GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon
Q&A Spotlight

Heating a Bathroom Floor

Is cork a good match for an electric radiant floor?

Schluter Ditra Heat is a polyethylene membrane that uncouples the subfloor from the finish floor and allows the addition of electric heat. It's one option under consideration for warming a bathroom floor, although the homeowner is considering the use of cork rather than tile as a finish surface.
Image Credit: Schluter Systems

A warm bathroom floor is a something to look forward to on a chilly winter morning, and C.L. is poking around for ideas on the best way of accomplishing that.

One option is installing a grid of electric cables beneath the finish floor in tandem with a polyethylene underlayment manufactured by Schluter Systems called Ditra. These installations are often topped with ceramic tile, which is impervious to water damage and readily transmits heat from the buried cables.

But C.L. has another idea.

“In regards to finish flooring over the heated floor in the non-shower part of the bath, cork looks like an interesting product,” C.L. says in a recent Q&A post. “Although cork is sometimes discussed as an insulator, the [manufacturers] of solid cork flooring propose it as an ideal finish floor for a heated floor; supposedly it heats fast.

“Does this sound reasonable, or is this just marketing hype?”

That’s the topic for this Q&A Spotlight.

First, how efficient is electric heat?

C.L. begins his post with a question on whether there’s a recognized metric for measuring the efficiency of electrically heated bathroom floors. Actually, there is, points out GBA Editor Martin Holladay.

“Yes,” he writes. “The metric is called ‘efficiency.’ It describes the conversion of electrical energy into heat energy. All electric-resistance floors have the same efficiency, namely 100%.”

This is technically true of all electric resistance heat, whether it’s in the floor or not. But Jon R suggests that definition doesn’t go far enough.

“A reasonable definition of efficiency would involve useful work,” he says, “so I’d exclude any heat lost to the underside of the floor. This efficiency will be less than 100% and cork will lower it (as compared to something more thermally conductive like tile).”

Would it make any difference if the radiant heat in the floor were provided by a hot water loop installed beneath the subfloor, C.L. asks. One advantage of this option, he adds, would be the opportunity to replace the finish floor in the bathroom at a later date without affecting the heat distribution system.

“I think this would be less efficient as the heat would need to transmit through the subfloor,” C.L. adds. “The subfloor material probably has an impact — temperaturei.e. Advantek vs. OSB vs. plywood. Is there any simple way to calculate the efficiency hit of going through that additional layer?”

There’s no need to make it that complicated, Holladay says.

“As long as your home’s thermal envelope has adequate insulation, an electric-resistance heating pad or PEX tubing installed as part of a radiant-floor heating system aren’t less ‘efficient’ if there is a thick subfloor or inappropriate flooring. The heat remains indoors, so it isn’t ‘lost.’ The problem is that a floor assembly with a thick subfloor or inappropriate flooring is less responsive, and takes longer to heat up, than a floor assembly with well-chosen materials. Moreover, the heat may end up in a different room than intended (the room below the floor assembly).”

Understanding heat delivery

It’s not a hydronic system C.L. system has in mind, but an electric under-floor mat made by SunTouch. How effective would that be in delivering heat to the floor?

Dana Dorsett suggests C.L. look for a chart provided by the manufacturer of the electric heat that describes the amount of heat it can deliver through different types of subflooring and flooring.

“If this system is not temperature-controlled but has a watts per square foot spec or watts per length spec, the R-value of the subfloor + floor don’t matter as much as the ratio of the R-value of the floor materials to the R-value of the insulation below,” Dorsett says “If it [is] R-1 of floor materials to R-20 of insulation, about 95% of the heat will be coming through the floor. If it’s R-2 flooring to R-11, something like 85% will be coming though the floor. To convert watts to BTU/hr, multiply by 3.412.”

He uses as an example the system C.L. has referenced, which delivers 72 watts for a 6-foot-long section or 12 watts (41 Btu/hour) per running foot.

“If it’s between joists 16 inches (1.33 feet) on center it can deliver 41/1.33 = 39 BTU/hr per square foot of floor going into the system, but only part if it is going up,” Dorsett says. “Some is going down through the insulation. If it’s only delivering 85% of it up, the rest [is] going through the insulation and joists. It’s about 33 BTU/hr per square foot coming through the floor, and the surface temperature of the floor will be about 16 F° warmer than the room temperature. In a 75°F bathroom that would be a bit north of 90°F, which is warmer than most people like on bare feet, but not super uncomfortable.”

At 50 Btu/hour, the surface temperature would be about 25 F° warmer than the room temperature. “On a tile or stone floor in a 75°F room that can be pushing the limits for barefoot comfort,” he adds.

Is cork an appropriate floor finish?

The Schluter Ditra system is typically used with tile, not cork, says Holladay. “I’ve never heard of anyone installing cork flooring above this type of heating mat,” he says, “and I’m skeptical as to whether it’s a good idea.”

Steven Knapp has similar concerns, adding that the cork products he’s researched are not recommended for bathrooms because they can swell and buckle when they get wet.

C.L., however, says that a manufacturer of cork flooring actually recommends the Schluter Ditra system topped with cementitious layer made by Ardex.

“This provides a waterproof membrane (Ditra) and a cementitious ‘subfloor’ (Ardex),” C.L. says. “Then install their cork tiles with contact adhesive onto the Ardex. This also allows for a future finish floor replacement without destroying the heated floor — you scrape the cork tiles off the Ardex.”

C.L. also makes a distinction between solid cork flooring and engineered cork flooring in which a top layer of cork has been applied on a backing made from a different materials.

“In regards to cork being unsuitable for wet areas due to swelling and buckling, wouldn’t that only apply to engineered cork products on a backing that would swell?” he asks “The [manufacturer] does warn that their engineered floating product is not suitable for wet area installation. They have no such warning on their solid cork product. Solid cork tiles have air pockets; would that preclude or reduce swelling?”

If the manufacturer warranties a glue-down, solid cork tile, there’s no need to worry about it, Knapp replies, although it would be smart to check whether the backer plus the tile will create an “awkward elevation change” between the bathroom floor and any adjacent flooring.

How much floor should be covered?

Whether C.L. uses cork or tile as a finish floor, there’s still the question of exactly what parts of the bathroom should be covered. Should the shower floor, for example, be included?

“We have Ditra heat under our bathroom tile floor,” says Stephen Sheehy. “It works very well. We generally turn it on 30 minutes or so before a shower. I suspect tiling your entire floor, with heat underneath, won’t be more costly than doing part with tile and part with something else. You wouldn’t need to manage the transition between the two floor types.”

In Sheehy’s bathroom, the shower is included. The shower is open to the rest of the room, with the floor in the shower sloped toward a linear drain near the wall. “There’s no shower door or partition,” he says. “The whole floor is heated. When I shower, I usually don’t bother with turning on the floor heat, but my wife likes it.”

One caution about the extent of under-floor heat comes from Dorsett: Do not put radiant floor under a toilet. The heat could melt the wax seal connecting the toilet with the drain line.

And then Peter Engle made this suggestion: “You do want to run the radiant at least under the toekick if you have standard vanities, or about 6 inches past the front if you have furniture-style vanities (open bottoms with furniture feet),” he says “Otherwise, your toes are touching cold tile when you brush your teeth. Voice of experience, here.”

Our expert’s opinion

GBA Technical Director Peter Yost added this:

Here are a few thoughts on this thread, particularly about thermal comfort:

Thermal comfort versus energy efficiency: For C. L., with a central HVAC system covering loads in the bathroom, the radiant floor heat is all about thermal comfort.

Thermal comfort of feet: ASHRAE Standard 55 gives this range for thermal comfort of feet in shoes as between 66.2° and 84.2° F. That’s not terribly helpful for a bathroom floor around a walk-in shower. This paper provides more information (see Table 5). It’s interesting that the “comfortable” temperature range for a concrete floor is narrower and higher (78.8° to 83.3°F) than for a cork floor (73.4° to 82.4°F) and even narrower and higher for a marble floor (perhaps the most like ceramic tile (82.4° to 85.1°F).

Flooring contact coefficient and foot thermal comfort: The most interesting paragraph to me in this paper came under the start of “Discussion” section:

“It is apparent from the series of experiments performed with 16 persons that it is not possible to find a floor temperature where all persons are satisfied. Neither is it possible to achieve less than 2% dissatisfied for short periods of occupancy (1 min) nor less than 11% dissatisfied for longer periods (10 min). These values can be attained when the temperature of the floor is optimal, i.e., that temperature which causes a group of persons occupying the floor on an average to evaluate foot comfort as neutral (voting = 0). If the floor temperature deviates from the optimal, the increase in the number of dissatisfied persons will depend on the flooring material. For floors with a small contact coefficient (e.g., cork, wood) the increase in the number of dissatisfied-persons will be moderate compared to floors with a large contact coefficient (e.g., concrete, stone).”

In additional experiments covered in this paper, with occupants standing and seated and with “light” clothing on, the change in clothing had little or no significant impact on comfort results.

Service life and performance of grouts: I have done quite a bit of ceramic tile work over the years and used both epoxy and polyurethane grouts. I have found both superior in terms of watertightness and stain/discoloration resistance but have found the polyurethane grouts a bit easier to work with.


  1. Eric Sandeen | | #1

    Efficiency aside, how much energy?
    We opted out of a heated bathroom floor during our remodel. I couldn't find any data on how many kWh/year we could expect it to consume, not even a ballpark. Anyone know?

  2. Malcolm Taylor | | #2

    If the floor heat is only used during the period when you are supplying heat to the rest of the house, wouldn't the only additional expense be the difference between the efficiency of the two heat sources? If the rest of the house was supplied by similar resistance heat,would there be any additional cost at all?

  3. Malcolm Taylor | | #3

    Product Warranties
    Relying on product warranties can bring a lot of grief. They generally cover manufacturer's defects and problems that occur when the product is installed according to the supplied instructions. But what they warranty is usually just the product itself. So you may get the damaged or defective material replaced, but the labour to do so, the work necessary to integrate the material into the rest of the house, and the cost of making good surrounding or supporting materials isn't generally covered.

  4. User avater GBA Editor
    Martin Holladay | | #4

    Response to Eric Sandeen
    Dana did the math for you -- the answer is 11.4 watts per square foot in the example he gave.

    For an annual estimate, you would simply multiply 11.4 watts times the square footage of floor heat, times the number of hours per year that your family wants to run it, divided by 1,000 to get kWh.

  5. User avater
    Dana Dorsett | | #5

    That's only it's maximum output @ Martin, Eric
    Assuming there is at least some thermostat in the controls it would still have a duty cycle. Even when in use, and would not be running the 11.4 watts per square foot continuously, but would be cycling on/off based on the room (or floor) whether it had reached the thermostat's setpoint.

    Michael's estimate of 2kwh/per hour would mean it's running a constant 2000 watts (=6824 BTU/hr), which is a HUGE amount of heat, several times the heat load of most bathrooms even at the 99% outside design temperature. The heat load of the main bathroom at my house is about 1100 BTU/hr when it's -10F outdoors, +70F indoors, and that's before subtracting out the heat emitted by the lights, or the shower/bath. OK, so I have a small bathroom, but even if the design heat load were 3-4x bigger, 2000 watts would be extreme overkill, and it would take (2000/11.4=) 175 square feet of electric radiant floor to emit that much heat.

    Note the arithmetic error in my quoted forum post creates overly optimistic expectations):

    " can deliver 41/1.33 = 39 BTU/hr per square foot..."

    Uhh... make that ~31 BTU/hr per square foot, not 39, which is ~9.1 BTU/hr per square foot, not 11.4, so it would really take about 220 square feet of radiant floor to deliver 2000 watts of heat into the room. The per square foot estimates that followed also have to be scaled back, since they're based on the original arithmetic error (eg, With only 85% of the heat going up and 15% losses to elsewhere would be 0.85 x 31= ~26 BTU/hr per square foot.)

    Whatever the real numbers for C.L.s bathroom are, even if it's heated solely by the electric floor and on continuously under thermostatic control it's not likely to hit 500 kwh /year unless it's a gia-normous bathroom with a large amount of window area (which isn't very common for bathrooms) adding to the heat loss numbers

  6. User avater GBA Editor
    Martin Holladay | | #6

    Response to Dana Dorsett
    Thanks for your corrections and improved math. Very helpful.

  7. Adam W | | #7

    A friend put in the Schluter system a while back and loves it. Just wishes he had upgraded his electrical line powering the heated floor to 240V.

  8. Eric Sandeen | | #8

    Thanks for the detailed answers on expected energy use. ~500kWh/ year isn't too bad I guess; I was hung up on electric resistance heat as a source compared to our efficient boiler... in retrospect it might have been a good idea. I don't think I'm going to revisit this decision with my wife... :)

  9. User avater
    Dana Dorsett | | #9

    But it really WON'T be using 500 kwh/year @ Eric Sandeen
    I know it's only my hacked version of 'merican, but parsing it correctly helps:

    "...even if it's heated solely by the electric floor and on continuously under thermostatic control it's not likely to hit 500 kwh /year unless it's a gia-normous bathroom with a large amount of window area..."

    The key phrase being "'s not likely to hit 500kwh /year..."

    The napkin math on this for MY house looks like this:

    At an indoor to outdoor temperature difference of 70F the heat load of my sub-code circa 1923 bathroom is about 1000 BTU/hr.

    1000 BTU/hr =293 watts or 0.293kw

    So in a 24 hour day at an indoor to outdoor temperature difference that big it that's (0.293 x 24=) 7 kwh.

    That's also about 65 heating degree days (base 65F), but for the sake of making the math easy call it 70 HDD.

    So the power use per HDD would be about (7kwh/70 HDD=) 0.1 kwh / HDD.

    If my bathroom were located a 5000 HDD climate it might hit (5000 x 0.1=) 500 kwh/year, if the radiant floor was the ONLY heat source, was on ALL the time, the lights were never on, and nobody ever entered or bathed there, and the door was always closed. (As it happens I'm in a ~7000 HDD climate, not that it really matters.)

    In a real bathroom in a house with central heating there are other heat sources- from the main heating system, hot water use, lights, and occupants, all of which would offset the power used by the floor. Thus the power used by the floor is "... not likely to hit 500kwh /year..." even if it might hit or even exceed that in a worst-case scenario. If the radiant floor is only on when the bathroom is occupied it will be less than 100 kwh/year, maybe even less than 50 kwh/yr even if the heat loss of your bathroom is twice mine.

Log in or create an account to post a comment.


Recent Questions and Replies

  • |
  • |
  • |
  • |