AB20-3_技术文档

application brief AB20-3

replaces AN1149-3

Electrical Design

Considerations for

SuperFlux LEDs

Table of Contents

Overview

Summary

2

Electrical Design Process Overview

Circuit Design Overview

3

4

Worst-Case Circuit Analysis and Validation

Comparison of Different Methods of Worst-Case Circuit Analysis

Characterization of Prototype LED Signal Lamp

Validation of LED Signal Lamp

5

7

Key Concepts for Electrical Design of LED Signal Lamps

Resistor-Limited Drive Circuits

8

Effect of LED String Length on Forward Current Regulation

Forward Current Variations between LED Emitters in an Array

EMC Transient Protection Circuits

9

10

12

Stop/Tail Drive Circuits

Theory

Electrical, Optical, and Thermal Characteristics of LED Emitters

LED Emitter Modeling

13

16

17

18

19

Linear Forward Voltage Model

Luminous Flux Models Versus Forward Current and Temperature

Thermal Resistance Models

Applications

Resistive Current Limiting

CHMSL Design Example

EMC Transient Protection

Special Considerations for Dual Luminous Intensity Operation

Luminous Flux Variations at Low Currents

PWM Drive Circuit for Tail Functions

Current and Voltage Regulator Circuits

Operation of Shunt, Series-Pass, and Switching Regulators

21

24

27

29

30

31

32

Overview

Summary

The electrical design is part of the overall signal

The electrical design of an LED signal lamp has

several objectives. The first objective is to

operate the individual LED emitters at sufficient

drive current in order to generate sufficient

luminous flux to meet the lighting requirements.

The second objective should be to limit the

forward current through the individual LED

emitters so as not to exceed their maximum

internal junction temperatures and maximum dc

forward currents under worst-case conditions of

ambient temperature and input voltage. In

addition, the electrical design should protect the

LED array from automotive EMC transients.

Finally, the electrical design should provide

good intensity matching within the LED array.

The result of achieving these objectives will be a

maintenance-free LED signal light that operates

reliably for the lifetime of the passenger vehicle

or truck.

used in LED signal lamps and related circuit

design issues.

In addition to AB20-3, two companion electrical

design application notes are also available.

AB20-3A, titled “Advanced Electrical Design

Models,” discusses forward voltage models that

are more accurate and usable over a larger range

of forward currents than the simple linear models

shown in the “LED Emitter Modeling” section. In

addition, AB20-3A derives several additional

thermal-modeling equations from the basic

equations shown in the “LED Emitter Modeling”

section.

Application Note titled AB20-3B “SuperFlux LED

Forward Voltage Data” gives worst-case forward

voltage data for SuperFlux and SnapLED70

emitters. There are several potential electrical

models for LED emitters (linear, diode equation

models, etc.), each one optimized for an

Application Brief AB20-3 has been written to

simplify the electrical design of LED signal

lamps and is part of the Application Brief AB20

series. This application note has been divided

into three major blocks—Overview, Theory, and

Applications. The Overview consists of five

sections that discuss the electrical design

process and key electrical design concepts.

Theory consists of two sections that give an

in-depth overview of the electrical, optical,

and thermal properties of LED emitters and

mathematical modeling of their operation.

Applications consist of four sections that cover

specific types of circuit designs commonly

expected range of forward currents, and various

levels of “worst-casing” (i.e. min/max, average,

average ± one, two, three standard deviations,

etc.). In order to accommodate these various

needs, the data presented in AB20-3B gives the

nominal forward voltage and expected forward

voltage range for each SuperFlux LED

characterized over a range of forward currents

up to 70 mA. From this data, the desired

electrical model can easily be generated. The

latest forward voltage data for SuperFlux LED

emitters is available from your local Lumileds

Lighting or Agilent Technologies sales engineer

or from the following URL:

forward voltage can cause luminous intensity

matching variations within groups of

http://www.lumileds.com

SuperFlux emitters from the same luminous

flux and forward voltage category. For this

reason, Lumileds Lighting does not warrant

LED performance at currents less than 20 mA

(40 mA for SnapLED 150) and strongly

discourages these designs.

Note: For best matching within an array,

SuperFlux and SnapLED 70 emitters should

be operated at forward currents over 20 mA

(40 mA for SnapLED 150). At forward

currents below 20 mA (40 mA for SnapLED

150), variations in luminous efficiency and

Electrical Design Process

The electrical design is part of the overall signal

lamp design process described in AB20-1.

exceeded, portions of the electrical or thermal

design may need to be iterated in order to

reduce the forward current and/ or junction

temperature. Finally, additional prototypes of the

signal lamps should be constructed and

subjected to the appropriate reliability validation

tests.

The electrical design consists of several discrete

steps. The first step is to determine the circuit

topology and to generate an electrical

schematic of the overall circuit. First, the circuit

topology must be determined. Circuit topology

refers to the arrangement of electrical

The most important circuit topology

considerations include:

components on the electrical schematic. Next,

the circuit must be designed. Circuit design is

the process where the electrical components

are selected and component values are

• Number of LED emitters in series

• Whether LED emitters are connected in

individual series-strings, or in cross-

connected series strings

determined. Third, the operation of the circuit

must be analyzed. Circuit analysis refers to the

mathematical analysis of the variations in

voltage and current through the electrical

components due to variations in applied voltage

and component tolerances. The fourth step is

to create a breadboard of the circuit and to

measure the forward current, light output, and

thermal properties of the entire signal. If the

maximum junction temperature or the maximum

DC forward current of any of the LED emitters is

exceeded, the reliability of the LED signal lamp

may be compromised. Thus, if the circuit

validation tests indicate that these limits are

• Method of current limiting (i.e. resistors or

active circuit)

• Method of EMC protection (if any)

• Method of dimming, such as for a combined

Stop/ Tail signal

Circuit design is the solution of several

simultaneous linear equations that model the

forward current through each loop or node of the

circuit. The solution of these equations

determines the values of electrical components

that drive the LED array at the desired forward

current at the specified external supply voltage.

3

Different mathematical models can be used to

model the forward voltage of the LED emitters

depending on the accuracy and dynamic range

needed.

and LED emitters. With this technique, the

voltages and currents through each component

can be determined based on the expected

minimum and maximum limits of component

values in the circuit. Another technique is to use

a Monte Carlo simulation. With this method the

voltages and currents through each component

are determined for random combinations of

component values in the circuit. Then the results

for a large number of Monte Carlo simulations

are statistically tabulated. The Monte Carlo

simulation method gives a better estimate of the

expected manufacturing variations for the circuit.

Circuit analysis uses the same types of

simultaneous linear equations used in the

electrical design. However, circuit analysis

generally assumes that the external voltage

applied to the circuit, the values of electrical

components, and the ambient temperature can

vary over some predetermined range. Circuit

analysis can be done using “worst-case”

electrical models for the electronic components

Circuit Design Overview

As stated earlier, the first step in the electrical

design is to pick one of the circuit topologies.

Next, the operation of the circuit can be

modeled with a series of simultaneous linear

equations that describe the current through

each electronic component as a function of

component values and applied voltage. For

circuit design, it is usually assumed that all

LED emitters have the same electrical

characteristics, which greatly simplifies the

mathematical modeling.

As will be shown in the section “Key Concepts

for the Electrical Design of LED Signal Lamps”

the different circuit topologies provide different

levels of forward current regulation and overall

electrical power consumption. Circuits with poor

forward current regulation would require the LED

emitters to be driven at a lower forward current

at the nominal input voltage than would circuits

with better forward current regulation (so as not

to exceed the maximum forward current at the

maximum input voltage). Circuits with higher

amounts of power consumption would tend to

have higher internal self-heating (unless the

circuitry is located outside the signal lamp

In order to ensure reliable operation, the

maximum forward current through the LED

emitters should not exceed the maximum value

obtained from Figure 4 in the HPWx-xx00 data

sheet. Note that the maximum forward current

of the LED emitters is based on the maximum

ambient temperature, T_A, the maximum input

voltage, and the thermal resistance, R q_JA, of the

LED signal light assembly.

housing), which would also tend to reduce the

maximum forward current of the LED emitters. At

this point, it may be desirable to evaluate several

different circuit topologies on paper and see

which one gives the “best” overall results.

If the signal lamp will be exposed to high-voltage

EMC transients, then the appropriate protection

circuitry can be added to the basic circuit

4

chosen. Worst-case forward and reverse

transient currents can be estimated using the

linear forward current model for positive

transients and the minimum breakdown voltage

specification, V_BR, on the data sheet for negative

transients.

Worst-Case Circuit Analysis and Validation

The next step in the electrical design should be

an analysis of the forward current through the

LED emitters at worst-case input voltage and

operating temperature extremes using worst-

case component tolerances. All of the active

and passive electronic components used in the

circuit design can be modeled with their worst-

case minimum and maximum values. This

analysis serves several purposes. First, it

determines whether the forward current is less

than the maximum dc forward current under all

operating conditions. Secondly, it determines

the change in light output of the signal lamp

under the same conditions. Finally, it can be

used to determine the worst-case matching

within the LED array.

circuit. Thus, the actual occurrence of these

worst-case conditions could be extremely small.

Another approach to worst-case analysis is to

characterize a number of LED emitters and

determine the appropriate forward voltage

model for each one. Then using a Monte-Carlo

simulation, random combinations of these

emitters can be assembled into a “paper”

circuit and the actual forward currents can

be calculated for the circuit based on the

corresponding forward voltage models. Then

the results from multiple simulations can be

tabulated. This approach provides a much better

understanding of the forward current variations

that would occur in actual practice.

This worst-case analysis can be done in several

different ways. One approach is to use worst-

case values for one or more LED emitters in the

array such as to cause worst-case current

matching between LED emitters or to establish

the maximum or minimum forward current

through individual LED emitters. The problem

with this approach is that probability of this

occurrence actually happening can be quite

low. If the probability of getting worst-case LED

emitters is very small, then the probability of

both minimum and maximum worst-case LED

emitters occurring in the same circuit assembly

is even lower. Furthermore, for the worst-case

variations in forward currents to actually occur,

these worst-case LED emitters must both be

randomly assembled into certain parts of the

In general, within arrays of LED emitters, the

maximum forward current occurs at the

maximum input voltage with the minimum value

of the current limiting resistor and minimum

forward voltage model for the LED emitters.

Likewise, the minimum forward current occurs

at the minimum input voltage with the maximum

value of the current limiting resistor and

maximum forward voltage model for the LED

emitters.

The worst-case forward current variations

for different LED emitters within the array is

determined by the circuit topology, the drive

current, and the variation in electrical

characteristics of the individual LED emitters

in the array. When several LED emitters are

5

connected in series, the worst-case minimum

forward current would occur when all LED

emitters in a given series string have the worst-

case maximum forward voltage. Likewise, the

worst-case maximum forward current would

occur when all LED emitters in another series-

string have the worst-case minimum forward

voltage. For these series-string circuits, the

likelihood of all LED emitters being at their

worst-case forward voltage extremes is quite

low. When LED emitters are connected in

parallel, the forward current through each LED

will vary somewhat from the average forward

current so as to generate the same forward

voltage across all LED emitters in the parallel

grouping. The worst-case forward current

variations occur when one LED emitter has

the worst-case minimum forward voltage and

another LED emitter in the same parallel

grouping has the worst-case maximum

forward voltage.

range of 35 to 70 mA (70 to 150 mA for the

SnapLED 150). These matching effects are

covered in more detail in the section “Key

Concepts for Electrical Design of LED Signal

Lamps.”

The reader will need to determine whether the

assumptions used for the worst-case designs are

reasonable. It is possible to design with such

large tolerances, that the worst-case design

results in an over-designed circuit. Over-

designing occurs if significant cost is added to

the assembly in order to protect against the

remote possibility of occurrences that might

never happen in practice. In the case of LED

signal lamps, over-designing might result in many

more LED emitters being added to the array than

needed. The best example might be where the

designer chooses an extremely high worst-case

input voltage and maximum ambient

temperature. Then, the use of the suggested

design process in the “Resistive Current Limiting”

section would result in a fairly small design

current at the design input voltage. This design

current would require a large number of LED

emitters to achieve the desired light output from

the array. Or the assumptions used for the worst-

case input voltage and ambient temperature

might require the use of a more expensive

constant current drive circuit, where with more

reasonable assumptions a resistive circuit could

have been used. These concerns about over-

designing by using excessive tolerances on input

voltage and ambient temperature also can be

applied to LED emitter tolerances. For example,

the probability of all LED emitters in a given array

being at their worst-case minimum or maximum

limits is very small but still is greater than zero. If

every LED emitter is assumed to be at the worst-

case minimum extreme, then the external current

SuperFlux and SnapLED 70 emitters are

categorized for forward voltage at 70 mA.

As might be expected, the smallest forward

current variations within an array of SuperFlux

or SnapLED 70 LED emitters occur at drive

current approaching 70 mA. Similarly, SnapLED

150 emitters are categorized for forward voltage

at 150 mA, so the best matching occurs at 150

mA. At lower forward currents, the variations in

forward current within the LED array become

larger—especially when LED emitters are

connected in parallel. For series-string circuits,

acceptable forward current variations can

usually be achieved over forward currents over

a range of 20 to 70 mA (40 to 150 mA for the

SnapLED 150). However, when LED emitters

are connected in parallel, acceptable forward

current variations can be achieved only over a

6

limiting resistor may be chosen to overly restrict

the forward current through the array. Another

possibility is that if the circuit is designed to

accommodate every LED emitter in a given

array being at their worst-case minimum or

maximum limits, then design might eliminate

several potential circuit topologies because of

excessive light output variations. In practice,

the likelihood of these conditions actually

occurring is so small, that the design process

might have eliminated more cost effective circuit

topologies. For this reason, Lumileds Lighting

recommends that worst-case design be used

as a development tool in conjunction with

characterization and validation of the signal

lamp assembly.

optical transmission losses. The prototype allows

these assumptions to be measured. Third, the

working prototype allows the thermal properties

of the LED signal light to be evaluated. The

thermal resistance, Rq_{PIN A}, can be measured by

attaching thermocouples to the cathode pins of

several LED emitters in the array.

Based on the electrical, optical, and thermal

measurements of the working prototype,

additional iterations of the design may be

required. These design iterations will further

refine the estimates for the electrical component

values, the number of LED emitters needed for

the signal lamp, and the thermal resistance of

the signal lamp. Also, it may be necessary to

evaluate improved optical designs and methods

for improving the thermal properties of the LED

signal lamp assembly. For more information on

thermal design, the reader is encouraged to

review AB20-4 “Thermal Management

Following this paper design, a working

prototype LED signal light can be constructed.

This prototype serves several purposes. First,

it allows verification of the electrical design. The

forward current can be measured at different

input voltages and compared with the paper

electrical design. Second, it allows verification

of the optical design. In AB20-1 section,

“Estimating the Number of LED Emitters

Needed,” assumptions were made for the

Considerations for SuperFlux LEDs.” For

more information on optics design, the reader

is encouraged to review AB20-5 “Secondary

Optic Design Considerations for SuperFlux

LEDs.”

7

Key Concepts for Electrical Design of LED Signal Lamps

Presently, most current LED signal lamp

designs use resistive current limiting for the LED

array. Since most LED signal lamps are driven

from 12 to 24 V dc and require several LED

emitters, the emitters can be connected in

series and share the same supply current.

Some of the most common circuit

Where:

V_IN= input voltage applied to the circuit

V_F= forward voltage of LED emitter at forward

current I_F

V_D= voltage drop across optional reverse

transient EMC protection diode

y = number of series connected LED emitters

x = number of paralleled strings

configurations are shown in Figure 3.1. The

series connected string circuit in Figure 3.1a

uses a separate current limiting resistor for each

string of y LED emitters. The paralleled-string

circuit in Figure 3.1b uses a single resistor for

the entire LED array. Note that this circuit uses x

strings with y LED emitters per string. The

cross-connected paralleled string circuit shown

in Figure 3.1c has one or more cross

Thus for a given I_F, the value of R depends both

on the number of LED emitters per string as well

as the number of paralleled strings.

As shown later in the section “LED Emitter

Modeling” the forward voltage of an LED emitter

can be mathematically modeled by the following

equation:

connections between the strings of LED

emitters. A mechanical analogy to this circuit is

that the circuit looks like a “ladder” with each

cross connection being a “rung” on the

V_F@?V _O+ R_SI_F

“ladder.” In the diagram, z, refers to the number

of series connected LED emitters between each

“rung” with 1 £ z £ y. Most CHMSL designs

use either several series connected strings

(Figure 3.1a) or several cross-connected series

strings (Figure 3.1c) with z =1. Note that in

order to obtain the same forward current for

all of LED emitters, both of the circuits

shown in Figure 3.1b and Figure 3.1c need

to have the same number of LED emitters in

each string and each “rung” for Figure 3.1c.

Where:

V_O= turn-on voltage of each LED emitter

R_S= series resistance of each LED emitter

This equation is known as the linear forward

voltage model since it models the forward

voltage of the LED with a straight line. In using

this model it is important to remember that it can

only be used over a specific range of forward

currents. Outside of this range, the model will

give misleading results. Using the linear forward

voltage model, this equation can be rewritten as:

Note that if all LED emitters have identical

electrical forward characteristics, then the value

of the external current limiting resistor, R, is

equal to:

8

The number of LED emitters selected per

string (y) affects the change in the forward

current of the LED emitters as the input

voltage varies over some range. In general,

as the value of y is increased, the change in

forward current becomes larger due to the

same input voltage variation. In addition, when

the input voltage is less than (yV_O) the forward

current through each LED emitter is

through each string of LED emitters as a function

of input voltage. Figure 3.2 shows string lengths

ranging from 2 to 6 emitters. Note that series-

strings with 5 and 6 emitters have the largest

change in forward current and highest “threshold

voltage.” Since the forward voltage of the LED

emitters varies slightly over temperature, then the

forward current through each LED string will

change slightly over temperature. Circuits with

longer string lengths will have a slightly larger

forward current change over temperature.

approximately equal to zero. As the value of y is

increased, this “threshold” voltage increases.

Figure 3.2 shows the change in forward current

Figure 3.1 Circuit Configuration of LED Arrays Used in LED Signal Lamps.

9

Figure 3.2 Forward Current Through String of

HPWA-xHOO LED Emitters versus Applied

Voltage.

Figure 3.3 Total Supply Current versus Supply

Voltage for Sixty HPWA-xHOO Emitter LED

Signal Lamp Array Driven in Series Strings

with Two, Three, Four, Five, or Six Emitters

per String.

The number of LED emitters per string also

affects the total supply current. For a fixed

number of LED emitters longer string lengths

result in fewer total strings and thus a lower

total supply current. Figure 3.3 shows the

total supply current for the series-string

configurations in Figure 3.2. Figure 3.3 assumes

a total of 60 LED emitters, thus 2-LED strings

would require 30 strings, 3-LED strings would

require 20 strings, etc. Note that the total

supply current is much higher for series strings

with 2 and 3 emitters.

Variations in the forward voltage characteristics

of the individual LED emitters can lead to forward

current variations within the LED array. These

forward current variations directly affect the

luminous flux output of each LED emitter and can

cause noticeable luminous intensity variations

within the LED array. These random variations in

forward voltage characteristics affect the three

circuits shown in Figure 3.1 differently.

The “series-string” circuit shown in Figure 3.1a

is least affected by random forward voltage

variations between the LED emitters because the

forward voltages of all LED emitters in a given

string are averaged together. In many cases, one

emitter with a high forward voltage can cancel

out another emitter in the same series-string with

a low forward voltage. In addition, the voltage

drop across the current limiting resistor, R, is

much higher than the combined voltage drops

across the series resistors, yR_S. Thus, small

variations in R_Sonly cause small variations in the

forward current through the series string.

As shown by Figures 3.2 and 3.3, the choice of

the number of emitters per string is a tradeoff

between the regulation of forward current due

to input voltage variations and the total supply

current for the LED array. Small string lengths

give excellent forward current regulation, but

require a higher supply current. Long string

lengths provide poor forward current regulation

but require less supply current. For these

reasons, most 12 V resistive limited designs

use three or four LED emitters per series

string (y = 3 or 4).

Because SuperFlux and SnapLED 70 emitters

are categorized for forward voltage at 70 mA

(150 mA for the SnapLED 150), the smallest

10

Figure 3.4 Worst-Case and Typical Variations

in Forward Current Between Two Strings of

HPWT-xHOO LED Emitters Driven with

Individual Current-Limiting Resistors per

String (Figure 1a Circuit, with y=4).

Figure 3.5 Worst-Case and Typical Variations

in Forward Current Between 16 HPWT-xHOO

LED Emitters Driven in a Cross-Connected

Paralleled String Configuration (Figure 1c

Circuit with x = y = 4, z = 1).

forward current variations between adjacent

series-strings of LED emitters occur at a drive

current of 70 mA (150 mA for the SnapLED

150). However, the forward current matching

between adjacent strings is quite good even at

forward currents as low as 10 mA (20 mA for

the SnapLED 150). Figure 3.4 shows the worst-

case forward current variation between two LED

strings constructed using four LED emitters per

string. The worst-case calculations assume that

all HPWT-xH00 emitters are from the same

forward voltage category and one string uses

four “minimum” forward voltage emitters and

the other string uses four “maximum” forward

voltage emitters. The typical calculations

SuperFlux and SnapLED 70 emitters and 40

mA for SnapLED 150 emitters (see the section

“Electrical, Optical, and Thermal

Characteristics of LED Emitters).

The “paralleled-string” circuit shown in Figure

3.1b and the “cross-connected series string”

circuit shown in Figure 3.1c do not regulate the

forward current as well as the “series-string”

circuit. When two or more LED emitters are

connected in parallel, the forward current

through each emitter will be somewhat higher or

lower than the average forward current through

them so as to force the forward voltage across

them to be the same. Again, because SuperFlux

LED and SnapLED 70 emitters are categorized

for forward voltage at 70 mA (150 mA for the

SnapLED 150), the smallest forward current

variations between adjacent LED strings (Figure

3.1b circuit) or between LED emitters in the

same “rung” (Figure 3.1c circuit) occur at a drive

current of 70 mA (150 mA for SnapLED 150).

The variations in forward currents become much

worse at lower drive currents. These variations in

forward currents can cause unacceptable

assume that all HPWT-xH00 emitters are from

the same forward voltage category and one

string consists of two “minimum” forward

voltage emitters and two “typical” forward

voltage emitters and the other string consists of

two “maximum “forward voltage emitters and

two “typical” forward voltage emitters. Due to

potential variations in luminous flux output at

low currents, Lumileds Lighting recommends

a minimum forward current of 20 mA for

11

luminous intensity variations even using LED

emitters from the same forward voltage and

luminous flux category. Figure 3.5 shows the

worst-case forward current variations within the

LED array when the array is constructed using

LED emitters from the same forward voltage

category. The worst-case calculations assume

that the LED array consists of 16 HPWT-xH00

emitters constructed using four “minimum” LED

emitters, four “maximum” LED emitters, and

eight “typical” LED emitters. Then each

SuperFlux LED emitters from only one forward

voltage category within the same LED array.

paralleled grouping consists of one “max,” one

“min,” and two “typical” LED emitters. The

typical calculations assume that the LED array

consists of 16 HPWT-xH00 emitters

Figure 3.6 Typical EMC Transient Protection Circuits

for LED Signal Lamps.

constructed using two “minimum” LED emitters,

two “maximum” LED emitters, and twelve

“typical” LED emitters. Then two of the

LED emitters are susceptible to permanent

damage due to high voltage automotive EMC

transients. The addition of a high-voltage

silicon diode in series with the LED array can

effectively protect the array from high-voltage

negative transients. The LED array can be

protected from positive “Load Dump”

paralleled groupings consist of one “max,” one

“min,” and two “typical” LED emitters, and the

other two paralleled groupings consist of four

“typical” LED emitters. Lumileds Lighting

recommends a minimum forward current of

35 mA (70 mA for the SnapLED 150), for the

“paralleled-string” circuit in Figure 3.1b or the

“cross-connected parallel-string” circuit

shown in Figure 3.1c. At drive currents less

than 35 mA (70 mA for SnapLED 150), the

“worst-case” forward current variations

between adjacent LED emitters can exceed

2:1. Because of the averaging effects of several

series-connected LED emitters, the circuit in

Figure 3.1b has somewhat lower typical forward

current variations than the circuit shown in

Figure 3.1c. Note that the forward current

matching can be improved with the addition of

a small resistor (R_OPT> R_S) in series with each

string for the circuit shown in Figure 3.1b or

“rung” for the circuit shown in Figure 3.1c. For

these circuits, it is important to use

transients with the addition of a transient

suppressor connected in parallel with the LED

array. Figure 3.6 shows the addition of EMC

protection circuitry to the LED array. EMC

transient protection is covered in more detail in

the following section “EMC Transient Protection.”

Some applications require the LED array to

operate at two levels of luminous intensity (i.e.

a rear Stop/ Tail signal). Generally, it is desirable

that the LED emitters should appear matched at

both drive conditions. SuperFlux and SnapLED

70 emitters are categorized for luminous flux at

70 mA (150 mA for the SnapLED 150). As shown

in the section “Electrical, Optical, and Thermal

Characteristics of LED Emitters,” the light output

matching for random combinations of LED

emitters gets progressively worse at lower

12

forward currents. For SuperFlux and SnapLED

emitters, the light output varies by a factor of

2:1 at a forward current of 20 mA (40 mA for

the SnapLED 150). Thus, even if all of the LED

emitters are driven at the same forward current,

there would likely be unacceptable light output

matching if the Tail signal is driven at a low DC

forward current.

reduced luminous intensity. The Stop signal

might operate the LED array at a high DC

forward current. Then for the Tail signal, the

array would be operated at the same peak

forward current with a low duty cycle (ratio of

“on” time to “on” plus “off” time). This

approach provides light output matching under

both levels of luminous intensity. A

recommended PWM circuit is shown in the

section “Special Considerations for Dual

Luminous Intensity Operation.”

For best matching, it is recommended that a

pulse width modulation (PWM) circuit be

designed to operate the signal lamp at

Theory

Overview of Electrical, Optical, and

Thermal Characteristics of an LED Emitter

In order to properly design an LED signal light,

it is important to have a basic understanding

of the electrical, optical, and thermal

the incremental forward voltage. Although this

graph has been used traditionally to describe the

forward characteristics of a diode, in reality LED

emitters are best thought of as current controlled

devices, not voltage controlled devices. For an

LED emitter, the optical properties are best

described as a function of current, not a function

of voltage. In addition, operation of the LED

emitter at a constant current gives the best

control of light output. In contrast, operation of

the LED emitter at a constant voltage allows a

larger variation in forward current and light output

from device to device.

characteristics of an LED emitter.

The typical forward current (I_F) versus forward

voltage (V_F) characteristic for an AlInGaP LED

emitter under positive (forward) bias is shown in

Figure 3.7. On a linear scale of forward current

versus forward voltage, negligible current flows

until a threshold voltage, also known as the

turn-on voltage (V_O), is exceeded. Above this

voltage, the current increases proportionally to

Figure 3.7 Typical Forward Current versus

Forward Voltage for HPWA-xHOO LET Emitter

(Linear Scale).

Figure 3.8 Typical Forward Current versus Forward

Voltage for HPWA-xHOO LED Emitter (Semi-Long

Scale.)

13

The low-current forward characteristics of the

same AlInGaP LED emitter are shown in Figure

3.8. This graph shows the forward voltage

versus the log of forward current. Note that a

small current flows through the emitter even at

low forward voltages below the turn-on voltage

shown in Figure 3.7. Due to the high optical

efficiency of AlInGaP material, a perceptible

amount of light is generated from the LED

emitters at forward currents as low as 10 mA.

Thus the inadvertent operation at low forward

currents can cause “ghosting” within an “off”

LED signal light.

HPWT-xx00 emitters. Operation of the LED

emitter in the reverse current region is not

recommended. Reverse currents in excess of 50

mA can cause permanent damage to the LED

junction, as discussed later in the section titled

“Electrical Transients.” The reverse breakdown

voltage is essentially constant over the –40ºC to

100ºC temperature range.

The change in luminous flux (F_V) as a function of

forward current (I_F) of an AlInGaP LED emitter is

shown in Figure 3.10. Note that the change in

luminous flux is roughly proportional to the

change in forward current. At forward currents

over 20 mA, the luminous flux increases at a

lower rate due to internal heating within the LED

emitter. The change in luminous flux due to a

change in forward current ( DF_V/ DI_F) varies

somewhat from unit to unit. Figure 3.11 shows

the expected range in light output for HPWT-

xH00 emitters that were matched at 70 mA.

Note that the light output varies by a factor of 2:1

at a 20 mA forward current. Since the SnapLED

150 emitter is matched at 150 mA, then the light

output can be expected to vary by a factor of 2:1

at a forward current of 40 mA.

The forward voltage of an AlInGaP LED emitter

changes by about –2 mV per °C over

temperature. Thus, the forward voltage at a

given current is slightly lower at elevated

temperatures and slightly higher at colder

temperatures.

The reverse characteristics of an AlInGaP LED

emitter are shown in Figure 3.9. Note that a

negligible amount of reverse current (< 1 mA)

flows through the LED until the reverse

breakdown voltage is reached. The reverse

current increases quickly at voltages higher than

the reverse breakdown voltage (defined as the

voltage across the LED at which the reverse

current reaches 100 mA). The reverse

The luminous flux of an AlInGaP LED emitter

varies inversely with temperature as shown in

Figure 3.12.

breakdown voltage for AlInGaP LED emitters is

typically in the range of 20 V. However, it can

be as low as 10 V for the HPWA-xx00 and

14

Figure 3.9 Typical Reverse Current versus

Reverse Voltage for HPWA-xHOO LED Emitter

(Linear Scale).

Figure 3.10 Typical Luminous Flux versus

Forward Current for HPWA-xHOO LED Emitter

(Logarithmic Scale).

At 85°C, the light output will be approximately

50% of the light output at 25°C. At –40°C, the

light output will be approximately twice the light

output at 25°C. This change is fully reversible.

and dominant wavelength over temperature are

fully reversible when the ambient temperature

returns to 25ºC.

As might be expected, the junction temperature

varies directly as a function of ambient

The peak and dominant wavelength of an

AlInGaP LED emitter changes by about 0.1 nm

per ºC. Thus, the color of the LED shifts slightly

toward the red at elevated temperatures.

temperature. In addition, the junction

temperature of the LED emitter is hotter than the

surrounding ambient temperature due to the

internal power dissipation (I_FV_F) within the LED

emitter. Figure 3.13 shows the internal

By now, it should be apparent that a number of

the electrical and optical characteristics of an

LED emitter vary as a function of ambient and

junction temperature. All of these changes in

forward voltage, luminous intensity, and peak

temperature rise, T_J– T_A, for an LED signal lamp

over a range of thermal resistance, as a function

of the forward current through the LED emitter.

Figure 3.11 Expected Variations in Luminous Flux

versus Forward Current for HPWT-xHOO LED

Emitters (Logarithmic Scale).

Figure 3.12 Typical Luminous Flux versus

Temperature for HPWT-xHOO LED Emitter

Driven at 60 mA (Linear Scale).

15

Figure 3.13 Expected Internal Temperature Rise

(T_J-T_A) for HPWA-xHOO LED Emitter versus

Forward Current (Linear Scale).

Figure 3.14 Straight-Line Forward Voltage Model

for LED Emitter (Linear Scale).

Besides affecting a number of the electrical and

optical parameters, the junction temperature

and maximum operating current also affect the

reliability of the LED emitter. For AlInGaP

SuperFlux LED emitters, operation of the emitter

near the maximum operating temperature limit

and maximum operating current limit can result

in a small amount of light degradation over time.

In addition, due to the different rates of thermal

expansion of the epoxy material used for the

emitter package and the metal pins and gold

bond-wire within the LED emitter, there are

upper and lower limits to the operating and

storage temperature ranges for each LED

package. Exceeding these limits, especially for

hundred’s of temperature cycles, can result in

premature catastrophic LED emitter failures.

These effects are covered in detail in AB20-6,

titled “Reliability Considerations for SuperFlux

LEDs.”

For these reasons, it is important to understand

the thermal properties of the individual LED

emitter as well as the thermal properties of all the

elements of the LED signal lamp (printed circuit

board, case, etc). AB20-4, titled “Thermal

Management Considerations for SuperFlux

LEDs” discusses the proper thermal design of an

LED signal lamp. In this application note, thermal

modeling will be covered only in sufficient detail

so as to allow proper circuit modeling to maintain

the LED junction temperature within the

recommended operating temperature limits. This

application note will also use thermal modeling to

estimate the change in electrical and optical

parameters of the LED emitters over temperature

and how these effect the operation of the LED

signal light. For additional detail on thermal

modeling please refer to AB20-4.

LED Emitter Modeling

The purpose of modeling is to represent the

electrical, optical, and thermal characteristics of

a component with equations that allow their

interactions with other electronic components to

be expressed mathematically. The process of

modeling also requires that mathematical

16

expressions be selected that best approximate

the actual measured data.

Thus, the equations for V_Oand R_Scan be written

as:

For operation over a restricted range of current,

say from 30 mA to 70 mA, the forward current

can be modeled with a linear model. As shown

in Figure 3.14, the linear model draws a straight

line between two points (I_F1, V_F1) and (I_F2, V_F2) at

two forward currents, I_F1< I_F2, to linearize the

electrical forward characteristics between these

forward currents. The linear model is shown

graphically in Figure 3.15 for the forward voltage

versus forward current curve shown in Figure

3.7. The equation for the forward current

becomes:

For most applications this linear model can be

used to model the forward characteristics of an

LED emitter. For best accuracy, the use of the

linear model should be restricted to a range of

forward currents, I_F2/ I_F1, less than 4:1. For

operation at a lower range of currents, different

points (I_F3, V_F3) and (I_F4, V_F4) can be selected to

bracket the approximate range of operating

current. However, it’s always important to

recognize that the linear model only works for

a specified range of forward currents (I_F1£?I_F

£?I_F2) as the accuracy of the linear model

degrades quickly outside of this range. It

should go without saying that the linear model

cannot be used at all for values of V_F< V_O.

Where:

V_O= turn-on voltage, the y-intercept of the

straight line (I_F= 0)

R_S= series resistance, the slope of the straight

line

Figure 3.15 Linear Forward Voltage Model for

HPWA-xHOO LED Emitter Shown in Figure 7.

Figure 3.16 Worst-Case Linear Forward Voltage

Models for LED Emitters.

17

Using the nominal forward voltage at the two

test currents in Equations #3.4 and #3.5 would

generate the typical linear forward voltage

model as shown below. The nominal linear

forward voltage model (V_{O nom}and R_{S nom}) is

based on the average forward voltages at two

test currents, I_F1and I_F2, for a large number of

SuperFlux LED emitters from the same forward

voltage category.

estimated with two permutations of the linear

model as shown in Figure 3.16:

V_{F min}= V_{O LL}+ R_{S LL}I_F@ V_{O min}+ R_{S min}I_F

V_{F max}= V_{O HH}+ R_{S HH}I_F@ V_{O max}+ R_{S max}I_F

In order to model the variation in electrical

forward characteristics over temperature,

another term can be added to the linear model

as shown in Equation #3.6. Note that the data

shown in AB20-3B represents the forward

voltage at 25°C with the units measured cold (i.e.

T_J= 25°C). Thus, the thermally stabilized forward

voltage at 25°C will be slightly lower than the

values shown in AB20-3B.

(I_F1, V_{F1 nom}), (I_F2, V_{F2 nom}) Þ ( V_{O nom}, R_{S nom}

)

Then:

V_{F nom}= V_{O nom}+ R_{S nom}I_F

The values of V_F(I_F1) and V_F(I_F2) vary for different

SuperFlux LED emitters from the same forward

voltage category. Statistical forward voltage

data for SuperFlux LED emitters is given in

AB20-3B. Then, the values of V_Oand R_Scan be

calculated using the desired limits (i.e. V_{F max}, V_F

_min, or V_{F average}± n s ). Worst-case circuit analysis

is concerned primarily with the highest and

lowest forward voltages over the range of I_F1£ I_F

£ I_F2. In most cases, the worst-case range of

forward current and forward voltage can be

Where:

T_J

= junction temperature, °C

D V_F/D = change in V_Fdue to temperature,

@ –2mV°C

V_O, R_S= measured at a junction temperature

of 25°C

Figure 3.17 Linear Model (m = 1) for

Luminous Flux versus Forward Current for

HPWA-xHOO LED Emitter Shown in Figure

3.10.

Figure 3.18 Exponential Model (k = -0.0110)

and Exponential Curve Fit (k = -0.0096) for

Luminous Flux versus Temperature for

HPWA-xHOO LED Emitter Shown in

Figure 3.12.

18

In general, the luminous flux output of LED

emitters varies as a function of the forward

current. Ignoring the effect of heating, the

relationship between luminous flux and forward

current can be modeled with the following

equation:

F_V(T_J) = luminous flux at forward current I_Fat

junction temperature, T_J

F_V(T_J= 25°C) = luminous flux at 25°C, without

heating

k

= thermal coefficient, k @ –0.01

(3.7)

Over the automotive operating temperature

range of –40°C to 85°C, this model matches the

actual data within ± 10%. Figure 3.18 shows

how the modeled data for F_Vas a function of

temperature compares to the actual data shown

in Figure 3.12. Note that the value selected for k

was chosen to improve the curve fit at elevated

temperatures than at temperatures below 25°C.

Typical values of k for AlInGaP and TS AlGaAs

SuperFlux LED emitters are shown in Table 3.1.

m

F_V(I_F,T_J= 25°C) @ F _V(I_{F TEST},T_J= 25°C)[I_F/I_{F TEST}

]

Where:

F_V(I_F,T_J= 25°C) = Luminous flux at forward

current, I_F, ignoring heating

F (I_{F TEST},T_J= 25°C) = Luminous flux at test

V

current, I_{F TEST}, ignoring heating

= forward current

I_F

I

_{F TEST}= forward current at data sheet test

conditions

m

= linearity factor, 1 £ m £ 2

Thermal resistance is a measurement of the

temperature rise within the LED signal lamp

caused by internal power dissipation as well as

other sources of heat in close proximity to the

LED (i.e. bulbs, resistors, drive transistors, etc).

For a detailed discussion of thermal resistance,

please refer to AB20-4. The units of thermal

resistance are ºC/W. For the same power

dissipation, the LED signal lamp with a higher

thermal resistance would have a larger internal

temperature rise. The basic thermal modeling

equation is shown below:

At forward currents less than 10 mA, m » 1.3 for

AlInGaP LED emitters. At forward currents over

30 mA, the linearity factor, m » 1.0 for AlInGaP

LED emitters. Figure 3.17 shows how the

modeled data for F_Vversus I_Fcompares to the

actual data shown in Figure 3.10.

For operation at forward currents over 30 mA,

Equation #3.7 can be simplified into a simple

linear equation:

F_V(I_F, T_J= 25°C) @ F_V(I_{F TEST}, T_J= 25°C)[ I_F/ I_F

T_J@ T_A+ Rq_JAP_D

(3.9)

]

TEST

Where:

The luminous flux varies exponentially with

T_J= internal junction temperature within the LED

emitter, °C

temperature. The simplest model is shown

below:

T_A= ambient temperature surrounding the LED

signal lamp, °C

(3.8)

F_V(T_J) @ F _V(T_J= 25°C) exp [k(T_J-25°C)]

Where:

Rq_JA= thermal resistance, junction to ambient,

°C/W

19

P_D= internal power dissipation within the LED

LED signal lamps typically use several LED

emitters. Each LED has a slightly different

thermal resistance, based on the proximity of

other heat sources (e.g. adjacent LED emitters,

resistors, power transistors, bulbs, etc) and

printed circuit board layout. Generally, the

thermal resistance value used for thermal

modeling is the highest thermal resistance,

Rq_JA, of any of the LED emitters within the LED

lamp assembly. Experience has shown that the

LED emitter with the highest thermal resistance

is usually either one of the emitters in the center

of the LED lamp assembly for an x-y

emitter (I_FV_F),W

Where:

Rq _JP= thermal resistance, junction to pin (LED

emitter package), °C/W

Rq _PA

= thermal resistance pin to air (printed

circuit and case), °C/W

T_P= LED cathode pin temperature on

underside of printed circuit board, °C

These equations are especially useful since the

thermal resistance junction to pin, Rq_JP, is

specified on the product data sheet and the LED

pin temperature can be measured directly by

attaching a thermocouple on the cathode lead of

the LED emitter on the underside of the printed

circuit board. Thus the junction temperature can

be estimated based on a measurement of the pin

temperature of the LED emitter (please refer to

AB20-4).

arrangement of emitters, the middle emitter in a

single row of emitters, or one of the emitters

adjacent to other heat sources (e.g. resistors,

power transistors, bulbs, etc).

The thermal modeling equation can be further

broken down by separately considering the

thermal resistance of each of the elements of

the LED signal lamp as shown below:

Usually thermal resistance measurements are

done at thermal equilibrium. For an LED signal

lamp, thermal equilibrium usually occurs after

30 minutes of continuous operation. In some

cases, it is important to calculate the junction

temperature under a transient condition

(e.g. 2 minutes at 24 V).

T_J@ T_A+ (Rq _JP+ Rq_PA) P_D

T_P@ T_A+ (Rq _PA) P_D

(3.10)

Table 3.1

Values of k for AlInGaP SuperFlux LED Emitters

Coefficient of

Dominant

Family

LED Material

Wavelength

F V, ( T ), k

HPWA-xHOO

HPWA-xLOO

HPWT-xDOO

HPWT-xHOO

HPWT-xLOO

AS AlInGaP

TS AlInGaP

618 nm

592 nm

630 nm

620 nm

594 nm

-0.0106

-0.0175

-0.0106

-0.0175

20

This can be done by further subdividing the

thermal modeling equation as shown below:

die will be the first element of the model to heat-

up, followed by the LED emitter package, and

then the rest of the LED signal light.

T_J@ T_A+ (Rq_{J LF}+ Rq_{LF P}+ Rq_{P A}) P_D(3.11)

This section discussed the key concepts of

modeling the electrical, optical, and thermal

performance of LED signal lights. Equations #3.3

and #3.7 can be used to model the operation of

an LED emitter at room temperature, ignoring

the effects of self-heating. Equations #3.6, #3.7,

#3.8, and #3.9 can be used together to model

the effects of self-heating of an LED emitter at

room temperature as well as to model the

operation of an LED emitter over temperature.

Equations #3.10 and #3.11 show the various

components of the overall thermal resistance,

Rq_JA, which can be useful in the thermal

modeling of an LED signal lamp assembly and

the thermal modeling of transient power

conditions.

Where:

Rq _{J LF}= thermal resistance, junction to lead

frame (LED die), °C/W

Rq _{LF P}= thermal resistance, lead frame to pin

(LED package excluding die), °C/W

Please note that each thermal resistance (Rq _{J LF}

Rq _{LF P}, and Rq _{P A}) has a different heating time

constant. The time constant associated with

heating of the LED die is in the order of one

millisecond. The time constant associated with

the heating of the LED emitter package is in the

order of one minute. The time constant

,

associated with the heating of the complete LED

signal lamp is in the order of 10 to 30 minutes.

Thus, for a transient heating condition, the LED

Applications

Resistive Current Limiting

As discussed previously in the section “Key

Concepts for Electrical Design of LED Signal

Lamps,” the choice of the number of LED

emitters per series-string has a large effect on

the forward current regulation and the overall

electrical power consumption of the LED signal

lamp. Most 12V designs commonly use either

three or four emitters per series-string, which is

a good balance of current regulation and

electrical power consumption. Then, the choice

of circuit topology (Figure 3.1 circuits) and the

design current determine the variation in

if the LED signal light will be subjected to

automotive EMC transients.

For a resistive current-limited circuit the electrical

design process consists mainly of picking the

proper value(s) for the current limiting resistor(s).

The key principles of worst-case design are

shown in Figure 3.19. The figure shows the

forward current through one LED string of four

emitters as a function of input voltage. The

equation for this graph (Equation #3.12) is equal

to Equation #3.2 solved for I_F:

forward currents for the LED emitters in the

array. Finally, protection circuitry can be added,

Equation #3.2, from “Key Concepts for Electrical

Design of LED Signal Lamps.”

21

#3.12, for the same value of external current

limiting resistor, R.

The maximum forward current through the LED

string is determined by the effects of thermal

resistance of the LED signal lamp, Rq_JA, and

maximum ambient temperature, T_{A MAX}, as shown

in Figure 4 of the SuperFlux LED Data Sheet.

Where:

R = external current limiting resistor

V_IN= input voltage applied to the circuit

V_D= voltage drop across optional reverse

transient EMC protection diode

I_F= design forward current through the circuit

V_O= turn-on voltage of the linear forward

voltage model

So, the worst-case design procedure is to

determine the maximum forward current, I_{F MAX}

,

based on parameters Rq_JAand T_{A MAX}from Figure

4 of the SuperFlux LED Data Sheet. Then the

value of the external resistor, R, is determined

R_S= series resistance of the linear forward

voltage model

y = number of series connected LED emitters

x = number of paralleled strings used with

external current limiting resistor

with Equation #3.2, using values of I_{F MAX}, V_{IN MAX}

and LED linear model parameters (V_{O LL}, R_{S LL}).

,

The nominal design current, I_{F DES}, which occurs

at the design voltage, V_{IN DES}, can be calculated

with Equation #3.12 using values of V_{IN DES}, R,

and LED emitters with a typical forward voltage

[i.e. with linear model parameters (V_{O NOM}, R_{S NOM}).

In practice, the LED signal lamp needs to be

designed to accommodate a range of forward

voltage categories and luminous flux categories.

In most cases, the overall goal is to design an

LED signal lamp with a fixed light output. For

LED emitters from the same luminous flux

category, this can be accomplished by

Figure 3.19 Worst-Case Variations in Forward

Current Through Several LED Strings as a

Function of the Applied Voltage.

calculating different values of external resistor, R,

using Equation #3.2, with the same design

current and voltage, I_{F DES}and V_{IN DES}, and using

the nominal forward voltage models (V_{O NOM}, R_S

_NOM) for each forward voltage category.

Note, for the example shown in this section:

x = 1 and y = 4.

Figure 3.19 shows how the variation in forward

voltage of the LED emitters affects the forward

current through the string. These curves are

generated by substituting the worst-case linear

forward voltage model parameters [(V_{O NOM}, R_S

_NOM), (V_{O LL}, R_{S LL}), or (V_{O HH}, R_{S HH})] into Equation

For LED emitters with higher luminous flux

categories, the design current would be reduced

to keep the light output constant. First, the

minimum luminous flux would be calculated for

the lowest expected luminous flux category at

22

the nominal design current. Then Figure 3 from

the SuperFlux LED Data Sheet could be used to

calculate the other design currents for higher

luminous flux categories. Then values of

design. Lumileds Lighting recommends that the

designer use realistic assumptions for these

parameters. It is very easy to overly guard-band

these assumptions, which results in an excessive

estimate of the number of LED emitters needed.

external resistor, R, can be calculated with

Equation #3.2 at the reduced design current

using the appropriate nominal forward voltage

models for each forward voltage category.

The luminous efficiency of AlInGaP technology

has significantly improved over the past few

years. Please consult with your Lumileds Lighting

or Agilent Technologies Field Sales Engineer for

the recommended minimum luminous flux

categories of SuperFlux LED emitters for given

future production dates.

After all of these “ideal” values of R are

computed, the designer would need to choose

the closest standard resistor values. In many

cases, the designer can use the same resistor

value for multiple LED emitter categories

provided that the maximum forward current is

not exceeded under worst-case conditions.

Many LED signal lamp requirements also include

operation at higher voltages for a limited duration

(i.e. 24 volts for two minutes). In analyzing the

performance of an LED signal lamp under these

conditions, it is important to analyze the transient

heating effects. Under these conditions the LED

emitters don’t reach thermal equilibrium so the

junction temperatures are lower than indicated

by Equation #3.9. Equation #3.11 can be used

to estimate the maximum junction temperature

Experience has shown that the worst-case

design occurs with the lowest expected

luminous flux category and the highest forward

voltage category. Thus, the lowest value of

external resistor, R, and the highest design

current, I_{F DES}, would be determined for this

particular category combination. For LED

emitters with the lowest expected luminous flux

category and lower forward voltage categories,

Equation #3.2 will generate higher values of R.

For LED emitters with higher luminous flux

categories, the values of R will be even larger

since the design current is reduced.

using the appropriate time constants for Rq_{J LF}

Rq_{LF P}, and Rq_{P A}. In addition, many test

specifications allow a different operating

temperature for these tests.

,

The assumptions used for maximum ambient

temperature, Rq_JA, maximum steady-state input

voltage, and the worst-case SuperFlux LED

categories (minimum expected luminous flux

and maximum expected forward voltage) have a

large effect on the nominal design current and

thus the luminous flux output. Thus these

parameters have a large effect on the number of

LED emitters needed for a given signal lamp

23

CHMSL Design Example

This worst-case design procedure will be

illustrated with an example. Let suppose that an

LED array is being designed using 4-LED

strings of HPWT-MH00 from luminous flux

categories F through L and forward voltage

categories 2 through 6. For this example Rq_JA=

350° C/W, T_{A MAX}= 70° C, V_{IN MAX}= 15.0 V, and

V_{IN DES}= 12.8 V. In addition, a silicon diode with

a forward voltage of 0.8 V is connected in series

with the circuit in order to provide protection

against negative EMC transients. Then, the

design steps are shown below:

For HPWT-MH00, forward voltage category 6:

V_{O LL}= 1.85 V, R_{S LL}= 14.4 ohm

Then:

3. Determine nominal design current, I_{F DES}

:

The nominal forward current through the LED

emitters at the design voltage is determined with

Equation #3.12. The equation should use the

value of R from the Step 2 at the design voltage

and the nominal forward voltage of the SuperFlux

LED emitters using the same forward voltage

category used in Step 2.

1. Determine I_{F MAX}

:

The maximum allowable DC forward current

through the SuperFlux LED emitters is

determined from the maximum ambient

temperature, T_{A MAX}, estimated overall thermal

resistance, Rq_JA, of the LED signal lamp, and

Figure 4 of the Data Sheet.

For HPWT-MH00, forward voltage category 6:

V_{O NOM}= 2.03 V, R_{S NOM}= 12.4 ohm

For T_{A MAX}= 70° C, and Rq_JA= 350° C/W:

Then: I_{F MAX}= 55 mA, from HPWT-MH00

DataSheet, Figure 4

2. Determine minimum value of current limiting

resistor, R:

The minimum value of R is determined with

Equation #3.2 at the maximum input voltage

and maximum forward current from Step 1 for

the maximum forward voltage category Super

Flux LED to be used in the assembly.

24

4. Determine value of external current limiting

The linear forward voltage models for the other

HPWT-MH00 forward voltage categories are

shown below:

resistor, R, for each forward voltage category:

For SuperFlux LED emitters from lower forward

voltage categories, the value of the external

resistor, R, will need to be increased (using

Equation #3.2) in order to maintain the same

nominal forward current.

HPWT-

MHOO

Voltage

Category 2

Voltage

Category 3

Voltage

Category 4

Voltage

Category 5

Voltage

Category 6

V_O

R_S

=

1.87 V

8.2 ohm

1.91 V

9.2 ohm

1.94 V

10.5 ohm

1.96 V

11.6 ohm

2.03 V

12.4 ohm

NOM

For V_{IN DES}= 12.8 V, V_D= 0.8 V, I_{F DES}= 33.6 mA, x = 1, y = 4:

HPWT-

MHOO

Voltage

Category 2

Voltage

Category 3

Voltage

Category 4

Voltage

Category 5

Voltage

Category 6

R =

102 ohm

93 ohm

84 ohm

77 ohm

66 ohm

5. Determine minimum thermally stabilized

luminous flux:

Design currents for SuperFlux LED emitters at

higher luminous flux categories can be

determined using Figure 3 from the SuperFlux

LED Data Sheet. This can be done by computing

a new “relative luminous flux” equal to the

desired luminous flux divided by the minimum

luminous flux category bin limit and then reading

a new value of forward current from Figure 3.

The thermally stabilized luminous flux of the

SuperFlux LED emitters from the lowest

expected luminous flux category can be

determined using Figure 3 from the SuperFlux

LED Data Sheet.

For HPWT-MH00, luminous flux category F (3.0

lm minimum) and I_{F DES}= 33.6 mA:

The same minimum luminous flux obtained from

a HPWT-MH00 luminous flux category F (3.0 lm)

driven at 33.6 mA can be obtained from a

HPWT-MH00 from the following luminous flux

categories when driven at the specified forward

current:

Then: F _{V MIN}= (3.0)(0.54) = 1.62 lm, (i.e. “relative

luminous flux” = 0.54) from HPWT-MH00 Data

Sheet, Figure 3

6. Determine design currents for brighter

SuperFlux LED emitters:

25

From Figure 3, HPWT-MH00 Data Sheet:

Luminous Flux Category

F

G

H

J

L

Minimum Luminous Flux

Desired “relative luminous flux”

Design Current, from Figure 3

3.0 lm

0.54

33.6 mA

3.5 lm

0.46

29 mA

4.0 lm

0.41

25 mA

5.0 lm

0.32

19 mA

6.0 lm

0.27

15 mA

Note: Due to the variations in forward voltage at

low currents, there is a practical limit to the use

of higher and higher luminous flux categories at

correspondingly lower dc drive currents. For the

series string circuit shown in Figure 3.1a, at

drive currents less than 20 mA, the “worst-

case” ratio of forward currents between two

strings of LED emitters can vary by over 2:1. As

shown by this example, in order to achieve the

same light output for all CHMSL arrays, designs

using HPWT-MH00 emitters from luminous flux

categories J and L would require drive currents

less than 20 mA. Thus, the designer needs to

establish whether it is better to limit the forward

current to 20 mA and allow the light output to

increase for these brighter luminous flux

categories, or to accept possible visible light

output mismatch within the array.

7. Determine values of R for expected luminous

flux and forward voltage categories:

Values of R can be determined (using Equation

#3.2) for each SuperFlux LED emitter forward

voltage category and luminous flux category at

the appropriate design current as calculated in

Step 6.

Note: Since the design current of the HPWT-

MH00 LED array is less than 32 mA for designs

using luminous flux bins G through K, a “low

current” linear forward voltage model (8 mA £ I_F

£?32 mA) was used. This model is shown below:

HPWT-

MHOO

Voltage

Category 2

Voltage

Category 3

Voltage

Category 4

Voltage

Category 5

Voltage

Category 6

V_O

R_S

=

1.80 V

10.4 ohm

1.83 V

11.7 ohm

1.85 V

13.2 ohm

1.88 V

14.3 ohm

1.94 V

15.1 ohm

NOM

For V_{IN DES}= 12.8 V, V_D= 0.8 V, I_{F DES}from Step 6, V_{O NOM}, R_{S NOM}, x = 1, y = 4:

HPWT-

MHOO

Design

Current

Voltage

Category 2

Voltage

Category 3

Voltage

Category 4 Category 5 Category 6

Flux, F

Flux, G

Flux, H

Flux, J

Flux, K

33.6 mA

29 mA

102 ohm

124 ohm

151 ohm

212 ohm

279 ohm

93 ohm

115 ohm

140 ohm

199 ohm

265 ohm

84 ohm

106 ohm

131 ohm

189 ohm

253 ohm

77 ohm

97 ohm

122 ohm

179 ohm

242 ohm

66 ohm

86 ohm

109 ohm

162 ohm

222 ohm

25 mA

19 mA^[1]

15 mA^[1]

Note 1: Operation at dc drive currents below 20 mA can cause noticeable light output differences within

the LED array.

26

8. Select “standard” resistor values:

Standard 5% Tolerance Resistors

HPWT-

MHOO

Design

Current

Voltage

Category 2

Voltage

Category 3

Voltage

Category 4 Category 5 Category 6

Flux, F

Flux, G

Flux, H

Flux, J

Flux, K

33.6 mA

29 mA

100 ohm

120 ohm

150 ohm

220 ohm

270 ohm

91 ohm

110 ohm

150 ohm

200 ohm

270 ohm

82 ohm

110 ohm

130 ohm

180 ohm

240 ohm

75 ohm

100 ohm

120 ohm

180 ohm

240 ohm

68 ohm

91 ohm

110 ohm

160 ohm

220 ohm

25 mA

19 mA^[1]

15 mA^[1]

Note 1: Operation at dc drive currents below 20 mA can cause noticeable light output differences within

the LED array.

9. Group “standard” adjacent cells in resistor matrix in Step 8 as desired:

Standard 5% Tolerance Resistors

HPWT-

MHOO

Design

Current

Voltage

Category 2

Voltage

Category 3

Voltage

Category 4 Category 5 Category 6

Flux, F

Flux, G

Flux, H

Flux, J

Flux, K

33.6 mA

29 mA

100 ohm

82 ohm

110 ohm

130 ohm

180 ohm

240 ohm

120 ohm

150 ohm

220 ohm

270 ohm

25 mA

19 mA^[1]

15 mA^[1]

Note 1: Operation at dc drive currents below 20 mA can cause noticeable light output differences within

the LED array.

10. Perform “worst-case” analysis to ensure

that maximum forward current is not exceeded

over temperature.

varies slightly over temperature as shown in

Equation #3.6. This thermal effect can be

included in Equation #3.12 as shown below:

Calculate maximum forward current (using

Equation #3.12) at “worst-case” conditions—

i.e. maximum input voltage, minimum resistor

values, and minimum forward voltages for each

SuperFlux LED emitter forward voltage

category. The forward voltage of LED emitters

EMC Transient Protection

Circuits designed for the automotive electrical

environment must be able to operate over a

wide range of input voltages and be able to

tolerate a number of different types of electrical

transients. These worst-case voltage ranges and

electrical transients have been characterized and

27

are defined in different automotive specifications

such as:

are turned off within the vehicle, switching

transients of electronic circuitry, alternator field

decay, or a fully discharged battery being

disconnected while the alternator is operating at

rated load. These transients consist of both

positive and negative pulses with different

amplitudes and decay times.

DIN 40839 Part 1

“Electromagnetic Compatibility (EMC) in Motor

Vehicles; interferences conducted along supply

lines in 12 V onboard system”

DIN 40839 Part 2

Limited reliability testing has been done with

AlInGaP LED emitters connected in typical LED

signal lamp configurations.

“Electromagnetic Compatibility (EMC) in Motor

Vehicles; interferences conducted along supply

lines in 24 V onboard system”

High-voltage negative transients in excess of the

reverse breakdown voltage can permanently

damage AlInGaP LED emitters. Sufficient energy

can be dissipated within the AlInGaP LED die to

cause localized damage to the p-n diode

structure. This damage can result in reduced

breakdown voltages, and degraded low-current

performance. Under extreme conditions, high

voltage negative transients can even destroy the

p-n junction, resulting in a short between anode

and cathode.

ISO 7647-1

“Road Vehicles—Electrical Disturbance Caused

by Conduction and Coupling; passenger cars

and light commercial vehicles with nominal 12 V

supply voltage”

ISO 7647-2

“Road Vehicles—Electrical Disturbance Caused

by Conduction and Coupling; commercial

vehicles with nominal 24 V supply voltage”

SAE J1113

Adding a high-voltage silicon diode in series with

the LED signal light array, such as previously

shown in Figure 3.6, can prevent potential

damage to high voltage negative transients. The

silicon diode should have a higher reverse

breakdown voltage than the amplitude of the

worst-case negative transient, which can be as

large as –300 V (–600 V for heavy trucks). Table

3.2 shows several recommended silicon diodes

for different LED signal lamp applications.

“Electromagnetic Susceptibility Measurement

Procedures for Vehicle Components (except

Aircraft)”

SAE J1211

“Recommended Environmental Practices for

Electronic Equipment Design”

SAE J1812

“Function Performance Status for EMC

Susceptibility Testing of Automotive Electronic

and Electrical Devices”

In addition, high-voltage positive transients can

permanently damage AlInGaP LED emitters.

Sufficient energy can be dissipated within the

AlInGaP LED die to cause permanent damage to

the p-n diode structure and cause epoxy

These specifications define several electrical

transient pulses that occur when inductive loads

28

delamination between the LED and surrounding

epoxy. Under extreme conditions, the epoxy

surrounding the LED die can be charred and

the gold bond wire and LED die can be

destroyed, resulting in an open circuit. The

AlInGaP LED die can tolerate non-recurring

peak current transients of several hundred

milliamperes for short time periods (t << 1 ms)

with minimal permanent effects. However,

longer transients can cause sufficient localized

heating to cause the various effects listed

earlier. The “Load Dump” transient pulse can be

especially damaging since the pulse duration

can be up to 400 ms.

limiting properties of the LED drive circuit,

determine the maximum peak current through

the LED array. For best results, the breakdown

voltage of the transient suppressor should fall

within the following range:

24 V < V_BR< 45 V

Note that the 24 V restriction is determined by

the “Jump Start” voltage condition. The 45 V

restriction is determined by the ability of the LED

array to withstand the peak current imposed by

the transient voltage. Since the 45 V limit

depends on the circuit topography of the LED

array and the maximum “Load Dump” transient

pulse duration, this voltage limit should be

established by reliability testing.

The effects of “Load Dump” transients can be

minimized by putting a surge-suppressor or

silicon transient suppressor in parallel with the

LED array as previously shown in Figure 3.6.

Note that the breakdown voltage of the

transient suppressor, as well as the current-

Table 3.2

Silicon Diodes for EMC Negative Transient Protection

Diode Part

Number

Maximum Continuous

Reverse Breakdown

Forward Current, I_O

Voltage, V_RRM

Applications

1N4005

1N5396

1N5406

1.0 A

1.5 A

3.0 A

600 V

CHMSL

Rear Combination Lamp

Rear Combination lamp,

& Front Turn Signal

Special Considerations for Dual Luminous Intensity Operation

Some applications, such as a Stop/Tail signal

lamp require two discrete levels of light output.

These applications require some additional

design considerations. In most cases, the ratios

in light output at the two signal conditions are

determined by the various signal lamp

ratios for Stop/Tail signals are 7:1 to 15:1.

Generally, the LED emitters should appear

matched in luminous flux at both drive

conditions. This implies that the forward currents

for the LED emitters in the array should be

matched at both drive currents. SuperFlux and

SnapLED 70 emitters are categorized for

specifications or regulations. Typical dimming

29

luminous flux and forward voltage at 70 mA

(150 mA for the SnapLED 150). Thus, their

matching is best when driven at high forward

currents. As discussed in the section “Electrical,

Optical, and Thermal Properties of an LED

Emitter,” Figure 3.11 shows the expected range

in light output for HPWT-xH00 emitters that are

matched at 70 mA. Note that the light output

varies by approximately 2:1 at a 20 mA forward

current. Note that since the SnapLED 150 is

matched at 150 mA, then the light output would

vary by about 2:1 at 40 mA. Thus, even if a

well-controlled constant current circuit drives

the LED emitters, the light output matching

might be unacceptable for the Tail function if

driven by a low DC forward current.

Figure 3.20 shows a PWM Tail circuit. When the

input voltage is applied to the Stop pin, the

cross-connected paralleled string circuit is

operated at a dc forward current determined

external resistor, R_STOP. The value for R_STOPcan be

determined using Equation #3.2. When the input

voltage is applied to the Tail pin, the NE555

oscillator is energized. Resistors, R₁and R₂, and

capacitor, C₁, determine the frequency and duty

cycle of the oscillator. The values shown

generate a 2K Hz frequency and a 10% duty

cycle (output low). When the output of the

NE555 (pin 3) is low, a high-current switching

transistor is turned-on, which supplies current to

the LED array. External resistor, R _TAIL, determines

the peak current of the Tail circuit. The value for

R _TAILcan be determined using a similar equation

as used previously for R_STOPand including the

extra voltage drop across the high-current

switching transistor. Thus for the same value of

peak forward current, R _TAIL< R_STOP. Diodes D₁

and D₂protect the circuit from negative EMC

transients. Zener diode D ₃protects the NE555

from positive EMC transients.

A PWM circuit is recommended for best

luminous intensity matching for the Tail function.

This circuit could drive the LED array at a high

forward current for the Stop function and drive

the LED array at the same peak forward current

but at a low duty cycle for the Tail function. With

this approach, the matching of the LED array

will be the same regardless of whether the

signal is operating in Stop or Tail mode.

Figure 3.20 Stop/Tail LED

Signal Lamp Circuit that Uses

a PWM Scheme to Generate

the Reduced Light Output of

the Tail Signal.

30

Current and Voltage Regulator Circuits

This section will discuss active circuits that are

designed to drive the LED emitter array at a

constant voltage or constant current despite

input voltage or load variations. These circuits

are called voltage or current regulator circuits

because they are designed to regulate the input

voltage to generate either a fixed output voltage

or current.

and compare the load voltage with a reference

voltage. Current regulator circuits usually

measure the current through the load by

measuring the voltage drop across a “sense”

resistor in series with the load. Then the voltage

across the sense resistor is compared with a

reference voltage.

Since most LED signal lamp designs consist of

several LED emitters, they are normally arranged

in one or more series-connected strings, such as

shown previously in Figure 3.1. While it is

possible to use one voltage or current regulator

per string, due to cost considerations, most

practical designs use a single voltage or current

regulator for the entire LED array. Note that when

only a single regulator is used for the entire array

it is possible to encounter the same type of

forward current variations as described earlier in

the section “Key Concepts for the Electrical

Design of LED Signal Lamps.” Since LED

emitters are current-controlled devices,

voltage regulator circuits should use current-

limiting resistors in series with each string of

LED emitters (Figure 3.1a circuit), paralleled

string of LED emitters (Figure 3.1b circuit), or

cross-connected paralleled string of LED

emitters (Figure 3.1c circuit). For voltage

regulator circuits, Equation #3.2 can be used to

calculate the value of the external current-limiting

resistor(s) if the regulated output voltage, V_OUT, is

substituted into the equation for V_IN. For current

regulator circuits, external current limiting

resistors are not required but their use can

reduce forward current variations within the LED

array.

The use of voltage or current regulation

improves the operation of the LED signal lamp.

Since the drive current of the LED array remains

constant despite variations in the supply

voltage, the light output is not affected by input

voltage variations. Since the drive current

doesn’t increase due to over voltage conditions,

the LED emitters can be driven at a higher

forward current at the design voltage without

exceeding the maximum allowable forward

current at the maximum input voltage. In

addition, if the circuit is located outside of the

LED signal lamp case, the voltage or current

regulator circuit can improve the thermal

properties of the signal lamp by reducing the

power consumption within the LED signal lamp.

Block diagrams of typical voltage and current

regulator circuits are shown in Figure 3.21. The

basic elements of all of these circuits consist of

a high gain amplifier and feedback circuit, which

vary the dynamic load of a power circuit that is

either in series or parallel with the LED emitter

array. The regulator circuit modulates the

dynamic load so as to provide either a constant

voltage or current to the LED emitter array

independent of input voltage or load variations

(over some specified range). Voltage regulator

circuits measure the voltage across the load

31

As shown in Figure 3.21, there are three basic

types of regulator circuits. The circuits shown in

Figures 3.21a and 3.21d are called “shunt”

regulators. They use a dynamic load in parallel

with the load being regulated that shunts some

of the supply current around the load. Shunt

regulators also have a power resistor in series

with both loads. The value of the power resistor

has been selected such that at the minimum

input voltage and maximum load condition, the

current through the dynamic load goes to zero.

At higher input voltages or smaller loads, the

current through the dynamic load is increased,

which increases the voltage drop across the

power resistor to keep the load current or

voltage constant. In this way, the shunt

due to the large variations in input voltage and

will not be covered further in this section.

The circuits shown in Figures 3.21b and 3.21e

are called “series-pass” regulators. They use a

dynamic load in series with the load being

regulated. At minimum input voltages the voltage

drop across the dynamic load goes to a

minimum value. This minimum voltage drop is

called the “drop-out” voltage. At higher input

voltages, the voltage drop across the dynamic

load increases so as to maintain either a fixed

current or voltage across the load. At voltages

below the drop-out voltage, the dynamic load

can no longer regulate the output voltage or

current. Thus, for proper voltage or current

regulation, the input voltage needs to be higher

than the sum of the drop-out voltage, the voltage

across the load, and the voltage drop across the

sense resistor (if applicable).

regulator maintains either a fixed current or

voltage across the load. Shunt regulators are

not very practical for automotive signal lamps

Figure 3.21 Block Diagrams of Several Active Drive Circuits for LED Signal Lamps.

32

The circuits shown in Figures 3.21c and 3.21f

are called “switching” regulators. They use a

dynamic load that is switched ON and OFF at

very high frequencies at a varying duty cycle.

The dynamic load supplies electrical power to

an energy storage element such as a capacitor

or an inductor or a combination of both. This

energy storage element then supplies power to

the load. The percentage of time the dynamic

load is ON is varied depending on the input

voltage and load requirements. The “switching”

regulator provides the highest power efficiency

of the three circuits. However, it is the most

complex of the three regulator circuits and has

the highest potential for creating unwanted EMI

(due to the high-frequency switching).

circuit shown in Figure 3.1b or “rung” for the

circuit shown in Figure 3.1c.

With a voltage regulator, the forward voltage

applied to the LED array voltage will be

independent of supply voltage variations as long

as the voltage regulator remains in its active

region. However, ambient temperature variations

and the use of different forward voltage

categories can affect the forward current through

the LED array unless provisions are made in the

design. As mentioned earlier, current limiting

resistors, R, are needed for each string of LED

emitters. With R > y DR_S, the forward current

through each string will primarily be determined

by the value of R. If the designer uses a voltage

regulator with a fixed output voltage, then the

values of these current-limiting resistors will need

to be varied for each of the different forward

voltage categories in order to compensate for the

different forward voltages at the design current.

Alternatively, the designer could use the same

value of current-limiting resistors for all forward

voltage categories. However, in this case, the

regulator output voltage would need to be varied

slightly for each different forward voltage

The performance of these different types of

regulators is compared with an example shown

in the sidebar “Comparison of Three Constant-

Current Circuits.”

The LED emitter array can be driven from either

a voltage regulator or a current regulator circuit.

With a current regulator, the total array current

will be independent of supply voltage,

temperature and forward voltage category

variations as long as the current regulator

remains in its active region. If the current

regulator is used with parallel-connected LED

emitters, such as shown in Figure 3.1b or 3.1c,

there can still be similar forward current

category to compensate for the different forward

voltages at the design current. Despite these

precautions, there will still be small variations in

the total current through the LED array due to

slightly different forward voltages of the individual

emitters. With only a small voltage drop across

the current limiting resistor, small variations in

the regulated voltage can cause large changes

in forward current through the LED emitters. In

addition, since the forward voltage of the LED

emitter varies with temperature, the forward

current through the LED array will increase at

elevated temperatures. However, it is possible

variations within the LED array as was

discussed in the section “Key Concepts for the

Electrical Design of LED Signal Lamps.” Note

that the forward current matching can be

improved with the addition of a small resistor

(R_OPT> R_S) in series with each string for the

33

to maintain fixed current through the LED array

if the output voltage of the regulator tracks

the DV_F/DT of the LED array (approximately

–2 mV/°C times the number of emitters in each

series string). Finally, if the voltage regulator is

used with parallel-connected LED emitters,

such as shown in Figure 3.1b or 3.1c, there can

still be similar forward current variations within

the LED array as was discussed in the section

“Key Concepts for Electrical Design of LED

Signal Lamps.” Note that the forward current

matching can be improved with the addition of a

small resistor (R_OPT> R_S) in series with each string

for the circuit shown in Figure 3.1b or “rung” for

the circuit shown in Figure 3.1c.

Comparison of Three Constant-Current Circuits

SETUP:

Suppose an LED signal lamp is being designed

using 30 HPWT-DH00 SuperFlux LED emitters

from forward voltage category 3. The circuit will

be designed to operate at 50 mA per emitter at

a design voltage of 12.8 V.

SOLUTION:

For the first two designs, the value of the external

current limiting resistor, R, can be determined

using Equation #3.2. Note, for forward voltage

category 3, the nominal linear forward voltage

model is V_{O NOM}= 1.91 V, and R_{S NOM}= 9.2 ohm.

Thus, for the three-LED string circuit, R = 114

ohm. For the four-LED string circuit, R = 66 ohm.

The detailed designs are shown in Figure 3.22.

Then over an input voltage range of 7 to 18 volts,

the forward current through each LED string

would vary as shown in Figure 3.23.

PROBLEM STATEMENT:

How does the overall power consumption

compare for the following 4 possible circuit

designs over an input voltage range of 9 V

to 18 V?

1. Resistive current limiting (Figure 3.1a circuit)

with ten strings of three emitters per string.

2. Resistive current limiting (Figure 3.1a circuit)

with eight strings of four emitters per string.

3. Series-pass constant-current regulator

driving ten strings of three emitters per string.

4. Switching constant-current regulator driving

ten strings of three emitters per string.

Figure 3.22 Two LED Signal Lamp Designs Using

Resistive Current Limiting.

34

Figure 3.23 Forward Current Through LED

Emitters as a Function of Applied Voltage

for Resistive Limited Circuits Shown in

Figure 3.22.

Figure 3.24 Block Diagram of LED Signal Lamp Design

Using Series-Pass Constant-Current Regulator.

The key elements of the series pass constant-

current regulator are shown in Figure 3.24. For

series strings of three forward voltage category

3 HPWT-DH00 emitters, the forward voltage of

the string is about 7.10 V at 50 mA. Assuming

a voltage drop across the sense resistor of 0.25

V, then at an input voltage of 9 V, the drop-out

voltage of the regulator would be (9 V – 7.1 V –

0.25 V), or 1.65 V. Since there are 10 strings of

LED emitters, the total LED array current would

be 50 mA times 10, or 500 mA. Thus, the sense

resistor would be (0.25 V / 0.500 A), or 0.5

ohms. Then over an input voltage range of 7

to 18 volts, the total load current of the circuit

would vary as shown in Figure 3.25. As

Figure 3.25 Forward Current Through LED Emitters

as a Function of Applied Voltage for Series-Pass

Constant-Current Regulator Circuit Shown in Figure

3.24.

The key elements of the switching constant-

current regulator are shown in Figure 3.26. There

are a number of different types of switching

regulators. Buck or Down Converters are

designed, this circuit maintains a constant

current through the LED array at input voltages

greater than 9 V. Suppose that the minimum

compliance voltage of the circuit is designed to

be 10 V, then an additional volt can be dropped

across the load or series pass regulator.

designed to generate a regulated output voltage

that is always less than the input voltage. Boost

or Up Converters are capable of generating a

regulated output voltage that is always higher

than the input voltage. Buck/Boost or Up/Down

Converters can generate a regulated output

voltage using any input voltage. By comparison,

35

current-limiting resistors and series-pass

input voltages, the efficiency of the switching

regulator is better than the series-pass regulator

and the resistor-limited circuits. Assuming a

0.25 V drop across the sense resistor, then for

the ten-string circuit, R_SENSEwould be equal to

(0.25 V / 0.500 A), or 0.5 ohms. Assuming a

power conversion efficiency of 80% and an input

voltage range of 7 to 18 volts, then the input

current and total load current of the circuit would

vary as shown in Figure 3.27.

regulators can only reduce the output voltage to

a lower value than the input voltage. Thus for

some types of switching regulator circuits the

number of LED emitters per string can be larger

than the number of LED emitters per string for a

resistor-limited or series-pass regulator circuit.

In general, the switching regulator converts the

average input power (V_INtimes I_IN) to the desired

output power (V_LOADtimes I_LOAD) with a relatively

fixed power conversion efficiency. At higher

Figure 3.26 Block Diagram

of LED Signal Lamp

Design Using Switching

Constant-Current

Regulator.

Figure 3.27 Forward Current Through LED

Emitters as a Function of Applied Voltage for

Switching Constant-Current Regulator Circuit

Shown in Figure 3.26.

Figure 3.28 Comparison of Total Supply

Current versus Applied Voltage for Circuit

Designs Shown in Figures 3.22, 3.24 and 3.26.

The total power consumption for the four

different LED signal lamp designs is shown in

Figure 3.28. The series pass and switching

regulator designs provide substantial power

savings compared to the resistor-controlled

circuits during over-voltage conditions. Note

that at an input voltage of 18 V, both resistor

limited circuits have an overall power consumption

of 15 W. The series pass current regulator circuit

has an overall power consumption of 9 W. The

switching current regulator has an overall power

consumption of 5 W.

36


型号：	AB20-3
厂家：	LUMILEDS LIGHTING COMPANY
描述：	Electrical Design Considerations for SuperFlux LEDs 电气设计考虑食人鱼LED灯
文件：	总37页 (文件大小：2017K)
中文：	中文翻译
下载：	下载PDF数据表文档文件

AB20-3 [LUMILEDS]

相关型号：

AB20-3A

AB20-3B

AB20-4

AB20-5

AB20-6

AB20-7

AB2020B-2

AB2022A

AB2025B

AB2036AF

AB2036B

AB2040B