The LM2650 is a step-down DC/DC converter featuring high
efficiency over a 3A to milliamperes load range. This feature
makes the LM2650 an ideal fit in battery-powered applications that demand long battery life in both run and standby
modes.
The LM2650 alsofeatures a logic-controlled shutdown mode
in which it draws at most 25µA from the input power supply.
The LM2650 employs a fixed-frequency pulse-width modulation (PWM) and synchronous rectification to achieve very
high efficiencies. In many applications, efficiencies reach
95%+ for loads around 1A and exceed 90%for moderate to
heavy loads from 0.2A to 2A.
A low-power hysteretic or ″sleep″ mode keeps efficiencies
high at light loads. The LM2650 enters and exits sleep mode
automatically as the load crosses ″sleep in″ and ″sleep out″
thresholds. The LM2650 provides nodes for programming
both thresholds via external resistors. A logic input allows the
user to override the automatic sleep feature and keep the
LM2650 in PWM mode regardless of the load level.
An optional soft-start feature limits current surges from the
input power supply at start up and provides a simple means
of sequencing multiple power supplies.
Typical Application
Features
n Ultra high efficiencies (95%possible)
n High efficiency over a 3A to milliamperes load range
n Synchronous switching of internal NMOS power FETs
n Wide input voltage range (4.5V to 18V)
n Output voltage adjustable from 1.5V to 16V
n Automatic low-power sleep mode
n Logic-controlled micropower shutdown (I
n Frequency adjustable up to 300 kHz
n Frequency synchronization with external signal
n Programmable soft-start
n Short-circuit current limiting
n Thermal shutdown
n Available in 24-lead Small-Outline package
QSD
≤ 25 µA)
Applications
n Notebook and palmtop personal computers
n Portable data terminals
n Modems
n Portable Instruments
n Global positioning devices (GPSs)
n Battery-powered digital devices
1, 12SUB: These pins make electrical contact with the substrate of the die. Ground them. For best thermal
performance, ground them to the same large, uninterrupted copper plane as the PGND pins.
2SLEEP LOGIC: Use this logic input to select the conversion mode; low selects PWM, high selects sleep, and
high impedance (open) permits the LM2650 to move freely and automatically between the modes, using PWM
for moderate to heavy loads and sleep for light loads.
3, 4, 9, 10PGND: The ground return of the power stage. The power stage consists of the two power switches Q1 and
Q2, the gate drivers DH and DL, and the linear voltage regulators VRegH and VRegL. For best electrical and
thermal performance, ground these pins to a large, uninterrupted copper plane.
5, 8SW: The output node of the power stage. It swings from slightly below ground to slightly below the voltage to
. To minimize the effects of switching noise on nearby circuitry, keep all traces originating from SW short
PV
IN
and to the point. Route all traces carrying signals well away from the SW traces.
6, 7PV
: The positive supply rail of the power stage. Bypass each PVINpin to PGND with a 0.1 µF capacitor. Use
IN
capacitors having low ESL and low ESR, and locate them close to the IC.
11BOOT: The positive supply rail of the high-side gate driver DH. Connect a 0.1 µF capacitor from this node to
SW. Bootstrapping action creates a supply rail about 9V above that at PV
the gate of the NMOS power FET Q1. Overriding ensures low R
13FB: The feedback input.
14V
: An internal regulator steps the input voltage down to a 4V rail used by the signal-level circuitry. VDDis the
DD
output node of this regulator. Bypass V
15COMP: The inverting input of the error amplifier EA.
16EA OUT: The output node of the error amplifier EA.
17SS: The soft start node. Connect a capacitor from SS to GND.
18GND: The ground return of the signal-level circuitry.
19V
: The positive supply rail of the internal 4V regulator. Bypass VINto GND close to the IC with a 0.1 µF
IN
capacitor.
20FREQ ADJ: The LM2650 switches at a nominal 90 kHz. Connect a resistor between FREQ ADJ and GND to
adjust the frequency up from the nominal. Use the graph under Typical performance Characteristics to select
the resistor.
21SYNC: The synchronization input. If the switching frequency is to be synchronized with an external clock
signal, apply the clock signal here.
22SD: Use this logic input to control shutdown; pull low for operation, high for shutdown.
23SLEEP OUT ADJ (SOA): The value of the resistor connected between SIA and ground programs the sleep-in
threshold. Higher values program lower thresholds.
24SLEEP IN ADJ (SIA): The value of the resistor connected between SIA and ground programs the sleep-in
threshold. Higher values program lower thresholds.
Top View
, and DH uses this rail to override
IN
.
DS(on)
to GND close to the IC with a 0.2 µF capacitor.
DD
www.national.com2
Page 3
Absolute Maximum Ratings (Note 1)
If Military/Aerospace specified devices are required,
please contact the National Semiconductor Sales Office/
Distributors for availability and specifications.
(All voltages are referenced to the PGND and GND pins.)
DC Voltage at PV
DC Voltage at SD, SLEEP LOGIC
and SYNC15V
DC current into SW
Junction TemperatureLimited by the IC
DC Power Dissipation (Note 2)1.28W
Storage Temperature−65˚C to +150˚C
Note 1: Absolute Maximum Ratings are limits beyond which damage to the device may occur. Operating Ratings are conditions under which the device operates
correctly. Operating ratings do not imply guaranteed performance limits. For guaranteed performance limits and associated test conditions, see the Electrical Characteristics.
Note 2: This rating is calculated using the formula P
junction temperature, and θ
78˚C/W for T
the safe dissipation of more power. See Application Notes on thermal management. The LM2650 actively limits its junction temperature to about 170˚C.
Note 3: For detailed information on soldering plastic small-outline packages, refer to the Packaging Databook published by National Semiconductor Corporation.
Note 4: ESD is applied using the human-body model, a 100pF capacitor discharged through a 1.5kΩ resistor.
Note 5: A typical is the center of characterization data taken at T
Note 6: All limits are guaranteed. The guarantee is backed with 100%testing at T
Note 7: V
Note 8: Quiescent current is the total current flowing into the P
nominal switching frequency. I
Note 9: Pulling 100µA out of FREQ ADJ simulates adjusting the oscillator frequency with a 12.5 kΩ resistor connected from FREQ ADJ to GND. The sleep mode
cannot be used at switching frequencies above 250 kHz.
Minimum Duty CycleV
Internal Rail VoltageI
Bootstrap Regulator Voltage
(VRegH)
Soft Start Current10
Hysteresis of the Sleep
Comparator (C2
of SD0.95V(max)
V
IL
V
of SD2.10V(min)
IH
V
of SLEEP LOGIC0.9V(max)
IL
V
of SLEEP LOGIC2.0V(min)
IH
V
of SYNC0.50V(max)
IL
V
of SYNC1.45V(min)
IH
Figure 2
TJfor Thermal Shutdown170˚C
is the junction ot ambient thermal resistance of the package. The P
JA
and θJArespectively.A θJAof 78˚C represents the worst condition of no heat sinking of the M24B small-outline package. Heat sinking allows
Jmax,TA
is measured at SLEEP OUT ADJ.
REF
includes no such current.
QS
=
0V unless superseded under Conditions. Typicals and limits appearing in plain
SD
=
+50 mV,
V
FB
REF
Not Adjusted
F
OSC
=1mA4.0
VDD
2.8
5
3.6/3.4
4.2/4.3
I
=1mA7.5
BOOT
6.5/6.0
13.5/20.0
V
)
SLEEPLOGIC
=
3V30
10
50
DCmax
=
)/θJA, where P
(T
Jmax−TA
= 25˚C.
A=TJ
and VINpins. IQincludes the current used to drive the gates of the two NMOS power FETsat the
VIN
A=TJ
is the absolute maximum power dissipation, T
DCmax
= 125˚C and statistical correlation for room temperature and cold limits.
rating of 1.28W results from substituting 170˚C, 70˚C and
DCmax
is the maximum
Jmax
%
V(min)
V(max)
V(min)
µA(max)
mV(min)
mV(max)
%
(min)
V
V
µA
mV
www.national.com4
Page 5
Typical Performance Characteristics
I
vs Input Voltage
QSD
IQSvs Input Voltage
IQvs Input Voltage
IQvs Oscillator Frequency
R
Low-Side vs Junction
DS(on)
Temperature
DS012848-3
DS012848-6
DS012848-9
R
Low-Side vs Input Voltage
DS(on)
R
High-Side vs Junction
DS(on)
Temperature
DS012848-4
DS012848-7
DS012848-10
DS012848-5
R
High-Side vs Input Voltage
DS(on)
DS012848-8
Oscillator Frequency vs Junction
Temperature
DS012848-11
Oscillator Frequency vs Adjusting Resistor
DS012848-12
Current Limit vs Junction Temperature
DS012848-13
www.national.com5
Page 6
Block Diagrams
FIGURE 1. The PWM Circuit with External Components in a Closed Control Loop
DS012848-15
DS012848-16
FIGURE 2. The Hysteretic or ″Sleep″ Circuit with External Components in a Closed Control Loop
DS012848-21
FIGURE 3. The Internal Voltage Regulator and Voltage Reference used by Both the PWM and Hysteretic Circuits
www.national.com6
Page 7
Operation
OVERVIEW
The LM2650 uses two step-down conversion modes:
fixed-frquency pulse-width modulation (PWM) and hysteretic. It moves freely and automatically between them, using
PWM for moderate to heavy loads and hysteretic for light
loads.
For clarity, separate block diagrams for each conversion
mode have been included. See
Blocks used in both modes appear in both diagrams with the
same label. For example, both modes use the input buffer B.
To keep the diagrams simple, most power supply rails have
been omitted. R3, C10, R
are outside the IC.
THE PWM CIRCUIT (
Figure 1
The PWM is a fixed-frequency, voltage-mode pulse-width
modulator. It consists of four functional blocks: an input
buffer, an error amplifier, a modulator, and a power stage.
1. The input buffer B: B is a voltage follower. A fraction of
the output voltage is fed back to its noninverting input
FB. Circumventing B by using the COMP input as the
feedback input will cause the IC to malfunction.
2. The error amplifier EA: EA is a voltage amplifier. It subtracts the feedback voltage from the 1.25V reference
and amplifies the difference to produce an error voltage
for the control loop. For the purpose of loop compensation, EA is typically configured as an integrator. In this
configuration, a capacitor C
nected in series between the inverting input COMP and
the output terminal EA OUT.The capacitor and the internal 6.5kΩ resistor create a pole, while the capacitor and
series resistor create a zero.
3. The modulator: The modulator is the heart of the PWM
circuit. It consists of the 90 kHz oscillator, the voltage
comparator C1, and output logic represented here as a
simple SR latch.
The modulator generates a continuous stream of rectangular, signal-level. It generates the pulses at a fixed frequency, and it modulates or varies their widths in response to variations in the error voltage. The pulses
appear at Q, the output of the SR latch. An increase in
the error voltage results in a proportional increase in the
pulse widths, and, conversely, a decrease in the error
voltage results in a proportional decrease in the pulse
widths.
The oscillator produces a 90 kHz sawtooth that ramps
between 1V and 2V. At the beginning of each ramp, the
oscillator sets the SR latch sending Q high. As the ramp
voltage surpasses the error voltage, C1 resets the SR
latch sending Q low. An increase in the error voltage increases the time between the setting and the resetting of
the SR latch which , in turn, results in an equal increase
in pulse widths: that is, an equal increase in the time Q
spends high in each cycle. A decrease in the error voltage has the opposite effect on the pulse widths as it decreases the time between the setting and resetting of the
SR latch.
4. The power stage: The power stage puts some punch between the output of the modulator by translating the
stream of signal-level pulses generated by the modulator into a stream of power pulses that swing from ground
up to the input voltage while sinking and sourcing as
much as 3.5A. The power stage consists of two gate
Figure 1
C,CC,CB
, L1, R1, R2, and C
)
and a resistor RCare con-
C
and
Figure 2
OUT
drivers DH and DL, two linear voltage regulators VRegH
and VRegL, and two NMOS power FETs Q1 and Q2.
The power pulses appear at the SW mode. When Q
goes high, DL drives the gate of Q2 low turning Q23 off.
While Q2 turns off, the SW potential may remain at just
below ground as the body diode of Q2 conducts what
was previously reverse current (source-to-drain) in Q2,
or the SW potential may swing up to just above the input
voltage as the body diode of Q1 conducts what was pre-
.
viously forward current (drain-to-source) in Q2. About 50
ns after Q goes high, DH drives the gate of Q1 high turning Q1 on. If the task remains, Q1 pulls the SW potential
up, if not, Q1 simply takes over the conduction responsibility from its own body diode. When Q goes low, the inverse action occurs resulting in the SW potential swinging from the input voltage to the ground. The 50 ns delay
between one switch beginning to turn off and the other
switch beginning to turn on prevents the switches from
″shooting through″ directly from the input supply to the
ground.
The PWM circuit drives the pulse stream into the
low-pass filter made up of L1 and C
passed the DC component of the stream and attenuates
the AC components. The output of the filter is the DC
voltage V
Since the DC component of any periodic waveforms the
average value of the waveform, V
ing:
superimposed with a small ripple voltage.
OUT
OUT
Here T is the switching period in seconds V(t) is the pulse
stream. Under DC steady-state conditions, (1) yields
Here V
is the input voltage, and therefore the height of the
IN
pulses, in volts, is the width of the pulses in seconds, and D
is the ratio of t
The output voltage is programmed using the resistive divider
to T, the duty or the duty cycle.
ON
made up for R1 and R2,
As Q1 turns on, its source voltage swings up to just below
the input voltage. The LM2650 uses a simple technique
called ″bootstrapping″ to pull the positive supply rail of DH
(at BOOT) up along with the source voltage of Q1, but to a
voltage above the input voltage. Because the source of Q1
and the positive supply rail of DH make the same voltage
swing together, DH maintains the positive gate-to-source
voltage required to turn Q1 on. Q12 plays an active role in
pulling the supply rail of DH up and is therefore said to pull itself up by its ″bootstraps″, thus the name of the technique
and of the BOOT pin.
In the typical application, a capacitor CB is connected outside the IC between the BOOT and SW pins. When Q2 is on,
the input supply charges CB through VRegH and the internal
diode D.
THE HYSTERETIC CIRCUIT AND LOOP (
Except for C2, the hysteretic circuit borrows all its circuit
blocks from the PWM circuit.
The hysteretic comparator C2 is a voltage comparator with
built-in hysteresis V
1.25V.
of typically 30mV centered at
HYST
. The filter
OUT
can be found us-
(1)
(2)
(3)
Figure 2
)
www.national.com7
Page 8
Operation (Continued)
The diode D2 is the body diode of Q2. The hysteretic circuit
uses D2 as a rectifier instead of switching Q2 as a synchronous rectifier.
When the load current drops below the prescribed sleep-in
threshold, the LM2650 shuts down the PWM loop and starts
up the hysteretic loop. The hysteretic loop supports light
loads more efficiently because it uses less power to support
its own operation; it uses less bias power because it’s a simpler loop having less circuit blocks to bias, and it switches
slower, so it incurs lower switching losses.
The hysteretic control loop does not switch at a constant frequency. Instead, it monitors V
V
reaches either side of a narrow window centered on
OUT
the desired output voltage. C2 directs the switching based
on its reading of the feedback voltage. Switching in this manner yields a regulated voltage consisting of the desired output voltage and an AC ripple voltage. The magnitude of the
AC component can be approximated using
and switches only when
OUT
back voltage just surpasses the lower hysteretic threshold of
C2, the output of C2 changes states from low to high, and
DH responds by pulling the gate of Q1 up turning Q1 on and
starting the hysteretic cycle over.
Note that as the load current decreases, it takes increasingly
longer periods for the load current to discharge C
through the hysteretic window, and as the load current in-
OUT
creases, the periods become even shorter. It can be seen
from the above observation that the switching frequency of
the hysteretic loop varies as the load varies. The switching
frequency can be approximated using
(6)
Here f is the switching frequency in hertz, I is the load current
in amperes, C
V
OUT_PP
Typical switching frequencies range anywhere from a few
is the value of the capacitor in farads, and
OUT
is the magnitude of the AC ripple voltage in volts.
hertz for very light loads to a few thousand hertz for light
loads bordering on the moderate level.
(4)
For example, with V
120mV,
OUT
set to 5V,V
OUT_PP
is approximately
(5)
When it starts up, the hysteretic loop turns Q1 on. While Q1
is on, the input power supply charges C
current to the load. Current from the supply reaches C and
and supplies
OUT
the load via the series path provided by Q1 and L1. As the
feedback voltage just surpasses the upper hysteretic threshold of C2, the output of C2 changes from high to low, and HD
responds by pulling the gate of Q1 down turning Q1 off. As
Q1 turns off, L1 generates a negative-going voltage transient
that D2 clamps at just below ground. D2 remains on only
briefly as the current in L1 runs out. While both Q1 and D2
are off, C
alone supplies current to the load. As the feed-
OUT
Application Circuits
Figure 4
is a schematic of the typical application circuit. use
the component values shown in the figure and those contained in
Table 1
DC/DC converter. As with the design of any DC/DC converter, the design of these circuits involved tradeoffs between efficiency, size, and cost. Here more weight was given
to efficiency than to size as evidenced by the low switching
frequency which keeps switching losses low but pushes the
value and size of the inductor up.
From a smaller circuit, use the component values shown in
Figure 4
slightly higher switching losses for a much smaller inductor.
Note,
Figure 4
adjust the switching frequency from 90 kHz up to 200 kHz.
Connect R
to build a 5V, 3A, or 3.3V, 3A step-down
and those contained in
Table3
. These circuits trade
does not show RFA, the resistor required to
between the FREQ ADJ pin and ground.
FA
FIGURE 4. The Typical 90 kHz Application Circuit
www.national.com8
DS012848-19
Page 9
Application Circuits (Continued)
TABLE 1. Components for the Typical 90 kHz Application Circuit
Input Voltage7V to 18V IN
Applicable Cell Stacks8 to 12 Cell NiCd or NiMh, 3 to 4 Cell Li Ion, 8 to 11 Cell Alkaline, 6 Cell
Output5V, 3A Out3.3V, 3A out
Input Capacitor C
IN
Inductor L140µH (See
Output Capacitor C
OUT
Feedback Resistors R1
and R2
Compensation
Components R
and C
Sleep Resistors R
and R
10
SOA
C,CC,R3
SIA
,
2 x 22 µF, 35V AVX TPS
Series or Sprague 593D Series
Table 2
3x220 µF, 10V AVX TPS
Series or Sprague 593D Series
R1=75kΩ,1%,
R2=24.9kΩ,1%,
=
R
37.4 kΩ,
C
=
4.7 nF,
C
C
=
3.57 kΩ,
R
3
=
5.6 nF
C
10
=
R
33 kΩ,
SIA
=
200 kΩ
R
SOA
TABLE 2. Toroidal Inductors Using Cores from MICROMETALS, INC.
NATIONAL’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT
DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT AND GENERAL
COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein:
1. Life support devices or systems are devices or
systems which, (a) are intended for surgical implant
into the body, or (b) support or sustain life, and
whose failure to perform when properly used in
accordance with instructions for use provided in the
2. A critical component is any component of a life
support device or system whose failure to perform
can be reasonably expected to cause the failure of
the life support device or system, or to affect its
safety or effectiveness.
labeling, can be reasonably expected to result in a
significant injury to the user.
National does not assume any responsibility for use of any circuitry described, no circuit patent licenses are implied and National reserves the right at any time without notice to change said circuitry and specifications.