The Big Chill

By Barbara Horwitz-Bennett, Contributing Editor July 1, 2005

Chiller experts predict that modular central chiller plants will continue to gain market share over field-built installations. The technology offers definite advantages—but MCCP isn’t always right for the job.

CSE: How novel is the concept of modular chillers and what is your assessment of their popularity?

LANDRY: The concept of packaging chiller plants is not new. In the 1980s, several companies made packaged plants, but the concept wasn’t yet commercialized on a large scale.

McAULIFFE: Actually, the modular central chiller plant (MCCP) approach grew out of the electric utility industry where MCCPs have been applied for more than a decade, usually to boost power plant efficiency via gas-turbine-inlet air-cooling. Success in this industry has encouraged fabricators to begin installing the modules in projects that require conventional air-conditioning and process-cooling duty.

CSE: What’s the adoption rate on the commercial side?

McAULIFFE: At this point, a very small percentage of conventional projects are going modular, but interest is growing as engineers and owners recognize the potential advantages. For example, in the Middle East, more than 250,000 tons of refrigeration via MCCP were specified this year. Many of these plants were originally designed as conventional stick-built, but designers eventually chose MCCP for reduced cost and time to completion.

PETERSON: Some of the popularity is due to the great part-load efficiency of these chillers. Since most chiller plants will spend the majority of the year operating at part load, this can sometimes provide greater efficiency.

CSE: What are the advantages—and disadvantages—of MCCP?

McAULIFFE: First, some background: The evolution of the MCCP is similar to the evolution of the chiller itself. Some may remember the days when large-tonnage chillers were engineered and erected in the field with condensers, heat exchangers and compressors being individually specified, delivered, installed and piped together. Today, the vast majority of chillers are packaged by the original manufacturer with the main components integrated, tested and shipped as a unit.

That same thinking is now being applied to the entire chiller plant. Specialized fabricators pre-engineer all the central plant equipment including chillers, pumps, cooling towers, starters and controls. These manufacturers package the equipment in an enclosed skid that is pre-piped, pre-wired, tested and shipped. Once the module is placed on-site, only water and power connections need to be made.

CSE: How does that help the consulting engineer?

McAULIFFE: A big benefit for consulting engineers is that total plant design time can be reduced, permitting more time and attention to interior space conditioning, ventilation, IAQ and overall system optimization.

PETERSON: It also allows a smaller footprint and lower cost and serviceability while running. There is also more redundancy, with overall efficiency the same as traditional plants.

As for the disadvantages, these mainly concern aesthetics and accessibility—and the fact that they show their age after 10 years.

GEISTER: Other benefits are the efficiency of single-source responsibility and quality control of ISO-9001 factory conditions. Also, constructing the plant off-site reduces the overall construction time and minimizes on-site conflicts in the case of retrofit applications.

On the flip side, designers can experience difficulty integrating into existing facilities or with the replacement of chillers.

LANDRY: Focusing specifically on the cost advantages, MCCP, via global sourcing of standardized components, as well as repetitive engineering and fabrication, is less costly, as it lowers manufacturing and assembly costs when compared to a traditional field-erected chiller plant.

With regard to efficiency, a traditional chiller plant is typically built at 0.8 to 0.9 kW/ton or higher, including the chiller, pumps, cooling tower and auxiliaries; a modular system can be optimized at 0.7 to 0.8 kW/ton. Essentially, the chiller plant components of an MCCP are optimized for plant efficiency rather than individual operation.

There are also the savings from reduced delivery, installation and commissioning time, not to mention that less time is spent on project management, compatibility and logistical issues.

There are, however, cost-related disadvantages. For example, the packaged system may actually cost more in some overseas installations where field labor costs are extremely low. For instance, in China, factory labor costs are often higher than field labor costs, although quality control advantages should still be considered.

Modular chillers also offer less flexibility in component selections. Because standardization and pre-engineering are two of the keys to lower costs in packaged plants, if one requires custom components, additional engineering costs reduce the savings.

CSE: Given all the pros and cons for MCCP, what are some ideal applications for this equipment?

PETERSON: My list would include the following: inlet cooling systems for turbines; retrofit buildings with limited access; noise-sensitive areas using turbocore or scroll chillers; cases where there is a limited amount of refrigerant per circuit, but that don’t require mechanical room ventilation if all requirements are satisfied; or an easily expandable plant.

LANDRY: I would add schools, universities, campuses and hospitals to the list. Also, MCCP is ideal for new construction on commercial and industrial applications; national/global chains where standard designs are needed in multiple locations; retrofit of existing or lower-efficiency air-cooled chiller plants; areas where skilled labor isn’t available, is overloaded or must work in high security or restricted access; and jobs where downtime is critical and bringing a system online quickly is crucial.

GEISTER: The market is also expanding to casinos, hospitality facilities and large district-cooling facilities. However, individual job requirements seem to drive interest more than the type of application.

Some retrofit applications welcome the idea of modular plants because conventional mechanical plants are built within a building, including the cooling tower, which has to be mounted outside. Packaged plants eliminate the central plant from the interior of the building altogether.

CSE: With manufacturers involved in the assembly of these modular systems, aren’t engineers much more reliant on them when specifying these systems?

PETERSON: Yes, because many times the project may be a custom configuration. Unfortunately, we have found that the product application support may be in a different time zone or country, or the local rep cannot lay out the equipment. All this can create a bottleneck delaying the normal design process.

LANDRY: I’d say yes and no. Yes, manufacturers have to be more involved to clearly define what they are offering—pre-selected components and configurations. They will also be more involved in reviewing the specifications.

On the other hand, a lot of the work has been done for the engineer, and detailed specifications have already been provided. Since this is a change for the general market, the engineer may spend more time initially because of the newness factor. But once a higher level of familiarity is reached, there will be less time involved. In addition, the engineer will initially need to rewrite the mechanical specifications to allow for competitive bids.

McAULIFFE: In many respects, it can make specification a simpler process for the consulting engineer, with significant responsibility—for delivery, testing and performance—consolidated with the fabricator.

Some of the fabricators will even guarantee the overall MCCP efficiency, tonnage and pressure drop, providing the consulting engineer with additional assurance of delivering specified performance to the owner.

But it should be pointed out that when engineers and building owners decide to go with a modular system, they don’t necessarily limit themselves to one manufacturer. This equipment can come from a variety of original equipment manufacturers. Consequently, specifiers and their end users still have choices.

GEISTER: With a modular plant, the engineer can write more of a performance specification, highlighting specific quality features desired. As a result, modular chiller manufacturers seem to prefer to act as an extension of the M/E/P firm. Since most projects involve many other types of mechanical equipment other than just the chiller plant, by utilizing the pre-engineered approach on as many products as possible the engineer can maximize time spent evaluating total building performance.

CSE: If an end user has specific customization needs, would it still be to his or her benefit to use such a pre-fabricated kind of product?

McAULIFFE: Customization is certainly possible. For example, vertical in-line pumps lend themselves quite well to the modular concept because of their small footprint. But if an engineer requests a traditional base-mounted pump, the module can be customized to include it.

However, it is important to note that customization can impact both the cost and the delivery schedule, and in many cases, it may not be necessary. In general, components have been carefully selected, based on the experiences of fabricators’ R&D teams.

LANDRY: Within the modular concept there will be significant options available and some customization will always be allowed. The degree of customization isn’t as great as with traditional equipment, but the benefits of lower cost and shorter commissioning time make it worthwhile.

For each specific customization the engineer can weigh the cost vs. benefit of a change. For example, changing the outer enclosure material may be easy and have minimal engineering costs. For a central plant, customization is generally restricted to the controls, starters and water connections. Thus, 80% is not customizable.


W. Ryan Geister , LaCrosse Water Chillers Product Manager, Field Sales Support Trane, La Crosse, Wis.

Christopher M. Landry Director Packaged HVAC Plants

AAF International, in cooperation with McQuay International, Louisville, Ky.

Bill McAuliffe , Strategic Account Manager, York International, Houston

Kent Peterson , P.E. Principal and Chief Engineer, P2S Engineering, Long Beach, Calif.

First Cost: Modular vs. Conventional

Most experts would agree that when all factors are considered, significant savings can be realized with modular chiller plants. “A typical pre-engineered package chiller plant costs approximately 15% less than field-built central plants,” says W. Ryan Geister, product manager with Trane, LaCrosse, Wis.

Others claim even better savings. “We also find that first cost is typically about 15% lower with an MCCP,” says Bill McAuliffe, a manager with York International, Houston. “However, we have seen it go as much as 30% lower, not including soft costs such as time saved by the engineer and contractor.”

Consulting engineer Kent Peterson, P.E., P2S, Long Beach, Calif., is even more optimistic: “I would estimate even greater savings—30% to 40%—depending on the exterior building requirements, serviceability, longevity and ease of replacement.”

But there are cases when field-built plants make economic sense. “In some [international] locations, field labor cost may be lower than factory labor,” says Christopher Landry, director of packaged HVAC plants, AAF Int’l., Louisville, Ky. “In this situation, the overall system cost may actually come out lower by using traditional methods.”

Ten Years After: The Future of Modular

So what do chiller experts have to say about the future of the modular central chiller plant concept? “In five years we won’t have to explain the term modular central chiller plant anymore,” says Bill McAuliffe, York International, Houston. “It will be well-known. In 10 years, the application will be commonplace.”

McAuliffe also predicts that we will see further enhancements to MCCPs, including architecturally enhanced enclosures and hybrid systems. These hybrids will deliver the chiller plant on modular skids, but the enclosure around the plant will be a tilt-up type of construction to blend into surrounding structures.

“I predict a significant increase over the next 10 years, potentially capturing upwards of 25% of the market,” says Ryan Geister with Trane, LaCrosse, Wis., “especially if additional manufacturers enter the markets and general acceptance of this practice increases.”

However, Kent Peterson, P.E., principal with P2S Engineering, Long Beach, Calif., is a bit less optimistic about the future of MCCP. “It all depends on the client and visibility of the project,” says Peterson. “First cost is not the primary concern for most chiller plants. Most owners want a chiller plant that can operate efficiently throughout the year for 20 to 30 years.”