Have just learned that my carefully laid plans to put tubes under the subfloor won't work (for reasons I won't bore you with). We are now looking at putting the tubes on top of the existing subfloor (no appetite for removing / replacing this). What products are out there as far as grooved plywood or insulated panels, that we can lay on top of the subfloor and put tubing in? Thanks!

Roth Panels

WarmBoard

With either you'll need at least R10 of fiber or something  in the joists below for zone isolation or you'll have radiant-ceiling effect in the rooms below.  (You can blow in some cellulose or fiberglass from the top before you put down the new floor, if there isn't easy access from below.) If it's over a ventilated unconditioned crawl space you'll want at least R20-R30.

Looks like my plumber is recommending Watts SmartTrac

http://www.wattsradiant.com/products/smarttrac/

Anyone ever used this? Any advice on whether it's any good for a second-floor retrofit?

Have a look at our product as well, RHT floor panel system,
surface mount system, works very well in low temp applications.
http://www.blueridgecompany.com/radiant/hydronic/527/rht-floor-panel-system-for-radiant-heat-systems
Application runs about $3.00 sq ft complete. (pipe, heat plates plywood, manifolds).

Dan

Ask your plumber for a room-by-room Manual J and for the output of the "smarttrac" and design water temperature needed to meet the heat load during design conditions. If he gets past the first question, you might have something there.

In cold climate the aluminum-free panels often come up short for output in retrofit applications making an accurate heat load imperative. More likely Dana is no track and Dan always has the competitive edge for the savvy DIYer.

If a name means anything, the location is in Vancouver (either BC or WA). In either Vancouver the outside design temp is about 24-25F, so the loads are probably around half that of a typical project for BadgerBoilerMN's MN climate.

But you still have to run at least the napkin math on any hydronic heating design to know what's really going to be needed, starting with the heat load of the room. That wasn't really his/her question, but it's a valid point.

It's not just plumbing project, and not all plumbers are up to the task of doing the math on heating design (which is why I gave up and started doing my own hydronic designs. Crude as my methods may be it's way better than a WAG. Taking a WAG seems to be a common design methodology in the "plumpin' an' heatin' " trades in my neighborhood. YMMV.

I personally wouldn't go with a no-aluminum version on any plumber's say so, without doing the math on it first, unless they had a printout of the design tool using that product that included realistic heat load numbers for the room, and a water temperature. The ones with aluminum get the heat out of the tubing and into the room more effectively, and in a temperate climate like Vancouver it's probably safe to just punt on that, unless you have unusually lossy rooms, say rooms with large single-pane picture windows & un-insulated 2x4 wall construction, etc. (In which case fixing those gross heat leaks would add more comfort than installing a radiant floor.)

Dana is correct, even an above average plumber will be challenged to design a below average hydronic system. There is plenty of free DIY hydronic design info and software around these days to avoid having to make that mistake.

why dont you just lay the pex on the floor (open floor, any floor ) that you have now, set 2x4 16oc and around the perimeter and lay 5/8 floor on top. boom done simple easy as 1 2 3

option 2

lay pex use 2x4 fill gaps with sand (concrete if the weight can be supported) set floor, laminate

What you suggest would be a very inefficient HR emitter and I would not recommend it. Warmboard or something like it would be a far better solution.

Posted By sailawayrb on 23 Nov 2014 10:34 AM
What you suggest would be a very inefficient HR emitter and I would not recommend it. Warmboard or something like it would be a far better solution.

care to elaborate more on that? the area between the floors will be warm heating the floor on top same thing as placing pex under the floor in this case your just sandwiching the pex and if u cover it with sand would work same way as the slab Ohh Myyyy

Concrete is a much better thermal conductor than dry sand.

I never could understand the concept. The sand doesn't add anything to a radiant floor but weight.

Look to the many sandwich and sub-floor systems available on the market.

One of my first radiant floors--installed in the mountains of Colorado circa 1988--was polybutylene stapled down to the sub-floor and 2x4 sleepers placed to allow for hardwood flooring over a very early Gypcrete. It worked perfectly with design temperatures of 105 degrees Fahrenheit.

Concrete is common also, but heavier and really should be poured 2" thick.

Really what the others have indicated. Concrete is a much better conductor of heat than sand. Anything that adds R-value above the PEX reduces the heat transfer capability and the overall performance of the HR emitter. This is why below-floor HR emitter designs have the worst performance, above-floor designs being better, and concrete slab design being the best. You can make all of these HR emitters "work", but you must design them to align with the heat loss of the room/building and you will have to live with the resulting operational performance forever.