The MAX1778/MAX1880–MAX1885 multiple-output
DC-DC converters provide the regulated voltages
required by active matrix thin-film transistor (TFT) liquid
crystal displays (LCD) in a low-profile TSSOP package.
One high-power step-up converter and two low-power
charge pumps convert the 2.7V to 5.5V input voltage
into three independent output voltages. A built-in linear
regulator and VCOM buffer complete the power-supply
requirements.
The main step-up converter accurately generates an
externally set output voltage up to 13V that can supply
the display’s row/column drivers. The converter’s high
switching frequency and current-mode PWM architecture provide fast transient response and allow the use
of small low-profile inductors and ceramic capacitors.
The low-power BiCMOS control circuitry and internal
14V switch (0.35Ω N-channel MOSFET) enable efficiencies up to 91%.
The dual low-power charge pumps (MAX1778/
MAX1880/MAX1881/MAX1882 only) independently regulate one positive output (V
POS
) and one negative out-
put (V
NEG
). These low-power outputs use external
diode and capacitor stages (as many stages as
required) to regulate output voltages up to +40V and
-40V. A unique control scheme minimizes output ripple
as well as capacitor sizes for both charge pumps.
A resistor-programmable, 40mA, low-dropout linear
regulator (MAX1778/MAX1881/MAX1883/MAX1884
only) provides preregulation or postregulation for any of
the supplies. For higher current applications, an external transistor can be added. Additionally, the VCOM
buffer provides a high current output that is ideal for
driving the capacitive backplane of TFT LCD panels.
The VCOM buffer’s output voltage is preset with an
internal 50% resistive-divider or can be externally
adjusted for other voltages.
The MAX1778/MAX1880–MAX1885 are protected
against output undervoltage and thermal overload conditions by a latched fault detection circuit that shuts
down the device. All devices are available in the ultrathin TSSOP package (1.1mm max height).
Applications
TFT LCD Notebook Displays
TFT LCD Desktop Monitor Panels
Features
♦ 500kHz/1MHz Current-Mode PWM Step-Up
Regulator
Up to +13V Main High-Power Output
±1% Accurate
High Efficiency (91%)
♦ Dual Regulated Charge-Pump Outputs
(MAX1778/MAX1880/MAX1881/MAX1882 only)
Up to +40V Positive Charge-Pump Output
Up to -40V Negative Charge-Pump Output
♦ Low-Dropout 40mA Linear Regulator
(MAX1778/MAX1881/MAX1883/MAX1884 only)
Up to +15V LDO Input
♦ Optional Higher Current with External Transistor
Typical Operating Circuit appears at end of data sheet.
Pin Configurations and Selector Guide appear at end of
data sheet.
For pricing, delivery, and ordering information, please contact Maxim/Dallas Direct! at
1-888-629-4642, or visit Maxim’s website at www.maxim-ic.com.
PARTTEMP. RANGEPIN-PACKAGE
MAX1778EUG-40°C to +85°C24 TSSOP
MAX1880EUG-40°C to +85°C24 TSSOP
MAX1881EUG-40°C to +85°C24 TSSOP
MAX1882EUG-40°C to +85°C24 TSSOP
MAX1883EUP-40°C to +85°C20 TSSOP
MAX1884EUP-40°C to +85°C20 TSSOP
MAX1885EUP-40°C to +85°C20 TSSOP
= 1µF, TA= 0°C to +85°C. Typical values are at TA= +25°C, unless otherwise noted.)
Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent damage to the device. These are stress ratings only, and functional
operation of the device at these or any other conditions beyond those indicated in the operational sections of the specifications is not implied. Exposure to
absolute maximum rating conditions for extended periods may affect device reliability.
IN, SHDN, TGND, FLTSET to GND...........................-0.3V to +6V
DRVN to GND .........................................-0.3V to (V
SUPN
+ 0.3V)
DRVP to GND..........................................-0.3V to (V
SUPP
+ 0.3V)
PGND to GND.....................................................................±0.3V
RDY, SUPB to GND ................................................-0.3V to +14V
LX, SUPP, SUPN to PGND .....................................-0.3V to +14V
SUPL to GND..........................................................-0.3V to +18V
LDOOUT to GND ....................................-0.3V to (V
SUPL
+ 0.3V)
INTG, REF, FB, FBN, FBP to GND...............-0.3V to (V
IN
+ 0.3V)
FBL to GND .............-0.3V to the lower of (V
SUP
L
+ 0.3V) or +6V
BUFOUT, BUF+, BUF- to GND ...............-0.3V to (V
Main Step-Up Regulator Feedback Input. Regulates to 1.25V
nominal. Connect a resistive divider from the output (V
to analog ground (GND).
Main Step-Up Integrator Output. When using the integrator,
connect 1000pF to analog ground (GND). To disable the
integrator, connect INTG to REF.
Main Supply Voltage. The supply voltage powers the control
circuitry for all of the regulators and may range from 2.7V to 5.5V.
Bypass with a 0.1µF capacitor between IN and GND, as close to
the pins as possible.
VCOM Buffer (Operational Transconductance Amplifier) Positive
Feedback Input. Connect to GND to select the internal resistive
divider that sets the positive input to half the amplifier’s supply
voltage (V
The MAX1778/MAX1880–MAX1885 are highly efficient
multiple-output power supplies for thin-film transistor
(TFT) liquid crystal display (LCD) applications. The
devices contain one high-power step-up converter, two
low-power charge pumps, an operational transconductance amplifier (V
COM
buffer), and a low-dropout linear
regulator. The primary step-up converter uses an internal N-channel MOSFET to provide maximum efficiency
and to minimize the number of external components.
The output voltage of the main step-up converter
(V
MAIN
) can be set from VINto 13V with external resis-
tors.
The dual charge pumps (MAX1778/MAX1880/
MAX1881/MAX1882 only) independently regulate a
positive output (V
POS
) and a negative output (V
NEG
).
These low-power outputs use external diode and
capacitor stages (as many stages as required) to regulate output voltages from -40V to +40V. A unique
control scheme minimizes output ripple as well as
capacitor sizes for both charge pumps.
A resistor-programmable 40mA linear regulator
(MAX1778/MAX1881/MAX1883/MAX1884 only) can
provide preregulation or postregulation for any of the
supplies. For higher current applications, an external
transistor can be added.
Additionally, the V
COM
buffer provides a high current
output that is ideal for driving capacitive loads, such as
the backplane of a TFT LCD panel. The positive feedback input features dual mode operation, allowing this
input to be connected to an internal 50% resistivedivider between the buffer’s supply voltage and
ground, or externally adjusted for other voltages.
Also included in the MAX1778/MAX1880–MAX1885 is a
precision 1.25V reference that sources up to 50µA,
logic shutdown, soft-start, power-up sequencing,
adjustable fault detection, thermal shutdown, and an
active-low, open-drain ready output.
During normal pulse-width modulation (PWM) operation, the MAX1778/MAX1880–MAX1885 main step-up
controllers switch at a constant frequency of 500kHz or
1MHz (see Selector Guide), allowing the use of lowprofile inductors and output capacitors. Depending on
the input-to-output voltage ratio, the controller regulates
the output voltage and controls the power transfer by
modulating the duty cycle (D) of each switching cycle:
On the rising edge of the internal clock, the controller
sets a flip-flop when the output voltage is too low, which
turns on the N-channel MOSFET (Figure 2). The inductor current ramps up linearly, storing energy in a magnetic field. Once the sum of the feedback voltage error
amplifier, slope-compensation, and current-feedback
signals trip the multi-input comparator, the MOSFET
turns off, the flip-flop resets, and the diode (D1) turns
on. This forces the current through the inductor to ramp
back down, transferring the energy stored in the magnetic field to the output capacitor and load. The MOSFET remains off for the rest of the clock cycle. Changes
in the feedback voltage-error signal shift the switch-current trip level, consequently modulating the MOSFET
duty cycle.
Under very light loads, an inherent switchover to pulseskipping takes place (Figure 3). When this occurs, the
controller skips most of the oscillator pulses in order to
reduce the switching frequency and gate charge losses. When pulse-skipping, the step-up controller initiates
a new switching cycle only when the output voltage
drops too low. The N-channel MOSFET turns on, allowing the inductor current to ramp up until the multi-input
comparator trips. Then, the MOSFET turns off and the
diode turns on, forcing the inductor current to ramp
down. When the inductor current reaches zero, the
diode turns off, so the inductor stops conducting current. This forces the threshold between pulse-skipping
and PWM operation to coincide with the boundary
between continuous and discontinuous inductor-current operation:
The switching waveforms will appear noisy and asynchronous when light loading causes pulse-skipping
operation; this is a normal operating condition that
improves light-load efficiency.
Dual Charge-Pump Regulator (MAX1778/
MAX1880/MAX1881/MAX1882 Only)
The MAX1778/MAX1880/MAX1881/MAX1882 controllers contain two independent low-power charge
pumps (Figure 4). One charge pump inverts the input
voltage and provides a regulated negative output voltage. The second charge pump doubles the input voltage and provides a regulated positive output voltage.
The controllers contain internal P-channel and N-channel MOSFETs to control the power transfer. The
internal MOSFETs switch at a constant frequency
(fCHP = fOSC/2).
Positive Charge Pump
During the first half-cycle, the N-channel MOSFET turns
on and charges flying capacitor C
X(POS)
(Figure 4).
This initial charge is controlled by the variable N-channel on-resistance. During the second half-cycle, the Nchannel MOSFET turns off and the P-channel MOSFET
turns on, level shifting C
X(POS)
by V
SUPP
volts. This
connects C
X(POS)
in parallel with the reservoir capaci-
tor C
OUT(POS)
. If the voltage across C
OUT(POS)
plus a
diode drop (V
POS
+ V
DIODE
) is smaller than the level-
shifted flying capacitor voltage (V
CX(POS)
+ V
SUPP
),
charge flows from C
X(POS)
to C
OUT(POS)
until the diode
(D3) turns off.
Figure 3. Discontinuous-to-Continuous Conduction Crossover
Point
During the first half-cycle, the P-channel MOSFET turns
on, and flying capacitor C
X(NEG)
charges to V
SUPN
minus a diode drop (Figure 4). During the second halfcycle, the P-channel MOSFET turns off, and the Nchannel MOSFET turns on, level shifting C
X(NEG)
. This
connects C
X(NEG)
in parallel with reservoir capacitor
C
OUT(NEG)
. If the voltage across C
OUT(NEG)
minus a
diode drop is greater than the voltage across C
X(NEG)
,
charge flows from C
OUT(NEG)
to C
X(NEG)
until the diode
(D5) turns off. The amount of charge transferred to the
output is controlled by the variable N-channel on-resistance.
Low-Dropout Linear Regulator (MAX1778/
MAX1881/MAX1883/MAX1884 Only)
The MAX1778/MAX1881/MAX1883/MAX1884 contain a
low-dropout linear regulator (Figure 5) that uses an
internal PNP pass transistor (QP) to supply loads up to
40mA. As illustrated in Figure 5, the 1.25V reference is
connected to the error amplifier, which compares this
reference with the feedback voltage and amplifies the
difference. If the feedback voltage is higher than the
reference voltage, the controller lowers the base current of QP, which reduces the amount of current to the
output. If the feedback voltage is too low, the device
increases the pass transistor base current, which
allows more current to pass to the output and increases
the output voltage. However, the linear regulator also
includes an output current limit to protect the internal
pass transistor against short circuits.
The low-dropout linear regulator monitors and controls
the pass transistor’s base current, limiting the output
current to 130mA (typ). In conjunction with the thermal
overload protection, this current limit protects the output, allowing it to be shorted to ground for an indefinite
period of time without damaging the part.
VCOM Buffer
The MAX1778/MAX1880–MAX1885 include a VCOM
buffer, which uses an operational transconductance
amplifier (OTA) to provide a current output that is ideal
for driving capacitive loads, such as the backplane of a
TFT LCD panel. The unity-gain bandwidth of this current-output buffer is:
GBW = gm/C
OUT
where gm is the amplifier’s transconductance. The
bandwidth is inversely proportional to the output
capacitor, so large capacitive loads improve stability;
however, lower bandwidth decreases the buffer’s transient response time. To improve the transient response
Figure 5. Low-Dropout Linear Regulator Block Diagram
times, the amplifier’s transconductance increases as
the output current increases (see Typical OperatingCharacteristics).
The VCOM buffer’s positive feedback input features
dual mode operation. The buffer’s output voltage can
be internally set by a 50% resistive divider connected
to the buffer’s supply voltage (SUPB), or the output voltage can be externally adjusted for other voltages.
Shutdown (SHDN)
A logic-low level on SHDN shuts down all of the converters and the reference. When shut down, the supply
current drops to 0.1µA to maximize battery life, and the
reference is pulled to ground. The output capacitance,
feedback resistors, and load current determine the rate
at which each output voltage will decay. A logic-level
high on SHDN power activates the MAX1778/
MAX1880–MAX1885 (see Power-Up Sequencing). Do
not leave SHDN floating. If unused, connect SHDN to
IN. A logic-level transition on SHDN clears the fault
latch.
Power-Up Sequencing
Upon power-up or exiting shutdown, the MAX1778/
MAX1880–MAX1885 start a power-up sequence. First,
the reference powers up. Then, the main DC-DC stepup converter powers up with soft-start enabled. The linear regulator powers up at the same time as the main
step-up converter; however, the power sequence and
ready output signal are not affected by the regulation of
the linear regulator. While the main step-up converter
powers up, the output of the PWM comparator remains
low (Figure 2), and the step-up converter charges the
output capacitors, limited only by the maximum duty
cycle and current-limit comparator. When the step-up
converter approaches its nominal regulation value and
the PWM comparator’s output changes states for the
first time, the negative charge pump turns on. When the
negative output voltage reaches approximately 90% of
its nominal value (V
FBN
< 110mV), the positive charge
pump starts up. Finally, when the positive output voltage reaches 90% of its nominal value (V
FBP
> 1.125V),
the active-low ready signal (RDY) goes low (see PowerReady), and the VCOM buffer powers up. The
MAX1883/MAX1884/MAX1885 do not contain the
charge pumps, but the power-up sequence still contains the charge pumps’ startup logic, which appears
as a delay (2 ✕4096/fOSC) between the step-up converter reaching regulation and when the ready signal
and VCOM buffer are activated.
Soft-Start
For the main step-up regulator, soft-start allows a gradual increase of the current-limit level during startup to
reduce input surge currents. The MAX1778/MAX1880–
MAX1885 divide the soft-start period into four phases.
During the first phase, the controller limits the current
limit to only 0.38A (see Electrical Characteristics),
approximately a quarter of the maximum current limit
). If the output does not reach regulation within
1ms, soft-start enters phase II, and the current limit is
increased by another 25%. This process is repeated for
phase III. The maximum 1.5A (typ) current limit is
reached within 3072 clock cycles or when the output
reaches regulation, whichever occurs first (see the
startup waveforms in the Typical OperatingCharacteristics).
For the charge pumps (MAX1778/MAX1880/
MAX1881/MAX1882 only), soft-start is achieved by controlling the rate of rise of the output voltage. Both
charge-pump output voltages are controlled to be in
regulation within 4096 clock cycles, irregardless of output capacitance and load, limited only by the charge
pump’s output impedance. Although the MAX1883/
MAX1884/MAX1885 controllers do not include the
charge pumps, the soft-start logic still contains the
4096 clock cycle startup periods for both charge
pumps.
Fault Trip Level (FLTSET)
The MAX1778/MAX1880–MAX1885 feature dual mode
operation to allow operation with either a preset fault
trip level or an adjustable trip level for the step-up converter and positive charge-pump outputs. Connect FLTSET to GND to select the preset 0.9 ✕V
REF
fault
threshold. The fault trip level may also be adjusted by
connecting a voltage divider from REF to FLTSET
(Figure 8). For greatest accuracy, the total load on the
reference (including current through the negative
charge-pump feedback resistors) should not exceed
50µA so that VREF is guaranteed to be in regulation
(see Electrical Characteristics Table). Therefore, select
R10 in the 100kΩ to 1MΩ range, and calculate R9 with
the following equation:
R9 = R10 [(V
REF
/ V
FLTSET
) - 1]
where V
REF
= 1.25V, and V
FLTSET
may range from 0.67
x V
REF
to 0.85 x V
REF
. FLTSET’s input bias current has
a maximum value of 50nA. For 1% error, the current
through R10 should be at least 100 times the FLTSET
input bias current (I
FLTSET
).
Fault Condition
Once RDY is low, if the output of the main regulator or
either low-power charge pump falls below its fault
detection threshold, or if the input drops below its
undervoltage threshold, then RDY goes high impedance and all outputs shut down; however, the reference
remains active. After removing the fault condition, toggle shutdown (below 0.8V) or cycle the input voltage
(below 0.2V) to clear the fault latch and reactivate the
device.
The reference fault threshold is 1.05V. For the step-up
converter and positive charge-pump, the fault trip level is
set by FLTSET (see Fault Trip Level). For the negative
charge pump, the fault threshold measured at the
charge-pump’s feedback input (FBN) is 140mV (typ).
Power Ready (RDY)
Power ready is an open-drain output. When the powerup sequence for the main step-up converter and lowpower charge pumps has properly completed, the 14V
MOSFET turns on and pulls RDY low with a 125Ω (typ)
on-resistance. If a fault is detected on any of these
three outputs, the internal open-drain MOSFET appears
as a high impedance. Connect a 100kΩ pullup resistor
between RDY and IN for a logic-level output.
Voltage Reference (REF)
The voltage at REF is nominally 1.25V. The reference
can source up to 50µA with good load regulation (see
Typical Operating Characteristics). Connect a 0.22µF
ceramic bypass capacitor between REF and GND.
Thermal-Overload Protection
Thermal-overload protection limits total power dissipation in the MAX1778/MAX1880–MAX1885. When the
junction temperature exceeds T
J
= +160°C, a thermal
sensor activates the fault protection, which shuts down
the controller, allowing the IC to cool. Once the device
cools down by 15°C, toggle shutdown (below 0.8V) or
cycle the input voltage (below 0.2V) to clear the fault
latch and reactivate the controller. Thermal-overload
protection protects the controller in the event of fault
conditions. For continuous operation, do not exceed
the absolute maximum junction-temperature rating of
T
J
= +150°C.
Operating Region and Power Dissipation
The MAX1778/MAX1880–MAX1885s’ maximum power
dissipation depends on the thermal resistance of the IC
package and circuit board, the temperature difference
between the die junction and ambient air, and the rate
of any airflow. The power dissipated in the device
depends on the operating conditions of each regulator
and the buffer.
The step-up controller dissipates power across the
internal N-channel MOSFET as the controller ramps up
the inductor current. In continuous conduction, the
power dissipated internally can be approximated by:
Charge-Pump Input Power and Efficiency
Considerations), linear regulator, and VCOM buffer.
The linear regulator generates an output voltage by dissipating power across an internal pass transistor, so
the power dissipation is simply the load current times
the input-to-output voltage differential:
When driving an external transistor, the internal linear
regulator provides the base drive current. Depending
on the external transistor’s current gain (β) and the
maximum load current, the power dissipated by the
internal linear regulator may still be significant:
The charge pumps provide regulated output voltages
by dissipating power in the low-side N-channel MOSFET, so they could be modeled as linear regulators followed by unregulated charge pumps. Therefore, their
power dissipation is similar to a linear regulator:
where N is the number of charge-pump stages, V
DIODE
is the diodes’ forward voltage, and V
SUPD
is the posi-
tive charge-pump diode supply (Figure 4).
The VCOM buffer’s power dissipation depends on the
capacitive load (C
LOAD
) being driven, the peak-to-
peak voltage change (V
P-P
) across the load, and the
load’s switching rate:
To find the total power dissipated in the device, the
power dissipated by each regulator and the buffer must
be added together:
The maximum allowed power dissipation is 975mW (24pin TSSOP) / 879mW (20-pin TSSOP) or:
P
MAX
= (T
J(MAX
) - TA) / ( θJB+ θBA)
where T
J
- TAis the temperature difference between
the controller’s junction and the surrounding air, θJB(or
θJC) is the thermal resistance of the package to the
board, and θBAis the thermal resistance from the printed circuit board to the surrounding air.
Design Procedure
Main Step-Up Converter
Output Voltage Selection
Adjust the output voltage by connecting a voltagedivider from the output (VMAIN) to FB to GND (see
Typical Operating Circuit). Select R2 in the 10kΩ to
50kΩ range. Calculate R1 with the following equations:
R1 = R2 [(V
MAIN
/ V
REF
) - 1]
where V
REF
= 1.25V. V
MAIN
may range from VINto 13V.
Inductor Selection
Inductor selection depends upon the minimum required
inductance value, saturation rating, series resistance, and
size. These factors influence the converter’s efficiency,
maximum output load capability, transient response time,
and output voltage ripple. For most applications, values
between 4.7µH and 22µH work best with the controller’s
switching frequency (Tables 1 and 2).
The inductor value depends on the maximum output
load the application must support, input voltage, output
voltage, and switching frequency. With high inductor
values, the MAX1778/MAX1880–MAX1885 source higher output currents, have less output ripple, and enter
continuous conduction operation with lighter loads;
however, the circuit’s transient response time is slower.
On the other hand, low-value inductors respond faster
to transients, remain in discontinuous conduction operation, and typically offer smaller physical size for a
given series resistance and current rating. The equations provided here include a constant LIR, which is the
ratio of the peak-to-peak AC inductor current to the
average DC inductor current. For a good compromise
between the size of the inductor, power loss, and output voltage ripple, select an LIR of 0.3 to 0.5. The
inductance value is then given by:
includes
the primary load current and the input supply currents
for the charge pumps (see Charge-Pump Input Powerand Efficiency Considerations), linear regulator, and
VCOM buffer. Considering the typical application circuit, the maximum average DC load current
(I
MAIN(MAX)
) is 300mA with an 8V output. Based on the
above equations and assuming 85% efficiency, the
inductance value is then chosen to be 4.7µH.
The inductor’s saturation current rating should exceed
the peak inductor current throughout the normal operating range. The peak inductor current is then given by:
Under fault conditions, the inductor current may reach
up to 1.85A (I
LIM(MAX)
), see Electrical Characteristics).
However, the controller’s fast current-limit circuitry
allows the use of soft-saturation inductors while still protecting the IC.
The inductor’s DC resistance may significantly affect
efficiency due to the power loss in the inductor. The
power loss due to the inductor’s series resistance (P
LR
)
may be approximated by the following equation:
where R
L
is the inductor’s series resistance. For best per-
formance, select inductors with resistance less than the
internal N-channel MOSFET on-resistance (0.35Ω typ).
Use inductors with a ferrite core or equivalent. To minimize radiated noise in sensitive applications, use a
shielded inductor.
Output Capacitor
Output capacitor selection depends on circuit stability
and output voltage ripple. A 10µF ceramic capacitor
works well in most applications (Tables 1 and 2).
Additional feedback compensation is required (see
Feedback Compensation) to increase the margin for
stability by reducing the bandwidth further. In cases
where the output capacitance is sufficiently large, additional feedback compensation will not be necessary.
Output voltage ripple has two components: variations in
the charge stored in the output capacitor with each LX
pulse, and the voltage drop across the capacitor’s
equivalent series resistance (ESR) caused by the current into and out of the capacitor:
where I
PEAK
is the peak inductor current (see Inductor
Selection). For ceramic capacitors, the output voltage
ripple is typically dominated by V
RIPPLE(C)
. The voltage
rating and temperature characteristics of the output
capacitor must also be considered.
Feedback Compensation
For stability, add a pole-zero pair from FB to GND in the
form of a compensation resistor (R
COMP
) in series with
a compensation capacitor (C
COMP
) as shown in Figure
2. Select R
COMP
to be half the value of R2, the low-side
feedback resistor.
Integrator Capacitor
The MAX1778/MAX1880–MAX1885 contain an internal
current integrator that improves the DC load regulation
but increases the peak-to-peak transient voltage (see
the load-transient waveforms in the Typical OperatingCharacteristics). For highly accurate DC load regulation, enable the current integrator by connecting a
470pF (ƒ
OSC
= 1MHz)/1000pF (ƒ
OSC
= 500kHz)
capacitor to INTG. To minimize the peak-to-peak transient voltage at the expense of DC regulation, disable
the integrator by connecting INTG to REF. When using
the MAX1883/MAX1884/MAX1885, connect a 100kΩ
resistor to GND when disabling the integrator.
Input Capacitor
The input capacitor (CIN) in step-up designs reduces
the current peaks drawn from the input supply and
reduces noise injection. The value of CINis largely
determined by the source impedance of the input supply. High source impedance requires high input capacitance, particularly as the input voltage falls. Since
step-up DC-DC converters act as “constant-power
”
loads to their input supply, input current rises as input
voltage falls. A good starting point is to use the same
capacitance value for CINas for C
Use a Schottky diode with an average current rating
equal to or greater than the peak inductor current, and
a voltage rating at least 1.5 times the main output voltage (V
MAIN
).
Charge Pumps (MAX1778/ MAX1880/
MAX1881/MAX1882 Only)
Selecting the Number of Charge-Pump Stages
The number of charge-pump stages required to regulate the output voltage depends on the supply voltage,
output voltage, load current, switching frequency, the
diode’s forward voltage drop, and ceramic capacitor
values.
For positive charge-pump outputs, the number of
required stages may be determined by:
where V
SUPD
is the positive charge-pump diode supply
(Figure 4), V
DIODE
is the diode’s forward voltage drop,
and RTXis the charge pump’s output impedance. The
charge pump’s output impedance may be approximated using the following equation:
where the charge pump’s switching frequency (f
CHP
) is
equal to 0.5 x f
OSC
, the P-channel MOSFET’s on-resis-
tance (R
PCH(ON)
) is 10Ω, and the N-channel MOSFET’s
on-resistance (R
NCH(ON
)) is 4Ω (see Electrical
Characteristics).
For negative charge pump outputs, the number of
required stages may be determined by:
The charge pumps in the MAX1778/MAX1880/
MAX1881/MAX1882 provide regulated output voltages
by controlling the voltage drop across the low-side Nchannel MOSFET, so they can be modeled as linear
regulators followed by an unregulated charge pump
when determining the input power requirements and
efficiency.
The charge pump only provides charge to the output
capacitor during half the period (50% duty cycle), so
the input current is a function of the number of stages
and the load current:
for the positive charge pump, and:
for the negative charge pump, where N is the number
of charge pump stages.
The efficiency characteristics of the MAX1778/
MAX1880/MAX1881/MAX1882 regulated charge
pumps are similar to a linear regulator. It is dominated
by quiescent current at low output currents and by the
input voltage at higher output currents (see Typical
Operating Characteristics). So the maximum efficiency
may be approximated by:
for the positive charge pump, and:
for the negative charge pump, where V
SUPD
is the pos-
itive charge pump’s diode supply (Figure 4).
Output Voltage Selection
Adjust the positive output voltage by connecting a voltage divider from the output (V
POS
) to FBP to GND (see
Typical Operating Circuit). Adjust the negative output
voltage by connecting a voltage-divider from the output
(V
NEG
) to FBN to REF. Select R4 and R6 in the 50kΩ to
100kΩ range. Higher resistor values improve efficiency
at low output current but increase output voltage error
due to the feedback input bias current. For the negative
charge pump, higher resistor values also reduce the
load on the reference, which should not exceed 50µA
for greatest accuracy (including current through the
FLTSET resistors) to guarantee that V
REF
remains in
regulation (see Electrical Characteristics Table).
Calculate the remaining resistors with the following
equations:
R3 = R4 [(V
POS
/ V
REF
) - 1]
R5 = R6 |V
NEG
/ V
R
EF
|
where V
REF
= 1.25V. V
POS
may range from V
SUPP
to
40V, and V
NEG
may range from 0V to -40V.
Flying Capacitor
Increasing the flying capacitor (CX) value increases the
output current capability. Above a certain point,
increasing the capacitance has a negligible effect
because the output current capability becomes dominated by the internal switch resistance and the diode
impedance. The flying capacitor’s voltage rating must
exceed the following:
for the positive charge pump, and:
for the negative charge pump, where N is the stage
number in which the flying capacitor appears, and
V
SUPD
is the positive charge pump’s diode supply
(Figure 4). For example, the two-stage positive charge
pump in the typical application circuit (Figure 1) where
V
SUPP
= V
SUPD
= 8V contains two flying capacitors.
The flying capacitor in the first stage (C4) requires a
voltage rating over 12V. The flying capacitor in the second stage (C6) requires a voltage rating over 24V.
Charge-Pump Output Capacitor
Increasing the output capacitance or decreasing the
ESR reduces the output ripple voltage and the peak-topeak transient voltage. With ceramic capacitors, the
output voltage ripple is dominated by the capacitance
value. Use the following equation to approximate the
required capacitor value:
where f
CHP
is typically f
OSC
/2 (see Electrical
Characteristics).
Charge-Pump Input Capacitor
Use a bypass capacitor with a value equal to or greater
than the flying capacitor. Place the capacitor as close
to the IC as possible. Connect directly to power ground
(PGND).
Charge-Pump Rectifier Diodes
Use Schottky diodes with a current rating equal to or
greater than two times the average charge-pump input
current, and a voltage rating at least 1.5 times V
SUPP
for the positive charge pump and V
SUPN
for the nega-
tive charge pump.
Low-Dropout Linear Regulator (MAX1778/
MAX1881/MAX1883/MAX1884 Only)
Output Voltage Selection
Adjust the linear-regulator output voltage by connecting
a voltage-divider from LDOOUT to FBL to GND (Figure
5). Select R8 in the 5kΩ to 50kΩ range. Calculate R7
with the following equation:
0.8µA (max). For less than 0.5% error due to FBL input
bias current (I
F
BL
), R8 must be less than 8kΩ.
Capacitor Selection and Regulator Stability
Capacitors are required at the input and output of the
MAX1778/MAX1881/MAX1883/MAX1884 for stable
operation over the full temperature range and with load
currents up to 40mA. Connect a 1µF input bypass
capacitor (C
SUPL
) between SUPL and ground to lower
the source impedance of the input supply. Connect a
ceramic capacitor between LDOOUT and ground,
using the following equation to determine the lowest
value required for stable operation:
For example, with a 5V linear regulator output voltage
and a maximum 40mA load, use at least 4µF of output
capacitance. Applications that experience high-current
load pulses may require more output capacitance.
The ESR of the linear regulator’s output capacitor
(C
LDOOUT
) affects stability and output noise. Use out-
put capacitors with an ESR of 0.1Ω or less to ensure
stability and optimum transient response. Surfacemount ceramic capacitors are good for this purpose.
Place C
SUPL
and C
LDOOUT
as close to the linear regulator as possible to minimize the impact of PC board
trace inductance.
External Pass Transistor
For applications where the linear regulator currents
exceed 40mA or where the power dissipation in the IC
needs to be reduced, an external NPN transistor can
be used. In this case, the internal LDO only provides
the necessary base drive while the external NPN transistor supports the load, so most of the power dissipation occurs across the external transistor’s collector
and emitter.
Selection of the external NPN transistor is based on
three factors: the package’s power dissipation, the current gain (β), and the collector-to-emitter saturation voltage (V
CE(SAT)
). First, the maximum power dissipation
should not exceed the transistor’s package rating:
Once the appropriate package type is selected, consider the NPN transistor’s current gain. Since the internal LDO cannot source more than 40mA (min), the
transistor’s current gain must be high enough at the
lowest collector-to-emitter voltage to support the maximum output load:
For stable operation, place a capacitor (C
LDOOUT
) and
a minimum load resistor (R5) at the output of the internal linear regulator (the base of the external transistor)
to set the dominant pole:
Since the LDO cannot sink current, a minimum pulldown resistor (R5) is required at the base of the NPN
transistor to sink leakage currents and improve the
high-to-low load-transient response. Under no-load
conditions, leakage currents from the internal pass
transistor supply the output capacitor (C
LDOOUT
), even
when the transistor is off. As the leakage currents
increase over temperature, charge may build up on
C
LDOOUT
, making the linear regulator’s output rise
above its set point. Therefore, R5 must sink at least
100µA to guarantee proper regulation. Additionally, the
minimum load current provided by R5 improves the
high-to-low load transients by lowering the impedance
seen by C
LDOOUT
after the transient occurs. Therefore,
if large load transients are expected, select R5 so that
the minimum load current is 10% of the transistor’s
maximum base current:
Alternatively, output capacitance placed on the external
linear regulator’s output (the emitter) adds a second pole
that could destabilize the regulator. A capacitive-divider
from the transistor’s base to the feedback input (C2 and
C3, Figure 7) circumvents this second pole by adding a
pole-zero pair. Furthermore, to minimize excessive overshoot, the capacitive-divider’s ratio must be the same as
the resistive-divider’s ratio. Once the output capacitor is
selected, using the following equations to determine the
required capacitive-divider values:
A linear regulator’s minimum input-to-output voltage differential (dropout voltage) determines the lowest useable supply voltage. Because the MAX1778/
MAX1881/MAX1883/MAX1884 use an internal PNP
transistor (or external NPN transistor), their dropout
voltage is a function of the transistor’s collector-to-emitter saturation voltage (see Typical OperatingCharacteristics). The linear regulator’s quiescent current increases when in dropout.
The internal linear regulator will try to start up once its
supply voltage (V
SUPL
) exceeds 4V. When the linear
regulator powers up, the linear regulator may be in
dropout if the linear regulator’s output set voltage is
higher than its input supply voltage. Therefore, during
this brief period, the linear regulator draws additional
supply current until the input supply voltage exceeds
the output set voltage plus the pass transistor’s saturation voltage (V
LDO(SET
) + V
CE(SAT)
).
VCOM Buffer (Operational
Transconductance Amplifier)
Buffer Output Voltage and Capacitor Selection
The positive input (BUF+) features dual mode operation. Connect BUF+ to GND for the preset VSUPB/2
output voltage, set by an internal 50% resistive-divider.
Adjust the amplifier’s output voltage by connecting a
voltage-divider from SUPB to BUF+ to GND (Figure 6).
Select R12 in the 10kΩ to 100kΩ range. Calculate R11
with the following equation:
where V
SUPB
may range from 4.5V to 13V, and V
BUF+
may range from 1.2V to (V
SUPB
- 1.2V). Connect a mini-
mum 1µF ceramic capacitor from BUFOUT to ground.
PC Board Layout and Grounding
Careful PC board layout is extremely important for
proper operation. Follow the following guidelines for
good PC board layout:
1) Place the main step-up converter output diode and
output capacitor less than 0.2in (5mm) from the LX
and PGND pins with wide traces and no vias.
2) Separate analog ground and power ground. The
ground connections for the step-up converter’s and
charge pump’s input and output capacitors should
be connected to the power ground plane. The linear regulator’s and VCOM buffer’s input and output
capacitors should be connected to a separate
power-ground path, star-connected to the PGND
pin to minimize voltage drops. When using multilayer boards, the top layer should contain the boost
regulator and charge-pump power ground plane,
and the inner layer should contain the analog
ground plane and power-ground plane/path for the
V
COM
buffer and LDO. Connect all three ground
planes together at one place near the PGND pin.
3) Locate all feedback resistive-dividers as close to
their respective feedback pins as possible. The
voltage-divider’s center trace should be kept short.
Avoid running any feedback trace near the LX
switching node or the charge-pump drivers. The
resistive-dividers’ ground connections should be to
analog ground (GND).
4) When using multilayer boards, separate the top signal layer and bottom signal layer with a ground
plane between to eliminate capacitive coupling
between fast-charging nodes on the top layer and
high-impedance nodes on the bottom layer. The
fast-charging nodes, such as the LX and chargepump driver nodes, should not have any other
traces or ground planes near by.
5) Keep the charge-pump circuitry as close to the IC
as possible, using wide traces and avoiding vias
when possible. Place 0.1µF ceramic bypass
capacitors near the charge-pump input pins (SUPP
and SUPN) to the PGND pin.
6) To maximize output power and efficiency and minimize output ripple voltage, use extra wide, power
ground traces, and solder the IC’s power ground
pin directly to it.
Refer to the MAX1778/MAX1880-MAX1885 evaluation
kit for an example of proper board layout.
Notebook applications generally require low-profile
components, potentially limiting the circuit’s performance. For example, low-profile inductors typically
have lower saturation ratings and more series resistance, limiting output current and efficiency. Low-profile
capacitors have lower voltage ratings for a given
capacitance value, so 3.3µF low-profile capacitors with
voltage ratings greater than 10V were not available at
the time of publication.
Desktop Monitors
Monitor applications do not have the same component
height restrictions associated with laptops, allowing
more flexibility in component selection (Figure 8).
Larger output capacitors with higher voltage ratings
allow configurations with output voltages above 10V.
Additionally, physically larger inductors with less series
resistance and higher saturation ratings provide more
output current and higher efficiency.
Input Voltage Above and
Below the Output Voltage
Combining the step-up converter and linear regulator
as shown in Figure 9 provides output voltage regulation
above and below the input voltage. Supplied by the
step-up converter, the linear regulator output provides
a constant output voltage (V
LDO
). When the input voltage exceeds the main step-up converter’s nominal output voltage, the controller stops switching but the linear
regulator maintains the output voltage. When the input
voltage drops below the output voltage, the step-up
Figure 9. Input Voltage Above and Below the Output Voltage
converter steps up the input voltage so that the linear
regulator will not drop out. Therefore, to guarantee that
the external pass transistor does not saturate, the stepup converter’s output voltage must be set above the linear regulator’s output voltage plus the transistor’s
saturation rating (V
MAIN
≥ V
LDO
+ V
SAT
).
Power-Up Sequencing and Fault Protection
The MAX1778/MAX1880–MAX1885’s fault protection
cannot be activated until the power-up sequence is
successfully completed and the power ready output
goes low. Therefore, faults on the main output or positive charge-pump output could damage the controller
or external components. Additional fault protection may
be added as shown in Figure 10. The external MOSFET
and PNP transistor isolate the positive outputs during
startup. When the controller finishes the power-up
sequence, the power-ready output goes low, turning on
the PNP transistor. Any fault on the positive chargepump output will pull down the charge pump’s output
voltage and trigger the fault protection; otherwise, the
MOSFET’s gate slow charges. Once the MOSFET turns
on, any faults on the main step-up converter’s output
will pull down the main output voltage and trigger the
fault protection.
VCOM Buffer Startup
The VCOM buffer does not include soft-start. Therefore,
once the VCOM buffer turns on, it draws high surge
currents while charging the output capacitance. In
some applications, the buffer’s high startup surge
current could potentially trip the fault detection circuit,
forcing the controller to shut down. In these cases,
adding a soft-start resistive divider between SUPB and
BUFOUT reduces the startup surge current and voltage
drops associated with this load (Figure 11), as shown in
Figure 10. Power-Up Sequencing and Fault Protection;
V
IN
INPUT
= 3.3V
C
4.7µF
C
REF
0.22µF
L1
6.8µH
IN
IN
C1
0.22µF
30kΩ
R10
100kΩ
SHDN
MAX1778
INTG
REF
R9
FLTSET
PGND
LX
FB
SUPP
DRVP
FBP
RDY
TGND
GND
274kΩ
49.9kΩ
C4
0.1µF
C6
0.1µF
R4
49.9kΩ
R1
R2
R3
750kΩ
STARTUP
POSITIVE
V
POS(START)
R
RDY
5.1kΩ
STARTUP MAIN
V
MAIN(START)
C10
0.1µF
C5
1.0µF
= 20V
C
OUT
(2) 3.3µF
= 8V
C7
1.0µF
Q3
Q2
100kΩ
SYSTEM MAIN
V
MAIN(SYS)
R7
10kΩ
SYSTEM
POSITIVE
V
POS(SYS)
R8
= 8V
C8
3.3µF
= 20V
INPUT
V
IN
= 3.3V
the Typical Operating Characteristics. Set the resistive
divider to precharge BUFOUT, matching the buffer’s
output set voltage:
These resistor values are selected to charge the output
capacitor close to the output set voltage before the
buffer starts up:
Maxim cannot assume responsibility for use of any circuitry other than circuitry entirely embodied in a Maxim product. No circuit patent licenses are
implied. Maxim reserves the right to change the circuitry and specifications without notice at any time.
38 ____________________Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600