MAXIM MAX5038AEAI18

19-3034; Rev 0; 10/03
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
Features
♦ +4.75V to +5.5V or +8V to +28V Input Voltage
Range
♦ Up to 60A Output Current
♦ Internal Voltage Regulator for a +12V or +24V
Power Bus
♦ True Differential Remote Output Sensing
♦ Two Out-Of-Phase Controllers Reduce Input
Capacitance Requirement and Distribute Power
Dissipation
♦ Average-Current-Mode Control
Superior Current Sharing Between Individual
Phases and Paralleled Modules
Accurate Current Limit Eliminates MOSFET and
Inductor Derating
♦ Limits Reverse-Current Sinking in Paralleled
Modules
♦ Integrated 4A Gate Drivers
♦ Selectable Fixed Frequency 250kHz or 500kHz per
Phase (Up to 1MHz for Two Phases)
♦ Fixed (MAX5038A) or Adjustable (MAX5041A)
Output Voltages
♦ External Frequency Synchronization from 125kHz
to 600kHz
♦ Internal PLL with Clock Output for Paralleling
Multiple DC-DC Converters
♦ Thermal Protection
♦ 28-Pin SSOP Package
Ordering Information
Applications
Servers and Workstations
Point-of-Load High-Current/High-Density
Telecom DC-DC Regulators
Networking Systems
Large-Memory Arrays
PART
TEMP RANGE
PINPACKAGE
OUTPUT
VOLTAGE
(V)
MAX5038AEAI12
-40°C to +85°C
28 SSOP
Fixed +1.2
MAX5038AEAI15
-40°C to +85°C
28 SSOP
Fixed +1.5
MAX5038AEAI18
-40°C to +85°C
28 SSOP
Fixed +1.8
RAID Systems
MAX5038AEAI25
-40°C to +85°C
28 SSOP
Fixed +2.5
High-End Desktop Computers
MAX5038AEAI33
-40°C to +85°C
28 SSOP
Fixed +3.3
MAX5041AEAI
-40°C to +85°C
28 SSOP
Adj +1.0 to
+3.3
Pin Configuration appears at end of data sheet.
________________________________________________________________Maxim Integrated Products
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.
1
MAX5038A/MAX5041A
General Description
The MAX5038A/MAX5041A dual-phase, PWM controllers
provide high-output-current capability in a compact
package with a minimum number of external components. The MAX5038A/MAX5041A utilize a dual-phase,
average-current-mode control that enables optimal use
of low RDS(ON) MOSFETs, eliminating the need for external heatsinks even when delivering high output currents.
Differential sensing enables accurate control of the output voltage, while adaptive voltage positioning provides
optimum transient response. An internal regulator
enables operation with input voltage ranges of +4.75V to
+5.5V or +8V to +28V. The high switching frequency, up
to 500kHz per phase, and dual-phase operation allow
the use of low-output inductor values and input capacitor
values. This accommodates the use of PC boardembedded planar magnetics achieving superior reliability, current sharing, thermal management, compact size,
and low system cost.
The MAX5038A/MAX5041A also feature a clock input
(CLKIN) for synchronization to an external clock, and a
clock output (CLKOUT) with programmable phase delay
(relative to CLKIN) for paralleling multiple phases. The
MAX5038A/MAX5041A also limit the reverse current in
case the bus voltage becomes higher than the regulated output voltage. The MAX5038A offers a variety of factory-trimmed preset output voltages (see Selector Guide)
and the MAX5041A offers an adjustable output voltage
between +1.0V to +3.3V.
The MAX5038A/MAX5041A operate over the extended
temperature range (-40°C to +85°C) and are available
in a 28-pin SSOP package. Refer to the MAX5037A and
MAX5065/MAX5067 data sheets for a VRM 9.0/VRM 9.1compatible, VID-controlled, adjustable output voltage
controller in a 44-pin MQFP/thin QFN or 28-pin SSOP
package.
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
ABSOLUTE MAXIMUM RATINGS
IN to SGND.............................................................-0.3V to +30V
BST_ to SGND ........................................................-0.3V to +35V
DH_ to LX_ ................................-0.3V to [(VBST_ - VLX_) + 0.3V]
DL_ to PGND ..............................................-0.3V to (VCC + 0.3V)
BST_ to LX_ ..............................................................-0.3V to +6V
VCC to SGND............................................................-0.3V to +6V
VCC to PGND............................................................-0.3V to +6V
SGND to PGND .....................................................-0.3V to +0.3V
All Other Pins to SGND...............................-0.3V to (VCC + 0.3V)
Continuous Power Dissipation (TA = +70°C)
28-Pin SSOP (derate 9.5mW/°C above +70°C) ..........762mW
Operating Temperature Range ...........................-40°C to +85°C
Maximum Junction Temperature .....................................+150°C
Storage Temperature Range .............................-60°C to +150°C
Lead Temperature (soldering, 10s) .................................+300°C
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.
ELECTRICAL CHARACTERISTICS
(VCC = +5V, circuit of Figure 1, TA = -40°C to +85°C, unless otherwise noted. Typical specifications are at TA = +25°C.) (Note 1)
PARAMETER
SYMBOL
CONDITIONS
MIN
TYP
MAX
UNITS
SYSTEM SPECIFICATIONS
8
28
4.75
5.5
Input Voltage Range
VIN
Quiescent Supply Current
IQ
EN = VCC or SGND
4
Efficiency
η
ILOAD = 52A (26A per phase)
90
Short IN and VCC together for +5V input
operation
10
V
mA
%
OUTPUT VOLTAGE
MAX5038A only, no load
Nominal Output Voltage
Accuracy (Note 4)
-0.8
+0.8
-1
+1
MAX5041A only, no load
0.992
1.008
MAX5041A only, no load, VIN = VCC =
+4.75V to +5.5V or VIN = +8V to +28V
0.990
1.010
MAX5038A only, no load, VIN = VCC =
+4.75V to +5.5V or VIN = +8V to +28V
(Note 2)
SENSE+ to SENSE- Voltage
Accuracy (Note 4)
%
V
STARTUP/INTERNAL REGULATOR
VCC Undervoltage Lockout
UVLO
VCC rising
4.0
VCC Undervoltage Lockout
Hysteresis
4.15
4.5
200
VCC Output Accuracy
VIN = +8V to +28V, ISOURCE = 0 to 80mA
4.85
V
mV
5.1
5.30
V
1
3
Ω
MOSFET DRIVERS
Output Driver Impedance
RON
Output Driver Source/Sink
Current
IDH_, IDL _
Nonoverlap Time
tNO
Low or high output
CDH_/DL _ = 5nF
4
A
60
ns
OSCILLATOR AND PLL
Switching Frequency
fSW
PLL Lock Range
fPLL
PLL Locking Time
tPLL
2
CLKIN = SGND
238
250
262
CLKIN = VCC
475
500
525
125
600
200
_______________________________________________________________________________________
kHz
kHz
µs
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
(VCC = +5V, circuit of Figure 1, TA = -40°C to +85°C, unless otherwise noted. Typical specifications are at TA = +25°C.) (Note 1)
PARAMETER
CLKOUT Phase Shift
(at fSW = 125kHz)
CLKIN Input Pulldown Current
SYMBOL
φCLKOUT
CONDITIONS
MIN
TYP
MAX
PHASE = VCC
115
120
125
PHASE = unconnected
85
90
95
PHASE = SGND
55
60
65
5
7
µA
0.8
V
ICLKIN
3
CLKIN High Threshold
VCLKINH
2.4
CLKIN Low Threshold
VCLKINL
CLKIN High Pulse Width
tCLKIN
200
PHASE High Threshold
VPHASEH
4
PHASE Low Threshold
VPHASEL
PHASE Input Bias Current
CLKOUT Output Low Level
VCLKOUTL
CLKOUT Output High Level
VCLKOUTH ISOURCE = 2mA (Note 2)
Degrees
V
ns
V
1
IPHASEBIA
UNITS
-50
ISINK = 2mA (Note 2)
V
+50
µA
100
mV
4.5
V
CURRENT LIMIT
Average Current-Limit Threshold
VCL
CSP_ to CSN_
45
Reverse Current-Limit Threshold
VCLR
CSP_ to CSN_
-3.9
Cycle-by-Cycle Current Limit
VCLPK
CSP_ to CSN_ (Note 3)
Cycle-by-Cycle Overload
Response Time
tR
90
VCSP_ to VCSN_ = +150mV
48
112
51
mV
-0.2
mV
130
mV
260
ns
CURRENT-SENSE AMPLIFIER
CSP_ to CSN_ Input Resistance
Common-Mode Range
Input Offset Voltage
RCS_
4
kΩ
VCMR(CS)
-0.3
+3.6
V
VOS(CS)
-1
+1
mV
Amplifier Gain
AV(CS)
18
V/V
3dB Bandwidth
f3dB
4
MHz
CURRENT-ERROR AMPLIFIER (TRANSCONDUCTANCE AMPLIFIER)
Transconductance
gmca
Open-Loop Gain
AVOL(CE)
No load
550
µS
50
dB
DIFFERENTIAL VOLTAGE AMPLIFIER (DIFF)
Common-Mode Voltage Range
VCMR(DIFF)
DIFF Output Voltage
VCM
Input Offset Voltage
VOS(DIFF)
Amplifier Gain
AV(DIFF)
3dB Bandwidth
f3dB
Minimum Output Current Drive
SENSE+ to SENSE- Input
Resistance
-0.3
VSENSE+ = VSENSE- = 0
+1.0
V
+1
mV
0.6
-1
V
MAX5038A (+1.2V, +1.5V, +1.8V output
versions), MAX5041A
0.997
1
1.003
MAX5038A (+2.5V and +3.3V output versions)
0.495
0.5
0.505
CDIFF = 20pF
3
IOUT(DIFF)
1.0
RVS_
50
V/V
MHz
mA
100
kΩ
_______________________________________________________________________________________
3
MAX5038A/MAX5041A
ELECTRICAL CHARACTERISTICS (continued)
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
ELECTRICAL CHARACTERISTICS (continued)
(VCC = +5V, circuit of Figure 1, TA = -40°C to +85°C, unless otherwise noted. Typical specifications are at TA = +25°C.) (Note 1)
PARAMETER
SYMBOL
CONDITIONS
MIN
TYP
MAX
UNITS
VOLTAGE-ERROR AMPLIFIER (EAOUT)
Open-Loop Gain
AVOL(EA)
Unity-Gain Bandwidth
fUGEA
EAN Input Bias Current
IB(EA)
Error-Amplifier Output Clamping
Voltage
70
dB
3
VEAN = +2.0V
VCLAMP(EA) With respect to VCM
MHz
-100
+100
nA
810
918
mV
THERMAL SHUTDOWN
Thermal Shutdown
TSHDN
Thermal-Shutdown Hysteresis
150
°C
8
°C
EN INPUT
EN Input Low Voltage
VENL
EN Input High Voltage
VENH
3
IEN
4.5
EN Pullup Current
Note 1:
Note 2:
Note 3:
Note 4:
4
1
V
5.5
µA
V
5
Specifications from -40°C to 0°C are guaranteed by characterization but not production tested.
Guaranteed by design. Not production tested.
See Peak-Current Comparator section.
Does not include an error due to finite error amplifier gain (see the Voltage-Error Amplifier section).
_______________________________________________________________________________________
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
80
f = 250kHz
70
60
VIN = +5V
VOUT = +1.8V
40
30
30
20
VOUT = +1.8V
fSW = 250kHz
0
IOUT (A)
EFFICIENCY vs. OUTPUT CURRENT
AND OUTPUT VOLTAGE
EFFICIENCY vs. OUTPUT CURRENT
AND OUTPUT VOLTAGE
SUPPLY CURRENT
vs. FREQUENCY AND INPUT VOLTAGE
80
VOUT = +1.5V
70
η (%)
VOUT = +1.1V
60
50
40
40
30
30
20
VOUT = +1.8V
12.0
11.5
11.0
10.5
ICC (mA)
VOUT = +1.8V
90
MAX5038A/41A toc05
VOUT = +1.5V
100
VOUT = +1.1V
20
VIN = +12V
fSW = 250kHz
10
0
VIN = +5V
fSW = 500kHz
VIN = +24V
10.0
9.5
9.0
8.5
8.0
7.5
7.0
VIN = +12V
VIN = +5V
6.5
0
MAX5038A/41A toc06
0 4 8 12 16 20 24 28 32 36 40 44 48 52
IOUT (A)
50
EXTERNALCLOCK
NO DRIVER LOAD
6.0
0 4 8 12 16 20 24 28 32 36 40 44 48 52
0 4 8 12 16 20 24 28 32 36 40 44 48 52
100 150 200 250 300 350 400 450 500 550 600
IOUT (A)
IOUT (A)
FREQUENCY (kHz)
SUPPLY CURRENT
vs. TEMPERATURE AND FREQUENCY
SUPPLY CURRENT
vs. TEMPERATURE AND FREQUENCY
SUPPLY CURRENT
vs. LOAD CAPACITANCE PER DRIVER
250kHz
70
600kHz
150
100
500kHz
90
80
70
ICC (mA)
125kHz
50
ICC (mA)
125
60
100
40
MAX5038A/41A toc09
90
80
175
MAX5038A/41A toc07
100
60
50
40
75
30
0
VIN = +24V
VOUT = +1.8V
fSW = 125kHz
10
IOUT (A)
60
10
50
0 4 8 12 16 20 24 28 32 36 40 44 48 52
70
20
60
40
0
80
η (%)
50
0 4 8 12 16 20 24 28 32 36 40 44 48 52
90
ICC (mA)
70
VIN = +5V
60
10
100
10
80
MAX5038A/41A toc08
40
90
20
MAX5038A/41A toc04
50
VIN = +12V
70
η (%)
η (%)
80
90
η (%)
f = 500kHz
EFFICIENCY vs. OUTPUT CURRENT
100
MAX5038A/41A toc02
90
100
MAX5038A/41A toc01
100
EFFICIENCY vs. OUTPUT CURRENT
AND INPUT VOLTAGE
MAX5038A/41A toc03
EFFICIENCY vs. OUTPUT CURRENT AND
INTERNAL OSCILLATOR FREQUENCY
VIN = +12V
CDL_ = 22nF
CDH_ = 8.2nF
-40
-15
50
30
20
VIN = +5V
CDL_ = 22nF
CDH_ = 8.2nF
25
10
35
TEMPERATURE (°C)
60
85
VIN = +12V
fSW = 250kHz
10
0
-40
-15
10
35
TEMPERATURE (°C)
60
85
1
3
5
7
9
11
13
15
CDRIVER (nF)
_______________________________________________________________________________________
5
MAX5038A/MAX5041A
Typical Operating Characteristics
(Circuit of Figure 1. TA = +25°C, unless otherwise noted.)
Typical Operating Characteristics (continued)
(Circuit of Figure 1, TA = +25°C, unless otherwise noted.)
OUTPUT VOLTAGE vs. OUTPUT CURRENT
AND ERROR AMP GAIN (RF / RIN)
53
52
VIN = +12V
VOUT = +1.8V
RF / RIN = 15
1.80
DIFFERENTIAL AMPLIFIER BANDWIDTH
90
45
3.0
PHASE
0
2.5
VOUT (V)
51
50
49
PHASE 2
48
PHASE 1
GAIN (V/V)
RF / RIN = 12.5
1.75
1.70
RF / RIN = 7.5
47
RF / RIN = 10
1.65
-45
2.0
-90
1.5
-135
GAIN
1.0
-180
0.5
46
1.3
1.4
1.5
1.6
1.7
0
DIFF OUTPUT ERROR
vs. SENSE+ TO SENSE- VOLTAGE
VCC LOAD REGULATION
vs. INPUT VOLTAGE
VIN = +24V
VIN = +12V
5.10
VCC (V)
0.100
5.00
0.075
4.95
0.050
4.90
0.025
4.85
0
4.80
5.25
5.20
VIN = +8V
ICC = 0
5.15
5.10
VCC (V)
5.05
0.125
10
1
VCC LINE REGULATION
5.20
5.15
0.1
FREQUENCY (MHz)
ILOAD (A)
0.150
-270
0.01
5 10 15 20 25 30 35 40 45 50 55
VOUT (V)
VIN = +12V
NO DRIVER
0.175
1.8
MAX5038A/41A toc15
1.2
MAX5038A/41A toc13
0.200
1.1
-225
0
1.60
1.0
MAX5038A/41A toc14
45
ERROR (%)
MAX5038A/41A toc12
3.5
ICC = 40mA
5.05
5.00
4.95
4.90
4.85
4.80
DC LOAD
0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0
VIN (V)
VCC LINE REGULATION
DRIVER RISE TIME
vs. DRIVER LOAD CAPACITANCE
DRIVER FALL TIME
vs. DRIVER LOAD CAPACITANCE
5.00
80
70
60
DL_
DH_
50
40
30
20
10
0
4.95
4.90
4.85
ICC = 80mA
9
10
11
VIN (V)
12
13
6
11
16
21
CDRIVER (nF)
90
26
31
80
70
60
DL_
DH_
50
40
30
20
10
0
VIN = +12V
fSW = 250kHz
1
MAX5038A/41A toc18
MAX5038A/41A toc17
90
tR (ns)
5.05
120
110
100
tF (ns)
5.10
6
120
110
100
MAX5038A/41A toc16
5.15
8
10 12 14 16 18 20 22 24 26 28
ICC (mA)
5.20
4.75
8
∆VSENSE (V)
5.25
4.80
4.75
15 30 45 60 75 90 105 120 135 150
36
VIN = +12V
fSW = 250kHz
1
6
11
16
21
CDRIVER (nF)
_______________________________________________________________________________________
26
31
36
PHASE (DEGREES)
54
(VCSP_ - VCSN_) (mV)
1.85
MAX5038A/41A toc10
55
MAX5038A/41A toc11
CURRENT-SENSE THRESHOLD
vs. OUTPUT VOLTAGE
VCC (V)
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
HIGH-SIDE DRIVER (DH_)
SINK AND SOURCE CURRENT
PLL LOCKING TIME
250kHz TO 350kHz AND
350kHz TO 250kHzMAX5038A/41A toc21
LOW-SIDE DRIVER (DL_)
SINK AND SOURCE CURRENT
MAX5038A/41A toc19
MAX5038A/41A toc20
CLKOUT
5V/div
350kHz
DH_
1.6A/div
PLLCMP
200mV/div
DL_
1.6A/div
250kHz
0
VIN = +12V
CDL_ = 22nF
VIN = +12V
CDH_ = 22nF
100ns/div
VIN = +12V
NO LOAD
100ns/div
PLL LOCKING TIME
250kHz TO 500kHz AND
500kHz TO 250kHzMAX5038A/41A toc22
100µs/div
PLL LOCKING TIME
250kHz TO 150kHz AND
150kHz TO 250kHzMAX5038A/41A toc23
HIGH-SIDE DRIVER (DH_)
RISE TIME
MAX5038A/41A toc24
CLKOUT
5V/div
CLKOUT
5V/div
DH_
2V/div
250kHz
PLLCMP
200mV/div
500kHz
PLLCMP
200mV/div
150kHz
0
250kHz
VIN = +12V
NO LOAD
0
VIN = +12V
CDH_ = 22nF
VIN = +12V
NO LOAD
100µs/div
100µs/div
HIGH-SIDE DRIVER (DH_)
FALL TIME
LOW-SIDE DRIVER (DL_)
RISE TIME
MAX5038A/41A toc25
MAX5038A/41A toc27
DL_
2V/div
VIN = +12V
CDL_ = 22nF
40ns/div
LOW-SIDE DRIVER (DL_)
FALL TIME
MAX5038A/41A toc26
DH_
2V/div
VIN = +12V
CDH_ = 22nF
40ns/div
DL_
2V/div
VIN = +12V
CDL_ = 22nF
40ns/div
40ns/div
_______________________________________________________________________________________
7
MAX5038A/MAX5041A
Typical Operating Characteristics (continued)
(Circuit of Figure 1, TA = +25°C, unless otherwise noted.)
Typical Operating Characteristics (continued)
(Circuit of Figure 1, TA = +25°C, unless otherwise noted.)
OUTPUT RIPPLE
ENABLE STARTUP RESPONSE
INPUT STARTUP RESPONSE
MAX5038A/41A toc28
MAX5038A/41A toc30
MAX5038A/41A toc29
VPGOOD
1V/div
VPGOOD
1V/div
VOUT
1V/div
VOUT
1V/div
VOUT
(AC-COUPLED)
10mV/div
VIN
5V/div
VIN = +12V
VOUT = +1.75V
IOUT = 52A
VIN = +12V
VOUT = +1.75V
IOUT = 52A
500ns/div
VIN = +12V
VOUT = +1.75V
IOUT = 52A
REVERSE-CURRENT SINK
AT INPUT TURN-ON
REVERSE-CURRENT SINK vs.
TEMPERATURE
LOAD-TRANSIENT RESPONSE
MAX5038A/41A toc31
MAX5038A/41A toc33
MAX5038A/41A toc32
2.8
R1 = R2 = 1.5mΩ
2.7
VIN = +12V
VOUT = +1.5V
VEXTERNAL = 2.5V
R1 = R2 = 1.5mΩ
VEXTERNAL = +3.3V
VOUT
50mV/div
VIN = +12V
VOUT = +1.75V
ISTEP = 8A TO 52A
tRISE = 1µs
VEN
2V/div
1ms/div
2ms/div
IREVERSE (A)
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
2.6
REVERSE
CURRENT
5A/div
0A
2.5
VEXTERNAL = +2V
2.4
2.3
40µs/div
VIN = +12V
VOUT = +1.5V
-40
-15
10
35
60
200µs/div
85
TEMPERATURE (°C)
REVERSE-CURRENT SINK
AT ENABLE TURN-ON
REVERSE-CURRENT SINK
AT INPUT TURN-ON
VIN = +12V
VOUT = +1.5V
VEXTERNAL = 3.3V
R1 = R2 = 1.5mΩ
VIN = +12V
VOUT = +1.5V
VEXTERNAL = 2.5V
R1 = R2 = 1.5mΩ
REVERSE
CURRENT
10A/div
8
MAX5038A/41 toc36
VIN = +12V
VOUT = +1.5V
VEXTERNAL = 3.3V
R1 = R2 = 1.5mΩ
REVERSE
CURRENT
5A/div
0A
0A
200µs/div
REVERSE-CURRENT SINK
AT ENABLE TURN-ON
MAX5038A/41 toc35
MAX5038A/41A toc34
200µs/div
200µs/div
_______________________________________________________________________________________
REVERSE
CURRENT
10A/div
0A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
PIN
NAME
FUNCTION
1, 13
CSP2,
CSP1
Current-Sense Differential Amplifier Positive Input. Senses the inductor current. The differential voltage
between CSP_ and CSN_ is amplified internally by the current-sense amplifier gain of 18.
2, 14
CSN2,
CSN1
Current-Sense Differential Amplifier Negative Input. Together with CSP_, senses the inductor current.
3
PHASE
Phase-Shift Setting Input. Connect PHASE to VCC for 120°, leave PHASE unconnected for 90°, or connect
PHASE to SGND for 60° of phase shift between the rising edges of CLKOUT and CLKIN/DH1.
4
PLLCMP
External Loop-Compensation Input. Connect compensation network for the phase-locked loop (see PhaseLocked Loop section).
5, 7
CLP2,
CLP1
Current-Error Amplifier Output. Compensate the current loop by connecting an RC network to ground.
6
SGND
Signal Ground. Ground connection for the internal control circuitry.
8
SENSE+
Differential Output Voltage-Sensing Positive Input. Used to sense a remote load. Connect SENSE+ to
VOUT+ at the load. The MAX5038A regulates the difference between SENSE+ and SENSE- according to the
factory preset output voltage. The MAX5041A regulates the SENSE+ to SENSE- difference to +1.0V.
9
SENSE-
Differential Output Voltage-Sensing Negative Input. Used to sense a remote load. Connect SENSE- to
VOUT- or PGND at the load.
10
DIFF
Differential Remote-Sense Amplifier Output. DIFF is the output of a precision unity-gain amplifier.
11
EAN
Voltage-Error Amplifier Inverting Input. Receives the output of the differential remote-sense amplifier.
Referenced to SGND.
12
EAOUT
15
EN
16, 26
BST1,
BST2
Boost Flying-Capacitor Connection. Reservoir capacitor connection for the high-side FET driver supply.
Connect a 0.47µF ceramic capacitor between BST_ and LX_.
17, 25
DH1,
DH2
High-Side Gate Driver Output. Drives the gate of the high-side MOSFET.
18, 24
LX1, LX2
19, 23
DL1,
DL2
Low-Side Gate Driver Output. Synchronous MOSFET gate drivers for the two phases.
20
VCC
Internal +5V Regulator Output. VCC is derived internally from the IN voltage. Bypass to SGND with 4.7µF
and 0.1µF ceramic capacitors.
21
IN
22
PGND
27
CLKOUT
Oscillator Output. CLKOUT is phase shifted from CLKIN by the amount specified by PHASE. Use CLKOUT
to parallel additional MAX5038A/MAX5041As.
28
CLKIN
CMOS Logic Clock Input. Drive the internal oscillator with a frequency range between 125kHz and 600kHz,
or connect to VCC or SGND. Connect CLKIN to SGND to set the internal oscillator to 250kHz or connect to
VCC to set the internal oscillator to 500kHz. CLKIN has an internal 5µA pulldown current.
Voltage-Error Amplifier Output. Connect to the external gain-setting feedback resistor. The external error
amplifier gain-setting resistors determine the amount of adaptive voltage positioning
Output Enable. A logic low shuts down the power drivers. EN has an internal 5µA pullup current.
Inductor Connection. Source connection for the high-side MOSFETs. Also serves as the return terminal for
the high-side driver.
Supply Voltage Connection. Connect IN to VCC for a +5V system. Connect the VRM input to IN through an
RC lowpass filter, a 2.2Ω resistor, and a 0.1µF ceramic capacitor.
Power Ground. Connect PGND, low-side synchronous MOSFET’s source, and VCC bypass capacitor returns
together.
_______________________________________________________________________________________
9
MAX5038A/MAX5041A
Pin Description
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
MAX5038A/MAX5041A
Functional Diagram
EN
IN
+5V
LDO
REGULATOR
UVLO
POR
TEMP SENSOR
VCC
TO INTERNAL CIRCUITS
CSP1
CSN1
CSP1
DRV_VCC
SHDN
CSN1
CLP1
DH1
CLP1
CLK
SGND
MAX5038A
MAX5041A
PHASE
LX1
PHASE 1
DL1
GMIN
PGND
PHASELOCKED
LOOP
CLKIN
BST1
RAMP1
CLKOUT
PLLCMP
RAMP
GENERATOR
DIFF
SENSE-
0.6V
DIFF
AMP
PGND
SENSE+
EAOUT
EAN
ERROR
AMP
DRV_VCC
VREF = VOUT for VOUT ≤ 1.8V (MAX5038A)
VREF = VOUT/2 for VOUT > 1.8V (MAX5038A)
VREF = +1.0V (MAX5041A)
CSN2
CSP2
10
PGND
RAMP2
GMIN
CLP2
SHDN
CLK
PHASE 2
DH2
LX2
CLP2
CSN2
CSP2
______________________________________________________________________________________
DL2
BST2
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
The MAX5038A/MAX5041A (Figures 1 and 2) averagecurrent-mode PWM controllers drive two out-of-phase
buck converter channels. Average-current-mode control improves current sharing between the channels
while minimizing component derating and size. Parallel
multiple MAX5038A/MAX5041A regulators to increase
SENSESENSE+
3 PHASE
CSN1
CSP1
9
8
14
13
VIN
15 EN
R1
VIN = +12V
C3–C7
21 IN
C1, C2
C39
DH1
LX1
VCC
28
DL1
17
Q1
R2
L1
18
19
C12
Q2
CLKIN
D1
MAX5038A
BST1 16
4
PLLCMP
R4
C25
D3
+1.8V AT 60A
VOUT
VCC 20
C34
C32
C26
VCC
R7
RX
10
11
R8
12
DIFF
VIN
C31
D4
C8–C11
EAOUT
DH2 25
R6
7
CLP1
DL2 23
C16–C24, LOAD
C33
Q1
L2
LX2 24
C29
C14,
C15
EAN
R3
C13
Q2
D2
C30
5
C28
C27
BST2 26
CLP2
R5
1
6
22
SGND
PGND
CSP2
CSN2
2
NOTE: SEE TABLE 1 FOR COMPONENT VALUES.
Figure 1. MAX5038A Typical Application Circuit, VIN = +12V
______________________________________________________________________________________
11
MAX5038A/MAX5041A
the output current capacity. For maximum ripple rejection at the input, set the phase shift between phases to
90° for two paralleled converters, or 60° for three paralleled converters. The paralleling capability of the
MAX5038A/MAX5041A improves design flexibility in
applications requiring upgrades (higher load).
Detailed Description
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
Dual-phase converters with an out-of-phase locking
arrangement reduce the input and output capacitor
ripple current, effectively multiplying the switching frequency by the number of phases. Each phase of the
MAX5038A/MAX5041A consists of an inner average
current loop controlled by a common outer-loop volt-
SENSESENSE+
3
CSN1
PHASE
CSP1
age-error amplifier (VEA). The combined action of the
two inner current loops and the outer voltage loop corrects the output voltage errors and forces the phase
currents to be equal.
9
8
14
13
VIN
15 EN
R1
VIN = +12V
C3–C7
21 IN
C1,
C2
C39
DH1
LX1
VCC
28
DL1
17
Q1
L1
18
19
C12
Q2
CLKIN
R2
D1
MAX5041A
BST1 16
4
PLLCMP
R4
C25
D3
+1.8V AT 60A
VOUT
VCC 20
C34
C32
C31
RH
C26
R7
VCC
10
11
R8
RX
12
DIFF
D4
VIN C8–C11
EAN
EAOUT
R6
7
CLP1
C16–C24, LOAD
C33
RL
DH2 25
Q1
L2
LX2 24
C29
C14,
C15
DL2 23
R3
C13
Q2
D2
C30
5
C28
C27
BST2 26
CLP2
R5
1
6
22
SGND
PGND
CSP2
CSN2
2
NOTE: SEE TABLE 1 FOR COMPONENT VALUES.
Figure 2. MAX5041A Typical Application Circuit, VIN = +12V
12
______________________________________________________________________________________
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
Internal Oscillator
The internal oscillator generates the 180° out-of-phase
clock signals required by the pulse-width modulation
(PWM) circuits. The oscillator also generates the 2VP-P
voltage ramp signals necessary for the PWM comparators. Connect CLKIN to SGND to set the internal oscillator
frequency to 250kHz or connect CLKIN to VCC to set the
internal oscillator to 500kHz.
CLKIN is a CMOS logic clock input for the phaselocked loop (PLL). When driven externally, the internal
oscillator locks to the signal at CLKIN. A rising edge at
CLKIN starts the ON cycle of the PWM. Ensure that the
external clock pulse width is at least 200ns. CLKOUT
provides a phase-shifted output with respect to the rising edge of the signal at CLKIN. PHASE sets the
amount of phase shift at CLKOUT. Connect PHASE to
VCC for 120° of phase shift, leave PHASE unconnected
for 90° of phase shift, or connect PHASE to SGND for
60° of phase shift with respect to CLKIN.
The MAX5038A/MAX5041A require compensation on
PLLCMP even when operating from the internal oscillator.
The device requires an active PLL in order to generate
the proper clock signal required for PWM operation.
ICC = IQ + fSW x (QG1 + QG2 + QG3 + QG4) (2)
where Q G1 , Q G2 , Q G3, and Q G4 are the total gate
charge of the low-side and high-side external
MOSFETs, IQ is 4mA (typ), and fSW is the switching frequency of each individual phase.
For applications utilizing a +5V input voltage, disable
the VCC regulator by connecting IN and VCC together.
Undervoltage Lockout (UVLO)/Soft-Start
The MAX5038A/MAX5041A include an undervoltage
lockout with hysteresis and a power-on reset circuit for
converter turn-on and monotonic rise of the output voltage. The UVLO threshold is internally set between
+4.0V and +4.5V with a 200mV hysteresis. Hysteresis
at UVLO eliminates “chattering” during startup.
Most of the internal circuitry, including the oscillator,
turns on when the input voltage reaches +4V. The
MAX5038A/MAX5041A draw up to 4mA of current before
the input voltage reaches the UVLO threshold.
The compensation network at the current-error amplifiers (CLP1 and CLP2) provides an inherent soft-start of
the output voltage. It includes a parallel combination of
capacitors (C28, C30) and resistors (R5, R6) in series
with other capacitors (C27, C29) (see Figures 1 and 2).
The voltage at CLP_ limits the maximum current available to charge output capacitors. The capacitor on
CLP_ in conjunction with the finite output-drive current
of the current-error amplifier yields a finite rise time for
the output current and thus the output voltage.
Control Loop
The MAX5038A/MAX5041A use an average-currentmode control scheme to regulate the output voltage
(Figures 3a and 3b). The main control loop consists of
an inner current loop and an outer voltage loop. The
inner loop controls the output currents (IPHASE1 and
IPHASE2) while the outer loop controls the output voltage. The inner current loop absorbs the inductor pole
reducing the order of the outer voltage loop to that of a
single-pole system.
The current loop consists of a current-sense resistor
(RS), a current-sense amplifier (CA_), a current-error
amplifier (CEA_), an oscillator providing the carrier
ramp, and a PWM comparator (CPWM_). The precision
CA_ amplifies the sense voltage across RS by a factor
of 18. The inverting input to the CEA_ senses the CA_
output. The CEA_ output is the difference between the
voltage-error amplifier output (EAOUT) and the amplified voltage from the CA_. The RC compensation network connected to CLP1 and CLP2 provides external
frequency compensation for the respective CEA_. The
start of every clock cycle enables the high-side drivers
and initiates a PWM ON cycle. Comparator CPWM_
compares the output voltage from the CEA_ with a 0 to
+2V ramp from the oscillator. The PWM ON cycle terminates when the ramp voltage exceeds the error voltage.
______________________________________________________________________________________
13
MAX5038A/MAX5041A
VIN and VCC
The MAX5038A/MAX5041A accept an input voltage
range of +4.75V to +5.5V or +8V to +28V. All internal
control circuitry operates from an internally regulated
nominal voltage of +5V (VCC). For input voltages of +8V
or greater, the internal VCC regulator steps the voltage
down to +5V. The VCC output voltage is a regulated +5V
output capable of sourcing up to 80mA. Bypass VCC to
SGND with 4.7µF and 0.1µF low-ESR ceramic capacitors
in parallel for high-frequency noise rejection and stable
operation (Figures 1 and 2).
Calculate power dissipation in the MAX5038A/
MAX5041A as a product of the input voltage and the
total VCC regulator output current (ICC). ICC includes
quiescent current (IQ) and gate drive current (IDD):
PD = VIN x ICC
(1)
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
CCF
CLP1
CSP1
CSN1
RCF
MAX5038A
CCFF
CA1
VIN
RF*
IPHASE1
CEA1
SENSE+
DRIVE 1
CPWM1
RS
RIN*
DIFF
AMP
VEA
SENSE-
VOUT
VIN
COUT
CEA2
VREF
CPWM2
IPHASE2
DRIVE 2
RS
LOAD
CLP2
CSN2
CSP2
CA2
CCF
RCF
*RF AND RIN ARE EXTERNAL TO MAX5038A
(RF = R8, RIN = R7, FIGURE 1).
CCCF
Figure 3a. MAX5038A Control Loop
CCF
MAX5041A
CLP1
CSP1
CSN1
RCF
CCFF
CA1
RF*
VIN
IPHASE1
CEA1
SENSE+
CPWM1
DIFF
AMP
DRIVE 1
RS
RIN*
VEA
SENSE-
VOUT
VIN
CEA2
VREF =
+1.0V
CPWM2
DRIVE 2
IPHASE2
RS
COUT
LOAD
CLP2
CSP2
CSN2
CA2
CCF
RCF
*RF AND RIN ARE
EXTERNAL TO MAX5041A
(RF = R8, RIN = R7, FIGURE 2).
CCCF
Figure 3b. MAX5041A Control Loop
14
______________________________________________________________________________________
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
than the average current limit (48mV). Proper inductor
selection ensures that only extreme conditions trip the
peak-current comparator, such as a broken output
inductor. The 112mV voltage threshold for triggering
the peak-current limit is twice the full-scale average
current-limit voltage threshold. The peak-current comparator has a delay of only 260ns.
Current-Error Amplifier
Current-Sense Amplifier
The differential current-sense amplifier (CA_) provides a
DC gain of 18. The maximum input offset voltage of the
current-sense amplifier is 1mV and the common-mode
voltage range is -0.3V to +3.6V. The current-sense amplifier senses the voltage across a current-sense resistor.
Each phase of the MAX5038A/MAX5041A has a dedicated transconductance current-error amplifier (CEA_)
with a typical gm of 550µS and 320µA output sink and
source current capability. The current-error amplifier
outputs, CLP1 and CLP2, serve as the inverting input to
the PWM comparator. CLP1 and CLP2 are externally
accessible to provide frequency compensation for the
inner current loops (Figures 3a and 3b). Compensate
CEA_ such that the inductor current down slope, which
becomes the up slope to the inverting input of the PWM
comparator, is less than the slope of the internally generated voltage ramp (see the Compensation section).
Peak-Current Comparator
The peak-current comparator provides a path for fast
cycle-by-cycle current limit during extreme fault conditions such as an output inductor malfunction (Figure 4).
Note that the average-current-limit threshold of 48mV
still limits the output current during short-circuit conditions. To prevent inductor saturation, select an output
inductor with a saturation current specification greater
PWM Comparator and R-S Flip-Flop
The PWM comparator (CPWM) sets the duty cycle for
each cycle by comparing the output of the current-error
amplifier to a 2VP-P ramp. At the start of each clock
cycle, an R-S flip-flop resets and the high-side driver
(DH_) turns on. The comparator sets the flip-flop as
soon as the ramp voltage exceeds the CLP_ voltage,
thus terminating the ON cycle (Figure 4).
DRV_VCC
PEAK-CURRENT
COMPARATOR
112mV
CLP_
CSP_
AV = 18
Gm =
550µS
CSN_
BST_
PWM
COMPARATOR
GMIN
S
Q
DH_
RAMP
LX_
2 x fs (V/s)
CLK
R
Q
DL_
PGND
SHDN
Figure 4. Phase Circuit (Phase 1/Phase 2)
______________________________________________________________________________________
15
MAX5038A/MAX5041A
The outer voltage control loop consists of the differential amplifier (DIFF AMP), reference voltage, and VEA.
The unity-gain differential amplifier provides true differential remote sensing of the output voltage. The differential amplifier output connects to the inverting input
(EAN) of the VEA. The noninverting input of the VEA is
internally connected to an internal precision reference
voltage. The MAX5041A reference voltage is set to
+1.0V and the MAX5038A reference is set to the preset
output voltage. The VEA controls the two inner current
loops (Figures 3a and 3b). Use a resistive feedback
network to set the VEA gain as required by the adaptive
voltage-positioning circuit (see the Adaptive Voltage
Positioning section).
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
Differential Amplifier
The differential amplifier (DIFF AMP) facilitates output
voltage remote sensing at the load (Figures 3a and 3b).
It provides true differential output voltage sensing while
rejecting the common-mode voltage errors due to highcurrent ground paths. Sensing the output voltage
directly at the load provides accurate load voltage
sensing in high-current environments. The VEA provides the difference between the differential amplifier
output (DIFF) and the desired output voltage. The differential amplifier has a bandwidth of 3MHz. The difference between SENSE+ and SENSE- regulates to the
preset output voltage for the MAX5038A and regulates
to +1V for the MAX5041A.
Voltage-Error Amplifier
The VEA sets the gain of the voltage control loop and
determines the error between the differential amplifier
output and the internal reference voltage (VREF).
VREF equals VOUT(NOM) for the +1.8V or lower voltage
versions of the MAX5038A and VREF equals VOUT(NOM)/2
for the +2.5V and +3.3V versions. For MAX5041A, VREF
equals +1V.
An offset is added to the output voltage of the
MAX5038A/MAX5041A with a finite gain (RF/RIN) of the
VEA such that the no-load output voltage is higher than
the nominal value. Choose R F and R IN from the
Adaptive Voltage Positioning section and use the following equations to calculate the no-load output voltage.
MAX5038A:
 R 
VOUT(NL) = 1 + IN  × VOUT(NOM )
RF 

(3)
MAX5041A:
 R  R +R 
L ×V
VOUT(NL) = 1 + IN  ×  H

REF
RF   RL 

(4)
where RH and RL are the feedback resistor network
(Figure 2).
Some applications require VOUT equal to VOUT(NOM) at
no load. To ensure that the output voltage does not
exceed the nominal output voltage (VOUT(NOM)), add a
resistor RX from VCC to EAN.
16
Use the following equations to calculate the value of RX.
For MAX5038A versions of VOUT(NOM) ≤ +1.8V:
RX = [VCC − (VNOM + 0.6)] ×
RF
VNOM
(5)
For MAX5038A versions of VOUT(NOM) > +1.8V:
RX = [2VCC − (VNOM + 1.2)] ×
RF
VNOM
(6)
For MAX5041A:
RX = [VCC − 1.6] ×
RF
(7)
VREF
The VEA output clamps to +0.9V (plus the commonmode voltage of +0.6V), thus limiting the average maximum current from individual phases. The maximum
average-current-limit threshold for each phase is equal
to the maximum clamp voltage of the VEA divided by
the gain (18) of the current-sense amplifier. This allows
for accurate settings for the average maximum current
for each phase. Set the VEA gain using RF and RIN for
the amount of output voltage positioning required as
discussed in the Adaptive Voltage Positioning section
(Figures 3a and 3b).
Adaptive Voltage Positioning
Powering new-generation processors requires new
techniques to reduce cost, size, and power dissipation.
Voltage positioning reduces the total number of output
capacitors to meet a given transient response requirement. Setting the no-load output voltage slightly higher
than the output voltage during nominally loaded conditions allows a larger downward voltage excursion when
the output current suddenly increases. Regulating at a
lower output voltage under a heavy load allows a larger
upward-voltage excursion when the output current suddenly decreases. A larger allowed, voltage-step excursion reduces the required number of output capacitors
or allows for the use of higher ESR capacitors.
Voltage positioning and the ability to operate with multiple
reference voltages may require the output to regulate
away from a center value. Define the center value as the
voltage where the output drops (∆VOUT/2) at one half the
maximum output current (Figure 5).
______________________________________________________________________________________
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
VOLTAGE-POSITIONING WINDOW
VCNTR + ∆VOUT/2
VCNTR
VCNTR - ∆VOUT/2
1/2 LOAD
NO LOAD
FULL LOAD
LOAD (A)
Figure 5. Defining the Voltage-Positioning Window
Set the voltage-positioning window (∆VOUT) using the
resistive feedback of the VEA. Use the following equations to calculate the voltage-positioning window for the
MAX5038A:
I
× RIN
∆VOUT = OUT
2 × GC × RF
GC =
0.05
(8)
(9)
RS
Use the following equation to calculate the voltage-positioning window for the MAX5041A:
∆VOUT =
IOUT × RIN
R + RL
× H
G
R
2
×
×
( C F ) RL
GC =
0.05
(10)
(11)
RS
where RIN and RF are the input and feedback resistors of
the VEA, GC is the current-loop transconductance, and
RS is the current-sense resistor or, if using lossless inductor current sensing, the DC resistance of the inductor.
The PLL synchronizes the internal oscillator to the external frequency source when driving CLKIN. Connecting
CLKIN to VCC or SGND forces the PWM frequency to
default to the internal oscillator frequency of 500kHz or
250kHz, respectively. The PLL uses a conventional
architecture consisting of a phase detector and a
charge pump capable of providing 20µA of output current. Connect an external series combination capacitor
(C25) and resistor (R4) and a parallel capacitor (C26)
from PLLCMP to SGND to provide frequency compensation for the PLL (Figure 1). The pole-zero pair compensation provides a zero at fZ defined by 1 / [R4 x
(C25 + C26)] and a pole at fP defined by 1 / (R4 x C26).
Use the following typical values for compensating the
PLL: R4 = 7.5kΩ, C25 = 4.7nF, C26 = 470pF. If changing the PLL frequency, expect a finite locking time of
approximately 200µs.
The MAX5038A/MAX5041A require compensation on
PLLCMP even when operating from the internal oscillator. The device requires an active PLL in order to generate the proper internal PWM clocks.
MOSFET Gate Drivers (DH_, DL_)
The high-side (DH_) and low-side (DL_) drivers drive
the gates of external N-channel MOSFETs (Figures 1
and 2). The drivers’ high-peak sink and source current
capability provides ample drive for the fast rise and fall
times of the switching MOSFETs. Faster rise and fall
times result in reduced cross-conduction losses. For
modern CPU voltage-regulating module applications
where the duty cycle is less than 50%, choose highside MOSFETs (Q1 and Q3) with a moderate RDS(ON)
and a very low gate charge. Choose low-side MOSFETs
(Q2 and Q4) with very low RDS(ON) and moderate gate
charge.
The driver block also includes a logic circuit that provides an adaptive nonoverlap time to prevent shootthrough currents during transition. The typical
nonoverlap time is 60ns between the high-side and lowside MOSFETs.
BST_
VCC powers the low- and high-side MOSFET drivers.
Connect a 0.47µF low-ESR ceramic capacitor between
BST_ and LX_. Bypass VCC to SGND with 4.7µF and
0.1µF low-ESR ceramic capacitors. For high-current
applications, bypass VCC to PGND with one or more
0.1µF, low-ESR ceramic capacitor(s). Reduce the PC
board area formed by these capacitors, the rectifier
diodes between V CC and the boost capacitor, the
MAX5038A/MAX5041A, and the switching MOSFETs.
______________________________________________________________________________________
17
MAX5038A/MAX5041A
Phase-Locked Loop: Operation and
Compensation
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
Overload Conditions
Average-current-mode control has the ability to limit the
average current sourced by the converter during a fault
condition. When a fault condition occurs, the VEA output clamps to +0.9V with respect to the common-mode
voltage (VCM = +0.6V) and is compared with the output
of the current-sense amplifiers (CA1 and CA2) (see
Figures 3a and 3b). The current-sense amplifier’s gain
of 18 limits the maximum current in the inductor or
sense resistor to ILIMIT = 50mV/RS.
Parallel Operation
For applications requiring large output current, parallel
up to three MAX5038A/MAX5041As (six phases) to triple
the available output current. The paralleled converters
operate at the same switching frequency but different
phases keep the capacitor ripple RMS currents to a minimum. Three parallel MAX5038A/MAX5041A converters
deliver up to 180A of output current. To set the phase
shift of the on-board PLL, leave PHASE unconnected for
90° of phase shift (two paralleled converters), or connect
PHASE to SGND for 60° of phase shift (three converters
in parallel). Designate one converter as master and the
remaining converters as slaves. Connect the master and
slave controllers in a daisy-chain configuration as shown
in Figure 6. Connect CLKOUT from the master controller
to CLKIN of the first slaved controller, and CLKOUT from
the first slaved controller to CLKIN of the second slaved
controller. Choose the appropriate phase shift for minimum ripple currents at the input and output capacitors.
The master controller senses the output differential voltage through SENSE+ and SENSE- and generates the
DIFF voltage. Disable the voltage sensing of the slaved
controllers by leaving DIFF unconnected (floating).
Figure 7 shows a detailed typical parallel application circuit using two MAX5038As. This circuit provides four
phases at an input voltage of +12V and an output voltage range of +1V to +3.3V at 104A.
Applications Information
Each MAX5038A/MAX5041A circuit drives two 180° outof-phase channels. Parallel two or three MAX5038A/
MAX5041A circuits to achieve four- or six-phase operation, respectively. Figure 1 shows the typical application
circuit for a two-phase operation. The design criteria for
a two-phase converter includes frequency selection,
inductor value, input/output capacitance, switching
MOSFETs, sense resistors, and the compensation network. Follow the same procedure for the four- and sixphase converter design, except for the input and output
capacitance. The input and output capacitance requirements vary depending on the operating duty cycle.
18
The examples discussed in this data sheet pertain to a
typical application with the following specifications:
VIN = +12V
VOUT = +1.8V
IOUT(MAX) = 52A
fSW = 250kHz
Peak-to-Peak Inductor Current (∆IL) = 10A
Table 1 shows a list of recommended external components (Figure 1) and Table 2 provides component supplier information.
Number of Phases
Selecting the number of phases for a voltage regulator
depends mainly on the ratio of input-to-output voltage
(operating duty cycle). Optimum output-ripple cancellation depends on the right combination of operating duty
cycle and the number of phases. Use the following
equation as a starting point to choose the number of
phases:
(12)
NPH ≈ K/D
where K = 1, 2, or 3 and the duty cycle is D = VOUT/VIN.
Choose K to make NPH an integer number. For example, converting V IN = +12V to V OUT = +1.8V yields
better ripple cancellation in the six-phase converter
than in the four-phase converter. Ensure that the output
load justifies the greater number of components for
multiphase conversion. Generally limiting the maximum
output current to 25A per phase yields the most costeffective solution. The maximum ripple cancellation
occurs when NPH = K/D.
Single-phase conversion requires greater size and power
dissipation for external components such as the switching MOSFETs and the inductor. Multiphase conversion
eliminates the heatsink by distributing the power dissipation in the external components. The multiple phases
operating at given phase shifts effectively increase the
switching frequency seen by the input/output capacitors,
thereby reducing the input/output capacitance requirement for the same ripple performance. The lower inductance value improves the large-signal response of the
converter during a transient load at the output. Consider
all these issues when determining the number of phases
necessary for the voltage regulator application.
Inductor Selection
The switching frequency per phase, peak-to-peak ripple current in each phase, and allowable ripple at the
output determine the inductance value.
______________________________________________________________________________________
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
MAX5038A/MAX5041A
CSN1
CSP1
SENSE+
VIN
SENSEDH1
VCC
LX1
PHASE
DL1
VCC
CLKIN
VIN
VIN
MAX5038A/
MAX5041A
DH2
LX2
IN
DL2
DIFF
EAN
CSP2
CSN2
EAOUT
PGND SGND CLKOUT
CSN1
CSP1
CLKIN
VIN
DH1
VCC
LX1
PHASE
DL1
MAX5038A/
MAX5041A
IN
VIN
DH2
DIFF
*
LX2
LOAD
*
DL2
EAN
CSP2
EAOUT
CSN2
PGND SGND CLKOUT
CSN1
CSP1
CLKIN
VIN
DH1
VCC
LX1
PHASE
DL1
MAX5038A/
MAX5041A
IN
VIN
DH2
DIFF
LX2
DL2
EAN
EAOUT
*FOR MAX5041A ONLY.
CSP2
CSN2
PGND SGND CLKOUT
TO OTHER MAX5038A/MAX5041As
Figure 6. Parallel Configuration of Multiple MAX5038A/MAX5041As
______________________________________________________________________________________
19
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
VIN = +12V
C1, C2
2 x 47µF
VCC
C31
R3
2.2Ω
C32
C39
0.1µF
R4
PLLCMP
CLKIN
IN
VIN
C3–C7
5 x 22µF
SENSE- SENSE+ CSN1 CSP1
DH1
Q1
L1
0.6µH
R1
1.35mΩ
LX1
DL1
C12
0.47µF
Q2
D1
BST1
VCC
VCC
EN
D3
C38
4.7µF
C40
0.1µF
D4
OVPIN
VCC
R7
VIN
MAX5038A
(MASTER)
DIFF
4 x 22µF
C8–C11
EAN
RX
DH2
R8
Q3
L2
0.6µH
EAOUT
R2
1.35mΩ
LX2
DL2
Q4
D2
CLP1
CLP2 PGND SGND CLKOUT PHASE PGOOD CSN2 CSP2
R6
C13
0.47µF
BST2
R9
R5
C36
C34
C35
PGOOD
VCC
C33
R18
C26–C30,
C37
6 x 10µF LOAD
C14, C15,
C41, C42
2 x 100µF
C62
R12
2.2Ω
C63
C47
0.1µF
R13
EN
C16–C25,
C43–C46
14 x 270µF
PLLCMP
IN
VIN
C48–C51
5 x 22µF
CLKIN SENSE-SENSE+ CSN1 CSP1
DH1
L3
0.6µH
Q5
R10
1.35mΩ
LX1
DL1
C57
0.47µF
Q6
D5
BST1
D7
VCC
C65
4.7µF
C64
0.1µF
D8
MAX5038A
(SLAVE)
R16
VCC
VIN
C52–C55
4 x 22µF
EAN
RX
DH2
R17
EAOUT
L4
0.6µH
Q7
R11
1.35mΩ
LX2
DIFF
DL2
Q8
D6
CLP1
CLP2
R15
R14
C59
C60
C58
PGND
C61
SGND
PHASE
CSN2 CSP2
C56
0.47µF
BST2
VCC
Figure 7. Four-Phase Parallel Application Circuit (VIN = +12V, VOUT = +1.2V to +3.3V at 104A)
20
______________________________________________________________________________________
VOUT = +1.2V TO
+3.3V AT 104A
R19
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
MAX5038A/MAX5041A
Table 1. Component List
DESIGNATION
QTY
C1, C2
2
47µF,16V X5R input-filter capacitors, TDK C5750X5R1C476M
DESCRIPTION
C3–C11
9
22µF, 16V input-filter capacitors, TDK C4532X5R1C226M
C12, C13
2
0.47µF, 16V capacitors, TDK C1608X5R1A474K
C14, C15
2
100µF, 6.3V output-filter capacitors, Murata GRM44-1X5R107K6.3
C16–C24, C33
10
270µF, 2V output-filter capacitors, Panasonic EEFUE0D271R
C25
1
4700pF, 16V X7R capacitor, Vishay-Siliconix VJ0603Y471JXJ
C26, C28, C30
3
470pF, 16V capacitors, Murata GRM1885C1H471JAB01
C27, C29
2
0.01µF, 50V X7R capacitors, Murata GRM188R71H103KA01
C31
1
4.7µF, 16V X5R capacitor, Murata GRM40-034X5R475k6.3
C32, C34, C39
3
0.1µF, 16V X7R capacitors, Murata GRM188R71C104KA01
D1, D2
2
Schottky diodes, ON Semiconductor MBRS340T3
D3, D4
2
Schottky diodes, ON Semiconductor MBR0520LT1
L1, L2
2
0.6µH, 27A inductors, Panasonic ETQP1H0R6BFX
Q1, Q3
2
Upper power MOSFETs, Vishay-Siliconix Si7860DP
Q2, Q4
2
Lower power MOSFETs, Vishay-Siliconix Si7886DP
R1
1
2.2Ω ±1% resistor
R2, R3
4
Current-sense resistors, use two 2.7mΩ resistors in parallel, Panasonic ERJM1WSF2M7U
R4
1
7.5kΩ ±1% resistor
R5, R6
2
1kΩ ±1% resistors
R7
1
4.99kΩ ±1% resistor
R8, R9
2
37.4kΩ ±1% resistors
Table 2. Component Suppliers
SUPPLIER
PHONE
FAX
WEBSITE
Murata
770-436-1300
770-436-3030
www.murata.com
ON Semiconductor
602-244-6600
602-244-3345
www.on-semi.com
Panasonic
714-373-7939
714-373-7183
www.panasonic.com
TDK
Vishay-Siliconix
847-803-6100
847-390-4405
www.tcs.tdk.com
1-800-551-6933
619-474-8920
www.vishay.com
______________________________________________________________________________________
21
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
Selecting higher switching frequencies reduces the
inductance requirement, but at the cost of lower efficiency. The charge/discharge cycle of the gate and drain
capacitances in the switching MOSFETs create switching
losses. The situation worsens at higher input voltages,
since switching losses are proportional to the square of
input voltage. Use 500kHz per phase for VIN = +5V and
250kHz or less per phase for VIN > +12V.
Although lower switching frequencies per phase increase
the peak-to-peak inductor ripple current (∆IL), the ripple
cancellation in the multiphase topology reduces the RMS
ripple current of the input and output capacitors.
Use the following equation to determine the minimum
inductance value:
LMIN =
(VINMAX − VOUT ) × VOUT
VIN × fSW × ∆IL
(13)
Choose ∆IL equal to about 40% of the output current
per phase. Since ∆IL affects the output-ripple voltage,
the inductance value may need minor adjustment after
choosing the output capacitors for full-rated efficiency.
Choose inductors from the standard high-current,
surface-mount inductor series available from various
manufacturers. Particular applications may require custom-made inductors. Use high-frequency core material
for custom inductors. High ∆IL causes large peak-to-peak
flux excursion increasing the core losses at higher frequencies. The high-frequency operation coupled with
high ∆IL, reduces the required minimum inductance
and even makes the use of planar inductors possible.
The advantages of using planar magnetics include lowprofile design, excellent current-sharing between phases due to the tight control of parasitics, and low cost.
For example, calculate the minimum inductance at
VIN(MAX) = +13.2V, VOUT = +1.8V, ∆IL = 10A, and fSW =
250kHz:
LMIN =
(13.2 − 1.8) × 1.8 = 0.6µH
13.2 × 250k × 10
(14)
The average-current-mode control feature of the
MAX5038A/MAX5041A limits the maximum peak inductor current and prevents the inductor from saturating.
Choose an inductor with a saturating current greater
than the worst-case peak inductor current.
22
Use the following equation to determine the worst-case
inductor current for each phase:
IL _ PEAK =
∆I
0.051
+ L
RSENSE
2
(15)
where RSENSE is the sense resistor in each phase.
Switching MOSFETs
When choosing a MOSFET for voltage regulators,
consider the total gate charge, RDS(ON), power dissipation, and package thermal impedance. The product of
the MOSFET gate charge and on-resistance is a figure of
merit, with a lower number signifying better performance.
Choose MOSFETs optimized for high-frequency switching applications.
The average current from the MAX5038A/MAX5041A
gate-drive output is proportional to the total capacitance
it drives from DH1, DH2, DL1, and DL2. The power dissipated in the MAX5038A/MAX5041A is proportional to the
input voltage and the average drive current. See the VIN
and VCC section to determine the maximum total gate
charge allowed from all the driver outputs combined.
The gate charge and drain capacitance (CV2) loss, the
cross-conduction loss in the upper MOSFET due to finite
rise/fall time, and the I2R loss due to RMS current in the
MOSFET RDS(ON) account for the total losses in the MOSFET. Estimate the power loss (PDMOS_) in the high-side
and low-side MOSFETs using following equations:
PDMOS − HI = (QG × VDD × fSW ) +
(16)
 VIN × IOUT × ( t R + t F ) × fSW 
2

 + 1.4RDS(ON) × I RMS − HI
4


where QG, RDS(ON), tR, and tF are the upper-switching
MOSFET’s total gate charge, on-resistance at +25°C,
rise time, and fall time, respectively:
IRMS−HI =
(I
)
D
2
2
DC + I PK + IDC × IPK ×
3
(17)
where D = V OUT /V IN , I DC = (I OUT - ∆I L )/2, and I PK =
(IOUT + ∆IL)/2.
______________________________________________________________________________________
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
Input Capacitors
2
2×C

(18)
2
OSS × VIN × fSW + 1.4R


DS(ON) × I RMS − LO
3


IRMS−LO =
(I
)
(
)
1− D
2
2
DC + I PK + IDC × IPK ×
3
(19)
where COSS is the MOSFET drain-to-source capacitance.
For example, from the typical specifications in the
Applications Information section with VOUT = +1.8V, the
high-side and low-side MOSFET RMS currents are 9.9A
and 24.1A, respectively. Ensure that the thermal impedance of the MOSFET package keeps the junction temperature at least 25°C below the absolute maximum
rating. Use the following equation to calculate maximum junction temperature:
TJ = PDMOS x θJ-A + TA
(20)
Table 3. Peak-to-Peak Output Ripple
Current Calculations
NO. OF
PHASES (N)
DUTY CYCLE
(D) (%)
2
< 50
2
> 50
EQUATION FOR ∆IP-P
V (1 − 2D)
∆I = O
L × fSW
∆I =
(VIN − VO )(2D − 1)
L × fSW
V (1− 4D)
∆I = O
L × fSW
4
0 to 25
4
25 to 50
V (1 − 2D)(4D − 1)
∆I = O
2 × D × L × fSW
4
> 50
V (2D − 1)(3 − 4D)
∆I = O
D × L × fSW
6
< 17
V (1− 6D)
∆I = O
L × fSW
The discontinuous input-current waveform of the buck
converter causes large ripple currents in the input
capacitor. The switching frequency, peak inductor current, and the allowable peak-to-peak voltage ripple
reflected back to the source dictate the capacitance
requirement. Increasing the number of phases increases the effective switching frequency and lowers the
peak-to-average current ratio, yielding a lower input
capacitance requirement.
The input ripple comprises ∆VQ (caused by the capacitor discharge) and ∆VESR (caused by the ESR of the
capacitor). Use low-ESR ceramic capacitors with high
ripple-current capability at the input. Assume the contributions from the ESR and capacitor discharge are
equal to 30% and 70%, respectively. Calculate the
input capacitance and ESR required for a specified ripple using the following equations:
ESRIN =
(∆VESR )
 IOUT ∆IL 
+


 N
2 
IOUT
× D(1 − D)
CIN = N
∆VQ × fSW
(21)
(22)
where IOUT is the total output current of the multiphase
converter and N is the number of phases.
For example, at V OUT = +1.8V, the ESR and input
capacitance are calculated for the input peak-to-peak
ripple of 100mV or less yielding an ESR and capacitance value of 1mΩ and 200µF.
Output Capacitors
The worst-case peak-to-peak and capacitor RMS ripple
current, the allowable peak-to-peak output ripple voltage, and the maximum deviation of the output voltage
during step loads determine the capacitance and the
ESR requirements for the output capacitors.
In multiphase converter design, the ripple currents from
the individual phases cancel each other and lower the
ripple current. The degree of ripple cancellation
depends on the operating duty cycle and the number of
phases. Choose the right equation from Table 3 to calculate the peak-to-peak output ripple for a given duty
cycle of two-, four-, and six-phase converters. The maximum ripple cancellation occurs when NPH = K / D.
______________________________________________________________________________________
23
MAX5038A/MAX5041A
PDMOS − LO = (QG × VDD × fSW ) +
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
The allowable deviation of the output voltage during the
fast transient load dictates the output capacitance and
ESR. The output capacitors supply the load step until
the controller responds with a greater duty cycle. The
response time (tRESPONSE) depends on the closed-loop
bandwidth of the converter. The resistive drop across
the capacitor ESR and capacitor discharge causes a
voltage drop during a step load. Use a combination of
SP polymer and ceramic capacitors for better transient
load and ripple/noise performance.
Keep the maximum output voltage deviation less than
or equal to the adaptive voltage-positioning window
(∆VOUT). Assume 50% contribution each from the output capacitance discharge and the ESR drop. Use the
following equations to calculate the required ESR and
capacitance value:
ESROUT =
∆VESR
ISTEP
(23)
I
×t
COUT = STEP RESPONSE
∆VQ
(24)
where I STEP is the load step and t RESPONSE is the
response time of the controller. Controller response
time depends on the control-loop bandwidth.
Current Limit
The average-current-mode control technique of the
MAX5038A/MAX5041A accurately limits the maximum
output current per phase. The MAX5038A/MAX5041A
sense the voltage across the sense resistor and limit
the peak inductor current (IL-PK) accordingly. The ON
cycle terminates when the current-sense voltage reaches 45mV (min). Use the following equation to calculate
maximum current-sense resistor value:
0.045
RSENSE =
IOUT
N
(25)
2.5 × 10−3
RSENSE
(26)
PDR =
where PDR is the power dissipation in sense resistors.
Select 5% lower value of RSENSE to compensate for any
parasitics associated with the PC board. Also, select a
noninductive resistor with the appropriate wattage rating.
Reverse Current Limit
The MAX5038A/MAX5041A limit the reverse current in
the case that VBUS is higher than the preset output voltage setting.
Calculate the maximum reverse current based on VCLR,
the reverse current-limit threshold, and the currentsense resistor:
(27)
IREVERSE =
2 × VCLR
RSENSE
Compensation
The main control loop consists of an inner current loop
and an outer voltage loop. The MAX5038A/MAX5041A
use an average-current-mode control scheme to regulate the output voltage (Figures 3a and 3b). IPHASE1 and
IPHASE2 are the inner average current loops. The VEA
output provides the controlling voltage for these current
sources. The inner current loop absorbs the inductor
pole reducing the order of the outer voltage loop to that
of a single-pole system.
A resistive feedback network around the VEA provides
the best possible response, since there are no capacitors to charge and discharge during large-signal excursions, R F and R IN determine the VEA gain. Use the
following equation to calculate the value for RF:
(28)
IOUT × RIN
RF =
N × GC × ∆VOUT
(29)
0.05
GC =
RS
where GC is the current-loop transconductance and N
is the number of phases.
When designing the current-control loop ensure that the
inductor downslope (when it becomes an upslope at the
CEA output) does not exceed the ramp slope. This is a
necessary condition to avoid subharmonic oscillations
similar to those in peak-current-mode control with insufficient slope compensation. Use the following equation
to calculate the resistor RCF:
(30)
2 × fSW × L × 102
RCF ≤
VOUT × RSENSE
For example, the maximum RCF is 12kΩ for RSENSE =
1.35mΩ.
24
______________________________________________________________________________________
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
CCF =
1
2 × π × fZ × RCF
Pin Configuration
TOP VIEW
CSP2 1
28 CLKIN
CSN2 2
27 CLKOUT
26 BST2
PHASE 3
(31)
25 DH2
PLLCMP 4
24 LX2
CLP2 5
1
CCFF =
2 × π × fP × RCF
(32)
SGND 6
MAX5038A
MAX5041A
23 DL2
22 PGND
CLP1 7
PC Board Layout
SENSE+ 8
21 IN
Use the following guidelines to lay out the switching
voltage regulator:
SENSE- 9
20 VCC
1) Place the VIN and VCC bypass capacitors close to
the MAX5038A/MAX5041A.
DIFF 10
19 DL1
EAN 11
18 LX1
EAOUT 12
17 DH1
CSP1 13
16 BST1
CSN1 14
15 EN
2) Minimize the area and length of the high-current
loops from the input capacitor, upper switching
MOSFET, inductor, and output capacitor back to
the input capacitor negative terminal.
3) Keep short the current loop formed by the lower
switching MOSFET, inductor, and output capacitor.
4) Place the Schottky diodes close to the lower
MOSFETs and on the same side of the PC board.
5) Keep the SGND and PGND isolated and connect
them at one single point close to the negative terminal of the input filter capacitor.
6) Run the current-sense lines CS+ and CS- very
close to each other to minimize the loop area.
Similarly, run the remote voltage sense lines
SENSE+ and SENSE- close to each other. Do not
cross these critical signal lines through power circuitry. Sense the current right at the pads of the
current-sense resistors.
7) Avoid long traces between the VCC bypass capacitors, driver output of the MAX5038A/MAX5041A,
MOSFET gates, and PGND. Minimize the loop
formed by the VCC bypass capacitors, bootstrap
diode, bootstrap capacitor, MAX5038A/MAX5041A,
and upper MOSFET gate.
SSOP
9) Distribute the power components evenly across the
board for proper heat dissipation.
10) Provide enough copper area at and around the
switching MOSFETs, inductor, and sense resistors
to aid in thermal dissipation.
11) Use 4oz copper to keep the trace inductance and
resistance to a minimum. Thin copper PC boards
can compromise efficiency since high currents are
involved in the application. Also, thicker copper
conducts heat more effectively, thereby reducing
thermal impedance.
Chip Information
TRANSISTOR COUNT: 5431
PROCESS: BiCMOS
8) Place the bank of output capacitors close to the load.
______________________________________________________________________________________
25
MAX5038A/MAX5041A
CCF provides a low-frequency pole while RCF provides a
midband zero. Place a zero (fZ) to obtain a phase bump
at the crossover frequency. Place a high-frequency pole
(fP) at least a decade away from the crossover frequency
to reduce the influence of the switching noise and
achieve maximum phase margin. Use the following
equations to calculate CCF and CCFF:
Package Information
(The package drawing(s) in this data sheet may not reflect the most current specifications. For the latest package outline information,
go to www.maxim-ic.com/packages.
2
SSOP.EPS
MAX5038A/MAX5041A
Dual-Phase, Parallelable, Average-Current-Mode
Controllers
1
INCHES
E
H
MILLIMETERS
DIM
MIN
MAX
MIN
MAX
A
0.068
0.078
1.73
1.99
A1
0.002
0.008
0.05
0.21
B
0.010
0.015
0.25
0.38
C
D
0.20
0.09
0.004 0.008
SEE VARIATIONS
E
0.205
e
0.212
0.0256 BSC
5.20
MILLIMETERS
INCHES
D
D
D
D
D
5.38
MIN
MAX
MIN
MAX
0.239
0.239
0.278
0.249
0.249
0.289
6.07
6.07
7.07
6.33
6.33
7.33
0.317
0.397
0.328
0.407
8.07
10.07
8.33
10.33
N
14L
16L
20L
24L
28L
0.65 BSC
H
0.301
0.311
7.65
7.90
L
0.025
0∞
0.037
8∞
0.63
0∞
0.95
8∞
N
A
C
B
e
L
A1
D
NOTES:
1. D&E DO NOT INCLUDE MOLD FLASH.
2. MOLD FLASH OR PROTRUSIONS NOT TO EXCEED .15 MM (.006").
3. CONTROLLING DIMENSION: MILLIMETERS.
4. MEETS JEDEC MO150.
5. LEADS TO BE COPLANAR WITHIN 0.10 MM.
PROPRIETARY INFORMATION
TITLE:
PACKAGE OUTLINE, SSOP, 5.3 MM
APPROVAL
DOCUMENT CONTROL NO.
21-0056
REV.
C
1
1
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.
26 ____________________Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600
© 2003 Maxim Integrated Products
Printed USA
is a registered trademark of Maxim Integrated Products.