Multifamily Research Gets in Hot Water

Figure 1. Three common multifamily water heating configurations are represented here. For the indirect-fired tank, the heat exchanger may be located in the tank as shown, or in the boiler. One other indirect-fired configuration involves the use of a dedicated boiler where the boiler water and the DHW are one and the same.

A recent study by the New York State Energy Research and Development Authority (NYSERDA) has demonstrated that the tankless coil is, in fact, slightly less efficient than two other common heating systems (the research was conducted by the EME Group and cofunded by Con Edison). The other two systems examined in the study were the direct-fired storage tank and the indirect-fired storage tank.

The direct-fired storage tank is a larger version of the water heater common to most detached homes across the country (see Figure 1). Indirect-fired tanks, which have been aggressively promoted but have found little use in single-family homes, consist of a boiler and a heavily insulated, nonfired storage tank. Water in these units is heated via a heat exchanger that is in the tank itself or in the boiler. Because of its extra capacity--the tank serves as a buffer during peak loads--the indirect unit can generally use a smaller boiler than a tankless coil.

The tankless coil is essentially a bundle of copper tubes situated inside a boiler. When there is a demand for DHW, cold water enters the coil, is heated by the boiler water to about 140°F-180°F, and is then mixed with cold water via a thermostatic tempering valve, typically down to a tap temperature of about 120°F.

The data showed small differences in efficiency among the three systems. The system with the tankless coil was the least efficient of the three, with overall efficiency at 34% (meaning that 2 out of every 3 Btu that went into heating the water were wasted). In comparison, direct- and indirect-fired tanks were slightly more efficient, with overall efficiencies of 35% and 37%, respectively. When the added first cost of these other two systems was factored in, the simple payback from choosing either one over the tankless coil was approximately 25 years! (The difference in first cost between the tankless coil and other systems in this project ranged between $3,000 and $4,000.)

The small difference in efficiency is not surprising. Modern, well-insulated, low water-volume boilers, especially those large enough to serve multifamily buildings, have relatively low jacket and off-cycle losses. Equipped with a tankless coil, they can compete effectively against the other two systems.

Figure 2. Distribution of DHW load with respect to flow rates. As shown in this graph, the direct-fired system was most efficient at high loads. At low-to-medium loads, the most frequent load type, efficiency was dismal.

Another interesting finding was the distribution of DHW flow. DHW systems are difficult to design and size, because of the wide range of flow from zero to peak demand. The DHW demand in the buildings ranged from 0 to 180 gallons per hour (gph). Of the total DHW flow, 70% was at 60 gph or less, 90% was at 100 gph or less, and the range of 120 to 180 gph represented a mere 2% of the total (see Figure 2).

When developing new systems, designers should strive for higher efficiency at low loads. One approach would be to downsize the main boiler significantly, and install a small booster boiler that would come on only during peak DHW demand. Overall first cost would be similar, but the system would operate closer to the full capacity of the downsized boiler, and thus more efficiently. Several manufacturers currently market water heating systems that are energy efficient at low and high loads.

Recirculation line losses must also be addressed--even if the energy codes do not address them. Recirculation flow can be optimized and the temperature reduced during periods of low draw using demand-sensitive, temperature-reducing controls (see Controlling Recirculation Loop Heat Losses, HE Jan/Feb '93, p. 9).

Piping losses can be reduced if lines are insulated. If piping is scheduled for replacement, or if it is accessible, installing insulation could reduce losses considerably--the N.Y. State Energy Code recommends a minimum of R-4 insulation for pipes less than 3/4 inch in diameter. NYSERDA is currently evaluating recirculation control strategies to identify and demonstrate the lowest-cost strategies for minimizing line losses.

Home Energy can be reached at: contact@homeenergy.org
Home Energy magazine -- Please read our Copyright Notice