The Weidt Group conducted a study of daylight harvesting projects to find out if daylight harvesting projects are living up to their expectations. The team—consisting of Eijadi, Prasad Vaidya, Tom McDougall, Jason Steinbock and Jim Douglas—reviewed dozens of completed projects, most of which were sidelit using windows, and separated the success stories—some of them operating for 25 years—from projects they considered to be failures.
They eventually focused on eight that represented different components of the success/failure spectrum. In some cases, the project was largely successful except for a single element which caused it to fail—a case of daylight harvesting systems being only as strong as their weakest component. (Like a roof, says Eijadi, success can be defeated by a series of small compromises or a single catastrophic failure.)
For the eight projects, The Weidt Group team reviewed the design intent, met with the project team, and made additional site visits. The results were presented at ACEEE 2004 and other conferences.
The team concluded:
* Savings from automatic daylighting control systems are often not realized fully when a building is turned over to users.
* Daylighting performance needs attention and evaluation from multiple design disciplines during the design development and construction process.
* Users are not educated about the installed control systems; when something doesn’t work, users often disable the system instead of getting it fixed.
“If energy efficiency through daylighting controls is to proliferate as a strategy, its success rate needs to be improved,” says Eijadi. “Though there are successes, the intention of our research is to throw light on the weak areas so that future research and development on improving the process can be more focused.”
He cites common examples of why daylight harvesting project fail:
* Lack of coordination or understanding between the design disciplines concerning the daylighting control system.
* Improper location of daylighting controls.
* Inadequate specification of the controls systems, component parameters and sequence of operations.
* Shop drawings made by contractors that detail the system are not checked, or the lighting designer does not know what to check.
* Field changes to tune a system are not documented and taken back to the designer to complete the feedback loop.
These problems result in common failure modes:
* under-dimming, which results in less than expected energy savings
* over-dimming, which results in user irritation
* frequent cycling of dimming or switching, which results in user irritation
* lights left on at night, which results in less than expected energy savings
Below is a summary of four of the eight projects studied by The Weidt Group:
#1 – College Dining Hall: Large windows on three sides and high diffuse glazing on the south side were to provide daylighting. The south windows had a deep interior light shelf which functioned as a shading device. Photosensors were placed near the windows, linked to controllers by zone and a central control system.
However, non-dimmable lighting systems were connected to the dimming system. Control zones were not matched to daylight patterns. The circuits were not wired as they were shown in the construction drawings. The photosensor was not calibrated during construction, resulting in lack of detection of the wiring errors.
The facilities staff noticed the problem and called for calibration. The control system was reprogrammed as switching instead of dimming. In some cases, the control zones matched the daylight pattern and the problem was resolved easily. In other cases, the control zones did not match the daylight, which required rewiring.
#2 – College Classroom Building: High clerestory windows with interior light-shelves were to provide daylight, two photosensors located near the windows monitored the daylight contribution to the space, and connected controllers dimmed the lights for the perimeter and two next interior rows as separate control zones.
Unfortunately, the windows were designed smaller than those planned and tested in modeling. The first row of pendant lighting obstructed the daylight and made the interior relatively darker. The interior designer had not been a part of the daylighting evaluation and further reduced the daylighting by using dark colors. The photosensors, which were not calibrated, received the upward component of the direct/indirect light fixtures; as a result, the lights did not dim. The location of the photosensors relative to the window and light fixtures had not been checked in the shop drawings submitted by the contractor. Eventually, the facility staff noticed these problems and scheduled a calibration.
#3 – Office Building: The 300,000-sq.ft. office building featured a deep perimeter open office arrangement and ribbon glazing to a 10-ft. ceiling height. A series of dimmable T5HO direct/indirect light fixtures were installed in rows parallel to the windows, each row controlled as a separate zone with its own photosensor.
At first, everything seemed to be done perfectly. The controls were calibrated and responded well to changes in daylight levels. However, furnishings colors were not selected to support the daylighting conditions; dark furnishings were installed, resulting in a light level of 25-40 footcandles on work surfaces when the daylight harvesting system was active. The occupants, accustomed to a brighter work environment that lacked daylight, generated so many complaints that the system was deactivated.
The Weidt Group also felt that the sensors were calibrated too aggressively considering occupant preferences. Most significantly, the users were not told about the control system and its benefits, so there was no buy-in. Because there was no problem-reporting process for the controls, the operations staff lost confidence in the system.
#4 – General Merchandise Retail: The sales floor is daylighted using uniformly distributed diffuse horizontal skylights covering 2% of the floor area. Shelving heights range from 8-12 ft. in a 17-ft.-high lay-in ceiling space. The daylight harvesting system uses one closed-loop photosensor mounted on the ceiling, looking downward. The sensor provides continuous input to a central controller that separately dims three control zones, established according to the daylight pattern.
The control system was calibrated and short-term monitoring demonstrated that the system was performing as expected, reducing lighting energy use by 35%. However, a year later, the store manager felt the system was over-dimming, making the store feel too dark. He ordered the disabling of the photosensor.
To solve the problem, further daylight monitoring was used to re-calibrate the system and the level of dimming in two of the control zones was reduced. However, the store manager nixed the plan, believing that the control system might hurt sales.
“Most of the process went well,” says Eijadi. “This case study illustrates the potential long-term persistence problems daylighting systems can have when users or operators change and new people are unfamiliar with or unaware of the system. Accurately calibrating the system in the beginning is very important, but continued user and operator education is also essential. Unless the occupants and operators request it specifically, calibration may need to be done conservatively. This may result in a reduction of energy savings, but an aggressively calibrated system may very well be deactivated in the future.”
Ensuring Success: Prasad Vaidya, International Assoc. AIA, The Weidt Group, says: “Lighting designers should develop a controls narrative that defines the behavior expected of the daylight harvesting system; a good control narrative can help the user and building operator understand the system, and it helps the controls contractor calibrate and test it adequately.”
Eijadi and Vaidya advise control system designers to follow these steps to help ensure a successful daylight harvesting project:
1. Conduct a daylight simulation and use these plans when designing the lighting system and its controls.
2. Prepare plans that document daylight zones and establish independent control zones that work optimally with these patterns.
3. Locate the photosensor on the reflected ceiling plans and interior elevations.
4. Identify light fixtures that are controlled by individual sensors or controllers.
5. Write a daylighting controls narrative.
6. Require the contractor to submit shop drawings based on design documents and control narrative for your review.
7. Include the requirement for calibration of controls in the specifications, and require calibration logs to be submitted by the contractor.
8. Provision building operator training by the controls manufacturer.
Eijadi points to the Iowa Association of Municipal Utilities Office and Training Headquarters in Ankeny, IA as a classic example of daylight harvesting done right and an ongoing demonstration of how green design is good business.
“The building is exemplary in its resource efficiency,” he explains. “Designed by RDG Planning and Design and built within a modest budget, its actual metered energy consumption is 65 percent less than a conventional design. Building occupants, who enjoy multiple views of the landscape and sky from any point inside the building, report being extremely content.”
Early in the project, the owner cited daylighting as an application need, which became a primary driver for the project requiring integration with building sitting, orientation and massing/shape; interior design; the lighting system and controls; and HVAC distribution. The project team conducted an integrated evaluation of energy and building system strategies, including DOE-2.1e modeling, incremental construction cost estimating for alternatives, and incorporation of utility financial incentives.
The result was selection of a number of cost-effective strategies, including photosensors and dimming ballasts, occupancy sensors, time-clocks for outdoor lighting scheduling, windows carefully tuned by orientation, overhangs and seasonal light baffles, wood-framed operable windows and others.
Daylight harvesting need not be complicated—it just needs attention. What led to success in this project, says Eijadi, was a commitment to daylighting and proper coordination between disciplines. “The typical contractual divisions of authority, responsibility and compensation provide all parties with reasonable deniability and blamelessness when almost any aspect of a project fails,” he says. “The only way to have successful daylight harvesting is to make the whole team responsible for the outcome. Anyone can drop the ball or pick it up and run with it.”