V25N3 - NOVEMBER

November 2015
I N
T H I S
I S S U E
CAN bus transceivers
operate from 3.3V or 5V
and withstand ±60V faults
13
low power IQ modulator
for digital communications
Volume 25 Number 3
Control Individual LEDs in Matrix
Headlights with Integrated 8-Switch
Flicker-Free Driver
Keith Szolusha
20
low profile supercapacitor
power backup with input
current limiting 25
boost 12V to 140V with a
single converter IC 30
LEDs combine design flexibility with practical, robust circuitry,
enabling automotive designers to produce striking headlight designs
matched by exceptionally long life and performance. Automobile
designers are increasingly incorporating LEDs in lighting because
they can be arranged in distinctive eye-catching designs—helping
distinguish new models from old, or high end from economy.
There is no question that automobile LED lighting has arrived, but it has not yet reached its full
potential. Future models will feature more LED lights, including new shapes and colors, and more
control over the individual LEDs. Simple strings of LEDs
will give way to matrices of LEDs that can be individually
dimmed via computer control, enabling unlimited real-time
pattern control and animation. The future has arrived:
Linear Technology’s LT®3965 matrix LED driver makes it
easy to take the next step in automotive lighting design.
I 2C CONTROL OF EIGHT POWER SWITCHES WITH A
SINGLE IC
A basic LED headlight design operates with uniform LED
current, and thus, uniform brightness. But this leaves much
of the LEDs’ potential on the table. Matrix headlights take
advantage of the innate abilities of LEDs by enabling control
of the brightness of individual LEDs within LED strings.
It is not difficult, in theory, to address the individual LEDs in
a matrix via computer-controlled power switches, allowing
individual LEDs to be turned on or off, or PWM dimmed,
The LT3965: limitless control of automotive matrix headlights and lighting systems
w w w. li n ea r.com
(continued on page 4)
Linear in the News
In this issue...
COVER STORY
LINEAR TECHNOLOGY CO-FOUNDERS RECEIVE
ACE LIFETIME ACHIEVEMENT AWARD
Control Individual LEDs in Matrix Headlights
with Integrated 8-Switch Flicker-Free Driver
Linear Technology co-founders Bob Swanson and Bob Dobkin received
UBM Canon’s EDN and EE Times ACE Lifetime Achievement Award
at a ceremony at the Embedded Systems Conference in Santa Clara,
California in July. The award is presented annually to individuals for
their contributions to the electronics industry, selected by a panel of
academic and business leaders, and editors of EDN and EE Times.
Keith Szolusha
1
DESIGN FEATURES
CAN Bus Transceivers Operate from 3.3V
or 5V and Withstand ±60V Faults
Ciaran Brennan
13
Low Power IQ Modulator for Digital Communications
Bruce Hemp and Sunny Hsiao
20
Low Profile Supercapacitor Power Backup
with Input Current Limiting
David Salerno
25
DESIGN IDEAS
What’s New with LTspice IV?
Gabino Alonso
28
Boost 12V to 140V with a Single Converter IC
Victor Khasiev
30
back page circuits
32
Bob Swanson and Bob Dobkin founded Linear Technology 34 years
ago, with a goal of building unique high performance analog ICs.
The company now employs 4,865 people in over 50 locations worldwide. Linear’s name has become synonymous with innovative, high
quality analog solutions in a broad range of application areas.
EDN Senior Technical Editor Steve Taranovich stated, “Bob Swanson and
Bob Dobkin are unique as individuals in their decades-deep contributions
to integrated electronics, as well as their longevity in Silicon Valley. They
are also exceptional as a team because of their unprecedented continuity of
management at a semiconductor company, as well as their refreshing style of
management, amplifying Linear Technology’s engineering talent. For these
reasons and many more, EE Times and EDN chose to honor Bob Swanson
and Bob Dobkin with the 2015 ACE Lifetime Achievement Award.”
Steve Taranovich continued, “The talents and experiences of these two leaders,
in my estimation, have brought about a corporate culture unlike most of
the companies in the electronics business today. This culture encourages
innovation and strongly values and recognizes the company’s engineering
talent so that when a good idea emerges, management recognizes it and
‘gets out of the way’ to allow the engineer to bring it to fruition.”
The following comments are excerpted from the EDN interview
with Bob Swanson and Bob Dobkin: www.linear.com/46876.
Bob Swanson: “If you were an Analog Aficionado, Linear Technology looked
like a bunch of great technical guys just doing analog and they wanted to
be part of that. After 34 years, we are still the same. We can hire as many
good people as we can afford to hire. So the great thing we have going for
us is that we have a disproportionate share of really innovative people. We
leverage that because the world needs innovative analog solutions.”
2 | November 2015 : LT Journal of Analog Innovation
Linear in the news
They are strong, talented and intelligent
leaders, but with a touch of humility and
compassion for their employees from
which other companies can learn a great
deal. The book entitled, The Company
That No One Leaves gives wonderful
insight into one of the key reasons this
company has had such success over the
last 34 years.” www.linear.com/46751
Linear Technology co-founders
Bob Swanson and Bob Dobkin
received UBM Canon’s EDN
and EE Times ACE Lifetime
Achievement Award at a
ceremony at the Embedded
Systems Conference.
Bob Dobkin: “One the things we do is
Bob Swanson: “Smart, happy people
to hire engineers who want to innovate and build products. And then we
don’t get in their way. So they build
products, they like what they are doing,
and that works well within Linear.”
probably can be innovators. I tell this
story—about five or six years ago somebody asked Steve Jobs: ‘You guys are
acknowledged to be the best at R&D and
innovation, and you’ve got the smallest
R&D budget.’ And he said, ‘All you need
is a handful of real innovators and you
can be successful. And I’ve got a handful
of really good innovators and that’s all I
need.’ So R&D is not an arms race. R&D
is about having innovators. Going from
100 engineers to 200 does not automatically double your innovation. If you have
25 innovators and you can get two or
three more a year, you’re in good shape.”
Bob Dobkin: “If you do some products right,
they will sell for over 30 years. If we do
some products as well as they can be
done, you never have to do them again.”
Bob Swanson: “Obviously we have great
products, but competitors have great products too. In this analog-challenged world,
we have been so good at transferring our
knowledge to customers. I will occasionally see the big customers at social events.
They always tell me how much they
depend upon our design and field people.
Our field people are brilliant FAEs.”
Bob Swanson: “One of the things that
Bob Dobkin told me was that power
was analog. So take an automobile
with hundreds of processors—every
one of them needs a power supply.
And it is typically a power supply that
is way more difficult than it was 20
years ago. The challenge in this ‘explosion of electronics’ in the automobile
seems to me to be as much about
solving analog issues as digital issues.”
Bob Dobkin: “One way we are unique is
that the engineering group comes up
with the product ideas. We don’t go
and do a marketing survey. Our engineers visit customers and figure out
new products. We’re excited to work
on things that the customers want.”
Steve Taranovich concluded: “During
this interview with these two industry
icons, I sensed that they know their place
in the industry and in their company,
and effectively use the talent of their
employees in a way that I personally have
never seen in my 42 years in electronics.
AWARDS
•The LTC®2983 digital temperature
measurement IC received the
EDN/EE Times Analog Ultimate
Product ACE Award.
•The LTC2000 16-bit, 2Gsps DAC
received the Best Product Award
from EDN China in the Analog
and Mixed Signal IC category.
CONFERENCES & EVENTS
The Wireless Congress: Systems & Applications,
Konferenzzentrum, Munich, Germany, Nov. 17–18—
Presenting energy harvesting and wireless sensor networks. Joy Weiss presents
“Low Power Wireless Sensor Networks
for IoT.” www.wireless-congress.com
Energy Harvesting & Storage Conference,
Santa Clara Convention Center, Santa Clara,
California, Nov. 18–19, Booth N28—Presenting
energy harvesting and wireless sensor
networks. Ross Yu on “Low Power
Wireless Sensor Networks for IoT”;
Tony Armstrong on “Energy Harvesting:
Battery Life Extension & Storage.”
www.idtechex.com/energy-harvesting-usa
3rd Annual Analog Gurus Conference, Tokyo
Conference Center Shinagawa, Tokyo, Japan,
Nov. 18—Linear’s analog gurus present.
CTO Bob Dobkin on“Inside Precision
Voltage References,” Steve Pietkiewicz,
VP, Power Management products on
“High Performance µModules®” and Bob
Reay, VP, Mixed Signal products on“High
Precision Temperature Measurement.” n
November 2015 : LT Journal of Analog Innovation | 3
When combined with a suitable constant-current LED driver, the matrix dimmer LED
driver allows the individual LEDs to be computer-controlled in headlights, daytime
running lights, brake and tail lights, side-bending lights, and other trim lighting.
(LT3965, continued from page 1)
to create unique patterns and functions. Each LED (or segment of LEDs)
requires either its own converter or its
own shunt power switch. It is possible
to build a matrix driver with traditional
driver/converter ICs that include a serial
communications feature, but once more
than two or three switches are needed
for a matrix of LEDs, designing a discrete
component solution becomes challenging, involving a matrix of components
that exceeds the size of the LED matrix.
The LT3965 I2C 8-switch matrix LED
dimmer makes it easy to control large
or small LED matrices (up to 512 LEDs).
Figure 1 shows the LT3965 in action on
Linear’s demonstration circuit DC2218.
Its highly integrated design (Figure 2)
minimizes component count. The individually addressable channels of the
LT3965 can be used to control LED
matrices in many ways, including:
•Each LT3965 can control eight dimming
channels—eight LEDs or eight clusters—within a string of LEDs.
•The eight channels can control the
individual red, green, blue and white
light on two RGBW LED modules for
adjustable brightness or changing
color of dashboard or trim lighting.
•Multiple LT3965s can be individually addressed on a single
communications bus to multiply
the strings in a large array.
4 | November 2015 : LT Journal of Analog Innovation
•The LT3965 can control multiple
LEDs per channel, or channels can
be combined to efficiently control
a single LED at higher current.
When combined with a suitable constantcurrent LED driver, the matrix dimmer
LED driver allows the individual LEDs
to be computer-controlled in headlights, daytime running lights, brake
and tail lights, side-bending lights,
dashboard display and other trim lighting. The LT3965’s built-in automatic
fault detection protects individual
LEDs in case of a failure and reports
failures to the microcontroller.
The 60V LT3965 includes eight integrated
330mΩ power switches, which can be
connected to one or more LEDs. The
power switches act as shunt devices
by turning off or PWM dimming the
LEDs on a particular channel. The
switches create eight individually
controlled brightness channels (up to
256:1 dimming ratio) and eight faultproof segments of an LED string.
The LT3965 can handle a string current
of 500m A when all eight power switches
are on at the same time (all LEDs off).
The switches can be connected in parallel and run at 1A through four channels
of LEDs as shown later in this article.
Regardless of the number of LEDs or
current, the LED string must be driven by
a properly designed converter that has the
bandwidth to handle the fast transients
of the matrix dimmer. Some reference
designs are included in this article.
LT3797 BOOST-THEN-DUAL-BUCK
MODE DRIVES TWO STRINGS, 16
LEDs AT 500mA WITH TWO LT3965s
The eight shunt power switches of the
LT3965 control the brightness of eight
channels of LEDs at 500m A. The string
voltage of the 8-LED matrix dimmer
system can be between 0V and 26V,
depending on how many LEDs are on or
off at a given time. The recommended
converter topology to drive these LEDs
is a 30V step-down converter with high
bandwidth and little or no output capacitor. This step-down topology requires
that 9V–16V automotive input is “preboosted” to a 30V rail from which the
step-down regulators can operate.
The triple output LT3797 LED controller
conveniently serves as a single-IC solution
for both the “pre-boost” and step‑down
functions—it can be configured as a
step-up voltage regulator on one channel,
followed by step-down LED drivers on the
other two channels. Each of two stepdown LED drivers can drive a string of
matrix-dimmed LEDs. This topology has
a number of advantages, most notably,
regardless whether the LED string voltages
are above or below the battery voltage, the
circuit continues to function optimally.
Figure 3 shows the schematic of the
demonstration board shown in Figure 1,
a boost-then-dual-buck mode LT3797 and
LT3965 matrix dimming headlight system
with 16 LEDs at 500m A. Each LED can
be individually controlled to be on, off
or PWM dimmed down to 1/256 brightness. The 350kHz switching frequency of
the LT3797 is outside the AM band (good
design features
Demonstration circuit DC2218 features a complete matrix LED dimmer system with
LT3797 boost-then-dual-buck mode LED drivers and two LT3965 matrix dimmers that
drive 16 LEDs at 500mA from a car battery. The board operates a matrix headlight
with an attached I2C microcontroller via DC2026, the Linduino One demo circuit.
12V 3A DC
POWER INPUT
BOARDS CAN BE EASILY CONNECTED IN SERIES,
ALL CONTROLLED BY A SINGLE µCONTROLLER
TWO LT3965 MATRIX
DIMMERS PER BOARD
TWO 26V, 500mA SUPPLIES
PRODUCED BY A SINGLE LT3797
BOOST-THEN-DUAL-BUCK MODE
CONVERTER
SYNC AT 350kHz
CONTROLLED BY LTC6900.
PWM AT 170Hz
POT ALLOWS
DIRECT PATTERN
INTERACTION
16 HIGH POWER LEDs
(EIGHT LEDs FOR EACH
LT3965) MOUNTED ON BOARD
WITH PROTECTIVE FILTER.
CONNECTIONS FOR MORE LEDs
PROVIDED.
PREPROGRAMMED
LINDUINO DC2026
INCLUDED.
USB FOR µCONTROLLER
REPROGRAMMING AND
CONTOL VIA GUI
SELECT FROM SEVEN
PRELOADED DIMMING
PATTERNS OR GUI
CONTROL
Photo:
Steven Tanghe
Figure 1. LT3965 LED matrix dimmer demonstration circuit DC2218 run as a Linduino™ shield (DC2026). This demonstration circuit runs headlight, turning light, tail light
and trim patterns and can be evaluated with Linear’s graphical user interface via a USB cable.
for EMI) and the resulting 170Hz PWM
dimming frequency of the LT3965, generated from the same 350kHz clock, is above
the visible range. With the system properly
synchronized, the LT3797 and LT3965
matrix headlight operates flicker-free.
The LT3797 buck mode converters are
optimized for extremely fast transients
with little or no output capacitor and
properly compensated control loops. These
>30kHz bandwidth converters tolerate fast
LED transients as the LEDs are turned on
and off and PWM dimmed at will. A filter
capacitor placed on the LED sense resistor replaces a pole in the control system
that is lost when the output capacitor is
reduced or removed for the fast transient
performance of the matrix dimmer.
November 2015 : LT Journal of Analog Innovation | 5
LED+
VIN
VIN
CH8
DRN8
LED FAULT
LED
SW FAULT
8
LED+
FAULT
DETECTOR
DRIVER
SRC8
EN/UVLO
+
–
IS1
2.7µA
CH7
1.24V
LED
SW FAULT
BANDGAP
REFERENCE
INTERNAL
BIAS
SRC7
CH6
LED
SW FAULT
DRN6
FAULT
DETECTOR
DRIVER
SDA
SDA
FAULT
DETECTOR
DRIVER
VDD
VDD
5V
DRN7
SRC6
CH5
DRN5
SCL
SCL
ADDR1
ADDR2
ADDR3
ADDR4
LED
SW FAULT
I2C
SERIAL
INTERFACE
ADDR1
ADDR2
FAULT
DETECTOR
DRIVER
ADDR3
REGISTERS
AND
CONTROL
LOGIC
ADDR4
SRC5
CH4
LED
SW FAULT
FAULT
DETECTOR
DRIVER
SRC4
TSD SET OVERHEAT FAULT
–170°C
ALERT
ALERT
CH3
LED
SW FAULT
SET LED FAULT
1.2V
0.6V
SRC3
CH2
+
–
LED
SW FAULT
0
PWMCLK
DRN3
FAULT
DETECTOR
DRIVER
+
–
DRN4
DRN2
FAULT
DETECTOR
DRIVER
SRC2
1
Figure 2. LT3965 60V 8-switch
LED matrix dimmer block
diagram reveals eight power
NMOS shunt switches
for brightness control, a
fault flag and I2C serial
communications interface.
CH1
INTERNAL
OSCILLATOR
0.9V
RTCLK
RT
A charge pump from the switch node
is used to power the LT3965 VIN pin
more than 7V above the LED+ voltage to
enable the top channel NMOS to be fully
enhanced when driven. The low RDS(ON)
NMOS switches in the LT3965 enable high
power operation without the IC getting
6 | November 2015 : LT Journal of Analog Innovation
+
–
LED
SW FAULT
Q1
FAULT
DETECTOR
DRIVER
GND
hot, even when all eight shunt switches are
on, turning the entire LED string off. In
this case, the LT3797 LED driver survives
the virtual output short created by all
eight shunt switches without any issues,
and is ready to quickly regulate 500m A
through the next LED that is turned on.
DRN1
SRC1
LEDREF
LED–
Demonstration circuit DC2218 (Figure 1)
features the system shown in Figure 3
and operates a matrix headlight with
an attached I2C microcontroller via
DC2026, the Linduino™ One demo circuit.
DC2218, operated as a large Linduino
shield, has up to 400kHz serial code that
design features
The LT3965 I2C 8-switch matrix dimmer, LED driver eases
the control of large or small LED matrices (up to 512 LEDs).
Its highly integrated design minimizes component count,
and built-in fault detection protects individual LEDs in case
of a failure and reports failures to the microcontroller.
Figure 3. LT3965 matrix LED dimmer system with LT3797 boost-then-dual-buck mode LED drivers and two LT3965 matrix dimmers that drive 16 LEDs at 500mA from a
car battery. I2C serial communications control the brightness of individual LEDs and check for LED and channel faults.
SKYHOOK
VOUT
ISP1
L1, L2: WURTH ELECTRONICS 7447789133
L3: WURTH ELECTRONICS 7443551151
L4: COOPER BUSSMANN SD25-470-R
D1, D2: DIODES DFLS260
D3: DIODES PDS340
D4, D5, D10, D11: NXP SEMI PMEG6010CEJ
D6, D7: NXP SEMI PMEG6010CEH
D8, D9: NXP SEMI PMEG4010CEH
M1, M2: VISHAY Si7308DN
M3: VISHAY Si7850DP
M4, M5: VISHAY Si7611DN
Q1, Q2: ZETEX FMMT593TA
5V
ALERT SDA
0.50Ω
0.50A
22µF
6.3V
ISP2
SCL
0.50Ω
0.50A
10k
ISN1
ISN2
10k
TG1
10k
D6
M4
1µF
50V
5V
TO ATTACHED
LINDUINO ONE
DC2026C
D10
D11
LED1+
1µF
D4
D5
0.50A
INTVCC
0.1µF
0.1µF
0.1µF
0.1µF
VLED
26V
SYNC
VIN
9V TO 36V
L3
15µH
+
33µF
50V
D3
10µF
50V
10µF
50V
×6
OPTIONAL
–
COUT LED1
0.1µF
VDD VIN
SCL
SCL
SDA
SDA
ALERT
49.9k
LED1+
LT3965
U1
49.9k
LED2+
1k
EN/UVLO
Q2
9.09k
ADDR1
M3
43.2k
D1
ADDR2
ADDR2
ADDR3
ADDR4
ADDR4
LEDREF
LEDREF
RTCLK
RTCLK
5V
10Ω
VIN
0.1µF
10V
57.6k
0.02Ω
1µF
50V
GATE3 SENSEP3 SENSEN3
VIN
1M
TG3 ISP3 ISN3
10Ω
FBH3
1M
10k
RT
SENSEN1 SENSEP1 GATE1
SS3
47.5k
D2
0.05Ω
SYNC GATE2 SENSEP2 SENSEN2
0.22µF
ISN1-2
FLT1-3 SW1 SW2
SS1-2
L4
47µH
1µF
ON/OFF
INTVCC
TG1-2
ISP1-2
LT3797
VREF CTRL3 CTRL1-2 PWM1-2
OPTIONAL
COUT
0.1µF
44.2k
FBH2
M2
DIV
0.05Ω
499k
EN/UVLO
PWM3
69.8k
LED2–
L2
33µH
OUT
SET
VLED
26V
D9
LTC6900
GND
0.50A
SRC1
GND
350kHz SYNC
L1
33µH
M1
ADDR1
ADDR3
D8
10Ω
EN/UVLO
9.09k
LED2+
DRN8
SRC8
DRN7
SRC7
DRN6
SRC6
DRN5
SRC5
DRN4
SRC4
DRN3
SRC3
DRN2
SRC2
DRN1
LT3965
U2
1k
Q1
GND
M5
1µF
50V
SKYHOOK
SRC1
44.2k
1M
FBH1
D7
1µF
ALERT
DRN8
SRC8
DRN7
SRC7
DRN6
SRC6
DRN5
SRC5
DRN4
SRC4
DRN3
SRC3
DRN2
SRC2
DRN1
TG2
5V
VIN VDD
LT3470
5V
REGULATOR
22µF
6.3V
BOOST
INTVCC
10µF
0.1µF
GND VC1-2
FBH1-2
VC3
5.7k
2.2nF
15k
1nF
22nF
November 2015 : LT Journal of Analog Innovation | 7
generates different headlight
patterns and interfaces with
Linear Technology’s graphical user interface (Figure 4).
Within the GUI shown in Figure 4,
LED brightness and fault protection functions can be examined with ALL CHANNEL
MODE and SINGLE CHANNEL MODE
commands, as well as FAULT CHECK read
and write commands to check for open
and short LEDs. Flicker-free operation,
fault protection and transient operation
can be examined with this demonstration
circuit system. DC2218 can be plugged
directly into a 12V DC source and it can
be controlled by a personal computer
running the GUI or reprogrammed
from a simple USB connection.
1A MATRIX LED DRIVER USING
PARALLEL CHANNELS
The LT3965 can be used to drive matrices
of 1A LED channels. It is easy to connect
the power switches of the LT3965 in
parallel so that two power switches split
1A of LED current and each LT3965
controls four 1A channels. One way to
use parallel power switches for higher
current is to run each of the anti-phase
parallel switches for only 50% of the
PWM period. By alternating and running
1A through a single NMOS power switch
for half the time, the effective heating is
about equal to running 500m A through
the same NMOS all of the time.
Figure 5 shows a 1A matrix headlight
system using eight LEDs driven by two
LT3965s and another boost-then-dual-buck
mode LT3797. When PWM dimming, the
LT3797 uses a unique 1/8-cycle phasing of
the eight switches, as shown in Figure 6.
In this 1A matrix system, LT3797 channels are combined in parallel pairs, so
Figure 4. The PC-based interface allows designers to access control and monitoring of the LEDs driven by the LT3965.
8 | November 2015 : LT Journal of Analog Innovation
design features
The LT3965 can be used to drive matrices of 1A LED
channels. It is easy to connect the power switches of the
LT3965 in parallel so that two power switches split 1A of
LED current and each LT3965 controls four 1A channels.
Figure 5. 1A matrix LED driver combines anti-phase
parallel channels for higher current applications in
high power LED headlight systems.
SKYHOOK
VOUT
ISP1
5V
D13
TG1
0.1µF
SYNC
VIN
9V TO 18V
+
33µF
50V
M4
D10
OPTIONAL
10k
1µF
50V
D4
1µF
DRN4
DRN8
SRC4
SRC8
DRN3
DRN7
SRC3
SRC7
DRN2
DRN6
SRC2
SRC6
DRN1
DRN5
SRC1
SRC5
LED1+
D7
1A
0.1µF
L3
10µH
VLED
15V
0.1µF
D3
4.7µF
25V
20V
OPTIONAL
0.1µF
LED1–
10µF
25V
×6
63.4k
1M
FBH1
10Ω
M3
66.5k
D1
SCL
SDA
SDA
LT3965
U1
0.015Ω
GATE3 SENSEP3 SENSEN3
VIN
100k
249k
17.8k
38.3k
TG3
1M
ISP3 ISN3
DRN4
DRN8
SRC4
SRC8
DRN3
DRN7
SRC3
SRC7
DRN2
DRN6
SRC2
SRC6
DRN1
DRN5
SRC1
SRC5
49.9k
LED2+
1k
LT3965
U2
1k
Q1
Q2
9.09k
EN/UVLO
9.09k
ADDR1
ADDR1
ADDR2
ADDR2
ADDR3
ADDR3
ADDR4
ADDR4
LEDREF
LEDREF
RTCLK
RTCLK
D8
GND
5V
VIN
0.1µF
10V
OUT
SET
1A
VLED
15V
LED2–
10Ω
M2
DIV
0.033Ω
FBH3
1M
L2
15µH
LTC6900
GND
LED2+
D9
350kHz SYNC
10Ω
D11
OPTIONAL
ALERT
SKYHOOK
EN/UVLO
L1
15µH
M1
D5
VDD VIN
SCL
57.6k
1µF
25V
1µF
50V
1µF
49.9k
LED1+
GND
TG2
M5
5V
ALERT
D6
22µF
6.3V
ISN2
VIN VDD
INTVCC
0.1µF
0.25Ω
1A
10k
5V
TO ATTACHED
LINDUINO ONE
DC2026C
D12
ISP2
SCL
10k
ISN1
L1, L2: WURTH ELECTRONICS 74437349150
L3: WURTH ELECTRONICS 74432510000
L4: COOPER BUSSMANN SD25-470-R
D1, D2: DIODES SBR2M30P1
D3: DIODES PDS1040
D4–D7, D10–D13: NXP SEMI PMEG4010CEJ
D8, D9: NXP SEMI PMEG4010CEH
M1, M2: VISHAY SiS402DN
M3: INFINEON BSC0901NS
M4, M5: VISHAY Si7611DN
Q1, Q2: ZETEX FMMT593TA
LT3470
5V
REGULATOR
ALERT SDA
0.25Ω
1A
22µF
6.3V
OPTIONAL
0.1µF
63.4k
FBH2
D2
0.033Ω
SENSEN1 SENSEP1 GATE1
SYNC GATE2 SENSEP2 SENSEN2
TG1-2
EN/UVLO
ISP1-2
LT3797
ISN1-2
FBH1-2
OVLO
PWM3
VREF CTRL3 CTRL1-2
100k
ANALOG DIM
PWM1-2 RT
SS3
52.3k
SS1-2
0.1µF
FLT1-3 SW1 SW2
L4
47µH
0.47µF
ON/OFF
INTVCC
BOOST
INTVCC
10µF
0.1µF
GND VC1-2
VC3
4.7k
2.2nF
8.2k
330pF
10nF
November 2015 : LT Journal of Analog Innovation | 9
Turning a high number of LEDs on or off presents a significant current
load step to the DC/DC converter. The converters presented here
handle these transients with grace, with a small output capacitor and
high bandwidth. An ACM write transitioning a high number of LEDs
produces no visible flicker or significant transient on the LED current.
Figure 6. 1/8 PWM flickerfree phasing of the eight
LT3965 power switches
limits transients during
PWM dimming brightness
control.
POR
1 DIMMING CYCLE = 2048 RTCLK CLOCK CYCLES
1/256 DIMMING (8 CLOCK CYCLES)
CH1
PHASE SHIFT OF 1/8 DIMMING CYCLE = 256 CLOCK CYCLES
CH2
CH3
CH4
CH5
CH6
CH7
CH8
that paired channels are anti-phase, 180°
from each other; specifically pairing
channels 8 and 4, 7 and 3, 6 and 2, and
5 and 1. Parallel channels alternate
shunting, effectively doubling the PWM
frequency, with the advantage of spreading out the shunted current and heat.
For this to work properly, the maximum
duty cycle for any single shunt power
switch is 50%, because two anti-phase
switches that are on 50% of the time
(each shunting an LED 50% of the time)
turns the LED off 100% of the time.
Each LT3965 controls the brightness of
four 1A LEDs that are driven by two 1A
buck mode LT3797 channels (from the
LT3797-boosted 20V channel). This high
power, robust system can be expanded
to power more LEDs with more LT3965s
or higher current LEDs with more channels in parallel. It is possible to drive two
LEDs per channel at 1A and drive up the
power of this flexible headlight system.
10 | November 2015 : LT Journal of Analog Innovation
MORE THAN ONE LED PER CHANNEL
The LT3965 can support one to four LEDs
per channel. Although it can be advantageous to individually control every single
LED for fault protection or high resolution
patterns, it is not always necessary. Using
more than one LED per channel reduces
the number of matrix dimmers in a system
and is enough to accomplish the patterns
or dimming required for some designs.
Segments of headlights, signal lights and
tail lights can have up to four LEDs with
the same brightness. Emergency LED lights
can have sets of three and four LEDs that
blink and wave with the same pattern.
The circuit in Figure 7 demonstrates a
two-LED-per-channel system—it has the
same number of LEDs as the circuit in
Figure 3, but uses only a single LT3965
matrix dimmer instead of two.
When an I2C command tells the LT3965 to
turn on, off, or dim a channel, it affects
the two LEDs that are controlled by that
channel’s shunt power switch. To stay
within the voltage limitations of the
LT3965, the 16 LEDs at 500m A still need
to be split into two series LED strings as
they are in Figure 2. The same LT3797
circuit in Figure 2 can be used, but only
a single LT3965 controls the brightness
of the two strings. This demonstrates
how each NMOS shunt power switch
inside the LT3965 can be configured
independently of the others, allowing
an endless variety of matrix designs.
ALL CHANNEL MODE AND SINGLE
CHANNEL MODE I 2C COMMANDS
WITH FLICKER-FREE PWM AND FADE
The I2C instruction set of the LT3965
includes 1-, 2- and 3-word commands.
These commands are sent over the serial
data line (SDA) alongside the mastergenerated clock line (SCL) at up to 400kHz
speed. The master microcontroller sends
all channel mode (ACM) or single channel
mode (SCM) write commands to control
the brightness, fade, open-circuit threshold and short-circuit threshold of the
LED channels and LT3965 addresses.
Broadcast mode (BCM), ACM and SCM
read commands request that the LT3965s
report the content of their registers,
design features
The LT3965 can support one to four LEDs per channel.
Segments of headlights, signal lights and tail lights
can have two to four LEDs with the same brightness.
Emergency LED lights can have sets of three and four
LEDs that blink and wave with the same pattern.
Figure 7. The flexible LT3965
can drive LED channels on
independent LED strings
and can drive between one
and four LEDs per channel.
(Complete driver circuit is
similar to Figure 3, but with
only one LT3956, as shown
here.)
LED1+
LED2+
SKYHOOK
5V
1µF
SKYHOOK
1µF
10k
VIN VDD
EN/UVLO
10k
10k
SCL
SCL
SDA
SDA
ALERT
DRN8
0.50A
VLED
26V
LT3965
LED1–
including open and short registers for
fault diagnostics. The LT3965 asserts an
ALERT flag when there is a new fault. The
micro can respond to the fault by determining which LT3965 reported the fault,
as well as the type and channel of fault.
In the case that multiple LT3965 ICs are
reporting faults, the LT3965s can sequence
fault reporting to the master to prevent
overlap errors. This makes the alert
response system reliable and conclusive. A
complete list of the registers and command
set is given in the LT3965 data sheet.
ACM write commands instantly turn
all of the eight channels of a single
LT3965 address on or off with just two
I2C words—the channels transition on
DRN4
SRC8
DRN7
SRC4
DRN3
SRC7
DRN6
SRC3
DRN2
SRC6
DRN5
SRC2
DRN1
SRC5
SRC1
0.50A
VLED
26V
SHORT AND OPEN LED FAULT
PROTECTION FOR EACH CHANNEL
ADDR1–4
RTCLK
350kHz
SYNC
ALERT
GND
LEDREF
LED2–
or off at the same time. Turning a high
number of LEDs on or off presents a
significant current voltage load step to
the DC/DC converter. The converters
presented here handle these transients
with grace, with little or no output
capacitor and high bandwidth.
As shown in Figure 8, an ACM write
transitioning a high number of LEDs
produces no visible flicker or significant transient on the LED current of
other channels. The high bandwidth
buck mode converter built around
the LT3797 is the reason for such a
small and controlled transient.
Single channel mode writes produce relatively small and fast single-LED transients.
SCM writes are used to set the brightness
of only one channel at a time to ON, OFF,
or PWM dimming with or without fade.
PWM dimming values between 1/256 and
255/256 are communicated in 3-word
writes while ON and OFF can be communicated in shorter, 2-word commands. A
fade bit on a single SCM write command
enables the LT3965 to move between two
PWM dimming levels with internally determined logarithmic fade and no additional
I2C traffic. The open and short thresholds
of each channel can be set between one
and four LEDs with SCM write commands.
Short- and open-circuit protection is an
inherent benefit of the matrix dimmer.
Each channel’s NMOS power switch
can shunt out between one and four
series LEDs. Traditional LED strings
have protection against the entire
string being open or shorted and only
some ICs have output diagnostic flags
to indicate these fault conditions. In
contrast, the LT3965 protects against,
and rides through, individual channel
shorts and opens, keeping operational
channels alive and running while recording and reporting the fault conditions.
When a fault occurs within a string,
the LT3965 detects the fault and asserts
its ALERT flag, indicating to the microcontroller that there is an issue to be
addressed. If the fault is an open-circuit,
the LT3965 automatically turns on its
November 2015 : LT Journal of Analog Innovation | 11
Short- and open-circuit protection is an inherent benefit of the matrix dimmer. Each
channel’s NMOS power switch can shunt out between one and four series LEDs.
Traditional LED strings have protection against the entire string being open or shorted
and only some ICs have output diagnostic flags to indicate these fault conditions.
In contrast, the LT3965 protects and rides through individual channel shorts, keeping
operational channels alive and running while recording and reporting the fault conditions.
corresponding NMOS power switch,
bypassing the faulty LED until a full diagnosis occurs or until the fault is removed.
The LT3965 maintains registers of open
and short faults for each channel and
returns the data to the microcontroller
during I2C fault read commands. The
command set includes reads that leave
the status register unchanged and those
that clear the fault registers, allowing
user-programmable fault diagnostics.
Registers can be read in the various modes
allowed for writes, SCM, ACM, BCM:
•Single channel mode (SCM) reads return
the open and short register bits for a
single channel. SCM reads also check
the open and short threshold register,
the mode control, and the 8-bit PWM
dimming value for that channel.
•All channel mode (ACM) reads return
the open and short register bits for all
channels of a given address without
clearing the bits, as well as the ACM ON
and OFF bits for all eight channels.
•The ACM and SCM reads can be
used to check and clear faults and to
read all of the registers for a robust
I2C communications system.
UP TO 16 ADDRESSABLE LT3965s ON
THE SAME BUS
Every LT3965 features four user-selectable
address bits, enabling 16 unique bus
addresses. Every ACM and SCM I2C
command is sent to the shared communications bus, but action is only taken by
the addressed LT3965. BCM commands
are followed by all ICs on the bus. The
4-bit address architecture allows a single
microcontroller and a single I2C 2-line
communications bus to support up to
8 × 16 = 128 individually controllable
channels. With the LT3965, for all but
the most ambitious lighting displays,
all individual LEDs in an automobile’s
Figure 8. The LED matrix driver designs shown in
this article feature minimal to no cross-channel
transient effects. For instance, transitioning half
the channels—here, simultaneously turning on two
and turning off two—has little to no transient effect
on the other four, untouched channels. The nontransitioned channels remain flicker free.
ILED1–4
500mA/DIV
(100% ON)
ILED5,6
500mA/DIV
(ON TO OFF)
ILED7,8
500mA/DIV
(OFF TO ON)
ALL CHANNEL MODE (ACM)
COMMAND
•In more complex systems with many
LT3965 matrix dimmers sharing the
same bus, a broadcast mode (BCM)
read first requests which, if any, LT3965
address has asserted the fault flag.
20µs/DIV
12 | November 2015 : LT Journal of Analog Innovation
headlight, tail light and trim lights can
be controlled by a single I2C communications bus and a single microcontroller.
Given that each channel can be connected
to up to four LEDs, one relatively
easy-to-implement system can support
matrix dimming for up to 512 LEDs.
CONCLUSION
The LT3965 matrix LED dimmer controls
eight LED-brightness channels on a single
LED string, giving lighting designers
unlimited access to sophisticated and
striking automotive lighting designs. The
I2C communications interface allows a
microprocessor to control the brightness
of individual LEDs in the string. Fault
protection in the I2C interface ensures LED
lighting system robustness. The channels
of the matrix dimmer are versatile: each
channel can control multiple LEDs; channels can be combined to support higher
current LEDs; or high LED-count systems
can be produced with up to 16 matrix
dimmer ICs on the same communications
bus. Take the next step in designing automotive headlights, tail lights, front, side,
dash and trim lights—the future is now. n
design features
CAN Bus Transceivers Operate from 3.3V or 5V and
Withstand ±60V Faults
Ciaran Brennan
The LTC2875 is a robust CAN bus transceiver that features ±60V overvoltage and ±25kV
ESD tolerance to reduce failures caused by electrical overstress. These transceivers
introduce several new capabilities for high voltage tolerant CAN bus transceivers: operation
from 3.3V or 5V supply voltages, up to 4Mbps data rate, ±36V common mode voltage
range, continuously variable slew rate and availability in 3mm × 3mm DFN packages.
The CAN bus forms the backbone of many
automotive, commercial and industrial
data communications systems. CAN bus
networks are used in a wide variety of
applications, including automotive and
transportation electronics, industrial
control systems, supervisory control and
data acquisition systems, building automation and security, HVAC control, and other
custom networked systems. Robustness
to electrical overstress is an important
attribute for CAN bus transceivers used in
these applications, which risk exposure
to wiring faults, ground voltage faults
and lightning induced surge voltages.
by contemporary network applications. The LTC2875 transceiver is Linear
Technology’s response to these requests.
However, few CAN transceivers capable of
operating from 3.3V supplies are available, and until now, none offer the high
voltage tolerance and wide common mode
operating range of the LTC2875. Many
customers have requested a robust CAN
bus transceiver with the performance
and the expanded capabilities demanded
3.3V OR 5V OPERATION
Figure 1. LTC2875 demo circuit with DFN and SO packages in foreground
Most high voltage tolerant CAN bus
transceivers can operate only from
a 5V supply, but 5V is rarely used by
most modern digital circuits. The
CAN bus transceiver may be the only
5V component in the system. A high
voltage tolerant CAN bus transceiver
that operates from a 3.3V supply reduces
design time and cost by eliminating
the need for a dedicated 5V supply.
The LTC2875 maintains compatibility with
the ISO 11898-2 CAN bus standards when
operating from a 3.3V supply, driving the
full specified differential bus voltage VOD
and maintaining the same receiver input
threshold voltages. The only difference
between 3.3V and 5V operation is that the
common mode bus voltage is reduced to
1.95V while operating at 3.3V, which falls
below the range of 2V to 3V specified by
ISO 11898-2. This minor shift in common
mode voltage falls within the minimum
common mode voltage range of −2V to
7V specified in the standard (and is truly
inconsequential to the ±25V common
mode voltage range of the LTC2875 when
operating at 3.3V), allowing the LTC2875
Ciaran Brennan
November 2015 : LT Journal of Analog Innovation | 13
The LTC2875 provides a continuously variable slew rate over an
approximate 20-to-1 range. The lowest slew rate is appropriate
for data rates of 200kbps or less. The slew rate is programmed
by a single resistor in series with the chip enable pin RS.
RSL
RS
CANH
LTC2875
TXD
VCC
RXD
15pF
47µF
TXD
RS
GND
CANH
VCC
CANL
RXD
SPLIT
RL/2
1%
VOD
CM
RL/2
1%
0.1µF
CANL
VOC
GND
RSL = 0Ω EXCEPT AS NOTED
(a)
Figure 3. Single resistor termination (a) and split termination (b)
RSL
RS
CANH
LTC2875
TXD
VCC
RXD
15pF
47µF
TXD
RS
GND
CANH
VCC
CANL
RXD
SPLIT
RL/2
1%
VOD
RL/2
1%
4.7nF
0.1µF
CANL
GND
RSL = 0Ω EXCEPT AS NOTED
(b)
to communicate seamlessly with any
other ISO11898-2 compliant transceivers. The LTC2875 is fully interoperable
with other transceivers on the same bus
that are powered by 5V when operating from either a 3.3V or 5V supply.
Not all systems require a high data rate.
In applications where lower data rates
suffice, the system designer may prefer a
CAN bus driver with low electromagnetic
60
4 Mbps DATA RATE WITH HIGH
SYMMETRY DRIVER AND
CONTINUOUS SLEW RATE CONTROL
14 | November 2015 : LT Journal of Analog Innovation
SLEW RATE (V/µs)
Modern CAN bus systems may operate
at data rates that exceed the capabilities of existing high voltage tolerant transceivers. For example, Linear
Technology’s LT1796 CAN transceiver
operates at a maximum of 125kbps.
The LTC2875 offers similar high voltage
tolerance to this predecessor, but can
communicate 32 times faster, up to 4Mbps.
50
VCC = 5V
40
VCC = 3.3V
30
20
10
0
1
10
RSL (kΩ)
100
Figure 2. Slew rate vs slew control resistor RSL
emissions (EME) slew controlled transitions. The LTC2875 provides a continuously
variable slew rate over an approximate
20-to-1 range. The lowest slew rate is
appropriate for data rates of 200kbps
or less. The slew rate is programmed by
a single resistor in series with the chip
enable pin RS, as plotted in Figure 2.
Considerable design effort was made to
keep the switching symmetry of the CAN
transmitter highly symmetrical (or more
accurately, anti-symmetrical) between
the CANH and CANL outputs, because
any asymmetry between the switching
waveforms of the two outputs produces
a change in the common mode voltage.
While the electromagnetic fields produced
by the differential voltage along a twisted
design features
The LTC2875 transceivers operating from a 3.3V supply can interoperate
with other CAN transceivers operating from a 5V supply on the same
bus. The only major difference between operating at 3.3V and 5V is that
the common mode voltages are ~1.95V and ~2.5V, respectively.
pair largely cancel and produce little EME,
the electromagnetic fields of the common
mode voltage on the pair add together
and may produce significant EME, particularly if the twisted pair is unshielded.
Therefore, good CAN transmitter switching symmetry results in lower EME.
The LTC2875 provides two features to
reduce the EME produced by fluctuations
of the common mode voltage during
switching: variable slew rate control and
split termination. The transmitter slew rate
can be programmed by a single resistor
in series with the enable pin RS. Reducing
the slew rate reduces the high frequency
content of the switching waveforms. Split
termination entails dividing the terminator
resistor at each end of the bus into two
equal, series resistors of half the termination resistance value, with the center point
of the resistors biased at the DC common
mode voltage supplied by the SPLIT pin
and a decoupling capacitor (Figure 3). Split
termination provides a low impedance
load for the common mode signal while
maintaining the proper termination for
the differential signals. The low impedance
common mode loading helps suppress
common mode voltage fluctuations.
The effectiveness of the split termination in reducing EME from the common
mode voltage fluctuations is illustrated
in Figure 4. In this figure, the voltages
at the CANH and CANL terminals and
the common mode voltage are recorded
for an LTC2875 transmitting at 1Mbps
over a 10-meter unshielded twisted pair,
with VCC = 3.3V, the slew rate set to
maximum, and 120Ω termination resistors placed on each end of the cable.
The FFT power spectra of the common
mode voltage waveforms are also
shown in Figure 4. The results using
split termination and those using single
resistor termination are both shown.
The waveforms with the single resistor
termination show a larger magnitude
of common mode transients during the
switching transition, as well as a damped
oscillation after the dominant to recessive
4V
transition. This damped oscillation is
the result of the inductance of the line
interacting with line and transceiver
capacitance after the transceiver switches
to its high impedance recessive state.
In this example, the common mode
voltage in the recessive state with the
single resistor termination is loaded only
by the four 40kΩ input resistors of two
LTC2875 devices, one on each end of the
cable, for a parallel resistance of 10k.
By contrast, the common mode voltage
4V
TRANSMITTER
SPLIT TERMINATION
MAXIMUM SLEW RATE
CANH
0.5V/DIV
CANH
0.5V/DIV
COMMON MODE
CANL
0V
0µs
0dB
COMMON MODE
CANL
200ns/DIV
2µs
FFT POWER SPECTRUM
SPLIT TERMINATION
MAXIMUM SLEW RATE
0V
0µs
0dB
200ns/DIV
2µs
FFT POWER SPECTRUM
SINGLE TERMINATION
MAXIMUM SLEW RATE
20dB/DIV
20dB/DIV
−160dB
0MHz
TRANSMITTER
SINGLE TERMINATION
MAXIMUM SLEW RATE
5MHz/DIV
50MHz
−160dB
0MHz
5MHz/DIV
50MHz
Figure 4. Transmitter waveforms and FFT power spectrum plots of the common mode voltage for split and
single bus terminations on 10m unshielded twisted pair cable; VCC = 3.3V, 1Mbps
November 2015 : LT Journal of Analog Innovation | 15
The use of the split termination results in a significant reduction in common
mode noise across the frequency spectrum when transmitting at 100kbps. At
these lower data rates, further reductions in the common mode noise spectrum
can be obtained by setting the LTC2875 to its minimum slew rate.
in the split termination case is also
loaded by the four 60Ω split termination resistors, for a parallel resistance
of 15Ω, in series with two parallel 4.7nF
capacitors. The common mode voltage
FFT power spectrum is lower in amplitude across a wide range of frequencies for the split termination compared
to the single resistor termination.
For transmitting at a lower data rate,
a slower slew rate may be used for
additional reduction in common mode
EME. Figure 5 illustrates four cases with
minimum and maximum slew rates,
combined with split or single resistor
termination. These measurements were
performed with the same test configuration as those shown in Figure 4, except
that the data rate was reduced to 100kbps
and waveforms at both the minimum and
maximum slew rates were recorded.
As in the 1Mbps waveforms of Figure 4,
the use of the split termination results in
a significant reduction in common mode
noise across the frequency spectrum
when transmitting at 100kbps. At these
lower data rates, further reductions in
the common mode noise spectrum can
be obtained by setting the LTC2875 to
its minimum slew rate. In this example,
combining both the split termination
and the minimum slew rate reduces the
common mode noise power by 20dB or
more over most of the recorded spectrum,
compared to the single resistor termination
combined with the maximum slew rate.
16 | November 2015 : LT Journal of Analog Innovation
Figure 5. Transmitter waveforms and FFT power spectrum plots of the common mode voltage for four
combinations of termination and slew rate; VCC = 3.3V, 100kbps, 10m unshielded twisted pair cable
4V
4V
TRANSMITTER
SPLIT TERMINATION
MAXIMUM SLEW RATE
TRANSMITTER
SPLIT TERMINATION
MINIMUM SLEW RATE
CANH
0.5V/DIV
CANH
0.5V/DIV
COMMON MODE
COMMON MODE
CANL
0V
0µs
CANL
2µs/DIV
4V
20µs
0V
0µs
4V
TRANSMITTER
SINGLE TERMINATION
MAXIMUM SLEW RATE
CANH
0.5V/DIV
COMMON MODE
COMMON MODE
CANL
0V
0µs
0dB
CANL
2µs/DIV
20µs
0V
0µs
0dB
FFT POWER SPECTRUM
MAXIMUM SLEW RATE
2µs/DIV
20µs
FFT POWER SPECTRUM
MINIMUM SLEW RATE
SINGLE TERMINATION
SINGLE TERMINATION
20dB/DIV
20dB/DIV
SPLIT TERMINATION
−160dB
0MHz
20µs
TRANSMITTER
SINGLE TERMINATION
MINIMUM SLEW RATE
CANH
0.5V/DIV
2µs/DIV
5MHz/DIV
SPLIT TERMINATION
50MHz
−160dB
0MHz
5MHz/DIV
50MHz
design features
Another technique to reduce EME from common mode voltage fluctuations is
using a common mode choke. The choke increases the source impedance of the
common mode signal, and in conjunction with capacitors added between CANH
and CANL and GND, forms a lowpass filter that attenuates high frequency noise.
REDUCING EME WITH A COMMON
MODE CHOKE
Another technique to reduce EME from
common mode voltage fluctuations is
using a common mode choke. The choke
increases the source impedance of the
common mode signal, and in conjunction
with capacitors added between CANH
and CANL and GND, forms a lowpass
filter that attenuates high frequency
noise. The effectiveness of a 100µ H
common mode choke in conjunction
with two 33pF capacitors in reducing
the common mode noise is shown in
Figure 6. In this example, VCC = 3.3V,
split termination was employed, the
twisted pair cable was 10 meters long
and the data rate was 100kbps.
Figure 6. Transmitter waveforms and FFT power spectrum plots of the common mode voltage, with or without
common mode choke
4V
4V
SPLIT TERMINATION
MAXIMUM SLEW RATE
NO COMMON MODE CHOKE
SPLIT TERMINATION
MINIMUM SLEW RATE
NO COMMON MODE CHOKE
CANH
0.5V/DIV
CANH
0.5V/DIV
COMMON MODE
COMMON MODE
CANL
0V
0µs
4V
CANL
2µs/DIV
20µs
SPLIT TERMINATION
MAXIMUM SLEW RATE
100µH COMMON MODE CHOKE
0V
0µs
4V
2µs/DIV
20µs
SPLIT TERMINATION
MINIMUM SLEW RATE
100µH COMMON MODE CHOKE
CANH
CANH
MIXED 3.3V AND 5V OPERATION
The LTC2875 transceivers operating
from a 3.3V supply can interoperate
with other CAN transceivers operating
from a 5V supply on the same bus. The
only major difference between operating at 3.3V and 5V is that the common
mode voltages are ~1.95V and ~2.5V,
respectively. The common mode of
the bus therefore fluctuates depending on the logical state of the bus.
When all transmitters are in the recessive
state, the common mode voltage settles to
some intermediate voltage depending on
all the resistive loads placed on the bus,
including receiver input resistors, and split
termination resistors (if present). When a
transmitter powered by 5V is dominant, it
pulls the common mode voltage toward
2.5V. When a transmitter is powered by
0.5V/DIV
0.5V/DIV
COMMON MODE
COMMON MODE
CANL
0V
0µs
0dB
CANL
2µs/DIV
20µs
0V
0µs
0dB
FFT POWER SPECTRUM
MAXIMUM SLEW RATE
20µs
FFT POWER SPECTRUM
MINIMUM SLEW RATE
NO COMMON MODE CHOKE
NO COMMON MODE CHOKE
20dB/DIV
20dB/DIV
100µH COMMON MODE CHOKE
−160dB
0MHz
2µs/DIV
5MHz/DIV
50MHz
100µH COMMON MODE CHOKE
−160dB
0MHz
5MHz/DIV
50MHz
November 2015 : LT Journal of Analog Innovation | 17
An ideal CAN bus transceiver would survive large common mode voltages and
continue to send and receive data without disruption. Accordingly, the receiver in the
LTC2875 is designed to operate over an expanded ±36V common mode voltage range
when operating from a 5V supply, and ±25V when operating from a 3.3V supply.
3.3V, it pulls the common mode voltage
toward 1.95V. The common mode voltage
fluctuates between 2.5V and 1.95V,
resulting in a modest increase in EME.
An example of mixed voltage operation
is shown in Figure 7. The experimental
setup consists of two LTC2875 transceivers,
each connected to the end of a 10-meter
twisted pair, with split termination
employed. Each transceiver alternates in
driving a dominant state on the bus. The
waveforms are recorded on the CANH and
CANL pins of the near side transceiver.
In the plots shown on the left, both the
near and far transceivers are powered by
a 3.3V supply. The common mode voltage
remains near 1.95V with only minor
perturbations. In the plots on the right, by
comparison, the near transceiver remains
powered by 3.3V, while the far transceiver
is powered by 5V. The recessive common
mode voltage settles to about 2.23V, the
average of 2.5V and 1.95V. When the
near side transceiver is dominant, the
common mode voltage is pulled down
close to 1.95V, whereas when the far side
transceiver is dominant, the common
mode voltage is pulled close to 2.5V.
The difference in EME resulting from
common mode voltage fluctuations can be
seen by comparing the FFT power spectra
of the common mode voltage recorded at
the terminals of the near transceiver. An
increase in power of approximately 8dB
is observed from 0MHz to 25MHz for the
mixed power supply voltage case, with the
difference rolling off above that frequency.
18 | November 2015 : LT Journal of Analog Innovation
4V
4V
NEAR SIDE
DOMINANT
3.3V
FAR SIDE
DOMINANT
3.3V
NEAR SIDE
DOMINANT
3.3V
0.5V/DIV
0.5V/DIV
SPLIT TERMINATION
MAXIMUM SLEW RATE
NEAR & FAR SIDES POWERED BY 3.3V
0V
0µs
0dB
FAR SIDE
DOMINANT
5V
SPLIT TERMINATION
MAXIMUM SLEW RATE
NEAR SIDE POWERED BY 3.3V; FAR SIDE 5V
5µs/DIV
50µs
0V
0µs
0dB
FFT POWER SPECTRUM
MAXIMUM SLEW RATE
5µs/DIV
50µs
FFT POWER SPECTRUM
MINIMUM SLEW RATE
NEAR: 3.3V; FAR: 5V
20dB/DIV
20dB/DIV
NEAR & FAR: 3.3V
−160dB
0MHz
5MHz/DIV
50MHz
NEAR & FAR: 3.3V
−160dB
0MHz
5MHz/DIV
50MHz
Figure 7. Transmitter waveforms and FFT power spectrum plots of the common mode voltage for two
transmitters powered by identical and mixed power supply voltages; 10m unshielded twisted pair cable,
100kbps, maximum slew rate
±36V COMMON MODE VOLTAGE
RANGE
Standard CAN bus transceivers operate
over a limited common mode voltage
range that extends from −2V to +7V. In
commercial or industrial environments,
ground faults, noise, and other electrical interference can induce common
mode voltages that greatly exceed these
limits. An ideal CAN bus transceiver
would survive large common mode voltages and continue to send and receive
data without disruption. Accordingly,
the receiver in the LTC2875 is designed
to operate over an expanded ±36V
common mode voltage range when
operating from a 5V supply, and ±25V
when operating from a 3.3V supply.
The receiver uses low offset bipolar
differential inputs, combined with high
precision resistor dividers, to maintain
precise receiver thresholds over the
wide common mode voltage range. The
design features
The LTC2875 is a groundbreaking ±60V overvoltage tolerant CAN bus transceiver
that operates from either 3.3V or 5V supplies. Its industrial robustness is matched
by superior performance, application flexibility and excellent EME characteristics.
transmitters operate up to the absolute
maximum voltages of ±60V, and will
sink or source current up to the limits
imposed by their current limit circuitry.
HOT PLUGGING, HOT SWAPPING,
AND GLITCH-FREE POWER-UP AND
POWER-DOWN
The LTC2875 features glitch-free powerup and power-down protection to meet
hot plugging (or hot swap) requirements. These transceivers do not produce
a differential disturbance on the bus
when they are connected to the bus
while unpowered, or while powered but
disabled. Similarly, these transceivers do
not produce a differential disturbance on
the bus when they are powered up in the
disabled state while already connected to
the bus. In all of these cases the receiver
output RXD remains high impedance
(with an internal 500k pull-up resistor),
while the CANH and CANL outputs remain
in the high impedance recessive state.
If the transceiver is powered up in the
enabled state, the chip goes active shortly
after the supply voltage passes through
the transceiver’s internal power good
detector threshold. The RXD output
reflects the state of the bus data when
the chip goes active, while the transmitter remains in the recessive state with
its outputs high impedance until the
first recessive-to-dominant transition
of TXD after the chip goes active.
If the transmitter is in the enabled state
when the chip is powered down, the chip
goes inactive shortly after the supply
voltage passes through the transceiver’s
internal supply undervoltage detector
threshold. If the transmitter is in the
dominant state at this time, the outputs
smoothly switch to the recessive state.
Regardless of whether it is outputting a
dominant or recessive state, the receiver
output RXD smoothly switches to a
high impedance state (weakly pulled up
through an internal 500k pull-up resistor).
±60V FAULT AND ±25KV ESD
TOLERANCE
CAN bus wiring connections in industrial
installations are sometimes made by
connecting the bare twisted wire to screw
terminal blocks. The apparatus containing the CAN bus interface may house
circuits powered by 24V AC/DC or other
voltages that are also connected with
screw terminals. The handling of exposed
wires and screw terminals by service
personnel introduces the risk of ESD
damage, while the possibility of wiring
the cables to the wrong screw terminals introduces the risk of overvoltage
damage. The high fault voltage and ESD
tolerance make the LTC2875 exceptionally
resistant to damage from these hazards.
LTC2875 is protected from ±60V faults even
with GND open, or VCC open or grounded.
The LTC2875 is protected from electrostatic
discharge from personnel or equipment
up to ±25kV (HBM) to the A, B, Y and
Z pins with respect to GND. On-chip
protection devices begin to conduct at
voltages greater than approximately
±78V and conduct the discharge current
safely to the GND pin. Furthermore, these
devices withstand up to ±25kV discharges
even when the part is powered up and
operating without latching up. All the
other pins are protected to ±8kV (HBM).
CONCLUSION
The LTC2875 is a groundbreaking
±60V overvoltage tolerant CAN bus
transceiver that operates from either
3.3V or 5V supplies. Its industrial
robustness is matched by superior
performance, application flexibility
and excellent EME characteristics. n
The ±60V fault protection of the LTC2875
is achieved by using high voltage
BiCMOS integrated circuit technology.
The naturally high breakdown voltage
of this technology provides protection in
powered-off and high impedance conditions. The driver outputs use a progressive
foldback current limit design to protect
against overvoltage faults while still
allowing high current output drive. The
November 2015 : LT Journal of Analog Innovation | 19
Low Power IQ Modulator for Digital Communications
Bruce Hemp and Sunny Hsiao
IQ modulators are versatile building blocks for RF systems. The most common
application is generating RF signals for digital communication systems. This article
illustrates the modulation accuracy of the LTC5599 low power IQ modulator, and shows
by simple example how to integrate the device into a digital communication system.
MODULATOR APPLICATIONS
Figure 1. Test setup to measure basic modulation accuracy
LO
ROHDE & SCHWARZ SMJ 100A
450MHz, 0 dBm
BASEBAND IQ SIGNAL GENERATOR
ROHDE & SCHWARZ SMJ 100A
OR EQUIVALENT
DAC (14-BIT) + LPF
DAC (14-BIT) + LPF
I+
I−
Q+
Q−
LTC5599
RF OUT
−4 dBm RMS
LOW POWER
IQ MODULATOR
Virtually any type of RF modulation
can be generated with IQ modulation,
within the center frequency, bandwidth and accuracy capabilities of the
modulator device. Table 1 shows some
of the applications of the LTC5599.
MODULATION: 16-QAM, PRBS 9,
ROOT COSINE ALPHA 0.35, 30k SYM./SEC
VBIAS = 1.4V, IQ DRIVE = 1.1VP–P(DIFF),
CREST FACTOR 5.4dB
Table 1. Some possible applications for the LTC5599 low power IQ modulator.
APPLICATION
MOD STD
MODULATION TYPE (REFERENCE 1)
MAX RF BW
Digital wireless microphones
Proprietary
QPSK, 16/32/64-DAPSK, Star-QAM
200kHz
802.11af
OFDM: BPSK, QPSK, 16/64/256-QAM
Up to 4× 6MHz channels
DOCSIS
16-QAM
6MHz
Custom
Wide programmability range
—
—
AM, FM/PM, SSB, DSB-SC
—
TETRA
π/4-DQPSK, π/8-D8PSK, 4/16/64-QAM
25kHz to 150kHz
• Commercial
TETRAPOL
GMSK
10kHz, 12.5kHz
• Industrial
P-25
C4FM, CQPSK
6.25kHz to 12.5kHz
DMR
4FSK
6.25kHz, 12.5kHz
Wireless networking
• White-space radios
• Cognitive radio
CATV upstream
Military radios (portable, manpack)
Software defined radios (SDR)
Portable test equipment
Analog modulation
2-way radios
• Public safety
20 | November 2015 : LT Journal of Analog Innovation
design features
Figure 2. LTC5599 EVM measured using
lab-grade baseband and LO signal
generators. Note that the MER measures
over 49dB, basically “Broadcast Quality.”
MEASURED IQ VECTOR
MEASURED EYE DIAGRAM
MODULATION ACCURACY AND EVM
The error vector magnitude or EVM is
a measure of modulation accuracy in
digital radio communication systems.
Modulation accuracy is important because
any error on the modulated signal can
cause reception difficulty or excessive
occupied bandwidth. If left unchecked,
the receiver could exhibit excessive bit
errors, the effective receiver sensitivity
could be degraded or the transmit adjacent
channel power (ACP) can become elevated.
EVM VS TIME
An error vector is a vector in the I-Q plane
between the actual received or transmitted
symbol and the ideal reference symbol.
EVM is the ratio of the average of the error
vector power over the average ideal reference symbol vector power. It is frequently
expressed in either dB or percentage.
Figure 1 is a test setup example showing
the modulation accuracy attainable with
the LTC5599 low power direct quadrature
modulator. Figure 2 shows the results. In
this test, precision lab equipment generates
a 30k symbol/second 16-QAM baseband
(120kbps), and 450MHz LO input signal
to the modulator. A vector signal analyzer
(VSA) examines the modulator output.
In Figure 2, the EVM vs time results
show EVM uniformly low across all
symbols, while the error summary
shows EVM approximately 0.24% RMS,
and 0.6% peak. This is indeed excellent performance, shown by a modulation error ratio (MER) of 49.6dB.
The LTC5599 has internal trim registers
that facilitate fine adjustments of I and
ERROR SUMMARY
TEST CIRCUIT SHOWN IN FIGURE 1
DUT: LTC5599
REGISTERS[0…8] = [0X31, 84, 80, 80, 80, 10, 50, 06, 00]
FREQUENCY SET TO 450MHz; OTHERWISE ALL DEFAULTS
VSA: AGILENT 89441A, ANALYSIS SPAN = 100kHz
MEASURING FILTER: ROOT COSINE ALPHA 0.35
REFERENCE FILTER: RAISED COSINE
What’s an IQ Modulator?
An IQ modulator is a device that converts baseband
information into RF signals. Internally, two doublebalanced mixers are combined as shown below. By
modulating with both in-phase (I) and quadrature (Q)
inputs, any arbitrary output amplitude and phase can
be selected.
By targeting specific points in amplitude and phase,
high order modulation is created. Shown below is 16QAM. There are four possible I values, which decodes
into two bits. Likewise for the Q axis. So each symbol
can convey four bits of information.
I-CHANNEL
MODULATOR
INPUT
LO
INPUT
Q+
90°
COMBINER
Q-CHANNEL
MODULATOR
INPUT
I−
I+
Q−
Fundamental architecture of an IQ modulator
November 2015 : LT Journal of Analog Innovation | 21
Q DC offset, amplitude imbalance, and
quadrature phase imbalance to further
optimize modulation accuracy—results are
even better if trim registers are adjusted.
In many ways, this test demonstrates the
best-case capabilities of the modulator
without optimization: baseband bandwidth is large, DAC accuracy and resolution are superb and digital filtering is
nearly ideal.1 While these test results are
useful for measuring the true performance
of the modulator, practical low power
wireless implementations necessitate
some compromises, as discussed below.
DRIVING FROM PROGRAMMABLE
LOGIC OR AN FPGA
Figure 3. Transmit exciter block diagram. (Full schematic is in Figure 4.)
LO
0 dBm
Many FPGAs and programmable devices
support digital filter block (DFB) functionality, an essential building block for
digital communications. Raw transmit
data is readily IQ mapped and digitally
filtered. Figure 3 shows an example
of how a device such as the Cypress
PSoC 5LP can be utilized to drive IQ
modulators such as the LTC5599.
MCU WITH
PROGRAMMABLE LOGIC
DFB
DFB
DAC
8-BIT
DAC
8-BIT
I
I
LT6238
I
SINGLE ENDED
TO DIFF
Q
Q
VOCM = 1.4V
MODULATION:
16-QAM, PRBS 9,
ROOT COSINE ALPHA 0.35,
30k SYM./SEC
LTC5599
LC
RECONSTRUCT
FILTERS
LOW POWER
IQ MODULATOR
RF OUT
−4 dBm RMS
Q
5th ORDER BESSEL
FULL SCHEMATIC IN FIGURE 4
OCCUPIED BW = 40.5kHz
Figure 4. Driving an IQ modulator with programmable logic and DACs. The passive Bessel filter attenuates DAC images and provides lowest RF output noise floor,
while imposing negligible symbol error vector.
VCC
MCU WITH
PROGRAMMABLE
LOGIC
UNMODULATED LO INPUT
0 dBm
VCC
VREF
1.74k
0.10µF
2.49k
VCC
133Ω
470µH
Vcc
I CH. DAC
0~1.024V
FULL-SCALE
1.47k
+1.4V
U2
LT6238
27nF
2.49k
2.49k
13nF
10nF
2.2nF
220µH
4.7nF
1nF
1nF
RL(I)
267Ω
VCC VCTRL EN
+1.4V
U2
133Ω
470µH
27nF
LT6238
GAIN SCALING
AND DC SHIFT
BBMI
SINGLE-ENDED
TO DIFFERENTIAL
DRIVER
VREF
1.74k
2.49k
133Ω
DAC LC RECONSTRUCTION FILTER
1.47k
+1.4V
U2
2.49k
LT6238
+1.4V
VOLTAGE
OUTPUT
DAC
VREF
2.49k
U2
LT6238
SPI
MASTER
2.49k
0.10µF
27nF
10nF
RF
2.2nF
SDI
SCLK
CSB
GNDRF
1.5nF
220µH
13nF
4.7nF
1nF
470µH
27nF
220µH
10nF
2.2nF
MODULATED
RF OUTPUT
RL(Q)
267Ω
133Ω
BESSEL RESPONSE, −3dB @ 50kHz
−0.7dB @ 20kHz
−44dB @ 220kHz
INDUCTORS: COILCRAFT DS1608C SERIES, OR EQUIVALENT
SPI BUS SETS MODULATOR CENTER FREQUENCY
22 | November 2015 : LT Journal of Analog Innovation
U3
LTC5599
BBPQ
1.82k
+1.4V
SDO
BBMQ
470µH
Q CH. DAC
0~1.024V
FULL-SCALE
LOL
TEMP
2.2nF
BBPI
U1
CY8C58LP
LOC
TTCK
220µH
10nF
39nH
15pF
VCC = 3.3V
10µF
GND
SPI BUS
54mA
design features
Digital interpolation is used to increase
the DAC clock frequency, and hence the
DAC image frequencies. This lowers the
filter order requirement of the LC reconstruct filter, which serves to attenuate DAC
images to acceptable levels, while minimizing phase error and wideband noise.
Figure 4 shows the complete circuit. The
differential baseband drive to the modulator, as opposed to single-ended baseband
drive, offers the highest RF output power
and lowest EVM. The LTC6238 low noise
amplifier, U2, converts the DAC singleended I and Q outputs to differential.2
Input amplifier U2 gain is designed to scale
the DAC out voltage range to the modulator input voltage range, after the 2:1 attenuation effect of filter terminating resistors
RL(I) and RL(Q) is taken into account. The
input amplifier U2 is also designed to
supply the required input common mode
voltage for the IQ modulator—important
for maintaining proper modulator
DC operating point and linearity.
Classical LC filter synthesis methods
are used for the DAC reconstruction
lowpass filter (LPF) design. Some of the
filter shunt capacitance is implemented
as common mode capacitors to ground.
This also reduces common mode noise,
which can find its way to the modulator output. If active filters are used here,
the final filter stage before the modulator should be a passive LC roofing filter
for lowest broadband RF noise floor.
Table 2, Figure 5 and Figure 6 show the
performance results. In this case, EVM
is limited by the digital accuracy of the
baseband waveforms, here determined by
MEASURED IQ VECTOR
EVM VS TIME
MEASURED EYE DIAGRAM
ERROR SUMMARY
Figure 5. EVM measurement detail. Two IC devices replace the lab signal generator. It’s not perfect, but is
usually ‘good enough.’
the number of U1 FIR filter taps (63), and
by the DAC resolution (eight bits). For
this reason, EVM does not substantially
improve when IQ modulator impairments are adjusted out, as shown in
Table 2. For lower EVM, use more FIR
filter taps and higher resolution DACs.
When comparing the results shown in
Figures 2 and 5, we see the price paid
for replacing a high grade lab signal
generator with a circuit composed of
programmable logic and op amp filters.
EVM increased from 0.24% RMS to 0.8%
RMS. The increased EVM is primarily due
to the fact that the waveforms generated
by the programmable logic IC are not as
accurate as the lab instrument. Such is
the case in a real world implementation,
but Figure 5 shows a fairly decent eye
diagram, and a summary measurement
that shows the modulation accuracy
is sufficient for most applications.
In Figure 6 we see the output spectrum is quite clean. The amplitude of
the DAC image spurs, relative to the
desired signal, is estimated by sin(x)/x,
where x = πf/fCLK , plus attenuation
afforded by the DAC LC reconstruct
filter. For lowest adjacent channel
power, a long FIR filter (many taps) is
essential, as is a low phase noise LO.
Table 2. EVM performance. Even with a 63-tap FIR filter design and 8-bit dual-DACs, the 0.8% RMS EVM achievement is entirely adequate for most applications.
TX FIR FILTER
DESIGN
INTERPOLATION
FACTOR
SYMBOL RATE
(ksps)
DATA RATE
(kbps)
63-tap RRC,
α = 0.35
8
30
120
EVM
(% RMS)
EVM
(% PEAK)
NOTES
0.8
2.0
LTC5599 Unadjusted (MER = 39.1dB)
0.8
1.8
LTC5599 Adjusted (MER = 39.8dB)
November 2015 : LT Journal of Analog Innovation | 23
Linear Technology’s LTC5599 IQ modulator is a versatile
RF building block, offering low power consumption,
high performance, wide frequency range and unique
optimization capabilities. It simplifies radio transmitter
design without sacrificing performance or efficiency.
Table 3. Output noise density levels off at
approximately 17dB over kTB.
0
−10
−20
AMPLITUDE (dBm)
Figure 6. Output spectrum.
In this design, the closest
image spurs are about 70dB
down, reasonably good for
most systems. Modulator
RMS output power measures
−4dBm. Harmonic filtering is
still required.
−30
RBW = 5kHz
DAC IMAGE SPURS, −70dBc:
−25dB FROM (sin x)/x
+ −45dB FROM LPF
−70dB TOTAL
−40
−50
−60
−70
FREQUENCY OFFSET
(MHz)
RF OUTPUT NOISE
DENSITY (dBM/Hz)
+6
−156.7
+10
−156.8
+20
−156.8
−80
−90
−100
449.5
450
FREQUENCY (MHz)
Higher frequency span sweeps show
no visible spurious products except
for the harmonics of the carrier,
which must be filtered as usual.
Low output noise floor is also important
in many cases, such as when a transmitter
and receiver are duplexed or co-located,
when high PA gain is used, or when
multiple transmitters run simultaneously.
Table 3 shows the measured output
noise density for the system of Figure 3,
while transmitting at a modulated carrier
frequency of 460MHz. The low U2 op amp
noise, combined with the 5th order roll-off
of the LC reconstruct filter, keeps the baseband noise contribution as low as possible.
450.5
Total current consumption at 3.3V
measures 96m A, as summarized in Table 4.
The majority of the DC power is consumed
by U1, the programmable logic device,
for which each DFB is specified to typically consume 21.8m A at the 67MHz
clock frequency of this application.3 In
summary, the DFBs account for 81% of
the digital power consumption. Clearly
the key to reduced current consumption for the digital section is optimization of the DFB architecture, which is
beyond the scope of this article.4
Linear Technology’s LTC5599 IQ modulator is a versatile RF building block,
offering low power consumption, high
performance, wide frequency range
and unique optimization capabilities. It
simplifies radio transmitter design without
sacrificing performance or efficiency. n
Notes
1Test
equipment FIR filters are synthesized in software,
so hundreds or thousands of filter taps are feasible and
preferred, since signal quality is most important, and
delay is inconsequential. In contrast, a real-time wireless
application typically requires trade-offs between filter
delay and EVM/ACP.
2For
lower symbol rate applications, the LTC1992 low
power fully differential input/output amplifier/driver could
also be used for this purpose, offering improved DC
accuracy and lower DC power consumption in exchange
for a higher transmit noise floor within the channel passband.
3In
Table 4. Total power consumption
STAGE
DESCRIPTION
ICC (mA)
POWER (mW)
U1
CY8C58LP Programmable System on Chip
54
178
U2
LT6238 Quad Op Amp
13
43
U3
LTC5599 Low Power IQ Modulator
29
96
Total:
96
317
24 | November 2015 : LT Journal of Analog Innovation
CONCLUSION
this example, the minimum DFB clock frequency =
30kHz symbol rate • 8x interpolation • 63 FIR filter taps
• 2 cycles for multiply and accumulate (MAC) • 2 cycles
for arithmetic logic (ALU) = 60.5MHz.
4DFBs
that are faster and more highly optimized are available from Altera and Xilinx.
References
1“Digital
Modulation in Communications Systems –
An Introduction,” Application Note 1298, Keysight
Technologies
design features
Low Profile Supercapacitor Power Backup with Input Current
Limiting
David Salerno
Supercapacitors are increasingly used as backup power
sources, due in large part to their continually improving
volumetric energy capacity and robust nature. Large output
capacitors can strain the load capabilities of an input
source, especially when that source is limited by protocol
(USB or PCMCIA) or a high source resistance. Input source
limitations can complicate designs. The LTC3128 simplifies
power backup by adding a programmable accurate input
current limit to a complete supercapacitor charger. Figure 1
shows that only a few components are needed to produce
a supercapacitor charger with a 3.0A input current limit.
The LTC3128 is a buck-boost DC/DC
supercapacitor charger with programmable accurate input current limit (up
to 3A) and active balancing, offered in
4mm × 5mm × 0.75mm QFN or 24-lead
TSSOP packages. The 1.2MHz switching
frequency, along with low resistance, low
gate charge integrated switches provide an
efficient, compact and low profile solution
for charging large output capacitors. The
high accuracy (±2%) of the programmable
input current limit allows designers to
limit the maximum current draw to just
below the capability of the input source.
Capacitor voltage monitoring and protection, combined with the integrated active
charge balancer, prevents mismatched
capacitors from being overvoltaged
and keeps capacitors with mismatched
leakages in balance. This makes the
LTC3128 ideal for backup or pulsed load
applications. Supercapacitors, because
of their long lifetime, large cycle capability (up to 10 years and 500,000 cycles)
and relatively straightforward charging
profiles, are ideal for backup solutions.
SUPERCAPACITOR CHARGE TIME
AND HOLDUP TIME
When designing a backup system,
two of the most important criteria
are charge time and holdup time. The
charge time determines the minimum
amount of time the system needs to be
in operation before it can withstand a
power failure, and holdup time determines how long a system can maintain
operation from its backup source.
Charge time is determined by a combination of programmed input current limit,
programmed output voltage, converter
efficiency and output capacitance. Figure 2
shows the charge time for a 1F output
capacitance at a programmed input
current of 3.0A. This curve takes into
account VIN , VOUT and the converter efficiency. If the output capacitance is larger
or smaller than 1F, the charge time scales
proportionally to the output capacitance.
At the end of charging, the LTC3128 dials
back the input current to top off the
Figure 2. LTC3128 charge time
5.5
5.0
3.3µH
SW1
VIN
2.4V TO 5.5V
3.0A
10µF
10µF
4.5
SW2
VOUT = 4.2V
RSENP VOUTP
RSENS VOUTS
LTC3128
VIN
RUN
MID
PFI
FB
PFO
PGOOD
MAXV
PROG
GND
10µF
4.0
TO LOAD
1.87M
100F
VOUT (V)
Figure 1. Complete
supercapacitor charging
circuit with input current limit
3.5
3.0
2.5
2.0
1.5
1.0
3.57k
470pF
301k
COUT = 1F
IIN = 3.0A
0.5
0
VIN = 2.4V
VIN = 3.3V
VIN = 4.2V
VIN = 5.0V
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
TIME (s)
November 2015 : LT Journal of Analog Innovation | 25
The active charge balancer uses the inductor of the LTC3128 to efficiently move
charge from one capacitor to another to balance them, maintaining the same
voltage across each capacitor. Active charge balancing eliminates the high
quiescent current and continuous power dissipation of passive balancing.
charge on the output capacitor stack. This
is done to prevent chattering in and out
of regulation due to the ESR of the output
capacitors. Figure 3 shows an example
of the charge current being dialed back
as the capacitor approaches full charge.
The current is typically dialed back at
95% of programmed output voltage,
and this is the voltage that should be
used for the charge time calculation.
The circuit of Figure 1 charges 100F to
4.2V with a programmed input current
of 3.0A and a VIN of 3.3V. Figure 2 shows
that it takes 1.3 seconds to charge a 1.0F
capacitor to 4.0V (4.0V ≈ 0.95 • 4.2V)
from 0V. Since the capacitor in this
example is one hundred times larger, it
will take roughly 130 seconds to charge
a 100F capacitor to 4.0V from 0V.
To determine how long backup power
can be provided to the system, the operational voltage range on the output needs
to be determined. For this application,
the operational output voltage is 4.2V
down to 1.0V. The stored energy on the
100F capacitor can be determined as:
Figure 4. LTC3122 boost
converter powered by the
LTC3128
1
2
2 1
W = COUT ( VINITIAL ) − COUT ( VFINAL )
2
2
1
2 1
2
= 100 • 4.2 − 100 • 1.0 = 832J
2
2
where W is the work done in joules,
COUT is the total output capacitance,
VINITIAL is the beginning voltage on
COUT, and VFINAL is the minimum
voltage COUT can run down to.
VIN
2.4V TO 5.5V
3.0A
10µF
10µF
tBACKUP =
VMID
5V/DIV
5.00s/DIV
Figure 3. Input current charging profile
capacitor, and PLOAD is the power
draw from the secondary converter.
BALANCING SUPERCAPACITORS
Achieving higher output voltages with
supercapacitors requires putting two
or more cells in series with each other
because the maximum voltage for each
capacitor is typically specified between
2.3V and 2.7V, depending on the manufacturer and type of capacitor. The life of
the capacitor is dependent on the voltage
across the capacitor. To extend capacitor lifetime the voltage on the capacitor
should be regulated below the rated
WSTORED 832J
=
= 554.66s
PLOAD
1.5W
where tBACKUP is the holdup time
of the system, WSTORED is the available stored energy on the output
3.3µH
SW2
SW
VOUT = 4.2V
RSENP VOUTP
RSENS VOUTS
LTC3128
VIN
RUN
MID
PFI
FB
PFO
PGOOD
MAXV
PROG
GND
10µF
1.87M
4.7µF
100F
VIN
OFF ON
BURST PWM
SD
470pF
301k
LTC3122
PWM/SYNC
100nF
CAP
FB
VCC
VC
SGND
4.7µF
VOUT
12V
800mA
VOUT
RT
57.6k
3.57k
26 | November 2015 : LT Journal of Analog Innovation
VOUT
2V/DIV
If a secondary boost converter is
connected to VOUT, it acts as a constant
power draw from the supercapacitor.
Figure 4 shows an example of a secondary
boost converter being powered by VOUT
of the LTC3128. The LTC3122 data sheet
shows that for a 12V output with a 100m A
load, the average converter efficiency
across a 1V to 4.2V input is approximately 80%, resulting in a 1.5W constant
power load on the holdup capacitor. The
holdup time can be determined by:
3.3µH
SW1
IIN
2A/DIV
PGND
1.02M
22µF
113k
210k
390pF
10pF
design features
supercapacitors. The LTC3128 allows the
secondary converter to pull its current
through a current sense resistor internal
to the LTC3128. This allows the secondary
converter to draw the required current
from the supply, up to 4A, and the
LTC3128 will charge the output capacitors with the programmed input current,
less the current drawn from the secondary converter. As long as the secondary
converter never draws more than the
programmed input current, the LTC3128
limits the total current draw from the
input supply to the programmed value,
while charging the backup capacitors
with the remaining available current.
3.3µH
SW1
VOUT = 4.2V
MAXV = 2.7V
SW2
RSENP VOUTP
RSENS VOUTS
LTC3128
VIN
MID
RUN
PFI
PFO
FB
PGOOD
PROG
MAXV
GND
VIN
2.4V TO 5.5V
3.0A
10µF
10µF
135k
10µF
1.87M
TO LOAD
200F
10µF
470pF
3.57k
maximum voltage. Capacitor vendors
typically specify how to derate the voltage
on their supercapacitors to extend life.
The LTC3128 integrates a programmable
maximum capacitor voltage comparator
and an efficient active charge balancer. The
maximum capacitor voltage comparators
look at the voltage across each individual
capacitor and ensure that the programmed
voltage is not exceeded while charging.
If the maximum programmed capacitor
voltage is reached on either capacitor,
the LTC3128 halts charging to balance
the cells and then resumes charging.
The active charge balancer uses the
inductor of the LTC3128 to efficiently
move charge from one capacitor to
another to balance them, so that the
capacitors maintain the same voltage
across them. This is important because
during a holdup event, if the capacitors
are far enough out of balance, the polarity of one of the cells could become
200F
Figure 5. LTC3128 with
charge balancer and
maximum capacitor
voltage protection
301k
reversed, damaging the capacitor. The
LTC3128 will only balance the cells if one
of the cells has violated its programmed
maximum capacitor voltage, or if the
output voltage is in regulation and the
capacitors are out of balance but the
maximum voltage has not been violated.
To extend backup time, the LTC3128
draws less than 1µ A from VOUT
when in shutdown, or less than 2µ A
when in input UVLO. Figure 6 shows
a power ride-through application
using the LTC3128 and LTC3122.
Active charge balancing eliminates the
high quiescent current and continuous
power dissipation of passive balancing.
Figure 5 shows the LTC3128 configured
with 100F of total output capacitance, a
programmed output voltage of 4.2V, and a
maximum capacitor voltage of 2.7V, each.
CONCLUSION
The LTC3128 3A buck-boost DC/DC
supercapacitor charger is a streamlined solution for efficiently charging
and protecting supercapacitors in
high reliability, long-life applications.
It features a ±2% accurate programmable input current limit, programmable
maximum capacitor voltage comparators and active charge balancing. n
POWER RIDE-THROUGH
APPLICATION
In a backup system, the ability to wait for
the storage capacitors to charge before
you begin operating is not always an
option. A power ride-through application
provides a means to power the secondary converter directly from the input
supply while simultaneously charging the
3.3µH
Figure 6. Power ride-through
application using the LTC3128
and the LTC3122 boost
converter
3.3µH
VOUT = 4.2V
MAXV = 2.7V
SW2
RSENP VOUTP
RSENS VOUTS
LTC3128
VIN
MID
RUN
PFI
PFO
FB
PGOOD
PROG
MAXV
GND
SW1
VIN
2.4V TO 5.5V
3.0A
10µF
10µF
135k
10µF
1.87M
200F
OFF ON
BURST PWM
10µF
4.7µF
470pF
3.57k
SW
VIN
200F
301k
SD
LTC3122
PWM/SYNC
100nF
CAP
RT
FB
VCC
VC
SGND
57.6k
4.7µF
VOUT
12V
800mA
VOUT
PGND
1.02M
22µF
113k
210k
390pF
10pF
November 2015 : LT Journal of Analog Innovation | 27
What’s New with LTspice IV?
Gabino Alonso
—Follow @LTspice at www.twitter.com/LTspice
—Like us at facebook.com/LTspice
BLOG BY ENGINEERS, FOR
ENGINEERS
Check out the LTspice® blog
(www.linear.com/solutions/LTspice)
for tech news, insider tips and
interesting points of view.
New Article: “Achieving Low
On-Resistance with Guaranteed SOA in
High Current Hot Swap Applications”
by Dan Eddleman
www.linear.com/solutions/5722
The requirement for live insertion and
removal in high current backplane applications demands MOSFETs that exhibit
both low on-resistance during steady
state operation and high safe operating area (SOA) for transient conditions.
Often, modern MOSFETs optimized for
low on-resistance are unsuitable for high
SOA hot swap applications. This article
overviews an application that provides
the best of both worlds by utilizing the
LTC4234 to satisfy SOA requirements and
an external low on-resistance MOSFET
reduces the overall power dissipation.
SELECTED DEMO CIRCUITS
For a complete list of simulations utilizing
Linear Technology’s devices, please visit
www.linear.com/democircuits.
Buck Regulators
• LT3697: 5V step-down converter with
cable drop compensation & output
current limit (8V–35V to 5V at 6A)
www.linear.com/solutions/5476
• LT8613: 5V step-down converter with 6A
output current limit (5.8V–42V to 5V at
6A) www.linear.com/solutions/5751
• LT8616: 5V, 3.3V, 2MHz step-down
converter (5.8V–42V to 5V at 1.5A & 3.3V
at 2.5A) www.linear.com/solutions/5753
• LT8640: 2MHz µPower ultralow
EMI
buck converter (5.7V–42V to 5V at 5A)
www.linear.com/solutions/5635
What is LTspice IV?
LTspice IV is a high performance SPICE
simulator, schematic capture and waveform
viewer designed to speed the process of power
supply design. LTspice IV adds enhancements
and models to SPICE, significantly reducing
simulation time compared to typical SPICE
simulators, allowing one to view waveforms for
most switching regulators in minutes compared
to hours for other SPICE simulators.
LTspice IV is available free from Linear
Technology at www.linear.com/LTspice. Included
in the download is a complete working version of
LTspice IV, macro models for Linear Technology’s
power products, over 200 op amp models, as
well as models for resistors, transistors and
MOSFETs.
28 | November 2015 : LT Journal of Analog Innovation
Buck-Boost Controllers
• LT8709: Negative buck-boost
regulator with output current
monitor and power good (−4.5V to
−38V input to −12V output at 5A)
www.linear.com/solutions/5719
SEPIC Converter
• LT8494: 450k Hz , 5V output
SEPIC
converter (3V–60V to 5V at 1A)
www.linear.com/solutions/5848
Isolated Regulator
• LTM®8057: 2kV AC isolated low noise
µModule regulator (3.1V–29V to 5V at
300m A) www.linear.com/solutions/5206
LED Driver
• LT3952: Short-circuit robust boost
LED
driver (7V–42V to 50V LED string at
333m A) www.linear.com/solutions/5749
SELECT MODELS
To search the LTspice library for a particular device model, choose Component
from the Edit menu or press F2. Since
LTspice is often updated with new
features and models, it is good practice to
update to the current version by choosing Sync Release from the Tools menu.
The changelog.txt file (see root installation directory) list provides a revision
history of changes made to the program.
Linear Regulators
• LT3042: 20V, 200m A, ultralow noise,
ultrahigh PSRR RF linear regulator
www.linear.com/LT3042
• LT3088: 800m A single resistor
rugged linear regulator
www.linear.com/product/LT3088
Buck Regulators
• LT8602: 42V quad monolithic
synchronous step-down regulator
www.linear.com/LT8602
• LTC3887: Dual output PolyPhase®
step-down DC/DC controller with
digital power system management
www.linear.com/LTC3887
• LTC7138: High efficiency, 140V
400m A step-down regulator
www.linear.comt/LTC7138
• LTM4622: Dual ultrathin 2.5A step-
down DC/DC µModule regulator
www.linear.com/LTM4622
design ideas
Be sure to check for recently added
demonstration circuit simulations, such as this
wide input voltage range boost/SEPIC/inverting
controller: 2.5V to 36V input, 12V/2A output
SEPIC converter (automotive 12V regulator).
Multitopology Controller
• LT8570: Boost/SEPIC/inverting
DC/DC
converter with 65V switch, soft-start and
synchronization www.linear.com/LT8570
Surge Stopper
• LTC7860: High efficiency switching surge
stopper www.linear.com/LTC7860
• LTM4675: Dual 9A or single 18A µModule
regulator with digital power system
management www.linear.com/LTM4675
• LTM4676A: Dual 13A or single 26A
µModule regulator with digital
power system management
www.linear.com/LTM4676A
Wake-Up Timer
• LTC2956: Wake-up timer with pushbutton
control www.linear.com/LTC2956 n
Power User Tip
TIGHTEN UP YOUR SCHEMATICS: COMBINE MULTIPLE MODEL INSTANCES INTO ONE SYMBOL
When you need multiple instances of a model, it is easy to copy and paste a symbol,
but sometimes you can tighten up your schematics by using a single symbol to define
multiple instances of same device. For instance, instead of placing four identical
capacitor symbols in parallel, use one symbol times four, “x4”. This feat can be
accomplished using the M (parallel units) or N (series units) parameters.
A number of intrinsic devices support the M (parallel units) parameter, such as the
capacitor, inductor, diode and MOSFET models. If you are not sure if the model
supports the M (parallel units) parameter, try it, and if you do not get an error
message, you should be good. The diode (including LED) model is the only intrinsic
model that supports N (series units) parameter.
To define multiple instances of a model in a device symbol:
1.Ctrl + right-click the symbol to edit the component attributes.
2.Insert “m=<number>” or “n=<number>” into the Value2 field. Note that
non-integer <number> values are allowed.
3.Make the multiple instances visible in your schematic by selecting the Value2
attribute and clicking the Vis column.
Parallel Capacitors
To match certain electrical schematic
standards you can define parallel
capacitors either using “m=<number>”
or “x<number>” syntax as in “x4”.
Series (String) of LEDs
Diodes are the only intrinsic models that
support the N (series units) parameter.
Happy simulations!
November 2015 : LT Journal of Analog Innovation | 29
Boost 12V to 140V with a Single Converter IC
Victor Khasiev
Generating a high voltage from a much lower voltage
presents a number of challenges for the classical single
stage boost topology. For instance, the maximum
duty cycle limitation of a boost controller may not
allow the required step-up ratio. Even if it does, there
is often a sharp decrease in efficiency at high duty
cycles. The duty cycle can be shortened by choosing
discontinuous mode of operation, but this leads to high
peak input current, higher losses and EMI challenges.
An alternative to a single boost converter
is a 2-stage boost converter, where the
first stage produces an intermediate
voltage and the second stage boosts
to the final high voltage. A 2-stage
converter can be produced with a single
controller IC, such as the LTC3788, a
high performance 2-phase dual output
synchronous boost controller, which
drives all N‑channel power MOSFETs.
The LTC3788 can be configured such that
the first boost stage takes advantage of its
synchronous rectification feature, which
maximizes efficiency, reduces power losses
Figure 1. Block diagram of LTC3788-based 2-stage boost converter
CINT
40V ABS MAX
TG1
SW1
PGOOD1
ILIM
SS1
ITH1
FB1
SENSE1+
Q1
RS1
Q2
SENSE1–
VIN
GND
BOOST1
FREQ
BG1
PHASMD
VBIAS
CLKOUT
PGND
LTC3788
BG2
RUN2
BOOST2
SS2
ITH2
FB2
SENSE2
TG2
RUN1
SW2
INTVCC 5V
PGOOD2
SGND
+
EXTVCC
SENSE2–
PLLIN/MODE
GND
L1
U1
30 | November 2015 : LT Journal of Analog Innovation
VOUT
COUT
D1
L2
DR
7V TO 10V
MOSFET
DRIVER
Q3
GND
RS2
GND
and eases thermal requirements. The
maximum output voltage of this controller
is 60V, when using synchronous rectification. If greater than 60V is required,
the second stage can be designed to run
non-synchronously, as described below.
2-STAGE BOOST PRODUCES 140V
FROM 12V
The block diagram in Figure 1 shows
the LTC3788 in a 2-stage boost configuration. This block diagram also reveals a
few caveats that must be observed in this
design:
•The output of the first stage (Q1,
CINT) is connected to the input of
second stage (RS2, L2). The output
of the first stage should not exceed
40V, because the maximum absolute
rating of the SENSE pins is 40V.
•The gate drive voltage of 5V is suitable
for logic level MOSFETs, but not for high
voltage standard MOSFETs, with typical
gate voltages of 7V to 12V. The external
gate driver DR, controlled by the BG2
signal can be used as shown here to
drive high voltage standard MOSFETs.
•To generate an output voltage
above maximum limit of 60V, the
synchronous rectification MOSFET
is replaced by a single diode D1.
Figure 2 shows the complete solution.
Transistors Q1, Q2 and inductor L1
compose the first stage, which generates an intermediate bus voltage of 38V.
The first stage employs synchronous
rectification for maximum efficiency.
The output of the first stage is connected
as input to the second stage, comprised
design ideas
12.1k
V_INT
2.21k
V_INT
PLLIN/MODE
EXTVCC
SGND
INTVCC
RUN1
BG2
SW2
SS2
PGOOD2
ITH2
FB2
SENSE2+
SENSE2–
GND
12.1k
DRIVER BIAS
CIRCUIT
5.1k
Q4
MMBTA42LT1G
8.2V
SNS2+
SNS2–
1μF
100pF
2.7k
6.98k
+
Q2
BSC028N06LS3G
VIN
82μF/50V
50HVH82M
GND
V_INT
VOUT 140V AT 1.0A
2x0.47μF/450V
C4532X7T2W474M
+
22μF/200V
EEVEB2D220SQ
VOUT
GND
D1
SBR10U200P5130
0.1μF
15nF
VIN 3V TO 36V
4x4.7μF
4.7μF
BOOST2
RUN2
309k
SENSE1– SENSE1+
L1, 6.8μF
SER2915H682 RS1, 0.002Ω
0.1μF
PGND
LTC3788EUH
82μF/50V
50HVH82M
Q1
BSC067N06LS3G
VBIAS
VBIAS
CLKOUT
VBIAS
BAS140W
BG1
PHASMD
+
4x4.7μF
0.1μF
BOOST1
FREQ
42.2k
TG1
SW1
ILIM
SS1
ITH1
SENSE1–
U1
TG2
SNS1–
FB1
SENSE1+
0.1μF
INTERMEDIATE BUS
15nF
PGOOD1
8.66k
SNS1+
100pF
374k
U2
VBIAS
LTC4440
BST
VCC
GND
TG
IN
TS
L2, 100μH
PCV210405L
SENSE2–
SENSE2+
RS2, 0.01Ω
Q3
BSC320N20NS3
806k
Figure 2. Full schematic of 2-stage 140V output, 1A boost converter
of Q3, D1, L2. The output of the
second stage produces 140V at 1A.
Transistor Q3 is a standard level
MOSFET, driven by the LTC4440.
Here, an LDO, based on transistor Q4,
biases the gate driver, but a switching
94
regulator can be employed instead
(such as one built around the LTC3536)
to further increase overall efficiency.
This solution features an input voltage
range from 3V to 36V, nominal 12V. To
decrease components’ thermal stress, the
VOUT = 140V
EFFICIENCY (%)
CONCLUSION
VIN = 12V
IOUT = 1A
92
LTC3788 is a high performance 2-phase
dual output synchronous boost controller, suitable for high power, high voltage
applications. Its dual outputs can be
used in tandem to achieve extremely
high step-up ratios to high voltages. n
90
VOUT
20V/DIV
88
86
84
output current should be reduced
when the input voltages fall below 10V.
Figure 3 shows measured efficiency,
and Figure 4 shows the start-up waveforms. A 93% efficiency is shown with
VIN = 24V and with the 140V output
loaded from 0.4A to 1A. This converter
can operate at full load with no airflow.
VIN = 24V
VIN = 12V
VIN = 8V
0
0.2
0.6
0.4
LOAD CURRENT (A)
0.8
Figure 3. Efficiency of the 2-stage
converter in Figure 2
1
5ms/DIV
Figure 4. Start-up waveforms
November 2015 : LT Journal of Analog Innovation | 31
highlights from circuits.linear.com
100V INPUT, 5V OUTPUT, LOW RIPPLE 1A STEP-DOWN CONVERTER
The LT8631 is a current mode PWM step-down DC/DC converter with internal
synchronous switches that provides current for output loads up to 1A. The
wide input range of 3V to 100V makes the LT8631 suitable for regulating power
from a wide variety of sources, including automotive and industrial systems
and 36V to 72V telecom supplies. Low ripple Burst Mode® operation enables
high efficiency operation down to very low output currents while keeping the
output ripple below 10mVP-P. Resistor programmable 100kHz to 1MHz frequency
range and synchronization capability enable optimization between efficiency
and external component size. The soft-start feature controls the ramp rate
of the output voltage, eliminating input current surge during start-up, while
also providing output tracking. A power good flag signals when the output
voltage is within ±7.5% of the regulated output. Undervoltage lockout can be
programmed using the EN/UV pin. Shutdown mode reduces the total quiescent
current to < 5μA. The LT8631 is available in a 20-lead TSSOP package with
exposed pad for low thermal resistance and high voltage lead spacing.
www.linear.com/solutions/5855
VIN
3.3V
VOUT1
0.1µF
10µF
60.4k
PGOOD1 PGOOD2
VIN
VOUT1
RUN1
RUN2 LTM4622 VOUT2
INTVCC
COMP1
SYNC/MODE
COMP2
TRACK/SS1
FB1
TRACK/SS2
FREQ
60.4k
10µF
10µF
VOUT1
1.5V, 2.5A
60.4k
VOUT2
1.2V, 2.5A
40.2k
4622 F27
4.35V TO 5V
INPUT SUPPLY
(PROTECTED
TO 40V)
5V BACKUP APPLICATION WITH OVERVOLTAGE PROTECTION
AND NON-BACKED-UP-LOAD OPTION (CHARGE CURRENT
SETTING: 2.5A, INPUT CURRENT LIMIT SETTING: 4A)
The LTC4040 is a complete 3.5V to 5.5V supply rail battery
backup system. It contains a high current step-up DC/DC
regulator to back up the supply from a single-cell Li-Ion or
LiFePO4 battery. When external power is available, the step-up
regulator operates in reverse as a step-down battery charger.
www.linear.com/solutions/5821
VPWR
VIN
2.2µF
BST
LT8631
EN/UV
0.1µF
SW
22µH
PG
IND
INTVCC
2.2µF
VOUT
47pF
RT
25.5k
1M
VOUT
5V, 1A
FB
SYNC/MODE
191k
TR/SS
GND
47µF ×4
1210, 16V
0.1µF
FSW = 400kHz
L: TDK SLF1O145T-22OM1R9
ULTRATHIN µModule 3.3V INPUT, 1.5V AND 1.2V OUTPUTS AT
2.5A DESIGN WITH OUTPUT COINCIDENT TRACKING
The LTM4622 is a complete dual 2.5A step-down switching mode µModule®
(micromodule) regulator in a tiny ultrathin 6.25mm × 6.25mm × 1.82mm LGA
package. Included in the package are the switching controller, power FETs, inductor
and support components. Operating over an input voltage range of 3.6V to 20V,
the LTM4622 supports an output voltage range of 0.6V to 5.5V, set by a single
external resistor. Its high efficiency design delivers dual 2.5A continuous, 3A peak,
output current. Only a few ceramic input and output capacitors are needed.
www.linear.com/solutions/5841
FB2
GND
VIN
6.5V TO 100V
RS
6mΩ
MN1
2.2µF
6.2k 1/4W
OVP OPT
178k
MN2
VIN CLN
OVSNS
PFI
60.4k
FAULT
PFO
RST
CHRG
CLPROG
TO NON-BACKED-UP
LOAD
VSYS
4.35V TO 5V
SW
BAT
LTC4040
100µF
1690k
VSYS
BSTFB
RSTFB
IGATE
TO BACKED-UP
SYSTEM LOAD
324k
2.2µH
10µF
VIN
RBIAS
NTC
CHGOFF BSTOFF GND F0 F1 F2 PROG
VSYS
800Ω
NTC
+
Li-Ion
BATTERY
4.1V
L1: COILCRAFT XAL-5030-222
MN1: VISHAY/SILICONIX SiS488DN
MN2: VISHAY/SILICONIX SIR424DP-T1-GE3
L, LT, LTC, LTM, Linear Technology, the Linear logo, Burst Mode, PolyPhase, LTspice and µModule are registered trademarks and Linduino is a trademark of Linear Technology Corporation. All other trademarks are the property of their respective owners.
© 2015 Linear Technology Corporation/Printed in U.S.A./71.5K
Linear Technology Corporation
1630 McCarthy Boulevard, Milpitas, CA 95035
(408) 432-1900
www.linear.com
Cert no. SW-COC-001530