Demonstration and Evaluation of PostPost - top Photovoltaic LED Luminaires Nicholas Skinner, MS Daniel Frering, LC Jennifer Brons, MS, LC Lighting Research Center R Rensselaer l Polytechnic P l t h i Institute I tit t © 2010 Rensselaer Polytechnic Institute All rights reserved Outdoor Lighting Outdoo g t g Pros › Safety, security, visibility › Economic E i development, d l t community pride, crime prevention ti Cons › Energy use › Maintenance costs › Light pollution © 2010 Rensselaer Polytechnic Institute All rights reserved New York State DOT Project oject Background ac g ou d Sullivan S lli Renaissance: R i Outdoor lighting project goals › Help the communities in the county p nighttime g safetyy and improve accessibility, develop local economies, pride and build communityy p › Without increased energy use, environmental pollution, pollution and reduced views of the night sky © 2010 Rensselaer Polytechnic Institute All rights reserved So ut o Solution Photovoltaic (PV) Lighting System Design › Based on historic post-top fixture › Philips Hadco and SolarOne Solutions l › PV panels integrated like banners › LED lighting – 50,000 hour expected life › Excellent light control and distribution © 2010 Rensselaer Polytechnic Institute All rights reserved So ut o Solution © 2010 Rensselaer Polytechnic Institute All rights reserved Demonstration e o st at o Funding received from NYSERDA th through competitive solicitation Three sites selected for demonstration › Diversity of potential applications pp and environmental conditions › High g degree g of replicability p y › Installations completed in late 2007 and earlyy 2008 © 2010 Rensselaer Polytechnic Institute All rights reserved Photo courtesy: C Emery Village g of Woodridge, g NY Planned to replace the existing electrical wiring along Green Avenue plaza 12 lamp posts installed along a downtown street in the village © 2010 Rensselaer Polytechnic Institute All rights reserved Town o of o Liberty be ty – S Swan a Lake a e Park a Six lamp posts at Swan Lake Park Positioned along a boardwalk next to a lake © 2010 Rensselaer Polytechnic Institute All rights reserved Bethel Woods Center for the Arts Country road, minimum ambient light Six lamp posts Parking area for the 1969 Woodstock Festival mon ment monument Phillips HADCO © 2010 Rensselaer Polytechnic Institute All rights reserved Evaluation a uat o Conducted by the LRC, included: › › › › › › › Performance of the systems and components Energy savings Lighting conditions provided Lif cycle Life l costs t Hours of operation Maintenance Residents’ opinions of the systems Residents Photo courtesy: C Emery 10 © 2010 Rensselaer Polytechnic Institute All rights reserved Survey Su ey Results esu ts All sites › Showed similar positive trends › Bethel Woods site p less positive • Less frequent visitors • Unfamiliar with site • Uneven ground materials 11 © 2010 Rensselaer Polytechnic Institute All rights reserved Co Comments e ts 12 © 2010 Rensselaer Polytechnic Institute All rights reserved Performance Monitoring Battery-powered light monitoring devices installed at all sites in: › Summer June/ July 2008 › Fall/ Winter September 2008 to February 2009 13 © 2010 Rensselaer Polytechnic Institute All rights reserved Fall/ Winter Monitoring g Results All units turned on at about the same time Earlier on shortest days of the year Typical Nightly Start Time (F ll/Wi t ) (Fall/Winter) Nightly Start Time e (PM) 7:45 7:15 6:45 6:15 5:45 5:15 4:45 10/6 10/20 11/311/17 12/1 12/1512/291/12 1/26 Date 14 © 2010 Rensselaer Polytechnic Institute All rights reserved 2/9 Fall/ Winter Monitoring g Results All units operated at full light output for 4-6 hours then stepped down to dimmed state Typical Operating Pattern ((as Measured with Battery-Powered y Lightlogger) g gg ) R Relative Light L Output O Nightly on-time, highest output Controller reduces output slightly over 4-6 46 hours After 4-6 hours, light output drops to dim level stays low for the rest of the night Lightlogger sees some light during the daytime: Sunny Cloudy Nt Day Nt Day Nt Day Nt Day Nt Time (Day/Night Cycle) 15 © 2010 Rensselaer Polytechnic Institute All rights reserved Fall/ Winter Monitoring Results Some units programmed to return to full light output before dawn Typical Operating Pattern – Morning Program ( M (as Measured d with ith Battery-Powered B tt P d Lightlogger) Li htl ) Re elative Liight Output Some units are programmed to come back to full output at the end of the night Nt Day Nt Day Nt Day Nt Day Nt D Time (Day/Night Cycle) 16 © 2010 Rensselaer Polytechnic Institute All rights reserved Occupancy Occupa cy Se Sensor so Activity ct ty ` Two in Bethel Woods parking lot Dim to low level when no occupancy is sensed Results: Typical Night/Day Cycle (October 2008) Typical Night/Day Cycle (October 2008) Occupancy Sensor Controlled Luminaire › Frequent switching at night › May not have been responding to actual human occupancy › False-offs Frequent switching Night 17 © 2010 Rensselaer Polytechnic Institute All rights reserved Frequent switching Day Night Fall/ a / Winter te Monitoring o to g Results esu ts For about a week in N November b and d December: D b › Luminaires turned off prematurely, t l or didn’t did ’t turn t on at all › Possible low voltage disconnect Most of fall/ winter, units operated t d as expected t d Some units may be staying on after sunrise (terrain?, shading?) 18 © 2010 Rensselaer Polytechnic Institute All rights reserved Photometric Performance Woodridge – similar before and after › 1-17 17 lux horizontal › 1-16 lux vertical › Residents surveyed rated LED system as bright and safe as previous HPS system Swan Lake › 3-20 lux horizontal › 1-15 lux vertical Bethel Woods › 1-10 lux horizontal › 1-9 lux vertical 19 © 2010 Rensselaer Polytechnic Institute All rights reserved Installation sta at o / Maintenance a te a ce Co Comments e ts Easy to install Some issues… A new floodlight impacted operation (one location) Water ingress; housing redesigned Different wiring connectors suggested More robust PV mounting g bracket design implemented 20 © 2010 Rensselaer Polytechnic Institute All rights reserved Photo courtesy: C Emery Energy e gy Ca Calculations cu at o s Save S energy b by avoiding conventional utility power At Woodridge Woodridge, actual savings compared to p e io s system previous s stem Hypothetical savings, compared to no PV and a conventional luminaire 21 © 2010 Rensselaer Polytechnic Institute All rights reserved Life-Cycle e Cyc e Cost Estimates st ates ((10-yr) 0y) Over twice the cost of conventional system If located further from power, costs would b comparable be bl 22 © 2010 Rensselaer Polytechnic Institute All rights reserved Co c us o s Conclusions Community residents liked the PVpowered po e ed LED lighting g t g syste system › Positive feedback on appearance and function › Feel that it provides appropriate illumination Systems generally operated as intended during Summer, Fall, and Winter A PV-powered LED outdoor lighting system may present a viable alternative to grid-connected systems in situations where grid-power is either unavailable unavailable, or costly to bring to a site 23 © 2010 Rensselaer Polytechnic Institute All rights reserved Acknowledgments c o edg e ts The New York State Energy Research and Development Authorityy ((NYSERDA)) The Gerry Foundation Sullivan Renaissance SolarOne Solutions Philips p Hadco Lighting g g Village of Woodridge Town of Libertyy Bethel Woods Center for the Arts 24 © 2010 Rensselaer Polytechnic Institute All rights reserved Publications ub cat o s “Demonstration and evaluation of sustainable photovoltaic powered LED sustainable, street lighting” Solar 2009 Conference, Buffalo/ Niagara, N.Y., May 2009 Field Test DELTA: “Post-top Photovoltaic Pathway Luminaire”, October 2009 › In print › On-line “Streetlights g for Collector Roads”,, 2010 › On-line http://www.lrc.rpi.edu/programs/NLPIP/publications.asp http://www lrc rpi edu/programs/DELTA http://www.lrc.rpi.edu/programs/DELTA 25 © 2010 Rensselaer Polytechnic Institute All rights reserved