1
4012fa
LTC4012/
LTC4012-1/LTC4012-2
Features
applications
Description
High Efficiency,
Multi-Chemistry Battery Charger
with PowerPath Control
The LTC4012 is a constant-current/constant-volt-
age battery charger controller. It uses a synchronous
quasi-constant frequency PWM control architecture
that will not generate audible noise with ceramic bulk
capacitors. Charge current is set by external resis-
tors and can be monitored as an output voltage.
With no built-in termination, the LTC4012 family charges
a wide range of batteries under external control.
The LTC4012 features fully adjustable output voltage, while
the LTC4012-1 and LTC4012-2 can be pin-programmed for
Lithium-Ion/Polymer battery packs of 1-, 2-, 3- or 4-series
cells. The LTC4012-1 provides output voltage of 4.1V/cell
and the LTC4012-2 is a 4.2V/cell version.
The device includes AC adapter input current limiting, which
maximizes charge rate for a fixed input power level. An
external sense resistor programs the input current limit,
and the ICL status pin indicates reduced charge current as
a result of AC adapter current limiting. Ideal diode control
at the adaptor input improves charger efficiency.
The CHRG status pin is active during all charging modes,
including special indication for low charge current.
n General Purpose Battery Charger Controller
n Efficient 550kHz Synchronous Buck PWM Topology
n ±0.5% Output Float Voltage Accuracy
n Programmable Charge Current: 4% Accuracy
n Programmable AC Adapter Current Limit:
3% Accuracy
n No Audible Noise with Ceramic Capacitors
n INFET Low Loss Ideal Diode PowerPath™ Control
n Wide Input Voltage Range: 6V to 28V
n Wide Output Voltage Range: 2V to 28V
n Indicator Outputs for AC Adapter Present, Charging,
C/10 Current Detection and Input Current Limiting
n Analog Charge Current Monitor
n Micropower Shutdown
n 20-Pin 4mm × 4mm × 0.75mm QFN Package
n Notebook Computers
n Portable Instruments
n Battery Backup Systems
Efficiency at DCIN = 20V
L, LT, LTC, LTM, Linear Technology and the Linear logo are registered trademarks and
PowerPath and ThinSOT are trademarks of Linear Technology Corporation. All other trademarks
are the property of their respective owners. Protected by U.S. Patents including 5723970.
CLP
FROM
ADAPTER
13V TO 28V 0.1µF 5.1k
25mΩ
3.01k
0.1µF
6.8µH
20µF
33mΩ
INFET
0.1µF
2µF
0.1µF
6.04k
26.7k
LTC4012
DCIN
ACP
CHRG
ICL
SHDN
ITH
PROG
CLN
BOOST
TGATE
SW
INTVDD
BGATE
TO/FROM
MCU
GND
CSP
CSN
BAT
FBDIV
VFB
20µF
POWER TO
SYSTEM
4.7nF
32.8k
4012 TA01
12.3V
Li-Ion
BATTERY
3.01k
301k +
CHARGE CURRENT (A)
0
70
EFFICIENCY (%)
POWER LOSS (mW)
75
80
85
90
100
100
1000
10000
0.5 1 1.5 2
4012 TA02
2.5 3
95
PART
LTC4009
LTC4012
INFET
X
DCDIV
X
PIN NAME
VOUT = 12.3V
RSENSE = 33mΩ
RIN = 3.01k
RPROG = 26.7k
EFFICIENCY
POWER LOSS
typical application
LTC4012/
LTC4012-1/LTC4012-2
2
4012fa
absolute MaxiMuM ratings
DCIN ............................................................14V to 30V
DCIN to CLP ................................................ 32V to 20V
CLP, CLN or SW to GND ............................. 0.3V to 30V
CLP to CLN ............................................................±0.3V
CSP, CSN or BAT to GND ........................... 0.3V to 28V
CSP to CSN ............................................................±0.3V
BOOST to GND ........................................... 0.3V to 36V
BOOST to SW...............................................0.3V to 7V
(Note 1)
orDer inForMation
LEAD FREE FINISH TAPE AND REEL PART MARKING* PACKAGE DESCRIPTION TEMPERATURE RANGE
LTC4012CUF#PBF LTC4012CUF#TRPBF 4012 20-Lead (4mm × 4mm) Plastic QFN 0°C to 85°C
LTC4012CUF-1#PBF LTC4012CUF-1#TRPBF 40121 20-Lead (4mm × 4mm) Plastic QFN 0°C to 85°C
LTC4012CUF-2#PBF LTC4012CUF-2#TRPBF 40122 20-Lead (4mm × 4mm) Plastic QFN 0°C to 85°C
LTC4012IUF#PBF LTC4012IUF#TRPBF 4012 20-Lead (4mm × 4mm) Plastic QFN –40°C to 125°C
LTC4012IUF-1#PBF LTC4012IUF-1#TRPBF 40121 20-Lead (4mm × 4mm) Plastic QFN –40°C to 125°C
LTC4012IUF-2#PBF LTC4012IUF-2#TRPBF 40122 20-Lead (4mm × 4mm) Plastic QFN –40°C to 125°C
Consult LTC Marketing for parts specified with wider operating temperature ranges. *The temperature grade is identified by a label on the shipping container.
Consult LTC Marketing for information on non-standard lead based finish parts.
For more information on lead free part marking, go to: http://www.linear.com/leadfree/
For more information on tape and reel specifications, go to: http://www.linear.com/tapeandreel/
SHDN, FVS0, FVS1 or VFB to GND ...............0.3V to 7V
ACP, CHRG or ICL to GND .......................... 0.3V to 30V
Operating Temperature Range
(Note 2) ............................................. 40°C to 125°C
Junction Temperature (Note 3) ............................. 125°C
Storage Temperature Range ..................65°C to 150°C
pin conFiguration
LTC4012 LTC4012-1
LTC4012-2
20 19 18 17 16
6 7 8
TOP VIEW
21
GND
UF PACKAGE
20-LEAD (4mm s 4mm) PLASTIC QFN
9 10
5
4
3
2
1
11
12
13
14
15
CLN
CLP
INFET
DCIN
ACP
CSP
CSN
PROG
ITH
BAT
BOOST
TGATE
SW
INTVDD
BGATE
SHDN
CHRG
ICL
VFB
FBDIV
TJMAX = 125°C, JA = 37°C/W
EXPOSED PAD (PIN 21) IS GND, MUST BE SOLDERED TO PCB
20 19 18 17 16
678
TOP VIEW
21
GND
UF PACKAGE
20-LEAD (4mm s 4mm) PLASTIC QFN
9 10
5
4
3
2
1
11
12
13
14
15
CLN
CLP
INFET
DCIN
ACP
CSP
CSN
PROG
ITH
BAT
BOOST
TGATE
SW
INTVDD
BGATE
SHDN
CHRG
ICL
FVS0
FVS1
TJMAX = 125°C, JA = 37°C/W
EXPOSED PAD (PIN 21) IS GND, MUST BE SOLDERED TO PCB
3
4012fa
LTC4012/
LTC4012-1/LTC4012-2
electrical characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. DCIN = 20V, BAT = 12V, GND = 0V unless otherwise noted. (Note 2)
SYMBOL PARAMETER CONDITIONS MIN TYP MAX UNITS
Charge Voltage Regulation
VTOL VBAT Accuracy (See Test Circuits) LTC4012
C-Grade
I-Grade
l
l
–0.5
–0.8
–1.0
0.5
0.8
1.0
%
%
%
LTC4012-1/LTC4012-2
C-Grade
FVS1 = 0V, FVS0 = 0V, I-Grade
FVS1 = 0V, FVS1 = 5V, I-Grade
FVS1 = 5V, FVS0 = 0V, I-Grade
FVS1 = 5V, FVS1 = 5V, I-Grade
l
l
l
l
l
–0.6
–0.8
–1.1
–1.15
–1.25
–1.35
0.6
0.8
1.1
1.15
1.25
1.35
%
%
%
%
%
%
IVFB VFB Input Bias Current VFB = 1.2V ±20 nA
RON FBDIV On Resistance ILOAD = 100µA l85 190 Ω
ILEAK-FBDIV FBDIV Output Leakage Current SHDN = 0V, FBDIV = 0V l–1 0 1 µA
VBOV VFB Overvoltage Threshold LTC4012 l1.235 1.281 1.32 V
BAT Overvoltage Threshold LTC4012-1/LTC4012-2, Relative to
Selected Output Voltage
l103 106 109 %
Charge Current Regulation
ITOL Charge Current Accuracy with RIN = 3.01k,
6V < BAT < 18V (LTC4012)
6V < BAT < 15V (LTC4012-1, LTC4012-2)
RPROG = 26.7k
C-Grade
I-Grade
l
l
–4
–5
–9.5
4
5
9.5
%
%
%
VSENSE = 0mV, PROG = 1.2V –12.75 –11.67 –10.95 µA
AICurrent Sense Amplifier Gain (PROG ∆I) with
RIN = 3.01k, 6V < BAT < 18V (LTC4012)
6V < BAT < 15V (LTC4012-1, LTC4012-2)
VSENSE Step from 0mV to 5mV,
PROG = 1.2V
–1.78 –1.66 –1.54 µA
VCS-MAX Maximum Peak Current Sense Threshold Voltage
per Cycle (RIN = 3.01k)
ITH = 2V, C-Grade
ITH = 2V, I-Grade
ITH = 5V
l
l
l
140
125
195
325
250
265
430
mV
mV
mV
VC10 C/10 Indicator Threshold Voltage PROG Falling 340 400 460 mV
VREV Reverse Current Threshold Voltage PROG Falling 180 253 295 mV
Input Current Regulation
VCL Current Limit Threshold CLP – CLN
C-Grade
I-Grade
l
l
97
96
92
100
100
103
104
108
mV
mV
mV
ICLN CLN Input Bias Current CLN = CLP ±100 nA
VICL ICL Indicator Threshold (CLP – CLN) – VCL –8 5 –2 mV
CLP Supply
OVR Operating Voltage Range 6 28 V
VUVLO CLP Undervoltage Lockout Threshold CLP Increasing l4.65 4.85 5.25 V
VUV(HYST) UVLO Threshold Hysteresis 200 mV
ICLPO CLP Operating Current CLP = 20V, No Gate Loads 2 3 mA
Shutdown
VACP AC Present Threshold Voltage DCIN – BAT, DCIN Rising
C-Grade
I-Grade
l
l
350
300
500
650
700
mV
mV
VACP(HYST) ACP Threshold Hysteresis Voltage 200 mV
VIL SHDN Input Voltage Low l300 mV
VIH SHDN Input Voltage High l1.4 V
LTC4012/
LTC4012-1/LTC4012-2
4
4012fa
electrical characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. DCIN = 20V, BAT = 12V, GND = 0V unless otherwise noted. (Note 2)
SYMBOL PARAMETER CONDITIONS MIN TYP MAX UNITS
RIN SHDN Pull-Down Resistance 40 kΩ
ICLPS CLP Shutdown Current CLP = 12V, DCIN = 0V
SHDN = 0V
l15
350
26
500
µA
µA
ILEAK-BAT BAT Leakage Current SHDN = 0V or DCIN = 0V,
0V ≤ CSP = CSN = BAT ≤ 18V
l–1.5 0 1.5 µA
ILEAK-CSN CSN Leakage Current SHDN = 0V or DCIN = 0V,
0V ≤ CSP = CSN = BAT ≤ 20V
l–1.5 0 1.5 µA
ILEAK-CSP CSP Leakage Current SHDN = 0V or DCIN = 0V,
0V ≤ CSP = CSN = BAT ≤ 20V
l–1.5 0 1.5 µA
ILEAK-SW SW Leakage Current SHDN = 0V or DCIN = 0V,
0V ≤ SW ≤ 20V
l–1 0 2 µA
INTVDD Regulator
INTVDD Output Voltage No Load l4.85 5 5.15 V
VDD Load Regulation IDD = 20mA 0.4 –1 %
IDD Short-Circuit Current (Note 5) INTVDD = 0V 50 85 130 mA
Switching Regulator
IITH ITH Current ITH = 1.4V 40/+90 µA
fTYP Typical Switching Frequency 467 550 633 kHz
fMIN Minimum Switching Frequency CLOAD = 3.3nF 20 25 kHz
DCMAX Maximum Duty Cycle CLOAD = 3.3nF 98 99 %
tR-TG TGATE Rise Time CLOAD = 3.3nF, 10% – 90% 60 110 ns
tF-TG TGATE Fall Time CLOAD = 3.3nF, 90% – 10% 50 110 ns
tR-BG BGATE Rise Time CLOAD = 3.3nF, 10% – 90% 60 110 ns
tF-BG BGATE Fall Time CLOAD = 3.3nF, 90% – 10% 60 110 ns
tNO TGATE, BGATE Non-Overlap Time CLOAD = 3.3nF, 10% – 10% 110 ns
PowerPath Control
IDCIN DCIN Input Current 0V ≤ DCIN ≤ CLP l–10 60 µA
VFTO Forward Turn-On Voltage (DCIN Detection Threshold) DCIN-CLP, DCIN rising l15 60 mV
VFR Forward Regulation Voltage DCIN-CLP l15 25 35 mV
VRTO Reverse Turn-Off Voltage DCIN-CLP, DCIN falling l45 –25 –15 mV
VOL(INFET) INFET Output Low Voltage, Relative to CLP DCIN-CLP = 0.1V, IINFET =1µA –6.5 –5 V
VOH(INFET) INFET Output High Voltage, Relative to CLP DCIN-CLP = –0.1V, IINFET =5µA –250 250 mV
tIF(ON) INFET Turn-On Time To CLP-INFET > 3V, CINFET = 1nF 85 180 µs
tIF(OFF) INFET Turn-Off Time To CLP-INFET < 1.5V, CINFET = 1nF 2.5 6 µs
Float Voltage Select Inputs (LTC4012-1/LTC4012-2 Only)
VIL Input Voltage Low 0.5 V
VIH Input Voltage High 3.5 V
IIN Input Current 0V ≤ VIN ≤ 5V –10 10 µA
Indicator Outputs
VOL Output Voltage Low ILOAD = 100µA, PROG = 1.2V 500 mV
ILEAK Output Leakage SHDN = 0V, DCDIV = 0V,
VOUT = 20V
l–10 10 µA
IC10 CHRG C/10 Current Sink CHRG = 2.5V l15 25 38 µA
5
4012fa
LTC4012/
LTC4012-1/LTC4012-2
Note 1: Stresses beyond those listed under Absolute Maximum Ratings
may cause permanent damage to the device. Exposure to any Absolute
Maximum Rating condition for extended periods may affect device
reliability and lifetime.
Note 2: The LTC4012C is guaranteed to meet performance specifications
over the 0°C to 85°C operating temperature range. The LTC4012I is
guaranteed to meet performance specifications over the –40°C to 125°C
operating temperature range.
Note 3: Operating junction temperature TJ (in °C) is calculated from
the ambient temperature TA and the total continuous package power
dissipation PD (in watts) by the formula TJ = TA + (θJA • PD). Refer to the
Applications Information section for details.
Note 4: All currents into device pins are positive; all currents out of device
pins are negative. All voltages are referenced to GND, unless otherwise
specified.
Note 5: Output current may be limited by internal power dissipation. Refer
to the Applications Information section for details.
test circuits
+
9
13 12
1.2085V
1.2085V
TARGET
PROG VFB ITH
40012 TC01
LTC4012
0.6V
EA
FROM ICL
(CLP = CLN)
+
LTC1055
+
11
13 12
1.2085V
TARGET VARIES
WITH FVSO,1
PROG BAT ITH
40012 TC02
LTC4012-1
LTC4012-2
0.6V
EA
FROM ICL
(CLP = CLN)
+
LTC1055
electrical characteristics
LTC4012/
LTC4012-1/LTC4012-2
6
4012fa
Battery Load Dump
BATTERY VOLTAGE
(500mV/DIV)
LOAD
STATE
TIME (1ms/DIV)
CLP = 20V
VOUT = 12.3V
4012 G06
DISCONNECT
RECONNECT
3A
2A
12.1V 1A
1A
typical perForMance characteristics
Efficiency at DCIN = 20V, BAT = 8V
VFB Line Regulation
CHARGE CURRENT (A)
0
80
EFFICIENCY (%)
POWER LOSS(mW)
85
90
100
100
1000
10000
0.5 1 1.5 2
4012 G01
2.5 3
95
POWER LOSS
EFFICIENCY
RSENSE = 33mΩ
RIN = 3.01k
Efficiency at DCIN = 20V, BAT = 12V
CHARGE CURRENT (A)
0
80
EFFICIENCY (%)
POWER LOSS(mW)
85
90
100
100
1000
10000
0.5 1 1.5 2
4012 G02
2.5 3
95 EFFICIENCY
POWER LOSS
RSENSE = 33mΩ
RIN = 3.01k
Efficiency at DCIN = 20V, BAT = 16V
CHARGE CURRENT (A)
0
80
EFFICIENCY (%)
POWER LOSS(mW)
85
90
100
100
1000
10000
0.5 1 1.5 2
4012 G03
2.5 3
95
EFFICIENCY
POWER LOSS
RSENSE = 33mΩ
RIN = 3.01k
CLP PIN VOLTAGE (V)
5
VFB ERROR (%)
0.02
0.06
0.10
25
4012 G04
–0.02
–0.06
0
0.04
0.08
–0.04
–0.08
–0.10 10 15 20 30
LTC4012 TEST CIRCUIT
BATTERY VOLTAGE (V)
0
75
RON (Ω)
100
150
175
200
10 20 25
300
4012 G05
125
5 15
225
250
275
CLP = BAT + 3V
(CLP ≥ 6V)
FBDIV Pin RON vs Battery Voltage
(TA = 25°C unless otherwise noted. L = IHLP-2525 6.8µH)
Charge Current Accuracy
7
4012fa
LTC4012/
LTC4012-1/LTC4012-2
typical perForMance characteristics
PWM Frequency vs Duty Cycle
Gate Drive Non-Overlap
Battery Shutdown Current
Charge Current Line Regulation Input Current Limit
DCIN PIN VOLTAGE (V)
5
–0.5
CHARGE CURRENT ERROR (%)
–0.4
–0.2
–0.1
0
0.5
0.2
10 15
4012 G08
–0.3
0.3
0.4
0.1
20 25
ICHG = 1A
ICHG = 2A
ICHG = 3A
BAT = 6V
RSENSE = 33mΩ
RIN = 3.01k
Charge Current Load Regulation
BATTERY VOLTAGE (V)
11.0
CHARGE CURRENT (A)
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
–0.5
12.6
4012 G09
11.4 11.8 12.2 13.0
ICHG = 3A
ICHG = 2A
ICHG = 1A
DCIN = 20V
RSENSE = 33mΩ
RIN = 3.01k
SYSTEM LOAD (A)
0
CURRENT (A)
0.5
1.0
1.5
1.5 2.5
4012 G10
0
–0.5
–1.0 0.5 1.0 2.0
2.0
2.5
3.0
IIN
ICHG
ICL STATE
2.5A BULK CHARGE
2.1A INPUT CURRENT LIMIT
PWM Soft-Start
ICHG
2A/DIV
TIME (500µs/DIV) 4012 G11
ITH
1V/DIV
PROG
1V/DIV
SHDN
5V/DIV
TIME (80ns/DIV)
EXTERNAL FET DRIVE (1V/DIV)
4012 G12
TGATE
BGATE
DUTY CYCLE (%)
0
0
PWM FREQUENCY (kHz)
100
200
300
400
500
600
20 40 60 80
4012 G13
100
CLP = 6V
CLP = 12V
CLP = 20V
CLP = 25V
ICHG = 750mA
PWM Frequency
vs Charge Current
CHARGE CURRENT (A)
0
0
PWM FREQUENCY (kHz)
100
200
300
400
600
0.5 1.0 1.5 2.0
4012 G14
2.5 3.0
500
BAT = 14.5V
BAT = 12V
BAT = 5V
CLP = 15V
RSENSE = 33mΩ
RIN = 3.01k
BATTERY VOLTAGE (V)
0
0
BATTERY CURRENT (µA)
5
10
15
25
5 10 15
4012 G15
20 25
20
DC1256-CLASS
APPLICATION
DCIN = 0V
LTC4012,
ALL PINS
DCIN = 0V
LTC4012,
BAT PINS
DCIN = 20V
(TA = 25°C unless otherwise noted. L = IHLP-2525 6.8µH)
INFET Response Time to
DCIN Short to Ground
0A
VGS = 0V
TIME (1µs/DIV)
DCIN = 15V
INFET = Si7423DN
IOUT = <50mA
VOUT = 12.3V
COUT = 0.27F
4012 G16
PFET VGS (1V/DIV)
IDCIN, REVERSE
(5A/DIV)
LTC4012/
LTC4012-1/LTC4012-2
8
4012fa
pin Functions
CLN (Pin 1): Adapter Input Current Limit Negative Input.
The LTC4012 senses voltage on this pin to determine if
less charge current should be sourced to limit total input
current. The threshold is set 100mV below the CLP pin. An
external filter should be used to remove switching noise.
This input should be tied to CLP if not used. Operating
voltage range is (CLP – 110mV) to CLP.
CLP (Pin 2): Adapter Input Current Limit Positive Input.
The LTC4012 also draws power from this pin, including
a small amount for some shutdown functions. Operating
voltage range is GND to 28V.
INFET (Pin 3): PowerPath Control Output. This output
drives the gate of a PMOS pass transistor connected
between the DC input (DCIN) and the raw system supply
rail (CLP) to maintain a forward voltage of 25mV when a
DC input source is present. INFET is internally clamped
about 6V below CLP. Maximum operating voltage is CLP,
which is used to turn off the input PMOS transistor when
the DC input is removed.
DCIN (Pin 4): DC Sense Input. One of two voltage sense
inputs to the internal PowerPath controller (the other input
to the controller is CLP). This input is usually supplied
from an input DC power source. Operating voltage ranges
from GND to 28.2V.
ACP (Pin 5): Active-Low AC Adapter Present Indicator
Output. This open-drain output pulls to GND when adequate
AC adapter (DCIN) voltage is present. This output should
be left floating if not used.
SHDN (Pin 6): Active-low Shutdown Input. Driving SHDN
below 300mV unconditionally forces the LTC4012 into the
shutdown state. This input has a 40kΩ internal pulldown
to GND. Operating voltage range is GND to INTVDD.
CHRG (Pin 7): Active-Low Charge Indicator Output. This
open-drain output provides three levels of information
about charge status using a strong pull-down, 25µA weak
pull-down or high impedance. Refer to the Operation and
Applications Information sections for further details. This
output should be left floating if not used.
ICL (Pin 8): Active-Low Input Current Limit Indicator Out-
put. This open-drain output pulls to GND when the charge
current is reduced because of AC adapter input current
limiting. This output should be left floating if not used.
VFB (Pin 9, LTC4012): Battery Voltage Feedback Input. An
external resistor divider between FBDIV and GND with the
center tap connected to VFB programs the charger output
voltage. In constant voltage mode, this pin is nominally
at 1.2085V. Refer to the Applications Information section
for complete details on programming battery voltage.
Operating voltage range is GND to 1.25V.
FVS0 (Pin 9, LTC4012-1/LTC4012-2): Battery Voltage
Select Input (LSB). This pin is one of two pins used on the
LTC4012-1 or LTC4012-2 to select one of four preset battery
voltages. Selection is done by connecting to either GND or
INTVDD. Operating voltage range is GND to INTVDD.
FBDIV (Pin 10, LTC4012): Battery Voltage Feedback
Resistor Divider Source. The LTC4012 connects this pin
to BAT when charging is in progress. FBDIV is an open-
drain PFET output to BAT with an operating voltage range
of GND to BAT.
FVS1 (Pin 10, LTC4012-1/LTC4012-2): Battery Voltage
Select Input (MSB). This pin is one of two pins used on
the LTC4012-1 or LTC4012-2 to select one of four preset
battery voltages. Selection is done by connecting to
either GND or INTVDD. Operating voltage range is GND
to INTVDD.
BAT (Pin 11): Battery Pack Connection. The LTC4012 uses
the voltage on this pin to control PWM operation when
charging. Operating voltage range is GND to CLN.
ITH (Pin 12): PWM Control Voltage and Compensation
Node. The LTC4012 develops a voltage on this pin to
control cycle-by-cycle peak inductor current. An external
R-C network connected to ITH provides PWM loop com-
pensation. Refer to the Applications Information section
for further details on establishing loop stability. Operating
voltage range is GND to INTVDD.
9
4012fa
LTC4012/
LTC4012-1/LTC4012-2
pin Functions
PROG (Pin 13): Charge Current Programming and Monitor-
ing Pin. An external resistance connected between PROG
and GND, along with the current sense and PWM input
resistors, programs the maximum charge current. The
voltage on this pin can also provide a linearized indicator
of charge current. Refer to the Applications Information
section for complete details on current programming and
monitoring. Operating voltage range is GND to INTVDD.
CSN (Pin 14): Charge Current Sense Negative In-
put. Place an external input resistor (RIN, Figure 1)
between this pin and the negative side of the charge
current sense resistor. Operating voltage ranges from
(BAT – 50mV) to (BAT + 200mV).
CSP (Pin 15): Charge Current Sense Positive Input.
Place an external input resistor (RIN, Figure 1) be-
tween this pin and the positive side of the charge
current sense resistor. Operating voltage ranges from
(BAT – 50mV) to (BAT + 200mV).
BGATE (Pin 16): External Synchronous NFET Gate Control
Output. This output provides gate drive to an external NMOS
power transistor switch used for synchronous rectification
to increase efficiency in the step-down DC/DC converter.
Operating voltage is GND to INTVDD. BGATE should be
left floating if not used.
INTVDD (Pin 17): Internal 5V Regulator Output. This pin
provides a means of bypassing the internal 5V regulator
used to power the LTC4012 PWM FET drivers. This supply
shuts down when the LTC4012 shuts down. Refer to the
Application Information section for details if additional
power is drawn from this pin by the application circuit.
SW (Pin 18): PWM Switch Node. The LTC4012 uses the
voltage on this pin as the source reference for its topside
NFET (PWM switch) driver. Refer to the Applications In-
formation section for additional PCB layout suggestions
related to this critical circuit node. Operating voltage range
is GND to CLN.
TGATE (Pin 19): External NFET Switch Gate Control Output.
This output provides gate drive to an external NMOS power
transistor switch used in the DC/DC converter. Operating
voltage range is GND to (CLN + 5V).
BOOST (Pin 20): TGATE Driver Supply Input. A bootstrap
capacitor is returned to this pin from a charge network
connected to SW and INTVDD. Refer to the Applications
Information section for complete details on circuit topol-
ogy and component values. Operating voltage ranges from
(INTVDD – 1V) to (CLN + 5V).
GND (Exposed Pad Pin 21): Ground. The package paddle
provides a single-point ground for the internal voltage
reference and other critical LTC4012 circuits. It must be
soldered to a suitable PCB copper ground pad for proper
electrical operation and to obtain the specified package
thermal resistance.
LTC4012/
LTC4012-1/LTC4012-2
10
4012fa
block DiagraM
+
+
19
EA
R1
TO INTERNAL
CIRCUITS
TO
INTERNAL
CIRCUITS
CC
+
CA
1.2085V
REFERENCE
5V
REGULATOR
PWM
LOGIC
FAULT
DETECTION
C/10
DETECTION
SHUTDOWN
CONTROL
OSCILLATOR
BAT
SHUTDOWN
CHARGE
INPUT
CURRENT
LIMIT
TGATE
20
BOOST
12
ITH
13
PROG
14
CSN
15
CSP
18
SW
21
GND
(PADDLE)
4012 BD01
16
BGATE
17
INTVDD
TO
INTERNAL
CIRCIUTS
11
VFB
9
CHRG
7
FBDIV
10
ACP
6SHDN
ICL
8
CLP
CLN
2
1
INFET
3
DCIN
4
+
IF
5
(LTC4012)
11
4012fa
LTC4012/
LTC4012-1/LTC4012-2
block DiagraM
(LTC4012-1/LTC4012-2)
+
+
19
EA
R1
TO INTERNAL
CIRCUITS
TO
INTERNAL
CIRCUITS CC
+
CA
1.2085V
REFERENCE
5V
REGULATOR
PWM
LOGIC
Output
Voltage
Select
FAULT
DETECTION
C/10
DETECTION
SHUTDOWN
CONTROL
OSCILLATOR
BAT
FVS0
FVS1
SHUTDOWN
CHARGE
INPUT
CURRENT
LIMIT
TGATE
20
BOOST
12
ITH
13
PROG
14
CSN
15
CSP
18
SW
P
GND
(PADDLE)
4012 BD02
16
BGATE
17
INTVDD
TO
INTERNAL
CIRCIUTS
CHRG
7
10
9
11
5
ACP
6SHDN
ICL
8
CLP
CLN
2
1
INFET
3
DCIN
4
+
IF
VFB
LTC4012/
LTC4012-1/LTC4012-2
12
4012fa
operation
Overview
The LTC4012 is a synchronous step-down (buck) current
mode PWM battery charger controller. The maximum
charge current is programmed by the combination of a
charge current sense resistor (RSENSE), matched input
resistors (RIN, Figure 1), and a programming resis-
tor (RPROG) between the PROG and GND pins. Battery
voltage is programmed with an external resistor divider
between FBDIV and GND (LTC4012) or two digital battery
voltage select pins (LTC4012-1/LTC4012-2). In addition,
the PROG pin provides a linearized voltage output of the
actual charge current.
The LTC4012 family does not have built-in charge termina-
tion and is flexible enough for charging any type of battery
chemistry. These are building block ICs intended for use
with an external circuit, such as a microcontroller, capable
of managing the entire algorithm required for the specific
battery being charged. Each member of the LTC4012 fam-
ily features a shutdown input and various state indicator
outputs, allowing easy and direct management by a wide
range of external (digital) charge controllers. Due to the
popularity of rechargeable Lithium-Ion chemistries, the
LTC4012-1 and LTC4012-2 also offer internal precision
resistors that can be digitally selected to produce one of
four preset output voltages for simplified design of those
charger types.
Shutdown
The LTC4012 remains in shutdown until DCIN is greater
than 5.1V and exceeds CLP by 60mV and SHDN is driven
above 1.4V. In shutdown, current drain from the battery
is reduced to the lowest possible level, thereby increasing
standby time. When in shutdown, the ITH pin is pulled to
GND and CHRG, ICL, FET gate drivers and INTVDD output
are all disabled. The charging can be stopped at any time
by forcing SHDN below 300mV.
AC Present Indication
The ACP status output correctly indicates sensed adapter
input voltage during all LTC4012 states. AC present is
indicated (ACP output low) as soon as DCIN exceeds
BAT by at least 500mV. Charging is not enabled until this
condition is first met. After this event, charging is no longer
gated by AC present detection. If battery voltage rises due
to ESR, or DCIN droops due to current load, AC present
may no longer be indicated by the IC if charging was
started with very low input overhead. However, charging
will remain enabled unless DCIN falls below the supply
voltage on CLP.
Input PowerPath Control
The input PFET controller performs many important func-
tions. First, it monitors DCIN and enables the charger
when this input voltage is higher than the raw CLP sys-
tem supply. Next, it controls the gate of an external input
power PFET to maintain a low forward voltage drop when
charging, creating improved efficiency. It also prevents
reverse current flow through this same PFET, providing
a suitable input blocking function. Finally, it helps avoid
synchronous boost operation during invalid operating
conditions by detecting elevated CLP voltage and forcing
the charger off.
If DCIN voltage is less than CLP, then DCIN must rise
60mV higher than CLP to enable the charger and activate
the ideal diode control. At this point, the ACPb status
output also transitions to low impedance to indicate
to the host system that an external adapter is present.
The gate of the input PFET is driven to a voltage sufficient
to regulate a forward drop between DCIN and CLP of about
25mV. If the input voltage differential drops below this
point, the FET is turned off slowly. If the voltage between
DCIN and CLP drops to less than –25mV, the input FET is
turned off in less than 6µs to prevent significant reverse
current from flowing back through the PFET. In this case,
ACPb also switches back to high impedance and the
charger is disabled.
Soft-Start
Exiting the shutdown state enables the charger and releases
the ITH pin. When enabled, switching will not begin until
DCIN exceeds BAT by 500mV and ITH exceeds a threshold
that assures initial current will be positive (about 5% to
25% of the maximum programmed current). To limit inrush
current, soft-start delay is created with the compensation
values used on the ITH pin. Longer soft-start times can be
realized by increasing the filter capacitor on ITH, if reduced
loop bandwidth is acceptable. The actual charge current at
13
4012fa
LTC4012/
LTC4012-1/LTC4012-2
the end of soft-start will depend on which loop (current,
voltage or adapter limit) is in control of the PWM. If this
current is below that required by the ITH start-up threshold,
the resulting charge current transient duration depends on
loop compensation but is typically less than 100µs.
Bulk Charge
When soft-start is complete, the LTC4012 begins sourc-
ing the current programmed by the external components
connected to CSP, CSN and PROG. Some batteries may
require a small conditioning trickle current if they are heav-
ily discharged. As shown in the Applications Information
section, the LT4012 can address this need through a variety
of low current circuit techniques on the PROG pin. Once
a suitable cell voltage has been reached, charge current
can be switched to a higher, bulk charge value.
End of Charge and CHRG Output
As the battery approaches the programmed output volt-
age, charge current will begin to decrease. The open-drain
CHRG output can indicate when the current drops to 10%
of its programmed full-scale value by turning off the strong
pull-down (open-drain FET) and turning on a weak 25µA
pull-down current. This weak pull-down state is latched
until the part enters shutdown or the sensed current rises
to roughly C/6. C/10 indication will not be set if charge
current has been reduced due to adapter input current
limiting. As the charge current approaches 0A, the PWM
continues to operate in full continuous mode. This avoids
generation of audible noise, allowing bulk ceramic capaci-
tors to be used in the application.
Charge Current Monitoring
When the LTC4012 is charging, the voltage on the PROG pin
varies in direct proportion to the charge current. Referring
to Figure 1, the nominal PROG voltage is given by
VI R R
RµA R
PROG CHRG SENSE PROG
IN PROG
= +
. 11 67
Voltage tolerance on PROG is limited by the charge current
accuracy specified in the Electrical Characteristics table.
Refer to the Applications Information section on program-
ming charge current for additional details.
operation
Figure 1. PWM Circuit Diagram
+
+
EA
CA
+
CC
19
2
11 PWM
LOGIC
OSCILLATOR
WATCHDOG
TIMER
LOOP
COMPENSATION
TGATE
SYSTEM
POWER
L1
RIN +
16
BGATE
15
CSP
14
CSN
13
PROG
9
VFB
12
ITH
RPROG
R1
FROM ICL
1.2085V
RSENSE VSENSE
ICHRG
4012 F01
CPROG
QS
CLOCK
LTC4012
CLP
BAT RD
RIN
+
LTC4012/
LTC4012-1/LTC4012-2
14
4012fa
Adapter Input Current Limit
The LTC4012 can monitor and limit current from the input
DC supply, which is normally an AC adapter. When the
programmed adapter input current is reached, charge
current is reduced to maintain the desired maximum input
current. The ITH and PROG pins will reflect the reduced
charge current. This limit function avoids overloading the
DC input source, allowing the product to operate at the
same time the battery is charging without complex load
management algorithms. The battery will automatically be
charged at the maximum possible rate that the adapter will
support, given the application’s operating condition. The
LTC4012 can only limit input current by reducing charge
current, and in this case the charger uses nonsynchro-
nous PWM operation to prevent boosting if the average
charge current falls below about 25% of the maximum
programmed current. Note that the ICL indicator output
becomes active (low) at an adapter input current level just
slightly less than that required for the internal amplifier to
begin to assert control over the PWM loop.
Charger Status Indicator Outputs
The LTC4012 open-drain indicator outputs provide valu-
able information about the IC’s operating state and can
be used for a variety of purposes in applications. Table 1
summarizes the state of the three indicator outputs as a
function of LTC4012 operation.
operation
Table 1. LTC4012 Open-Drain Indicator Outputs
ACP CHRG ICL CHARGER STATE
Off Off Off No DC Input (Shutdown)
On Off Off Shutdown or Reverse Current
On On Off Bulk Charge
On 25µA Off Low Current Charge or Initial
DCIN – BAT <500mV
On On On Input Current Limit During Bulk
Charge
On 25µA On Input Current Limit During Low
Current Charge
Off On or
25µA
On or
Off
Indicated Charge with DCIN - BAT
< 300mV. Bulk charge may be less
than programmed value.
PWM Controller
The LTC4012 uses a synchronous step-down architec-
ture to produce high operating efficiency. The nominal
operating frequency of 550kHz allows use of small filter
components. The following conceptual discussion of basic
PWM operation references Figure 1.
The voltage across the external charge current sense
resistor RSENSE is measured by current amplifier CA. This
instantaneous current (VSENSE/RIN) is fed to the PROG pin
where it is averaged by an external capacitor and converted
to a voltage by the programming resistor RPROG between
PROG and GND. The PROG voltage becomes the aver-
age charge current input signal to error amplifier EA. EA
also receives loop control information from the battery
voltage feedback input VFB and the adapter input current
limit circuit.
Figure 2. PWM Waveforms
ON
OFF
OFF
INDUCTOR
CURRENT
TOP FET
BOTTOM FET
ON
tOFF
THRESHOLD
SET BY ITH
VOLTAGE
4012 F02
15
4012fa
LTC4012/
LTC4012-1/LTC4012-2
The ITH output of the error amplifier is a scaled control
voltage for one input of the PWM comparator CC. ITH
sets a peak inductor current threshold, sensed by R1, to
maintain the desired average current through RSENSE. The
current comparator output does this by switching the state
of the RS latch at the appropriate time.
At the beginning of each oscillator cycle, the PWM
clock sets the RS latch and turns on the external top-
side NFET (bottom-side synchronous NFET off) to
refresh the current carried by the external inductor L1.
The inductor current and voltage across RSENSE begin
to rise linearly. CA buffers this instantaneous voltage
rise and applies it to CC with gain supplied by R1.
When the voltage across R1 exceeds the peak level set by
the ITH output of EA, the top FET turns off and the bottom
FET turns on. The inductor current then ramps down lin-
early until the next rising PWM clock edge. This closes the
loop and sources the correct inductor current to maintain
the desired parameter (charge current, battery voltage,
or input current). To produce a near constant frequency,
the PWM oscillator implements the equation:
tCLP BAT
CLP kHz
OFF =
550
Repetitive, closed-loop waveforms for stable PWM opera-
tion appear in Figure 2.
PWM Watchdog Timer
As input and output conditions vary, the LTC4012 may need
to utilize PWM duty cycles approaching 100%. In this case,
operating frequency may be reduced well below 550kHz.
An internal watchdog timer observes the activity on the
TGATE pin. If TGATE is on for more than 40µs, the watchdog
activates and forces the bottom NFET on (top NFET off)
for about 100ns. This avoids a potential source of audible
noise when using ceramic input or output capacitors and
prevents the boost supply capacitor for the top gate driver
from discharging. In low drop out operation, the actual
charge current may not be able to reach the programmed
full-scale value due to the watchdog function.
Overvoltage Protection
The LTC4012 also contains overvoltage detection that
prevents transient battery voltage overshoots of more than
about 6% above the programmed output voltage. When
battery overvoltage is detected, both external MOSFETs are
turned off until the overvoltage condition clears, at which
time a new soft start sequence begins. This is useful for
properly charging battery packs that use an internal switch
to disconnect themselves for performing functions such
as calibration or pulse mode charging.
Reverse Charge Current Protection (Anti-Boost)
Because the LTC4012 always attempts to operate syn-
chronously in full continuous mode (to avoid audible
noise from ceramic capacitors), reverse average charge
current can occur during some invalid operating condi-
tions. INFET PowerPath control avoids boosting a lightly
loaded system supply during reverse operation. However,
under heavier system loads, CLP may not boost above
DCIN, even though reverse average current is flowing. In
this case a second circuit monitors indication of reverse
average current on PROG.
If either of these circuits detects boost operation, The
LTC4012 turns off both external MOSFETs until the reverse
current condition clears. At that point, a new soft-start
sequence begins.
operation
LTC4012/
LTC4012-1/LTC4012-2
16
4012fa
applications inForMation
Programming Charge Current
The formula for charge current is:
IR
R
V
RµA
CHRG IN
SENSE PROG
=
. .
1 2085 11 67
The LTC4012 operates best with 3.01k input resistors,
although other resistors near this value can be used to
accommodate standard sense resistor values. Refer to
the subsequent discussion on inductor selection for other
considerations that come into play when selecting input
resistors RIN.
RSENSE should be chosen according to the following
equation:
RmV
I
SENSE MAX
=100
where IMAX is the desired maximum charge current ICHRG.
The 100mV target can be adjusted to some degree to obtain
standard RSENSE values and/or a desired RPROG value, but
target voltages lower than 100mV will cause a proportional
reduction in current regulation accuracy.
The required minimum resistance between PROG and GND
can be determined by applying the suggested expression
for RSENSE while solving the first equation given above for
charge current with ICHRG = IMAX:
RV R
V µA R
PROG MIN IN
IN
( )
.
. .
=+
1 2085
0 1 11 67
If RIN is chosen to be 3.01k with a sense voltage of 100mV,
this equation indicates a minimum value for RPROG of
26.9k. Table 6 gives some examples of recommended
charge current programming component values based
on these equations.
The resistance between PROG and GND can simply be
set with a single a resistor, if only maximum charge cur-
rent needs to be controlled during the desired charging
algorithm.
However, some batteries require a low charge cur-
rent for initial conditioning when they are heav-
ily discharged. The charge current can then be safely
switched to a higher level after conditioning is complete.
Figure 3 illustrates one method of doing this with 2-level
control of the PROG pin resistance. Turning Q1 off reduces
the charge current to IMAX/10 for battery conditioning.
When Q1 is on, the LTC4012 is programmed to allow
full IMAX current for bulk charge. This technique can be
expanded through the use of additional digital control
inputs for an arbitrary number of pre-programmed cur-
rent values.
For a truly continuous range of maximum charge current
control, pulse width modulation can be used as shown
in Figure 4.
Figure 3. Programming 2-Level Charge Current
13
Q1
2N7002
4012 F03
R2
53.6k
PROG
LTC4012
R1
26.7k CPROG
4.7nF
BULK
CHARGE
PRECHARGE
Figure 4. Programming PWM Current
13
Q1
2N7002
4012 F04
PROG
LTC4012
RPROG RMAX
511k
CPROG
0V
5V
17
4012fa
LTC4012/
LTC4012-1/LTC4012-2
applications inForMation
The value of RPROG controls the maximum value of charge
current which can be programmed (Q1 continuously on).
PWM of the Q1 gate voltage changes the value of RPROG
to produce lower currents. The frequency of this modula-
tion should be higher than a few kHz, and CPROG must be
increased to reduce the ripple caused by switching Q1. In
addition, it may be necessary to increase loop compensa-
tion capacitance connected to ITH to maintain stability or
prevent large current overshoot during start-up. Selecting
a higher Q1 PWM frequency (≈10kHz) will reduce the need
to change CPROG or other compensation values. Charge
current will be proportional to the duty cycle of the PWM
input on the gate of Q1.
Programming LTC4012 Output Voltage
Figure 5 shows the external circuit for programming the
charger voltage when using the LTC4012. The voltage is
then governed by the following equation:
VV R R
RR R A R B
BAT =+
( )
= +
1 2085 1 2
22 2 2
. ,
See Table 2 for approximate resistor values for R2.
R R VR R A R B1 2 1 2085 1 2 2 2=
= +
VBAT
. ,
Selecting R2 to be less than 50k and the sum of R1 and
R2 at least 200k or above, achieves the lowest possible
error at the VFB sense input. Note that sources of error
such as R1 and R2 tolerance, FBDIV RON or VFB input
impedance are not included in the specifications given in
the Electrical Characteristics. This leads to the possibil-
ity that very accurate (0.1%) external resistors might be
required. Actually, the temperature rise of the LTC4012 will
rarely exceed 50°C at the end of charge, because charge
current will have tapered to a low level. This means that
0.25% resistors will normally provide the required level of
overall accuracy. Table 2 gives recommended values for
R1 and R2 for popular lithium-ion battery voltages. For
values of R1 above 200k, addition of capacitor CZ may
improve transient response and loop stability. A value of
10pF is normally adequate.
Table 2. Programming Output Voltage
VBAT
(V)
R1 (0.25%)
(kΩ)
R2A (0.25%)
(kΩ)
R2B (1%)*
(Ω)
4.1 165 69
4.2 167 67.3 200
8.2 162 28
8.4 169 28.4
12.3 301 32.8
12.6 294 31.2
16.4 284 22.6
16.8 271 21
20.5 316 19.8
21 298 18.2
24.6 298 15.4
25.2 397 20
*To Obtain Desired Accuracy Requires Series Resistors For R2.
Figure 5. Programming Output Voltage
11
10
BAT
FBDIV
85Ω
TYPICAL
9
VFB
LTC4012 R1
R2A
R2B*
4012 F05
CZ
21
GND
(EXPOSED PAD)
*OPTIONAL TRIM RESISTOR
+
LTC4012/
LTC4012-1/LTC4012-2
18
4012fa
applications inForMation
Programming LTC4012-1/LTC4012-2 Output Voltage
The LTC4012-1/LTC4012-2 feature precision internal bat-
tery voltage feedback resistor taps configured for common
lithium-ion voltages. All that is required to program the
desired voltage is proper pin programming of FVS0 and
FVS1 as shown in Table 3.
Table 3. LTC4012-1/LTC4012-2 Output Voltage Programming
VBAT VOLTAGE
LTC4012-1 LTC4012-2 FVS1 FVS0
4.1V 4.2V GND GND
8.2V 8.4V GND INTVDD
12.3V 12.6V INTVDD GND
16.4V 16.8V INTVDD INTVDD
Programming Input Current Limit
To set the input current limit ILIM, the minimum wall
adapter current rating must be known. To account for the
tolerance of the LTC4012 input current sense circuit, 5%
should be subtracted from the adapters minimum rated
output. Refer to Figure 6 and program the input current
limit function with the following equation.
RmV
I
CL LIM
=100
where ILIM is the desired maximum current draw from
the DC (adapter) input, including adjustments for tolerance,
if any.
Figure 6. Programming Input Current Limit
21
RCL
CDC CF 0.1µF
CLP
LTC4012
CLN
RF
5.1k
4012 F06
10k
FROM DC
POWER
INPUT
TO
REMAINDER
OF SYSTEM
3INFET
2
1
CLP
CLN
17
INTVDD
LTC4012 RCL
1%
R3 = R1
1%
R1
1%
Q2
2SC2412
RF
2.49k 1%
R2
Q1
IMX1
4012 F07
CF
0.22µF
TO REMAINDER
OF SYSTEM
FROM INFET
Figure 7. Adjusting Input Current Limit
Often an AC adapter will include a rated current output
margin of at least +10%. This can allow the adapter cur-
rent limit value to simply be programmed to the actual
minimum rated adapter output current. Table 4 shows
some common RCL current limit programming values.
A lowpass filter formed by RF (5.1k) and CF (0.1µF) is
required to eliminate switching noise from the LTC4012
PWM and other system components. If input current limit-
ing is not desired, CLN should be shorted to CLP while
CLP remains connected to power.
Table 4. Common RCL Values
ADAPTER
RATING
(A)
RCL VALUE (1%)
(Ω)
RCL POWER
DISSIPATION
(W)
RCL POWER
RATING
(W)
1.00 0.100 0.100 0.25
1.25 0.080 0.125 0.25
1.50 0.068 0.150 0.25
1.75 0.056 0.175 0.25
2.00 0.050 0.200 0.25
2.50 0.039 0.244 0.50
3.00 0.033 0.297 0.50
3.50 0.027 0.331 0.50
4.00 0.025 0.400 0.50
Figure 7 shows an optional circuit that can influence
the parameters of the input current limit in two ways.
19
4012fa
LTC4012/
LTC4012-1/LTC4012-2
applications inForMation
The first option is to lower the power dissipation of RCL at
the expense of accuracy without changing the input current
limit value. The second is to make the input current limit
value programmable.
The overall accuracy of this circuit needs to be better
than the power source current tolerance or be margined
such that the worse-case error remains under the power
source limits.
The accuracy of the Figure 7 circuit is a function of the
INTVDD, VBE, RCL, RF
, R1 and R3 tolerances. To improve
accuracy, the tolerance of RF should be changed from
5.1k, 5% to a 2.49k 1% resistor. RCL and the programming
resistors R1 and R3 should also be 1% tolerance such
that the dominant error is INTVDD (±3%). Bias resistor R2
can be 5%. When choosing NPN transistors, both need
to have good gain (>100) at 10µA levels. Low gain NPNs
will increase programming errors. Q1 must be a matched
NPN pair. Since RF has been reduced in value by half, the
capacitor value of CF should double to 0.22µF to remain
effective at filtering out any noise.
If you wish to reduce RCL power dissipation for a given
current limit, the programming equation becomes:
R
mV k
R
I
CL LIM
=
100 5 2 49
1
.
If you wish to make the input current limit programmable,
the equation becomes:
I
mV k
R
R
LIM CL
=
100 5 2 49
1
.
The equation governing R2 for both applications is based
on the value of R1. R3 should always be equal to R1.
R2 = 0.875 • R1
Figure 8. PROG Voltage Buffer
17
13
INTVDD
PROG
<30nA
LTC4012
4012 F08
TO SYSTEM
MONITOR
+
In many notebook applications, there are situations
where two different ILIM values are needed to allow two
different power adapters or power sources to be used.
In such cases, start by setting RLIM for the high power
ILIM configuration and then use Figure 7 to set the lower
ILIM value. To toggle between the two ILIM values, take
the three ground connections shown in Figure 7, combine
them into one common connection and use a small-signal
NFET (2N7002) to open or close that common connec-
tion to circuit ground. When the NFET is off, the circuit
is defeated (floating) allowing ILIM to be the maximum
value. When the NFET is on, the circuit will become active
and ILIM will drop to the lower set value.
Monitoring Charge Current
The PROG pin voltage can be used to indicate charge cur-
rent where 1.2085V indicates full programmed current (1C)
and zero charge current is approximately equal to RPROG
11.67µA. PROG voltage varies in direct proportion to the
charge current between this zero-current (offset) value and
1.2085V. When monitoring the PROG pin voltage, using a
buffer amplifier as shown in Figure 8 will minimize charge
current errors. The buffer amplifier may be powered from
the INTVDD pin or any supply that is always on when the
charger is on.
LTC4012/
LTC4012-1/LTC4012-2
20
4012fa
applications inForMation
Table 5. Digital Read Back State (IN, Figure 10)
LTC4012
CHARGER STATE
OUT STATE
Hi-Z 1
Off 1 1
C/10 Charge 0 1
Bulk Charge 0 0
Input and Output Capacitors
In addition to typical input supply bypassing (0.1µF) on
DCIN, the relatively high ESR of aluminum electrolytic ca-
pacitors is helpful for reducing ringing when hot-plugging
the charger to the AC adapter. Refer to LTC Application
Note 88 for more information.
The input capacitor between system power (drain of top
FET, Figure 1) and GND is required to absorb all input PWM
ripple current, therefore it must have adequate ripple current
rating. Maximum RMS ripple current is typically one-half
of the average battery charge current. Actual capacitance
value is not critical, but using the highest possible voltage
rating on PWM input capacitors will minimize problems.
Consult with the manufacturer before use.
Figure 10. Microprocessor Status Interface
33k
200k
4012 F10
VDD
3.3V
µP
IN
OUT
LTC4012
CHRG 7
C/10 CHRG Indicator
The value chosen for RPROG has a strong influence
on charge current monitoring and the accuracy of the
C/10 charge indicator output (CHRG). The LTC4012
uses the voltage on the PROG pin to determine when
charge current has dropped to the C/10 threshold.
The nominal threshold of 400mV produces an accu-
rate low charge current indication of C/10 as long as
RPROG = 26.7k, independent of all other current pro-
gramming considerations. However, it may sometimes
be necessary to deviate from this value to satisfy other
application design goals.
If RPROG is greater than 26.7k, the actual level at which
low charge current is detected will be less than C/10. The
highest value of RPROG that can be used while reliably
indicating low charge current before reaching final VBAT
is 30.1k. RPROG can safely be set to values higher than
this, but low current indication will be lost.
If RPROG is less than 26.7k, low charge current detection
occurs at a level higher than C/10. More importantly, the
LTC4012 becomes increasingly sensitive to reverse cur-
rent. The lowest value of RPROG that can be used without
the risk of erroneous boost operation detection at end of
charge is 26.1k. Values of RPROG less than this should not
be used. See the Operation section for more information
about reverse current.
The nominal fractional value of IMAX at which C/10 indica-
tion occurs is given by:
I
I
mV R µA
V R
C
MAX
PROG
PROG
10 400 11 67
1 2085
=
( )
.
. .11 67µA
( )
Direct digital monitoring of C/10 indication is possible
with an external application circuit like the one shown in
Figure 9.
By using two different value pull-up resistors, a micropro-
cessor can detect three states from this pin (charging, C/10
and not charging). See Figure 10. When a digital output port
(OUT) from the microprocessor drives one of the resistors
and a second digital input port polls the network, the charge
state can be determined as shown in Table 5.
Figure 9. Digital C/10 Indicator
17
7
INTVDD
CHRG
Q1
TP0610T
Q2
2N7002
100k
LTC4012
4012 F09
100k
VLOGIC
100k
C/10
CHRG
100k
Q3
2N7002
21
4012fa
LTC4012/
LTC4012-1/LTC4012-2
The output capacitor shown across the battery and ground
must also absorb PWM output ripple current. The general
formula for this capacitor current is:
I
VV
V
L f
RMS
BAT BAT
CLP
PWM
=
0 29 1
1
.
For example, IRMS = 0.22A with:
VBAT = 12.6V
VCLP = 19V
L1 = 10µH
fPWM = 550kHz
High capacity ceramic capacitors (20µF or more) available
from a variety of manufacturers can be used for input/out-
put capacitors. Other alternatives include OS-CON and
POSCAP capacitors from Sanyo.
Low ESR solid tantalum capacitors have high ripple cur-
rent rating in a relatively small surface mount package,
but exercise caution when using tantalum for input or
output bulk capacitors. High input surge current can be
created when the adapter is hot-plugged to the charger
or when a battery is connected to the charger. Solid tan-
talum capacitors have a known failure mechanism when
subjected to very high surge currents. Select tantalum
capacitors that have high surge current ratings or have
been surge tested.
EMI considerations usually make it desirable to minimize
ripple current in battery leads. Adding Ferrite beads or
inductors can increase battery impedance at the nominal
550KHz switching frequency. Switching ripple current splits
between the battery and the output capacitor in inverse
relation to capacitor ESR and the battery impedance. If
the ESR of the output capacitor is 0.2Ω and the battery
impedance is raised to 4Ω with a ferrite bead, only 5%
of the current ripple will flow to the battery.
Inductor Selection
Higher switching frequency generally results in
lower efficiency because of MOSFET gate charge
losses, but it allows smaller inductor and capacitor
values to be used. A primary effect of the inductor
value L1 is the amplitude of ripple current created.
The inductor ripple current ∆IL decreases with higher
inductance and PWM operating frequency:
I
VV
V
L f
L
BAT BAT
CLP
PWM
=
1
1
Accepting larger values of ∆IL allows the use of low in-
ductance, but results in higher output voltage ripple and
greater core losses. Lower charge currents generally call
for larger inductor values.
The LTC4012 limits maximum instantaneous peak inductor
current during every PWM cycle. To avoid unstable switch
waveforms, the ripple current must satisfy:
ImV
RI
L
SENSE
MAX
<
2150
so choose:
LV
fmV
RI
CLP
PWM SENSE MAX
10 125
150
>
.
For C-grade parts, a reasonable starting point for setting
ripple current is ∆IL = 0.4 IMAX. For I-grade parts, use
∆IL = 0.2 IMAX only if the IC will actually be used to
charge batteries over the wider I-grade temperature range.
The voltage compliance of internal LTC4012 circuits also
imposes limits on ripple current. Select RIN (in Figure 1)
to avoid average current errors in high ripple designs. The
following equation can be used for guidance:
R I
µA RR I
µA
SENSE L IN SENSE L
50 20
applications inForMation
LTC4012/
LTC4012-1/LTC4012-2
22
4012fa
RIN should not be less than 2.37k or more than 6.04k. Val-
ues of RIN greater than 3.01k may cause some reduction in
programmed current accuracy. Use these equations and
guidelines, as represented in Table 6, to help select the cor-
rect inductor value. This table was developed for C-grade
parts to maintain maximum ∆IL near 0.6 IMAX with fPWM at
550kHz and VBAT = 0.5 VCLP (the point of maximum ∆IL),
assuming that inductor value could also vary by 25% at
IMAX. For I-grade parts, reduce maximum ∆IL to less than
0.4 IMAX, but only if the IC will actually be used to charge
batteries over the wider I-grade temperature range. In that
case, a good starting point can be found by multiplying
the inductor values shown in Table 6 by a factor of 1.6 and
rounding up to the nearest standard value.
Table 6. Minimum Typical Inductor Values
VCLP L1
(Typ)
IMAX RSENSE RIN RPROG
<10V ≥10µH 1A 100mΩ 3.01k 26.7k
10V to 20V ≥20µH 1A 100mΩ 3.01k 26.7k
>20V ≥28µH 1A 100mΩ 3.01k 26.7k
<10V ≥5.1µH 2A 50mΩ 3.01k 26.7k
10V to 20V ≥10µH 2A 50mΩ 3.01k 26.7k
>20V ≥14µH 2A 50mΩ 3.01k 26.7k
<10V ≥3.4µH 3A 33mΩ 3.01k 26.7k
10V to 20V ≥6.8µH 3A 33mΩ 3.01k 26.7k
>20V ≥9.5µH 3A 33mΩ 3.01k 26.7k
<10V ≥2.5µH 4A 25mΩ 3.01k 26.7k
10V to 20V ≥5.1µH 4A 25mΩ 3.01k 26.7k
>20V ≥7.1µH 4A 25mΩ 3.01k 26.7k
To guarantee that a chosen inductor is optimized in any
given application, use the design equations provided and
perform bench evaluation in the target application, par-
ticularly at duty cycles below 20% or above 80% where
PWM frequency can be much less than the nominal value
of 550kHz.
TGATE BOOST Supply
Use the external components shown in Figure 11 to de-
velop a bootstrapped BOOST supply for the TGATE FET
driver. A good set of equations governing selection of the
two capacitors is:
CQ
VC C
G
120
4 5 2 20 1= =
.,
where QG is the rated gate charge of the top external NFET
with VGS = 4.5V. The maximum average diode current is
then given by:
ID = QG • 665kHz
To improve efficiency by increasing VGS applied to the
top FET, substitute a Schottky diode with low reverse
leakage for D1.
PWM jitter has been observed in some designs operating
at higher VIN/VOUT ratios. This jitter does not substantially
affect DC charge current accuracy. A series resistor with a
value of 5Ω to 20Ω can be inserted between the cathode
of D1 and the BOOST pin to remove this jitter, if present.
A resistor case size of 0603 or larger is recommended to
lower ESL and achieve the best results.
applications inForMation
Figure 11. TGATE Boost Supply
20
17
BOOST
INTVDD
18
SW
LTC4012
4012 F11
C2
2µF
C1
0.1µF
L1 TO
RSENSE
D1
1N4148
23
4012fa
LTC4012/
LTC4012-1/LTC4012-2
applications inForMation
FET Selection
Two external power MOSFETs must be selected for use
with the charger: an N-channel power switch (top FET)
and an N-channel synchronous rectifier (bottom FET).
Peak gate-to-source drive levels are internally set to
about 5V. Consequently, logic-level FETs must be used.
In addition to the fundamental DC current, selection
criteria for these MOSFETs also include channel resis-
tance RDS(ON), total gate charge QG, reverse transfer
capacitance CRSS, maximum rated drain-source voltage
BVDSS and switching characteristics such as td(ON/OFF).
Power dissipation for each external FET is given by:
PV I T R
V
k V
D TOP BAT MAX DS ON
CLP
C
( ) ( )
=+
( )
+
21δ
LLP MAX RSS
D BOT CLP BAT M
I C kHz
PV V I
2665
( ) =
( )
AAX DS ON
CLP
T R
V
21 ( )
+
( )
δ
where δ is the temperature dependency of RDS(ON),
∆T is the temperature rise above the point specified in
the FET data sheet for RDS(ON) and k is a constant in-
versely related to the internal LTC4012 top gate driver.
The term (1 + δT) is generally given for a MOSFET in the
form of a normalized RDS(ON) curve versus temperature,
but δ of 0.005/°C can be used as a suitable approxima-
tion for logic-level FETs if other data is not available.
CRSS = ∆QGD/VDS is usually specified in the MOSFET
characteristics. The constant k = 2 can be used in estimat-
ing top FET dissipation. The LTC4012 is designed to work
best with external FET switches with a total gate charge
at 5V of 15nC or less.
For VCLP < 20V, high charge current efficiency generally
improves with larger FETs, while for VCLP > 20V, top gate
transition losses increase rapidly to the point that using
a topside NFET with higher RDS(ON) but lower CRSS can
actually provide higher efficiency. If the charger will be
operated with a duty cycle above 85%, overall efficiency
is normally improved by using a larger top FET.
The synchronous (bottom) FET losses are greatest at high
input voltage or during a short circuit, which forces a low
side duty cycle of nearly 100%. Increasing the size of this
FET lowers its losses but increases power dissipation in the
LTC4012. Using asymmetrical FETs will normally achieve
cost savings while allowing optimum efficiency.
Select FETs with BVDSS that exceeds the maximum VCLP
voltage that will occur. Both FETs are subjected to this level
of stress during operation. Many logic-level MOSFETs are
limited to 30V or less.
The LTC4012 uses an improved adaptive TGATE and
BGATE drive that is insensitive to MOSFET inertial delays,
td(ON/OFF), to avoid overlap conduction losses. Switching
characteristics from power MOSFET data sheets apply
only to a specific test fixture, so there is no substitute for
bench evaluation of external FETs in the target application.
In general, MOSFETs with lower inertial delays will yield
higher efficiency.
Diode Selection
A Schottky diode in parallel with the bottom FET and/or
top FET in an LTC4012 application clamps SW during the
non-overlap times between conduction of the top and
bottom FET switches. This prevents the body diode of the
MOSFETs from forward biasing and storing charge, which
could reduce efficiency as much as 1%. One or both diodes
can be omitted if the efficiency loss can be tolerated. A 1A
Schottky is generally a good size for 3A chargers due to the
low duty cycle of the non-overlap times. Larger diodes can
actually result in additional efficiency (transition) losses
due to larger junction capacitance.
Loop Compensation and Soft-Start
The three separate PWM control loops of the LTC4012
can be compensated by a single set of components at-
tached between the ITH pin and GND. As shown in the
typical LTC4012 application, a 6.04k resistor in series
with a capacitor of at least 0.1µF provides adequate loop
compensation for the majority of applications.
LTC4012/
LTC4012-1/LTC4012-2
24
4012fa
Figure 12. High Speed Switching Path
4012 F12
VBAT
L1 RSENSE
HIGH
FREQUENCY
CIRCULATING
PATH BAT
ANALOG
GROUND
SYSTEM
GROUND
SWITCH NODE
CIN
SWITCHING GROUND
COUT
VIN
GND
D1 +
The LTC4012 can be soft-started with the compensation
capacitor on the ITH pin. At start-up, ITH will quickly rise
to about 0.25V, then ramp up at a rate set by the com-
pensation capacitor and the 40µA ITH bias current. The
full programmed charge current will be reached when ITH
reaches approximately 2V. With a 0.1µF capacitor, the time
to reach full charge current is usually greater than 1.5ms.
This capacitor can be increased if longer start-up times
are required, but loop bandwidth and dynamic response
will be reduced.
INTVDD Regulator Output
Bypass the INTVDD regulator output to GND with a low
ESR X5R or X7R ceramic capacitor with a value of 0.47µF
or larger. The capacitor used to build the BOOST supply
(C2 in Figure 11) can serve as this bypass. Do not draw
more than 30mA from this regulator for the host system,
governed by IC power dissipation.
Calculating IC Power Dissipation
The user should ensure that the maximum rated junction
temperature is not exceeded under all operating conditions.
The thermal resistance of the LTC4012 package (θJA) is
37°C/W, provided the Exposed Pad is in good thermal
contact with the PCB. The actual thermal resistance in the
application will depend on forced air cooling and other heat
sinking means, especially the amount of copper on the PCB
to which the LTC4012 is attached. The following formula
may be used to estimate the maximum average power dis-
sipation PD (in watts) of the LTC4012, which is dependent
upon the gate charge of the external MOSFETs. This gate
charge, which is a function of both gate and drain voltage
swings, is determined from specifications or graphs in the
manufacturer’s data sheet. For the equation below, find the
gate charge for each transistor assuming 5V gate swing and
a drain voltage swing equal to the maximum VCLP voltage.
Maximum LTC4012 power dissipation under normal op-
erating conditions is then given by:
PD = DCIN(3mA + IDD + 665kHz(QTGATE + QBGATE))
– 5IDD
applications inForMation
where:
IDD = Average external INTVDD load current, if any
QTGATE = Gate charge of external top FET in Coulombs
QBGATE = Gate charge of external bottom FET in
Coulombs
PCB Layout Considerations
To prevent magnetic and electrical field radiation and
high frequency resonant problems, proper layout of the
components connected to the LTC4012 is essential. Refer
to Figure 12. For maximum efficiency, the switch node
rise and fall times should be minimized. The following
PCB design priority list will help insure proper topology.
Layout the PCB using this specific order.
1. Input capacitors should be placed as close as possible
to switching FET supply and ground connections with
the shortest copper traces possible. The switching
FETs must be on the same layer of copper as the input
capacitors. Vias should not be used to make these
connections.
2. Place the LTC4012 close to the switching FET gate
terminals, keeping the connecting traces short to
produce clean drive signals. This rule also applies to IC
supply and ground pins that connect to the switching
FET source pins. The IC can be placed on the opposite
side of the PCB from the switching FETs.
25
4012fa
LTC4012/
LTC4012-1/LTC4012-2
3. Place the inductor input as close as possible to the
switching FETs. Minimize the surface area of the switch
node. Make the trace width the minimum needed to
support the programmed charge current. Use no cop-
per fills or pours. Avoid running the connection on
multiple copper layers in parallel. Minimize capacitance
from the switch node to any other trace or plane.
4. Place the charge current sense resistor immediately
adjacent to the inductor output, and orient it such
that current sense traces to the LTC4012 are not long.
These feedback traces need to be run together as a
single pair with the smallest spacing possible on any
given layer on which they are routed. Locate any filter
component on these traces next to the LTC4012, and
not at the sense resistor location.
5. Place output capacitors adjacent to the sense resistor
output and ground.
6. Output capacitor ground connections must feed into
the same copper that connects to the input capacitor
ground before connecting back to system ground.
7. Connection of switching ground to system ground,
or any internal ground plane, should be single-point.
If the system has an internal system ground plane,
a good way to do this is to cluster vias into a single
star point to make the connection.
8. Route analog ground as a trace tied back to the LTC4012
GND paddle before connecting to any other ground.
Avoid using the system ground plane. A useful CAD
technique is to make analog ground a separate ground
net and use a 0Ω resistor to connect analog ground
to system ground.
9. A good rule of thumb for via count in a given high
current path is to use 0.5A per via. Be consistent when
applying this rule.
10. If possible, place all the parts listed above on the same
PCB layer.
11. Copper fills or pours are good for all power connections
except as noted above in Rule 3. Copper planes on
multiple layers can also be used in parallel. This helps
with thermal management and lowers trace inductance,
which further improves EMI performance.
12. For best current programming accuracy, provide a
Kelvin connection from RSENSE to CSP and CSN. See
Figure 13 for an example.
13. It is important to minimize parasitic capacitance on
the CSP and CSN pins. The traces connecting these
pins to their respective resistors should be as short
as possible.
applications inForMation
Figure 13. Kelvin Sensing of Charge Current
TO CSP
RIN
4012 F13
DIRECTION OF CHARGING CURRENT
RSENSE
TO CSN
RIN
LTC4012/
LTC4012-1/LTC4012-2
26
4012fa
package Description
4.00 ± 0.10
4.00 ± 0.10
NOTE:
1. DRAWING IS PROPOSED TO BE MADE A JEDEC PACKAGE OUTLINE MO-220
VARIATION (WGGD-1)—TO BE APPROVED
2. DRAWING NOT TO SCALE
3. ALL DIMENSIONS ARE IN MILLIMETERS
4. DIMENSIONS OF EXPOSED PAD ON BOTTOM OF PACKAGE DO NOT INCLUDE
MOLD FLASH. MOLD FLASH, IF PRESENT, SHALL NOT EXCEED 0.15mm ON ANY SIDE
5. EXPOSED PAD SHALL BE SOLDER PLATED
6. SHADED AREA IS ONLY A REFERENCE FOR PIN 1 LOCATION
ON THE TOP AND BOTTOM OF PACKAGE
PIN 1
TOP MARK
(NOTE 6)
0.40 ± 0.10
2019
1
2
BOTTOM VIEW—EXPOSED PAD
2.00 REF
2.45 ± 0.10
0.75 ± 0.05 R = 0.115
TYP
R = 0.05
TYP
0.25 ± 0.05
0.50 BSC
0.200 REF
0.00 – 0.05
(UF20) QFN 01-07 REV A
RECOMMENDED SOLDER PAD PITCH AND DIMENSIONS
APPLY SOLDER MASK TO AREAS THAT ARE NOT SOLDERED
0.70 ±0.05
0.25 ±0.05
0.50 BSC
2.00 REF 2.45 ± 0.05
3.10 ± 0.05
4.50 ± 0.05
PACKAGE OUTLINE
PIN 1 NOTCH
R = 0.20 TYP
OR 0.35 s 45°
CHAMFER
2.45 ± 0.10
2.45 ± 0.05
UF Package
20-Lead Plastic QFN (4mm × 4mm)
(Reference LTC DWG # 05-08-1710 Rev A)
27
4012fa
LTC4012/
LTC4012-1/LTC4012-2
Information furnished by Linear Technology Corporation is believed to be accurate and reliable.
However, no responsibility is assumed for its use. Linear Technology Corporation makes no representa-
tion that the interconnection of its circuits as described herein will not infringe on existing patent rights.
revision history
REV DATE DESCRIPTION PAGE NUMBER
A 3/10 I-Grade Parts Added. Reflected Throughout the Data Sheet 1 to 28
LTC4012/
LTC4012-1/LTC4012-2
28
4012fa
Linear Technology Corporation
1630 McCarthy Blvd., Milpitas, CA 95035-7417
(408) 432-1900 FAX: (408) 434-0507 www.linear.com
LINEAR TECHNOLOGY CORPORATION 2009
LT 0610 • PRINTED IN USA
relateD parts
PART NUMBER DESCRIPTION COMMENTS
LTC4006 Small, High Efficiency, Fixed Voltage, Lithium-Ion
Battery Chargers with Termination
Complete Charger for 3- or 4-Cell Li-Ion Batteries, AC Adapter Current
Limit and Thermistor Sensor, 16-pin SSOP Package
LTC4007 High Efficiency, Programmable Voltage, Lithium-Ion
Battery Charger with Termination
Complete Charger for 3- or 4-Cell Li-Ion Batteries, AC Adapter Current
Limit, Thermistor Sensor and Indicator Outputs
LTC4008/LTC4008-1 High Efficiency, Programmable Voltage/Current Battery
Chargers
Constant-Current/Constant-Voltage Switching Regulator, Resistor
Voltage/Current Programming, Thermistor Sensor and Indicator
Outputs, AC Adapter Current Limit (Omitted on 4008-1)
LTC4009/LTC4009-1
LTC4009-2
High Efficiency, Multichemistry Battery Charger Constant-Current/Constant-Voltage Switching Regulator in a 20-Lead
QFN Package, AC Adapter Current Limit, Indicator Outputs
LTC4411 2.6A Low Loss Ideal Diode No External MOSFET, Automatic Switching Between DC sources, 140mΩ
On Resistance in ThinSOT
TM
package
LTC4412/LTC4412HV Low Loss PowerPath Controllers Very Low Loss Replacement for Power Supply ORing Diodes Using
Minimal External Complements, Operates up to 28V (36V for HV)
LTC4413 Dual 2.6A, 2.5V to 5.5V Ideal Diodes Low Loss Replacement for ORing Diodes, 100mΩ On Resistance
LTC4414 36V, Low Loss PowerPath Controller for Large PFETs Low Loss Replacement for ORing Diodes, Operates up to 36V
LTC4416 Dual Low Loss PowerPath Controllers Low Loss Replacement for ORing Diodes, Operates up to 36V, Drives
Large PFETs, Programmable, Autonomous Switching
typical application
12.6V 4 Amp Charger
CLP
FROM
ADAPTER
15V AT 4A
BULK
CHARGE
C4
0.1µF R8
5.1k R14
100k
Q5
R15
0Ω*
D1 7
4 2
3
1
20
19
18 D3
17
16
21
15
14
11
10
Q1
9
R12
294k
C10
10pF
5
8
12
13
6
R7
25mΩ
R9 3.01k
C5
0.1µF
C6
2µF
Q2
Q3
D4
L1
4.7µH
R11
25mΩ
12.6V
Li-Ion
BATTERY
DCIN
CHRG
C2
0.1µF
R1
3k
R
R4
6.04k
R5
26.7k
R6
53.6k
LTC4012
ACP
ICL
SHDN
ITH
PROG
CLN
BOOST
INFET
TGATE
SW
INTVDD
BGATE
TO/FROM
MCU
GND
CSP
CSN
BAT
FBDIV
VFB
C8
10µF
POWER TO SYSTEM
TO POWER SYSTEM LOAD
WHEN ADAPTER IS NOT
PRESENT, USE
SCHOTTKY DIODE D5 OR
THE COMBINATION OF R14,
D6 AND Q4
D6
18V
ZENER
C1
0.1µF
C3
4.7nF
C9
10µF
R10 3.01k
R13
31.2k
4012 TA03
D3: CMDSH-3
D4: MBR230LSFT1
Q1: 2N7002
Q2, Q3: Si7212DN OR SiA914DJ
OR Si4816BDY (OMIT D4)
Q4, Q5: Si7423DN
L1: 1HLP-2525CZER4R7M11
*: SEE TGATE BOOST SUPPLY IN
APPLICATIONS INFORMATION
+
D5
Q4
OR