Metal Architecture Home
Columns

Generative Design with Daylight Analysis

Constructive Insights1 Dec18 Ma
Circadian Curtainwall. Image courtesy of HOK

The evidence quantifying both the real and perceived value of natural light continues to mount. Recently published research by Dr. Alan Hedge at Cornell University identified some specific physiological wellness benefits of natural light in work environments including reductions in the occurrence of headaches, eyestrain, and drowsiness of 63 percent, 51 percent, and 56 percent respectively. People also instinctively understand these benefits—as reported recently in the Harvard Business Review, natural light was chosen as the number one office perk! It is preferred over fitness centers, on-site childcare, nap pods and other popular amenities.

As we know, there’s such a thing as too much of a good thing. An all too common design response to the demand for natural light has been buildings with floor-to-ceiling, wall-to-wall glass. This provides plentiful daylight and expansive views, but also tremendous heat gain/loss, accompanied by thermal discomfort and excessive glare. To provide effective daylight in buildings, it is important to consider contextual factors and potential strategies as the design is conceived, including:

  • Building massing and optimal floor plate depth, considering the solar access available given surrounding buildings and vegetation.
  • Planning of spaces based in daylight requirements (e.g., which spaces need more daylight or a particular quality of light, and at which hours of the day).
  • Façade design, optimizing the window-to-wall ratio and design of external shading to balance daylight entry and protection from direct sun to support thermal and visual comfort, and energy performance.

Experience and intuition can go a long way to informing these strategies, but cannot anticipate every potential daylight and glare issue presented by complex designs and surrounding contexts. Daylight metrics provide an objective way to measure the anticipated daylight performance of a building design. LEED v4 and WELL both include requirements to increase spatial daylight autonomy (sDA, a measure of adequate daylight illumination without electric light) while controlling annual daylight exposure (ASE, an indicator of excessive daylight and glare).

Circadian Curtainwall Daylight Analysis. Image courtesy of WSP

It is common to use daylight modeling to evaluate how a design performs against these metrics, but this often occurs after the design is well defined. LEED v4 and WELL metrics can be difficult to achieve without a conscious effort to ensure good daylight design. By the time an evaluative daylight model has been completed, major design decisions may be locked in and it can be difficult to make changes if daylighting is found to be suboptimal. Evaluative models can benchmark the design against daylight metrics, but come too late to determine the most influential performance factors or the best solutions to meet goals.

A more effective approach involves the proactive use of daylight modelling as an integral part of the design process. This will help guide designers toward high performing daylight solutions that optimize visual and thermal comfort and energy use. Computational design, the application of computer scripting applied to design processes, can be utilized to easily and efficiently explore multiple design alternatives and find the solution with the best performance. Custom algorithms can be written to solve complex project-specific problems related to building geometry and materiality, varying sun position, and dynamic shading systems.

Analysis can be integrated in early design phases, evaluating different massing options for daylight potential and revealing how climate (sun angles, sky conditions) and context affect daylight access and inform the optimal floor plate depths and façade parameters needed to respond to those conditions.

Circadian Curtainwall Daylight Analysis. Image courtesy of WSP

Parametric studies can quickly generate numerous façade configuration options, graphing the impact of different window-to-wall ratios and exterior shading configurations on the LEED-related metrics of sDA and ASE, and guiding the selection of optimal solutions. Analysis tools can also be used to produce generative shading studies. Rather than designing exterior shading devices and then testing them for effectiveness, a generative study will ray trace all annual solar vectors at each opening to generate the optimal solar shading configuration that would block direct sun while admitting indirect light.

My colleague and co-author, Elliot Glassman, recently put these computation analysis tools to the test in support of HOK’s development of the Circadian Curtainwall, a novel concept where the shape of the building was conceived to maximize daylighting, views, and visual comfort while reducing energy use. A typical rectilinear building always has one façade facing the sun, meaning that for a good portion of the day, all the shades will be down along that façade to protect from direct sun, restricting daylight and views. The curved lobes that make up each segment of the Circadian Curtainwall design would always have one side facing the sun and the other facing away. Each lobe would have a bifurcated set of automated blinds in a closed cavity (between glazing), with one deployed to protect from direct sun, while the other adjacent blind remains open, preserving daylight and views for occupants.

Circadian Curtainwall Daylight Analysis. Image courtesy of WSP

Computational daylight modelling of a typical floor proved out the concept, programming 122 individual annual shading schedules based on the changing solar exposure. The additional glass layers of the closed cavity façade improve the thermal performance of the assembly, but lower the visual light transmittance. Yet, once the dynamic shade operation was incorporated into annual daylight studies, the results revealed better performance than a rectilinear scheme with a more transparent façade. The added thermal performance and lighting energy savings contribute to an estimated 16 percent energy savings, demonstrating that with good design there does not need to be a tradeoff between daylight and energy performance.

While evaluative daylight analysis can validate if daylight and visual comfort criteria have been met, it doesn’t typically inform design decisions when the building form and fenestration are still fluid. The powerful analytical tools now available to designers are best used for proactive daylight analysis and generative design studies, to guide design decisions toward solutions that achieve optimal daylight, visual comfort and views. Whether pursuing bold new forms or improving on more conventional structures, designing around the fundamental human need and desire for natural light and connections to the outdoors is essential for high-performance buildings.


Alan Scott, FAIA, LEED Fellow, LEED AP BD+C, O+M, WELL AP, CEM, is an architect with over 30 years of experience in sustainable building design. He is a senior associate with WSP in Portland, Ore. Elliot Glassman, AIA, NCARB, LEED BD+C, CPHD, is an architect who specializes in computational design and analysis for daylight and energy in passive buildings. He is an associate with WSP in New York City. To learn more, visit www.wsp.com/en-US/services/built-ecology and follow Scott on Twitter @alanscott_faia.