© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Part IPart ISSL Technology UpdateSSL Technology Update
Nadarajah Narendran, Ph.D.Lighting Research Center
Rensselaer Polytechnic InstituteTroy, NY 12180
Solid State Lighting WorkshopAlbany, NYMay 1, 2008
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
http://upload.wikimedia.org/wikipedia/commons/b/b4/Gluehlampe_01_KMJ.png
1809: Humphry Davyinvented the first electric light.
1879-1880: Thomas Edisondeveloped the first practical incandescent, electric light source. http://inventors.about.com/library/inventors/bledison.htm
Thomas Edison (1847-1931)
Electric lightingElectric lighting
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
LightingLighting
Over the past 100 years,incandescent and gas dischargetechnologies have provided many light fixtures for a variety of lighting applications.
www.sportsvenue-technology.com/contractors/flood/abacus/abacus2.htmlimages.google.com/images?q=hotel%20lighting&hl=en&lr=&ie=UTF-8&sa=N&tab=wi
10x lm
100x lm
10000x lm
1000x lm
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Solid-State Light Sources
Light Emitting Polymer (LEP) orOrganic Light Emitting Diode (OLED)
Light-Emitting Diode (LED)
UniversalUniversal--displaydisplay
Evolving new light sourcesEvolving new light sources
Cree XLamp® LED
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Industrial14%
Outdoor Stationary
8%
Commercial51%
Residential27%
Lighting22%
Other78%
.
Motivation for solidMotivation for solid--state lightingstate lighting
About 22% of the total energy use in the U.S.Demand for energy keeps increasing
Source: U.S. DOE Website
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
White LED promiseWhite LED promise
15 lm/W
90 lm/W
120 lm/W
150 lm/WEnergy savings
1000 hrs
20,000 hrs30,000 hrs
100,000 hrsLower maintenance cost
By 2012
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Banning of the light bulbBanning of the light bulb
United StatesDecember 2007 – Clean Energy Act of 2007 was
signed into law.
This legislation effectively banned (by January 2014) incandescent bulbs that produce 310 - 2600 lumens of light. Bulbs outside this range (roughly, light bulbs currently less than 40 Watts or more than 150 Watts) are exempt from the ban. Also exempt are several classes of specialty lights, including appliance lamps, "rough service" bulbs, 3-way, colored lamps, and plant lights.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Lighting market forecastLighting market forecastForecast from Optoelectronics Industry Development Association (OIDA)
Incandescent will be the worst affectedOpportunity for SSL systems
http://images.google.com/imgres?imgurl=http://wrtassoc.com/__oneclick_uploads/2007/11/ssl-forecast.jpg&imgrefurl=http://wrtassoc.com/&h=346&w=575&sz=31&hl=en&start=7&um=1&tbnid=ziWTcKTuWLmgAM:&tbnh=81&tbnw=134&
prev=/images%3Fq%3Devolution%2Bof%2Bhigh%2Bpower%2Blight%2Bemitting%2Bdiodes%26um%3D1%26hl%3Den%26sa%3DX
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Life after lightingLife after lightingLight bulb aquarium
http://www.itsnicethat.com/images/507.jpg
http://blog.makezine.com/archive/retro/2.html
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Junction(depletion region)
Light
P N-+
- --++
+
+ ---
-Electrons
--
LightLight--emitting diodeemitting diode
P material: Has a slight “deficiency” of electrons for molecular bonding when forming a crystal. N material: Has excess electrons left over from the crystal bonding process, which can move and carry current.Photons (light) are generated when the positive and negative charges recombine.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
+ + + + + + + + + + + + + + + ++ + + + + + + + + + + + + + + + + + + +
pn Junction at bias VF
- - - - - - - - - - - - - -- - - - - - - - - - - - - - - - -
VFTrap
LightEg
ElectronElectron--hole recombinationhole recombination
Not all recombinations result in lightCharges trapped in defects result in heatBandgap energy depends on the semiconductor materialColor of the light output depends on the bandgap energy
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
LED: StateLED: State--ofof--thethe--artart
LumiLeds Lighting
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Mixing different colored LEDs (red, green, and blue) in the right proportions produces white light.
Combining blue (or UV) LEDs with phosphors produces white light.
White light with LEDsWhite light with LEDs
White LED
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
380 430 480 530 580 630 680 730
Wavelength, nm
Rel
ativ
e ou
tput
RGB High
0.0
0.5
1.0
400 500 600 700 800Wavelength (nm )
Rel
ativ
e O
utpu
t
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
White LEDWhite LED
To achieve higher luminous efficacy with white LEDs, improvements are needed at several stages:
internal quantum efficiency extraction efficiency from the die phosphor-conversion efficiency extraction efficiency from the package
DieEpoxy
& phosphor
Reflector
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Light extraction from the chipLight extraction from the chip
In traditional LEDs, more than 70% of the light generated by an LED is trapped within the device.
Total Internal Reflections (TIR) Fresnel reflection limits light extraction
θc = sin-1 (n2/n1)For θ > θc TIR
n1 = 2.4 to 3.7
n2 = 1.5
Fresnel reflection
r = {(n1 - n2) / (n1 + n2)}2
n2 = 1.5
n1 = 2.4 to 3.7
TIR
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Methods for improving chip light extractionMethods for improving chip light extraction
RCLED
Alter surface finishto improve extraction efficiency
OSRAM Opto’s thin film device
LumiLeds’ (TIP) LED
http://web.mit.edu/cmse/www/IRG-I.nug02.htmlPhotonics Bandgap
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
EncapsulantsEncapsulants
Encapsulants can affectlight extraction (refractive index mismatch)life of the LED (yellowing, photodegradation)
The industry is still looking for new phosphors and down-conversion materials, like quantum dots, for LEDs.
Air = 1.0008Encapsulant = 1.4 ~ 1.6 Phosphor = 1.85 Die = 2.4 ~ 3.7
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
HighHigh--flux LEDsflux LEDs
For state-of-the-art LEDs, the extraction efficiency is in the range 40 to 50%.
Lumileds
http://seoulsemicon.co.kr/_homepage/home_eng/product/product.asp?topCODE=1&midCODE=25
Cree
Lamina
Seoul
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
UV pcUV pc--LEDLED
405-nm LED + RGB phosphors Potential for better color stability1-W and 4-W versions (~162 lm)
http://www.lumination.com/literature/VioDataSheetWEB8_7_07.pdf
70 & 85 CRI3000 K to 4100 K
Manufacturer’s data:
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
AC LEDAC LED
AC operation No need for drivers
Flux2W: 65 to 80 lm4W: 150 to 195 lm
http://seoulsemicon.co.kr/_homepage
Manufacturer’s data:
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Commercial White LEDs Commercial White LEDs –– (March 2008)(March 2008)
Warm White
Vio Vio
OstarOstar
Z-Power P4
Acriche
XRE(P3)
Luxeon Rebel
Luxeon ILuxeon K2
Rigel
Platinum DRAGON
Golden DRAGON
Titan
Moonstone
Luxeon K2
0
10
20
30
40
50
60
70
80
90
100
0 200 400 600 800 1000Luminous flux (lm)
Lum
inou
s ef
ficac
y (lm
/W)
Cool White
Moonstone
Diamond DRAGON
Z-Power P4
Acriche
XRE (Q5)
Luxeon Rebel
Luxeon ILuxeon K2
Luxeon III
Rigel
Platinum DRAGON
Golden Dragon
Ostar
Titan
0
10
20
30
40
50
60
70
80
90
100
0 200 400 600 800 1000Luminous flux (lm)
Lum
inou
s ef
ficac
y (lm
/W)
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
White LED Color IssuesWhite LED Color Issues
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
White LightWhite Light
General illumination:CCT: 2800 K to 5500 K
CIE x,y values close to the blackbody locus
Good color-rendering properties
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
CCT & CRICCT & CRI
New phosphors have enabled white LEDs with a variety of CCT and CRI values.
Now most commercial vendors are offering at least two CCTs (warm and cool white)Some of the new white LEDs have CRI values greater than 92
The industry is still looking for new phosphors and down-conversion materials, like quantum dots, for LEDs.
Nichia White LEDs
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
LEDLED--toto--LED color variationLED color variation
1931 CIE Chromaticity DiagramMacAdam EllipsesRepresent the loci of just-noticeable color difference
ANSI SpecificationCalls for a 4-step MacAdam ellipse for certain types of lamps
Wyszecki and Stiles, “Color Science,” 1982
Large color variations between similar light sources is an undesirable feature.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
LED color binsLED color bins
Most LED companies bin their LEDs Products are tested and sorted into performance bins, and color is one of them.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Life of LEDLife of LED
I = Io exp – α(t)
70%
80%
90%
100%
0 10000 20000 30000 40000Hours
Ligh
t Out
put
Thermocouple
R2 = 0.96
010000
2000030000
4000050000
35 40 45 50 55 60T-point Temperature (deg C)
Life
(hrs
)
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
LED lifeLED life
All LEDs are not created the sameLife varies significantly
High power white LED
70%
75%
80%
85%
90%
95%
100%
100 1000 10000 100000Time (hours)
Rel
ativ
e lig
ht o
utpu
t
F
C
A D
E B
1W white LEDs operated at 35 deg C, 350 mA
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Present industry trendPresent industry trendGrowing number of LEDs and LED fixtures
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Ballast (Driver)
Heat Sink
Optics (Lens)
LED Array
Optics (Diffuser)
Electrical Grid
Fixture Housing
Controls
Circuit Board
LED lighting systemLED lighting system
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Examples of commercial systemsExamples of commercial systems
http://www.colorkinetics.com/ls/rgb/colorburst6/
Lamina
6 Golden DRAGON™ LEDs connected in series
enLux LED R30
OptiLED's DESIGNER models
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
LED systemLED systemOptimized system
LEDElectricalOpticalMechanical (heat sink / housing)
Luxeon
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
http://www.display-optics.com/pdf/tech_papers_oct2002.pdf
OpticsOptics
Options for LEDs:Conventional refractiveConventional reflectiveDiffractiveMicrostructured refractive
Small source size of LEDs allows for more efficient optics
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
LED lighting systemLED lighting system
LED
Optical
Thermal
Electrical
System efficiency ~ 58%
Lighting Research Center
System efficacy ~ 55% of LED efficacy
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Lighting systems performance Lighting systems performance
0
50
100
150
200
250
2000 2004 2008 2012Year
Perf
orm
ance
(Lm
/W)
White LEDR&D Results
Liner Fluorescent Systems
CFL SystemsLED Systems Incand. Systems
(2007)
To the end user, system performance matters…not source performance
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
System reliabilitySystem reliability
Reduced replacement cost is one of the promises of LEDs
It is the final system performance that matters to the end user
http://members.shaw.ca/sagelighting
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Market spoilersMarket spoilers
Over-promised, under-delivered products
Not good for the LED lighting industry
Taipei
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Energy output of electric light sources
65 %34 %18 %Heat(Conduction- Convection)
15 %13 %Ballast
Low29 %72 %IR
20 %23 %10 %Light
LED(estimate)
Fluorescent*(T-8 F32 rare earth)
Incandescent*(100 W)
* IESNA Handbook – 9th Edition
Why is thermal management important?Why is thermal management important?
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Impact of heat on LEDsImpact of heat on LEDs
Heat affects LED performanceLight outputColorLife
1 Watt White LEDs
60%70%80%
90%100%110%
40 60 80 100 120Tj
Rel
ativ
e lig
ht o
utpu
t
LED A
LED B
LED C
R2 = 0.96
010000
2000030000
4000050000
35 40 45 50 55 60T-point Temperature (deg C)
Life
(hrs
)
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Time
Ligh
t O
utpu
t
100%
0%
Heat at the p-n junction increases the degradation rate
Individual LED vs. SystemIndividual LED vs. System
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Ceiling tiles (Poor thermal conductor)
Halogen / Incandescent Fixture
LED Fixture
Conducted heat
Radiant heat
Insulation material above the ceiling area
Application issuesApplication issues
Even a properly designed LED fixture can perform badly if installed incorrectly
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
ApplicationsApplications
To date, the most popular application for LED technology is colored lighting.Compared to traditional light sources, LEDs offer greater benefits for colored light applications.
Less energyLonger lifeMore control optionsShallow-profile fixtures
www.tirsys.com/resource-center/showcase/structures-bridges.htm
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
ApplicationsApplications
White light applications are just starting Not as popular as colored lightingEnergy savings only in certain niche applications
Two most-touted applications for white LEDs:
Downlights – interior applicationsStreet and parking lot lights – outdoor applications
Are LEDs ready for these applications?
stylmark.com
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
DownlightingDownlighting
IncandescentHalogen CFL LED
One of the most-touted applications for white LEDs is downlighting
Now, many technologies can cater to the same application.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Commercial LED downlightCommercial LED downlight
System efficacy = 15 lm/W
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Commercial CFL downlightCommercial CFL downlight
System efficacy = 33 lm/W
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Commercial LED downlightCommercial LED downlight
System efficacy = 54 lm/W
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Initial costInitial cost
At the present time, most LED lighting fixtures have a higher initial purchase cost than incandescent or fluorescent fixtures (approximately 3 to 6 times higher).
Downlights: Incandescent 75W ~$50CFL ~$90 to $140LEDs (PC) ~$300 to $500
(*LLF LED fixture under $100)
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Life cycle costLife cycle cost
In hospitality applications, the cost of using LED systems is approaching the cost of using traditional lighting systems.
The following cost estimates are based on the assumption that LED systems would last 50,000 hrs or longer.
Residential Application
$0$500
$1,000$1,500$2,000$2,500$3,000$3,500
Inc 1
Inc 2
Inc 3
CFL 1
CFL 2
LED 1
LED 2
LED 3
Tota
l Cos
t for
10y
rs
Hospitality Application
$0$1,000$2,000$3,000$4,000$5,000$6,000$7,000
Inc 1
Inc 2
Inc 3
CFL 1
CFL 2
LED 1
LED 2
LED 3
Tota
l Cos
t for
20y
rs
(LLF was not considered in this calculation)
$670$1400
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
SummarySummary
Rapid development of LED technologyAchieving 150 lm/W by 2012 appears feasibleImprovements needed
Internal quantum efficiency Extraction efficiency from the die Phosphor-conversion efficiency Extraction efficiency from the package
Achieving 100 lm/W system efficacy is challengingImprovements needed
OpticsThermal management componentDrivers and other control devices
Growing applications for white LEDs
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Part IIPart IITesting and Evaluating SSL SystemsTesting and Evaluating SSL Systems
Nadarajah Narendran, Ph.D.Lighting Research Center
Rensselaer Polytechnic InstituteTroy, NY 12180
Solid State Lighting WorkshopAlbany, NYMay 1, 2008
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Need for MetricsNeed for Metrics
Rapid development of LED technologyApplications community interested in using LEDsMany commercial products for general illumination
Some products have exaggerated claims Insufficient performance data availableLack of measurement standards is one reasonMeasurements made at standard conditions may not represent performance in applications
Failed applications can hurt the entire industryMany agencies are actively working on standards
Insufficient understanding of technology can lead to bad standards
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
ASSISTASSIST recommendsrecommends
LRC conducts research on LED systems to develop and disseminate information that is useful to users and standards-setting bodies
The project goal is to develop a series of publications
ASSIST recommendsRecommendations for testing and evaluating LED systems Application guides
• Recommendations for using LED light fixtures in applications
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Introduction: Introduction: ASSISTASSIST
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
History & backgroundHistory & background
ASSISTASSIST: Alliance for Solid-State Illumination Systems and Technologies Established: 2002Goal: To support the development and widespread application of LEDs for general illumination
Identify and reduce the major technical hurdles currently facing solid-state lighting
Activities: Industry collaboration, research, demonstration, and education
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
What is What is ASSIST recommendsASSIST recommends??
When standard definitions and metrics for LED technology are not available, ASSIST develops and publishes recommendations.
The recommendations are developed through research conducted on behalf of ASSIST by the LRC.
ASSIST recommends helps manufacturers present information to end users in a consistent manner.
ASSIST also publishes application guidelines to help end users select and apply LED technology successfully.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Industry activitiesIndustry activities
Standards-setting organizations such as NEMA, CIE, ANSI, …….and others have been working on certain standards.
ASSIST is conducting research to develop information that can be useful for metrics and setting standards.
ASSIST recommends
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Jan 2004 June 2004 Jan 2005 June 2005 Jan 2006
ASSISTASSIST
Extracted from March 30, 2004 ASSIST roundtable meeting— Las Vegas
Roadmap to standardsRoadmap to standards
IES NEMA ANSI
Industry Adopts
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
ASSIST recommendsASSISTASSIST recommendsrecommends
LED Life for General Lighting (Released in 2004)
Life definition: 70% lumen maintenance (50% for indicators or decoration)
Life reported by both LEDand fixture manufacturers
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Current statusCurrent status
Some of the LED manufacturers who already have collected data are in the process of compiling information per ASSIST recommends.
Several organizations have referenced ASSIST recommends: LED Life
NEMA DOEIES
Several additional ASSIST recommends published
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
LED applicationsLED applications
LED performances have been steadily improving.Two potential illumination applications in the near term are:
Under-cabinet lightingDirectional lighting
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
DocumentsDocuments
Three partsGeneral guide to applicationsGuide to selecting LED fixturesTesting and evaluation recommendations
AudienceGuides – Homeowners, general contractorsTesting and evaluation – Manufacturers, standards-setting bodies, state and federal agencies, public benefit program administrators, independent test labs
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Under-cabinet Lighting
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
UnderUnder--cabinet Lightingcabinet Lighting
Application efficacy = Total lumens on the taskTotal fixture power
18-inch
12-inch
Under-cabinet luminaire
Application efficacy in realistic conditions
φi = Ei . Ai
φv = ∑ Ev . Av ; φh = ∑ Eh . Ah
4” x 4”
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Test Data per Test Data per ASSISTASSIST recommendsrecommends
05
10152025303540
Hal. T5 A T5 B T8 C T2 D LED A LED B LED C LED D LED E LED F
App
licat
ion
Effic
acy
(lm/W
)
Hal. T5 A T5 B T8 C T2 D LED A LED B LED C LED D LED E LED FFixture length (in) 12 12 12 20 18 24 24 12 12 21 12
Horizontal flux (lm) 53 91 72 180 199 95 77 87 111 172 173Horizontal average (lux) 96 163 129 277 307 128 104 155 199 252 311
Horizontal uniformity (max:average) 4:1 2:1 2:1 2:1 2:1 2:1 4:1 2:1 2:1 2:1 2:1Verticalflux (lm) 23 107 97 256 286 64 21 65 69 199 109
Vertical average (lux) 49 230 210 473 528 103 34 140 149 350 235Vertical uniformity (max:average) 4:1 3:1 4:1 4:1 3:1 3:1 3:1 4:1 2.5:1 2:1 3:1
(Fixture+Driver) Input power (W) 18.1 8.2 6.9 13.8 14.7 13.5 8.0 8.8 7.7 10.78 7.6(Fixture+Driver) Voltage (V) 119.0 119.5 119.0 118.8 119.2 118.3 118.9 24.0 119.8 119.8 120.0(Fixture+Driver) Current (A) 0.16 0.11 0.11 0.20 0.22 0.12 0.13 0.37 0.07 0.199 0.06
Ambient temperature (C) 23 23 22.9 23 23 23 23 23 23 24.6 23Fixture operating temperature (C) 38.2 33.6 40.1 44.8 41.9 37.7 28.2 41.4 30.6 35.2 35.5
Application flux (lm/ft) 76 198 169 262 324 80 49 152 180 212 283Application Efficacy (lm/W) 4 23 23 30 33 11 12 17 23 34 37
Fixture light output (lm) 88 281 420 623 616 194 151 222 417 420Fixture Efficacy (lm/W) 5 33 57 42 42 14 18 29 38 56
CCT 2591 3044 3965 2813 3223 2943 2868 5887 3500 7542CRI 100 87 86 82 78 73 65 76 73 71
Driver input power (W) 18.4 8.6 7.4 14.7 14.6 13.9 8.2 7.8 10.9 7.6
Grid measure
Sphere measure
ASSIST Reccomends MethodASSIST Recommends Method
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
UnderUnder--cabinet Lightingcabinet Lighting
67%
81%74%
54%
85%0
102030405060
Halogen F8T5 - 1 F8T5 - 3 LED 1 LED 2
Effic
acy
(lm/W
)
Fixture Efficacy Application Efficacy
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
SummarySummary
Application efficacy is a more meaningful metric than light source efficacy
Near-field photometry is useful to determine the task lumensSome manufacturers already provide illuminance data on the task
Application efficacy = Total lumens on the taskTotal fixture power
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
ASSISTASSIST recommendsrecommends
Recommendations for Testing and Evaluating Recommendations for Testing and Evaluating Luminaires Used in Directional LightingLuminaires Used in Directional Lighting
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Directional lighting test methodDirectional lighting test method
ASSIST Recommends proposed three environmental conditions to test fixtures:
Open air: Here the light source and the driver have plenty of ventilation around them.
Semi-ventilated: Here the light source and the driver have limited ventilation around them.
Enclosed: Here the light source and the driver have almost no ventilation around them.
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Sphere photometrySphere photometry
Temperature, Ts, is measured while operating the fixture in the three environments.
Fixture is placed inside a heated enclosure which is place inside the integrating sphere.
Data gathered once the temperature, Ts, reaches application temperature.
Heater
Lamp
Driver
Heated enclosure
Feedbackcontrol
Ts
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Luminaire testing Luminaire testing
Several commercial LED fixtures are being tested in three environments (per ASSIST recommends)
Open airNon-ICIC
Short-term testingFlux and color
Long-term testingLumen depreciation and life (L70)Color shift
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Flux (lumens)Flux (lumens)
Well-designed luminaires maintain light output even in hotter environments.Poorly designed luminaires have more than 30% lower light output in IC-condition.
236
649583
263212
678
446
223
643
396
183
0100200300400500600700800
Fixture A26W
Fixture B26W
Fixture C12W
Fixture D30W
Flux
(lum
ens)
Open air Non-IC IC
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Efficacy (lm/W)Efficacy (lm/W)
Generally, system efficacy values are 30% to 50% lower than LED efficacy values. However, well-designed luminaires have achieved over 50 lm/W.
10
22
54
8917
7715
5457
0
10
20
30
40
50
60
70
Fixture A26W
Fixture B26W
Fixture C12W
Fixture D30W
Effic
acy
(lm/W
)
Open air Non-IC IC
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Board temperature ( Board temperature ( °°C)C)
With increasing Tj the life shortens Generally half the life for every 10°C increase
83 °C 87 °C
42 °C
80 °C95 °C
107 °C
50 °C
90 °C
115 °C 119 °C
60 °C
-0
20
40
60
80
100
120
140
Fixture A26W
Fixture B26W
Fixture C12W
Fixture D30W
Boa
rd T
empe
ratu
re (d
eg C
)
Open air Non-IC IC
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Lumen depreciation & color shiftLumen depreciation & color shift
In the IC condition:Life (L70) is less than 3000 hrsThe color shift is greater than a 36-step MacAdam ellipse (reached within 3000 hrs)
Fixture A - 26W LED Downlight
50%
60%
70%
80%
90%
100%
110%
100 1,000 10,000Time (hours)
Rel
ativ
e Li
ght O
utpu
t
Enclosed Semi-ventilated Open air
83 °C95 °C115 °C
Open airNon-ICIC
Fixture A - 26W LED Downlight
0
10
20
30
40
50
100 1,000 10,000Time (hours)
Mac
Ada
m E
llips
es
Enclosed Semi-ventilated Open air
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Fixture B - 26W LED Downlight
50%
60%
70%
80%
90%
100%
110%
100 1,000 10,000Time (hours)
Rel
ativ
e Li
ght O
utpu
t
Enclosed Semi-ventilated Open air
Lumen depreciation & color shiftLumen depreciation & color shift
In the IC condition,Life (L70) is less than 3000 hrsThe color shift is greater than a 19-step MacAdam ellipse (reached within 4000 hrs)
87 °C107 °C119 °COpen airNon-ICIC
Fixture B - 26W LED Downlight
0
10
20
30
40
50
100 1,000 10,000Time (hours)
Mac
Ada
m E
llips
es
Enclosed Semi-ventilated Open air
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Fixture C - 12W LED Downlight
50%
60%
70%
80%
90%
100%
110%
100 1,000 10,000Time (hours)
Rel
ativ
e Li
ght O
utpu
t
Enclosed Semi-ventilated Open air
Lumen depreciation & color shiftLumen depreciation & color shift
Even in the IC condition:Life (L70) seems very longThe color shift is within a 4-stepMacAdam ellipse (in the 3000 hrs)
42 °C50 °C60 °C
Open airNon-ICIC
Fixture C - 12W LED Downlight
0
2
4
6
8
10
100 1,000 10,000Time (hours)
Mac
Ada
m E
llips
es
Enclosed Semi-ventilated Open air
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Fixture D - 30W LED Downlight
50%
60%
70%
80%
90%
100%
110%
100 1,000 10,000Time (hours)
Rel
ativ
e Li
ght O
utpu
t
Enclosed Semi-ventilated Open air
80 °C90 °C
Open airNon-IC
Even in the IC condition:Life (L70) seems very longThe color shift is within a 3-stepMacAdam ellipse (in the 3000 hours)
Lumen depreciation & color shiftLumen depreciation & color shift
Fixture D - 30W LED Downlight
0
2
4
6
8
10
100 1,000 10,000Time (hours)
Mac
Ada
m E
llips
es
Semi-ventilated Open air
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
SummarySummaryOut of the 4 fixtures presented here, only one showed results acceptable for general lighting, considering:
Light outputEfficacyLumen depreciationColor shift over time
“ASSIST recommends” test methods were designed to:Provide more useful information for selecting and using LED directional lighting luminairesHelp differentiate between good and poor performing LED luminaires in terms of light output and life
Additional “ASSIST recommends” in preparation:LED light engineFreezer case luminairesOutdoor (parking lot) luminaire
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Standards for LEDsStandards for LEDs
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Product Performance and Measurement StandardsProduct Performance and Measurement Standards
ANSI - American National Standards Institute(www.ansi.org)
C78.377: Specifications for the Chromaticity of Solid State Lighting Products
In preparation:C82.SSL1: Power Supply C82.77-2002: Harmonic Emission Limits – Related Power Quality Requirements for Lighting
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Product Performance and Measurement StandardsProduct Performance and Measurement Standards
UL – Underwriters Laboratories(www.ul.com)
"UL 8750: Outline of Investigation for Light Emitting Diode Light Sources for Use in Lighting Products" is now published and will now be used in product investigations involving LEDs for use in Lighting applications.
UL 1993 - Self-Ballasted Lamps and Lamp Adapters
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Product Performance and Measurement StandardsProduct Performance and Measurement Standards
IESNA - Illuminating Engineering Society of North America(www.IESNA.org)
LM79: IESNA Approved Method for the Electrical and Photometric Measurements of Solid-State Lighting Products TM-16-05: IESNA Technical Memorandum on Light Emitting Diode (LED) Sources and Systems
In preparation:RP-16:Nomenclature and Definitions for Illuminating Engineering Addendum (revision)LM-80: IESNA Approved Method for Measuring Lumen Depreciation of LED Light Sources
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
U.S. DOE SSL ProgramU.S. DOE SSL Program
Information onENERGY STARCALiPERStandards DevelopmentTechnical Information NetworkTechnology DemonstrationsDesign Competition
http://www.netl.doe.gov/ssl
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
AcknowledgmentsAcknowledgments
Linc, NYSERDA and CEG
Sponsors of ASSIST Program
LRC faculty, staff, and students
© 2008 Rensselaer Polytechnic Institute. All rights reserved.
Thank you
www.lrc.rpi.edu/programs/solidstate