19-3099; Rev 0; 12/07 Dual-Output Buck Controller with Tracking/Sequencing The MAX15002 is a dual-output, pulse-width-modulated (PWM), step-down DC-DC controller with tracking and sequencing options. The device operates over the input voltage range of 5.5V to 23V or 5V ±10%. Each PWM controller provides an adjustable output down to 0.6V and delivers at least 15A of load current with excellent load and line regulation. The MAX15002 is optimized for highperformance, small-size power management solutions. The options of Coincident Tracking, Ratiometric Tracking, and Output Sequencing allow the tailoring of the power-up/power-down sequence depending on the system requirements. Each of the MAX15002 PWM sections utilizes a voltage-mode control scheme with external compensation, allowing for good noise immunity and maximum flexibility with a wide selection of inductor values and capacitor types. Each PWM section operates at the same, fixed switching frequency that is programmable from 200kHz to 2.2MHz and can be synchronized to an external clock signal using the SYNC input. Each converter operating at up to 2.2MHz with 180° out-of-phase, increases the input capacitor ripple frequency up to 4.4MHz, thereby significantly reducing the RMS input ripple current and the size of the input bypass capacitor requirement. The MAX15002 includes internal undervoltage lockout with hysteresis, digital soft-start/soft-stop for glitch-free power-up and power-down of the converter. The poweron reset (RESET) with an adjustable timeout period monitors both outputs and provides a RESET signal to the processor when both outputs are within regulation. Protection features include lossless valley-mode current limit and hiccup mode output short-circuit protection. The MAX15002 is available in a space-saving, 6mm x 6mm, 40-pin TQFN-EP package and is specified for operation over the -40°C to +125°C automotive temperature range. See the MAX15003 data sheet for a triple version of the MAX15002. Features o 5.5V to 23V or 5V ±10% Input Voltage Range o Dual-Output Synchronous Buck Controller o Selectable In-Phase or 180° Out-of-Phase Operation o Output Voltages Adjustable from 0.6V to 0.85VIN o Lossless Valley-Mode Current Sensing or Accurate Valley Current Sensing Using RSENSE o External Compensation for Maximum Flexibility o Digital Soft-Start and Soft-Stop o Sequencing or Coincident/Ratiometric VOUT Tracking o Individual PGOOD Outputs o RESET Output with a Programmable Timeout Period o 200kHz to 2.2MHz Programmable Switching Frequency o External Frequency Synchronization o Hiccup Mode Short-Circuit Protection o Space-Saving (6mm x 6mm) 40-Pin TQFN Package Ordering Information PART TEMP RANGE MAX15002ATL+ -40°C to +125°C PINPACKAGE 40 TQFN-EP* (6mm x 6mm) PKG CODE T4066-3 +Denotes a lead-free package. Applications PCI Express® *EP = Exposed pad. Host Bus Adapter Power Supplies Networking/Server Power Supplies Point-of-Load DC-DC Converters Pin Configuration appears at end of data sheet. PCI Express is a registered trademark of PCI-SIG Corp. ________________________________________________________________ Maxim Integrated Products For pricing, delivery, and ordering information, please contact Maxim Direct at 1-888-629-4642, or visit Maxim’s website at www.maxim-ic.com. 1 MAX15002 General Description MAX15002 Dual-Output Buck Controller with Tracking/Sequencing ABSOLUTE MAXIMUM RATINGS Continuous Power Dissipation (TA = +70°C) 40-Pin TQFN (derate 37mW/°C above +70°C) .............2963mW* θJA ..................................................................................27°C/W θJc .................................................................................1.4°C/W Operating Junction Temperature Range ...........-40oC to +125°C Maximum Junction Temperature .....................................+150°C Storage Temperature Range .............................-60°C to +150°C Lead Temperature (soldering, 10s) .................................+300°C IN, LX_, CSN_ to SGND..........................................-0.3V to +30V BST_ to SGND ........................................................-0.3V to +30V BST_ to LX_ ..............................................................-0.3V to +6V REG, DREG_, SYNC, EN_, RT, CT, RESET, PHASE, SEL to SGND ...............................-0.3V to +6V ILIM_, PGOOD_, FB_, COMP_, CSP_ to SGND .......-0.3V to +6V DL_ to PGND_.......................................-0.3V to (VDREG_ + 0.3V) DH_ to LX_ ...............................................-0.3V to (VBST_ + 0.3V) PGND_ to SGND, PGND_ to Any Other PGND_.......-0.3V to +0.3V *As per JEDEC51 standard (multilayer board). 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 (VIN = 5.5V to 23V or VIN = VREG = 4.5V to 5.5V, VDREG_ = VREG, VPGND_ = VSYNC = VPHASE = VSEL = 0V, CREG = 2.2µF, RRT = 100kΩ, CCT = 0.1µF, RILIM_ = 60kΩ, TA = TJ = -40°C to +125°C, unless otherwise noted. Typical values are at VIN = 12V, TA = TJ = +25°C.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS SYSTEM SPECIFICATIONS Input-Voltage Range Input Undervoltage Lockout Threshold VIN VUVLO 5.5 23.0 V VIN = VREG = VDREG (Note 2) 4.5 5.5 V VIN rising 3.95 4.15 V Input Undervoltage Lockout Hysteresis 4.05 0.35 V Operating Supply Current VIN = 12V, VFB_ = 0.8V 4.3 6.0 mA Shutdown Supply Current VIN = 12V, EN_ = 0V, PGOOD_ unconnected 150 300 µA REG VOLTAGE REGULATOR Output-Voltage Setpoint VREG Load Regulation VIN = 5.5V to 23V 4.9 IREG = 0 to 120mA, VIN = 12V 5.2 V 0.2 V DIGITAL SOFT-START/SOFT-STOP Soft-Start/Soft-Stop Duration Reference Voltage Steps 2048 Clocks 64 Steps ERROR TRANSCONDUCTANCE AMPLIFIER FB_, TRACK_ Input Bias Current FB_ Voltage Setpoint -250 VFB +250 nA TA = TJ = 0°C to +85°C 0.593 0.600 0.605 V TA = TJ = -40°C to +125°C 0.590 0.600 0.608 V FB_ to COMP_ Transconductance COMP_ Output Swing 2.1 0.75 mS 3.50 V Open-Loop Gain 80 dB Unity-Gain Bandwidth 10 MHz 2 _______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing (VIN = 5.5V to 23V or VIN = VREG = 4.5V to 5.5V, VDREG_ = VREG, VPGND_ = VSYNC = VPHASE = VSEL = 0V, CREG = 2.2µF, RRT = 100kΩ, CCT = 0.1µF, RILIM_ = 60kΩ, TA = TJ = -40°C to +125°C, unless otherwise noted. Typical values are at VIN = 12V, TA = TJ = +25°C.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS DRIVERS DL_, DH_ Break-Before-Make Time DH1 On-Resistance DH2 On-Resistance DL1 On-Resistance DL2 On-Resistance LX_ to PGND_ On-Resistance CLOAD = 5nF 20 Low, sinking 100mA 0.9 High, sourcing 100mA 1.3 Low, sinking 100mA 0.9 High, sourcing 100mA 1.3 Low, sinking 100mA 0.9 High, sourcing 100mA 1.3 Low, sinking 100mA 0.9 High, sourcing 100mA 1.3 Sinking 10mA ns Ω Ω Ω Ω Ω 8 CURRENT-LIMIT AND HICCUP MODE Cycle-By-Cycle Valley CurrentLimit Adjustment Range VCL VCL_ = VILIM_/10 50 300 Cycle-By-Cycle Valley CurrentLimit Threshold Tolerance VILIM_ = 0.5V 44 54 VILIM_ = 3V 288 312 ILIM_ Reference Current VILIM_ = 0 to 3V, TA = TJ = +25°C ILIM_ Reference Current Temperature Coefficient CSP_, CSN_ Input Bias Current Number of Cumulative CurrentLimit Events to Hiccup Number of Consecutive NonCurrent-Limit Cycles to Clear NCL VCSP_ = 0V, VCSN_ = -0.3V mV 20 µA 3333 ppm/°C -20 +20 NCL 8 NCLR 3 Hiccup Timeout mV µA Clock periods 4096 ENABLE/PHASE/SEL EN1 Threshold VEN-TH EN1 rising 1.19 EN1 Threshold Hysteresis 1.215 1.24 0.12 EN1 Input Bias Current -1 PHASE Input High 2 PHASE Input Low V +1 -1 SEL Threshold SEL Input Bias Current -1 µA V 0.8 PHASE Input Bias Current V V +1 µA 20 %VREG +1 µA _______________________________________________________________________________________ 3 MAX15002 ELECTRICAL CHARACTERISTICS (continued) MAX15002 Dual-Output Buck Controller with Tracking/Sequencing ELECTRICAL CHARACTERISTICS (continued) (VIN = 5.5V to 23V or VIN = VREG = 4.5V to 5.5V, VDREG_ = VREG, VPGND_ = VSYNC = VPHASE = VSEL = 0V, CREG = 2.2µF, RRT = 100kΩ, CCT = 0.1µF, RILIM_ = 60kΩ, TA = TJ = -40°C to +125°C, unless otherwise noted. Typical values are at VIN = 12V, TA = TJ = +25°C.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS 0.54 0.555 0.57 V 0.1 V PGOOD, RESET OUTPUTS FB_ for PGOOD Threshold FB_ falling RESET, PGOOD_ Output Low Level Sinking 3mA RESET, PGOOD_ Leakage -1 CT Charging Current 1.8 CT Output Low 2 Sinking 3mA CT rising CT Threshold for RESET Delay 1.8 CT falling +1 µA 2.2 µA 0.1 V 2.6 1.2 V OSCILLATOR Switching Frequency Range (Each Converter) fSW Switching Frequency Accuracy (Each Converter) 200 2200 fSW ≤ 1500kHz -5 +5 fSW > 1500kHz -7 +7 VPHASE = 0V (DH1 rising to DH2 rising) Phase Delay RT Voltage VSYNC = 0V, fSW = 1.5 x 1011/RRT + 2k VPHASE = VREG (DH1 rising to DH2 rising) VRT 40kΩ < RRT < 500kΩ kHz % 180 degrees 0 degrees 2 V Minimum Controllable On-Time tON(MIN) 75 ns Minimum Off-Time tOFF(MIN) 150 ns SYNC High-Level Voltage 2 V SYNC Low-Level Voltage 0.8 SYNC Internal Pulldown Resistor SYNC Frequency Range 50 (Note 3) 100 0.4 V 200 kΩ 4.6 MHz SYNC Minimum On-Time SYNC Minimum Off-Time 30 30 ns ns PWM Ramp Amplitude (Peak-Peak) 2 V PWM Ramp Valley 1 V Note 1: 100% production tested at TA = TJ = +25°C and TA = TJ = +125°C. Limits at other temperatures are guaranteed by design. Note 2: For 5V applications, connect REG directly to IN. Note 3: The switching frequency is 1/2 of the SYNC frequency. 4 _______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing CONVERTER 2 EFFICIENCY vs. LOAD CURRENT VIN = 12V VIN = 16V 70 VOUT1 = 3.3V 70 VIN = 12V 60 VIN = 16V 50 40 30 20 VOUT1 = 3.3V fSW = 300kHz 40 0.1 1 0.50 0.25 0 -0.25 -0.50 -1.00 0.1 1 100 10 0 5 15 10 LOAD CURRENT (A) LOAD CURRENT (A) LOAD CURRENT (A) CONVERTER 2 LOAD REGULATION INTERNAL VOLTAGE REGULATION (REG) CONVERTER-SWITCHING FREQUENCY vs. RRT 4.98 0.50 10,000 4.97 VREG (V) 0.25 0 4.96 4.95 4.94 -0.25 4.93 -0.50 4.92 -0.75 15 10 0 LOAD CURRENT (A) 20 40 60 200 8 6 4 2 0 -2 -4 -6 -8 600 800 VALLEY CURRENT-LIMIT THRESHOLD vs. VILIM MAX15002 toc07 10 400 RRT (kΩ) TEMPERATURE (°C) SWITCHING FREQUENCY ACCURACY vs. TEMPERATURE SWITCHING FREQUENCY ACCURACY (%) 0 100 80 fSW = 300kHz -10 350 MAX15002 toc08 5 100 10 4.90 0 1000 VIN = 12V CREG = 2.2μF 4.91 -1.00 MAX15002 toc06 4.99 SWITCHING FREQUENCY (kHz) VOUT2 = 1.8V MAX15002 toc05 5.00 MAX15002 toc04 1.00 0.75 0.75 -0.75 0 100 10 VOUT2 = 1.8V fSW = 300kHz 10 MAX15002 toc03 1.00 MAX15002 toc02 80 60 50 OUTPUT-VOLTAGE ACCURACY (%) VIN = 6V VALLEY CURRENT-LIMIT THRESHOLD (mV) EFFICIENCY (%) 80 90 EFFICIENCY (%) VIN = 6V 90 CONVERTER 1 LOAD REGULATION 100 MAX15002 toc01 100 OUTPUT-VOLTAGE ACCURACY (%) CONVERTER 1 EFFICIENCY vs. LOAD CURRENT 300 250 200 150 100 50 -50 -25 0 25 50 75 TEMPERATURE (°C) 100 125 150 500 1000 1500 2000 2500 3000 3500 VILIM (mV) _______________________________________________________________________________________ 5 MAX15002 Typical Operating Characteristics (VIN = 12V, referenced to Figure 8, TA = TJ = +25°C, unless otherwise noted.) Typical Operating Characteristics (continued) (VIN = 12V, referenced to Figure 8, TA = TJ = +25°C, unless otherwise noted.) VALLEY CURRENT-LIMIT THRESHOLD vs. TEMPERATURE SWITCHING CURRENT vs. FREQUENCY MAX15002 toc11 MAX15002 toc10 90 14 SWITCHING CURRENT (mA) RILIMc = 25.5kΩ RATIOMETRIC STARTUP 15 MAX15002 toc09 100 VALLEY CURRENT-LIMIT THRESHOLD (mV) MAX15002 Dual-Output Buck Controller with Tracking/Sequencing 80 70 60 50 40 13 VIN 12 1V/div 10 9 1V/div 8 VIN = 12V DL_, DH_ UNCONNECTED VFB_ = 0V 6 TEMP COEFFICIENT (nom.) = 3,333ppm/°C 20 0V 11 7 30 10V/div 0V VOUT1, 2 VEN2 = 0V, SEL = REG 5 -50 -25 0 25 50 75 100 125 150 200 700 1200 2200 1700 TEMPERATURE (°C) FREQUENCY (kHz) RATIOMETRIC SHUTDOWN CHANNEL 2 SHORT CIRCUIT (RATIOMETRIC MODE) MAX15002 toc12 2ms/div CHANNEL 1 SHORT CIRCUIT (RATIOMETRIC MODE) MAX15002 toc14 MAX15002 toc13 VOUT2 10V/div VIN 500mV/div V OUT2 0V VOUT1 10V/div VIN 0V VOUT1 1V/div 2V/div 0V 0V 500mV/div VOUT1 VOUT2 1V/div 2V/div 0V 0V 0V VEN2 = 0V, SEL = REG VEN2 = 0V, SEL = REG 1ms/div VEN2 = 0V, SEL = REG 1ms/div 1ms/div COINCIDENT STARTUP COINCIDENT SHUTDOWN MAX15002 toc15 MAX15002 toc16 VOUT1 10V/div 0V VIN 1V/div 500mV/div VOUT2 500mV/div 1V/div VOUT1, 2 0V 0V CIRCUIT OF FIGURE 8, SEL = REG 2ms/div 6 2ms/div _______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing CHANNEL 2 SHORT CIRCUIT (COINCIDENT MODE) CHANNEL 1 SHORT CIRCUIT (COINCIDENT MODE) MAX15002 toc17 10V/div VIN MAX15002 toc19 10V/div VIN 10V/div VIN 0V VOUT2 SEQUENCING STARTUP MAX15002 toc18 VOUT1 0V 0V 1V/div 2V/div 0V 0V 1V/div 1V/div VOUT2 VOUT1 2V/div 1V/div 0V 0V 0V VOUT1, 2 SEL = REG 1ms/div 4ms/div 1ms/div CHANNEL 1 OUTPUT SHORT CIRCUIT (SEQUENCING MODE) CONVERTER 2 OUTPUT SHORT CIRCUIT (SEQUENCING MODE) SEQUENCING SHUTDOWN MAX15002 toc20 MAX15002 toc22 MAX15002 toc21 VOUT1 VIN 10V/div 500mV/div V OUT2 0V VIN 10V/div 0V VOUT1 2V/div 1V/div VOUT2 0V 0V 500mV/div VOUT1 2V/div VOUT2 1V/div 0V SEL = GND EN/TRACK2 = PGOOD1 SEL = REG 1ms/div 0V SEL = GND EN/TRACK2 = PGOOD1 0V 1ms/div 1ms/div RESET AT STARTUP (SEQUENCING MODE) RESET AT SHUTDOWN (SEQUENCING MODE) MAX15002 toc23 MAX15002 toc24 5V/div 5V/div VRESET 0V VRESET 0V 1V/div VOUT1 1V/div VOUT2 1V/div 1V/div VOUT1, 2 SEL = GND EN/TRACK2 = PGOOD1 20ms/div 0V SEL = GND EN/TRACK2 = PGOOD1 0V 1ms/div _______________________________________________________________________________________ 7 MAX15002 Typical Operating Characteristics (continued) (VIN = 12V, referenced to Figure 8, TA = TJ = +25°C, unless otherwise noted.) MAX15002 Dual-Output Buck Controller with Tracking/Sequencing Typical Operating Characteristics (continued) (VIN = 12V, referenced to Figure 8, TA = TJ = +25°C, unless otherwise noted.) CONVERTER 1 SHORT-CIRCUIT CONDITION (HICCUP MODE) 180° OUT-OF-PHASE OPERATION MAX15002 toc26 MAX15002 toc25 VOUT1 500mV/div 5V/div 0V VSYNC IOUT1 10V/div 10A/div VLX1 VLX1 0V VLX2 0V 10V/div VDL1 5V/div VPGOOD1 1V/div 10V/div 1μs/div 4ms/div IN-PHASE OPERATION BREAK-BEFORE-MAKE TIMING MAX15002 toc27 MAX15002 toc28 VLX1 5V/div 5V/div 0V VSYNC 0V 10V/div 0V VLX1 VDL1 2V/div 10V/div 0V VLX2 0V 1μs/div 20ns/div LOAD-TRANSIENT RESPONSE (IOUT2 = 100mA TO 10A) LOAD-TRANSIENT RESPONSE (IOUT2 = 5A TO 10A) MAX15002 toc29 VOUT2 MAX15002 toc30 100mV/div AC-COUPLED 100mV/div AC-COUPLED VOUT2 IOUT2 IOUT2 5A/div 5A/div 0 200μs/div 8 0 200μs/div _______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing PIN NAME 1 REG 5V Regulator Output. Bypass with a 2.2µF ceramic capacitor to SGND. FUNCTION 2 SEL Track/Sequence Select Input. At startup, connect SEL to REG to configure as a dual tracker or connect SEL to SGND to configure as a dual sequencer. Note: When configured as a dual sequencer, each rail is independently controlled by EN_. 3 PGND1 Controller 1 Power-Ground Connection. Connect the input filter capacitor’s negative terminal, the source of the synchronous MOSFET, and the output filter capacitor’s return to PGND1. Connect externally to SGND at a single point near the input capacitor return terminal. 4 DL1 5 DREG1 6 LX1 7 DH1 Controller 1 High-Side Gate Driver Output. DH1 drives the gate of the high-side MOSFET. 8 BST1 Controller 1 High-Side Gate Driver Supply. Connect BST1 to the cathode of the boost diode and to the positive terminal of the boost capacitor. 9 CSN1 Controller 1 Negative Current-Sense Input. Connect CSN1 to the synchronous MOSFET drain (connected to LX1). When using a current-sense resistor, connect CSN1 to the junction of a low-side MOSFET’s source and the current-sense resistor. See Figure 10. 10 CSP1 Controller 1 Positive Current-Sense Input. Connect CSP1 to the synchronous MOSFET source (connected to PGND1). When using a current-sense resistor, connect CSP1 to the PGND1 end of the current-sense resistor. 11 ILIM1 Controller 1 Valley Current-Limit Set Output. Connect a 25kΩ to 150kΩ resistor, RILIM1, from ILIM1 to SGND to program the valley current-limit threshold from 50mV to 300mV. ILIM1 sources 20µA out to RILIM1. The resulting voltage divided by 10 is the valley current-limit threshold. When using a precision current-sense resistor, connect a resistive divider from REG to ILIM1 to SGND to set the valley current limit. See Figure 10. 12 COMP1 Controller1 Error Transconductance Amplifier Output. Connect COMP1 to the compensation feedback network. 13 EN1 Controller 1 Enable Input. EN1 must be above 1.24V, VEN-TH, for the PWM controller to start Output 1. Controller 1 is the master. Use the master as the highest output voltage in a coincident tracking configuration. 14 FB1 Controller 1 Feedback Regulation Point. Connect to the center tap of a resistive divider from the converter output to SGND to set the output voltage. The FB1 voltage regulates to VFB (0.6V). 15 PGOOD1 Controller 1 Power-Good Output. Open-drain PGOOD1 output goes high impedance (releases) when FB1 is above 0.925 x VFB (0.555V). 16 PGND2 Controller 2 Power Ground Connection. Connect the input filter capacitor’s negative terminal, the source of the synchronous MOSFET, and the output filter capacitor’s return to PGND2. Connect externally to SGND at a single point near the input capacitor return terminal. 17 DL2 18 DREG2 19 LX2 Controller 1 Low-Side Gate Driver Output. DL1 is the gate driver output for the synchronous MOSFET. Controller 1 Low-Side Gate Driver Supply. Connect externally to REG and the anode of the boost diode. Connect a minimum of 0.1µF ceramic capacitor from DREG1 to PGND1. Controller 1 High-Side MOSFET Source Connection/Synchronous MOSFET Drain Connection. Connect the inductor and the negative side of the boost capacitor to LX1. Controller 2 Low-Side Gate Driver Output. DL2 is the gate driver output for the synchronous MOSFET. Controller 2 Low-Side Gate Driver Supply. Connect externally to REG and the anode of the boost diode. Connect a minimum of a 0.1µF ceramic capacitor from DREG2 to PGND2. Controller 2 High-Side MOSFET Source Connection/Synchronous MOSFET Drain Connection. Connect the inductor and the negative side of the boost capacitor to LX2. _______________________________________________________________________________________ 9 MAX15002 Pin Description MAX15002 Dual-Output Buck Controller with Tracking/Sequencing Pin Description (continued) PIN NAME 20 DH2 Controller 2 High-Side Gate Driver Output. DH2 drives the gate of the high-side MOSFET. 21 BST2 Controller 2 High-Side Gate Driver Supply. Connect BST2 to the cathode of the boost diode and to the positive terminal of the boost capacitor. 22 CSN2 Controller 2 Negative Current-Sense Input. Connect CSN2 to the synchronous MOSFET drain (connected to LX2). When using a current-sense resistor, connect CSN2 to the junction of the low-side MOSFET’s source and the current-sense resistor. See Figure 10. 23 CSP2 Controller 2 Positive Current-Sense Input. Connect CSP2 to the synchronous MOSFET source (connected to PGND2). When using a current-sense resistor, connect CSP2 to the PGND2 end of the current-sense resistor. 24 ILIM2 Controller 2 Valley Current-Limit Set Output. Connect a 25kΩ to 150kΩ resistor, RILIM2, from ILIM2 to SGND to program the valley current-limit threshold from 50mV to 300mV. ILIM2 sources 20µA out to RILIM2. The resulting voltage divided by 10 is the valley current-limit threshold. When using a precision current-sense resistor, connect a resistive divider from REG to ILIM2 to SGND to set the valley current limit. See Figure 10. 25 COMP2 Controller 2 Error Transconductance Amplifier Output. Connect COMP2 to the compensation feedback network. 26 EN/TRACK2 Controller 2 Enable/Tracking Input. See Figure 2. When sequencing, EN/TRACK2 must be above 1.24V for the PWM controller 2 to start. Coincident tracking—connect the same resistive divider used for FB2, from Output 1 to EN/TRACK2 to SGND. Ratiometric tracking—connect EN/TRACK2 to analog ground. 27 FB2 28 PGOOD2 29–33 N.C. 34 SYNC Synchronization Input. Drive with a frequency at least 20% higher than two times the frequency programmed using the RT pin. The switching frequency is 1/2 the SYNC frequency. Connect SYNC to SGND when not used. 35 SGND Analog Ground Connection. Connect SGND and PGND_ together at one point near the input bypass capacitor return terminal. 36 RT 37 PHASE Phase Select Input. Connect PHASE to SGND for 180° out-of-phase operation between the controllers. Connect to REG for in phase operation. 38 RESET RESET Output. Open-drain RESET output releases after all PGOODs are released and timeout programmed by CT finishes. 39 CT RESET Timeout Capacitor Connection. Connect a timing capacitor from CT to analog ground to set the RESET delay. CT sources 2µA into the timing capacitor. When the voltage at CT passes 2V, open-drain RESET goes high impedance. 40 IN Supply Input Connection. Connect to an external voltage source from 5.5V to 23V. For 4.5V to 5.5V input application, connect IN and REG together. — EP Exposed Pad. Solder the exposed pad to a large SGND plane. 10 FUNCTION Controller 2 Feedback Regulation Point. Connect to the center tap of a resistive divider from the converter output to SGND to set the output voltage. The FB2 voltage regulates to VFB (0.6V). Controller 2 Power-Good Output. Open-drain PGOOD2 output goes high impedance (releases) when FB2 is above 0.925 x VFB (0.555V). No Connection. Not internally connected. Oscillator Timing Resistor Connection. Connect a 750kΩ to 68kΩ resistor from RT to SGND to program the switching frequency from 200kHz to 2.2MHz. ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing PWM CONTROLLER 1 SEL IN EN1 RESET TIMEOUT 1.24VON 1.12VOFF CONFIG SELECTOR LDO CT RESET MAX15002 REG EN 1.24V 1.12V SGND SEQ_ PGPD_ SHDN 0.6V REF CSP1 SEQ_ VREGOK EN1 CSN1 DOWN1 VREF DIGITAL SOFT-START AND STOP OVL CONFIG OVL1 VR1 IMAX1 RES OVERLOAD MANAGEMENT E/A CLK1 OVL_ CURRENTLIMIT SET CLK1 BST1 FB1 DH1 R CPWM EN OSC LX1 Q SET DOMINANT COMP1 SYNC RT PHASE ILIM1 DREG1 S RAMP DL1 LEVEL CLK1 SHIFT CLK2 PGND1 0.925 x VREF PGPD1 FB1 PGOOD1 ______________________________________________________________________________________ 11 MAX15002 Functional Diagrams Dual-Output Buck Controller with Tracking/Sequencing MAX15002 Functional Diagrams (continued) PWM CONTROLLER 2 EN/TRACK2 1.24VON 1.12VOFF MAX15002 SEQ_ CSP2 SEQ_ VREF EN CONFIG DOWN2 DIGITAL SOFT-START AND STOP EN1 SHDN EN2 CSN2 OVL CONFIG OVL_ CURRENTLIMIT SET CLK2 SEL_ OVL2 RES OVERLOAD IMAX2 MANAGEMENT VREF VR2 EN/ TRACK2 E/A CLK2 ILIM2 BST2 FB2 DH2 R COMP2 CPWM CLK2 RAMP LEVEL SHIFT CLK2 0.925 x VREF LX2 Q SET DOMINANT DREG2 S DL2 PGND2 PGPD2 FB2 PGOOD2 Detailed Description The MAX15002 is a dual-output, pulse-width-modulated (PWM), step-down, DC-DC controller with tracking and sequencing options. The device operates over the input voltage range of 5.5V to 23V or 5V ±10%. Each PWM controller provides an adjustable output down to 0.6V and delivers at least 15A of load current with excellent load and line regulation. Each of the MAX15002 PWM sections utilizes a voltage-mode control scheme for good noise immunity and offers external compensation allowing for maximum flexibility with a wide selection of inductor values and capacitor types. The device operates at a fixed switching frequency that is programmable from 200kHz to 2.2MHz and can be synchronized to an external clock signal using the SYNC input. Each converter, operating at up to 2.2MHz with 180° out-of-phase, increases the input capacitor ripple frequency up to 4.4MHz, reducing the RMS input ripple current and the size of the input bypass capacitor requirement significantly. 12 The MAX15002 provides Coincident Tracking, Ratiometric Tracking, and Sequencing. This allows tailoring of the power-up/power-down sequence depending on the system requirements. The MAX15002 features lossless valley-mode currentlimit protection by monitoring the voltage drop across the synchronous MOSFET’s on-resistance to sense the inductor current. The MAX15002’s internal current source exhibits a positive temperature coefficient to help compensate for the MOSFET’s temperature coefficient. Use an external voltage-divider when a more precise current limit is desired. This divider along with a precision shunt resistor allows for more accurate current limit. The MAX15002 includes internal undervoltage lockout with hysteresis, digital soft-start/soft-stop for glitch-free power-up and power-down of the converter. The power-on reset (RESET) with adjustable timeout period monitors both outputs and provides a RESET signal to the processor indicating when the outputs are within regulation. Protection features include lossless valleymode current limit and hiccup mode output short-circuit protection. ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing Digital Soft-Start/Soft-Stop The MAX15002 soft-start feature allows the load voltage to ramp up in a controlled manner, eliminating outputvoltage overshoot. Soft-start begins after VIN exceeds the undervoltage lockout threshold and the enable input is above 1.24V. The soft-start circuitry gradually ramps up the reference voltage. This controls the rate of rise of the output voltage and reduces input surge currents during startup. The soft-start duration is 2048 clock cycles. The output voltage is incremented through 64 equal steps. The output reaches regulation when soft-start is completed, regardless of output capacitance and load. Soft-stop commences when the enable input falls below 1.12V. The soft-stop circuitry ramps down the reference voltage controlling the output voltage rate of fall. The output voltage is decremented through 64 equal steps in 2048 clock cycles. Internal Linear Regulator (REG) REG is the output terminal of a 5V LDO powered from IN which provides power to the IC. Connect REG externally to DREG to provide power for the low-side MOSFET gate driver. Bypass REG to SGND with a minimum 2.2µF ceramic capacitor. Place the capacitor physically close to the MAX15002 to provide good bypassing. REG is intended for powering only the internal circuitry and should not be used to supply power to external loads. REG can source up to 120mA. This current, I REG , includes quiescent current (IQ) and gate drive current (IDREG): IREG = IQ + [fSW x Σ(QGHS_ + QGLS_)] where QGHS_ + QGLS_ is the total gate charge of each of the respective high- and low-side external MOSFETs at VGATE = 5V. fSW is the switching frequency of the converter and IQ is the quiescent current of the device at the switching frequency. MOSFET Gate Drivers DREG_ is the supply input for the low-side MOSFET driver. Connect DREG_ to REG externally. Everytime the low-side MOSFET switches on, high peak current is drawn from DREG for a short amount of time. Adding an RC filter (1Ω to 3.3Ω and 2.2µF//0.1µF ceramic capacitors are typical) from REG to DREG_ filters out high-peak currents. Alternatively, DREG can be connected to an external source (VDREG-EXT). Note that the DREG voltage should be high enough to fully enhance the low-side MOSFET. To avoid partial enhancing of the MOSFETs, use the VDREG-EXT to set the UVLO externally using EN1. BST_ supplies the power for the high-side MOSFET drivers. Connect the bootstrap diode from BST_ to DREG_ (anode at DREG_ and cathode at BST_). Connect a bootstrap 0.1µF or higher ceramic capacitor between BST_ and LX_. Though not always necessary, it may be useful to insert a small resistor (4.7Ω to 22Ω) in series with the BST_ pin and the cathode of the bootstrap diode for additional noise immunity. The high-side (DH_) and low-side (DL_) drivers drive the gates of the external n-channel MOSFETs. The drivers’ 2A peak source- and sink-current capability provides ample drive for the fast rise and fall times of the switching MOSFETs. Faster rise and fall times result in reduced switching losses. The gate driver circuitry also provides a break-beforemake time (20ns typ) to prevent shoot-through currents during transition. Oscillator/Synchronization Input/Phase Staggering (RT, SYNC, PHASE) Use an external resistor at RT to program the MAX15002 switching frequency from 200kHz to 2.2MHz. Choose the appropriate resistor at RT to calculate the desired output switching frequency (fSW): fSW (Hz) = 1.5 x 1011/(RRT + 2000)Ω Connect an external clock at SYNC for external clock synchronization. A rising clock edge on SYNC is interpreted as a synchronization input. If the SYNC signal is lost, the internal oscillator takes control of the switching rate, returning the switching frequency to that set by RRT. This maintains output regulation even with intermittent SYNC signals. For proper synchronization, the external frequency must be at least 20% higher than twice the frequency programmed through the RT input. The switching frequency is 1/2 the SYNC frequency. Connect SYNC to SGND when not used. Connect PHASE to SGND for 180° out-of-phase operation between the controllers. Connect PHASE to REG for in-phase operation. ______________________________________________________________________________________ 13 MAX15002 Internal Undervoltage Lockout (UVLO) VIN must exceed the default UVLO threshold before any operation can commence. The UVLO circuitry keeps the MOSFET drivers, oscillator, and all the internal circuitry shut down to reduce current consumption. The UVLO rising threshold is 4.05V with 350mV hysteresis. MAX15002 Dual-Output Buck Controller with Tracking/Sequencing Coincident/Ratiometric Tracking (SEL, EN/TRACK2) VOUT1 The enable/tracking input in conjunction with digital soft-start and soft-stop provides coincident/ratiometric tracking. See Figure 1. Track an output voltage by connecting a resistive divider from the output being tracked to the enable/tracking input. For example, for VOUT2 to coincidentally track VOUT1, connect the same resistive divider used for FB2, from OUT1 to EN/TRACK2 to SGND. See Figure 2 and the Coincident Startup and Coincident Shutdown graphs in the Typical Operating Characteristics. Track ratiometrically by connecting EN/TRACK2 to SGND. This synchonizes the soft-start and soft-stop of all the controllers’ references, and hence their respective output voltages will track ratiometrically. See Figure 2 and the Ratiometric Startup and Ratiometric Shutdown graphs in the Typical Operating Characteristics. Connect SEL to REG to configure as a dual tracker. When the MAX15002 converter is configured as a tracker, the output short-circuit fault situations at master or slave output is handled carefully so that either the master or slave output does not stay on when the other output is shorted to the ground. When the slave is shorted and enters in hiccup mode, the master will softstop. When the master is shorted and the part enters in hiccup mode, the slave will ratiometrically soft-stop. Coming out of the hiccup, all outputs will soft-start coincidently or ratiometrically depending on their initial configuration. See the Typical Operating Characteristics for the output behaviour during the fault conditions. During the thermal shutdown or power-off, when the input falls below its UVLO, the output voltages fall down at the rate depending on the respective output capacitor and load. See Figure 1. Output-Voltage Sequencing (SEL, EN/TRACK2, PGOOD) Referring to Figure 1c, when sequencing, the enable/tracking input must be above 1.24V for each PWM controller to start. The PGOOD_ outputs and EN/TRACK2 inputs can be daisy-chained to generate power sequencing. Open-drain PGOOD_ outputs go high impedance when FB_ is above the PGOOD_ threshold (555mV typ). 14 VOUT2 SOFT-START SOFT-STOP A) COINCIDENT TRACKING OUTPUTS VOUT1 VOUT2 SOFT-START SOFT-STOP B) RATIOMETRIC TRACKING OUTPUTS VOUT1 VOUT2 SOFTSTART SOFT-STOP C) SEQUENCED OUTPUTS Figure 1. Graphical Representation of Coincident Tracking, Ratiometric Tracking, and PGOOD Sequencing Connect the power-good output to the enable/tracking input to set when the other controller will start. See Figure 2. Connect SEL to SGND to configure as a dual sequencer. ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing MAX15002 COINCIDENT TRACKING RATIOMETRIC TRACKING PGOOD SEQUENCING VIN VIN VIN EN1 EN1 EN1 VOUT1 EN/TRACK2 REG RA EN/TRACK2 SEL REG PGOOD1 RB EN/TRACK2 SEL VOUT2 RA FB2 RB SEL REG Figure 2. Ratiometric Tracking, Coincident Tracking, PGOOD Sequencing Configurations Error Amplifier The output of the internal error transconductance amplifier (COMP_) is provided for frequency compensation (see the Compensation Design Guidelines section). The inverting input is FB_ and the output COMP_. The error transamplifier has an 80dB open-loop gain and a 10MHz GBW product. Output Short-Circuit Protection (Hiccup Mode) The current-limit circuit employs a valley current-limiting algorithm that either uses a shunt or the synchronous MOSFET’s on-resistance as the current-sensing element. Once the high-side MOSFET turns off, the voltage across the current-sensing element is monitored. If this voltage does not exceed the current-limit threshold, the high-side MOSFET turns on normally at the start of the next cycle. If the voltage exceeds the current-limit threshold just before the beginning of a new PWM cycle, the controller skips that cycle. During severe overload or short-circuit conditions, the switching frequency of the device appears to decrease because the on-time of the low-side MOSFET extends beyond a clock cycle. If the current-limit threshold is exceeded for more than eight cumulative clock cycles (NCL), the device shuts down (both DH and DL are pulled low) for 4096 clock cycles (hiccup timeout) and then restarts with a softstart sequence. If three consecutive cycles pass without a current-limit event, the count of NCL is cleared (see Figure 3). Hiccup mode protects against a continuous output short circuit. ______________________________________________________________________________________ 15 MAX15002 Dual-Output Buck Controller with Tracking/Sequencing CURRENT LIMIT IN COUNT OF 8 NCL INITIATE HICCUP TIMEOUT NHT CLR The minimum input voltage is limited by the maximum duty cycle and is calculated using the following equation: VIN(MIN) ≥ ( VOUT 1 − t OFF(MIN) × fSW ) where tOFF(MIN) typically is equal to 150ns. Inductor Selection IN COUNT OF 3 NCLR CLR Figure 3. Hiccup-Mode Block Diagram PWM Controller Design Procedures Setting the Switching Frequency Connect a 750kΩ to 68kΩ resistor from RT to SGND to program the switching frequency from 200kHz to 2.2MHz. Calculate the switching frequency using the following equation: fSW (Hz) = 1.5 x 1011/(RRT + 2000)Ω Higher frequencies allow designs with lower inductor values and less output capacitance. Consequently, peak currents and I 2 R losses are lower at higher switching frequencies, but core losses, gate-charge currents, and switching losses increase. Effective Input Voltage Range Although the MAX15002 converters can operate from input supplies ranging from 5.5V to 23V, the input voltage range can be effectively limited by the MAX15002 duty-cycle limitations for a given output voltage. The maximum input voltage is limited by the minimum ontime (tON(MIN)): VIN(MAX) ≤ where tON(MIN) is 75ns. 16 VOUT t ON(MIN) × fSW Three key inductor parameters must be specified for operation with the MAX15002: inductance value (L), peak inductor current (IPEAK), and inductor saturation current (ISAT). The minimum required inductance is a function of operating frequency, input-to-output voltage differential, and the peak-to-peak inductor current (ΔIP-P). Higher ΔIP-P allows for a lower inductor value. A lower inductance value minimizes size and cost and improves large-signal and transient response. However, efficiency is reduced due to higher peak currents and higher peak-to-peak output voltage ripple for the same output capacitor. A higher inductance increases efficiency by reducing the ripple current, however resistive losses due to extra wire turns can exceed the benefit gained from lower ripple current levels especially when the inductance is increased without also allowing for larger inductor dimensions. A good rule of thumb is to choose ΔIP-P equal to 30% of the full load current. Calculate the inductance using the following equation: L = VOUT (VIN − VOUT ) VIN × fSW × ΔIP −P VIN and VOUT are typical values so that efficiency is optimum for typical conditions. The switching frequency is programmable between 200kHz and 2.2MHz (see Oscillator/Synchronization Input/Phase Staggering (RT, SYNC, PHASE) section). The peak-to-peak inductor current, which reflects the peak-to-peak output ripple, is worst at the maximum input voltage. See the Output Capacitor Selection section to verify that the worst-case output current ripple is acceptable. The inductor saturation current (ISAT) is also important to avoid runaway current during continuous output short-circuit conditions. Select an inductor with an ISAT specification higher than the maximum peak current. ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing ⎛V ⎞ ILOAD(MAX) × ⎜ OUT ⎟ ⎝ VIN ⎠ CIN = ΔVQ × fSW where: ΔIP −P = (VIN − VOUT ) × VOUT VIN × fSW × L ILOAD(MAX) is the maximum output current, ΔIP-P is the peak-to-peak inductor current, and fSW is the switching frequency. For the condition with only one converter on, calculate the input ripple current using the following equation: ICIN(RMS) = ILOAD _ MAX × VOUT × (VIN − VOUT ) VIN The MAX15002 includes UVLO hysteresis to avoid possible unintentional chattering during turn-on. Use additional bulk capacitance if the input source impedance is high. At lower input voltage, additional input capacitance helps avoid possible undershoot below the undervoltage lockout threshold during transient loading. Output Capacitor Selection The allowed output voltage ripple and the maximum deviation of the output voltage during load steps determine the required output capacitance and its ESR. The output ripple is mainly composed of ΔVQ (caused by the capacitor discharge) and ΔVESR (caused by the voltage drop across the equivalent series resistance of the output capacitor). The equations for calculating the output capacitance and its ESR are: COUT = ESR = ΔIP −P 8 × ΔVQ × fSW 2 × ΔVESR ΔIP −P ΔVESR and ΔVQ are not directly additive because they are out of phase from each other. If using ceramic capacitors, which generally have low ESR, ΔVQ dominates. If using electrolytic capacitors, ΔVESR dominates. The allowable deviation of the output voltage during fast load transients also affects the output capacitance, its ESR, and its equivalent series inductance (ESL). The output capacitor supplies the load current during a load step until the controller responds with a greater duty cycle. The response time (tRESPONSE) depends on the gain bandwidth of the converter (see the Compensation Design Guidelines section). The resistive drop across the output capacitor’s ESR, the drop across the capacitor’s ESL, and the capacitor discharge cause a voltage droop during the load-step (ISTEP). Use a combination of low-ESR tantalum/aluminum electrolyte and ceramic capacitors for better load-transient and voltage-ripple performance. Nonleaded capacitors and capacitors in parallel help reduce the ESL. Keep the maximum output voltage deviation below the tolerable limits of the electronics being powered. Use the following equations to calculate the required ESR, ESL, and capacitance value during a load step: ΔVESR ISTEP ×t I COUT = STEP RESPONSE ΔVQ ESR = ESL = ΔVESL × t STEP ISTEP where ISTEP is the load step, tSTEP is the rise time of the load step, and tRESPONSE is the response time of the controller. ______________________________________________________________________________________ 17 MAX15002 Input Capacitor Selection The discontinuous input current of the buck converter causes large input ripple currents and therefore the input capacitor must be carefully chosen to withstand the input ripple current and keep the input voltage ripple within design requirements. The 180° ripple phase operation increases the frequency of the input capacitor ripple current to twice the individual converter switching frequency. When using ripple phasing, the worst-case input capacitor ripple current is when the one converter with the highest output current is on. The input voltage ripple is comprised of ΔVQ (caused by the capacitor discharge) and ΔVESR (caused by the ESR of the input capacitor). The total voltage ripple is the sum of ΔVQ and ΔVESR that peaks at the end of the on-cycle. Calculate the input capacitance and ESR required for a specified ripple using the following equations: ΔVESR ESR = ΔIP −P ⎞ ⎛ ⎜ ILOAD(MAX) + ⎟ ⎝ 2 ⎠ Setting the Current Limit The MAX15002 uses a valley current-sense method for current limiting. The voltage drop across the low-side MOSFET due to its on-resistance is used to sense the inductor current. The voltage drop (VVALLEY) across the low-side MOSFET at the valley point and at ILOAD is: ΔI ⎛ ⎞ VVALLEY = RDS(ON) × ⎜ ILOAD − P −P ⎟ ⎝ 2 ⎠ RDS(ON) is the on-resistance of the low-side MOSFET, ILOAD is the rated load current, and ΔIP-P is the peakto-peak inductor current. The RDS(ON) of the MOSFET varies with temperature. Calculate the RDS(ON) of the MOSFET at its operating junction temperature at full load using the MOSFET datasheet. To compensate for this temperature variation, the 20µA ILIM reference current has a temperature coefficient of 3333ppm/°C. This allows the valley current-limit threshold (VCL) to track and partially compensate for the increase in the synchronous MOSFET’s RDS(ON) with increasing temperature. Use the following equation to calculate RILIM: ΔI ⎛ ⎞ RDS(ON) × ⎜ ICL(MAX) − P −P ⎟ ×10 ⎝ 2 ⎠ RILIM = 20 ×10 −6 ⎡1+ 3.333 ×10 −3 (T − 25°C)⎤ ⎣⎢ ⎦⎥ where ICL(MAX) is the maximum current limit. Figure 4 illustrates the effect of the MAX15002 ILIM reference current temperature coefficient to compensate for the variation of the MOSFET RDS(ON) over the operating junction temperature range. Power MOSFET Selection When choosing the MOSFETs, consider the total gate charge, RDS(ON), power dissipation, the maximum drainto-source voltage 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 that are optimized for high-frequency switching applications. The average gatedrive current from the MAX15002’s output is proportional to the frequency and gate charge required to drive the MOSFET. The power dissipated in the MAX15002 is proportional to the input voltage and the average drive current (see the Power Dissipation section). 18 VALLEY CURRENT-LIMIT THRESHOLD AND RDS(ON) vs. TEMPERATURE 1.5 1.4 RDS(ON) 1.3 MAX15002 fig04 Connect a 25kΩ to 150kΩ resistor, RILIM, from ILIM to SGND to program the valley current-limit threshold (VCL) from 50mV to 300mV. ILIM sources 20µA out to RILIM. The resulting voltage divided by 10 is the valley current-limit threshold. VILIM AND RDS(ON) (NORMALIZED) MAX15002 Dual-Output Buck Controller with Tracking/Sequencing 1.2 1.1 VILIM 1.0 0.9 0.8 0.7 0.6 RILIM = 25.5kΩ 0.5 -50 -30 -10 10 30 50 70 90 110 130 150 TEMPERATURE (°C) Figure 4. Current-Limit Trip Point and VRDS(ON) vs. Temperature Compensation Design Guidelines The MAX15002 uses a fixed-frequency, voltage-mode control scheme that regulates the output voltage by differentially comparing the output voltage against a fixed reference. The subsequent error voltage that appears at the error amplifier output (COMP) is compared against an internal ramp voltage to generate the required duty cycle of the pulse-width modulator. A second order lowpass LC filter removes the switching harmonics and passes the DC component of the pulse-width-modulated signal to the output. The LC filter, which has an attenuation slope of -40dB/decade, introduces 180° of phase shift at frequencies above the LC resonant frequency. This phase shift, in addition to the inherent 180° of phase shift of the regulator’s self-governing (negative) feedback system, poses the potential for positive feedback. The error amplifier and its associated circuitry are designed to compensate for this instability to achieve a stable closed-loop system. The basic regulator loop consists of a power modulator (comprised of the regulator’s pulse-width modulator, associated circuitry, and LC filter), an output feedback divider, and an error amplifier. The power modulator has a DC gain set by VIN/VRAMP, where VRAMP’s amplitude is typically 2VP-P. The output filter is effectively modeled as a double pole and a single zero set by the output inductance (L), the output capacitance (COUT), the DC resistance of the inductor (DCR), and its equivalent series resistance (ESR). ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing 1 ≈ ⎛R +ESR ⎞ 2π L × COUT × ⎜ OUT ⎟ ⎝ ROUT +DCR ⎠ fESR = 1 2π L × COUT 1 2π ×ESR× COUT The switching frequency is programmable between 200kHz and 2.2MHz using an external resistor at RT. Typically, the crossover frequency (fCO), which is the frequency when the system’s closed-loop gain is equal to unity (crosses the 0dB axis)—should be set at or below one-tenth the switching frequency (fSW/10) for stable, closed-loop response. The MAX15002 provides an internal transconductance amplifier with its inverting input and its output available to the user for external frequency compensation. The flexibility of external compensation for each converter offers wide selection of output filtering components, especially the output capacitor. For cost-sensitive applications, use aluminum electrolytic capacitors and for space-sensitive applications, use low-ESR tantalum or multilayer ceramic chip (MLCC) capacitors at the output. The higher switching frequencies of the MAX15002 allow the use of MLCC as the primary filter capacitor(s). POWER MODULATOR AND TYPE II COMPENSATOR GAIN AND PHASE RESPONSE WITH LOSSY BULK OUTPUT CAPACITORS (ALUMINUM ELECTROLYTICS) POWER MODULATOR GAIN AND PHASE RESPONSE WITH LOSSY BULK OUTPUT CAPACITORS (ALUMINUM ELECTROLYTICS) MAX15002 fig05a 40 fLC |GMOD| ASYMPTOTE 60 < GMOD -20 0 fLC fESR -45 -40 -90 -60 -135 90 |GEA| 20 fCO 0 10 100 1k 10k 100k 1M -180 10M FREQUENCY (Hz) Figure 5a. Power Modulator Gain and Phase Response (Large, Bulk COUT) 45 0 -20 < GMOD fESR -45 -40 -80 180 135 <GEA 40 |GMOD| PHASE (DEGREES) 0 MAX15002 fig05b 80 90 45 20 MAGNITUDE (dB) Closed-Loop Response and Compensation of Voltage-Mode Regulators The power modulator’s LC lowpass filter exhibits a variety of responses, depending on the value of the L and C (and their parasitics). One such response is shown in Figure 5a. In this example, the power modulator’s uncompensated crossover is approximately 1/6th the desired crossover frequency, fCO. Note also, the uncompensated roll-off through the 0dB plane follows the double-pole, -40dB/dec slope and approaches 180° of phase shift, indicative of a potentially unstable system. Together with the inherent 180° of phase delay in the negative feedback system, this can lead to near 360° or positive feedback—an unstable system. The desired (compensated) roll-off follows a -20dB/dec slope (and commensurate 90° of phase shift), and, in this example, occurs at approximately 6x the uncompensated crossover frequency, fCO. In this example, a Type II compensator provides for stable closed-loop operation, leveraging the +20dB/dec slope of the capacitor’s ESR zero (see Figure 5b). MAGNITUDE (dB) fLC = VIN V = IN VRAMP 2V PHASE (DEGREES) GMOD(DC) = First, select the passive and active power components that meet the application’s output ripple, component size, and component cost requirements. Second, choose the small-signal compensation components to achieve the desired closed-loop frequency response and phase margin as outlined below. -90 -60 |GMOD| -135 1M -180 10M -80 10 100 1k 10k 100k FREQUENCY (Hz) Figure 5b. Power Modulator (Large, Bulk COUT) and Type II Compensator Responses ______________________________________________________________________________________ 19 MAX15002 Below are equations that define the power modulator: The Type II compensator’s mid-frequency gain (approximately 4dB shown here) is designed to compensate for the power modulator’s attenuation at the desired crossover frequency, fCO (GE/A + GMOD = 0dB at fCO). In this example, the power modulator’s inherent -20dB/decade roll-off above the ESR zero (f ESR ) is leveraged to extend the active regulation gain-bandwidth of the voltage regulator. As shown in Figure 5b, the net result is a 2x increase in the regulator’s gain bandwidth while providing greater than 55° of phase margin (the difference between G E/A and G MOD respective phases at crossover, fCO). Other filter schemes pose their own problems. For instance, when choosing high-quality filter capacitor(s), e.g., MLCCs, and inductor, with minimal parasitics, the inherent ESR zero can occur at a much higher frequency, as shown in Figure 5c. As with the previous example, the actual gain and phase response is overlaid on the power modulator’s asymptotic gain response. One readily observes the more dramatic gain and phase transition at or near the power modulator’s resonant frequency, fLC, versus the gentler response of the previous example. This is due to the component’s lower parasitics (OCR and ESR) and corresponding higher frequency of the inherent ESR As demonstrated in Figure 5d, the Type III’s mid-frequency double-zero gain (exhibiting a +20dB/dec slope, noting the compensator’s pole at the origin) is designed to compensate for the power modulator’s double-pole -40dB/decade attenuation at the desired crossover frequency, fCO (again, GE/A + GMOD = 0dB at fCO). See Figure 5d. In the above example, the power modulator’s inherent (mid-frequency) -40dB/decade roll-off is mitigated by the mid-frequency double zero’s +20dB/decade gain to extend the active regulation gain-bandwidth of the voltage regulator. As shown in Figure 5d, the net result is an approximate doubling in the regulator’s gain bandwidth while providing greater than 60° of phase margin (the difference between GE/A and GMOD respective phases at crossover, fCO). Design procedures for both Type II and Type III compensators are shown below. POWER MODULATOR GAIN AND PHASE RESPONSE WITH LOW-PARASITIC OUTPUT CAPACITORS (MLCCs) POWER MODULATOR AND TYPE III COMPENSATOR GAIN AND PHASE RESPONSE WITH LOW PARASITIC OUTPUT CAPACITORS (MLCCs) |GMOD| 20 zero frequency. In this example, the desired crossover frequency occurs below the ESR zero frequency. In this example, a compensator with an inherent midfrequency double-zero response is required to mitigate the effects of the filter’s double-pole. Such is available with the Type III topology. 90 MAX15002 fig05d 80 < GEA 60 45 |GEA| -20 fESR < GMOD -45 -40 -90 -60 -135 MAGNITUDE (dB) 0 fLC PHASE (DEGREES) 40 0 203 135 fLC 20 68 0 -20 fCO 0 |GMOD| -68 < GMOD -40 -80 10 100 1k 10k 100k 1M -180 10M FREQUENCY (Hz) Figure 5c. Power Modulator Gain and Phase Response (HighQuality COUT) 20 -135 -60 |GMOD| ASYMPTOTE 270 PHASE (DEGREES) MAX15002 fig05c 40 MAGNITUDE (dB) MAX15002 Dual-Output Buck Controller with Tracking/Sequencing fESR -80 10 100 1k 10k 100k 1M -203 -270 10M FREQUENCY (Hz) Figure 5d. Power Modulator (High-Quality COUT) and Type III Compensator Responses ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing 1) Calculate the fZERO,ESR and LC double pole, fLC: fESR = VOUT R1 fLC = FB - COMP gM VREF R2 1 2π ×ESR× COUT 1 2π × L × COUT 2) Calculate the unity-gain crossover frequency as: + f fCO ≤ SW RF CF MAX15002 Type II: Compensation When fCO > fESR 10 CCF 3) Determine RF from the following: V (2π × fCO ×L)VOUT RF = RAMP VOUT × VIN × gm ×ESR Figure 6a. Type II Compensation Network Note: RF is derived by setting the total loop gain at crossover frequency to unity, e.g., GEA(fCO) x GM(fCO) = 1V/V. The transconductance error amplifier gain is GEA(fCO) = gM x RF while the modulator gain is: GAIN (dB) GMOD (fCO ) = VIN V ESR × × FB VRAMP 2π × fCO ×L VOUT The total loop gain can be expressed logarithmically as follows: 1ST ASYMPTOTE GMODVREFVOUT-1(ωCF)-1 [ ] 20log10 gmRF + 2ND ASYMPTOTE GMODVREFVOUT-1RF 1ST POLE (AT ORIGIN) 1ST ZERO RFCF ⎡ ⎤ ESR× VIN × VFB ⎥ = 0 dB 20log10 ⎢ ⎢⎣ (2π × fCO ×L ) × VOUT × VRAMP ⎥⎦ 3RD ASYMPTOTE GMODVREFVOUT-1(ωCCF)-1 2ND POLE RFCCF ω (rad/sec) where V RAMP is the peak-to-peak ramp amplitude equal to 2V. 4) Place a zero at or below the LC double pole, fLC: Figure 6b. Type II Compensation Network Response When the fCO is greater than fESR, a Type II compensation network provides the necessary closed-loop response. The Type II compensation network provides a midband compensating zero and high-frequency pole (see Figures 6a and 6b). R F C F provides the midband zero f MID,ZERO , and RFCCF provides the high-frequency pole fHIGH,POLE. Use the following procedure to calculate the compensation network components. CF = 1 2π ×RF × fLC 5) Place a high-frequency pole at or below fP = 0.5 x fSW: CCF = 1 π ×RF × fSW 6) Choose an appropriately sized R1 (connected from OUT_ to FB_, start with a 10kΩ). Once R1 is selected, calculate R2 using the following equation: V FB R2 = R1 × VOUT − VFB where VFB = 0.6V. ______________________________________________________________________________________ 21 MAX15002 Dual-Output Buck Controller with Tracking/Sequencing Type III: Compensation when fCO < fESR As indicated above, the position of the output capacitor’s inherent ESR zero is critical in designing an appropriate compensation network. When low-ESR ceramic output capacitors are used, the ESR zero frequency (fESR) is usually much higher than unity crossover frequency (fCO). In this case, a Type III compensation network is recommended (see Figure 7a). VOUT R1 CI FB R2 CF RF COMP + GAIN (dB) 4TH ASYMPTOTE RFRIC 3RD ASYMPTOTE ωRFCI 1 1 = 2π × RF × (CF || CCF ) 2π × R × CF × CCF F CF + CCF 5TH ASYMPTOTE ωRICCF-1 Set the ratios of fCO-to-fZ and fP-to-fCO equal to one another, e.g., fCO = fP = 5 is a good number to get about fZ fCO 60° of phase margin at fCO. Whichever technique, it is important to place the two zeros at or below the double pole to avoid the conditional stability issue. The following procedure is recommended: 1) Select a crossover frequency, fCO, at or below onetenth the switching frequency: 2ND ASYMPTOTE RFRI-1 1ST POLE (AT ORIGIN) 1ST ZERO 2ND POLE RFCF RICI 2ND ZERO RICI f fCO ≤ SW 10 3RD POLE ω (rad/sec) RFCCF 2) Calculate the LC double-pole frequency, fLC : fLC = Figure 7b. Type III Compensation Network Response As shown in Figure 7b, the Type III compensation network introduces two zeros and three poles into the control loop. The error amplifier has a low-frequency pole at the origin, two zeros, and two higher frequency poles at the following frequencies: 1 fZ1 = 2π × RF × CF 1 fZ2 = 2π × CI × (R1 + RI) 22 1 2π × RI × CI fP3 attenuates the high-frequency output ripple. The locations of the zeros and poles should be such that the phase margin peaks around fCO. Figure 7a. Type III Compensation Network 1ST ASYMPTOTE ωRICF-1 fP2 = fP3 = gM VREF fP1 = at the origin (0Hz) fP1 introduces a pole at zero frequency (integrator) for nulling DC output-voltage errors. Depending on the location of the ESR zero (fESR), fP2 can be used to cancel it, or to provide additional attenuation of the high-frequency output ripple. CCF RI Two midband zeros (fZ1 and fZ2) are designed to cancel the pair of complex poles introduced by the LC filter. 1 2π× L × COUT 3) Select RF ≥ 10kΩ. 1 4) Place compensator’s first zero fZ1 = R 2 π × F × CF at or below the output filter’s double pole, fLC, as follows: CF = 1 2π × RF × 0.5 × fLC ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing GainMOD = 4 × 1 2 (2π × fCO ) × L × COUT The gain of the error amplifier (GainE/A) in midband frequencies is: GainE/A = 2π x fCO x CI x RF If a ceramic capacitor is used, the capacitor ESR zero, fESR, is likely to be located even above onehalf of the switching frequency, that is fLC < fCO < fSW/2 < fESR . In this case, the frequency of the second pole (fP2) should be placed high enough not to significantly erode the phase margin at the crossover frequency. For example, it can be set at 5 x fCO, so that its contribution to phase loss at the crossover frequency fCO is only about 11°: fP2 = 5 x fCO Once fP2 is known, calculate RI: The total loop gain as the product of the modulator gain and the error-amplifier gain at fCO should be equal to 1, as follows: GainMOD × GainE / A = 1 So : 4× 1 2 (2π × fCO ) × COUT × L SolvingforCI : CI = × 2π × fCO × CI × RF = 1 (2π × fCO × L × COUT ) 4 × RF 6) For those situations where f LC < f CO < f ESR < fSW/2—as with low-ESR tantalum capacitors—the compensator’s second pole (fP2) should be used to cancel fESR. This provides additional phase margin. Viewed mathematically on the system Bode plot, the loop gain plot maintains its +20dB/decade slope up to 1/2 of the switching frequency verses flattening out soon after the 0dB crossover. Then set: fP2 = fESR RI = 1 2π × fP2 × CI 7) Place the second zero (fZ2) at 0.2 x fCO or at fLC, whichever is lower and calculate R1 using the following equation: R1 = 1 2π × fZ2 × CI − RI 8) Place the third pole (fP3) at 1/2 the switching frequency and calculate CCF from: CCF = 1 2π × 0.5 × fSW × RF 9) Calculate R2 as: R2 = R1 × VFB VOUT − VFB where VFB = 0.6V. ______________________________________________________________________________________ 23 MAX15002 5) The gain of the modulator (GainMOD)—comprised of the regulator’s pulse-width modulator, LC filter, feedback divider, and associated circuitry—at crossover frequency is: PGND CIN 100μF 100nF BST2 2.2kΩ 47μF 1μH DH2 SEL FOM58660 270μF MAX15002 CSP2 71.5kΩ 10kΩ PGND2 SGND 499kΩ FOM58660 LX2 PHASE 680pF 100nF 1.58kΩ 2.7nF 11.0kΩ 100kΩ 44.2kΩ EN/TRACK2 CT 1.8V COMP2 EP 46.4kΩ 22.1kΩ 100pF SGND PGOOD2 30.1kΩ PGOOD1 ILIM1 COMP1 FB1 EN1 PGND1 CSP1 DL1 CSN1 LX1 DH1 BST1 DREG1 2.7nF 100nF 47μF 30.1kΩ 11.0kΩ FOM58660 2.2Ω 1.4μH FOM58690 100pF (1/2) CMFSH-31 100nF 150μF IN CSN2 SYNC 200kΩ 49.1kΩ DL2 RT ______________________________________________________________________________________ REG Figure 8. Coincident Dual Tracker with Lossless Current Sense 2.3μF 44.2kΩ 560pF 47.6kΩ 24 RESET 3.3V MAX15002 Dual-Output Buck Controller with Tracking/Sequencing Typical Operating Circuits 1.91kΩ 10kΩ ILIM2 FB2 (1/2)CMFSH-31 DREG2 IN Dual-Output Buck Controller with Tracking/Sequencing PGOOD1 ILIM1 COMP1 FB1 EN1 PGND1 CSP1 DL1 CSN1 VOUT1 LX1 DH1 BST1 DREG1 PGOOD2 ILIM2 EP RESET COMP2 FB2 CT EN/TRACK2 SGND SYNC MAX15002 RT CSP2 DL2 CSN2 LX2 PHASE DH2 SEL VOUT2 REG PGND2 BST2 DREG2 PGND CIN IN SGND IN Figure 9. Dual Sequencer with Lossless Current Sense ______________________________________________________________________________________ 25 MAX15002 Typical Operating Circuits (continued) Dual-Output Buck Controller with Tracking/Sequencing PGOOD1 ILIM1 COMP1 FB1 EN1 CSP1 PGND1 CSN1 DL1 LX1 DH1 BST1 DREG1 VOUT1 MAX15002 Typical Operating Circuits (continued) PGOOD2 ILIM2 EP RESET COMP2 FB2 EN/TRACK2 CT PGND2 REG RT MAX15002 SGND CSP2 DL2 LX2 PHASE DH2 SEL VOUT2 CSN2 SYNC BST2 DREG2 PGND CIN IN AGND IN Figure 10. Ratiometric Dual Tracker with Accurate Valley-Mode Current Sense 26 ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing Power Dissipation The 40-pin TQFN thermally enhanced package can dissipate up to 2.96W. Calculate power dissipation in the MAX15002 as a product of the input voltage and the total REG output current (IREG). IREG includes quiescent current (I Q ) and the total gate drive current (IDREG): PD = VIN x IREG IREG = IQ + [fSW x (QG1 + QG2 + QG3 + QG4)] where QG1 to QG4 are the total gate charge of the lowside and high-side external MOSFETs. f SW is the switching frequency of the converter and IQ is the quiescent current of the device at the switching frequency. Use the following equation to calculate the maximum power dissipation (PDMAX) in the chip at a given ambient temperature (TA): PDMAX = 37 x (150 - TA)……….mW PCB Layout Guidelines Use the following guidelines to layout the switching voltage regulator. 1) Place the IN, REG, and DREG_ bypass capacitors close to the MAX15002. 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 the current loop formed by the lower switching MOSFET, inductor and output capacitor short. 4) Keep SGND and PGND isolated and connect them at one single point close to the negative terminal of the input filter capacitor. 5) Run the current-sense lines CS+ and CS- close to each other to minimize the loop area. 6) Avoid long traces between the REG/DREG_ bypass capacitors, driver output of the MAX15002, MOSFET gates, and PGND. Minimize the loop formed by the REG_ bypass capacitors, bootstrap diode, bootstrap capacitor, high-side driver output of the MAX15002, and upper MOSFET gates. 7) Place the bank of output capacitors close to the load. 8) Distribute the power components evenly across the board for proper heat dissipation. 9) Provide enough copper area at and around the switching MOSFETs, and inductor to aid in thermal dissipation. 10) Connect the MAX15002 exposed paddle to a large copper plane to maximize its power dissipation capability. Connect the exposed paddle to SGND. Do not connect the exposed paddle to the SGND pin (pin 35) directly underneath the IC. 11) Use 2oz copper to keep the trace inductance and resistance to a minimum. Thin copper PCBs compromise efficiency because high currents are involved in the application. Also, thicker copper conducts heat more effectively, thereby reducing thermal impedance. ______________________________________________________________________________________ 27 MAX15002 PWM Controller Applications Information Pin Configuration Chip Information PROCESS: BiCMOS BST2 CSN2 CSP2 ILIM2 COMP2 EN/TRACK2 FB2 PGOOD2 N.C. N.C. TOP VIEW 30 29 28 27 26 25 24 23 22 21 N.C. 31 20 DH2 N.C. 32 19 LX2 N.C. 33 18 DREG2 17 DL2 SYNC 34 16 PGND2 SGND 35 MAX15002 RT 36 15 PGOOD1 14 FB1 PHASE 37 RESET 38 CT 39 13 EN1 EP* + 12 COMP1 11 ILIM1 7 8 9 10 CSN1 CSP1 6 DH1 5 BST1 PGND1 4 LX1 3 DL1 2 DREG1 1 SEL IN 40 REG MAX15002 Dual-Output Buck Controller with Tracking/Sequencing THIN QFN (6mm x 6mm) *EP = EXPOSED PAD. 28 ______________________________________________________________________________________ Dual-Output Buck Controller with Tracking/Sequencing QFN THIN.EPS ______________________________________________________________________________________ 29 MAX15002 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.) MAX15002 Dual-Output Buck Controller with Tracking/Sequencing Package Information (continued) (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.) 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. 30 ____________________Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600 © 2007 Maxim Integrated Products is a registered trademark of Maxim Integrated Products, Inc.