100,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Serie: IEEE Press Series on Power Engineering
- Sprache: Englisch
The essential how-to guide to designing and building LED systems, revised and updated The second edition of Practical Lighting Design with LEDs has been revised and updated to provide the most current information for developing light-emitting diodes products. The authors, noted authorities in the field, offer a review of the most relevant topics including optical performance, materials, thermal design and modeling and measurement. Comprehensive in scope, the text covers all the information needed to design LEDs into end products. The user-friendly text also contains numerous drawings and schematics that show how things such as measurements are actually made, and show how circuits actually work. Designed to be practical, the text includes myriad notes and illustrative examples that give pointers and how-to guides on many of the book's topics. In addition, the book's equations are used only for practical calculations, and are kept at the level of high-school algebra. This thoroughly expanded second edition offers: * New chapters on the design of an LED flashlight, USB light, automotive taillight, and LED light bulbs * A practical and user-friendly guide with dozens of new illustrations * The nitty-gritty, day-to-day engineering and systems used to design and build complete LED systems * An essential resource on the cutting-edge technology of Light-Emitting Diodes Practical Lighting Design with LEDs helps engineers and managers meet the demand for the surge in usage for products using light-emitting diodes with a practical guide that takes them through the relevant fields of light, electronic and thermal design.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 634
Veröffentlichungsjahr: 2017
Ähnliche
CONTENTS
Cover
Series Page
Title Page
Copyright
Dedication
About the Authors
Preface
List of Figures
Chapter 1: Practical Introduction To LEDs
What is an LED?
Small LEDs Versus Power Devices
Phosphors Versus RGB
Inside an LED
Is an LED Right for my Application?
Haitz's Law(s)
The Wild West
LEDs and OLEDs and…?
Chapter 2: Light Bulbs and Lighting Systems
Light Sources
Characteristics of Light Sources
Types of Bulbs
History of Lighting
Governments
Lighting Systems
Chapter 3: Practical Introduction To Light
The Power of Light
Radiometric Versus Photometric
Luminous Intensity, Illuminance, and Luminance (or Candela, Lux, and Nits)
What Color White?
Color Rendition: How the Light Looks Versus how Objects Look
Chapter 4: Practical Characteristics of LEDs
Current, Not Voltage
Forward Voltage
Reverse Breakdown
Not Efficiency—Efficacy!
LED Optical Spectra
Overdriving LEDs
Key Datasheet Parameters
Binning
The Tolerance Game
Chapter 5: Practical Thermal Performance of LEDs
Mechanisms Behind Thermal Shifts
Electrical Behavior of LEDs with Temperature
Optical Behavior of LEDs with Temperature
Other Performance Shifts with Temperature
LED Lifetime: Lumen Degradation
LED Lifetime: Catastrophic Failure
Paralleling LEDs
Chapter 6: Practical Thermal Management of LEDs
Introduction to Thermal Analysis
Calculation of Thermal Resistance
The Ambient
Practical Estimation of Temperature
Heat Sinks
Fans
Radiation Enhancement
Removing Heat from the Drive Circuitry
Chapter 7: Practical DC Drive Circuitry For LEDs
Basic Ideas
Battery Basics
Overview of SMPS
Buck
Boost
Buck-Boost
Input Voltage Limit
Dimming
Ballast Lifetime
Arrays
Chapter 8: Practical AC Drive Circuitry For LEDs
Safety
Which AC?
Rectification
Topology Selection
Nonisolated Circuitry
Isolation
Component Selection
EMI
Power Factor Correction
Lightning
Dimmers
Ripple Current Effects on LEDs
Lifetime
UL, Energy Star, and All That
Chapter 9: Practical System Design with LEDs
PCB Design
Getting the Light out
LEDs in Harsh Environments
Designing with the Next-Generation LED in Mind
Lighting Control
Chapter 10: Practical Design of an LED Flashlight
Initial Marketing Input
Initial Analysis
Specification
Power Conversion
Selecting Components
Thermal Model
PCB
Design Rule Check
Gerber Files
Panelization
Final Design
Chapter 11: Practical Design of a USB Light
Initial Marketing Input
Initial Analysis
Specification
Power Conversion
Efficiency
Thermal Model
MTTF
PCB
Final Design
Chapter 12: Practical Design of an Automotive Tail Light
Initial Marketing Input
Initial Analysis
Specification
Load Dump
Power Conversion
Thermal Model
WCA
Testing
PCB
Final Design
Chapter 13: Practical Design of an Led Light Bulb
Initial Marketing Input
Initial Analysis
Specification
Power Conversion
THD
Flicker Index and Percent Flicker
Utilization
PCB
Spacing
Final Design
Chapter 14: Practical Measurement of LEDs and Lighting
Measuring Light Output
Special Considerations in Measuring Light Output of LEDs
LED Measurement Standards
ASSIST
Measuring LED Temperature
Measuring Thermal Resistance
Measuring Power, Power Factor, and Efficiency
Accelerated Life Tests
Chapter 15: Practical Modeling of LEDs
Preliminaries
Practical Overview of Spice Modeling
What Not to Do
What to Do
Modeling Forward Voltage
Reverse Breakdown
Modeling Optical Output
Modeling Temperature Effects
Modeling the Thermal Environment
A Thermal Transient
Some Comments on Modeling
References
Index
IEEE Press Series on Power Engineering
End User License Agreement
List of Tables
Table 1.1
Table 3.1
Table 3.2
Table 3.3
Table 3.4
Table 3.5
Table 3.6
Table 6.1
Table 8.1
Table 8.2
Table 8.3
Table 8.4
Table 10.1
Table 10.2
Table 10.3
Table 10.4
Table 10.5
Table 11.1
Table 11.2
Table 11.3
Table 11.4
Table 11.5
Table 12.1
Table 12.2
Table 12.3
Table 13.1
Table 13.2
Table 15.1
Table 15.2
Table 15.3
Table 15.4
Table 15.5
Table 15.6
Table 15.7
Table 15.8
Table 15.9
Table 15.10
Table 15.11
Table 15.12
Table 15.13
Table 15.14
Table 15.15
Table 15.16
Table 15.17
Table 15.18
Table 15.19
List of Illustrations
Figure 1.1
Figure 1.2
Figure 1.3
Figure 1.4
Figure 2.1
Figure 2.2
Figure 3.1
Figure 3.2
Figure 3.3
Figure 3.4
Figure 3.5
Figure 3.6
Figure 3.7
Figure 3.8
Figure 3.9
Figure 3.10
Figure 3.11
Figure 3.12
Figure 3.13
Figure 3.14
Figure 3.15
Figure 3.16
Figure 4.1
Figure 4.2
Figure 4.3
Figure 4.4
Figure 4.5
Figure 4.6
Figure 4.7
Figure 4.8
Figure 4.9
Figure 4.10
Figure 5.1
Figure 5.2
Figure 5.3
Figure 6.1
Figure 6.2
Figure 6.3
Figure 6.4
Figure 6.5
Figure 6.6
Figure 7.1
Figure 7.2
Figure 7.3
Figure 7.4
Figure 7.5
Figure 7.6
Figure 7.7
Figure 7.8
Figure 7.9
Figure 7.10
Figure 7.11
Figure 7.12
Figure 8.1
Figure 8.2
Figure 8.3
Figure 8.4
Figure 8.5
Figure 8.6
Figure 8.7
Figure 8.8
Figure 8.9
Figure 8.10
Figure 8.11
Figure 8.12
Figure 8.13
Figure 8.14
Figure 8.15
Figure 8.16
Figure 8.17
Figure 8.18
Figure 8.19
Figure 8.20
Figure 8.21
Figure 8.22
Figure 8.23
Figure 8.24
Figure 9.1
Figure 9.2
Figure 9.3
Figure 9.4
Figure 9.5
Figure 10.1
Figure 10.2
Figure 10.3
Figure 10.4
Figure 10.5
Figure 10.6
Figure 11.1
Figure 11.2
Figure 11.3
Figure 11.4
Figure 11.5
Figure 11.6
Figure 12.1
Figure 12.2
Figure 12.3
Figure 12.4
Figure 12.5
Figure 12.6
Figure 12.7
Figure 13.1
Figure 13.2
Figure 13.3
Figure 13.4
Figure 13.5
Figure 13.6
Figure 13.7
Figure 14.1
Figure 14.2
Figure 14.3
Figure 14.4
Figure 14.5
Figure 14.6
Figure 14.7
Figure 15.1
Figure 15.2
Figure 15.3
Figure 15.4
Figure 15.5
Figure 15.6
Figure 15.7
Figure 15.8
Figure 15.9
Figure 15.10
Figure 15.11
Figure 15.12
Figure 15.13
Guide
Cover
Table of Contents
Begin Reading
Chapter 1
Pages
i
ii
iii
iv
v
vi
ix
x
xi
xii
xiii
xiv
xv
xvi
xvii
xviii
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
IEEE Press445 Hoes LanePiscataway, NJ 08854
IEEE Press Editorial BoardTariq Samad, Editor in Chief
Giancarlo Fortino
Xiaoou Li
Ray Perez
Dmitry Goldgof
Andreas Molisch
Linda Shafer
Don Heirman
Saeid Nahavandi
Mohammad Shahidehpour
Ekram Hossain
Jeffrey Nanzer
Zidong Wang
Second Edition
Practical Lighting Design with LEDs
Ron Lenk
Carol Lenk
Copyright © 2017 by The Institute of Electrical and Electronics Engineers, Inc. All rights reserved.
Published by John Wiley & Sons, Inc., Hoboken, New Jersey.
Published simultaneously in Canada.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.
Library of Congress Cataloging-in-Publication Data is available.
ISBN: 978-1-119-16531-6
To our children, for being so patient
About the Authors
Ron Lenk is an authority in the fields of power electronics, power systems and LED drivers. The author of the bestselling Practical Design of Power Supplies (Wiley), he has spent the last twelve years working on LEDs and lighting. Lenk co-founded and was CEO at Switch Light, Inc. which made general-service LED light bulbs, and now is a consultant in the fields of power and LEDs. He is a Senior Member of the IEEE and has 35 issued US patents.
Carol Lenk was the co-founder and Director of Engineering at Switch Light, Inc. She earned a B.S. in electrical engineering from MIT and a master's in math and science education. One of the pioneers in applying LEDs to general lighting, Lenk has ten years' experience combining theoretical concepts with practical engineering in fields as diverse as optics, thermal modeling, material science, electronics and mechanical design. She is now a consultant in the field of LEDs and has nine issued US patents, all relating to LED lighting.
Preface
The Lighting Revolution
LEDs are bringing in a new era in lighting. Similar to the evolution of computing power that computers went through, from vacuum tubes to the silicon-based semiconductor brains of modern-day computers, lighting is now riding an exponential growth wave in efficacy. From oil lamps to the invention of the Edison light bulb 100 years ago to the fluorescent lights of 50 years ago to the LEDs of today, lighting technology is finally joining the modern world of solid-state technology.
LED-based lighting is increasingly becoming the efficient light source of choice, replacing both incandescent and fluorescents. The hurdles that have kept consumers from adopting energy-efficient lighting, such as shape, color quality, the presence of toxic mercury, and limited lifetime, are all better addressed by LEDs.
In the long term, LED-based lighting will be better and cheaper than every other light source. It will become the de facto light of choice. LED lighting will be cheap, efficient, and used in ways that haven't been imagined yet. It will transform the $100 billion lighting industry, and with transformation comes opportunity.
The Last Vacuum Tube
Lighting is the last field that still uses vacuum tubes. All electronics today use integrated circuits because of the enormous benefits in performance and cost. But a fluorescent tube is a type of vacuum tube. LEDs are solid-state devices, the same as the rest of electronics. The amount of light that an LED can convert from 1 W of power is already nearly double that of the best fluorescent tubes. The future is even brighter as LEDs are anticipated to continue that growth in the next decade, and then soon go on to reach the physical limits of electricity to light conversion. We look forward to seeing the last ceiling-mounted vacuum tube in the not-too-distant future.
Green Lighting
The benefits of using LEDs for lighting are many. The most obvious is their efficiency. Lighting accounts for 20% of total electricity use throughout the world today. Using LEDs could cut this down to 4% or less. As LEDs become the dominant light source over the next decade, the reduction of energy used and greenhouse gases emitted will benefit everyone. Each consumer will save hundreds of dollars every year from reduced energy use. Building owners will save even more. Utilities will be better equipped to manage growth. And the earth will experience the accumulation of fewer greenhouse gases, as well as a reduction in the emission of toxic mercury found in fluorescent lighting.
A Lifetime of Lighting
As solid-state devices, LEDs have extremely long lifetimes. They have no filaments to break. They can't leak air into their vacuum because they don't use a vacuum. In fact, they don't really break at all; they just very gradually get dimmer. Imagine changing your light bulb only once or twice in your entire lifetime!
Lighting the World with LEDs
Just as microprocessors got cheaper and more powerful, LEDs are also benefiting from the cost-reduction techniques developed in the semiconductor industry. LED light prices are already on par with fluorescent tubes. And with lower prices will come the ability to tailor light to the specific needs of the consumer. Taken together with LEDs' reduced energy usage, this will enable the universal availability of lighting. Imagine every child in the poorest village having a light to read by.
The design of LED-based lighting systems is an exciting field, but these systems are fairly technical. With this book, we hope to enable the readers to do great things with lighting, both for themselves and for the world.
Marietta, Georgia
Ron Lenk
Carol Lenk
List of Figures
Figure 1.1
T1¾ (5 mm) LEDs.
Figure 1.2
Fluorescent tube's spectral power distribution. (
Source:
http://www.gelighting.com/na/business_lighting/education_resources/learn_about_light/pop_curves.htm?1
.)
Figure 1.3
LEDs can be used everywhere. (
Source:
Kaist, KAPID.)
Figure 1.4
Haitz's law. (
Source:
http://i.cmpnet.com/planetanalog/2007/07/C0206-Figure3.gif
. Reprinted with permission from Planet Analog/EE Times, copyright United Business Media, all rights reserved.)
Figure 2.1
Currents in a fluorescent tube.
Figure 2.2
Various bulb shapes. (Courtesy of Halco Lighting Technologies.)
Figure 3.1
The electromagnetic spectrum.
Figure 3.2
Scotopic vision is much more sensitive than photopic vision. (
Source:
Kalloniatis and Luu (2007).)
Figure 3.3
Emission spectra of four common light sources.
Figure 3.4
Solar radiation spectrum. (
Source:
http://en.wikipedia.org/wiki/File:Solar_Spectrum.png
under license
http://creativecommons.org/licenses/by-sa/3.0/
. Accessed January 2011.)
Figure 3.5
One steradian intersects 1 m
2
of area of a 1-m radius ball. (
Source:
http://commons.wikimedia.org/wiki/File:Steradian.png
under license
http://creativecommons.org/licenses/by-sa/3.0/
.)
Figure 3.6
Solid angle in steradians (sr) versus half beam angle in degrees (°).
Figure 3.7
Definition of beam angle.
Figure 3.8
Typical Lambertian radiation pattern. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Co., 2007.)
Figure 3.9
Dimensions for a USB keyboard light design.
Figure 3.10
Spectra of neutral-white (a) and warm-white (b) LEDs. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Co., 2007.)
Figure 3.11
CIE 1931 (
x
,
y
) chromaticity space, showing the Planck line and lines of constant CCT. (
Source:
http://en.wikipedia.org/wiki/Color_temperature
under license
http://creativecommons.org/licenses/by-sa/3.0/
.)
Figure 3.12
(
x
,
y
) Chromaticity diagram showing CCT and seven-step MacAdam ellipses. (
Source:
http://www.photonics.com/Article.aspx?AID=34311
.)
Figure 3.13
(a) Cool white fluorescent 4100 K, CRI 60; (b): Incandescent, 2800 K, CRI 100; (c): Reveal® incandescent 2800 K, CRI 78. (
Source:
http://www.gelighting.com/eu/resources/learn_about_light/pop_color_booth.html
.)
Figure 3.14
Approximate Munsell test color samples. (
Source:
http://en.wikipedia.org/wiki/Color_rendering_index
under license
http://creativecommons.org/licenses/by-sa/3.0/.
)
Figure 3.15
Circadian rhythm sensitivity. (
Source:
“Visibility, Environmental and Astronomical Issues Associated with Blue-Rich White Outdoor Lighting,” May 2010, IDA. Image copyright of IDA.)
Figure 3.16
Identical gray boxes look different depending on their background.
Figure 4.1
Reverse bias protection.
Figure 4.2
LEDs with reverse bias protection.
Figure 4.3
Light output as a function of current. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Company, 2007.)
Figure 4.4
Forward voltage as a function of current. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Company, 2007.
Figure 4.5
Efficacy versus drive current.
Figure 4.6
Light output as a function of wavelength. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Co., 2007.)
Figure 4.7
Many LEDs have poor R9. (
Source:
http://www.yegopto.co.uk/LightingLEDs/CRI_Seoul_Semi
.)
Figure 4.8
(
x
,
y
) as a function of current. (
Source:
C6060-16014-CW/NW Datasheet, Intematix Technology Center Corp., 3/2008.)
Figure 4.9
Different output light distributions are available. (
Source:
http://www.philipslumileds.com/technology/radiationpatterns.cfm
.)
Figure 4.10
Neutral-white bin structure. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Company, 2007.)
Figure 5.1
Brightness as a function of temperature. (
Source:
“Technical Datasheet DS56, Power Light Source Luxeon Rebel,” Philips Lumileds Lighting Company, 2007.)
Figure 5.2
LED temperature profile for parameters given in the text.
Figure 5.3
Forward voltage as a function of current. (
Source:
“Technical Datasheet DS56, Power Light Source Luxeon Rebel”, Philips Lumileds Lighting Company, 2007.)
Figure 6.1
Thermal model for LED example.
Figure 6.2
Thermal model of two parallel thermal paths.
Figure 6.3
LED temperature as a function of time.
Figure 6.4
There are many thermal paths to ambient.
Figure 6.5
Estimating temperature rise from power density.69
Figure 6.6
An LED heat sink. (
Source:
http://www.aavidthermalloy.com/cgi-bin/stdisp.pl?Pnum=569000b00000g
. Courtesy Aavid Thermalloy.)
Figure 7.1
I
–
V
curve of 12 V battery. (
Source:
Lenk (1998).)
Figure 7.2
Alkaline cell battery voltage as a function of time with a resistive load. (
Source:
Rayovac, OEM 151 (R-3/99), “Application Notes & Product Data Sheet,” “Primary Batteries—Alkaline & Heavy Duty,” Figure 1. Property of Spectrum Brands, Inc.)
Figure 7.3
Operating a transistor in linear mode is inefficient.
Figure 7.4
When the Transistor (
t
) is on, current in the inductor (
I
) increases; when the transistor is off, current in the inductor decreases.
Figure 7.5
LM3405 schematic for buck. (
Source:
LM3405 datasheet, National Semiconductor, February 2007.)
Figure 7.6
FAN5333A schematic for boost. (
Source:
FAN5333A datasheet, Fairchild Semiconductor, August 2005.)
Figure 7.7
HV9910 schematic for buck-boost. (
Source:
HV9910 datasheet, Supertex Inc., 2006.)
Figure 7.8
Pulse width modulation turns the current rapidly on and off to get an average current.
Figure 7.9
Dimming circuit.
Figure 7.10
The effect of the current sense resistor is compensated by putting one in series with each string.
Figure 7.11
LED forward voltage variation can be compensated at the cost of additional power.
Figure 7.12
Ballasting LED strings with total current sensing.
Figure 8.1
Block diagram of AC SMPS for LED lighting.
Figure 8.2
A bridge rectifier.
Figure 8.3
Half-wave rectification.
Figure 8.4
Reducing the ripple from a bridge rectifier with a capacitor.
Figure 8.5
Running LEDs directly off-line.
Figure 8.6
How the off-line buck works.
Figure 8.7
A nonisolated off-line LED driver.
Figure 8.8
Adding a transformer makes the converter into a forward.
Figure 8.9
Adding a transformer makes the converter into a flyback.
Figure 8.10
Protecting the HV9910 from high voltages.
Figure 8.11
Resistors balance voltages for series capacitors.
Figure 8.12
Normal mode EMI filtering for a two-wire input.
Figure 8.13
Common mode EMI filtering added for a three-wire input.
Figure 8.14
Current loops may cause EMI problems: reducing loop area helps.
Figure 8.15
A big capacitor maintains constant voltage during the line cycle, generating large peak currents and bad power factor.
Figure 8.16
A smaller and cheaper PFC.
Figure 8.17
Simple power factor correction circuit.
Figure 8.18
Adding an MOV to the design protects it moderately well from lightning.
Figure 8.19
Output waveform of a triac dimmer.
Figure 8.20
Keeping an IC's power alive during the off-time of a dimmer.
Figure 8.21
As ripple current increases, power loss in the LED also increases. (
Source:
Betten and Kollman (2007). Used by permission of Electronics Technology, a Penton Media publication.)
Figure 8.22
Forward voltage increases with increasing current.
Figure 8.23
Increasing the current does not proportionally increase the light. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Co., 2007.)
Figure 8.24
200 mApp ripple current on a 350 mA DC drive.
Figure 9.1
A good schematic.
Figure 9.2
A schematic that could be improved. (
Source:
HV9910 datasheet, Supertex Inc., 2006.)
Figure 9.3
Poor grounding layout (a) and improved layout (b).
Figure 9.4
Soda-lime glass optical transmission. (
Source:
http://en.wikipedia.org/wiki/File:Soda_Lime.jpg
under license
http://creativecommons.org/licenses/by-sa/3.0/
.)
Figure 9.5
DALI topologies.
Figure 10.1
FAN5333A schematic for flashlight.
Figure 10.2
FAN5333A final schematic for flashlight.
Figure 10.3
Thermal model.
Figure 10.4
Schematic for LED flashlight.
Figure 10.5
Layout of LED flashlight.
Figure 10.6
Panelization of LED flashlight.
Figure 11.1
First-cut USB light schematic.
Figure 11.2
LM3405 schematic for USB light.
Figure 11.3
MP-3030 V-I curve.
Figure 11.4
MIL-HDBK-217F calculation for USB light.
Figure 11.5
Final LM3405 schematic for USB light.
Figure 11.6
USB light PCB, top view and X-ray bottom view.
Figure 12.1
HV9910 schematic for tail light.
Figure 12.2
Final HV9910 LED automobile tail light schematic.
Figure 12.3
WCA table.
Figure 12.4
HV9910 LED automobile tail light schematic.
Figure 12.5
LED tail light. Whole board.
Figure 12.6
LED tail light. Top view.
Figure 12.7
LED Tail light. Bottom view.
Figure 13.1
BR bulbs.
Figure 13.2
Initial thermal model for LEDs in a BR40.
Figure 13.3
Initial design for a BR bulb.
Figure 13.4
Design for a BR bulb.
Figure 13.5
Schematic for a BR40 bulb.
Figure 13.6
BR40 bulb PCB.
Figure 13.7
BR40 bulb PCB zoom.
Figure 14.1
Typical lux meter. (
Source:
http://www.p-mastech.com/products/06_ifep/ms6610.html
.)
Figure 14.2
Sketch of operation of an integrating sphere. (
Source:
http://www.aaccuracy.com/optical.htm
.)
Figure 14.3
Walk-in integrating sphere.
Figure 14.4
Gooch & Housego 6 in. integrating sphere with OL770-LED spectroradiometer. (
Source:
http://www.olinet.com//content/file/1210967318B220_770-LED_Brochure.pdf
.)
Figure 14.5
Location of thermocouple to measure LED temperature. (
Source:
Thermal Design Using Luxeon® Power Light Sources, Lumileds AB05, 2006.)
Figure 14.6
High current drawn at AC line peak gives a high crest factor.
Figure 14.7
Phase shift between current and voltage also gives low power factor.
Figure 15.1
Luxeon Rebel I–V curve. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Co., 2007.)
Figure 15.2
First model of Luxeon Rebel.
Figure 15.3
Setting the GMIN option to 1.0E-20 fixes the current at 0 V.
Figure 15.4
LED model including reverse breakdown.
Figure 15.5
With the breakdown modeled, the complete
I
–
V
curve is shown. The
y
-axis is the negative of anode-to-cathode current.
Figure 15.6
Normalized light output versus current. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Co., 2007.)
Figure 15.7
Data versus equation for optical output versus drive current.
Figure 15.8
LED model with light output, temperature not yet included.
Figure 15.9
The effect of temperature on light output. (
Source:
Technical Datasheet DS56, Power Light Source Luxeon Rebel, Philips Lumileds Lighting Co., 2007.)
Figure 15.10
LED model with temperature effect on both forward voltage and optical output.
Figure 15.11
Complete LED model.
Figure 15.12
The thermal response of two LEDs to current pulse.
Figure 15.13
The optical response of two LEDs to current pulse.
Chapter 1Practical Introduction To LEDs
Light bulbs are everywhere. There are over 20 billion light bulbs in use around the world today. That is, three for each person on the planet! We expect that within the next 10 years, the majority of these bulbs will be light-emitting diodes (LEDs). This is because LEDs can provide efficiency dozens of times higher than incandescent light bulbs. They can be as efficient as the theoretical limit for electricity to light conversion set by physics. This book is all about the practical aspects of LEDs and how you can make practical lighting designs using them.
What is an LED?
The purpose of this book is to tell you practical things about LEDs. So in this section, we're not going to regale you with jargon about “direct bandgap GaInP/GaP strained quantum wells” or such. Let's directly address the question: What is an LED?
The name “light-emitting diode” tells you a lot already. In the first place, the noun tells you that it is a diode. A diode conducts current in one direction and not the other. And that's what an LED does. While we'll explore the details of its electrical behavior in Chapter 4, the only thing to note for the moment is that it has a much higher forward voltage than the diodes usually used in electronics. While a 1N4148 has a drop of about 700 mV, an LED may drop 3.6 V. This is because LEDs are not made from silicon, but from other semiconductors. But other than that, an LED's electrical characteristics are very much like those of other diodes.
The words “light-emitting” tell you a lot more. Now all diodes emit at least a little bit of light. You can open up an integrated circuit (IC) and use a scanner to see which parts of the circuit are emitting light. This tells you which parts are conducting current. IC designers use this to help debug their ICs. However, the amount of light emitted by ICs is very small. Since the purpose of LEDs is to emit light, they have been carefully designed to optimize this performance. That's why, for example, they have a much higher forward voltage than normal, rectifier diodes. Rectifiers have been optimized to minimize their forward voltage while maximizing reverse breakdown voltage. LEDs are optimized to produce the most light of the right color at the lowest power, and things such as forward voltage (by itself) don't matter. Of course, forward voltage does enter into how much power the LED dissipates, and we'll see in Chapter 5 how to characterize the light emitted versus the power dissipated.
Small LEDs Versus Power Devices
Present-day thinking divides LEDs into two classes: small devices and power devices. Small LEDs became widely used in the 1970s. They come in all different colors, such as red, orange, green, yellow, and blue. They are the small T1¾ (5 mm) devices shown in Figure 1.1. Nowadays, there are literally tens of billions of them sold each year. They go into cell phone backlights, elevator pushbuttons, flashlights, incandescent bulb replacements, fluorescent tube replacements, road signage, truck taillights, traffic lights, automobile dashboards, and so on.
Figure 1.1 T1¾ (5 mm) LEDs.
What characterizes these small devices is their power level, or as the industry thinks of it, their drive current. The typical red small LED, for example, has a drive current of 20 mA. At a forward voltage of 2.2 V, this is only 44 mW of power. (The efficacy is so low that this is just about equal to the heat dissipation as well.) Small white LEDs have a higher forward voltage (3.6 V, corresponding to 72 mW), and some small LEDs can be run as high as 100 mA. But fundamentally, this type of LED is used as an indicator, not a real light source. It takes 14 of them to make a somewhat reasonable 1 W flashlight, and hundreds of them to make a (dim) fluorescent tube replacement.
While the information in this book is applicable to these small LEDs, the main focus is on power devices. Power devices are typically 1–3 W devices that are usually run at 350 mA. Their dice (the actual semiconductor, as opposed to its package) are substantially larger than those of small LEDs, although their footprint need not be. These devices are typically used in places requiring lighting, rather than as indicators. Applications include flashlights, incandescent bulb replacements, large-screen TVs, projector lights, automotive headlights, airstrip runway lighting, and just about everywhere lighting is used. Of course, not all of these applications have yet seen widespread adoption of power LEDs, but they will soon.
Phosphors Versus RGB
Most lighting designs are going to be made with white light (which includes incandescent “yellow” light). For this reason, this book concentrates primarily on white LEDs. However, what is described here for white LEDs can be straightforwardly applied to color LEDs. Color LEDs are very similar to white, albeit with differing forward voltage. The reason for the varying forward voltages is that the colored light (red, yellow, blue, etc.) is generated directly by the semiconductor material. The material is varied to get differing colors and the differences in material in turn cause differences in forward voltages.
However, white light cannot be directly generated by a single material (we are ignoring special types of engineered materials that are not yet in production). White light consists of a mixture of all of the colors. You already know this because white light can be separated into its constituent colors with a prism. White light thus has to be created. There are currently two main methods of generating white light with LEDs. In one method, an LED that emits blue light is used, and the blue light is converted to white by a phosphor. In the other method, a combination of different color LEDs is used.
The first method is the most common. A typical wavelength for the blue light generated by the LED is 435 nm. Why use blue light? This has to do with the physics of the way the white light is generated. The blue light is absorbed by a phosphor, and re-emitted as a broad spectrum of light approximating white. For the phosphor to be able to absorb and re-emit the light, the light coming out has to be lower in energy than the light going in. That's just like any electronic component. Energy goes in, some is dissipated as heat, and the rest comes out again, transformed. So to get all of the colors in the spectrum that humans can see, the phosphor needs to have input at a higher energy (shorter wavelength) than the shortest color's energy. For humans, this is about 450 nm, and so a 435 nm blue LED is the most energy-efficient way of generating white light using a phosphor.
Before turning to the second method of generating white light, we should say a few more words about the phosphor. There are various types of phosphors. Phosphors are designed to absorb one specific wavelength of light, and re-emit it at either one or more different wavelengths or in a band of wavelengths. LED phosphors are typically designed to do the latter. But there are limits to how broad a band of colors a phosphor can emit. So many LEDs use bi-band or tri-band phosphors to better cover the spectrum of light needed to approximate white. These phosphors are mixtures of two or three primary phosphors. These more complicated phosphors are typically used when better color rendition is needed (see the discussion of color rendering index (CRI) in Chapter 3).
As a side note, we can comment briefly on fluorescent lights. In some ways, a fluorescent light is quite similar to an LED, but its fundamental mechanism of light emission is different. It generates a high-temperature plasma inside a tube, which emits light in the ultraviolet (UV) range (254 nm) rather than in the blue range. But after that, it too uses a phosphor to absorb the light and re-emit it in the visible range. Note that since the wavelength of the light is considerably farther away from the visible spectrum than the 435 nm generated electrically by the LED die, the efficiency ultimately possible for a fluorescent is intrinsically lower than that possible for an LED. (At the moment, fluorescent lights and LEDs have roughly the same efficiency.)
But also interesting is the type of phosphor the typical fluorescent light uses. These phosphors are of the type that re-emits in just one or two narrow wavelengths, not in a band of colors. The specific wavelengths emitted have been very carefully chosen to make the light emitted give a good specification for the CRI. But the spiky nature of the emission spectrum (see Fig. 1.2) means that colors at wavelengths other than these are poorly reproduced by the fluorescent lamp. Of course, there is no reason (we know of) that fluorescents can't have the same spectra that LEDs do. But for the present moment at least, LEDs have the potential to give much better color rendition than do fluorescent lamps.
Figure 1.2 Fluorescent tube's spectral power distribution. (Source:http://www.gelighting.com/LightingWeb/na/resources/tools/lamp-and-ballast/pop_curves.jsp?12.)
Inside an LED
This book is about designing lighting with LEDs, not about how to make them. Nonetheless, some aspects of their construction are worth knowing. It helps to understand some of the design aspects of different manufacturers' products. It also helps to understand some of their claimed improvements in lifetime. We'll be talking about white LEDs made with phosphors, although much of the information is the same for other types.
The first thing to realize is that while almost all of the devices currently used by engineers—diodes, transistors, logic gates, microprocessors—are made of silicon, LEDs are not made of silicon. (There used to be some germanium devices around, but they don't work very well when they get hot, and so were abandoned.) However, it has proven difficult to get silicon to emit light. Thus, a number of different semiconductors have been put to use. While it's not important to know the details, you should realize that there are a variety of different materials being tried. Not all of the physics is understood yet, and the aging processes are unclear as well. Different types are in use for different devices from different manufacturers. What this means practically is that you should expect changes ahead. The device you buy today will probably be different from what is available tomorrow.
The fundamental semiconductor device in an LED is relatively large, a few square millimeters. This device emits blue light (for white LEDs), and two things must be done to it: the blue light has to be converted to white light with high efficiency, and the white light has to come out without being blocked. So the normal ceramic package that ICs come in won't work, because it (intentionally) doesn't let any light through.
What most manufacturers do is to add some transparent silicone (a rubbery polymer) on top of the die. This lets the light come out without much absorption or color change, bending the light as needed, and providing a degree of mechanical protection for the die. At least one manufacturer then adds a piece of glass on top of the silicone, although it's not clear to us that this offers much advantage.
To accomplish the color conversion, a phosphor is used, which is a complex molecule that absorbs the blue light that the LED is emitting and radiates it out over a band of other colors. It takes two or three different phosphors to make a reasonable white color; you should expect to see phosphor blends with even more components in the future.
Some manufacturers put the phosphor directly on top of the die, with the silicone going on top of that. Others stir it into the silicone before putting the mixture on top of the die. Putting it directly on the die increases the amount of blue light that is absorbed, but makes the phosphors sit at the same temperature as the die. Phosphors tend to degrade with high temperature. Indeed, phosphor degradation is one of the major reasons why LED light output decreases with age. Putting the phosphor in the silicone reduces the temperature the phosphors have to survive, but decreases the amount of blue light that is absorbed and converted. You could add more phosphor to compensate for this, except that phosphors are relatively expensive.
The die, phosphor, and silicone are all in a package. (And every manufacturer has its own package and footprint.) The package includes bond wires that connect the die to the leads so that you can put current through the LED. Even though it's just a single device, multiple bond wires are used in parallel to accommodate the relatively high currents (Fig. 1.3).
Figure 1.3 LEDs can be used everywhere. (Source: Kaist, KAPID.)
Now the package has an unwanted side effect. Since the LED emits light over a broad angle, some of the light is intercepted by the package. This affects efficacy somewhat, but also some of the intercepted light is reflected and emitted. That's okay, except that as the package ages (it's sitting at 85 °C for 50,000 h), it yellows. As the package yellows, the absorption of light by the package increases, which decreases the efficacy. And the reflected light is also yellowed, causing the correlated color temperature (CCT) and CRI of the emitted light to shift. In some devices, this package aging is one of the major reasons why the LED time to 70% light output is 50,000 h and not longer.
Some LEDs also include some optics in their package. This may take the form of a lens and/or a mirror. The optics may be used to increase light extraction or to shape the emission direction of the light. If you don't care about the emission direction of the light (e.g., if you're building an omnidirectional light bulb), you should try to avoid using devices with extra optics. (Why pay for the extra cost?)
Thus, LEDs are complicated devices. It's well worth your while to ask detailed questions of your vendor about how the devices are made and how they will stand up to high-temperature aging. You may even need to speak to people at the factory to get sufficient information.
Is an LED Right for my Application?
To listen to enthusiastic marketing, it seems that LEDs can be used everywhere. But even though this book is about LEDs, we have to acknowledge that not every application will be best served by them. As LEDs continue to increase in efficacy and drop in price, more and more applications will benefit from them. We expect that ultimately fluorescent tubes will become obsolete. But we also expect that incandescent bulbs will be around for a long, long time. Here's a checklist of things to think about in deciding whether an LED solution is right for your application (Table 1.1).
Table 1.1 Checklist of Considerations on Whether to Use LEDs for an Application
Question
LED
Fluorescent
Incandescent
Is energy efficiency top priority?
LEDs are probably best
Is cost an important factor?
Fluorescents should be considered
Is cost the
only
thing that matters?
Best to use an incandescent
Does the application need long life?
LEDs, properly designed, are the best choice
Fluorescents may be good enough
Are there lots of on/off cycles?
LEDs should definitely be used
Are there temperature extremes?
LEDs are better than fluorescents, and usually good enough
For really extreme conditions, incandescent bulbs are even better
Is the heat generated used for other purposes?
LEDs may not dissipate enough heat, for example, to melt snow off a traffic light
Fluorescents also may not dissipate enough heat, for example, to melt snow off a traffic light
Incandescent bulbs may remain a good choice
Is good color rendition needed?
LEDs are sometimes good enough
Fluorescents almost never are
Incandescent bulbs remain the best
Do colors need to be changed in operation?
LEDs are the only choice
Is a new form factor needed?
LEDs are the only choice
Haitz's Law(s)
You've probably heard of Moore's law. This was the prediction by Moore in 1965 that the performance of microprocessors would double every 2 years. It was based on observations, but proved to be remarkably accurate for the next 40 years. It is only now that it has finally slowed, as ICs reach some fundamental physical limits.
A similar prediction for LEDs was made by Roland Haitz (2006). This is backed by much more historical data (see Fig. 1.4). As currently stated, it predicts that the luminous output of individual LED devices is increasing at a compound rate of 35% per year and that the cost per lumen is decreasing at 20% per year. To the extent that current manufacturers seem to have settled on 3 W as the maximum practical power in a small device, we can read this as also meaning an increase in efficacy of 35% per year.
Figure 1.4 Haitz's law. (Source:http://i.cmpnet.com/planetanalog/2007/07/C0206-Figure3.gif. Reprinted with permission from Planet Analog/EE Times, copyright United Business Media, all rights reserved.)
This predicted rate of performance increase would be utterly unbelievable, except that it appears to be true. The authors began tracking the prediction a number of years ago, calculating where efficacy would be each month. Year after year, we have verified the numbers, and efficacy indeed continues to increase.
We talked to Haitz a couple of years ago about his law. His opinion was that it still had a long run ahead of it. And while he may be right that the lumens per device will continue to increase, in the next few years the efficacy will certainly start deviating, of course due to fundamental physical limitations.
To understand Haitz' law, we need to consider the meaning of “lumens” (see Chapter 3 for more detailed information). Lumens is not exactly a measure of light, but is rather a measure of how much light humans see with their eyes. As such, it very much depends on how eyes work. In particular, human eyes are most sensitive to green light. Thus, if you produce 1 W of light at 555 nm, you have 683 lumens. There's no possible way to increase this number; it is really almost a definition. The same is of course true for LEDs. If an LED gets 1 W of power, and converts it entirely to light at 555 nm, it will have no heat power dissipation at all. (Obviously, this is not really possible because of the Second Law of Thermodynamics.) All of that light then is equal to 683 lumens. So efficacy is limited to 683 Lm/W no matter what.
Now the reality is that we don't normally want intense green light. We want white light. And since white light consists of many different colors, the lumens and efficacy must be less than 683 Lm/W. What then is the real limit on efficacy?
There are two different limits, depending on how the white light is generated. Recall that white light can be made either by directly combining lights of different colors or by emitting low-wavelength light (such as blue or UV) and converting it with phosphors into white. The phosphor method is limited by the physical efficiency of phosphors. Since they absorb low-wavelength (=high energy) light, and emit higher wavelength (=lower energy) light, the difference in energy is lost as heat. This is described by Stokes' law. While the exact limit is subject to details (such as what CRI light is acceptable), phosphor conversion of white light from LEDs is limited to about 238 Lm/W. Note that since fluorescent tubes are also phosphor-converted, but starting from 235 nm rather than 435 nm, their ultimate efficacy is considerably lower than that of LEDs. While they too have room for improvement currently, ultimately LEDs will be more efficient than fluorescent lighting.
Direct emission of various color lights can be more efficient, because there is no absorption and re-emission involved. But since colors other than green are needed, the human eye response means that 683 Lm/W isn't achievable this way either. A seminal paper by Ohno (2004) shows that to get acceptable CRI, white light cannot be made at higher efficacy than about 350 Lm/W.
Haitz's law as extrapolated to efficacy thus has several more years to run. As the 200 Lm/W limit is reached, blue converted by phosphors will plateau in efficacy. To continue increasing in efficacy, red-green-blue (RGB) systems will need to be implemented. But if 35% per year is continued, only 2 years will remain before the ultimate limit in efficacy is achieved. After that, Haitz's law may still apply to the cost per lumen. Indeed, the figure shows that it is not until after 2015 that the cost per lumen of LEDs will approach that of 60 W incandescent bulbs.
Ultimately, then, LEDs can be expected to reach the theoretical limit of efficacy, and their cost can ultimately drop below that of incandescent bulbs. And what happens after that? Since efficacy can't be increased because of physics, it might be reasonable to suppose that LEDs are here to stay for the long term. Nothing can be better than LEDs, only cheaper.
The Wild West
The LED lighting industry, and the LED industry in particular, is currently like the “Wild West”: There aren't many rules, and most people aren't paying attention to them anyway. All sorts of claims are being made that are obviously wrong, and plenty more that you need special equipment to detect.
Looking first at LED device production, we should start out by saying that there are some reputable manufacturers. These tend to be the largest ones, although you can't assume that that's true either. They produce what they say they do, and their datasheet contains information from measurements they've taken. The problems start rather with their marketing departments.
The biggest players are presently in a contest to demonstrate that they have higher luminous efficacy white LEDs than their competitors do. As a result, they routinely release press announcements proclaiming their progress. Now everyone in the industry measures efficacy at a temperature of 25 °C. That's just a given. But actual operation is always at elevated temperatures, since LEDs heat up in operation. And the press announcements never mention how much that wonderful efficacy rolls off at higher temperatures. Different manufacturers' processes have different roll-offs, so you don't know what you would get from this new device. What's more, it's routine to announce results from a single lab device. It's not in production, and very possibly not producible without major changes. So it's all a bit of a cheat.
Moving on, even the big manufacturers tend to have problems with efficacy roll-off with aging, which is to say, lifetime. The truth is that the various manufacturing processes appear to create LEDs that age differently. And the aging varies greatly depending on the drive current, the die temperature, and even the package temperature. The fundamental problem is that 50,000 h is 8 continuous years. There's a new LED process a couple of times a year (recall the 34% increase in efficacy per year). So there isn't time to collect data before the part is obsolete. You would think that you could extrapolate data from, say, the first 1000 h. But the truth is that this works so poorly that the committee writing the specification for LED aging gave up on it. LED lifetime? It's anybody's guess.
Further, many of the LED manufacturers have problems with data. We've seen datasheets for products that have been in production for a year that still have forward voltage copied from a competitor's datasheet. We see efficacy numbers that came from handheld meters. In some cases, the parts don't match the datasheets either in color or in efficacy. The sad story goes on and on. Thus, “Caveat emptor!” The only way to be sure of what you get is to measure it yourself. Read Chapter 14 to find out how.
Moving on now to LED bulb manufacturers, the situation is even worse, if possible. We tested a couple dozen different bulbs. Only 5% of them generated the lumens they claimed, with a majority of them being wildly off! In some cases, it was apparent that no measurement had been made at all. They calculated that each LED is rated at 60 lumens, and they put three of them in the bulb, and so the package says it is 180 lumens! No thought had been given to the drive current, the optics, the packaging, not to mention the temperature effects. The U.S. Department of Energy is making efforts to clean this up. We hope for progress in this area.
We feel that all of these problems are characteristic of an infant industry. Doubtless. all of this will improve. We just hope that consumers aren't so disappointed early on that the industry never gets to maturity.
LEDs and OLEDs and…?
Incandescent bulbs replaced candles and kerosene lamps. Fluorescent tubes replaced incandescent bulbs for many purposes. It seems likely that LEDs will replace both fluorescent tubes and incandescent bulbs. What's next after LEDs?
There's been a lot of talk about OLEDs being the next big thing in lighting. The “O” in the front of the acronym OLED stands for “organic.” But it's really still an LED. The difference is that this particular type uses organic rather than inorganic material. The OLEDs' claim to fame is that they are more mechanically flexible than inorganic LEDs. Perhaps they could be made directly into light bulb shapes or printed onto mechanical forms of light bulb shape.
As we indicated in the section on Haitz's law, LEDs are probably going to reach the maximum theoretical limit for efficacy of any light source. So if OLEDs are going to supplant LEDs, it can't be on the basis of efficacy, because it's impossible to be better. The same is true for any other new light source. Once the theoretical limit is reached, nothing can be better.
The way that OLEDs could supplant LEDs is if they were cheaper. Once there are a variety of possible ways of achieving the maximum efficacy, the market will ensure that the cheapest one is the one that dominates. In our view, OLEDs are really just another type of LED, and their progress is part of Haitz's law. So we don't know if OLEDs or LEDs will prove the eventual cost winner. But our opinion is that there probably won't be any newer technologies for lighting that end up completely replacing LEDs. LEDs will end up being so inexpensive that cheaper won't matter to consumers. We think LEDs are here to stay.
Chapter 2Light Bulbs and Lighting Systems
This book is about lighting design with LEDs. While the rest of the book is about the LED part, in this chapter we present some background on the lighting part. The reason for this is that light bulbs have been around for more than 100 years. In that time, there have been many people working on them, and much technology has been developed. While we can't claim that this is a comprehensive survey, there's probably information in this chapter that you'll be happy you have.
A few words about terminology are in order. Wikipedia1 says that “A lamp is a replaceable component such as an incandescent light bulb, which is designed to produce light from electricity.” As you can see, there is a general confusion about what to call light-producing devices. Most consumers call the device a light bulb, and the unit that holds it a lamp. Manufacturers usually call the device a lamp, and the unit holding it a fixture. In this book, we will usually try to follow consumer usage. But the reader should be aware of the difference when reading publications.
Light Sources
Incandescent
Light-emitting diodes are merely the newest in a long list of different types of lighting devices. Ignoring truly ancient devices such as candles, all of them use electricity. The first and still most common light source is incandescent. An incandescent bulb works by heating a piece of metal, the filament, until it glows. By adjusting the power level, it can be made to glow different colors. The typical incandescent filament runs at about 2850 K, resulting in the familiar yellow color.2 When you dim the bulb it receives less power. This not only produces less light, it also reduces the temperature of the filament. This is why dimmed incandescent bulbs look reddish.
Note that the glass shell in an incandescent bulb is used to maintain a partial vacuum, preventing the filament from oxidizing and failing. There has been some research into altering the mixture of remaining gas in the shell to enhance bulb life.
The incandescent bulb runs very hot. The surface temperature of a 40 W bulb runs about 120 °C. That's why you have to wait a bit after turning it off to touch it. The common failure mode for an incandescent is for its filament to break. This typically happens after about 1000 h of operation. Switching incandescent bulbs on and off a lot can also cause the filament to fail, but in typical operation this is not the dominant reason for failure.
Before leaving the topic of incandescent light sources, a comment on safety is in order. If you unscrew an incandescent bulb from its socket without turning off the light switch, sticking your finger in the socket will connect you with 120 VAC. This is life-threatening. If you try to hold an incandescent bulb when it's on, it will burn you. It's hard to imagine a device with these sorts of extreme problems being introduced today. Conversations with engineers at UL suggest that incandescent bulbs are “grandfathered in.” They were there before regulations existed, and so they can't be easily eliminated. But it certainly seems like the time has come for engineers to come up with something better.
Halogen
Halogen bulbs are also incandescent. The difference between halogen and normal incandescent bulbs is that halogens contain a small amount of halogen. The halogen makes the filament burn hotter, which slightly increases the efficacy of the bulb. It also makes the CCT higher than in a normal incandescent. An additional benefit is that the halogen helps the filament to survive longer (by redepositing the filament material).
Fluorescent
Fluorescent bulbs work entirely differently from incandescent bulbs. They too have a partial vacuum inside a glass tube. In this case, though, the tube intentionally has some mercury vapor in it. When the filament inside the bulb is heated, it emits electrons. These ionize the mercury, forming a plasma arc at about 1100 K. The mercury emits UV light to go back to its normal state. The UV light hits a phosphor coated on the glass tube. This is the white coating on fluorescents. The phosphor absorbs the UV light, and emits visible light, which is the output of the bulb. The phosphor is carefully designed to produce just the color light that is desired, and is usually a mix of different phosphors.
To run this complicated device requires a special circuit called a ballast. The ballast is connected to the AC line as input, normally either 120 or 277 VAC in the United States. At its output the typical one bulb ballast has two pairs of wires. Each pair is heating one of the two bulb filaments. Additionally, current flows from one pair to the other. This latter is the current that produces the plasma arc. Figure 2.1 shows the currents in this lighting system.
Figure 2.1 Currents in a fluorescent tube.
Fluorescent tubes run much cooler than incandescent bulbs. The typical surface temperature is about 40 °C. You can easily touch them and pull them out of their fixture while they're running. For this reason, they typically have an electrical interlock system. If the tube is not present, the ballast typically is designed to turn off to avoid shocking you if you stick your finger into the socket.
