*
In an earlier post, Mongo said he had great results stapling PEX under the subfloor and not using those reflective metal plates the manufacturers want to sell you for big $. Instead, he used reflective rigid insulation under the stapled-up PEX.
That’s great, but I have concerns of the PEX’s being punctured by the nails used to install the hardwood floor. We can’t install the PEX after the hardwood floor is installed, and I don’t really want to go to the expense of having an exceptionally thick subfloor.
Now, very timely, I see on page 83 of the July 99 issue of FHB a drawing of PEX stapled to the sides of the joists, rather than the subfloor. If this application is as efficient for conducting heat, it seems to be the solution for my situation. Any opinions? The heat still has to go up, but is there a loss if the PEX isn’t right up against the subfloor? And if I have to have a thicker subfloor in order to staple the PEX to the subfloor, I’ll have some loss of heat there, too, won’t I?
This installation will be in the Portland, OR area, where’s it’s not all that cold in the winter anyway, so my heating requirements aren’t as great as those for you guys in Canada and Vermont, etc.
Replies
*
Use shorter flooring nails.
*Forgot to point out the savings in labor if the PEX can be installed from above before the subfloor goes down.
*Rufus, I've never installed PEX as shown in that drawing. I'm sure there is a loss in efficiency in the installation, how much of a loss I do not know.With a staple-up installation you can install the tubing runs 8" on center (typical), or as required. In this intallation it looks like they're placed 16" OC, on the joists. Do they put a tube on each side of the joist to bring it to an average of 8" OC? I don't know, as the schematic is incomplete. Maybe with their "energy-efficient" construction they require less tube. That I can't answer. As to the tubing being an increased distance from the subfloor, I'm sure that generates a bit of additional heat loss. With the tubing on the subfloor you get a bit of conduction along with convective and radiant heat transfer. In this application you lose the conduction. Is it great? Will you notice it? Once again, the loss will be there, but as to the amount and will it be noticeable, I can't put a number on it. If the thermostat is set at 70 degrees, the room will be 70 degrees. Their application may result in the furnace running a few extra minutes a day, but the comfort should still be there.In essence, the proof of application is in their success with the installation. It works for them. They may have lost a bit in efficiency, but it still works. They claim that they use 550 gal of oil per year to heat about 1700 feet of living space. As a comparison, in my own house I used about 930 gal over the past 12 months for 3200 feet. (I have 4100 feet total, but the attic, 900', while "heated" is unfinished. A few, but not too many BTU's are spent up there.) As a very rough comparison, halving my house size to 1600' (theirs was 1700') would halve my oil consumption to about 465 gal per year, compared to their 550 gallons.My house walls are "standard", cedar claps, 1/2 ply sheathing, 2x6 w FG, poly, 5/8 drywall. Windows are Marvin woodies, double pane w argon, etc. A tight envelope, but not as fancy as theirs. My house is also in CT vs VT. Not as high a heating demand for me or the other houses I've done down here.So...It works for them, I'm sure it would work for you. A slight loss in heat transfer, I'm sure, but at what cost I don't know. I'm also not sure if you'd notice "warm" and "less warm" strips on tile with the tube 16" OC. One other idea, which may seem wasteful or ridiculous, but I'll throw it out to you regardless: Increase the "thickness" of the subfloor by adding strips of 3/4" ply under the subfloor and attach the tubing to these strips. It will effectively increase the thickness of your subfloor to 1.5" (3/4 plus 3/4) in the areas where the tubing is. Use, 3/4", 5/8", whatever you need for nail protection. Cost-wise, in a 1600' house at 40 feet square, you'd have about 1200 linear feet of joist bays with the joist 16"OC. You'd need 2400' of strips at two per bay. Using 3" wide strips would consume about 20 sheets of ply or OSB. You can figure the cost depending on what you use. It may be a total waste of money, but it's an idea nonetheless.One final item, if you do go with the tubing on the sides of the joists, keep them as high as possible so other trades don't drill or sawzall them when installing the mechanicals.Let me know how it all works out in the end.
*Mongo,Thanks so much for responding. I've been throwing all sorts of literature and print-outs of threads in this forum at my plumber and builder. Now we just have to pick a method. I'll be sure to let you (and anyone else who might be interested) know what happens. I've already sold them both on the staple method...
*Rufus, My own opinion, but to me a "compromised" radiant installation is still better than forced hot air. You'll love it.
*Mongo and Rufus,I posted a response yesterday to your question Rufus. Somehow it did not appear or was lost in a Taunton server glitch.To make matters short (I will explain my position below (and much more) before the weekend is over.Here goes:The amount of tubing, whether you use single-tube between the joist[regardless of OD], reverse-return[8" OC), plate, plateless nor the supply water temperature determines the efficiency of the system as far as the amount of fuel consumed. Efficiency is based upon the DeltaT(temperature difference between the fluid temperature exiting the HEAT SOURCE and the temperature of the fluid returning to the HEAT SOURCE multiplied by the burner efficiency and amount of fuel consumed reduced by the standby and or/stack loss of the heating appliance).A radiant panel that is sufficiently insulated against "backlosses" could operate at 5 or even a 50 degree temperature difference and not affect "fuel usage". All is relative to the the load of the building and the capacity of the heating plant. Panels taking advantage of "greater conductivity and heat transfer" only satisfy the demand sooner. Faster conductivity may result in greater comfort(satisfy demand) or overshoot(over satisfy demand-decreasing comfort) sooner but may also? result in a shorter burn cycle thereby increasing stack losses and in essence lower efficiency of the heating plant and result in fewer BTUH's actually remaining within the liveable structure. Most heating appliance efficiencies are based on "steady-state firing cycles". Holy sh*t, it's been quite a day and I am may get bushed[if I'm lucky or beg]. If anyone really gives a hoot, let me know. I will try to do a better and more complete job of explaining why "everyone may be right" based upon particular/specific scenarios. Be forewarned though, I do not speculate on generalities. Radiant floor heat is subject to many different variables. I have a rather modest experience and understanding of those variables.I dislike "rules of thumb" just as much as carpenters and other true craftsmen disapprove of "eyeballing it" and using a carpenter's square to layout a 40' x 60' foundation.Jeff