Figure 1. Dewpoint and drybulb design condition outdoor air loads (Neutral = 75 degrees F/50% rh) - sensible heat ratios.
As our elderly population increases, so does the need for buildings that meet their special requirements. Research suggests that when designing facilities for seniors, hvac engineers should pay special attention to humidity control because of its impact on comfort, potential mold and mildew growth, and the bottom line. A recent independent field study provides useful insights into ventilation air system design and confirms the effectiveness of actively regenerated desiccant dehumidification, shedding light on operating costs and savings potential.

The Gas Research Institute (GRI)-funded study, conducted at a nursing home in Wilmington, DE, analyzed ventilation air strategies, their impact on worker and resident comfort, and energy implications. The facility's two resident wings were retrofitted with new rooftop ventilation air systems, one with a high-efficiency DX unit and the other with a combination unit that integrated both DX and actively regenerated desiccant rotor technology. Resident room air conditioning units were reconfigured to recirculation mode only. Weather data, internal temperatures, and humidity levels, energy use, and equipment performance parameters were recorded and analyzed.

The findings were conclusive and dramatic. The wing with the desiccant unit maintained consistently lower dewpoints and rh levels regardless of ambient conditions. This translates into superior comfort control and reduced potential for mold and mildew growth. The "standard design" wing, which served as the base-case for comparison, frequently experienced uncomfortable dewpoints ranging between 65 degrees and 69 degrees F with humidity levels regularly exceeding 60% rh and sometimes 70% rh.

The gas-fired desiccant equipment configuration tested in this study provided tighter humidity control for only $0.10/sq ft more for an entire 120-day cooling season or, put another way, a daily total of about $0.25/resident. "That's a reasonable cost tradeoff for the benefits," says Steve Ungar, corporate project liaison and regional director of plant operations for HCR/Manor Care, a leading nursing home chain and owner-operator of the Wilmington facility.

"The comfort of residents is a top priority for us," he says. "The challenge is to keep costs down and keep it simple to operate and maintain." According to the desiccant unit's manufacturer, Munters DryCool, a newer, simpler equipment configuration, developed in part using information from this study, eliminates the premium while delivering similar benefits.

Figure 2. Wilmington, DE weather data in joint frequency bins.

Humidity's Impact on Comfort, Health

Nursing home workers' and residents' comfort perceptions and needs are quite different due to a phenomenon commonly referred to as "metabolic mismatch." Elderly people often suffer from circulatory problems and degradation of their thermal sensing and control mechanisms, both of which diminish their ability to regulate body temperature. Generally, they're less active, have lower metabolism and therefore produce less body heat and prefer warmer temperatures, typically 77 degrees to 80 degrees.

On the other hand, hard-working nurses and housekeepers have higher metabolic rates and produce much more heat than the residents they serve. They must rely more on evaporation (perspiration) to supplement their bodies' release of heat by convection and radiation. The key to satisfying residents' and workers' divergent comfort requirements is to provide a higher drybulb temperature and lower dewpoint than might be more commonly specified for other commercial space conditioning applications.

This view is supported by Lew Harriman, lead author of the recently published Humidity Control Design Guide for Commercial and Institutional Buildings (ASHRAE, Inc.: 2001). He suggests that the standard design point of 75 degrees /50% rh (55 degrees dewpoint) probably is not appropriate for nursing homes and that 78 degrees with a dewpoint of 50 degrees might be a more suitable target condition. "At a lower dewpoint, the active body releases heat more efficiently through fast evaporation of perspiration. At a higher drybulb temperature, the older, less active, and less responsive body is more comfortable," says Harriman. "The summer temperature difference across the building envelope is reduced. This reduces the cooling load, which can save both construction and operating costs when the designer can resist the temptation to oversize the cooling equipment," he added.

Lower rh also helps improve IAQ and occupant health. Nursing home living spaces are cleaned and disinfected on a regular basis to reduce odors caused by incontinence, minimize microbial growth, and reduce the risk of infection. This frequent cleaning adds to internal latent loads which, along with large outdoor ventilation air volumes and infiltration, can cause rh to rise, especially during spring and summer. Humidity levels above 60% rh promote growth of mold and mildew, which release odors and mycotoxins as well as other particulate matter. These are believed to create and/or aggravate allergies and other health problems in susceptible individuals, especially the elderly who have reduced respiratory capacity.

To maintain comfort and compensate for internally generated moisture loads, the GRI study recommends that ventilation air be delivered between 72 degrees to 75 degrees and a 48 degrees to 50 degrees dewpoint, depending on delivery location. This is several degrees cooler and several grains per pound (gr/lb) drier than the target space condition. It supplements sensible cooling provided by resident room packaged terminal air conditioners (PTACs) while "pulling" moisture from carpets, furnishings, wall and window coverings, and other surrounding materials, thus reducing potential mold and mildew growth.

Figure 3. Ambient and internal conditions during high latent/high sensible load period.

New Perspective on Latent Loads

In the 1997 ASHRAE Handbook - Fundamentals, outdoor air latent loads received long-overdue recognition. For the first time, the chapter on climatic design information included design dewpoint conditions in addition to the more prevalently used cooling design drybulb conditions. Most notable in this new perspective on existing data is that, in nonarid climates, total outdoor air enthalpy at the design dewpoint condition is more than at the design drybulb condition.

This point is illustrated in Figure 1, which shows, for several U.S. cities including Wilmington, Delaware (field test site), the Btu necessary to bring 1,000 scfm of outdoor air from both dewpoint and drybulb design conditions to a "neutral" 75 degrees and 50% rh.

Also shown is the sensible heat ratio (SHR) for each design condition, which is the ratio of sensible load to total cooling load. Doug Kosar, principal research engineer at the Energy Resources Center at University of Illinois - Chicago explains the importance of SHR in coil performance and humidity control. "Most DX cooling coil performance specifications are based on ARI ratings of 95 degrees db/75 degrees mcwb [mean coincident wetbulb] outside condensing temperature and 80 degrees db/67 degrees mcwb entering air condition. This rating represents about 20% outside air volume mixing with return air of 75 degrees db/55% rh. Assuming an apparatus dewpoint of about 55 degrees , the SHR of the load on the coil is about 0.75," says Kosar. "But if the outside air condition is 80 degrees db/75 degrees mcwb, the SHR drops to about .65. This creates a mismatch between load and coil SHRs, which allows moisture loads to pass unmet through the cooling coil and internal humidity climbs."

While design conditions are important and, by definition, represent only a limited number of hours, an analysis of yearly weather data and ventilation air cooling loads confirms that, in most areas of the country, total latent loads are far greater than total sensible loads. For example, while Tampa, Florida has a larger total cooling load than that of Wilmington per constant scfm of outdoor air, the latent-to-sensible load ratios of these two cities are quite similar at about 6.1:1.

A review of Wilmington's weather data illustrates this point (Figure 2). This 3-D graph displays cooling season ambient conditions using joint frequency data bins. Each bar on the y-axis represents the number of hours that outdoor air meets both the temperature (degrees F) and humidity ratio (gr/lb) parameters defined in the x- and z-axes, respectively. The chart shows the large number of hours that outdoor air humidity levels are high when temperatures are fairly moderate.

To achieve humidity control, especially during periods of high latent loads, hvac design professionals must address three ventilation air challenges. The first is to determine the amount of fresh air needed, the second is to determine the best air distribution strategy to maintain pressurization and balance requirements, and the third is to select the right equipment to precondition it.

Figure 4. Control rh by drying or dehumidifying air, then ducting it directly to each room.

Ventilation Air Requirements, Strategies

Nursing home ventilation air volumes and air change rates are determined by local code requirements. Most reference or incorporate ventilation tables from ASHRAE Standard 62, "Ventilation for Acceptable Indoor Air Quality," the revised 2001 "AIA Guidelines for Design and Construction of Hospital and Healthcare Facilities," and/or the "International Mechanical Code." Generally, the trend in the latest code revisions has been to increase fresh air volumes and air change rates. Historically, the nursing home industry has more than met these minimums in order to achieve odor control.

For the past 30 years, the industry's "standard" hvac design has comprised use of conventional or heat pump PTACs in resident rooms with either rooftop DX units or split systems for corridors, dining rooms, lounges, and other service areas. In large facilities with central chiller and heater plants, the approach is similar with fancoil units in resident rooms and remote air handlers serving common areas. Ventilation air has been introduced primarily through rooftop or central AHUs, with dampers set as high as 75% or more open, supplemented by individual PTACs set at 10% outside air.

This ventilation air strategy leads to humidity control problems in most areas of the country because PTACs and fancoil units are rarely wired and controlled to supply a constant supply of ventilation air. When residents turn off units, (e.g., due to noise or drafts), they cause the air balance of the space to go negative. That's because central exhaust systems typically run 24/7/365 for odor control, with fans pulling a constant volume of air from resident bathrooms and soiled linen closet(s).

A related concern is integrity of the building envelope and air distribution systems. "Many of the moisture problems in eldercare facilities can be traced to exhaust ducts, return air ducts, and wall-mounted cooling units that create suction in building cavities because they're not sealed up tight," says Harriman. "When the building cavities are under suction, they pull humid outdoor air into the building through cracks in the exterior walls. Moisture in that air condenses when it gets inside the building and contacts a cool surface, and mold and mildew problems result."

In the rare case where PTACs are wired with fans in constant ventilation mode, they still allow large volumes of humid air to pass through the coil untreated. Most PTACs on the market are oversized for this size space and quickly cycle off after satisfying sensible load (temperature). This is an underlying problem with ventilating with PTACs or fancoil units, a common design practice in senior living and other lodging applications.

A better approach is to use PTACs in recirculation mode only and to bring all ventilation air through a unit dedicated either solely or primarily to that task. "We've definitely gone that direction with new facilities and several older problem sites," says David Schoonmaker of H.T. Bernsdorff (Toledo, OH), an HCR-Manor Care engineering consultant. "PTACs are fine for cooling or heating, but when it comes to outside air, they present more problems than they're worth like water puddling, spillage, drainage systems, and the related mildew problems," he says. "It's better to handle the moisture in one place, keep things pressurized and reduce the mechanical problems for the on-site maintenance person."

Dick Kelley, principal at I.C.E. (Little Rock, AR), consulting engineer to another major health care provider, concurs. "For the past five or six years, we've opted to deliver conditioned ventilation air directly to patient use areas ... and not just the rooms, but for the central units serving the dining room and other spaces too," says Kelley. "Most of the new homes are designed to look more like a residence so they don't have flat roofs," he adds. "We'll use split systems with a separate unit supplying all the makeup air to the returns of the air handlers."

"Unfortunately, standard DX units have similar difficulty in handling large volumes of outside air," says UIC's Kosar. "To compensate, a design engineer may be inclined to increase tonnage and coil size, but this merely exacerbates the problem," he notes. "Bigger systems satisfy sensible loads faster and cycle off sooner providing even less dehumidification performance. Engineers now recognize the magnitude of latent loads and are beginning to realize that conventional systems just can't get it done."

New Equipment Options

To combat outdoor air latent loads and address the shortcomings of conventional systems, a variety of commercial ventilation air equipment have been introduced in the market. DX units with condenser heat reclaim, for example, use deeper coils to overcool air to achieve a low dewpoint.

Then, on the discharge side of the cooling coil, another coil circulating hot refrigerant from the high-pressure side of the system provides "free" reheat, bringing the supply air temperature up to a more acceptable level. Other units incorporate passive desiccant (enthalpy) wheels to supplement DX coils, relying on a supply of exhaust air to mitigate outdoor air latent load. While both these approaches can be effective during hot and humid periods, each presents it own challenges during cooler, high-humidity periods, which, as noted above, occur much more frequently.

Actively regenerated desiccant dehumidification systems control humidity across a wider range of conditions. Unlike enthalpy wheels, they introduce additional heat to "stretch" the vapor pressure differential between the desiccant material and the incoming humid air stream for deeper grain depression. In addition, they don't require exhaust air for regeneration.

This is an advantage in situations where ductwork hurdles make this option unavailable or where use of exhaust air is undesirable, as is the case in nursing homes. The sources of regeneration heat are many and include gas burners, steam or hot water coils, electric resistance elements (although this option is rarely economically justified), recovered process heat, or condenser heat reclaim.

Active desiccant dehumidification systems have long been specified in a variety of industrial applications where their deep-drying capability is critical to improving process control and/or produces other economic benefits. They're also established in niche commercial markets like grocery stores, ice arenas, cold storage warehouses, and other refrigeration applications where lower humidity translates directly into improved economics. Over the last seven years or so, they've been gaining favor in other commercial and institutional makeup air applications, a trend that continues as manufacturers introduce new products specifically engineered to meet the market's performance and cost expectations.

Field Test

HCR/Manor Care Unit #532 is a two-story 58,000-sq-ft skilled nursing facility located about 1.5 miles from the Delaware River just outside of downtown Wilmington. Constructed in 1982, the block and brick structure comprises three sections, the 27,000-sq-ft main "core" building and the "premium" 17,000-sq-ft Heritage-Arcadia and "standard" 13,800-sq-ft Dover-New Castle resident wings. The test involved comparison of the resident wings only as their design and construction facilitated isolation of hvac loads and energy use.

The Heritage section (the upper floor) features 26 private rooms, each about 220 sq ft including private half-bath, carpeted hallways and rooms, upgraded cloth-upholstered furniture, wallpaper, and cloth curtains throughout. These moisture-absorbent materials are noted because they impact capacitance and humidity control of the space. The Arcadia section (the lower floor) includes 28 private and shared resident rooms (most with half-bath), carpeted hallways, and fewer upgraded finishes. Each resident room and the two TV lounges are equipped with a heat pump PTAC. On both floors of the Dover-New Castle wing, two or three persons share the 285 sq ft resident rooms and there are only 18 rooms per floor. Corridors and rooms have tile floors and vinyl curtains. As in the adjacent wing, Dover-New Castle resident rooms and TV lounges are also equipped with PTACs.

In late 1998, an 18-year-old 15-ton heat pump serving the Heritage-Arcadia wing was replaced by a Munters DryCool 4,800-cfm prototype package with gas-fired desiccant rotor, heat pipe, 10-ton DX post-cooling coil, and a gas-fired heating section. A thermostat and humidistat mounted in the Heritage corridor controlled the unit's operation. On the Dover-New Castle wing, a new, high-efficiency, 10-ton DX unit with gas-fired heat replaced an inefficient, old, 10-ton heat pump. A Dover corridor thermostat controlled it. All ventilation air for both wings was reconfigured so that it comes through the two rooftop packages, which feed their respective wing's upper and lower corridors; all PTAC outdoor air dampers were shut.

Temperature and humidity sensors were installed throughout both wings and on the roof. Each wing's electricity use was submetered; watt transducers isolated rooftop unit, PTAC, and plug-load energy use. In addition, current transformers captured compressor and fan run-time data and gas submeters recorded dehumidification and heating loads. Site-mounted dataloggers continuously collected information which was downloaded regularly for analysis, and frequent site visits were made to check instrumentation, assess comfort perceptions, and record observations.

Analysis of the 1999 cooling season data indicates that the desiccated wing had far better comfort and humidity control than the base-case wing. For example, Figure 3 shows ambient temperature and humidity ratio (gr/lb) during a hot, humid 48-hour period in August 1999 as well as the resident wings' temperatures and rh levels. While temperatures in both wings were near identical at 78 degrees to 80 degrees , Dover-New Castle humidity hovered between 60% to 70% rh the entire time while the desiccated Heritage-Arcadia wing remained comfortably between 40% to 50% rh. That week alone, moisture removal in the desiccated wing averaged as much as 435 gal of water per day, or about 8 1-gal buckets/resident room/day!

Data analyses further indicated that this prototype desiccant unit configuration consumed more fan energy than necessary, so several modifications were made and additional data was collected in summer 2001. Once again, space conditioning control in the Heritage-Arcadia wing was excellent. During peak dewpoint and drybulb design conditions, both units easily held drybulb temperatures in their respective wings between 75 degrees to 77 degrees but the Heritage-Arcadia wing dewpoint averaged 52 degrees while the dewpoint in the adjacent wing averaged 57 degrees and wandered as high as 64 degrees. The new configuration's dehumidification performance was equally impressive, removing over 575 gal of water during one 24-hr period.

Furthermore, when the Heritage-Arcadia wing hvac energy data was normalized against the adjacent wing using square footage and measured plug-loads, analysis confirmed that the additional cost of superior humidity control for the 2001 cooling season was about $1,000 for natural gas and $688 for electricity based on $0.50/therm and $0.06/kWh. For a 17,000 sq ft space, this equates to less than $0.10/ sq ft for the mid-May through mid-September dehumidification season.

Figure 5. View of "base-case" Dover-New Castle wing shows conventional rooftop DX unit, resident room PTACs, and central exhaust fan.

Conclusion

The good news for nursing home industry leaders is that the cost of better humidity control is coming down even more, according to Jeff Siemasko, marketing manager for Munters Dehumidification. "We recently introduced the HCU model, which precools makeup air with a DX coil prior to the desiccant rotor," says Siemasko. "It reclaims free condenser heat to regenerate that moisture off the rotor in the opposing airstream. This way, the unit provides similar dehumidification performance as the prototype measured in the field study, at an even lower operating cost and less first cost."

"Humidity affects nursing home life in so many ways," says Schoonmaker, "from resident comfort to worker satisfaction to building and interior remodeling costs. That's a small price to pay for better control as long as we - the consulting engineer - can communicate these benefits to the client. We have to work within the available budget." ES