Retrofit Savings for Dual Integrated Appliances in Small Commercial and Multifamily Buildings

Dual integrated appliances are simply water heaters that are used to provide both domestic hot water (DHW) and space heating. Dual integrated appliance (DIA) systems provide a number of advantages over separate water heater and space heating systems. Among these are the opportunity for increased energy efficiency, reduced system size, and ease of space heat zoning. DIA systems have had substantial market penetration in the warmer climates of the mid-atlantic states, but there are appropriate selected markets in colder climates as well. The primary goal of this study is to assess the applicability of DIA systems in small commercial and multi-family buildings in a cold climate. Ten systems are installed at five small commercial (SC) and five multi-family (MF) locations. In addition to examining the energy savings of and occupant response to DIA systems, a secondary purpose of this project is to identify and address barriers for the implementation of DIAs in the Twin Cities metropolitan area.

When this project began some code officials were reluctant to allow the installation of DIA systems. There were concerns of possible deleterious health effects from the stagnant water in the heating coil, reduced heater life span, lack of state codes, and that heaters were being used in an boiler application. Many of these concerns have been addressed and there is presently better acceptance of the systems. The State of Minnesota Plumbing Code (1990) has been amended to include guidelines for the installation of DIA systems. The guidelines specify that mixing valves are required to guard against injury from elevated DHW temperatures and that a method for purging the stagnant water in the heating coil be used to reduce the likelihood of Legionella growth. Two field studies have shown that using a water heater in a DIA system may increase the life span of the heater (BR Laboratories 1986, Thrall 1989). In most areas of the country where code officials have ruled that DIAs must meet boiler specifications, the rulings have been reversed.

The courts have consistently found that adding a space heating load to water heaters does not transform water heaters irto boilers (ACHRN 1991). The availability of DIA equipment somewhat limited the selection of test sites. A total of 56 heaters were identified as being manufacturer approved for use in DIA systems. Of the 15 heaters which have recovery efficiencies of 78% or above, only four have inputs greater than 40,000 Btu/h, six are stocked locally, and four can be vented through a side wall using either a direct or power venting method. Greater use of DIA systems in the Twin Cities metropolitan area will require that a larger percentage of the higher efficiency systems be locally stocked and that more heaters with higher input become available.

There were no occupant complaints of inadequate space heat or reduced DHW temperature during periods of high space heating loads. The lack of complaints confirms the appropriateness of the Pietsch and Talbert (1989) guidelines that were followed in sizing these systems. On the other hand, the dog grooming shop owner felt that the DHW capacity was marginal in the heating season. The manager of the sandwich shop had some complaints of inadequate space heat during high DHW draws, but found that recovery time from DHW draws could be decreased by turning off the space heater. Overall, the building occupants were satisfied with the operation of the DIA systems.

There were no repairs performed on any of the four atmospheric burner, vertically vented DIAs. Five of the Six power draft heaters had repairs which were related to the DIA system operation. Repairs to two of the three Mor¬Flow Polaris heaters included replacing two ignitor coils and completely changing out one heater that developed a leak. The original vent terminations of the three side-vented, Mor-Flow Integra heaters had to be modified to prevent the build-up of large ice formations on the termination. The terminations were modified and the heaters have since worked without further problems. Two of the three Integra heaters have been replaced due to high concentrations of combustion gas CO and to a destroyed burner. All three heaters produced moderate levels of CO which may be due to inadvertent flame quenching. The concerning trend of Polaris and Integra equipment failures are consistent with the experiences from other Twin City metropolitan area installations of these heaters. However, since these heaters are ideally suited for use in high efficiency DIA systems, it is encouraging that the manufacturer has typically been quick to modify troublesome components.

The savings of the DIA systems are determined by systematically comparing the measured gas use of the DIA systems to that of the existing systems. The PRISM pre/post analysis for four of the multi-family units resulted in a range of savings from 11.3 to 27.5% with an average of 18.0%. All of the savings results are highly significant and follow the expected general trend. The DIA system that is oversized and has the lowest recovery efficiency (#60) had the lowest savings. The savings are nearly the same for the Integra system in unit 70 and the DIA system in #64 that is of similar size and efficiency ratings but has a separate air handler/heating coil. Although initial measurements indicated an energy savings, a change in occupants in the fifth MF unit invalidated its energy savings analysis.
The two medical offices achieved statistically significant savings of 18.7 and 16.8% using an alternating mode type test. The use of the DIA system installed in the sandwich shop is 14.5% greater than that of the separate furnace and water heater system. Because of the similarity of the efficiencies of the two systems, the significantly greater energy use (p-value equal to 0.033) and heating slope are unexpected. It is possible that unforeseen interactions with the space heating system in the front of the store or differences in occupant system operation could have biased the results. The DIA system at the dog grooming shop energy use was not significantly different than that of the separate system. Extreme pre/post differences in sales volume did not allow valid energy savings results at the fifth SC site.
Approximate values of the combined annual efficiency for the separate systems are computed using the PRISM estimated building loads and the manufacturer specified EF and AFUE ratings. The low AFUE of the existing furnaces (60.3%) in the MF units caused the CAEs to average 56.1%. The furnaces at the SC sites are somewhat more efficient. The CAEs ranged from 59.2 to 64.4% and are, on average, 6.9% higher than that for the MF units. Equivalent CAEs of the DIA systems are computed by replacing the furnace AFUE with an estimate of the heater steady state efficiency. The computed CAEs of the MF DIAs range from 68.2 to 78.4% and average 74.7%. The DIA system CAEs at the SC sites range from 68.2 to 94.6% and average 83.5%. The higher average CAE of the DIA systems is primarily a result of the DIAs having higher recovery efficiencies than the corresponding AFUEs of the existing furnaces.

Based on the CAEs of the two types of systems, the estimated savings of the MF DIA systems range from 17.7 to 27.4% and average 24.7%. The estimated savings for the SC DIA systems range from 6.1 to 32.0% and average 23.2%. A comparison of the measured and estimated savings shows the expected trend of relatively higher measured savings for higher estimated savings. However, for only one MF unit does the measured savings plus its uncertainty equal or exceed the estimated savings. The average measured savings for the MF sites is 6.7% below the estimate and for the SC sites the measured is 19.0% below the estimate. The large discrepancy between the estimated and measured savings is either due to the separate system operating at a higher than rated efficiency or the DIA system operating at a lower than rated efficiency.

Measurements of heater input/output and subsequent regression analysis yielded a steady state efficiency of 80% and CAE e 67.4% for site 3920. The Ess is approximately 14% lower than the heater rated Er. The low value of Ess led to additional short-term measurements which confirmed the accuracy of the measured efficiencies. Preliminary results from another CEUE study of Mor¬Flow Polaris heaters indicate Esss of 96 and 83% for heaters with operating temperatures of 130 and 140F respectively. These findings strongly suggest that the lower Ess at #3920 is due to the elevated operating temperature (160F) that may not allow the water vapor in the combustion exhaust to condense. A reduction of 5 to 10% in the Ess could explain much of the difference in the estimated versus measured annual savings and rated versus measured steady state efficiency.

Intensive measurements and analysis at site 4207 yielded a Ess of 70% which is 13% below the rated Er of 83%. Vent gas measurements show that the combustion efficiency is above 85%. Short-term measurements were not able to reconcile the difference between the measured steady state efficiency and combustion efficiency. The analysis of the intensive data also show a DIA CAE of 62.1% which is 14.6% below the estimated value. The measured efficiency and savings results indicate that the unexpected low savings of the DIA systems are more likely due to lower than rated operating efficiencies of the DIA systems than to higher rather than rated operating efficiencies of the separate systems.

Full report (PDF)
Retrofit Savings for Dual Integrated Appliances in Small Commercial and Multifamily Buildings