The Best Boiler
and
Water Heating
Retrofits

When the Minneapolis Energy Office (now the Center for Energy and Environment) first addressed the issue of improving energy efficiency in multifamily buildings in 1981, the task was a little daunting. For one thing, there was no documented research on multifamily retrofits, making it difficult to identify recommendations, quantify savings, and recognize the specific conditions under which savings could be expected. Most information was in the form of case studies published by manufacturers in sales literature.

We also found considerable suspicion among multifamily building owners about energy retrofits. They had pretty much heard it all: every new device or piece of equipment on the market was an amazing breakthrough, "guaranteed to reduce energy bills by 25% with a one-year payback." Property owners lacked the technical expertise to distinguish truly cost-effective retrofits from snake oil. An additional challenge was the fact that multifamily building owners tend to avoid long-term investments.

Since then, we and other groups have systematically fieldtested a wide range of energy conservation measures for boilers and domestic water heaters in low-rise multifamily buildings. We avoided "way-out" retrofits, and concentrated on measuring the actual performance of widely recommended measures. This fieldwork has provided independent and objective test data that building owners can trust. This article presents a quick review of what we have learned over the years.

In our research, we focused on the Minneapolis-St. Paul area, where the vast majority of multifamily buildings are two- to seven-story walk-ups (no elevators) of 6 to 60 units. We identified two major groups of multifamily buildings: steam-heated buildings built between the turn of the century and World War II, and hydronic-heated buildings built since World War II (see "Minneapolis-St. Paul Building Types").

In all buildings, research emphasized mechanical system retrofits, since most envelope measures had already been implemented, were physically impossible to implement, or exceeded owners' payback criteria (typically one to three years). Research methods ranged from simple analysis of pre- and post-retrofit billing data to long-term intensive monitoring.

Research in steam buildings has covered four main retrofit options (see Table 1):

We did not study boiler replacements because the options for high-efficiency steam boiler replacements are limited. (Some condensing steam boilers are available now, but most of them are designed for industrial applications.) We also wanted to focus on the most cost-effective measures that owners can afford without rebates or other assistance.

The worst source of inefficiency we found in steam buildings was uneven heating. To minimize complaints from residents in cooler areas, building owners grossly overheat other areas, leading tenants to open their windows for relief. (In fact, many older contractors have told us that in the original design of steam systems, the idea was to overheat the building and let individual tenants regulate their heat by opening and closing their windows!)

We intensively monitored boiler cycles and the movement of steam through the distribution system at a prototypical test building. Through this we learned that the primary causes of uneven heating were large differences in steam arrival times combined with short boiler cycles (see "The Art and Science of Balancing Single-Pipe Steam Systems," EA&R Mar/Apr '87, p. 24).

We developed a strategy for balancing steam distribution that involved installing a thermostat with an adjustable dead band, and adding very high capacity (orifice of ⁵/₁₆-in) main-line air vents, as well as larger radiator air vents in some locations. This steam balancing reduced total building gas use by an average of 10% with a median payback of 1.3 years in 13 pilot buildings. Extensive work on steam balancing in Chicago has shown similar results.

Many of these older steam buildings have boilers that have been converted from coal to natural gas, making them good candidates for tune-ups for several reasons. First, the flue gas passages were grossly oversized for gas, resulting in significant excess air and poor heat transfer. Second, local codes offered flexibility in modifying conversion boilers. And third, maintenance on most of this equipment had been deferred.

Techniques used to reduce excess air and stack temperature included adjusting the fixed or motorized draft louvers, installing flue restrictors, installing baffles on the bridge walls of brickset boilers, sealing uncontrolled secondary air leaks into the combustion chamber, and adjusting the firing rate. (Turbulators are a common retrofit that we did not install except as a last resort. We found that the other measures cost less and often work better than turbulators to reduce stack temperature.) Savings for tune-ups ranged from 0% to 14% of space heating use, with typical paybacks of about six months.

While tune-ups sound trivially simple, they require a well-trained pool of contractors with access to electronic equipment, and we found that we had to identify and teach this pool ourselves (see "Boiler Tune-Up: Improving the `MPG' of Multifamily Buildings," HE Sept/Oct '89, p. 21).

The large thermal mass and vent size of coal-to-gas conversion boilers, combined with the massive masonry chimneys common in multifamily buildings, suggest large off-cycle losses and significant potential for vent dampers (see "Do Vent Dampers Work in Multifamily Buildings?" HE Mar/Apr '90, p. 27). We also expected the presence of barometric dampers for draft relief to make vent dampers more effective in conversion boilers, since the barometric damper swings shut when the vent damper closes, reducing air flow through the boiler and retaining heat.

The brickset boilers we tested showed average savings of 8.6% of total building gas use, with an average payback of 2.2 years. Oddly, when vent dampers were added to the water heaters as well, one of our test buildings showed greater savings while another showed almost no savings, a result we were unable to explain.

The final area we investigated for steam buildings was conversion to hot water heat, which produced savings averaging 18% of total gas use for single-pipe steam (SPS) buildings and 27% for two-pipe steam (TPS) buildings. The SPS conversions required new supply and return lines and new radiation, and costs averaged $58,000 ($3/ft2). Comparatively, TPS conversions reused the existing piping and radiators (with minor modifications), and costs averaged $28,000 ($1.50/ft2). While the median payback for this retrofit is long (10 years), conversion from steam to hot-water distribution provides compelling advantages: improved tenant comfort and system reliability, increased building resale value, and lower maintenance costs (for example, hydronic systems do not require regular water treatment, blowdown, or trap maintenance).

Research in buildings heated with hot water has covered five main options (see Table 2):

• Resets and cutouts.
• Vent dampers.
• Tune-ups of gas-designed boilers.
• Energy cost allocation systems.
• Front-end modular boilers.

In early energy audits, we found many hydronic buildings operating with constant boiler water temperature or with manual resetting only. As a result, one of the first measures we tested was a reset control that varied boiler water temperature between 110deg.F and 185deg.F as a function of outdoor temperature, combined with a cutout control that shut the boiler off on warm days. (Although manufacturers of cast-iron boilers suggest that their equipment should not operate below 140deg.F, common practice in Minneapolis and Saint Paul is to reset temperatures as low as 110deg.F. We have put resets on them for the past ten years and have never had a problem, and contractors have been doing it for 30 years or more.)

Tests of resets and cutouts on cast-iron boilers showed savings of 18% of space heating use with an average payback of 1.2 years (see "Outdoor Resets and Cutouts: Quick Fixes for Hot Water Heating," HE Nov/Dec '88, p.15). Before-and-after tests of automatic versus manual reset, both with automatic cutouts, showed savings of 10% of total gas use (about 14% of space heating use) with paybacks of less than one year.

These results were confirmed in a later pilot project, which achieved savings of 9% of total gas use (about 13% of space heating use) in eight buildings. A Wisconsin study found somewhat less savings (7% of space heating use) for a group of six four- to nine-unit buildings, but in most of these cases water did not circulate through the main distribution system unless at least one zone was calling for heat, whereas the larger buildings studied in Minnesota all had constant circulation.

On steel fire tube boilers, resetting the boiler water temperature directly carries greater risks of thermal shock and corrosion. In these buildings, we tested resetting of the distribution water using a three-way mixing valve, which is more expensive due to both the piping involved and the need for a modulating control. Savings averaged only 9.5% of space heating gas use, lower than for direct resetting of boiler water, and paybacks were considerably longer.

Part of the difference in performance for resets on steel fire tube boilers results from the fact that resetting the distribution water temperature does not improve the seasonal efficiency of the boiler, but resetting the boiler water temperature does. Based on direct input-output measurements CEE has done with cast-iron gas-designed boilers, we would expect seasonal efficiency to improve by 1.5%-3% in going from constant temperature control to reset control.

An alternative strategy to using a mixing valve for steel fire tube boiler applications is to use a reset that has the capability of setting a minimum boiler water temperature. Savings for this have not been measured to date but would be expected to be lower, since resetting of boiler water temperatures takes place in a narrower range (for instance, 140deg.F-185deg.F).

We anticipated lower savings from vent dampers in newer hydronically heated buildings because of the lower thermal mass of the boilers and because draft relief is provided by draft hoods. This allows air to flow continuously through the boiler (picking up heat and spilling it out the draft hood) when the vent damper closes. In addition, the vent dampers were tested on buildings that already had resets, and thus lower off-cycle losses.

Savings averaged 6.5% of total gas use with both boiler and water heater dampers active, but only 1.9% with only the boiler dampers active. These results are consistent with those from tests in single-family homes done by the Institute of Gas Technology in 1976 and 1980, which showed vent dampers on the water heater to be critical to overall savings.

To maximize the savings, we installed quick-closing, tight-fitting vent dampers and electronic ignitions, which inflated the cost considerably over a more typical retrofit with thermal dampers on the water heaters and slower dampers on the boilers.

Tune-ups of gas-designed boilers were limited by code restrictions on the installation of flue restrictors and on adjustments to input, as well as by the fact that the secondary air openings were fixed. Tune-ups by both CEE and a Wisconsin group showed essentially no savings.

A more novel retrofit strategy tested on hydronic buildings was the use of energy cost allocation systems to apportion gas costs among individual apartments on the basis of use. We studied nine buildings that had run-time meters installed to measure the length of time that zone valves were open. Savings of 16% of total gas use were observed, simply from making residents directly responsible for energy costs, with a payback of 1.4 years. These results agree with a 1983 study by Lou McClelland at the University of Colorado at Boulder, which compared 83 properties before and after tenant payment was introduced and found average energy savings of 10%-20%. Similar data is also available from Europe, where energy cost allocation is widespread.

The final heating system measure tested in newer hydronic buildings was to supplement existing boiler systems with small "front-end" boilers that could heat the buildings more efficiently under low load conditions (see "Front-End Modular Boilers: Lessons from the Real World," HE Mar/Apr '91, p. 19). For two systems installed by CEE and six installed by a contractor and monitored by CEE, savings ranged from -3.7% to 18.5% of total building energy use and averaged 7.7%. These savings were much lower than expected.

Somewhat better results were achieved by the Energy and Environmental Resource Center for four installations that were more closely monitored and precisely operated than the CEE cases, since they were part of a shared-savings program. Savings for these cases ranged from 6.8% to 18.3% of total building energy use and averaged 13.3%.

The best candidates for this retrofit appear to be existing boilers that have low seasonal efficiencies (such as atmospheric boilers) and are grossly oversized for the building load. Using the front-end boiler to provide domestic hot water also increases savings potential.

In spite of the lower-than-expected savings and the fact that front-end systems were relatively complicated to install and operate, other benefits might make front-end boilers attractive to an owner. These include availability of backup and the ability to forestall replacing an aging, existing boiler.

Domestic hot-water research has focused on two main areas (see Table 3):

• Cost-effective options at the time of construction or equipment replacement.
• Controls for systems with constant recirculation loops.
  

In order to identify the most cost-effective equipment options at the time of construction or replacement, CEE tested two types of higher-efficiency water heater against standard tank-type heaters. The first was a tank-type heater with an integral flue damper upstream of the draft diverter. Savings from installing this type of heater were small enough that paybacks were over 10 years, in spite of its modest incremental cost of $560. This is a somewhat moot point, since most commercial tank-type heaters manufactured after January 1994 must have integral flue dampers in order to meet new efficiency standards. However, these results are still worth noting, since local supplies of tank-type heaters without integral dampers may still dominate actual sales.

The second type of high-efficiency heater tested was a power-vented condensing unit. It produced dramatic savings of 28% of service water heating energy use, but at such a high incremental cost ($2,400) that paybacks were close to 20 years.

The only true retrofit measure tested was the strategy of resetting the tank and recirculation loop temperature from about 145deg.F to 110deg.F in buildings with constant recirculation of domestic hot water (see "Controlling Recirculation Loop Heat Losses," HE Jan/Feb '93, p. 9). Two controls were tested--one that resets based on the time of day and one that resets based on actual demand on the service hot water heater.

The time-based control saved about 10% but had more operation and maintenance problems. The demand-based control reduced water heating energy use by 16% and had fewer operational problems; as a result, it is recommended over the time-based control. Both controls had an average payback of about two years.

Confirmed energy savings ensure credibility and retrofit performance, but they are only half of the formula. The other half is setting up a comprehensive implementation program to ensure that the appropriate retrofits are actually installed and that they are installed correctly.

The first step involves working actively with trade allies, such as suppliers and contractors, to introduce new technologies into the marketplace and promote them. Our experience has also shown that contractors need assistance and training in the technical aspects of completing retrofits properly, both for unusual or innovative technologies (such as steam to hot water conversion and front-end modular boilers) and for ones that have been around for a while (such as boiler tune-ups, resets, and modular boiler installations).

Multifamily building owners also need a convenient, one-stop service that not only identifies the appropriate retrofits, but makes the installation of those retrofits as easy for them as possible. This means specifying particular equipment, using prescreened and trained contractors, and making postinstallation inspections to ensure quality. In addition, to minimize confusion about the installed retrofits on the part of building caretakers and maintenance staff, we developed simple operating instructions and guidelines for efficient operation, which are laminated and posted in the boiler room.

An effective program should also include financial incentives, such as rebates or low-interest loans, to ensure a high rate of installation in a market that is driven largely by first costs. However, we have found that financial incentives need not always be large. Throughout the 1980s, using this one-stop approach and only limited utility rebates, we audited over 1,500 buildings, 40% of which implemented measures through our program and an additional 30% of which implemented the measures using our information and their own contractors.

For some retrofits, our extensive work with multifamily owners has so transformed the market in the Twin Cities that it is difficult today to find buildings in which these measures have not been implemented. For instance, since 1990 we have audited about 1,200 additional buildings, but the implementation rate has dropped to about 30% as a result of this market saturation.

For central boilers with inputs larger than 300,000 Btu/h, one of the big unresolved technical issues concerns quantifying seasonal efficiency. Unfortunately there is currently no standardized test procedure for determining the seasonal efficiency of commercial boilers of this size, and most manufacturers only give steady-state thermal efficiency, which is a poor predictor of annual performance.

ASHRAE has recently set up a standards committee, headed by CEE staff, which is working on developing a standardized test procedure to determine annual fuel efficiency. Once completed, it will be of enormous benefit in helping practitioners to determine the cost-effectiveness of various boiler replacement options and to pin down more precisely case-by-case savings estimates for boiler retrofits such as resets and vent dampers.

For new construction in the multifamily sector, one issue that needs to be addressed is the trend toward increased use of electric heat--a trend driven mostly by first costs and by the desire to have tenants pay their own heat. (As discussed above, electric heat is not the only way to have tenants pay for what they use; energy cost allocation systems can apportion fuel use among tenants.) According to the U.S. Energy Information Administration, in 1975 19% of the multifamily housing stock nationwide was heated with electricity, but by 1990 this figure had risen to 34%. As this new housing stock ages, this will become a retrofit issue, especially since the majority of multifamily buildings house low-income people.

Finally, as utility funding for demand-side management programs declines in the coming years, rebates and incentives are also expected to decline. This trend will make it necessary to come up with alternate financing scenarios that still provide building owners with a positive cash flow and low up-front costs. One option that has been tried in some commercial markets is tying loans for cost-effective retrofits to the building meter or account, rather than to a specific customer. This strategy encourages an owner to install retrofits that are beneficial in the long run even if that particular owner sells the building. Each new owner becomes responsible for taking over the repayment of the loan as the building changes hands. Since the new owner is presumably reaping the benefits of the retrofit, the strategy is fair to all players.

Most of the research reported in this article was supported by Minnegasco, a Division of Arkla, Incorporated. The authors gratefully acknowledge Minnegasco's long-term commitment to research. Tests of domestic hot water measures were supported largely by the St. Paul Energy and Environmental Resource Center, through a grant of oil overcharge funds distributed through the Minnesota Department of Administration.

Biederman, N. and Katrakis, J. Space Heating Improvements in Multi-Family Buildings, GRI-88/0111. Chicago: Gas Research Institute, 1989.

Ewing, G., et al. "Effectiveness of Boiler Control Retrofits on Small Multifamily Buildings in Wisconsin," Proceedings of the American Council for an Energy Efficient Economy 1988 Summer Study on Energy Efficiency in Buildings, V2, p. 2.51-2.56. ACEEE, 2140 Shattuck Ave., Suite 202, Berkeley, CA 94704.

Goldman, C., Greely, K., Harris, J. Retrofit Experience in Multifamily Buildings: Energy Savings, Costs and Economics, LBL-25248. Berkeley, CA: Lawrence Berkeley Laboratory, 1988.

Hewett, M., Emslander, H., Koehler, M. "Heating Cost Allocation in Multi-family Buildings: Energy Savings and Implementation Standards." ASHRAE Transactions 95, pt. 1 (1989): 789-797.

Landry, R., et al. "Field Validation of Diagnostic Techniques for Estimating Boiler Part-load Efficiency." ASHRAE Transactions, 100, pt. 1 (1994): 859-875.

Lobenstein, M. S., Dunsworth, T., Hewett, M. "Energy Savings and Field Experience from Converting Steam-heated Buildings to Hydronic Heat." ASHRAE Transactions 99, pt. 1 (1993): 1282-1290.

Lobenstein, M.S., et al. "Field Testing of Various Energy-saving Measures for Domestic Hot Water Heating in Multifamily Buildings." In Proceedings of the American Council for an Energy Efficient Economy 1992 Summer Study on Energy Efficiency in Buildings, 2, 2.145-2.155. Berkeley, CA: ACEEE, 1992.

Mary Sue Lobenstein is an engineering analyst and Martha Hewett is a senior research analyst at the Center for Energy and Environment, a Minneapolis nonprofit organization focused on applied research and on the design, implementation, and monitoring of innovative energy efficiency programs.