Letting the Light Back In: Daylighting that Works

The best architects have always used light as a fine building material.  Now they are thinking in terms of energy, and since most of our kilowatt-hours came from the sun, light is more material – more real – than ever.  Not only does light drape the world so we can see it, it now acts as an indicator of human impact, and our satellites can map us by our tracks of light across the globe.


It’s a little ironic that there is a special term for non-artificial lighting (“Daylighting”), given that we have only had an electric alternative for the briefest flicker of our history.  The rewards of daylighting are tremendous, particularly in commercial buildings, and they come in terms of dollars, productivity and contentment.  Lighting can amount to 70% of electricity used in commercial buildings – making lighting efficiency the best way to reduce electricity use.  It’s no wonder that so much scholarly effort is put to designing daylighting strategies.  At WREF 2012, a Daylighting Technical Session illuminated great daylighting tips, and some of the contradictions between simulation and observation, design and practice.

Scanning Sky Simulator [Photo Credit: North Carolina State University

Scanning Sky Simulator Photo Credit: North Carolina State University

Presenters had some great rules of thumb.  Windows always have a positive effect on user experience, and should be incorporated for aesthetic reasons, but toplighting was the clear winner for pure function.  The team from the Queensland University of Technology compared light measurements with results of a user perception survey.  They found that toplighting has better consistency and angle on work space than sidelighting, and is most effectively positioned behind a workstation.  Photometric control systems enhance the user experience, and diffuse or redirected light is most pleasing.  The team from North Carolina State University found that the benefits of toplighting are maximized by integrating the skylight into the roof structure and splaying the sides of the wells.  Users need consistency in lighting, across the space and during the course of the day, as eyes tire of constantly adjusting.

The real trick to designing great daylighting is a problem as old as modeling (computer or physical):  the dicrepancy between simulation and performance.  The presenters each found a way around these obstacles.  Another team from North Carolina State University found that computer modeling is inadequate to predict the behavior of the sun and clouds, and of surfaces in the building.  So they stuck with physical models, replicated a 60-degree sector of sky in the lab and correlated its predictions with those of a scanner on the roof.  The Australian team found that even standard measurements in the field often fail to reflect user perception.  Of these measures, CIE Glare Index (CGI) agrees best with reported discomfort in open plan skylit offices and can be used to aid design.

Even when the daylighting solution is mocked up in full scale to the satisfaction of designers, the users may not take advantage of this free lighting.  The University of Oregon team voiced the designers’ frustration: “The daylight is great, but the lights are on!”  As with other best-laid architectural plans, occupant behavior, not design, drives energy use.  The Australian team found that lights were often left on during the day just to satisfy the light consistency needs of the most demanding user of the space.  Moveable dampers for daylight control were rarely used, and were left open or shut.  Occupants need comfort and consistency in lighting, but occupants vary and few take advantage of controls beyond the lightswitch, so the default state of the system settles at “lights on.”

Office Toplighting [Photo Credit: North Carolina State University

Office Toplighting [Photo Credit: North Carolina State University

Successful daylighting designs must have controls that empower multiple user types and reset themselves to “lights off” so that daylight is the default.  The system must be fast-operating – or at least fast enough to beat the user to the lightswitch.   Simulation is a great way to test options, but there is no substitute for reality and for listening to the actual users of the space.  The faster we get back to these basics, the faster we will return to lighting our world the old-fashioned way – for free.

Share on FacebookShare on LinkedInTweet about this on TwitterDigg thisEmail this to someone

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>