So-called radiant floors have an excellent reputation. Many customers report that this type of heating system is comfortable and quiet. Moreover, some suppliers of radiant floor materials and equipment claim that these systems can save energy.
In spite of the purported benefits of this type of heating system, few green homes include a radiant floor heating system. This article will explore why.
What shall we call these systems?
We’ve all seen ads for this type of heating system, including photos showing a barefoot mom watching her baby crawl across the floor. Using photos like this as a guide, is it possible to describe the heat transfer mechanisms in such rooms?
The mother’s bare feet are being heated by conduction. The air near the floor is also being heated; as the warm air rises to the ceiling, it creates a convective loop. So the room is being heated by convection. And, if the floor is warm enough, the mother’s bare arms are being heated by radiation.
In other words, all three heat transfer mechanisms are at work. So why is this a “radiant floor”? The phrase “radiant floor” is misleading, and should be abandoned. It’s more accurate to say that this floor has “in-floor hydronic tubing.”
Three ways to warm up your flooring
There are three types of heated floors:
A hydronic heating system has a boiler rather than a furnace. (Boilers heat water, while furnaces heat air.) While most hydronic systems distribute heat using fin-tube baseboard units or wall-mounted radiators, some use in-floor hydronic tubing.
It’s also possible for a hydronic heating system to use a water heater rather than a boiler to heat the water.
The main disadvantage of hydronic heating systems is that they don’t provide a convenient way to integrate air conditioning. (Most air-conditioned homes in the U.S. use ductwork to distribute cool air; since these homes need ducts for cooling, they usually use the same ducts for heating as well.)
Ironically, some homes with in-floor hydronic tubing use ductless minisplit heat pumps for cooling — which raises the question, “Why not just use the ductless minisplits for heating, too?”
Where does the tubing go?
There are at least six different ways to install an in-floor hydronic system:
- Slab-on-grade systems use hydronic tubing that is embedded in a full-thickness concrete slab placed over rigid foam insulation.
- Systems with thin slabs over wood framing have tubing that is attached to the top of the plywood subfloor, and is then covered with a thin slab (usually 1 1/2 inches thick) of Gyp-Crete or conventional concrete.
- Systems with aluminum heat-transfer plates use metal plates with a groove designed to accept tubing. In these systems, the plates are stapled up against the underside of the subfloor, working from below.
- Plateless staple-up systems use tubing that has been stapled directly to the underside of the subfloor, without any aluminum heat-transfer plates.
- Above-floor tube-and-plate systems use heat-transfer plates which are installed on 3/4-inch sleepers nailed above the subfloor.
- Radiant subfloor panel systems use grooved plywood or composite panels; the panel grooves are designed to accept snap-in tubing. There are several brands of these panels, including , , and .
Alex Wilson’s landmark article
The best analysis of in-floor hydronic systems from a green perspective was written over a decade ago by Alex Wilson. His classic article, was published in the January 2002 issue of Environmental Building News.
I am indebted to Wilson, as well as to Marc Rosenbaum and Andy Shapiro, the two energy consultants whom Wilson interviewed when researching the article, for laying the technical groundwork for most of the arguments I’ll be making in this article.
Is there any evidence that these systems save energy?
Proponents of in-floor hydronic tubing have suggested several mechanisms by which these systems could contribute to energy savings.
Do occupants of homes with in-floor hydronic tubing choose lower thermostat settings? Some people claim that radiant floors are so comfortable that occupants voluntarily set their thermostats to a lower-than-usual setting, thereby saving energy. Suffice it to say that there is no evidence that this is true; the best research study on this issue () found no evidence to support it.
Except in very cold weather, this type of heating system only has a “warm floor” for a few hours a day, or will circulate water with a relatively low temperature, so it stands to reason that most occupants will adjust their thermostats until the indoor air temperature feels comfortable — just like occupants of homes with different heating systems.
A house with in-floor hydronic tubing can select a lower boiler temperature than a house that uses fin-tube baseboard. That’s true, but a lower boiler temperature won’t result in significant energy savings. As Alex Wilson wrote in his 2002 article, lower boiler temperatures “might reduce heat loss into unconditioned space if boiler and piping are located in an unheated basement, but experts … suggest that the savings would be very small at best — especially because of the additional electricity consumption to operate pumps for long hours.”
Homes with hydronic systems may have lower rates of air leakage that homes with forced air systems. Strictly speaking, this point applies to hydronic heating systems in general, rather than specifically to homes with in-floor tubing. The point is nevertheless worth addressing.
A forced-air system with leaky ductwork can easily interact with a leaky building envelope in a way that increases infiltration and exfiltration. When this happens, the problem is due to the leaky duct system and the leaky thermal envelope rather than the forced-air system itself. It’s perfectly possible to design duct systems and thermal envelopes that don’t suffer from this type of leakage.
On the other hand, maybe these homes require more energy
If a house has in-floor hydronic tubing installed in a slab on grade, the house may have more heat loss to the ground than a house with a forced-air heating system — especially if the contractor didn’t install enough rigid foam under the slab (an all-too-common problem). After all, the concrete will be warmer than usual — for several hours a day, the temperature of the concrete will be above the room air temperature — potentially increasing the rate of heat loss from the slab to the soil.
Fortunately, there is a relatively simple solution to this problem: make sure to include lots of sub-slab insulation under heated slabs — between 4 and 6 inches, depending on your climate.
Disadvantages to in-floor hydronic tubing
These systems are slow to warm up and slow to cool down. Most homeowners prefer a heating system that responds quickly to thermostat changes. If your house is chilly or overheated, there is no advantage to having to wait several hours for the heating system to respond to a thermostat adjustment.
Equipment manufacturers have cleverly designed complex controls to address the fact that in-floor hydronic systems are slow to respond to thermostat adjustments. In many cases, these controls include outdoor temperature sensors as well as sensors embedded in the concrete slabs; the information from these sensors is used to adjust the boiler water temperature in anticipation of future conditions. While these sophisticated controls go a long ways towards addressing the slow-response problem, they add complexity and expense to this type of hydronic heating system.
Why isn’t my floor warm? Homeowners who look forward to walking barefoot over warm floors are often disappointed by homes with in-floor hydronic heating. That’s because these floors are rarely as warm as most homeowners expect.
Wilson described this problem in his 2002 article. “Heat is transferred from an exposed slab to the [indoor] space at a rate of about 2 Btu/ft2•hr•°F,” Wilson wrote. “In a well-insulated house, this rate of heat flow means that even when it is very cold outside, the slab can only be a few degrees warmer than the rest of the room or the room will keep heating up. For a concrete slab to feel warm, however, it needs to be about 80°F. Thus, for most of the heating season, the greatest feature of radiant-floor heat — a warm floor — won’t occur.”
In many well-insulated homes, a “radiant” floor may be maintained at only 75°F — which is less than the temperature of your bare feet.
These systems are incompatible with passive solar design principles. Passive solar principles established in the late 1970s recommended the installation of a dark-colored concrete slab or dark tile flooring on the south side of a house. In theory, morning sun streaming through the south-facing windows would strike the cool concrete floor, heating up the concrete. Over the next 12 hours, the warm concrete floor would slowly lose heat to the indoor air. The intent was for the mass of the concrete to act as a thermal storage system — a flywheel — to stabilize indoor air temperatures.
This approach works best (is most efficient) when a floor slab is cool during the early morning hours. Cool slabs can store more solar heat than warm slabs. Traditional passive solar homes work best when the occupants accept a wide range of indoor temperatures — from perhaps 60°F in the early morning to 80°F in the middle of the afternoon. If homeowners can tolerate these swings in indoor temperatures, and if they live in a location with lots of winter sunshine (for example, in Colorado or New Mexico), a passive solar home will have low energy bills.
If a home has in-floor hydonic tubing, however, the heating system is likely to turn on at 3:00 a.m., when outdoor temperatures are cold. When the sun strikes the floor at 9:30 a.m., the concrete is already warm, and therefore unable to store as much solar heat as it would have if it had started out cold.
Wilson discussed the problem in his 2002 article. “If a concrete slab is ‘charged’ with heat [via hydronic tubing] during the early morning hours and the surface is warmed to the point where it cannot readily absorb solar radiation striking it, that solar heat will more directly heat the air, increasing the risk of overheating,” Wilson wrote. “This isn’t a huge problem with radiant-floor heating systems, but it may mean that homeowners will have to open windows periodically in the winter and their overall energy savings from solar energy will not be as great. [Andy] Shapiro counsels against the use of radiant slabs in areas of houses with passive solar heat. ‘It’s a waste of energy,’ he says, though just how much waste occurs is unclear.”
These systems are expensive. In-floor hydronic systems cost significantly more than forced-air systems. That’s why they are more likely to be found in custom homes than in tract homes.
If you’re building a house, and are considering spending an extra $5,000 or $8,000 to upgrade from a forced-air heating system to a hydronic heating system, a strong argument can be made that you’d be better of spending the extra money for more insulation, improved air sealing, or better windows. Once you’ve done that, your heating needs will be lower, and it will be easier than ever to satisfy your home’s heating needs with a simple HVAC system.
Over a decade ago, Marc Rosenbaum explained the principle this way: “It just doesn’t make sense to put in a $10,000 heating system to provide $100 worth of heat per year.”
If you take some of the money you saved by not installing in-floor radiant tubing and use that money to buy thicker subslab insulation, you’ll get most of the warm-floor benefits you seek. As Andy Shapiro noted, “A house with a good enough envelope to be called green — well-insulated and tight — will have a very high level of comfort no matter what type of heating system is used, as long as that heating system is well designed.”
A case study in Massachusetts
Many designers of low-energy homes have learned the hard way about the disadvantages of using in-floor hydronic tubing for space heating. One relevant example is a net-zero energy house built in 2007 in Colrain, Massachusetts, by Rural Development Incorporated (RDI), a nonprofit developer of homes for low-income and moderate-income families. The 1,350-square-foot slab-on-grade home includes a 3.2-kW photovoltaic array and 57 square feet of solar thermal collectors.
The designers of the home specified a hydronic heating system with in-floor tubing. The water for the hydronic system is heated by solar thermal collectors, with back-up heat provided by a propane-fueled instantaneous water heater. “On the design heating day [the coldest day of the year], the water flowing through the tubes in the floor will only be at 85 degrees,” said Robb Aldrich, an engineer at Steven Winter Associates in Norwalk, Connecticut, who helped design the heating system. “So the radiant floor allows excellent utilization of the solar hot water. If the house had a hydro-coil forced-air system, you’d need at least 120-degree water.”
Yet Aldrich is still not convinced that the in-floor distribution system was a wise choice. “Because of the radiant slab, storing direct solar gain is out of the picture,” he noted. “Having invested so much in the envelope, they could have gone with a really simple, cheaper, lower-cost heating system. In their next project, RDI does not plan to do a radiant floor because it is just too expensive.”
Anne Perkins, RDI’s director of homeownership programs, echoed some of Aldrich’s points. “I don’t think a radiant floor is appropriate for such a well insulated house,” she said. “You have such a low heat load it doesn’t make sense to spend a lot of money on a heating system. Most of the time the water running through the pipes in that expensive radiant floor will be so cool that the floor won’t even feel warm.”
Before deciding on in-floor radiant heat, the system’s “parasitic” energy load — the electricity required for pumping — must be calculated. “We looked at pumping energy a lot,” said Aldrich. “It’s a big concern of mine. I was fairly rigorous on all the friction and pressure calcs. On the solar thermal system, we decided to use a PV-powered DC pump. We chose the smallest possible circulators for the radiant floor. There were some compromises there, though. We ended up with a constant circulation system — basically, the pumps operate continuously for the entire heating season.”
The Colrain heating system has two circulators that together draw 173 watts. Since the circulators will operate for 4,000 to 5,000 hours per year, they consume between 19% and 23% of the annual output of the home’s PV array. “The radiant pumping energy will really be a significant load, which definitely bothers me,” Aldrich said.
These systems make sense for auto-repair shops
Many authors have quoted Alex Wilson’s conclusion that a radiant floor heating system is “a great heating system for lousy houses.” These heating systems also make sense for auto-repair shops in cold climates. If you have to work on cars that are sometimes covered in ice and snow, you will certainly appreciate a heated slab in your garage.
When it comes to designing a green home, however, it’s important to follow these simple principles:
- Design your house to be small and compact.
- Aim for a thermal envelope that is highly insulated and close to airtight.
- Choose windows with appropriate glazing, and don’t make your windows too big.
The main benefit of these principles is that you’ll end up with a house that is comfortable. You’ll also end up with a house that is very easy to heat and cool, without spending an arm and a leg for a complicated HVAC system.
“But I really want one!”
Occasionally, after the arguments presented in this article are explained to someone planning to build a new home, the client says, “I understand why a radiant floor heating system doesn’t make much sense for the house I’m about to build. But I love the idea of radiant floor heat, and I can afford to install it. I just really, really want this type of heating system.”
If you are an engineer or a designer who has a client like this, the best response is clearly, “OK. We’ll help you design the system you want.”
Martin Holladay’s previous blog: “Thermal Barriers and Ignition Barriers for Spray Foam.”