My introduction into the art of leak detection in slab-based hydronic radiant heating systems was in our town's version of Levittown, N.Y., in an area called Fireside. Slab-on-grade homes with ½-in. copper tubing embedded in concrete. Back then (1972), homes in Fireside had radiant heating systems that were springing holes like a metal street sign sprayed with buckshot! The first one I tagged along on as an ap -prentice was easy to find because it could be plainly heard below the thin concrete. It had begun at a soldered coupling and thermal stress had driven the tubing into the coupling, breaking the solder joint. As the leak worsened, it had eroded a sink-hole in the earth — almost large enough for me to crawl into!

Fireside's boilers were shoe-horned into the space below the stairs to the second floor or, for single-story homes, the boiler was out in the open and looked like a stove without any burners. First time I encountered one of those stove-like boilers on a no-heat call, I walked right past it to look for one hidden in a closet.

At the time, we used several methods for finding leaks. The first was to stop, look and listen. Appliances were unplugged or turned o˛ and the boiler disabled to keep the home library quiet. While walking through the home's rooms, you kept on the lookout for damp spots or windows that were dripping wet. If that yielded no clues, a return to the boiler was in order to feel the outgoing ½-in. tubes between the manifold and floor. The idea was to find out which lines were hot to the touch. The boilers held domestic coils and maintained temperature, so a hydronic system at rest would quickly reveal which circuits were affected.

In areas where ceramic tile or floor surface materials impervious to water damage were installed, we would turn off the radiant heating system and close its manifold valves for a few hours or a day (if the weather wasn't too cold) and return to wet-mop the area while turning on the radiant system. Most Fireside homes did not include outdoor reset or mixing strategies, as we would today, and the sudden influx of hot water would create what looked like an x-ray as the tubing would dry out the flooring directly over its pathway. Leaks often appeared as a splotch of dried concrete, screaming out "dig here!"

We also used an old stethoscope-like stereo device with twin metal discs. A door slammed or hard shoes stepping on concrete sounded like gunshots, while the whine of a circulator or refrigerator motor could be clearly heard. For those of us with young knees, there was also an old-style hand-crank-like phone device with a contact-pin sticking out its opposite end for direct contact with piping or the floor to listen for leaks.

Locate, repair, patch or abandon.

Basically, those were the choices and many opted to cap off lone circuits. If multiple circuits were involved, many chose to abandon the radiant heating and install baseboard. Almost all of the Fireside systems have long been abandoned — victims of thermal stress.

That was then, this is now!

Here's a six-step leak-detection process:

1. Is there a leak: A hydrostatic pressure test will reveal if there is a leak and how quickly the system is losing water. Long-term leak detection can be monitored by installing a water meter.

2. Isolation: If loops/manifolds are valved on both the supply and return, you can begin the process of elimination.

3. As-built records: If the owners have an accurate as-built record for embedded tubing, you can narrow your search area considerably.

4. Detection methods include:

Stop, look & listen: windows dripping with condensation; pets seeking out a single specific warm spot; wet, moldy areas; damp carpeting and noises — like a roiling boil as small stones are rattled against the underside of concrete or a hissing spray-like sound. You could try old-style listening devices.

Infrared thermometer: A hand-held, point-and-shoot, inexpensive infrared heat detector and bare feet (ask permission first) can make for a great investigative combination. While the detector can "see" tubing heat-lines and larger areas where leaks may be present, your feet will often notice broader warmth zones. This is also a great way to ˚nd air-bound loops that aren't circulating.

Ultrasonic leak detection: If you've ever purged a hydronic system by using a garden hose, you've probably learned to double-over the hose to regulate pressure during the process. Water escaping through the partially pinched-off hose crackles as air bubbles pass from high to low pressure and expand. Ultrasonic leak detectors listen for noises generated by air or gas you've introduced in the system as it escapes through a crack or pin-hole leak and expands. Expect to spend $800 to $1,500 for a detector.

Infrared thermal imaging cameras: If you've got about $30,000 that you don't need, you can buy one of these. If not, you might be able to rent or borrow one. They show temperature variations by color and different density materials will sometimes appear in varying shades that can be deceiving.

Gas detection: Helium and hydrogen detectors can be used if they're sensitive enough to measure at or below 50 parts per million. Higher than 50-PPM devices can work too, but they're said to be less precise. Evacuate the affected loop by forcibly ejecting the hydronic fluid with compressed nitrogen or air instead of simply using an air compressor to move the fluid. When filling with helium or hydrogen, be careful not to flood the area with gas or you'll contaminate the site. Start with a single low-pressure charge and do not continue to feed gas into the system. It takes time for the tiny molecules to seep up through concrete but, if you're patient, the results can be spot-on and location of minute leaks can be pinpointed.

5. Repairs: There are some piping oddities awaiting your repair attention and some unwanted and unpleasant surprises. What looks like copper at the manifold can turn out to be copper-coated steel tubing below the floor. The age and type of piping materials should be considered. Demolition and removal of floor coverings, concrete and potential for interference from utilities and plumbing may all contribute to the go/no-go repair decision.

6. Restoration: Communication and a full understanding of the work-site responsibilities will ultimately de˚ne how satisfied a client will or won't be at completion.