Radiant Floor heating within a finished ceiling/ floor.

[Brad_bb]

Here's a pic.  Where I have the wood bracket for holding the ceiling, you could also a)cut a groove in the timber to insert the panels, or b) space the floor up enough(like with 3 or 4 inch spacers) that you can set the ceiling panels on top of the floor timber.  Check out January and February 2007 of Thomas Massie's blog.  His floor system is slightly different from what I described - a framed 24 system on 12 in centers, but as he states, has zero bounce and gives him the space needed to run the tubing.  I don't see how he finished the ceiling below, but we can ask him. Brad

[ReidH]

I will be doing the following on my TF addition for heated flooring this year.

2x6 TnG subfloor on top of the joists.
 
3/4 inch plywood strips about 6.5 to 7 inches wide spaced to mount the heat transfer plates on top.

Heat transfer plates, such as Thermofin on top of the plywood with Pex-AL-Pex tubing.  The reason for the Pex-AL-Pex is that it has a coefficient of expansion that is equivalent to aluminum, which is what the heat transfer plates are constructed of.  If you use regular Pex, the expansion rate is much higher than the plates and when the tubing trys to shift position, you will hear squeaks.

Hardwood finish floor on top of that.

For insulation I am adding the following on the underside between the joists:

2 inches of rigid insulation protected with drywall on the underside of the subfloor.  The joists will look to be a little shallower than they really are, but you can use the 2"  insulation space to conceal runs of wiring for ceiling mounted lights/chandeliers as well.  You could use TnG instead of drywall for the finished ceiling as long as your rigid insulation is provided with adequate fire protection.

Reid

[Holmes]

Brad_bb  Pex tubing will probably outlast the house. The rubber hose type tubing will not.  Your design is a good one as is ReidH's. 
Heat transfer plates are great at getting the heat out of the tubing and into the wood and is the best way to go if you are not using concrete or gyp-crete.  Holmes

[Brad_bb]

I've learned the hard way that if you don't build in a way to service it, then for sure there WILL be a problem and you'll have to tear it all up.  I wouldn't trust it for over 100 years.  We have houses that have lasted longer than that, and if it got to that point, I'd want to be able to service or replace the system.  Though, we'll probably have home fusion heaters by that point that are smaller than a toaster.

[witterbound]

Do they still make radiant hot water heaters that hook up to a boiler, like in the old days?  It sure seems like this would be a better way to heat your upstairs, than going through all you're planning under the floor.

[Thehardway]

Radiant hydronic heating wood floors makes for a very comfortable house if done correctly, however, since wood is a relatively poor conductor of heat, it is not a very efficient or economical means of heating and unless you have a low cost fuel source (such as waste wood products or solar collector).  Locating piping too close to the floor surface can causes a "striping" effect.  The heat does not spread out effectively as it conducts through the wood, this is often the case with floors that utilize channels or strapping on the sub-floor and directly under the finish floor.  Water temps must be kept very high for radiant heat through wood 120 ^ because of thermal inefficiency.  This can cause excessive shrinkage and movement or warping.   Too low in the floor and it bleeds through the ceiling below as previously mentioned.  This causes more inefficiency and causes rooms below to get too warm in the head space whilst cold at the feet.  This is bad.  You can see a delicate balance is required.

You would probably be better served and it would be more economical to use hydronic radiators for such an application and strategically place them in the space to be heated and allow natural convection to do the rest.

I like radiant floors under concrete, gypcrete or tile but I have reservations about under wood. 

At temps needed for radiant through wood, you will need to use oxygen barrier pipe or use all non ferrous circ. and heating devices to avoid problems as well.

[DouginUtah]

I agree with most of what Thehardway has said . I used PEX in my basement slab spaced at 1', which turned out to be overkill--it would have been overkill even at 18". But for my main floor I used slantfin  (www.slantfin.com) baseboard heaters (at about 3/4 the lengths recommended by their heat loss program) and if I set the thermostat at 70 it never varies more than 1/2 degree. This is for new construction with very good insulation.

[Holmes]

 I do not agree with Thehardway. Radiant heat is a very efficient way to heat a house if it is installed properly.  Striping is caused by poor installation. water temps in the tubing above 120* f are the result of poor installation. Systems with these problems need heat emission plates or suplemental  heat, like X amount footage of baseboard  radiation.   Radiant heat does not cause shrinkage in the flooring, a lack of moisture in the wood causes shrinkage.  Unless you install humidification in the dwelling the wood will shrink no matter what you have for heat.
  Baseboard  Radiation is a good way to heat.. It cost a lot less than radiant heat to install and if you want to use lower water temperatures just install twice the amount of base board radiation and run 130* water thru it...For every 3 degrees the water temp. is lowered you have a potential for 1% savings in fuel.  Holmes
 

[beenthere]

I'd have to side more with TheHardway on this. Holmes has missed a few points in his side of the discussion. But thought TheHardway outlined it quite well.
No offense meant to Holmes, just sayin.

[Thehardway]

Holmes,

Not trying to pick a fight here.  I agree with most of what you have said but there are a few key things that don't ring true.  Efficient radiant heat depends on goodly amounts of thermal mass.  Aluminum, steel, concrete, tile etc. all do very well at absorbing, conducting distributing and radiating heat over a period of time and space.  In radiant applications this means that the tubing can heat the floor that the tubing is in contact with and once it is up to temp it will stay there for a significant period of time.  It does not lose it's heat rapidly like air, and therefore does not require constant heat input and circulation.

Hardwood works better as an insulator than as a conductor or thermal storage mass. It does not absorb heat well nor conduct it well.  It burns before it heats through. This is precisely why they used to put wood handles on pots and pans.  Barbender relates an experience representative of this.  Can it be done? Certainly, but it is not near as efficient as it should be and not from lack of proper install, rather the property of materials.  You won't find many wood radiators out there nor wood heat sinks because of this particular inefficiency.  If one insists on warm floors, then it can be done but it is not near as economical as some other heating methods and they must be willing to accept the tradeoffs.

The other point I might make is that any honest wood floor installer or radiant heat installer will tell you up front that you can expect significant seasonal movement in the floors and there are several reasons for this.

The most obvious one you have already pointed out but with the wrong conclusion.  Humidity.

In a house heated with hydronic radiant heat, there is no consistant source of humidity.  If it is properly installed, the water stays in the tubing.  This means that the cool air in the home becomes extremely dry in the winter and heated by the floor rises as natural convection carries away the moisture in the wood very rapidly.  Conversely, when the heat is turned off, warm moist air in the house deposits its moisture at the floor which stays slightly cooler and condenses and absorbs the moisture like a sponge, swelling as it does so.

In order to adequately control the seasonal humidity changes, a secondary humidification/de-humidification system would be necessary.

Wood being hygroscopic in nature does not dry on its own past a EMC of approx. 10-12%.   EMC does change after EMC is obtained with changes in both relative humidity (a function of temperature) and to a lesser degree with temperature alone.  If relative humidity is not elevated as air temp is increased, wood will give up its moisture and become dimensionally unstable.
Those who kiln wood are keenly aware of these properties. 

Guidelines for installing hardwood floors over radiant heat do exist but meeting them to ensure a stable floor adds significantly to the cost of the floor.

Here are a few of the reccommendations:

EMC between 6-8%
t&g boards less than 3" in width
beveled edges
quatersawn boards
thinner boards better than thicker
acclimate wood and install at average seasonal relative humidity
provide for seasonal alternative humidity control system

This is not my opinion but that of flooring manufacturers.

I agree you can run water temps lower to avoid striping effects but you also limit your overall output and increase cycle times once again decreasing efficiency.


I love radiant heat, I hate squeaky or warped hardwood floors ;D

[Don_Papenburg]

Rooster , My plan is for 3/4 sub floor  5/8 strips 5.4" wide . that gives me a 6" tube spacing  using 1/2 pex.       I would not space tubes more than 6" apart .  As was mentioned previous it would leave a thermal striping effect.  I would like to run 3" under wood but I can't get a bend tighter than 6" .      I have my basement on 6" centers over 2" EXP with 4" concrete over.  I keep the basement at 70 F. and it will keep the first floor 60 F.
The six inch tube space lets you run a lower water temp.  I try to keep mine at 120 in and 100 out. 

5/8 ply is about the dia. of 1/2 pex .  I think that you could lay the all foil over the top with the 5/8 ply strips  because that will let the foil contact the tube without the U shape.

[Rooster]

I want to thank everyone for the "lively" debate.  I will be working with a local heating contractor, and I have asked him to take a look at this thread.  I will post the details of the the system that we will be using, and I encourage others to post their radiant floor projects along with the pros and cons of each application.

Thanks again,

Rooster

[stumpy]

Hi Craig. Go on line to a company called Radiantec.  They are very helpful and they will put together a heating "scheme" including parts list for free.  Granted, yours is not a typical installation, but I found them very helpful and accommodating.  They also sent me a packet of info along with the parts list.  There's a booklet that is very informative about the different method of radiant heat.  I was going to build a shop building with radiant heat in the floor.  They highly recommended using a standard LP water heater.

[Holmes]

I was going to build a shop building with radiant heat in the floor.  They highly recommended using a standard LP water heater.

The water heater can do the job but they are only 70% efficient. A good boiler can be 95% efficient. You could be losing 25% of your btus out the flue, so very every $4 you spend on a gallon of fuel $1 goes out the flue. Give it time and cold weather and a lot of money will be thrown away. Holmes

[D Hagens]

  I've installed miles of pex for radiant heating and it's an awesome and very efficient way to heat ones home 8) 8) This is a very detailed discussion and there's some very good points brought up.
Now here's something to think about.........I would never install this in my house ever The reason why is because I don't like the feel of the heat, it's a different feel to me compared to forced or the above floor radiant that I've been used to.
It's always a question I will ask a customer before I will install the product.

[frwinks]

ha, I found it.  The detail I tried explaining in my earlier post came from Rupert Newman's Oak Framed Buildings.  I just stumbled across it again this weekend and remembered this thread. What I like about it is the no nonsense approach, just the stuff you need without redundant layers of ply, stapping, nailers, etc..

[Brad_bb]

Too small Rafaelfrwinks.  We need a cropped close up please.

[frwinks]

hope this one is better ;)