Renewable Hydronic Heating
March 21, 2014
You can use renewably made hot water for your hydronic system—but designing for low water temperatures is critical to good performance.
Hydronic heating is the technology of moving heat using water. It has been used for decades in millions of North American homes, most of which have a gas-fired or oil-fired boiler as their hydronic heat source. Good hydronic design is also the “glue” that holds together renewable energy thermal systems that provide space heating and domestic hot water. In other words, pick a renewable heat source, do a good job with the underlying hydronics, and you’ll likely be pleased with the results. Treat the hydronics as “whatever,” and you’re likely to be disappointed.
In the past, solar hydronic heating meant using solar collectors as sunny-day substitutes for conventional boilers or water heaters. Designers focused on the collectors, storage, and control aspects of the solar subsystem, but devoted little thought to a compatible means of distributing solar-derived heat within the building.
Most hydronic distribution was designed around high-temperature supply water. Residential systems commonly used fin-tube baseboard heaters with water temperatures sometimes exceeding 200°F.
But those high water temperatures were beyond what solar collectors could produce consistently. Sure, there was an occasional “perfect solar day” in winter when the storage tank got hot enough to heat a home during the following night. However, performance over a typical northern heating season was often disappointing. As a result, after investing thousands of dollars in collectors, storage tanks, and controls, many early systems spent much of their time distributing heat generated by conventional fuels rather than by the sun.
The North American heating industry has a tendency to focus on heat sources rather than overall heating systems. This mindset continues to limit the performance of not only solar thermal, but also heating systems supplied by sources such as geothermal heat pumps and wood-fired boilers.
Low Temperatures = High Efficiency
All renewable heat sources yield better performance when combined with low-temperature distribution systems. To see why, take a look at the thermal performance characteristics of a solar collector and a geothermal water-to-water heat pump. The “Solar Collector” graph below shows how the thermal efficiency of a flat-plate solar collector is affected by the temperature of the fluid entering its absorber plate. On this typical sunny, midwinter day in the northern United States, the thermal efficiency of the collector drops rapidly with increasing inlet fluid temperature.
For example: If the fluid entering the collector is 90°F, the outdoor air temperature is 30°F, and the sun is bright (solar intensity is 250 Btu/hr./ft.2), the graph indicates that the fluid gathers about 56% of the solar energy striking the collector. However, if the entering fluid temperature is 160°F, while the other conditions remain unchanged, the collector’s efficiency falls to 33%—a significant “penalty” when the collector operates at the higher inlet temperature. It’s the result of greater heat loss from a collector to outside air, much like the increased heat loss associated with keeping your house at 75°F rather than 68°F.
The relationship between efficiency and the entering water’s temperature also holds true for hydronic heat pumps. The “Heat Pump” graph shows a similar effect for a modern water-to-water geothermal heat pump operating with a constant earth-loop inlet temperature (at the condenser side of the heat pump) of 45°F.
The coefficient of performance (COP) is the heat pump equivalent of efficiency: the ratio of the heat output divided by the electrical input. A COP of 4.0 means that the heat output is four times greater than the electrical energy required to operate it. The higher a heat pump’s COP, the lower its operating cost.
The graph shows the heat pump’s COP dropping rapidly as the hydronic heating system’s water temperature increases. Thus, for the highest possible COP, the water temperature supplied to the hydronic heat emitters should be kept as low as possible.
Wood-fired boilers can produce higher water temperatures, even up to 200°F, but that doesn’t negate the benefits of matching a wood-fired boiler to a low-temperature distribution system. These heat sources are best used with a thermal storage tank. They add heat to the storage tank, and the heating distribution system draws heat out. The lower the tank temperature can go and still supply sufficient heat to the building, the less often the boiler has to be stoked.
For the best efficiency, design hydronic heating systems supplied by either solar collectors or hydronic heat pumps so that the supply water temperature to the load (under maximum load conditions) doesn’t exceed 120°F. This temperature is a reasonable compromise between maintaining good heat source performance, while not overly increasing the cost and space requirements of the heat emitter.
Making It Happen
It’s critical to understand what determines the supply water temperature (the temperature being supplied to the load via heat emitters) in a hydronic heating system. Some designers think that it is determined by the heat source—because many boilers come with a dial or digital control that “sets” the temperature produced. Unfortunately, that’s not how it works. That setting is only a high-temperature limit on the heat source output. It does not guarantee that the set water temperature will ever be reached. The water temperature in any operating hydronic heating system climbs only high enough for thermal equilibrium—where the rate of heat release from the heat emitter balances the rate of heat input from the source. Once thermal equilibrium is reached, there is no thermodynamic incentive for the water temperature to climb higher, and it won’t!
It’s the hydronic distribution system’s design, rather than the high-limit setting, that determines the system’s operating water temperature. Almost everyone who designs heating systems wants to maximize thermal efficiency. For hydronic systems, this means moving away from high water temperatures by using heat emitters with larger active surfaces, or other details, such as internal microfans, that increase both convective and radiant heat transfer. This allows thermal equilibrium to occur at relatively low water temperatures during both maximum and partial load conditions.
There are several ways to design modern hydronic distribution systems around RE’s lower water temperatures, starting with the heat emitters—any device that removes heat from water flowing through it, and releases it into the room.
Many homeowners are used to the look of fin-tube baseboard heaters. From the outside, modern baseboard heaters look similar to older ones—but what’s inside is very different. Fins can be about three times larger than traditional baseboard heaters, with multiple copper tubes running through those fins. Tubes can be piped for either parallel or series flow. In the latter case, the hottest water flows in the upper tube, makes a U-turn at the end of the heater, and flows back in the lower tube.
Assuming an average water temperature of 110°F, this baseboard releases about 290 Btu/hr./ft.2 at 1 gpm when the tubes are configured for parallel flow. With 4 gpm, the output increases to about 345 Btu/hr./ft.2 If the two tubes are configured for series flow, the output drops about 10%.
Consider a 12- by 16-foot room in a well-insulated home, with a maximum heating load of 2,880 Btu per hour (e.g., 15 Btu/hr./ft.2). This load could be met using a 10-foot length of Heating Edge baseboard operating at an average water temperature of 110°F at 1 gpm. A 10-foot length of conventional residential baseboard heater would require an average water temperature of about 150°F. This temperature is well above what a typical geothermal heat pump can produce, and would significantly lower the efficiency of solar thermal collectors. An additional 14 feet of conventional baseboard would be required (24 feet total) to deliver the same output at an average water temperature of 110°F.
The key to low temperature operation is large emitter surface area. The greater the heated surface area, the lower the required water temperature for a given rate of heat output. By embedding tubing in floors, walls, and ceilings, it’s possible to create very large heated surfaces within a room, called hydronic radiant panels. Radiant floor heating is the best-known form, and can be installed in several ways that allow it to operate at relatively low water temperatures.
For a heated slab-on-grade floor, the tubing should be placed about mid-depth within the slab, and the underside and edge of the slab should be well-insulated. These details are crucial for low-temperature performance. It’s also important to use low-resistance (or even no-resistance) floor coverings. A painted, stained, or stamped concrete slab surface is ideal. If that doesn’t suit your tastes, consider ceramic tile or vinyl flooring. If you must have carpet, it’s best to use only 1/4-inch-thick commercial-grade loop carpet, and glue it directly to the slab. Covering a radiant slab with a pad and carpet will insulate your living space from the slab’s heat.
The “Radiant Heat Output” graph can help guide tube spacing and floor covering selections. Assuming you’re satisfied with a 70°F indoor air temperature, this graph gives you the required average water temperature in the floor tubing based on tube spacing, the R-value of the floor covering (if any), and the rate at which heat must leave the floor under maximum heating conditions (in Btu/hr./ft.2).
For example, assume the room’s maximum heating load (which is calculated manually or by using one of several available software packages) divided by the heated floor area is 15 Btu/hr./ft.2. You’ve decided to space the tubing at 6 inches, and finish the concrete slab with a sealed stain (RFF = 0). The required average water temperature in the embedded tubing is only 83°F. It’s common to design floor circuits with overall temperature drops of 16°F to 20°F. Thus, the supply water temperature is half this overall temperature drop (8°F to 10°F) above the average water temperature. The temperature drop along the circuit is due to the heat being released from the water flowing through the circuit. Under these assumptions, the slab needs to be supplied with 93°F water to cover the maximum heating load. The supply water temperature can be lower under partial load conditions.
Such temperatures give excellent performance from a renewable energy heat source. Although the floor surface temperature will likely only be in the low- to mid-70s, that’s as hot as the floor needs to be to release all the heat the room needs. I always stress this point so that occupants understand that a “toasty warm” floor implied by some advertisements simply doesn’t occur in low heating load conditions, and in systems optimized for low water temperature operation.
The graph only reflects a maximum finish flooring resistance of 1.5 and a maximum average water temperature of 110°F (with the assumption of a maximum supply water temperature of 120°F). If you’re serious about building a good performing system, don’t exceed these limits. Here are some more suggestions for a heated slab floor with a renewable energy heat source:
- Tube spacing within the slab should never exceed 12 inches.
- The slab should have a minimum of R-10 for both underside and edge insulation.
- Tubing should be placed at approximately half the slab depth. Doing so decreases the required water temperature for a given rate of heat output. For a 4-inch concrete slab, the average water temperature would need to be about 7°F higher if the tubing is at the bottom of the slab. Placing the tubing closer to the top of the slab risks putting it in the path of sawn control joints.
Another common method of hydronic floor heating uses a 1.5-inch “thin slab” poured over a wooden deck. The slab can be either concrete or poured gypsum, but should never be lightweight concrete, which uses vermiculite or polystyrene beads instead of stone aggregate, and has significantly higher thermal resistance.
Because the slab is thinner, it has somewhat poorer heat dispersion characteristics, needing a slightly higher water temperature for a given rate of heat output—but this difference is slight: A 1.5-inch concrete thin-slab with 12-inch tube spacing and covered with a finish flooring resistance of 0.5°F/hr./ft.2/Btu yields about 8% less heat output than a 4-inch-thick slab with the same tube spacing and finish flooring. To get the same efficiency, 9-inch, rather than 12-inch, tube spacing can be used.
The following guidelines are suggested for thin slabs supplied by renewable heat sources:
- Tube spacing for a thin-slab application should not exceed 9 inches.
- Floors under thin-slabs should have minimum of R-19 underside insulation.
- Floor finishes should have a total R-value of 1.5 or less (lower is always better).
Heated Walls & Ceilings
Walls and ceilings can also be turned into low-temperature hydronic radiant panels. These radiant walls are indistinguishable from a standard interior wall. Its low thermal mass lets it respond quickly to changing room load conditions or zone setback schedules. This fast response is especially important in homes with low heat loss or significant internal heat gain because such spaces can quickly overheat.
The panel’s rate of heat emission is approximately 0.8 Btu/hr./ft.2 for each 1°F the average water temperature in the tubing exceeds room air temperature. For example, if the average water temperature in the tubing is 110°F in a room with 70°F air temperature, each square foot of wall releases about 32 Btu per hour [0.8 × (110°F – 70°F)] . This average water temperature is well within the range of what most renewable energy heat sources can supply.
If you plan to install this system on the inside of an exterior wall, make sure the R-value of that wall is 50% higher than that of unheated exterior walls. That keeps the rate of heat loss to the outside about the same as for an unheated wall. If you’re installing this on an inside partition wall, use 3.5-inch fiberglass batt in the stud cavities behind the heated wall. Finally, radiant wall panels work best constructed no higher than 3 to 4 feet above floor level. These heights bias the radiant heat output into the occupied zone of rooms, and thus improve comfort.
Radiant ceilings use the same construction as a radiant wall. The only difference is that the materials are fastened to the ceiling framing rather than the studs. The infrared thermograph shows such a ceiling as it warms up. The red areas on the left side indicate that the aluminum heat transfer plates are dissipating heat away from the tubing and across the adjacent ceiling surfaces.
Like the radiant wall, a radiant ceiling has low thermal mass and can respond quickly to interior temperature changes. Heated ceilings also have the advantage of not being covered by rugs or furniture, and thus are likely to retain good performance over the building’s life, but can be a bit more expensive relative to a heated slab-on-grade floor.
The rate of heat emission from a ceiling panel constructed as shown is about 0.71 Btu/hr./ft.2 for each 1°F the average water temperature exceeds room air temperature. Thus, if the ceiling tubing operated with an average water temperature of 110°F in a room with 70°F air temperature, each square foot of ceiling would release about 28.4 Btu/hr./ft.2 [0.71 × (110°F – 70°F)] . Although not as high as the radiant wall due to lower convection, this performance is still very acceptable for use with most renewable heat sources.
Generously sized panel radiators can also provide good low-temperature performance. The suggested guideline for sizing panels to deliver the maximum heat output is to use a supply water temperature no higher than 120°F.
Manufacturers provide output ratings for their panel radiators in tables or graphs. Most list heat output for high water temperatures such as 180°F. Correction factors are provided to determine heat output at lower water temperatures. As an approximation, a panel radiator operating with an average water temperature of 110°F in a room maintained at 68°F provides about 27% of the heat output it would yield at 180°F. Larger panels increase surface area to compensate for lower operating temperatures.
From Here to There
None of these hydronic heat systems will deliver as expected without a well-planned distribution system. Although there are several piping layouts that may serve your purpose, the simplest, easiest to install, and most flexible approach is a “home run” distribution system, which starts with a manifold station—the same kind as used for radiant panel heating.
Two lengths of 1/2-inch PEX or PEX-AL-PEX tubing provide the supply and return from the manifold station to each heat emitter. The flexible tubing allows routing through framing cavities much like an electrical cable. This is particularly helpful in a retrofit situation, where the use of rigid tubing would be difficult.
A variable-speed, pressure-regulated circulator pump provides flow through the home run distribution system. Available from Grundfos, Taco, Wilo, and Bell & Gossett, these circulators can operate over a wide range of speeds, and in different control modes, depending on the application. For a home run distribution system, the circulator is set to “constant differential pressure” mode. Its responsibility is to maintain a constant (installer-set) pressure differential between its inlet and outlet ports. It does this by varying speed in response to changes in the distribution system’s hydraulic resistance.
At full speed, the motors in these “intelligent” circulators operate on approximately 50% of the electric power required by standard hydronic circulators of equal capacity. This characteristic, in combination with speed control, delivers annual electrical energy savings of 60% to 80% relative to standard hydronic circulators.
A thermostatic radiator valve (TRV) is used on each panel radiator. Each TRV constantly monitors the room’s air temperature. As that temperature drops slightly below the TRV’s setting, the valve slowly increases water flow through that panel. This causes a very slight drop in the distribution system’s hydraulic resistance, a change that’s quickly detected by the pressure-regulated circulator which responds by increasing pump speed to restore the set differential pressure. This yields a slight flow increase through the panel that needs it, and virtually no flow changes in the other panels—a “cruise control” for system flow.
The Big Picture
The main component in the schematic (above) is a well-insulated drainback/storage tank equipped with an electrical element or integrated gas burner for backup, with an internal heat exchanger. The element/burner keeps the water at the top of the tank warm enough to provide domestic hot water (typically 120°F to 130°F).
A drainback-protected solar collector feeds the hydronic heat distribution system. The collector circulator runs when the collectors are a few degrees warmer than the water near the bottom of the storage tank. No antifreeze is required in this system, and no heat exchanger is needed between the collectors and the storage tank. These features reduce cost and increase collector efficiency. The same water that flows through the collectors also flows through the heating distribution system. The system is completely “closed” from the atmosphere.
The captive air at the top of the tank is under slight positive pressure. This airspace provides a drainback reservoir, and acts as an expansion tank. The water in the tank provides thermal storage for the solar collectors, and it provides thermal mass to buffer the zoned space-heating distribution system. The latter function protects the burner against short operating cycles, which would otherwise decrease efficiency and increase maintenance. Short cycle protection is very important in a hydronic system with extensive zoning.
A flow switch detects whenever domestic hot water is being drawn at a flow rate of 0.5 gpm or higher, activating a small circulator that moves hot water from the top of the thermal storage tank through a plate heat exchanger. Cold domestic water is heated as it passes through the other side of this heat exchanger, and sent to the taps.
An antiscald thermostatic mixing valve protects against high domestic water temperature when the tank is very hot, like at the end of a sunny, warm day. For the fastest possible response, the piping between the thermal storage tank and heat exchanger should be short and insulated. Combination isolation/flushing valves should be installed on the domestic water inlet and outlet of this heat exchanger. They allow the heat exchanger to be isolated and flushed if necessary to remove scale.
A single variable-speed pressure-regulated circulator feeds the home run distribution system for space heating. One circulator can supply the entire distribution system in a typical 2,500-square-foot house using no more than 40 watts under maximum heating load.
Each panel radiator has an adjustable thermostatic valve that monitors room temperature, and adjusts the flow rate to maintain that temperature. No thermostats, batteries, transformers, or programming—just simple, effective, and reliable room-by-room temperature control.
The mixing valve upstream of the manifold station protects the distribution system from what could be a very hot storage tank following a sunny spring or fall day. It also adjusts the water temperature supplied to the panels based on outdoor temperature, known as “outdoor reset,” and stabilizes room temperature for optimum comfort.