Improving Hot Water Delivery in Multifamily Buildings
Usually the most efficient way to distribute hot water in a multifamily building is with a recirculation loop. In recirculation loop systems, one or more long loops of large-diameter pipe carry hot water around the whole building to within a few feet of each point of use. Small-diameter pipes—typically 1/2 inch but preferably 3/8 inch—connect the loop to the faucets and other end uses. The flow in the loops is driven by one or more recirculation pumps.
Recirculation loop systems are usually extremely water efficient compared to trunk-and-branch systems or individual water heaters. This is because the water in the distribution pipes is not allowed to cool, and therefore tenants don’t have to draw and dump a large amount of cool water before the hot water arrives. Water consumption in buildings has become a major—and growing—public cost and environmental concern, especially in the western states. Technologies that reduce demand or provide for water recycling offer the best chance for coping with increasing water demands as our population grows.
As well as being water efficient, recirculation loops meet two other essential goals: They can reduce energy use in larger buildings, and they provide quicker hot water service to tenants, compared to trunk-and-branch systems or to individual water heaters. The break-even point for energy savings over trunk-and-branch is typically around eight dwelling units. Compared to individual water heaters in each apartment, recirculation systems also reduce maintenance costs and flood risk from leaking individual water heaters.
On paper, recirculation systems have many benefits, but in practice there are at least four issues that designers and facilities managers must address to reap the full potential for energy and water savings. These are piping design, recirculation pump reliability, crossover flow, and control of pump on-time or water temperature.
These issues are the subject of ongoing research, but in the meanwhile there are steps that building owners and operators can take to minimize risks to performance. In 2005-06, the Heschong Mahone Group (HMG), conducted a field study of multifamily hot water recirculation system performance for the California Energy Commission. For this study, HMG monitored and analyzed recirculation systems in three multifamily buildings, retrofitting each with four different control systems over the course of the project.
Recirculation Pump Reliability
Anecdotal evidence, which HMG’s recent field study appears to confirm, suggests that recirculation pump failures are a frequent occurrence in multifamily buildings. The field study found that the recirculation pump was not working in one of the three buildings studied, and that the pump controls were disabled in another building. HMG also analyzed a few days of data obtained from EDC Technologies, Incorporated, a manufacturer of domestic hot water controls, on recirculation system performance at an additional 36 buildings. EDC has been installing controls systems on multifamily central hot water loops since 1985. For over a decade, their systems have typically included real-time monitoring with communication via the Internet. The monitoring systems send an e-mail alert when certain monitored indicators fall outside the expected norms. EDC then communicates with the building maintenance staff to determine the exact type of failure that has occurred. Their data showed that, at any given time, 12% of those systems are experiencing component failure—not controls failure. The recirculation pumps were the element that failed most often.
Pump failures are likely to increase both energy use and water use, because when water is not being circulated through the hot water storage tank, it cools in the loop below a usable temperature between tenant uses. Tenants then have to draw and discard this lukewarm water before they receive water that is hot enough to use. If the pump is working, the water (which is generally warmer than the cold makeup water) is circulated back to the storage tank—preserving the embodied heat energy. Of course, tenant dissatisfaction about long wait times for water may be more costly to a building owner—by increasing vacancy rates, for example—than the increased energy or water use.
When a pump fails, the reason for the increase in water use is clear. The reason for increased energy use is more complex. On the one hand, pump energy is saved because the pump is off, and between-use heat loss through the pipe and insulation is reduced. On the other hand, the heat energy of the lukewarm water is wasted when it is dumped down the drain. Which way this balance tips depends on many factors that are difficult to model, and a valid generalization is difficult. The most important “energy factor” may be the response of the building maintenance staff.
Results from the HMG field study (again supported by anecdotal evidence) suggest that maintenance staff’s most common response to tenant complaints about long wait times for hot water is to turn up the hot water temperature. They typically do this without realizing that the problem is caused, not by low temperature at the water heater, but by failure of the pump or some other element of the system. In the buildings monitored, supply temperatures of as much as 150ºF were recorded, leading, obviously, to very high energy use. Ensuring that pumps operate reliably, or detecting and repairing failed pumps, needs to be a priority.
The root causes of pump failure are not always clear, but candidates include the presence of air in the hot water return pipe, the presence of corrosive substances in the recirculation loop, and the buildup of mineral deposits on pipe and pump surfaces. There are several strategies for reducing possible pump failures. Adding an 18-inch vertical pipe with an air release valve just upstream of the recirculation pump will remove air bubbles that might otherwise cause the pump to cavitate, overheat, and eventually fail. Remote monitoring of the pump can provide early warning of failures, which reduces system downtime.
Crossover—hot water flowing thorough the cold water pipes or vice versa—is a problem that is unique to recirculation systems. Crossover generally occurs because the recirculation pump creates differences in pressure around the recirculation loop and between the water in the hot and cold water piping. These pressure differences, in turn, drive crossover flow through faucets. The water flows through worn valve seats or single-lever valves that were not designed for pressure differentials between the hot and cold water lines. Crossover flows increase water consumption because they lead to cooling of hot water in the recirculation loop. For the same reason, they can also increase energy use.
The crossover effect is widely known among plumbers who work with central hot water recirculation loops, but HMG could not find any data on the magnitude of heat loss and water waste caused by crossover flows. Consequently, they measured these flows as part of the study. The data showed crossover flow in two of the three buildings, with a magnitude of 15 to 35 gallons per day per dwelling unit. These flows accounted for around 4% of total domestic hot water energy use. The study measured only one of the many possible routes for crossover flow, so the estimated waste is, at best, a minimum condition for these buildings.
The crossover flow the study was able to measure is the flow of hot water from the storage tank back along the cold water supply pipe, and into cold water pipes throughout the building (see Figure 1). This can cause tenants to receive warm or hot water through their cold water faucets, as well as cold water from their hot water faucets. More importantly for the economics of the building owner, the hot water in the recirculation loop is eventually cooled by the crossover of cold water, and when it returns to the hot water tank, it has to be heated again.
The study was not able to measure the second type of crossover flow—flow that takes place between the hot and cold pipes but does not pass through the storage tank. Many experts believe that this type of crossover flow is caused mainly by leaking of worn single-lever valves and pressure differentials between floors of a multistory building (see Figure 2). Certain single-lever valves, even when new, may also permit crossover due to an “open-barrel” design. Single-lever valves seldom cause crossover flows in buildings without recirculation pumps, because the pressure on the hot and cold sides of the valve is generally more equal.
Preventing Crossover Flow
The best way to prevent crossover flow is to install check valves to prevent flow between the hot and cold supply lines, or to retrofit worn or poorly designed valves with third-party valve cartridges designed to fix this problem (see photo, p. 45). Nomix, Incorporated, supplies retrofit valve cartridges for Mixet and Moen valves.
Many of the newer low-flow showerheads are equipped with shutoff taps, which allow the user to reduce water flow without changing the temperature settings of the valve handles. For example, while soaping up, one can reduce the shower from 2.5 gallons per minute to just 0.5 gallons per minute. Unfortunately, some showerhead shutoff taps, shut the flow off entirely, allowing crossover flow through the shower mixing valve. Shutoff taps are an energy-saving feature only when a recirculation loop is not present, or when the flow is not completely shut off.
The first type of crossover flow is preventable with a check valve on the pipe that supplies the hot water system with cold water. That way, the hot water cannot enter the cold water supply line. The second type of crossover flow can be prevented by installing check valves on both the hot and cold water pipes leading into each leaking appliance or single-lever valve. The only other alternative is to replace leaking valves with new leakproof ones.
Design of the Recirculation Loop
In every central domestic hot water system, more hot water is drawn from the heater than is actually used by the occupants. As hot water cools in the pipes between uses, some amount of cooled water is inevitably drawn through the faucet before hot water arrives. The amount of wasted energy is determined largely by the routing and location of pipes, pipe material, and insulation. Careful routing of the recirculation loop should minimize its length, the number of right-angle turns, and the length of the final delivery pipes. Single-family homes with extensive piping have been found to consume as much as 200% more energy than the best systems; presumably, there is an equivalent increase in wasted water.
For recirculation systems, energy and water use is also affected by elapsed time between draws, which in turn is affected by the number of apartments served by each loop, the time of day or night, and the demographics and behavior of the occupants. Additional factors include the response time of the water-heating system, length and diameter of pipes, and crossover flow. These factors all affect the savings that can be achieved by the various types of control that are available for recirculation systems.
Another important aspect of recirculation loop design is the sizing of the pump or pumps. Large pumps, when coupled with demand controls, can save energy by reducing temperature differentials around the pipe and providing a faster response so that, for example, tenants get hot water more quickly after a long period without demand. However, large pumps also increase the rate of crossover flow because differential pressures increase. Additionally, they create faster and more turbulent flows that can lead to increased pipe wall erosion and the potential for leaks.
Smaller pumps, coupled with controls that modulate the temperature of the hot water being delivered based on the amount of demand at a particular time, don’t shut the pump off. They save energy by reducing the volume of water moved and the water temperature, thus reducing heat losses from the recirculation loop during periods when no tenants are using hot water. In some building designs, a pump that moves as little as 1 gallon per minute can keep the whole line hot enough that tenants receive hot water almost instantly. This obviously helps to conserve water.
Erosion of piping usually happens just downstream of a 90º bend in the pipe, where fast, turbulent water erodes the outermost section of the pipe wall. Therefore, pipe layouts should be designed without 90º bends whenever possible. This can be done by straightening the pipe route, changing the location of faucets, or replacing 90º bends with more gradual curves (perhaps using PEX tubing). The sizing and routing of pipes, and the choice of other equipment, is a complex issue; industry magazines and forums can be useful sources of advice.
In many recirculation systems, the pump needs only a few seconds to fully charge the recirculation loop with hot water. In systems with intermittent use, a pump control that switches the pump on only in response to a demand for hot water (as measured by a pressure transducer) can save a significant amount of energy. A well-tuned demand control will switch off the pump once the temperature of the hot water return pipe reaches a threshold value, indicating that the loop is fully charged. Demand controls save energy because they reduce the electricity use of the pump, and because they reduce the heat loss from water in the recirculation loop during extended periods between uses (see “Benefits of Demand-Controlled Pumping,” p. 48).
Demand controls in multifamily buildings typically work differently from demand controls in homes. In single-family homes, a manual switch or an occupancy sensor turns on the local pump, drawing hot water from the tank and dumping the unacceptably cool water back into the cold water line until the water at the outlet reaches an acceptable temperature. When controlled by a manual switch, these systems often save energy as well as water by reducing the wait time. When controlled by an occupancy sensor, hot water can often be drawn from the tank when it is not required—when someone just walks into the bathroom for some reason other than to use water, for example—leading to an increase in energy use.
The demand control systems that HMG tested in multifamily buildings sensed demand by a different mechanism—a pressure transducer in the recirculation loop. In the field study, advanced controls, including demand controls and temperature modulation controls, saved an average of 27% of the domestic hot water energy compared to the systems they replaced. In buildings that are vulnerable to crossover flows caused by leaking single-lever valves or other devices, demand controls can be especially effective, because pump-induced crossover flows do not occur when the pump is switched off. Demand controls may also extend the life of the hot water pipes by reducing erosion of the pipe walls caused by high-velocity flows and turbulence.
Another less effective but occasionally worthwhile form of pump control is the simple time clock. The time clock saves energy by switching off the pump overnight when there is likely to be no demand for hot water. The off periods should be carefully chosen so that the loop is charged before the first tenant requires hot water. Time clocks should also shut the recirculation pump off at times when there is enough water usage to keep the hot water pipes fully charged with hot water, such as midmorning and late afternoon. Monitoring hot water demand for a few weeks before installing the controls is a good way to ensure adequate service.
Even with a well-chosen off period, however, occasionally a tenant will need hot water in the middle of the night. In buildings that have a lot of apartments served by a single loop, occupants may experience a delay of many minutes before hot water arrives. Night-time off periods are inappropriate for very large loops unless the off periods are brief, and then there are little or no energy savings.
Commissioning of demand control systems is very important—especially the choice of the supply water temperature. If the supply temperature is only just above the required use temperature (for instance a supply temperature of 108ºF), then too often the water will cool below the use temperature before the next draw event. The whole loop will have to be recharged with fresh hot water. However, if the supply temperature is too high, then heat loss from the loop and from the storage tank will be excessive. The ideal balance point between hotter and cooler supply temperatures varies with the system, the building layout, and the occupants. In practice, that balance point is determined by the system
installer based on experience, and modified based on tenant satisfaction.
Control Of Water Temperature
An alternative to using demand control is to run the recirculation pump continuously, but to change the temperature of the water based on usage patterns. This is known as temperature modulation control. When demand is low, the hot temperature can be lower—108ºF, for example—and still meet all needs. When demand increases, the hot water temperature gets raised to perhaps 125ºF so less hot water is needed at every faucet to provide the right mix. Lowering the hot water supply temperature when demand is low reduces heat loss from the loop and the storage tank during those times.
Temperature modulation controls save the most energy when set-back times and temperatures are closely aligned with tenants’ draw patterns. Because draw patterns vary depending on climate, demographics, and other factors, temperature modulation systems typically include a mechanism that allows the setback to be adjusted. This mechanism can be based on remote monitoring and regular review of setpoints by the controls manufacturer or the installer, or it can be an automatic learning algorithm incorporated into the control system. Both mechanisms appear to have advantages and disadvantages. So far, there is no definitive third-party research to indicate when one mechanism should be preferred over the other.
Temperature modulation controls may be especially compatible with condensing boilers and heaters, because condensing systems work most efficiently at part load. Temperature modulation controls reduce the amount of time that heaters spend at full load by creating a buffer of hot water in the storage tank before periods of high demand.
Both types of advanced control system—demand controls and temperature modulation controls—often incorporate remote monitoring to allow faults in the system to be detected, diagnosed, and reported to the building manager. Quickly and accurately identifying faults is especially valuable to owners of multiple properties, to government agencies, and to organizations such as health care providers for whom hot water failures are a very serious problem.
Controls manufacturers claim that crossover flows, pump failure, and boiler or heater failure can be detected just by monitoring temperature at several points in the system, or by monitoring pressures or flow at a few spots (see Figure 3). The accuracy of fault detection has not yet been independently researched.
Some manufacturers offer fault detection services on a subscription basis. In addition to identifying faults quickly, this type of service gives building managers a level of insight into the functioning of the system that would otherwise be unavailable to them, and gives them a point of contact for technical questions. Such real-time information may result in better and more frequent maintenance of domestic hot water systems, and therefore, reduced energy and water consumption.
Because a well-designed and maintained central hot water system can save a lot of energy and a lot of water, several California utility programs provide design assistance, developer incentives, and training in the operation and maintenance of central domestic hot water systems.
Two such incentive programs are Pacific Gas and Electric Company’s Multifamily New Homes program (MFNHP) and Southern California Edison’s California New Homes Multifamily program. Both programs provide incentives for buildings that exceed the Title 24, California’s building energy code, by at least 15%. One of the easiest ways to achieve high energy savings in multifamily buildings is to improve the central hot water system. The programs also offer design assistance to developers and design teams to help identify the most cost-effective energy efficiency measures for their project. Installing central systems and controls is often found to be one such measure. The new-construction programs provide incentives for the developer of up to $200 per unit, along with $50 per unit for the services of an energy consultant, and $60 for a HERS (home energy rating system) rater. There are also nonincentive programs, which focus on providing information and design support. One such program is the Affordable Housing Energy Efficiency Alliance (AHEEA).
Builders, designers, and architects who want to take advantage of incentives and other services should contact a program representative as early as possible in the design and building process.
Existing multifamily buildings also offer major opportunities for savings. In many cases, building managers can improve tenant satisfaction with a well-designed and -controlled central water-heating system. This will reduce complaints, lower maintenance costs, and increase the value of the property. Building owners and managers should check with their local utility to see what kinds of design assistance or incentives are available.
In this study, HMG found a high degree of variation in water and energy usage between different buildings. This underscores the benefit of monitoring the performance of domestic hot water systems in individual buildings to minimize energy and water use, given the occupancy and usage patterns in each building. The California Energy Commission will be funding further studies of the performance of recirculation loop systems and associated equipment, including controls. In Southern California, Sempra Energy, through its utilities Southern California Gas and San Diego Gas and Electric, is funding research into savings realized from retrofitting recirculation controls in existing systems. One of the aims of both studies is to help owners and developers install the systems that will perform the best.
Owen Howlett is a project manager at Heschong Mahone Group, Incorporated, in Fair Oaks, California. Nehemiah Stone is director of demand side management (DSM) implementation at KEMA Services, Incorporated, in Oakland, California.
For more information:
Preliminary results of this study were presented in Stone, Nehemiah, and Owen Howlett. Central Hot Water Distribution Systems in Multifamily Buildings: 2008 California Building Energy Efficiency Standards. Prepared for the California Energy Commission by the Heschong Mahone Group, June 23, 2006.
Results of a previous study were presented in Stone, Nehemiah, Mudit Saxena, and Jon McHugh. “Code Change Proposal for Multifamily Water Heating.” Codes and Standards Enhancement Initiative. Prepared for Pacific Gas and Electric Company by the Heschong Mahone Group, May 2002. A PDF version of this report can be downloaded from
Ally, M.R., J. J. Tomlinson, and B. T. Ward. Water and Energy Savings Using Demand Hot Water Recirculating Systems in Residential Homes: A Case Study of Five Homes in Palo Alto, California, RNL/TM-2002/245. Oak Ridge, Tennessee: Oak Ridge National Laboratory, October 21, 2002.
Baskin, Evelyn, and Robert Wendt. “Numerical Evaluation of Alternative Residential Hot Water Distribution Systems.” in Minutes of Annual Meeting of American Society of Heating, Refrigerating, and Air-Conditioning Engineers 110, pt. 2 (2004): 1–11.
Rittlemann, William. “Field Investigation of 18 Solar-Assisted Domestic Hot Water Systems with Integral Collector Storage.” 2004 ACEEE Summer Study on Energy Efficiency in Buildings conference in Pacific Grove, California, August 22-27, 2004.
For more information on pipe internal erosion and corrosion, see www.bsee.co.uk/news/fullstory.php/aid/2321/Avoiding_corrosion.html.
For more information on pipe sizing and layout, see www.cuh2a.com/downloads/inthenews/DesignerNotebookMayJune04.pdf.
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)
While we will do our best to monitor all comments and blog posts for accuracy and relevancy, Home Energy is not responsible for content posted by our readers or third parties. Home Energy reserves the right to edit or remove comments or blog posts that do not meet our community guidelines.