Chiller selection: Much more than capacity
Selecting and specifying the right chiller is generally dictated by capacity. There are many ways to specify chillers and chilled-water systems, depending on the application.
- Review types of chillers by general range of capacity.
- Understand the variables of chiller selection.
- Learn about the value of metering and instrumentation to chiller selection and chilled-water system investment.
Chiller selection and system design are more art than engineering, using the variables of demand, capacity, service requirements, building type and site characteristics to form a picture that captures capital investment and value of utility systems. Site is intended as a universal term pertaining to a chiller installation whether as an individual unit dedicated to a single building or as part of a central plant.
The chilled-water system is a significant capital investment but one that offers a benefit or return realized through enhanced value of a property with comfortable living or work environment and/or protect and improve processing and manufacturing. Other variables might include energy efficiency, meeting demand and client requirements.
Selection of chilled-water system components is commonly and reasonably dependent upon a review of capacity requirements, redundancy requirements, relative capital cost, production efficiencies and anticipated operating expense. Metering and instrumentation, while not a direct variable of chiller selection, are often overlooked during the selection process. A means to accurately record and store chilled-water system data can provide a basis for monitoring system performance, diagnosing probable equipment failures and managing system operations and expense, ultimately supporting the inherent value derived from the capital investment.
A chilled-water system is the equipment, piping and controls that provides cooling for space conditioning or process loads. The chilled-water system typically consists of three types of equipment: chillers, cooling towers and pumps.
A chiller is simply a means of removing heat from a space or process load using a vapor compression cycle. Conceptually, the chiller consists of four basic components: compressor, condenser, expansion valve and evaporator. After compression, the working fluid (refrigerant) is a superheated vapor and can be condensed through the coils or tubes of the condenser, using either air or water for cooling. The condenser is where heat is rejected from the system.
The condensed liquid refrigerant is then expanded, thereby reducing its pressure and temperature, yielding a mixture of fluid and vapor. The temperature of this mixture is colder than the space or process to be cooled. Warm air or warm water circulates through the evaporator, evaporating the liquid portion of the refrigerant mixture. The evaporation process cools the circulating warm air or warm water. Simultaneously, the refrigerant is absorbing and removing heat from the space or process load served by the chiller. The refrigerant vapor leaving the evaporator is then compressed to repeat the vapor compression cycle.
Air circulated and cooled through the evaporator typically is used to directly condition space and is usually arranged for an individual building or selected space within a building. Water circulated through the evaporator is in a pumped, closed loop distribution piping system with control valves to deliver cooling where needed, whether cooling coils or heat exchangers. Chiller water systems can be configured to serve individual buildings, selected spaces within a building, group of buildings.
Types of chillers
Chillers often represent the largest capital cost of chilled-water systems. Chillers are generally classified by the type of compressor used in the refrigeration cycle. The predominate types of chiller compressors are scroll, screw and centrifugal. The capacity of scroll chillers nominally ranges from 10 to 300 tons. To provide larger capacity, a scroll chiller will typically have multiple blocks of smaller compressors (e.g. 20 tons each).
The primary response to changing loads is staging blocks of capacity rather than a controlled modulation of total capacity. Significant chiller capacity remains if one of the compressors fails. Screw chillers are available with capacity ranging from 30 to 800 tons. Screw chillers typically maintain design efficiency (kW/ton) at partial loads. The screw compressor can vary capacity in response to changing load, from 20 percent to 100 percent of capacity.
Scroll and screw chillers are positive displacement technologies, relying on compression of refrigerant gas. Centrifugal chillers incorporate dynamic technology that uses a change in velocity as a means to increase pressure for compression of the refrigerant gas. Centrifugal chillers are available in capacity ranging from hundreds to thousands of tons. Centrifugal chiller capacity can be seriously compromised at low-load conditions. At low-load conditions the flow of refrigerant may not be sufficient to maintain flow through the compressor causing a stall or a surge if flow is momentarily reversed until pressure is sufficient for flow to resume. However, this flow issue can be mitigated with variable speed drives. Because of the wide range of capacity and potential applications, the centrifugal chiller compressors can be coupled with alternate drives: electric motor, steam turbine and even reciprocating engines.
Absorption chillers offer an alternative to the compressor-based chillers. The compressor is replaced by a heat cycle, which changes the pressure and phase of the working fluid, such as ammonia or lithium bromide solution. The energy source for the heat cycle may be steam, hot water, recovered waste heat or direct fired natural gas. Steam absorption chillers can use low pressure (15 psi) and high-pressure steam (125 psi). Absorption chillers using hot water require typically a temperature of 230 degrees F. and possibly as low as 160 degrees F. for waste heat recovery absorption chillers. Absorption chillers typically do not respond well to rapidly varying loads or chilled water flows. Absorption chillers also tend to have higher maintenance expense compared to other chiller types.
Chilled-water systems must reject the heat from the building or process load to a heat sink. Air-cooled chillers use air as the medium to dissipate collected heat. As the rejected heat approaches the ambient dry-bulb temperature conditions, the chiller must operate at a high refrigerant temperature and pressure to reject that heat and continue to serve the load demand. Contrast a water-cooled chilling cycle, which circulates water through a cooling tower or a lake, river, or other water source, to reject heat from the condenser. Using evaporation, the water circulated through the cooling tower approaches the ambient wet-bulb temperature, which is generally much lower than the dry-bulb temperature. Depending on the project location, they may approach each other and cause other variables to affect the chiller type selection.
The compressor of water-cooled chiller needs less energy to provide the same amount of cooling. Air-cooled chillers are typically lower capacity, less than 400 tons, although in areas where water is not available to make-up evaporation losses from tower operation, large centrifugal chillers have been designed for operation with air-cooled heat rejection. Cooling towers are generally used for larger centrifugal chillers, 400 tons and greater. The higher capital cost of a water cooled chilled-water system should be evaluated in the context of longer service life (attributable to operation at lower temperature and pressures) and higher overall chilled water production efficiency.
However, the differential of production efficiency between water cooled and air-cooled chilling cycles is dependent on the ambient conditions of chiller location. Note that steam turbine drives and absorption chillers will require larger cooling towers than electric drive centrifugal chillers to condense the steam as well as the refrigerant.
Carefully choosing the best or most suitable chiller and associated system equipment should be predicated upon objective evaluation of the specific characteristics and service requirements of the site (see Figure 2).
Site service requirements includes maximum demand for chilled water service, but also the relative variability of chilled water demand, minimum demand for chilled water as well as length of the cooling season and annual chilled water production (ton-hours). Scroll chillers or screw chillers are more suitable for smaller loads with high variability. A central chilled water plant configuration and large load can affect this generalization. A central plant with multiple units can serve the peak cooling demand and non-peak loads regardless of chiller type.
Primary site characteristics include space for installation of chillers and towers as well as access to the space, considerations for noise of plant operation, local water use restrictions/availability for tower operation and ability of the site and capacity of site utility infrastructure to accommodate the operation of the chiller plant. These site characteristics may influence the type of chillers and configuration of the chilled-water system.
For example, restricted space or restricted access to the plant may drive a selection toward smaller scroll or screw chillers. (Consideration of physical space necessary to maintain individual chillers also should be included in this review. The cumulative space for maintenance and multiple small chillers can exceed the corresponding footprint of a single large unit offering equivalent capacity.) Similarly, a noise restriction requiring quiet chiller operations would favor scroll or absorption chillers rather than screw or centrifugal chillers. Water use restrictions or sensitivities to seasonal water consumption and water conservation might preclude water-cooled chillers.
Also, the site might not be able to tolerate the occasional plume from cooling tower operation. In some instances, the associated infrastructure of the site, such as the electric distribution system, may not have sufficient capacity to serve the operation of a large number of electric chillers; possible alternates would include a steam turbine drive for centrifugal chiller or an absorption chiller. The inherent capital premium associated with these alternates is then evaluated relative to the capital cost of upgrading the electric distribution system and differential of operating expenses between chiller alternates. The magnitude of this operating differential is a function of cooling season length and annual ton-hours of chilled water production.
Consideration of alternate drives or chillers such as absorption chillers should not be limited to cases of limited infrastructure. Local electric rates, especially time-of-use rates, could provide a justification for these alternates. Chilled water production with steam turbine drives or absorption chillers could be price competitive with electric chillers during on-peak rate periods. Another correlation in support of these alternates may be cogeneration. Steam production necessary to support the turbine drive or absorption chiller may improve the capacity utilization of the cogeneration plant, further improving economic performance of the cogeneration investment.
A life cycle cost evaluation of the system selection can then truly incorporate the characteristics and requirement of the site as manifested by capital cost, capacity, configuration and annual chilled water production and production expense, including maintenance and replacement costs. However, this life cycle cost evaluation is essentially a static forecast based largely on load profiles (historic or forecast) and equipment performance data.
This evaluation should be updated periodically to confirm performance of the chilled-water system and components under actual field conditions. From this perspective, the chiller and chilled-water system design should include significant metering and instrumentation and storage of data that facilitate performance analysis of chilled water production. Metered data also can reveal significant variations from vendor performance sheets. These variations are often attributable to operating decisions or may be symptomatic of maintenance issues. Periodic evaluation of system and component performance can provide the diagnostic basis for modifying and maintaining system equipment and operations to realize expected efficiencies and financial performance.