AN1261

Dimming Power LEDs Using a SEPIC Converter and

MCP1631 PIC Attach PWM Controller
The PWM driver is used to dim an LED and is based
Author: Valentin C. Constantin upon the persistence of vision of the human eye. The
Microchip Technology Inc. current does not flow through the LED continuously.
The PWM period is normally in the range of 100 Hz to
INTRODUCTION 250 Hz. Dimming is obtained by changing the PWM
duty cycle. Variations in supply voltage will directly
This application note covers the general concept, affect the LED current.
design and some source code modifications for using
the MCP1631HV Digitally Controlled Programmable The Constant Current Source driver is used to supply
Current Source Reference Design for an LED lighting the LED with a constant current. The constant current
application. The document offers additional insight into results in a consistent LED color output. A high
the MCP1631HV Digitally Controlled Programmable frequency PWM signal may be used to control the
Current Source Reference Design. A Microchip current that is flowing through the LED. The PWM
MCP1631 PWM controller and a low-cost PIC16F616 signal is used to set the LED current and thus the LED
microcontroller are used to manage a SEPIC (Single intensity.
Ended Primary Inductive Converter) powertrain. The The Constant Current Source driver is based upon
SEPIC topology is then used to drive and dim Light driving the LED with a constant current source to obtain
Emitting Diodes (LEDs). the desired brightness and color. The current source is
The LED technology continues to change and evolve. controlled by a high frequency PWM signal. The PWM
The energy efficiency (lumens/watt) of today’s LEDs modulated constant current source is set to the desired
allows them to replace the classical tungsten filament LED current. By changing the PWM duty cycle, the
and gas vapor lighting systems with solid-state LED current can be adjusted for desired intensity and
technology. The most efficient LEDs are able to to offset aging.
perform as well as Compact Fluorescent Lamps (CFL). Constant current driving is recommended to eliminate
The most important advantage of LEDs over tungsten light flickering and electrical stress due to voltage
and gas vapor bulbs is the cycle life. LED lifetimes fluctuations. An important observation is the LED is not
currently show a 30% improvement over their affected by the variation of input power supply voltage
counterparts. LEDs are being used in the automotive that can vary as much as +/- 30% of the nominal value.
and commercial lighting industries with good results. The design topology used for this application note will
LEDs must be carefully integrated into lighting systems be the Single Ended Primary Inductive Controller or
because of their sensitivity to thermal and electrical SEPIC topology. The SEPIC topology allows both buck
stress. The LED current must be optimized for both and boost operation. This allows for a constant current
thermal and electrical characteristics. Excessive drive output with variations in input voltage. The SEPIC
driving current will deteriorate LED performance, uses a current sense feedback loop for efficient control.
shorten the lifetime, and cause permanent reduction in Microchip’s MCP1631HV Digitally Controlled Program-
the luminous intensity. mable Current Source Reference Design is used for
LEDs may be driven using constant current sources in this application note.
order to maintain a consistent color output. One low
cost but inefficient solution for driving LEDs is using
series resistors. The series resistor method is very
sensitive to power supply variation. Constant LED
current depends on a constant supply voltage. Any
variation of the voltage supplying the resistor and LED
combination will result in a change in current and thus
a change in output color. There are more efficient
solutions for driving LEDs that save energy and provide
good current control. Two methods of driving an LED
are Constant Current and Pulse Width Modulation
(PWM) current control.

© 2009 Microchip Technology Inc. DS01261A-page 1

AN1261
LED BACKGROUND The typical forward voltage drop of a red LED (ultra red,
super red, or amber) ranges from 1.8V to 2.2V. Yellow
There are a few important parameters that must be and green LEDs range from 2.1V to 2.2V. Blue and
considered when designing an LED driver. white (cool or warm) LEDs range from 3.6V to 3.8V.
The user should consult the manufacturers data sheet
LED Power, Maximum Forward Current for the proper design parameters when using high
and Voltage power LEDs.
Another important parameter is the Maximum Reverse
Generally, the LED manufacturers recommend driving
Voltage (Vrmax), which is typically 5V. Applying reverse
LEDs at constant current. For power type LEDs, the
voltage in excess of the manufacturers
drive current covers hundreds of milliamps. Standard
recommendation will degrade the LED and may
values which are currently used for LED drive current
destroy it.
are: 350 mA, 700 mA, 1A and higher. Some standard
power LEDs are rated for 1W, 3W, and 5W. LED Power
is Vf x If, where Vf is the forward voltage of the LED and Series and Parallel Connection
If is the forward current. If the user intends to connect two or more power LEDs
Power LED applications utilizing higher current levels in series, it is necessary to calculate the sum of the
waste energy and create excessive heat when supplied individual forward voltages. The color of the LEDs used
by series resistors. Any power supply voltage variation may be different when using a series topology.
will result in a change in LED current and luminous
intensity. It is important to note that the LED’s forward
voltage varies from device to device and also varies
significantly with temperature.

400
White & Blue
350
Forward Current (mA)

Green
300 Amber
250
200
150
100
50
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Forward Voltage (V) FIGURE 2: Topology of an LED String
(Series and Parallel Connection).
FIGURE 1: Typical Forward Current vs.
Forward Voltage for a White, Blue, Green, and The LED string can be driven by a Constant Current
driver without series resistors. The sum of the forward
Amber Power LEDs (P3 series) at +25°C.
voltage of each diode in a series string must not exceed
the maximum voltage capability of the current source.
The Direct Relation between Forward
Due to LED tolerances, parallel connections using a
Voltage and the Color of the LED,
single current sense may not be able to maintain a
Maximum Reverse Voltage uniform intensity and color. Better control is done using
Different color LEDs have different forward voltage (Vf) separate current sensing for each LED rung in a
drops based upon the diode chemistry. A red LED parallel system.
dissipates less energy than a white LED when The advantage of paralleling LEDs is that more LEDs
operated at the same current level. That is because the can be supplied at lower output voltage levels on
red LED has a lower Forward Voltage drop than the constant current supplies. If slight variations in
white LED. LED manufacturers datasheets will provide luminous intensity are tolerable, a single current sense
the maximum forward voltage drop (Vfmax) at a given can be used to control the current loop. However, if one
maximum forward current (Ifmax). The higher Vfmax rung of the parallel LED string fails, the current through
values are typically for white and blue LEDs and the the other LEDs will increase and possibly cause
lower are typically for red LEDs. undesirable consequences. Therefore, it is
recommended to monitor current through each rung
and drive each rung as if the string was a series string.

DS01261A-page 2 © 2009 Microchip Technology Inc.

 AN1261
When LEDs are connected in series, the maximum Minimum Heat Sink Requirements,
reverse voltage increases with each additional LED Thermal Modeling
added to the string. The group of LEDs may make up a
series connected system or they may make up one When designing a lighting system with power LEDs, a
rung of a parallel connected system. big challenge is thermal management. Thermal
management includes the selection of the cooling
Operating Temperature and Cooling the method or combination of methods. Thermal
management methods may consist of PCB material,
Power LED
fan, and heat sinks and their respective fastening
The forward voltage of an LED is temperature processes. After selecting the cooling method, the
dependent. The temperature coefficient of forward cooling system may then be designed.
voltage is published in the data sheet and its value is Thermal modeling is used to predict the junction
around 3 to 5 mV/°K, depending on the LED’s color. temperature of the LED in order to avoid exceeding the
This parameter shows the changes in forward voltage LED thermal limits specified in the datasheet. The
with the ambient temperature. The temperature thermal limit for power LEDs is typically between
increases directly with forward current. The Operating +120°C and +135°C.
Temperature Range is generally between -20°C and
+85°C. It is important to keep the LED case tempera- The thermal model used in thermal management
ture around 30°C to 40°C for good results. The LED design is the concept of Thermal Resistance (Rt), given
junction can typically work at a maximum of +125°C. in °C/W. Thermal resistance is defined as the ratio of
temperature difference to the corresponding electrical
The LED manufacturers recommend in their power dissipation.
datasheets that a heatsink be used with the power
LEDs. A successful design starts with the transfer of
EQUATION 1:
LED heat away from the device. Two typical methods
used to provide proper heat sinking are an external o
R tJA [ C ⁄ W ] = ΔT JA ⁄ P dLED
heatsink or the printed circuit board (PCB) layout.
Where:
A PCB on FR4 substrate with a large copper area and
thermal vias under a surface mount (SMD) LED is a RtJA = represents the thermal resis-
low-cost but inefficient solution. For good thermal tance from junction to ambient.
results, the PCB needs to be double-sided. PCB PdLED = the electrical power dissipated
heat-sinking is not indicated for power LEDs that by LED.
require more than 1W. Modern lighting lamps use Metal
ΔTJA = the difference between junction
Core PCBs (MCPCB) with special substrates like
and ambient temperatures.
aluminum or copper. Copper offers the best thermal
conductivity. These PCBs offer better cooling because
of the lower thermal resistance of the core metal EQUATION 2:
compared to FR4 fiberglass. The heat is dissipated in
all directions on the metal surface. Where the space P dLED [ W ] = V f × I f
does not permit a large heat sink or a large metal PCB,
Where:
a small electric fan may be used to force the air across
the heat sink. Vf = the forward voltage.
It is recommended to use extra thermal protection If = the forward current.
devices to avoid accidental overheating due to wrong
drive current, high ambient temperature or fan fault.
The extra protection can be managed by an electronic EQUATION 3:
controller. A small NTC thermistor or a semiconductor o
ΔT JA [ C ] = T J – T A
sensor may be placed on the LED heat sink or on the
metal PCB. The temperature trip point may be set Where:
around 30°C to 40°C.
TJ = the junction temperature.
Note that the short term effect of excessive heating is
TA = the ambient temperature.
color shifting, which is reversible. The long term effect
of excessive heating is a permanent decrease in
luminous intensity and LED life cycle. The ideal LED manufacturers typically specify in their data
operating temperature is typically 25°C. The luminous sheets the thermal resistance from the diode junction to
intensity will be at a maximum around this temperature. the ambient RtJA (solder point or LED case).

The red and amber LEDs are affected the most by Thermal resistances RtJC (Thermal Resistance from
heating. The white LED is least affected by heating. junction to case) and RtCA (Thermal Resistance from
case to air) should be optimized for each application.

© 2009 Microchip Technology Inc. DS01261A-page 3

AN1261
Since RtJC is determined by the device construction, There is a dependence between the working junction
only RtCA may be optimized for thermal efficiency. RtJC temperature and the PCB or heat sink thickness.
typically ranges from 3°C/W to 4°C/W, depending on Thicker metal provides a lower working junction
the LED manufacturer. Increasing the heat sink surface temperature. This dependence on thickness can be
area will improve the heat dissipation. seen in Figure 4 for a typical 1W P3 series LED.
The typical thermal model of an LED package consists
of:
35
• LED power dissipation (Equation 2) which is 34 Heat Sink
modeled as a current source

Temperature (°C)
33 Metal PCB
32
• thermal resistance which is modeled as a resistor 31
Chip

(Equation 1) 30
• the difference between ambient temperature and 29
28
junction temperature which is modeled as a 27
voltage source (Equation 3) 26
25
These are solved by Ohm's law following the model in
1 2 3 4 5 6 7 8
Figure 3. The working junction temperature, TJ, must
Base Thickness (mm)
be kept lower than the TJmax of the LED.

FIGURE 4: Base Thickness

Dependence of Temperature for Power LEDs.
Pd = Vf X If
Luminous Intensity. Viewing Angle
ΔTJA The unit of measurement commonly used to describe
the LED intensity is the millicandela (mcd).
RtJA 1000 mcd = 1 candela.
Part of the SI system of measurement, one candela
(cd) is the monochromatic radiation of 540 × 1012 Hz
which has a radiant intensity of 1/683 Watt per
steradian in the same direction. The unit of
FIGURE 3: Typical Thermal Model of an measurement commonly used for most other light
LED. sources is the Lumen. Lumens are units of Luminous
Flux. The Luminous Flux (lumens) from a light source
RtJA is a sum of RtJC and RtCA (Equation 4). RtCA is the is equal to the Luminous Intensity (candelas) multiplied
sum of the case-to-sink thermal resistance (RtCS) and by the solid angle over which the light is emitted.
the sink-to-air thermal resistance (RtSA).
The solid angle is the angle (here defined in 3D space)
Substituting for RtCA and solving Equation 1, we obtain
that an object subtends at a point. It is a measure of
the minimum RtSA heat sink requirement for the
how big an object appears to an observer who is
application (Equation 4). There are manufacturers that
looking from that object. The Lumens measure how
offer spreadsheets with several heat sink models and
much light actually falls on a surface.
their RtSAvalues. References for a few power LED
manufacturers are listed in the References section [9; Two optical parameters that are very important when
10]. The manufacturers web sites offer additional we choose a power LED for a lighting application are
information about thermal management. the luminous intensity and the viewing angle. It may
take some extra calculations to determine which
EQUATION 4: manufacturer’s power LED is right for your application,
o as each manufacturer tends to test their LEDs under
R tJA [ C ⁄ W ] = R tJC + R tCA different operating conditions. For lighting applications,
an LED with a 100° to 130° viewing angle is typically
o
R tJA [ C ⁄ W ] = R tJC + R tCS + R tSA recommended. Lumens per watt (lm/W) is a measure
of efficiency in converting electrical energy to light.
Mounting a power LED ensemble on a heat sink may Multiply this by the watts dissipated in the LED to get
require a thermal adhesive such as tape or thermal lumens.
grease. Thermal adhesive tape is recommended for
wide and flat surfaces. Avoid any air pockets between
the thermal adhesive and the attached face when
applying the tape.

DS01261A-page 4 © 2009 Microchip Technology Inc.

 AN1261
POWER LED DIMMING SOLUTIONS
There are applications that have the requirement to 1.2

Relative Luminous Flux

1.1 P3 Blue LED
control the luminous intensity of LEDs. The modalities 1 P3 White LED
for adjusting the brightness of LEDs were summarized 0.9
0.8
in the previous section: 0.7
0.6
• Adjusting the forward current value 0.5
0.4
• Dimming by Pulse-Width-Modulation (PWM) 0.3
0.2
0.1
Adjusting the Forward Current Value 0
0 50 100 150 200 250 300 350 400
Generally, there are two methods to drive power LEDs Forward Current, If (mA)
by constant current. One method is to use a constant
current source and the other method is to use constant
voltage source. The constant current source method FIGURE 5: Relative Luminous Flux is
can drive the LED with a programmed current at the Lineary Dependent on the Direct Forward
expense of a few more circuit devices. Dimming is Current.
obtained by modifying the value of current supplied by
the current source. Dimming by PWM
The constant voltage source method is simply As explained in the previous section, luminous intensity
connecting the LED to the constant voltage power can be modified by controlling the current to the LED.
supply in series with a resistor chosen for the desired However, this causes a color shift that for some LED
current. It is the simplest and lowest-cost solution and applications is not tolerated. The PWM control changes
thus a popular way to drive LEDs when efficiency and the luminous intensity without changing the color of the
color range are not critical. Dimming would require LED. PWM control consists of maintaining a constant
either stepping the input voltage or switching different current setpoint value to maintain LED color while vary-
resistor taps to obtain a new current through the LED. ing the duty cycle to control the LED brightness. The
From an efficiency and performance standpoint, the forward current through the LED is kept at a constant
best choice is a current source that has the ability to value and only the duty cycle (D) is changed. When the
generate digitally programmable current levels as current is flowing through the LED during the “ON” por-
needed. A digitally programmable current source may tion of the duty cycle, the LED color operating point is
use a current sense resistor in the control loop to set. When the current flow is cut off for the “OFF” por-
monitor the LED current. The current sense would be tion of the duty cycle, the LED is also off. The persis-
used by the control loop to maintain regulation about tence of the human eye causes the appearance of
the programmed current setpoint. The current sense brightness changes relative to duty cycle. The duty
resistor is part of the output loop in a SEPIC topology cycle expresses the ratio between pulse “ON” duration
setup. The resistor senses the loop current of the (tPULSE) and the pulse period (T).
SEPIC topology consisting of the secondary coil,
rectifier and LED. The loop current of the SEPIC EQUATION 5:
topology consisting of the primary coil, switch and
source power supply is sensed by a separate resistor. D = tPULSE ⁄ T
The primary sense monitors peak current in the primary
loop. The secondary sense monitors average current in
the secondary loop. All these key features are covered
The human eye typically does not notice the variation
by the MCP1631HV Digitally Controlled Programmable
in light above 120 Hz. The eye integrates and interprets
Current Source Reference Design.
the light pulses in terms of brightness that can be
Note that the relative luminosity is directly related to the changed by varying the duty cycle of the source.
driving current which is a quasi-constant linear
PWM control also reduces electrical power
function. The dimming method by variable current
consumption because no series voltage dropping
control has the disadvantage of a shift in color as the
resistors are required. The brightness of the LED may
current level is changed. Most shift of color is around
also be changed linearly by varying the PWM duty
base color and under certain circumstances may be
cycle. The drawbacks of the PWM method are that it
tolerated. The color of an LED at rated forward current
may produce switching spikes and the square pulses
will be different than a dimmed LED operating at 50%
may electrically stress the die.
of rated forward current.

© 2009 Microchip Technology Inc. DS01261A-page 5

AN1261
High-Speed Analog PWM Operation
1 The high-speed analog PWM is used to drive the
Relative Luminous Flux

0.9
SEPIC powertrain switch to regulate the output of the
0.8
0.7 converter. Voltage or current may be regulated
0.6 depending on what is being sensed. For the SEPIC
0.5 LED driver application, the MCP1631HV is always
0.4
0.3 regulating current. A PIC microcontroller is providing
0.2 the oscillator and current references required to obtain
0.1 the desired output current.
0
0 10 20 30 40 50 60 70 80 90 100 The PWM signal starts with the PIC generated
Duty Cycle, D (%) oscillator input. When the oscillator input is low, the
VEXT driver output pin is pulled high, turning on the
FIGURE 6: Linearity of Brightness vs. external N-Channel MOSFET switch.
Duty Cycle in PWM LED drive. The current begins to ramp up in the primary coil
current sense resistor until the voltage exceeds 1/3 of
MCP1631 PWM CONTROLLER the VREF reference value supplied by the
microcontroller. The VREF voltage is internally limited to
The MCP1631/MCP1631V is a high-speed pulse width 0.9V. The 0.9V limit is used as an overcurrent limit. A
modulator (PWM) used to develop intelligent power passive filter is used on the current sense input (CS) to
systems. When combined with a microcontroller, the remove the leading edge turn-on spike associated with
MCP1631/MCP1631V will control the powertrain duty the turn-on of the external power MOSFET. The
cycle to provide output voltage or current regulation. A MCP1631/MCP1631V internal P-Channel MOSFET
functional block diagram of the PWM controller is output driver is powered using a separate PVDD pin to
shown in Figure 7. keep switching noise off of the analog AVDD pin and
The MCP1631/MCP1631V inputs were developed to sensitive current sense circuitry.
be attached to the I/O pins of a microcontroller. By The error amplifier (A1 on datasheet diagram) is used
combining the MCP1631/MCP1631V with a low cost to compare the switch current with the programmed
microcontroller, intelligent LED lighting designs can be reference current. When the switch current is less than
easily developed. the programmed current, the PWM switch driver duty
The microcontroller is used to adjust the output current, cycle increases. When the switch current is greater
switching frequency and maximum duty cycle. It than the programmed current, the PWM switch driver
provides additional features that make the power duty cycle decreases. The PWM duty cycle will track
system more intelligent, robust and adaptable. changes in output current and adjust the duty cycle
accordingly. The external feedback (FB) and
Additional features integrated into the MCP1631HV/ compensation (COMP) pins are used to control the
MCP1631VHV provide signal conditioning and speed of the error amplifiers output response.
protection features for constant current source
applications. The MCP1631HV contains a 10V/V inverting gain
amplifier (A2) for general purpose use. The A2 current
The MCP1631 is available in a 20-Lead 4x4 mm QFN, sense amplifier may be used to amplify the sense
20-Lead TSSOP, or SSOP packages. current in the secondary side of the SEPIC converter or
freewheeling current in a Buck converter.
Powering the MCP1631
For applications that operate from a high voltage input, Sensing Battery Voltage
the MCP1631HV and MCP1631VHV device options Using the microcontroller A/D converter to sense load
may be used. They will operate directly from a +3.5V to voltage requires a low source impedance to perform
+16V input. For these applications, an additional low accurate readings. Low source impedance requires
drop out +5V or +3.3V regulated output is available on low resistance values that would draw excessive
chip and can provide up to 250 mA of current to power current at the converter output, decreasing the
a microcontroller and auxiliary circuits. efficiency. The MCP1631HV integrates another
amplifier (A3) configured as a unity gain buffer. The
buffer provides a high impedance input to the voltage
sense divider while also providing a low impedance
source for the microcontroller A/D converter input.

DS01261A-page 6 © 2009 Microchip Technology Inc.

 AN1261
Overvoltage Protection
Overvoltage (OV) protection is a common voltage
convertor protection feature. It is used to protect the
power switch from excessive voltage if the load is
removed. OV protection also prevents accidental
overvoltage at the load if the programmed output
current exceeds the absorption capability of the load.
OV protection is typically required for any current
source application such as battery chargers and LED
drivers.
The MCP1631HV integrates an internal high speed OV
comparator (C2) that uses a 1.2V reference. When the
voltage on the OVIN pin exceeds the 1.2V reference
threshold, the VEXT output is immediately driven low.
The VEXT output will be re-enabled once the OVIN
voltage has dropped below the 1.2V reference by
50 mV. The 50 mV is the typical hysteresis of the OV
comparator.

OSC Disable Feature

The oscillator disable input (OSCDIS) is used to
asynchronously terminate the PWM VEXT output. This
can be used along with the shutdown (SHDN) pin and
a separate PWM input to modulate current into an LED
for lighting applications.

© 2009 Microchip Technology Inc. DS01261A-page 7

AN1261

MCP1631HV/VHV High Speed PIC PWM

Internal Regulator for MCP1631HV and MCP1631VHV

Options Only; For MCP1631 and MCP1631V AVDD_IN is input

VIN +3.3V or +5.0V VDD Internal

LDO 1.2V VREF
250 mA
VDD
AVDD_OUT / AVDD_IN Overvoltage Comp
- w/ Hysteresis
C2
Shutdown Control +
SHDN
A3 Remains On

OVIN PVDD

VDD OSCDIS

100 kΩ
0.1 µA OT

OSCIN VEXT

UVLO

S PGND
VDD Q

VDD
100 kΩ
CS/VRAMP 10R
+ Q
R
COMP C1
- VDD
- R ISIN
VDD
A2 R
+
FB
-
VREF A1
+ 2R ISOUT
Remove for MCP1631V
R
and MCP1631VHV Options VDD
AGND
2.7V Clamp + VSIN

A3
-

Note 1: On Shutdown, amplifier A3 remains functional.

VSOUT
2: For HV options, internal Low Drop Out Regulator provides
+3.3V or +5.0V bias to VDD.

FIGURE 7: MCP1631 PWM Controller - Functional Block Diagram.

DS01261A-page 8 © 2009 Microchip Technology Inc.

 AN1261
Shutdown Feature We can visualize the waveforms for a better
understanding of how the SEPIC works. We will
The MCP1631/MCP1631V shutdown (SHDN) feature consider a coupled inductor SEPIC operating in
is used to disable the device and to minimize the Continuous Current (CC) mode and refer to Figure 9
quiescent current draw. While shutdown, amplifier A3 for this example. L1 and L2 are wound on the same
remains operational. During shutdown, the device core. The SEPIC switch (Q1) is turned on at the start of
typically draws 4.4 µA from the supply source. the cycle. The current in the primary winding L1 (IL1)
will ramp up at a rate of VIN / L1. The current in the sec-
UnderVoltage Lockout and Thermal ondary winding L2 (IL2) ramps at VCc/L2.
Shutdown The DC voltage across the isolation capacitor CC (VCc)
The MCP1631/MCP1631V has built-in UnderVoltage is equal to VIN. The switch current (Q1) is equal to the
Lockout (UVLO) that ensures the output VEXT pin is sum of IL1 and IL2 during the switch on time. The input
forced to a low state when the input voltage or AVDD is current through L1 is continuous, minimizing the input
below a specified value. This prevents the main ripple current compared to other topologies.
MOSFET switch from being turned on during a
power-up or power-down sequence. The MCP1631/
MCP1631V provides a thermal shutdown protection
feature. If the internal junction temperature of the
device exceeds the overtemperature setpoint limit, the
overtemperature protection feature will pull the VEXT
output low to turn off the MOSFET switch.

SEPIC TOPOLOGY
We shall examine the SEPIC topology and
demonstrate the capability of the MCP1631 family in
current control applications using the SEPIC topology.
The SEPIC Topology (Single Ended Primary Inductive FIGURE 8: SEPIC Topology.
Converter) follows the flyback design, adding a When Q1 turns off, the stored energy in L1 forces the
coupling capacitor between the two windings of a current to continue flowing through the coupling cap
transformer. (Cc) and the rectifier (D1) to the output. A second
The output voltage may be lower or higher than the current flow exists due to L2 releasing it’s stored
input voltage, resulting in buck or boost operation. energy. Current flows from L2 through the rectifier (D1)
to the output. The total rectifier current is the sum of the
This topology can use two separate inductors or a currents flowing through L1 and L2. During the
single transformer with two coupled windings. The switch-off time, current flowing through the rectifier to
coupled inductor solution saves PCB space and the output must also replenish the output capacitor
reduces radiated EMI. Another big advantage of using (COUT). Looking at the output capacitor current
coupled inductors is the fact that only half of the calcu- waveform (ICc), it can be seen that the output current of
lated inductance needs to be used if the inductors are the SEPIC converter is supplied by the output capacitor
coupled. A capacitor connected between the primary when the switch is on and supplied by the inductors
inductor and secondary inductor offers DC isolation when the switch is off.
and protection against a shorted load. The capacitor
clamps the winding leakage inductance energy and The transfer function for the SEPIC converter is derived
eliminates the need for a snubber circuit. The inductor in a similar fashion as other switching power
input smooths the current draw and reduces the topologies. The volt-time product of the magnetics
required input filtering. must be balanced and the charge-time product on all
capacitors must be balanced for each switch period.
A single low-side switch reduces MOSFET drive
voltage requirements and current limit protection The input block components L1 and Q1 act like a
complexity. The switch current can be sensed using a standard boost converter input stage. The output block
ground referenced sense resistor connected to the components L2, D and COUT act like an inverted
MOSFET source pin. buck-boost converter. The goal is to determine the
volt-time product on the magnetics during the switch on
A typical SEPIC converter design that is used with the time and set that equal to the volt-time product on the
MCP1631HV PWM controller is shown in Figure 8. magnetics during the switch off time.
Basic definitions for design are:
• TSW = Switching Period

© 2009 Microchip Technology Inc. DS01261A-page 9

AN1261
• fSW = Switching Frequency As mentioned previously, duty cycle D = tON / (tON +
• tON = Switch on time tOFF) or (1 - D) = tOFF / (tON + tOFF). If we multiply both
• tOFF = Switch off time sides of Equation 8 with 1 / (tON + tOFF), we will
introduce D into the following equation:
• Duty Cycle D = tON / TSW = tON / (tON + tOFF)
As D can have values between 0 and 1, the previous EQUATION 9:
equation can be written as: 1 - D = tOFF / (tON + tOFF).
Balance the inductor volt-time product in the boost V IN × D = ( V Cc + V OUT – V IN ) × ( 1 – D )
stage (L1).
For Q1 Turned on (+ Slope):
EQUATION 10:
EQUATION 6:
V Cc = V IN × ⎛⎝ -------------⎞⎠ – V OUT
1
ΔI L1 ⁄ t ON = V IN ⁄ L 1 1–D

The secondary (L2) inductor current area defined by

For Q1 Turned off (- Slope): the slopes must also be equal. The volt-time product is
solved similar to the primary inductor volt-time product.
EQUATION 7: For Q1 Turned on (+ Slope):

V Cc + V OUT – V IN EQUATION 11:

ΔI L1 ⁄ t OFF = -------------------------------------------
-
L1
ΔI L2 V Cc
The inductor current area defined by the slopes must ----------- = --------
-
t ON L2
be equal for volt-time balance.
Solving Equation 6 and Equation 7 by extracting ΔIL1,
we will obtain VCc.

EQUATION 8:

V IN × t ON = ( V Cc + V OUT – V IN ) × t OFF

FIGURE 9: SEPIC Converter Waveforms.

DS01261A-page 10 © 2009 Microchip Technology Inc.

 AN1261
For Q1 Turned off (- Slope): EQUATION 18:

EQUATION 12: D max = ( V OUT + V D ) ⁄ ( V INmin + V OUT + V D )

ΔI L2 V OUT
----------
- = ------------- The first step in calculating the inductor current is to
t OFF L2
know the maximum output power required. An
As above, we’ll solve Equation 11 and Equation 12 for efficiency estimate of 85% for a SEPIC converter can
VCc. be used to approximate the input current. The average
input current is equal to the input power divided by the
Inductor slope's must be equal for volt-time balance,
input voltage.
so:

EQUATION 19:
EQUATION 13:
P OUT = V OUT × I OUT
t ON × V Cc = t OFF × V OUT

As shown above, duty cycle D can be introduced in EQUATION 20:

Equation 13 by multiplying both sides with 1 / (tON +
tOFF): P IN = P OUT ⁄ ( Efficiency )

EQUATION 14:
EQUATION 21:
V Cc = V OUT × ⎛⎝ -------------⎞⎠
1–D
D
I IN ( AVG ) = P IN ⁄ V IN
Setting Equation 10 equal to Equation 14 yields:
The inductor value is calculated by:
EQUATION 15:
EQUATION 22:
V Cc = V IN × ⎛ -------------⎞ – V OUT = V OUT × ⎛ -------------⎞
1 1–D
⎝ 1 – D⎠ ⎝ D ⎠
L1 = L2 = ( V INmin × D max ) ⁄ ( ΔI L × f SW )
The solution for VOUT/VIN is the transfer function of the
SEPIC converter in continuous mode:
ΔIL is the desired inductor peak to peak ripple current.
A good ΔIL approximation is 20% of the output current.
EQUATION 16: The approximate ripple current for a 1A output current
would be 200 mA.
V OUT
------------- = ⎛ -------------⎞
D
⎝ 1 – D⎠ If L1 and L2 are wound on the same core, which is the
V IN
case of this SEPIC design, the value of inductance in
By rearranging Equation 16 we get VIN = VOUT x (1-D)/ the equation above is replaced by (1/2)L due to mutual
D. Substituting that result into Equation 14 results in: inductance.

EQUATION 17: EQUATION 23:

L
V Cc = V IN --- = ( V INmin × D max ) ⁄ ( 2 ( ΔI L × f SW ) )
2

The voltage across the coupling capacitor (Cc) is equal The maximum inductor current (ripple and peaks for L1
to VIN. This occurs as long as the capacitance of COUT and L2) can be calculated to avoid saturating the
is adequate to supply the output current during the inductor.
switch ON time and the system operates in continuous
current mode. EQUATION 24:
Until now, the diode forward voltage drop (VD) has
been ignored at the output because of its low value. If ΔI L1ON = ( V IN × t ON ) ⁄ ( L ⁄ 2 )
the output voltage is low, the forward voltage drop of
the diode may be significant. When the diode drop is
considered, the maximum duty cycle will be:

© 2009 Microchip Technology Inc. DS01261A-page 11

AN1261
EQUATION 25: The selection of the SEPIC capacitor, Cc, depends on
the RMS current, which is passing through it. The RMS
I L1peak = I IN ( AVG ) + ΔI L1ON ⁄ 2 current in the SEPIC coupling capacitor is mainly
dependant upon output power with some influence by
inductor ripple current. The RMS current in Cc can be
calculated by solving Equation 31 for both states of
EQUATION 26:
switching, tON and tOFF. Equation 32 is the RMS
expression of a trapezoid waveform.
ΔI L2ON = ( V IN × t ON ) ⁄ ( L ⁄ 2 )
EQUATION 31:

EQUATION 27: 2 2
I CcRMS = I CcRMS – ON + I CcRMS – OFF

I L2peak = I OUT ( AVG ) + ΔI L2ON ⁄ 2

EQUATION 32:
The coupled inductor winding currents calculated
above are used to determine the wire size and core of I CcRMS –ON ( OFF ) = D × ( I 12 + ( I 1 × I 2 ) + I 22 ) ⁄ 3
the inductor. High switching frequency has several
advantages: smaller ripple current, lower peak and
RMS current and lower volt-time product on the where I1 is the lowest waveform current and I2 is the
inductor core. This leads to smaller and lower-cost highest waveform current: I1=IAVG + (Iripple / 2) and
solutions. I2=IAVG - (Iripple / 2). For Q1 turned on IAVG = IOUT and
Iripple = tON x 2VIN / (4L). For Q1 turned off IAVG = -IIN
The switch current (IQ1) is equal to the combination of and Iripple = tOFF x 2VIN / (4L).
the winding currents during the switch on time. When
the switch is turned on, it conducts the current from L1 To estimate the size for the coupling capacitor, the
and L2. capacitor derivative equation can be used.

EQUATION 28: EQUATION 33:

Δt
I switch ( Q1 ) = I L1 + I L2 C = I C × -------
ΔV

The peak Q1 switch voltage is equal to VIN + VOUT for For the SEPIC converter, the coupling capacitor
the SEPIC converter. Any leakage inductance voltage voltage is approximated as a DC value when deriving
spike is clamped through the output diode by the output the duty cycle. The ripple voltage should be no more
capacitor. A switch voltage rating for applications than a few percent (5% in practice) of the voltage
should be at least: across the capacitor or the input voltage. The SEPIC
capacitor must be rated for a large RMS current in
EQUATION 29: relation to the output power. As the output power
increases, the capacitor ripple current will increase as
well. This makes the SEPIC converter better suited for
V switch ( Q1 ) = V INmax + V OUTmax
lower power applications where the RMS current
through the capacitor is relatively small. The worst case
The maximum reverse voltage across the SEPIC diode RMS current in the coupling capacitor will occur at
occurs during the switch ON time. The cathode of the maximum output power and minimum input voltage.
schottky diode is connected to VOUT, the anode of the
schottky diode is connected to the SEPIC coupling EQUATION 34:
capacitor. The voltage across the coupling capacitor
voltage is equal to VIN. When Q1 is ON, the voltage I OUT D max
C c = ⎛ -------------⎞ × ⎛ -------------⎞
across the diode is: ⎝ ΔV Cc⎠ ⎝ f SW ⎠

EQUATION 30: The same algorithm should to be applied to the output

capacitor. Ripple voltage (ΔV or Vripple) should be no
V D = V OUT + V IN = ( V OUT + V Cc ) more than a few percent of the voltage across the
capacitor. When the switch is on (tON = D / fSW), the
inductor L1 is charging and the output current is
A schottky diode is recommended for low voltage supplied by the output capacitor COUT.
applications because of the low forward voltage drop.
The lower forward drop improves converter efficiency.

DS01261A-page 12 © 2009 Microchip Technology Inc.

 AN1261
EQUATION 35: The MCP1631HV Digitally Controlled Programmable
Current Source board (MCP1631RD-DCPC1) is used
I OUT to drive and dim one or more power LEDs in series or
C OUT = ⎛⎝ ----------------⎞⎠ × ⎛⎝ --------⎞⎠
D
V ripple f SW parallel topology. The board uses the PIC16F616
microcontroller to generate the proper dimming ratio.
The input capacitor CIN should be capable of handling The default dimming ratio is set at 10:1. The dimming
the input RMS current. The input current waveform is ratio and maximum current are selected before
continuous and triangular. The inductive input ensures compiling the source code. The firmware on the
that the input capacitor sees low ripple currents from development board is programmed for 700 mA of LED
the power supply. The input capacitor provides a low drive current. An LED load capable of handling 700 mA
impedance source for the SEPIC converter in cases of current is required to use the supplied firmware.
where the power source is not immediately adjacent to Lower drive currents may be obtained by downloading
the SEPIC powertrain. the firmware source code from Microchip’s website and
changing the drive current parameter in the header file.
MCP1631HV PROGRAMMED The source code is then compiled and downloaded to
the development board. The minimum programmed
CURRENT SOURCE REFERENCE current that the board can generate is 70 mA. An LED
DESIGN FOR DIMMING POWER LEDS from XLamp® XR-E series (Cree®) can be used to
evaluate the board. The LED can be supplied with
Microchip has developed a reference design for a
700 mA or 1000 mA, based on the selected model. As
dimmable power LED lighting application to
example, the XRWWHT-L1-0000-009E7 warm white
demonstrate the capability of the MCP1631HV PWM
LED can be driven at 700 mA maximum.
Controller.
LED indicators on the reference design provide status
and fault indication (i.e.: overvoltage, overtemperature,
thermal shutdown).
TABLE 1: DEFAULT OUTPUT CAPABILITY OF MCP1631HV LED DRIVER REFERENCE BOARD
Programmed output Max. Output power Over voltage Thermal
Dimming ratio
current (Note 1) (Note 1) protection shutdown (Note 2)
700 mA 10:1 8.5W 9V 40°C
Note 1: Maximum programmed output current in the source code is 1A. Total power output should not exceed
8.5W.
2: This option is not enabled in default compiled .hex file available on the Microchip website.
The reference board is capable of driving 1A maximum reference voltage, the feedback voltage and the current
@ 8.5W output power. Efficiency is typical 85% at sense. The output drive signal is a square-wave pulse
maximum power. The default parameters are displayed used to drive the gate of the powertrain MOSFET
in Table 1. switch. The power train topology used for the
An input terminal block is provided for connecting the MCP1631HV Programmable Current Source is a
input source voltage to the board. An output header is Single-Ended Primary Inductive Converter (SEPIC).
also provided for attaching the load LED and an The MCP1631HV Programmable Current Source has
external 10k thermistor. The thermistor should be the following features:
mounted on the LED heat sink near the LED. The • Input Operating Voltage Range: +3.5V to +16V
temperature monitoring feature must be enabled in the
• Maximum output current 700 mA with dimming
source code before compiling. The default value for
ratio 10:1 (70 mA/step for pre-compiled.hex file)
overtemperature shutdown is 40oC. Temperature
• Drive one or more power LEDs. Limited by OVP
monitoring is not enabled in the default .hex file. The
and 8.5W power capability
recommended thermistor is EPCOS Inc. PN
B57703M0103G040. • Optional battery charger software on Microchip
website for charging Li-Ion, NiMH, and NiCd
A Microchip ICD2 / PICkit 2 compatible programming battery packs. Charge 1 or 2 single cell Li-Ion
header is available for updating the reference board batteries in series or 1 to 4 single cell NiMH or
firmware contained within the PIC16F616. The NiCd batteries in series
PIC16F616 microcontroller is used to regulate output • Maximum output capability for all modes of
voltage, current, switching frequency and maximum operation: 1A @ 8.5W
duty cycle. The MCP1631HV generates the required
• ON/OFF/Dimming switch
switch duty cycle and provides fast overcurrent
• Status Indication
protection based upon the various external inputs.
External signals include the reference oscillator, the • Hardware and software overvoltage shutdown set
at 9.0V

© 2009 Microchip Technology Inc. DS01261A-page 13

AN1261
compared with the switch current sense and an artificial
compensation ramp to modulate the switch drive
0.7 signal. The artificial compensation ramp is generated
0.6 from the PIC16F616 OSC output signal to prevent the
Output Current (A)

0.5 system from switching between continuous and

0.4 discontinuous operation. The bi-modal operation would
0.3 cause the duty cycle to vary when D > 50%. The
3W White LED @ 700mA
0.2 3W Amber LED @ 700mA
artificial ramp voltage supplements the switch current
0.1 3W Amber LED @ 350mA sense voltage to guarantee the CS signal rises above
0
2 series 3W White LEDs @ 700mA
the reference signal before the end of the integrator
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 period. When the CS signal rises above the reference
Input Voltage (V) signal, the MOSFET switch will be turned off, reseting
the integrator. The artificial ramp slope for the SEPIC
FIGURE 10: Output Current Regulation will be the same as the slope of the current during the
vs. Input Voltage in Driving LED Mode. OFF time of the oscillator. The artificial ramp circuit
block consists of C10, R9, R13, and Q2 which is driven
Operation with the OSC signal.
The MCP1631HV Digitally Controlled Programmable A resistor divider network composed of R5, R8, and R15
Current Source Reference Design has a red/green dual is used for load voltage sensing. R5 senses
color LED (D2) to indicate status and faults. The green overvoltage conditions and feeds directly into the
LED will flash with a one second period when the MCP1631HV overvoltage input pin. R8 senses the load
system is activated and current is being supplied to the voltage. The load voltage tap at R8 is connected to the
load. MCP1631HV internal voltage follower amplifier via the
Press button S1 to enable the output and drive the LED VSIN pin. The corresponding VSout signal is fed to the
with the minimum programmed current of 70 mA. Each PIC16F616 A/D RA2 port.
additional press of S1 will increment the drive current When a thermistor is used, it is placed across J2-4 and
by 10% of the final programmed value. Press and hold J2-5. The thermistor forms a voltage divider with R1, a
S1 for at least 2 seconds until the red LED indicator is 10 kΩ resistor. The voltage across the thermistor is fed
activated to switch the output current off. If a fault to the A/D port RA4 of the PIC16F616. If an
condition such as no load, overvoltage, or overtemperature condition is detected, the
overtemperature is present when attempting to turn the microcontroller will disable the PWM signal of
output on, the system will stay off and the fault MCP1631.
condition will be indicated by the flashing red indicator
LED. The faults are coded as following: The OSC signal is generated directly by the
PIC16F616 PWM hardware at port RC5.
• Red LED flashes at 1 Hz rate if an overtempera-
ture of the LED case has been detected. The fault
is Thermal Shutdown MCP1631HV Digitally Controlled
• Red LED flashes at 2 Hz rate if an overvoltage or Programmable Current Source Reference
a missing/open load has been detected. The fault Design Firmware
is Overvoltage Shutdown
The source code for Microchip’s MCP1631HV Digitally
An overvoltage/overtemperature fault condition will be
Controlled Programmable Current Source Reference
verified 5 times by the microcontroller firmware before
Design was developed using mikroElektronica’s Mik-
the fault is validated and the system shuts down.
roC Compiler for PIC Processors. The software has
Two parallel resistors, R10 and R11, are used to sense
conditional compiler options for user selectable
the MOSFET switch (Q1) current (Figure 11). The
parameters and function. The LED driver is the default
current sense signal is applied to the CS comparator
compiler option. The NiMH/NiCd and Li-Ion battery
input of MCP1631HV. The load current is equal to the
charger options are selectable at compile time by
current in the secondary inductor L1:B. Sense resistors
modifying a parameter in the header file. The mikroC
R2 and R3 are used to sense the load current. The load
compiler may be downloaded at: http://
sense signal is amplified by the ISin current sense
www.mikroe.com.
amplifier before entering the feedback pin (FB). A
reference voltage signal from the PIC16F616 supplies Please consult the mikroC website and compiler Help
the VREF current reference voltage for the for more information about the compiler.
MCP1631HV. The pulse width modulated VEXT signal
that drives the MOSFET switch is generated by the
MCP1631HV using the microcontroller current
reference PWM signal and the load current feedback
signal. The current feedback signal is integrated with
the current reference signal. The integrated signal is

DS01261A-page 14 © 2009 Microchip Technology Inc.

 AN1261

M
8 3
7 2
6 1
5
4

2
3
1

FIGURE 11: MCP1631HV Programmable LED Current Source Reference Design Schematic.

© 2009 Microchip Technology Inc. DS01261A-page 15

AN1261
The C source code file is 00234R1.c and the header
file is 00234R1.h. These two files are referenced by
the mikroC project file: 00234R1.ppc.
To setup the parameters, the user starts with the
00234R1.h file. Set the function of interest for the
reference board as enabled. The example below is for
an LED driver.
/* Enable support for Lithium Ion batteries */
#define LION_SUPPORT DISABLED

/* Enable support for Nickel Metal Hydride batteries */

#define NIMH_SUPPORT DISABLED

FIGURE 12: Top Silk and Pads of /* Enable support for LED Driver */
#define LED_DRIVER_SUPPORT ENABLED
Reference Design.
The charging/driving current profile parameters are set
in separate sections of the header file for the project.
The LED driver overvoltage value is set to 9V by
parameter LED_DRIVER_OVER_VOLTAGE. The LED
driver starting current value is set to 70 mA by
parameter LED_DRIVER_CONDITION_CURRENT.
The LED maximum current value is set to 700 mA by
parameter LED_DRIVER_CURRENT. To evaluate the
board, a Cree XREWHT-L1-0000-009E7 with warm
white LEDs that can be driven at 700 mA maximum
should be used. The overtemperature setpoint is
enabled by setting OVERTEMPERATURE_SUPPORT to
ENABLED. The default overtemperature set point is
FIGURE 13: Top Layer of Reference
+40°C.
Design.
A 10k NTC thermistor must be connected between
J2-4 and J2-5 of the J2 connector when
overtemperature support is enabled. The thermistor
should be placed as close as possible to the LED
mounting area. The number of LEDs value is set to 1
by parameter NUMBER_OF_CELLS_DEFAULT. These
default values are for a typical 3W power LED.
#if (LED_DRIVER_SUPPORT == ENABLED)
// define the LED Driver characteristics in millivolts and
// milliamps). Default values are for one 3W/750mA Power LED
// with max forward voltage 4.5V

/* LED overvoltage */
#define LED_DRIVER_OVER_VOLTAGE ((int)(9000.0/BATTERY_MV_PER_BIT))

/* cell conditioning/startup current = 70 mA */

#define LED_DRIVER_CONDITION_CURRENT ((int)(70.0/BATTERY_MA_PER_BIT))
FIGURE 14: Bottom Layer of Reference
Design. /* total charge current = 700 milliamps */
#define LED_DRIVER_CURRENT ((int)(700.0/BATTERY_MA_PER_BIT))

The reference board PIC16F616 microcontroller has /* LED overtemperature limit = 40 deg C (40 / 0.1447) = 311 */
been pre-programmed with the LED driver option. The #define LED_DRIVER_OVTEMP_SETPOINT ((float)(THERMISTOR_OHMS_40C))

reference board may also be operated as a battery /* ENABLE or DISABLE the overtemperture detection */

charger. The user may re-configure and compile the #define OVERTEMPERATURE_SUPPORT DISABLED

source code for Li-Ion or NiMH/NiCd battery charging

/* default number of LEDs in series topology */
instead of LED driving. #define NUMBER_OF_CELLS_DEFAULT 1

The firmware packet has several files for LED dimming

#endif /* LED_DRIVER_SUPPORT */
and battery charging.
• LED = 00234R1.hex
• Li-Ion = 00234R1-LiIon_Charger.hex
• NiMH = 00234R1-NiMH_NiCd_Charger.hex
• NiCd = 00234R1-NiMH_NiCd_Charger.hex

DS01261A-page 16 © 2009 Microchip Technology Inc.

 AN1261
The OSC reference signal supplied to the MCP1631HV Example:
PWM controller is set by the PIC16F616. The The MCP1631RD-DCPC1 Reference Board needs to
frequency may be set to 200 kHz or 500 kHz. The drive an LED string composed of 2 white LEDs in series
parameters are set in the main C file 00234R1.c. topology with a 350 mA maximum forward current. The
/* define Oscillator Output PWM frequency Prescale Value board is powered from a 12V source.
for PR2 */
#define OSC_500KHZ_PR2_REG 0x03 Solution:
#define OSC_200KHZ_PR2_REG 0x09
#define OSC_PR2_REG OSC_500KHZ_PR2_REG
The system will need to provide 1 LED x 350 mA =
350 mA of drive current, because the LEDs are
The oscillator duty cycle is set to 25% if the SEPIC
connected in series topology.
reference board must work in buck-boost mode. The
25% duty cycle allows the system to boost the output IREF_INCREMENTS is set to 10 to allow 10
voltage. The parameter is in the 00234R1.h header incremental current steps.
file. LED_DRIVER_CONDITION_CURRENT is set to the
/* charge oscillator duty cycle */ startup current value of 35 mA.
#define OSCILLATOR_DUTY_CYCLE 25 LED_DRIVER_CURRENT is set to the final current of
The number of incremental LED current steps is set to 350 mA.
10 by the parameter IREF_INCREMENTS. Each S1 LED_DRIVER_OVER_VOLTAGE is set to 8V which is
button press will increment the output current by a the sum of maximum forward voltages of the two white
factor of (1 / IREF_INCREMENTS) times the LEDs.
programmed current value up to the programmed
current value. The system will only need to operate in buck mode
because the 12V power supply voltage is greater than
/* define Reference Current Increments when ramping
up current at startup */
the 8V load voltage.
/* rate is 1 increment every second in Charger Mode OSCILATOR_DUTY_CYCLE is set to 50% for buck
or every button push for LED Driver */ mode.
#define IREF_INCREMENTS 10
OVERTEMPERATURE_SUPORT is set to DISABLED
because cooling the LED string is not an issue.
The flowchart of the default source code is given in
Figure 15. When powered on, the microcontroller
initializes the ports and the default system parameters.
If push-button S1 is pressed, the load is supplied with
condition current. Each additional press of S1 will
increase the current until the maximum programmed
current is reached. Further presses will not increase
the current beyond the maximum value. When S1 is
pressed and held for more than 2 seconds, the red LED
will turn on and the output will turn off.

© 2009 Microchip Technology Inc. DS01261A-page 17

AN1261

START

Initialize: Processor, Parameters

(ILEDmin = 70 mA, ILEDmax = 700 mA),
Timers, Sample ADC()

No

S1 Pressed?

Yes

ILEDmin

Yes OVP Test

VOUT> 9.0V?

No

Shutdown
No
S1 Pressed?

Yes

No
No > 2s?
OVP S1 pressed =
time retries
>5
Yes Increment IREF(70 mA)

Yes
Shutdown
Fault

END

FIGURE 15: Firmware Flow Diagram for LED Driver With Dimming.

DS01261A-page 18 © 2009 Microchip Technology Inc.

 AN1261
CONCLUSION REFERENCES
For applications that require intelligent power 1. Terry Cleveland, “Using the MCP1631 Family to
management solutions like LED drivers, the develop Low-Cost Battery Chargers”, AN1137,
combination of a microcontroller and the MCP1631 Microchip Technology Inc.
high-speed PWM is very powerful. It brings the 2. Terry Cleveland, “New components and design
programmability benefits of the microcontroller and methods bring intelligence to battery chargers”,
adds the performance of a high speed analog PWM. AN960, Microchip Technology Inc.
The analog PWM will respond to changes in input
voltage and output current very quickly. The 3. ***, “MCP1631/HV/MCP1631V/VHV Data Sheet”,
microcontroller is used for programmability, for DS22063, Microchip Technology Inc.
establishing LED current, for monitoring the circuit for 4. Markus Hofmann, “Comparation of LED Circuits”,
fault conditions, and for taking the appropriate action in Application Notes, OSRAM Opto Semiconductor.
the event of a specific fault.
5. ***, “Controlling Luminous Intensity of LEDs”,
Application Notes, STS-KSE3692, NIKIA.
6. Hubert Ott, Ludwig Plotz, “Dimming InGaN LEDs”,
Application Notes, OSRAM Opto Semiconductor.
7. J. Reill, M. Wicke, “Dimming LEDs with respect of
grouping current”, Application Notes, OSRAM Opto
Semiconductor.
8. ***, “Thermal design consideration for Luxeon 5W
Power Light sources”, AB23, Lumileds Lighting.
9. ***, “Thermal Management guide”, Rev. 3,
Application Notes, Seoul Semiconductor.
10. Rainer Huber, “Thermal Management of SMT
LED”, Application Notes, OSRAM Opto
Semiconductor.
11. ***, “Electrical Drive Management”, Rev. 1,
Application Notes, Seoul Semiconductor.
12. ***, The National Technology Laboratory, website:
www.netl.doe.gov.

© 2009 Microchip Technology Inc. DS01261A-page 19

AN1261
NOTES:

DS01261A-page 20 © 2009 Microchip Technology Inc.

Note the following details of the code protection feature on Microchip devices:
• Microchip products meet the specification contained in their particular Microchip Data Sheet.

• Microchip believes that its family of products is one of the most secure families of its kind on the market today, when used in the
intended manner and under normal conditions.

• There are dishonest and possibly illegal methods used to breach the code protection feature. All of these methods, to our
knowledge, require using the Microchip products in a manner outside the operating specifications contained in Microchip’s Data
Sheets. Most likely, the person doing so is engaged in theft of intellectual property.

• Microchip is willing to work with the customer who is concerned about the integrity of their code.

• Neither Microchip nor any other semiconductor manufacturer can guarantee the security of their code. Code protection does not
mean that we are guaranteeing the product as “unbreakable.”

Code protection is constantly evolving. We at Microchip are committed to continuously improving the code protection features of our
products. Attempts to break Microchip’s code protection feature may be a violation of the Digital Millennium Copyright Act. If such acts
allow unauthorized access to your software or other copyrighted work, you may have a right to sue for relief under that Act.

DS01261AInformation contained in this publication regarding Trademarks

device applications and the like is provided only for your conve- The Microchip name and logo, the Microchip logo, dsPIC,
nience and may be superseded by updates. It is your responsi-
KEELOQ, KEELOQ logo, MPLAB, PIC, PICmicro, PICSTART,
bility to ensure that your application meets with your
rfPIC and UNI/O are registered trademarks of Microchip
specifications. MICROCHIP MAKES NO REPRESENTA- Technology Incorporated in the U.S.A. and other countries.
TIONS OR WARRANTIES OF ANY KIND WHETHER
EXPRESS OR IMPLIED, WRITTEN OR ORAL, STATUTORY FilterLab, Hampshire, HI-TECH C, Linear Active Thermistor,
OR OTHERWISE, RELATED TO THE INFORMATION, MXDEV, MXLAB, SEEVAL and The Embedded Control
INCLUDING BUT NOT LIMITED TO ITS CONDITION, Solutions Company are registered trademarks of Microchip
QUALITY, PERFORMANCE, MERCHANTABILITY OR Technology Incorporated in the U.S.A.
FITNESS FOR PURPOSE. Microchip disclaims all liability Analog-for-the-Digital Age, Application Maestro, CodeGuard,
arising from this information and its use. Use of Microchip dsPICDEM, dsPICDEM.net, dsPICworks, dsSPEAK, ECAN,
devices in life support and/or safety applications is entirely at ECONOMONITOR, FanSense, HI-TIDE, In-Circuit Serial
the buyer’s risk, and the buyer agrees to defend, indemnify and Programming, ICSP, ICEPIC, Mindi, MiWi, MPASM, MPLAB
hold harmless Microchip from any and all damages, claims, Certified logo, MPLIB, MPLINK, mTouch, Omniscient Code
suits, or expenses resulting from such use. No licenses are Generation, PICC, PICC-18, PICkit, PICDEM, PICDEM.net,
conveyed, implicitly or otherwise, under any Microchip PICtail, PIC32 logo, REAL ICE, rfLAB, Select Mode, Total
intellectual property rights. Endurance, TSHARC, WiperLock and ZENA are trademarks
of Microchip Technology Incorporated in the U.S.A. and other
countries.
SQTP is a service mark of Microchip Technology Incorporated
in the U.S.A.
All other trademarks mentioned herein are property of their
respective companies.
© 2009, Microchip Technology Incorporated, Printed in the
U.S.A., All Rights Reserved.
Printed on recycled paper.

Microchip received ISO/TS-16949:2002 certification for its worldwide

headquarters, design and wafer fabrication facilities in Chandler and
Tempe, Arizona; Gresham, Oregon and design centers in California
and India. The Company’s quality system processes and procedures
are for its PIC® MCUs and dsPIC® DSCs, KEELOQ® code hopping
devices, Serial EEPROMs, microperipherals, nonvolatile memory and
analog products. In addition, Microchip’s quality system for the design
and manufacture of development systems is ISO 9001:2000 certified.

© 2009 Microchip Technology Inc. DS01261A-page 21

 WORLDWIDE SALES AND SERVICE
AMERICAS ASIA/PACIFIC ASIA/PACIFIC EUROPE
Corporate Office Asia Pacific Office India - Bangalore Austria - Wels
2355 West Chandler Blvd. Suites 3707-14, 37th Floor Tel: 91-80-3090-4444 Tel: 43-7242-2244-39
Chandler, AZ 85224-6199 Tower 6, The Gateway Fax: 91-80-3090-4080 Fax: 43-7242-2244-393
Tel: 480-792-7200 Harbour City, Kowloon India - New Delhi Denmark - Copenhagen
Fax: 480-792-7277 Hong Kong Tel: 45-4450-2828
Tel: 91-11-4160-8631
Technical Support: Tel: 852-2401-1200 Fax: 45-4485-2829
Fax: 91-11-4160-8632
http://support.microchip.com
Fax: 852-2401-3431 France - Paris
Web Address: India - Pune
Australia - Sydney Tel: 91-20-2566-1512 Tel: 33-1-69-53-63-20
www.microchip.com
Tel: 61-2-9868-6733 Fax: 91-20-2566-1513 Fax: 33-1-69-30-90-79
Atlanta Fax: 61-2-9868-6755
Duluth, GA Japan - Yokohama Germany - Munich
China - Beijing Tel: 81-45-471- 6166 Tel: 49-89-627-144-0
Tel: 678-957-9614
Tel: 86-10-8528-2100 Fax: 49-89-627-144-44
Fax: 678-957-1455 Fax: 81-45-471-6122
Fax: 86-10-8528-2104 Italy - Milan
Boston Korea - Daegu
China - Chengdu Tel: 39-0331-742611
Westborough, MA Tel: 82-53-744-4301
Tel: 86-28-8665-5511 Fax: 82-53-744-4302 Fax: 39-0331-466781
Tel: 774-760-0087
Fax: 86-28-8665-7889 Netherlands - Drunen
Fax: 774-760-0088 Korea - Seoul
China - Hong Kong SAR Tel: 82-2-554-7200 Tel: 31-416-690399
Chicago
Itasca, IL Tel: 852-2401-1200 Fax: 82-2-558-5932 or Fax: 31-416-690340
Tel: 630-285-0071 Fax: 852-2401-3431 82-2-558-5934 Spain - Madrid
Fax: 630-285-0075 China - Nanjing Tel: 34-91-708-08-90
Malaysia - Kuala Lumpur
Tel: 86-25-8473-2460 Fax: 34-91-708-08-91
Cleveland Tel: 60-3-6201-9857
Independence, OH Fax: 86-25-8473-2470 Fax: 60-3-6201-9859 UK - Wokingham
Tel: 216-447-0464 China - Qingdao Tel: 44-118-921-5869
Malaysia - Penang
Fax: 216-447-0643 Tel: 86-532-8502-7355 Fax: 44-118-921-5820
Tel: 60-4-227-8870
Dallas Fax: 86-532-8502-7205 Fax: 60-4-227-4068
Addison, TX China - Shanghai Philippines - Manila
Tel: 972-818-7423 Tel: 86-21-5407-5533 Tel: 63-2-634-9065
Fax: 972-818-2924 Fax: 86-21-5407-5066 Fax: 63-2-634-9069
Detroit China - Shenyang Singapore
Farmington Hills, MI Tel: 86-24-2334-2829 Tel: 65-6334-8870
Tel: 248-538-2250 Fax: 86-24-2334-2393 Fax: 65-6334-8850
Fax: 248-538-2260
China - Shenzhen Taiwan - Hsin Chu
Kokomo Tel: 86-755-8203-2660 Tel: 886-3-6578-300
Kokomo, IN Fax: 86-755-8203-1760
Tel: 765-864-8360 Fax: 886-3-6578-370
Fax: 765-864-8387 China - Wuhan Taiwan - Kaohsiung
Tel: 86-27-5980-5300 Tel: 886-7-536-4818
Los Angeles Fax: 86-27-5980-5118 Fax: 886-7-536-4803
Mission Viejo, CA
Tel: 949-462-9523 China - Xiamen Taiwan - Taipei
Fax: 949-462-9608 Tel: 86-592-2388138 Tel: 886-2-2500-6610
Fax: 86-592-2388130 Fax: 886-2-2508-0102
Santa Clara
Santa Clara, CA China - Xian Thailand - Bangkok
Tel: 408-961-6444 Tel: 86-29-8833-7252 Tel: 66-2-694-1351
Fax: 408-961-6445 Fax: 86-29-8833-7256 Fax: 66-2-694-1350
Toronto China - Zhuhai
Mississauga, Ontario, Tel: 86-756-3210040
Canada Fax: 86-756-3210049
Tel: 905-673-0699
Fax: 905-673-6509

03/26/09

DS01261A-page 22 © 2009 Microchip Technology Inc.