Case study: Office building meets energy requirements

An office building client in the Midwest had very strict energy standards. This demonstrates how the design team managed the process.
By Patrick Dempsey, PE, LEED AP BD+C, BEMP; CannonDesign, Chicago June 22, 2018

Figure 3: Though rudimentary, most energy-modeling software packages generate simple renderings to allow the team to evaluate whether most of the geometry being designed is being captured. Courtesy: CannonDesignA confidential client in Indianapolis had strict heating and cooling system-selection requirements for "Office Building I." The structure is a 200,000-sq-ft, 3-story office building for a client with very stringent performance standards for both design quality and energy. Some of the initial project parameters included:

  • No major mechanical equipment could be located on the roof.
  • The campus had a central plant providing chilled water containing glycol 24/7.
  • No basement was permitted due to water table issues.
  • The program efficiency (usable program space for the project) exceeded 80% (where 70% is typical).
  • There was a hard cap on glazing, specifically 60% on all elevations; exceeding the glazing for some elevations while reducing it in others was permitted only by demonstrating exceptional energy performance elsewhere. Energy performance was benchmarked against a baseline building that met the client’s design standard.
  • All design decisions had to be evaluated for first cost, lifecycle cost, energy-use intensity (EUI), and maintenance cost throughout design. Local code compliance was an afterthought because EUI compared with the client’s peer buildings was the goal.
  • Each major design issuance had an energy-model report that summarized all energy-performance calculations to date with an updated EUI target.
  • Energy performance would be evaluated by comparing the EUI target with peer buildings across the world in the client portfolio, in addition to the energy-modeled baseline building.
  • The client had a prohibition on any refrigerants that either deplete the ozone layer or add to global warming potential.
  • The client had strict standards regarding duct sizing, resulting in medium-pressure ductwork being sized for 1,300 fpm or less. For large supply mains, 2,000 fpm or less is more typical.

Cooling generation was not a consideration—the campus chilled-water system was going to be used. Due to the size of the building, a central boiler plant was selected. A condensing boiler plant was selected because the office building required no steam for process needs, and the water-temperature requirements were not very high.

The HVAC design decisions were affected in conceptual design by the program-efficiency requirement and the requirement for no rooftop equipment and no basement. The air handling systems would have to occupy usable space, pushing the program efficiency down. When considering the other requirements including conveyance, stairs, shafts, and general utility space, the mechanical systems for the building had to fit in approximately 3,000 sq ft or less-effectively 1.5% of the gross square footage of the building. For a typical project with no rooftop equipment or basement equipment, the mechanical space would occupy 5% to 7.5% of the gross square footage.

Figure 4: An early rendering of the office building provided context for what was being evaluated for energy performance in the initial design stages. Initial models are often built off of simple early concepts. Courtesy: CannonDesignBecause of the strict space requirements, a dedicated outdoor-air system (DOAS) was selected, reducing the central air handling systems feeding the building to roughly 54,000 cfm, or 0.27 cfm/sq ft. Typical office buildings using a standard air-distribution system would typically supply 0.7 cfm/sq ft or more. The reduced airflow made it easier to run ductwork above the ceilings with the 1,300-fpm air-velocity cap.

Once the air handling units (AHUs) were laid out, decisions had to be made regarding how heating and cooling would be distributed throughout the spaces via the chilled-water and hot-water loops. The decision came down to a discussion between fan-coil units and chilled beams. The energy analysis showed slightly better cooling performance and measurably better fan performance for the chilled beams versus the standard fan-coil units. The chilled beams had a slightly higher first cost; they were selected due to improved energy and acoustic performance.

Once the decision had been made regarding chilled beams, a comparison was made between active chilled beams connected to the DOAS so the ventilation air could help increase the capacity of the beams versus a "passive" chilled-beam system with no ductwork connected. Due to the high glazing ratios in some areas, nearly 70% on some elevations, the active-beam solution was selected to decrease the number of beams and the first cost of the system.

To reduce the airflow to the beams when not needed, the supply ducts for the beams were coordinated together with variable air volume (VAV) boxes that would reduce airflow when cooling/heating was not needed and ventilation quantities were satisfied. Ventilation could be reduced by using carbon dioxide sensors in tandem with space-temperature sensors.

The perimeter chilled beams were tied into both the heating-water and chilled-water systems, while interior beams were tied into only the cooling system. Because the campus plant was chilled glycol, and the client did not want glycol running in the building, a heat exchanger was provided between the campus system and the building hydronic system. The chilled-beam loop, which runs at a higher temperature (58°F) to avoid condensing, was tied into the return leg of the building’s chilled-water loop with a separate pump and three-way valve control to prevent low temperatures in the loop.

Other energy efficiency measures included installing the two DOAS AHUs on opposite ends of the building to reduce the distance between duct runs, essentially splitting the building in half. Each AHU included energy-recovery wheels to overcome the 100% outside air supplied by the DOAS, mandated by energy code. Local heating and cooling fan-coil units were included in support areas, such as electrical rooms, and in areas with exceptional ceiling heights, like the lobby and open stairwells. All pumping systems are provided with variable-speed drives, and the building included an aggressive system for maintaining pressurization to allow the outside air to be as low as possible when temperature requirements and ventilation needs were met.

The building was functionally occupied in 2017, and initial utility bills have the building operating with a targeted (partial-year) annual energy cost of $123,701, as compared with a modeled cost of $110,350. The projected annual energy cost of $0.61/sq ft is extremely low when compared with typical peer benchmarks. Using Energy Star’s Portfolio Manager tool, the median (existing) property of this type would have an annual energy cost of $450,126. A building with an Energy Star score of 90 (on a scale of 0 to 100) would have an annual energy cost of $249,452, or $1.24/sq ft.

Once a full year’s data has been collected, the design team will collaborate with the client and determine what changes can be made to performance, if any. The purpose of the design standards from the client was to provide a process for evaluating the energy performance relative to its own high-performance standards, which were stricter than the criteria for a typical building pursuing an Energy Star rating. A code-compliant building would typically have an Energy Star score in the 80 to 90 range, but it would not achieve nearly the energy savings or performance that this client required, even though it has a code-compliant design and is energy-efficient from the perspective of existing metrics.


Patrick Dempsey is an associate vice president at CannonDesign. He is a 2014 40 Under 40 award winner.