area. Another option would be to purge the entire system and pressurize it with air, but there's about a dozen zones, and thousands of feet of tubing to deal with. On the other hand, I was planning to flush the whole system anyway, so maybe it wouldn't be too bad to include an air test too.

 

Regards,

Ed


stefan_trethan
 

Too bad TDR will probably not work, now that would be really sweet.
(Although even if it would be possible a length from the end does not nail the location).

I have considered resistance measurements, but can't come up with a useful method to narrow down the leak.


BTW PCB layout software is really useful in planning floor heating. The track length feature especially.


Perhaps something like freon, or whatever is the environmentally acceptable replacement, could be put into the pipe. It should boil off at the leak and stick out on a thermal camera. Be aware that most of the current refrigerants are flammable (like propane for example). You may be left with a bigger hole in your house ;-)


ST


On Sat, Feb 22, 2014 at 6:45 PM, <edbreya@...> wrote:

If that doesn't work, I'll try the ultrasonic/compressed air method, but I was preferring to not have to cut the original lines to splice in test fittings - there's not much room around the zone valves to tubing connections area. Another option would be to purge the entire system and pressurize it with air, but there's about a dozen zones, and thousands of feet of tubing to deal with. On the other hand, I was planning to flush the whole system anyway, so maybe it wouldn't be too bad to include an air test too.

 

Regards,

Ed




 

What about detecting moisture saturation with induction? This would pick up the
different thicknesses of material like the joists but it should still be
possible to detect which are wet and which are dry.

I imaging something like an LR or LC wien bridge oscillator where the gain of
the amplifier is varied to maintain a constant output so the gain control signal
will precisely show the loss of the inductive winding.

On 21 Feb 2014 16:29:09 -0800, you wrote:

We just acquired a property that has a house with radiant hydronic heating, and there is a leak in the PEX tubing somewhere in one of the upper floor heating zones. The water ultimately drips out through an opening in the ceiling of the lower floor. I don't want to tear up too much drywall or flooring to locate the actual leak point, which could be a few feet away in any direction. Before resorting to that, I figured I'd try an experimental setup with an old Solfan microwave doppler motion detector module.

I have collected a number of these over the years, so that's covered, and all I need is to set up power and an audio amp to listen in on the signals from the detector. I had planned on making a rig like this years ago, for locating a yellow jacket nest somwhere in a wall, but they wisely departed, causing the project to be mothballed. In that case, the sound would have been from body motion and wings buzzing, and easily audible, but for a water leak, it will depend on whether there's gurgling, spraying, or dripping action to get an audible signal.

Before I get started, I figured I'd check here to see if anyone has experimented with this stuff for this sort of pupose, or knows of commercial equipment that I could get ideas from. The Solfans run around 10.5 GHz, with Po maybe 10-40 mW, and normally rely on water absorption and motion of bodies up to a few meters away to provide an AC doppler signal. The detectors are DC coupled, so can in fact reach way down in frequency and even indicate position/phase of an object close to the horn - but not with high stability.

In this case the desired target is small streams or dripping water at maybe 0.1 to 0.5 GPM, from a distance of up to around a couple of feet, through layers of material depending on the view. From below, looking through the ceiling, there's 1/2" hydrated gypsum drywall, then about a foot (2x12" floor joists) of hollow space and possibly some fiberglas insulation, then 3/4" plywood subfloor, then the hollow PEX cavities formed with 3/4" x (3-6" W?) lathe strips. From above, there's 3/4" hardwood flooring, mounted on a 3/4" plywood substrate that rests on the lathe. The water loop is 1/2" PEX in a serpentine pattern, laid between the interdigitated lathe strips. Normal system pressure is about 15 PSIG. The objective is to locate the point of the leak in the tubing, as close as possible, and quickly, to minimize excess water damage. I think the water will be easiest to detect from below, and more accurately located from above, once the approximate location is narrowed down.

So, does anyone have any info, knowledge, or experience in this, or is there a better way? I know that there are ultrasonic leak detectors for gases, so I could rig up the loop to be tested with air instead, but then the ultrasonic emissions would have to be detected through layers of wood either way.

Ed


Dave Daniel
 

Looking online, it seems that the most common answer is borrow an infrared thermal imager.

DaveD

On 2/22/2014 11:17 AM, Stefan Trethan wrote:
 

Too bad TDR will probably not work, now that would be really sweet.
(Although even if it would be possible a length from the end does not nail the location).

I have considered resistance measurements, but can't come up with a useful method to narrow down the leak.


BTW PCB layout software is really useful in planning floor heating. The track length feature especially.


Perhaps something like freon, or whatever is the environmentally acceptable replacement, could be put into the pipe. It should boil off at the leak and stick out on a thermal camera. Be aware that most of the current refrigerants are flammable (like propane for example). You may be left with a bigger hole in your house ;-)


ST

On Sat, Feb 22, 2014 at 6:45 PM, <edbreya@...> wrote:

If that doesn't work, I'll try the ultrasonic/compressed air method, but I was preferring to not have to cut the original lines to splice in test fittings - there's not much room around the zone valves to tubing connections area. Another option would be to purge the entire system and pressurize it with air, but there's about a dozen zones, and thousands of feet of tubing to deal with. On the other hand, I was planning to flush the whole system anyway, so maybe it wouldn't be too bad to include an air test too.

 

Regards,

Ed





bobvines00
 

Sat Feb 22, 2014 10:17 am (PST) . Posted by:

"Stefan Trethan" stefan_trethan
[snip]

Perhaps something like freon, or whatever is the environmentally acceptable
replacement, could be put into the pipe. It should boil off at the leak and
stick out on a thermal camera. Be aware that most of the current
refrigerants are flammable (like propane for example). You may be left with
a bigger hole in your house ;-)

ST
================

At work we've just had an IVD (aluminum ion vapor deposition) system
upgraded. The Contractor checked for leaks in the cooling coils with
helium and a sniffer. Of course, they used to use Freon, but can't
use the "easiest-to-use" gasses like that anymore.


Bob


vollumscope <perls@ime.net>
 

I wonder if it's possible to do something like an acoustic  delay line using an ultrasonic transducer at the tubing input and listening for a return ping because of a density aberration at the leak point-- kind of like TDR...  ?

In the end you'll probably have to rip up the floor to find the leak-- I hate to be a kill-joy.

Good luck.

Vollumscope


Phil Sittner <sittners@...>
 

Gentlemen-
 
A friend had  a similar situation and hired a service that specializes in this kind of detection. I believe the most accurate method is with a specialized ultrasonic detection apparatus. The good news is that these guys exist, the bad news is that these leaks can get really ugly quickly. Good luck with your quest.
 
Phil

----- Original Message -----
From: Bob Vines
Sent: Saturday, February 22, 2014 3:12 PM
Subject: [TekScopes] Re: OT: X-band doppler water leak detection

 

Sat Feb 22, 2014 10:17 am (PST) . Posted by:

"Stefan Trethan" stefan_trethan
[snip]

Perhaps something like freon, or whatever is the environmentally acceptable
replacement, could be put into the pipe. It should boil off at the leak and
stick out on a thermal camera. Be aware that most of the current
refrigerants are flammable (like propane for example). You may be left with
a bigger hole in your house ;-)

ST
================

At work we've just had an IVD (aluminum ion vapor deposition) system
upgraded. The Contractor checked for leaks in the cooling coils with
helium and a sniffer. Of course, they used to use Freon, but can't
use the "easiest-to-use" gasses like that anymore.

Bob


Dave Daniel
 

On another note, many hydronic systems built around 2000 used Kitec fittings to join the PEX tubing. It is now known that these fittings slowly disintegrate over time when exposed to high temperature water. There is a class-action lawsuit in flight to reimburse homeowners for replacing these fittings (I don't have hydronic heating, but my entire house is plumbed with PEX tubing joined by Kitec fittings, both baseboard heating lines and domestic hot water).

Can you tell whether the leak is near a place where there is likely to be a fitting?

DaveD

On 2/22/2014 5:46 PM, Phil Sittner wrote:
 

Gentlemen-
 
A friend had  a similar situation and hired a service that specializes in this kind of detection. I believe the most accurate method is with a specialized ultrasonic detection apparatus. The good news is that these guys exist, the bad news is that these leaks can get really ugly quickly. Good luck with your quest.
 
Phil
----- Original Message -----
From: Bob Vines
Sent: Saturday, February 22, 2014 3:12 PM
Subject: [TekScopes] Re: OT: X-band doppler water leak detection

 

Sat Feb 22, 2014 10:17 am (PST) . Posted by:

"Stefan Trethan" stefan_trethan
[snip]

Perhaps something like freon, or whatever is the environmentally acceptable
replacement, could be put into the pipe. It should boil off at the leak and
stick out on a thermal camera. Be aware that most of the current
refrigerants are flammable (like propane for example). You may be left with
a bigger hole in your house ;-)

ST
================

At work we've just had an IVD (aluminum ion vapor deposition) system
upgraded. The Contractor checked for leaks in the cooling coils with
helium and a sniffer. Of course, they used to use Freon, but can't
use the "easiest-to-use" gasses like that anymore.

Bob



Phil Sittner <sittners@...>
 

I'm not knowledgeable in the specifics of that industry. However, I do know that the ultrasonic method can detect the liquid (water) leaking. The repair itself is straight forward; break out the concrete (a small pneumatic chipping hammer available at home centers for $40.00 will work fine), repair the damage pipe/fittings and patch the concrete. I wouldn't wait too long to fix it.
 
Phil

----- Original Message -----
Sent: Saturday, February 22, 2014 4:55 PM
Subject: Re: [TekScopes] Re: OT: X-band doppler water leak detection

 

On another note, many hydronic systems built around 2000 used Kitec fittings to join the PEX tubing. It is now known that these fittings slowly disintegrate over time when exposed to high temperature water. There is a class-action lawsuit in flight to reimburse homeowners for replacing these fittings (I don't have hydronic heating, but my entire house is plumbed with PEX tubing joined by Kitec fittings, both baseboard heating lines and domestic hot water).

Can you tell whether the leak is near a place where there is likely to be a fitting?

DaveD

On 2/22/2014 5:46 PM, Phil Sittner wrote:
 

Gentlemen-
 
A friend had  a similar situation and hired a service that specializes in this kind of detection. I believe the most accurate method is with a specialized ultrasonic detection apparatus. The good news is that these guys exist, the bad news is that these leaks can get really ugly quickly. Good luck with your quest.
 
Phil
----- Original Message -----
From: Bob Vines
Sent: Saturday, February 22, 2014 3:12 PM
Subject: [TekScopes] Re: OT: X-band doppler water leak detection

 

Sat Feb 22, 2014 10:17 am (PST) . Posted by:

"Stefan Trethan" stefan_trethan
[snip]

Perhaps something like freon, or whatever is the environmentally acceptable
replacement, could be put into the pipe. It should boil off at the leak and
stick out on a thermal camera. Be aware that most of the current
refrigerants are flammable (like propane for example). You may be left with
a bigger hole in your house ;-)

ST
================

At work we've just had an IVD (aluminum ion vapor deposition) system
upgraded. The Contractor checked for leaks in the cooling coils with
helium and a sniffer. Of course, they used to use Freon, but can't
use the "easiest-to-use" gasses like that anymore.

Bob