How to find existing radiant heat tubing in cement slab

[Reid]

There's a newer method to locate PEX tubing in concrete -- ground penetrating radar (GPR). GPR is a geophysical method which was primarily used for geology and archaeology in the past, but over the last decade has become very significant in construction and industrial services.

There aren't very many GPR service providers, but most major metro areas should have at least a couple. Locating PEX with GPR isn't an entry level service, it requires high-end equipment and reasonable expertise, but with these conditions met it's a very routine task.

As an example, we recently mapped out all of the PEX tubing in a 18' x 60' area in just over half a day. The area was not at the middle of a parallel tubing run, the tube layout ran around existing machinery in a complex (and cool) pattern. I've created a gallery album in my profile with more photos from the job.

Extending the service, highly qualified service providers can create 3D maps of large areas and recreate as-built documentation when it has been lost (or is insufficiently precise for destructive work, as is often the case with PEX). As another example, in the past I spearheaded a project in which we mapped tens of thousands of square feet in an abandoned / unfinished building prior to remodel. The as-builts were long gone with the original owner's bankruptcy. With the piping layout, they could proceed with architecture and design.

I don't mean to solicit business with this post, but to raise awareness that this technology is available. If you have any questions about it, please visit our site or ask here and I'll be happy to answer or try to help you find a qualified service provider in your area.

[barbender]

I used Liquid Nails on all of my treated sill plates to attach them to the radiant slabs (other than my exterior walls). I ended up making a mistake and having to pull one up, it was splintered wood and concrete chunks, about 1/4" of the concrete surface was still stuck to the sill when I finally got it pried off. For non-loadbearing walls, I would use this method again. Otherwise, like someone else suggested, fire a 3" nail into the concrete- it will hit PEX for sure :D

[Kbeitz]

Put an E-bay super magnet on a wire snake and push it down
the pex line and follow it with a stud finder...

[Den Socling]

This is a suggestion for new PEX installations. When I put up the bridge crane in my shop, sure enough one lag bolt hit a tube. However, I had made multiple loops so I just don't use that one. It makes no difference.

Also, looking at old building plans don't give you much because when you're tying the PEX down, it has a mind of it's own.

[Saga Dan]

I know this is a resurfaced thread, but in my opinion the best way to find hydronic or electric heating or cooling loops in concrete floors or ceilings is to use an infrared camera.  This would be an actual camera not the thermometer type that is a point and shoot laser type with a digital temperature read out.  (much deeper discussion than necessary here and now)  The camera has the ability to "see" where the loops are based upon the the heat conduction through the floor or ceiling material.  This can be used in real time during a project to locate to can be used to located and mark the locations for a future time.  A drawback of this is the need to have the surface in a changing state of temperature.  For best results the system should be turned off for a day or so to allow the entire make-up of the components to come to a state that is different than what the intention of the system is, heating or cooling.  Once this has been achieved and while the person with the IR camera is present the system can be turned up.  The actual time it takes to really show up on camera is a short period of time, many times less than a minute.  Here is an image of electric heating coils under a tile floor.  From the time the system was turned on until this image was taken was less than 20 seconds.  Granted the wires are only in self leveling coat of concrete below the tile and thin set.