MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution General Description PIN-PACKAGE Pin Configuration DL BST LX DH TON CSH TOP VIEW 18 17 16 15 14 13 VDD 19 12 CSL PGND1 20 11 FB 10 REFIN 9 VTTI 8 VTT 7 PGND2 AGND 21 MAX17000A SKIP 22 VCC 23 *EP *EXPOSED PAD 3 4 5 6 VTTR 2 VTTS SSTL Memory Supplies 1 STDBY SHDN 24 DDR, DDR2, and DDR3 Memory Supplies Quick-PWM is a trademark of Maxim Integrated Products, Inc. TEMP RANGE PGOOD2 Notebook Computers PART MAX17000AETG+ -40C to +85C 24 Thin QFN-EP* +Denotes a lead(Pb)-free/RoHS-compliant package. *EP = Exposed pad. PGOOD1 Applications Ordering Information OVP The MAX17000A pulse-width modulation (PWM) controller provides a complete power solution for notebook DDR, DDR2, and DDR3 memory. It comprises a stepdown controller, a source-sink LDO regulator, and a reference buffer to generate the required VDDQ, VTT, and VTTR rails. The VDDQ rail is supplied by a step-down converter using Maxim's proprietary Quick-PWMTM controller. The high-efficiency, constant-on-time PWM controller handles wide input/output voltage ratios (low duty-cycle applications) with ease and provides 100ns response to load transients while maintaining a relatively constant switching frequency. The Quick-PWM architecture circumvents the poor load-transient timing problems of fixed-frequency current-mode PWMs while also avoiding the problems caused by widely varying switching frequencies in conventional constant-on-time and constant-off-time PWM schemes. The controller senses the current to achieve an accurate valley current-limit protection. It is also built in with overvoltage, undervoltage, and thermal protections. The MAX17000A can be set to run in three different modes: power-efficient SKIP mode, low-noise forced-PWM mode, and standby mode to support memory in notebook computer standby operation. The switching frequency is programmable from 200kHz to 600kHz to allow small components and high efficiency. The VDDQ output voltage can be set to a preset 1.8V or 1.5V, or be adjusted from 1.0V to 2.5V by an external resistor-divider. This output has 1% accuracy over line-and-load operating range. The MAX17000A includes a 2A source-sink LDO regulator for the memory termination VTT rail. This VTT regulator has a 5mV deadband that either sources or sinks, ideal for the fast-changing load burst present in memory termination applications. This feature also reduces output capacitance requirements. The VTTR reference buffer sources and sinks 3mA, providing the reference voltage needed by the memory controller and devices on the memory bus. The MAX17000A is available in a 24-pin, 4mm x 4mm, thin QFN package. Features o SMPS Regulator (VDDQ) Quick-PWM with 100ns Load-Step Response Output Voltages--Preset 1.8V, 1.5V, or Adjustable 1.0V to 2.5V 1% VOUT Accuracy Over Line and Load 26V Maximum Input Voltage Rating Accurate Valley Current-Limit Protection 200kHz to 600kHz Switching Frequency o Source/Sink Linear Regulator (VTT) 2A Peak Source/Sink Low-Output Capacitance Requirement Output Voltages-Preset VDDQ/2 or REFIN Adjustable from 0.5V to 1.5V o Soft-Start/Soft-Shutdown o SMPS Power-Good Window Comparator o VTT Power-Good Window Comparator o Selectable Overvoltage Protection o Undervoltage/Thermal Protections o 3mA Reference Buffer (VTTR) THIN QFN 4mm x 4mm For pricing, delivery, and ordering information, please contact Maxim Direct at 1-888-629-4642, or visit Maxim's website at www.maximintegrated.com. 19-4307; Rev 3; 4/13 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution ABSOLUTE MAXIMUM RATINGS TON to PGND1 .......................................................-0.3V to +28V VDD to PGND1..........................................................-0.3V to +6V VCC to VDD ............................................................-0.3V to +0.3V OVP to AGND ...........................................................-0.3V to +6V SHDN, STDBY, SKIP to AGND .................................-0.3V to +6V REFIN, FB, PGOOD1, PGOOD2 to AGND ................................-0.3V to (VCC + 0.3V) CSH, CSL to AGND ....................................-0.3V to (VCC + 0.3V) DL to PGND1..............................................-0.3V to (VDD + 0.3V) BST to PGND1...........................................................-1V to +34V BST to LX..................................................................-0.3V to +6V DH to LX ....................................................-0.3V to (VBST + 0.3V) BST to VDD .............................................................-0.3V to +28V VTTI to PGND2 .........................................................-0.3V to +6V VTT to PGND2 ............................................-0.3V to (VTTI + 0.3V) VTTS to AGND............................................-0.3V to (VCC + 0.3V) VTTR to AGND ..........................................-0.3V to (VCSL + 0.3V) PGND1, PGND2 to AGND.....................................-0.3V to +0.3V Continuous Power Dissipation (TA = +70C) 24-Pin, 4mm x 4mm Thin QFN (derated 27.8mW/C above +70C) ..........................2222mW Operating Temperature Range ...........................-40C to +85C Junction Temperature ......................................................+150C Storage Temperature Range .............................-65C to +150C Lead Temperature (soldering, 10s) .................................+300C Soldering Temperature (reflow) .......................................+260C 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 = 12V, VCC = VDD = V SHDN = VREFIN = 5V, VCSL = 1.8V, STDBY = SKIP = AGND, TA = 0C to +85C, unless otherwise noted. Typical values are at TA = +25C.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS PWM CONTROLLER Input Voltage Range Output-Voltage Accuracy Output-Voltage Range VIN 3 26 VCC, VDD 4.5 5.5 VCSL VIN = 4.5V to 26V, SKIP = VCC FB = AGND 1.485 1.500 1.515 FB = VCC 1.782 1.800 1.818 FB = Adj 0.99 1.000 1.01 VCSL 1 2.7 V V V Load Regulation Error VCSH - VCSL = 0 to 18mV, SKIP = VCC 0.1 % Line Regulation Error VDD = 4.5V to 5.5V, VIN = 4.5V to 26V 0.25 % Soft-Start Ramp Time t SSTART Rising edge of SHDN 1.4 Soft-Stop Ramp Time t SSTOP Falling edge of SHDN 2.8 ms 25 mV Soft-Stop Threshold On-Time Accuracy (Note 2) 2 t ON VIN = 12V, VCSL = 1.2V 2.1 RTON = 96.75k (600kHz), 167ns nominal -15 +15 RTON = 200k (300kHz), 333ns nominal -10 +10 RTON = 303.25k (200kHz), 500ns nominal -15 +15 ms % Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution ELECTRICAL CHARACTERISTICS (continued) (VIN = 12V, VCC = VDD = V SHDN = VREFIN = 5V, VCSL = 1.8V, STDBY = SKIP = AGND, TA = 0C to +85C, unless otherwise noted. Typical values are at TA = +25C.) (Note 1) PARAMETER Minimum Off-Time Quiescent Supply Current (VDD) Quiescent Supply Current (VCC) Shutdown Supply Current (VDD + VCC) TON Shutdown Current SYMBOL t OFF(MIN) IDD ICC ICC + IDD ITON CONDITIONS MIN TYP MAX UNITS (Note 2) 250 350 ns FB forced above 1.0V, STDBY = AGND or VCC, TA = +25C 0.01 1.00 A FB forced above 1.0V (PWM, VTT, and VTTR blocks); STDBY = VCC 2 4 mA FB forced above 1.0V (PWM and VTTR blocks); STDBY = AGND 900 1500 A SHDN = AGND, TA = +25C 0.01 5 A SHDN = AGND, VIN = 26V, VDD = 0 or 5V, TA = +25C 0.01 1.00 A LINEAR REGULATOR (VTT) VTTI Input Voltage Range VTTI VTTI Supply Current IVTTI SHDN = AGND, TA = +25C VTTI = 2.8V, REFIN = 1.4V, TA = +25C VTTI Shutdown Current REFIN Input Bias Current REFIN Range 1.0 VREFIN VTT Output-Accuracy Source Load VTT Output-Accuracy Sink Load V 50 A 10 A -50 +50 nA 0.5 1.5 V 10 VCC 0.3 REFIN Disable Threshold VTT Internal MOSFET 2.8 VTTI = 2.8V, REFIN = 1.4V, no load V High-side on-resistance (source, I VTT = 0.1A) 0.12 0.25 Low-side on-resistance (sink, I VTT = 0.1A) 0.18 0.36 (VREFIN - 5mV) or (VCSL/2 - 5mV) to VTTS, VTT = VTTS (VREFIN + 5mV) or (VCSL/2 + 5mV) to VTTS, VTT = VTTS VREFIN = 1V, I VTT = +50A -5 +5 mV VREFIN = 0.5V to 1.5V, I VTT = +300mA VREFIN = 1V, I VTT = -50A -5 -5 +5 mV VREFIN = 0.5V to 1.5V, I VTT = -300mA +5 VTT Load Regulation -50A to -1A I VTT +50A to +1A 13 VTT Line Regulation 1.0V VTTI 2.8V, I VTT = 100mA 1 VTT Current Limit 17 Source 2 4 -4 -2 With respect to internal VTT_EN signal VTT Discharge MOSFET OVP = VCC 16 VTTS Input Current TA = +25C 0.1 mV/A mV Sink VTT Current-Limit Soft-Start Time Maxim Integrated 160 A s 1.0 A 3 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution ELECTRICAL CHARACTERISTICS (continued) (VIN = 12V, VCC = VDD = V SHDN = VREFIN = 5V, VCSL = 1.8V, STDBY = SKIP = AGND, TA = 0C to +85C, unless otherwise noted. Typical values are at TA = +25C.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS REFERENCE BUFFER (VTTR) VTTR Output Accuracy (Adj) REFIN to VTTR VTTR Output Accuracy (Preset) VCSL/2 to VTTR VTTR Maximum Recommended Current Source/sink I VTT = 1mA -10 +10 I VTT = 3mA -20 +20 I VTT = 1mA -10 +10 I VTT = 3mA -20 +20 5 mV mA FAULT DETECTION (SMPS) SMPS OVP and PGOOD1 Upper Trip Threshold 12 SMPS OVP and PGOOD1 Upper Trip Threshold Fault-Propagation Delay t OVP SMPS Output Undervoltage Fault-Propagation Delay tUVP SMPS PGOOD1 Lower Trip Threshold PGOOD1 Lower Trip Threshold Propagation Delay FB forced 25mV above trip threshold Measured at FB, hysteresis = 25mV t PGOOD1 PGOOD1 Output Low Voltage -12 FB forced 50mV below PGOOD1 trip threshold 15 I PGOOD1 FB = 1V (PGOOD1 high impedance), PGOOD1 forced to 5V, TA = +25C TON POR Threshold VPOR(IN) Rising edge, PWM disabled below this level; hysteresis = 200mV % 10 s 200 s -15 -18 10 I SINK = 3mA PGOOD1 Leakage Current 18 % s 0.4 V 1 A 3.0 V FAULT DETECTION (VTT) PGOOD2 Upper Trip Threshold Hysteresis = 25mV 8 10 13 % PGOOD2 Lower Trip Threshold Hysteresis = 25mV -13 -10 -8 % VTTS forced 50mV beyond PGOOD2 trip threshold 10 s PGOOD2 Fault Latch Delay VTTS forced 50mV beyond PGOOD2 trip threshold 5 ms PGOOD2 Output Low Voltage I SINK = 3mA PGOOD2 Propagation Delay PGOOD2 Leakage Current t PGOOD2 I PGOOD2 VTTS = VREFIN (PGOOD2 high impedance), PGOOD2 forced to 5V, TA = +25C 0.4 V 1 A FAULT DETECTION Thermal-Shutdown Threshold VCC Undervoltage Lockout Threshold CSL Discharge MOSFET 4 T SHDN Hysteresis = 15C Rising edge, IC disabled below this level VUVLO(VCC) hysteresis = 200mV OVP = VCC C 160 3.8 4.1 16 4.4 V Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution ELECTRICAL CHARACTERISTICS (continued) (VIN = 12V, VCC = VDD = V SHDN = VREFIN = 5V, VCSL = 1.8V, STDBY = SKIP = AGND, TA = 0C to +85C, unless otherwise noted. Typical values are at TA = +25C.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS 17 20 25 mV CURRENT LIMIT Valley Current-Limit Threshold VLIMIT VCSH - VCSL Current-Limit Threshold (Negative) VNEG VCSH - VCSL, SKIP = VCC Current-Limit Threshold (Zero Crossing) VZX VPGND1 - VLX DH Gate-Driver On-Resistance RDH BST - LX forced to 5V 1.5 DL Gate-Driver On-Resistance RDL DL high 1.5 5.0 DL low 0.6 3.0 DH Gate-Driver Source/ Sink Current IDH -23 mV 1 mV SMPS GATE DRIVERS DL Gate-Driver Source/ Sink Current DH forced to 2.5V, BST - LX forced to 5V 1 IDL(SRC) DL forced to 2.5V 1 IDL(SNK) DL forced to 2.5V 3 Dead Time tDEAD Internal BST Switch On-Resistance RBST LX, BST Leakage Current DL rising, TA = +25C 10 25 DL falling, TA = +25C 15 35 IBST = 10mA, VDD = 5V internal design target 5.0 A A ns 4.5 VBST = VLX = 26V, SHDN = AGND, TA = +25C 0.001 20 A 1.65 2.00 V -1 +1 A -1 +1 A 100 A INPUTS AND OUTPUTS Logic-Input Threshold SHDN, STDBY, SKIP, OVP, rising edge hysteresis = 300mV/600mV (min/max) Logic-Input Current SHDN, STDBY, SKIP = 0 or VCC, TA = +25C Input Leakage Current Input Bias Current Maxim Integrated CSH = 0 or VCC, TA = +25C CSL = 0 or VCC 1.30 55 5 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution ELECTRICAL CHARACTERISTICS (VIN = 12V, VCC = VDD = VSHDN = VREFIN = 5V, VCSL = 1.8V, STDBY = SKIP = AGND, TA = -40C to +85C, unless otherwise noted.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN MAX UNITS PWM CONTROLLER Input Voltage Range Output-Voltage Accuracy On-Time Accuracy (Note 2) Minimum Off-Time Quiescent Supply Current (VCC) VIN 3 26 VCC, VDD 4.5 5.5 VCSL t ON t OFF(MIN) ICC VIN = 4.5V to 26V, SKIP = VCC VIN = 12V, VCSL = 1.2V FB = AGND 1.485 1.520 FB = VCC 1.782 1.820 FB = Adj 0.990 1.020 RTON = 96.75k (600kHz), 167ns nominal -15 +15 RTON = 200k (300kHz), 333ns nominal -10 +10 RTON = 303.25k (200kHz), 500ns nominal -15 +15 (Note 2) V V % 350 ns FB forced above 1.0V (PWM, VTT, and VTTR blocks); STDBY = VCC 4 mA FB forced above 1.0V (PWM and VTTR blocks); STDBY = AGND 1500 A LINEAR REGULATOR (VTT) VTTI Input Voltage Range VVTTI VTTI Supply Current IVTTI REFIN Range 1.0 VTTI = 2.8V, REFIN = 1.4V, no load VREFIN 0.5 VTT Load Regulation 6 V 50 A 1.5 V VCC 0.3 REFIN Disable Threshold VTT Internal MOSFET 2.8 V High-side on-resistance (source, IVTT = 0.1A) 0.25 Low-side on-resistance (sink, I VTT = 0.1A) 0.36 -50A to -1A I VTT +50A to +1A 17 mV/A Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution ELECTRICAL CHARACTERISTICS (continued) (VIN = 12V, VCC = VDD = VSHDN = VREFIN = 5V, VCSL = 1.8V, STDBY = SKIP = AGND, TA = -40C to +85C, unless otherwise noted.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN MAX I VTT = 1mA -10 +10 I VTT = 3mA -20 +20 I VTT = 1mA -10 +10 I VTT = 3mA -20 +20 UNITS REFERENCE BUFFER (VTTR) VTTR Output Accuracy (Adj) REFIN to VTTR VTTR Output Accuracy (Preset) VCSL/2 to VTTR mV mV FAULT DETECTION (SMPS) PGOOD1 Output Low Voltage I SINK = 3mA 0.4 V I SINK = 3mA 0.4 V FAULT DETECTION (VTT) PGOOD2 Output Low Voltage FAULT DETECTION VCC Undervoltage-Lockout Threshold VUVLO(VCC) Rising edge, IC disabled below this level; hysteresis = 200mV 4.0 4.4 V VCSH - VCSL 15 25 mV BST - LX forced to 5V 5 DL high 5 DL low 3 CURRENT LIMIT Valley Current-Limit Threshold VLIMIT SMPS GATE DRIVERS DH Gate-Driver On-Resistance DL Gate-Driver On-Resistance Dead Time RDH RDL tDEAD DL rising 10 DL falling 15 SHDN, STDBY, SKIP OVP, rising edge hysteresis = 300mV/600mV (min/max) 1.3 ns INPUTS AND OUTPUTS Logic-Input Threshold 2 V Note 1: Limits are 100% production tested at TA = +25C. Maximum and minimum limits over temperature are guaranteed by design and characterization. Note 2: On-time and off-time specifications are measured from 50% point at the DH pin with LX = GND, VBST = 5V, and a 250pF capacitor connected from DH to LX. Actual in-circuit times might differ due to MOSFET switching speeds. Maxim Integrated 7 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Typical Operating Characteristics (MAX17000A Circuit of Figure 1, VIN = 12V, VDD = VCC = 5V, SKIP = GND, TA = +25C, unless otherwise noted.) 50 PWM MODE STDBY = HIGH OR LOW 40 30 20 0.01 0.1 1 PWM MODE STDBY = HIGH OR LOW 40 20 VIN = 7V SKIP MODE STDBY = HIGH 50 30 10 70 60 SKIP MODE STDBY = HIGH 50 PWM MODE STDBY = HIGH OR LOW 40 20 VIN = 12V 0.01 0.1 1 VIN = 20V 10 10 0.01 0.1 1 10 LOAD CURRENT (A) LOAD CURRENT (A) SMPS 1.5V OUTPUT VOLTAGE vs. LOAD CURRENT SMPS SWITCHING FREQUENCY vs. LOAD CURRENT SMPS VALLEY-CURRENT LIMIT vs. INPUT VOLTAGE 300 250 200 150 SKIP MODE 0.01 0.1 1 10.25 100 VIN = 12V VOUT = 1.5V 0 0 10 10.00 9.75 2 4 6 8 9.50 4 10 8 10 PWM MODE, IIN STDBY = HIGH, SKIP MODE, ICC + IDD 1 STDBY = LOW, SKIP MODE, ICC + IDD 0.1 50 SAMPLE SIZE = 150 SAMPLE PERCENTAGE (%) PWM MODE, ICC + IDD MAX17000A toc07 NO LOAD 16 20 24 28 PRESET 1.5V OUTPUT VOLTAGE DISTRIBUTION NO-LOAD SUPPLY CURRENT vs. INPUT VOLTAGE 100 12 INPUT VOLTAGE (V) LOAD CURRENT (A) LOAD CURRENT (A) SUPPLY CURRENT (mA) RSENSE = 2m MAX17000A toc08 VIN = 12V MAX17000A toc06 PWM MODE CURRENT LIMIT (A) PWM MODE 10.50 MAX17000A toc05 MAX17000A toc04 1.50 350 50 TA = +85C TA = +25C 40 30 20 10 SKIP MODE, IIN 0 0.01 4 8 12 16 20 INPUT VOLTAGE (V) 8 80 LOAD CURRENT (A) SKIP MODE 1.49 0.001 SKIP MODE STDBY = LOW 90 30 10 10 1.51 OUTPUT VOLTAGE (V) 70 60 100 EFFICIENCY (%) SKIP MODE STDBY = HIGH 60 80 EFFICIENCY (%) 70 SKIP MODE STDBY = LOW 90 SWITCHING FREQUENCY (kHz) EFFICIENCY (%) 80 SMPS 1.5V EFFICIENCY vs. LOAD CURRENT MAX17000A toc02 SKIP MODE STDBY = LOW 90 100 MAX17000A toc01 100 SMPS 1.5V EFFICIENCY vs. LOAD CURRENT MAX17000A toc03 SMPS 1.5V EFFICIENCY vs. LOAD CURRENT 24 28 1.490 1.495 1.500 1.505 1.510 OUTPUT VOLTAGE (V) Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Typical Operating Characteristics (continued) (MAX17000A Circuit of Figure 1, VIN = 12V, VDD = VCC = 5V, SKIP = GND, TA = +25C, unless otherwise noted.) SHUTDOWN WAVEFORM (DISCHARGE MODE ENABLED) STARTUP WAVEFORM (HEAVY LOAD) VDDQ VTTR VDDQ TON VTT DL PGOOD2 VTT MAX17000A toc11 IVTT = 50mA STDBY VDDQ VTT DL SHDN STANDBY TRANSITION WAVEFORM MAX17000A toc10 MAX17000A toc09 PGOOD1 VTTR PGOOD1 LX SHDN ILX ILX ILX DL DL: 5V/div VDDQ: 2V/div VTT: 1V/div VTTR: 1V/div MAX17000A toc12 TON: 1V/div DL: 5V/div LX: 10V/div ILX: 2A/div SMPS LOAD-TRANSIENT RESPONSE (SKIP MODE) MAX17000A toc14 MAX17000A toc13 VDDQ VDDQ DL STDBY: 5V/div VDDQ: 1V/div VTT: 1V/div PGOOD2: 5V/div PGOOD1: 5V/div SHDN: 10V/div ILX: 2A/div SMPS LOAD-TRANSIENT RESPONSE (PWM MODE) STANDBY TRANSITION WAVEFORM STDBY VDDQ VTT TON 100s/div 400s/div 200s/div SHDN: 5V/div PGOOD1: 2V/div RLOAD = 0.25 VDDQ: 500mV/div ILX: 5A/div SKIP = GND VTT: 500mV/div DL: 5V/div VTTR: 500mV/div LX LX ILOAD ILOAD ILX ILX LX ILX 10s/div STDBY: 5V/div VDDQ: 1V/div VTT: 1V/div TON: 1V/div Maxim Integrated 20s/div 20s/div DL: 5V/div LX: 10V/div ILX: 2A/div VDDQ: 50mV/div LX: 10V/div ILOAD: 5A/div ILX: 5A/div VDDQ: 50mV/div LX: 10V/div ILOAD: 5A/div ILX: 5A/div 9 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Typical Operating Characteristics (continued) (MAX17000A Circuit of Figure 1, VIN = 12V, VDD = VCC = 5V, SKIP = GND, TA = +25C, unless otherwise noted.) VTT VOLTAGE vs. SOURCE/SINK LOAD CURRENT VTT OFFSET VOLTAGE DISTRIBUTION AT 300mA LOAD DL MAX17000A toc16 0.79 0.78 VDDQ VTT VOLTAGE (V) 0.77 VTT VTTR PGOOD2 PGOOD1 0.76 0.75 0.74 50 SAMPLE SIZE = 150 MAX17000A toc17 MAX17000A toc15 SAMPLE PERCENTAGE (%) OUTPUT OVERLOAD WAVEFORM TA = +85C TA = +25C 40 30 20 10 ILX 0.73 VTTI = 1.5V 0.72 400s/div 0 -2.0 -1.5 -1.0 -0.5 PGOOD2: 2V/div PGOOD1: 2V/div ILX: 10A/div TA = +85C TA = +25C 2.0 -15.0 40 30 20 10 -10.0 -7.5 -5.0 VTT OVERLOAD FAULT WAVEFORMS (5ms TIMER) MAX17000A toc20 SAMPLE SIZE = 150 TA = +85C TA = +25C 40 DL ILX 30 VDDQ VTT 20 VTTR 10 0 -12.5 OFFSET VOLTAGE (mV) 50 SAMPLE PERCENTAGE (%) MAX17000A toc18 SAMPLE PERCENTAGE (%) 1.5 VTT SINK CURRENT LIMIT 50 PGOOD1 PGOOD2 0 2.0 2.5 3.0 3.5 CURRENT LIMIT (A) 10 1.0 LOAD CURRENT (A) VTT SOURCE CURRENT LIMIT SAMPLE SIZE = 150 0.5 MAX17000A toc19 DL: 5V/div VDDQ: 1V/div VTT: 1V/div VTTR: 1V/div 0 4.0 -4.0 -3.5 -3.0 -2.5 CURRENT LIMIT (A) -2.0 1ms/div DL: 5V/div ILX: 2A/div VDDQ: 2V/div VTT: 1V/div VTTR: 1V/div PGOOD1: 2V/div PGOOD2: 2V/div Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Typical Operating Characteristics (continued) (MAX17000A Circuit of Figure 1, VIN = 12V, VDD = VCC = 5V, SKIP = GND, TA = +25C, unless otherwise noted.) VTT LOAD-TRANSIENT RESPONSE (SOURCE) IVTT BETWEEN 10mA AND 1.5A VTT LOAD-TRANSIENT RESPONSE (SINK) MAX17000A toc21 MAX17000A toc22 IVTT IVTT VTT_ac VTT_ac VDDQ = 1.5V VDDQ = 1.5V 20s/div 20s/div IVTT: 1A/div VTT: 20mV/div IVTT: 1A/div VTT: 20mV/div VTT LOAD-TRANSIENT RESPONSE (SOURCE-SINK) VTTR OUTPUT VOLTAGE vs. LOAD CURRENT MAX17000A toc23 MAX17000A toc24 0.79 0.78 0.77 OUTPUT VOLTAGE (V) IVTT VTT_ac 0.76 0.75 0.74 0.73 0.72 VDDQ = 1.5V 0.71 0.70 20s/div IVTT: 1A/div VTT: 50mV/div Maxim Integrated -6 -4 -2 0 2 4 6 LOAD CURRENT (mA) 11 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Pin Description PIN FUNCTION OVP OVP Mode Control. This input selectively enables/disables the SMPS OV protection feature and output discharge mode. When enabled, the SMPS OV protection feature is enabled. Connect OVP to the following voltage levels for the desired function: High (> 2.4V) = Enable SMPS OV protection, and SMPS and VTT discharge FETs. Low (GND) = Disable SMPS OV protection, and SMPS and VTT discharge FETs. PGOOD1 Open-Drain Power-Good Output. PGOOD1 is low when the SMPS output voltage is more than 15% (typ) beyond the normal regulation point, in standby, in shutdown, and during soft-start. After the soft-start circuit has terminated, PGOOD1 becomes high impedance if the SMPS output is in regulation. 3 PGOOD2 Open-Drain Power-Good Output. PGOOD2 is low when the VTT output voltage is more than 10% (typ) beyond the normal regulation point, in standby, in shutdown, and during soft-start. After the SMPS soft-start circuit has terminated, PGOOD2 becomes high impedance if the VTT output is in regulation. 4 STDBY Standby Control Input. When SHDN is high and STDBY is low, the MAX17000A turns off the VTT output (high-Z). When STDBY is high, normal SMPS operation resumes and the VTT output is enabled. 5 VTTS Sense Pin for Termination Supply Output. Normally connected to the VTT pin to allow accurate regulation to VCSL/2 or the REFIN voltage. 6 VTTR Termination Reference Buffer Output. VTTR tracks VCSL/2 when REFIN is connected to VCC. VTTR tracks VREFIN when a voltage between 0.5V to 1.5V is set at REFIN. Decouple VTTR to AGND with a 0.33F ceramic capacitor. 7 PGND2 8 VTT Termination Power-Supply Output. Connect VTT to VTTS to regulate the VTT voltage to the VTTS regulation setting. 9 VTTI Termination Power-Supply Input. VTTI is the input power supply to the VTT linear regulator. Normally connected to the output of the SMPS regulator for DDR applications. 10 REFIN External Reference Input. REFIN sets the feedback regulation voltage (VTTR = VTTS = VREFIN) of the MAX17000A. Connect REFIN to VCC to use the internal VCSL/2 divider. Connect a 0.5V to 1.5V voltage input to set the adjustable output for VTT, VTTS, and VTTR. 11 FB 1 2 12 NAME Power Ground for VTT. Connect PGND2 externally to the underside of the exposed pad. Feedback Input for SMPS Output. Connect to VCC for a fixed +1.8V output or to AGND for a fixed +1.5V output. For an adjustable output (1.0V to 2.7V), connect FB to a resistive divider from the output voltage. FB regulates to +1.0V. 12 CSL Negative Input of the PWM Output Current-Sense and Supply Input for VTTR. Connect CSL to the negative side of the output current-sensing resistor or the filtering capacitor if the DC resistance of the output inductor is utilized for current sensing. CSL is also the path for the internal 16 discharge MOSFET when VCC UVLO occurs with OVP enabled. 13 CSH Positive Input of the PWM Output Current Sense. Connect CSH to the positive side of the output current-sensing resistor or the filtering capacitor if the DC resistance of the output inductor is utilized for current sensing. Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Pin Description (continued) PIN NAME FUNCTION Switching Frequency Setting Input. An external resistor between the input power source and this pin sets the switching frequency per phase according to the following equation: 14 TON T SW = CTON x (RTON + 6.5k) where CTON = 16.26pF. TON is high impedance in shutdown. 15 DH High-Side Gate-Driver Output. Swings from LX to BST. DH is low when in shutdown or UVLO. 16 LX Inductor Connection. Connect LX to the switched side of the inductor as shown in Figure 1. 17 BST 18 DL Synchronous-Rectifier Gate-Driver Output. DL swings from VDD to PGND1. 19 VDD Supply Voltage Input for the DL Gate Driver and 3.3V Reference/Analog Supply. Connect to the system supply voltage (+4.5V to +5.5V). Bypass VDD to power ground with a 1F or greater ceramic capacitor. 20 PGND1 Power Ground. Ground connection for the low-side MOSFET gate driver. 21 AGND Analog Ground. Connect backside exposed pad to AGND. Boost Flying Capacitor Connection. Connect to an external 0.1F, 6V capacitor as shown in Figure 1. The MAX17000A contains an internal boost switch. 22 SKIP Pulse-Skipping Control Input. This input determines the mode of operation under normal steadystate conditions and dynamic output-voltage transitions: High (> 2.4V) = Forced-PWM operation Low (AGND) = Pulse-skipping mode 23 VCC Controller Supply Voltage. Connect to a 4.5V to 5.5V source. Bypass VCC to AGND with a 1F or greater ceramic capacitor. Shutdown Control Input. Connect to VCC for normal operation. When SHDN is pulled low, the MAX17000A slowly ramps down the output voltage to ground. When the internal target voltage reaches 25mV, the controller forces DL low, and enters the low current (1A) shutdown state. 24 SHDN When discharge mode is enabled by OVP (OVP = high), the CSL and VTT internal 16 discharge MOSFETs are enabled in shutdown. When discharge mode is disabled by OVP (OVP = low), LX, VTT, and VTTR are high impedance in shutdown. A rising edge on SHDN clears the fault OV protection latch. -- Maxim Integrated EP Exposed Pad. Connect backside exposed pad to AGND. 13 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Standard Application Circuits The MAX17000A standard application circuit (Figure 1) generates the VDDQ, VTT, and VTTR rails for DDR, DDR2, or DDR3 in a notebook computer. See Table 1 for component selections. Table 2 lists the component manufacturers. Table 3 is the operating mode truth table. Table 1. Component Selection for Standard Applications COMPONENT VOUT = 1.5V TO 1.8V AT 10A VOUT = 1.5V TO 1.8V AT 6A VIN = 7V TO 20V (300kHz) VIN = 7V TO 16V (500kHz) Input Capacitor (2x) 10F, 25V Taiyo Yuden TMK432BJ106KM 10F, 25V Taiyo Yuden TMK432BJ106KM Output Capacitor (2x) 330F, 2.5V ,12m (C2 case) SANYO 2R5TPE330MCC2 (2x) 220F, 2.5V, 21m (B2 case) SANYO 2R5TPE220MLB Inductor 1.4H, 12A, 3.4m (typ) Sumida CDEP105(L)NP-1R4 1.4H, 12A, 3.4m (typ) Sumida CDEP105(L)NP-1R4 Current-Sensing Resistor 2m, 0.5W (2010) Vishay WSL20102L000FEA 3m, 0.5W (2010) Vishay WSL20103L000FEA MOSFETs 30V, 20A n-channel MOSFET (high side) Fairchild FDMS8690; 30V, 40A n-channel MOSFET (low side) Fairchild FDMS8660S 30V 20A n-channel MOSFET (high side) Fairchild FDMS8690; 30V 40A n-channel MOSFET (low side) Fairchild FDMS8660S Table 2. Component Suppliers SUPPLIER PHONE WEBSITE INDUCTORS Dale (Vishay) NEC/TOKIN America, Inc. Panasonic Corp. 402-563-6866 (USA) www.vishay,com 510-324-4110 (USA) www.nec-tokinamerica.com 65-231-3226 (Singapore), 408-749-9714 (USA) www.panasonic.com Sumida Corp. 408-982-9660 (USA) www.sumida.com TOKO America, Inc. 858-675-8013 (USA) www.tokoam.com 843-448-9411 (USA) www.avxcorp.com 408-986-0424 (USA) www.kemet.com CAPACITORS AVX Corp. KEMET Corp. Panasonic Corp. 65-231-3226 (Singapore), 408-749-9714 (USA) www.panasonic.com SANYO Electric Co., Ltd. 81-72-870-6310 (Japan), 619-661-6835 (USA) www.sanyodevice.com Taiyo Yuden TDK Corp. 03-3667-3408 (Japan), 408-573-4150 (USA) 847-803-6100 (USA), 81-3-5201-7241 (Japan) www.t-yuden.com www.component.tdk.com SENSING RESISTORS Vishay 402-563-6866 (USA) www.vishay,com 800-341-0392 (USA) www.fairchildsemi.com Central Semiconductor Corp. 631-435-1110 www.centralsemi.com Nihon Inter Electronics Corp. 81-3-3343-84-3411 (Japan) MOSFET Fairchild Semiconductor DIODES 14 www.niec.co.jp Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Table 3. Operating Mode Truth Table SHDN 1 2 3 4 5 6 LH LH H H H H STDBY LH L LH H H L SKIP OPERATION X SMPS output ramps up in skip mode with a 1.4ms (typ) ramp time. PGOOD1 is held low until the SMPS output is in regulation. VTT and VTTR ramp up to the final voltage based on VCSL/2 or VREFIN. PGOOD2 is held low until VTT is in regulation. X SMPS output ramps up in skip mode with a 1.4ms ramp time. PGOOD1 is held low until the SMPS output is in regulation. VTT remains off throughout since STDBY is low. PGOOD2 stays low throughout. VTTR ramps up to the final voltage based on VCSL/2 or VREFIN. X Standby mode is exited and the full current capability of the MAX17000A is available. VTT ramps up after the internal SMPS block is ready. VTT ramps to the final voltage based on VCSL/2 or VREFIN. PGOOD2 goes high when VTT is in regulation. H SMPS is in forced-PWM mode. VTT and VTTR are enabled. PGOOD1 is high when the SMPS output is in regulation. PGOOD2 is high when VTT is in regulation. L SMPS is in skip mode. VTT and VTTR are enabled. PGOOD1 is high when the SMPS output is in regulation. PGOOD2 is high when VTT is in regulation. H SMPS is in forced-PWM mode. VTT is off and is in high impedance. PGOOD2 is forced low. VTTR is active and regulates to VCSL/2 or VREFIN. 7 H L L SMPS is in skip mode. VTT is off and is high impedance. PGOOD2 is forced low. VTTR is active and regulates to VCSL/2 or VREFIN. 8 HL H X Skip mode is exited as the MAX17000A ramps the output down to zero. VTTR tracks VCSL/2 or VREFIN during shutdown. After the SMPS output reaches 25mV, DL goes low. 9 HL L X Skip mode is exited as the MAX17000A ramps the output down to zero. VTTR tracks VCSL/2 or VREFIN during shutdown. After the SMPS output reaches 25mV, DL goes low. VTT is not enabled throughout soft-shutdown. 10 L X X DL low. Internal16 discharge MOSFETs on CSL and VTT enabled if OVP is high, but disabled if OVP is low. Maxim Integrated 15 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution 1 +5V R3 100k R2 100k BST 2 3 +5V 19 CVDD 1F TON OVP DH PGOOD1 PGOOD2 LX VDD DL PGND1 R1 10 PGND 23 5V VCC CVCC 1F 21 VCC ON/OFF 24 22 10 NL REQ RC D1 COUT 20 CEQ 11 RFBB FB OPTIONS: 1. CONNECT FB TO 5V FOR FIXED +1.8V. 2. CONNECT FB TO GND FOR FIXED +1.5V. 3. USE FB RESISTOR-DIVIDER FOR ADJUSTABLE OUTPUT VOLTAGES. STDBY SHDN VTTI SKIP PGND2 VCC VDDQ +1.8V OR 1.5V L1 RFBA FB 4 NH CBST 0.1F 16 18 RC RDCR REQ + RC RDCR = L1 x ( 1 + 1 ) CEQ REQ RC CIN 15 MAX17000A AGND RCS = 17 13 CSH 12 CSL AGND SLP_S3# VIN 7V TO 20V RTON 14 9 +1V TO + 2.5V CVTTI 7 CVTT REFIN VTT VTTS VTTR EP 8 VTT = VDDQ/2 5 6 VTTR = VDDQ/2 CVTTR 0.33F Figure 1. MAX17000A Standard Application Circuit Detailed Description The MAX17000A complete DDR solution comprises a step-down controller, a source-sink LDO regulator, and a reference buffer. Maxim's proprietary Quick-PWM pulsewidth modulator in the MAX17000A is specifically designed for handling fast load steps while maintaining a relatively constant operating frequency and inductor operating point over a wide range of input voltages. The Quick-PWM architecture circumvents the poor load-transient timing problems of fixed-frequency current-mode PWMs, while also avoiding the problems caused by widely varying switching frequencies in conventional constant-on-time and constant-off-time PWM schemes. Figure 1 is the MAX17000A standard application circuit and Figure 2 is the MAX17000A functional diagram. 16 The MAX17000A includes a 2A source-sink LDO regulator for the memory termination rail. The source-sink regulator features a dead band that either sources or sinks, ideal for the fast-changing short-period loads presenting in memory termination applications. This feature also reduces the VTT output capacitance requirement down to 1F, though load-transient response can still require higher capacitance values between 10F and 20F. The reference buffer sources and sinks 3mA, generating a reference rail for use in the memory controller and memory devices. Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution TON ON-TIME COMPUTE CSL tOFF(MIN) Q TON TRIG TRIG ONE-SHOT BST Q ONE-SHOT S DH Q R ERROR AMP LX VDD DL S R Q PGND1 SKIP 1.2V SMPS FAULT DETECTION OVF SMPS FAULT ZERO CROSSING INT_FB SMPS FAULT LATCH OVP VTT FAULT UVF 1mV CSL VALLEY CURRENT LIMIT 20mV 0.7V 10ms TIMER RUN SMPS RUN EA SHDN SOFT-START/ SOFT-STOP 1V REF INT_FB POWER-GOOD1 PGOOD1 CSH FB DECODE FB VCC OVF 1.15V VTT FAULT VTTR WINDOW COMPARATOR VTTR INT_REF MAX17000A AGND POWER-GOOD2 PGOOD2 1.4ms VTT WINDOW COMPARATOR VTTS VTTI VTT SMPS FAULT 5ms TIMER VTT FAULT VTT POS CURRENT LIMIT VDD VTT SS CURRENT LIMIT 5mV VTT STDBY VTT NEG CURRENT LIMIT PGND2 VDD - 0.3V VTTI VTT_EN VTT VDD REFIN 5mV VTT_EN CSL VTT CSL VCC R VTTR 16 UVLO RUN OVP PGND2 PGND2 CSL R 16 PGND2 PGND1 Figure 2. MAX17000A Functional Diagram Maxim Integrated 17 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution +5V Bias Supply (VDD, VCC) The MAX17000A requires an external 5V bias supply in addition to the battery. Typically, this 5V bias supply is the notebook's 95% efficient 5V system supply. Keeping the bias supply external to the IC improves efficiency and eliminates the cost associated with the 5V linear regulator that would otherwise be needed to supply the PWM circuit and gate drivers. If stand-alone capability is needed, the 5V supply can be generated with an external linear regulator such as the MAX1615. The 5V bias supply powers both the PWM controller and internal gate-drive power, so the maximum current drawn is: IBIAS = IQ + fSWQG(MOSFETs) = 2mA to 20mA (typ) where IQ is the current for the PWM control circuit, fSW is the switching frequency, and QG(MOSFETs) is the total gate-charge specification limits at VGS = 5V for the internal MOSFETs. Free-Running Constant-On-Time PWM Controller with Input Feed-Forward The Quick-PWM control architecture is a pseudo-fixedfrequency, constant on-time, current-mode regulator with voltage feed-forward. This architecture utilizes the output filter capacitor's ESR to act as a current-sense resistor, so the output ripple voltage can provide the PWM ramp signal. In addition to the general QuickPWM, the MAX17000A also senses the inductor current through DCR method or with a sensing resistor. Therefore, it is less dependent on the output capacitor ESR for stability. The control algorithm is simple: the high-side switch on-time is determined solely by a oneshot whose pulse width is inversely proportional to input voltage and directly proportional to output voltage. Another one-shot sets a minimum off-time (250ns typ). The on-time one-shot is triggered if the error comparator is low, the low-side switch current is below the valley current-limit threshold, and the minimum off-time oneshot has timed out. On-Time One-Shot The heart of the PWM core is the one-shot that sets the high-side switch on-time. This fast, low-jitter, adjustable one-shot includes circuitry that varies the on-time in response to battery and output voltages. The high-side switch on-time is inversely proportional to the battery voltage as measured by the VIN input, and proportional to the output voltage. An external resistor between the input power source and TON pin sets the switching frequency per phase according to the following equation: 18 tON = CTON x (RTON + 6.5k) x (VCSL + 0.075V) VIN fSW = 1 CTON x (RTON + 6.5k) where CTON = 16.26pF, and 0.075V is an approximation to accommodate for the expected drop across the low-side MOSFET switch. This algorithm results in a nearly constant switching frequency despite the lack of a fixed-frequency clock generator. For loads above the critical conduction point, where the dead-time effect is no longer a factor, the actual switching frequency is: fSW = VOUT + VDIS tON x (VIN - VCHG + VDIS ) where VDIS is the sum of the parasitic voltage drops in the inductor discharge path, including synchronous rectifier, inductor, and PCB resistances; VCHG is the sum of the parasitic voltage drops in the charging path, including the high-side switch, inductor, and PCB resistances; and t ON is the on-time calculated by the MAX17000A. Automatic Pulse-Skipping Mode (SKIP = AGND) In skip mode (SKIP = AGND), an inherent automatic switchover to PFM takes place at light loads. This switchover is affected by a comparator that truncates the low-side switch on-time at the inductor current's zero crossing. DC output-accuracy specifications refer to the threshold of the error comparator. When the inductor is in continuous conduction, the MAX17000A regulates the valley of the output ripple, so the actual DC output voltage is higher than the trip level by 50% of the output ripple voltage. In discontinuous conduction (SKIP = AGND and IOUT < ILOAD(SKIP)), the output voltage has a DC regulation level higher than the error-comparator threshold by approximately 1.5% due to slope compensation. However, the internal integrator corrects for most of it, resulting in very little load regulation. The MAX17000A always uses skip mode during startup, regardless of the SKIP and STDBY setting. The SKIP and STDBY controls take effect after soft-start is done. See Figure 3. Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution VIN - VOUT L INDUCTOR CURRENT I = t IPEAK ILOAD = IPEAK/2 reverses at light loads while DH maintains a duty factor of VOUT/VIN. The benefit of forced-PWM mode is to keep a fairly constant switching frequency. However, forcedPWM operation comes at a cost: the no-load 5V bias current remains between 2mA to 20mA, depending on the switching frequency. STDBY = AGND overrides the SKIP pin setting, forcing the MAX17000A into standby. The MAX17000A switches to forced-PWM mode during shutdown, regardless of the state of SKIP and STDBY levels. Standby Mode (STDBY) 0 ON-TIME It should be noted that standby mode in the MAX17000A corresponds to computer system standby operation, and is not referring to the MAX17000A shutdown status. TIME Figure 3. Pulse-Skipping/Discontinuous Crossover Point Forced-PWM Mode (SKIP = VCC) The low-noise forced-PWM mode (SKIP = VCC) disables the zero-crossing comparator, which controls the lowside switch on-time. This forces the low-side gate-drive waveform to constantly be the complement of the highside gate-drive waveform, so the inductor current When standby mode is enabled (STDBY = AGND), VTT is disabled (high impedance) but VTTR remains active. When standby mode is disabled (STDBY = VCC), the VTT block is enabled and the VTT output capacitor is charged. The VTT soft-start current limit increases linearly from zero to its maximum current limit in 160s (typ), keeping the input VTTI inrush low. See Figure 4. STDBY SMPS OUTPUT VTTR OUTPUT VTT OUTPUT VTT HIGH-IMPEDANCE VTT CURRENT LIMIT 160s PGOOD1 PGOOD2 STANDBY TIMING Figure 4. MAX17000A Standby Mode Timing Maxim Integrated 19 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Valley Current-Limit Protection The MAX17000A uses the same valley current-limit protection employed on all Maxim Quick-PWM controllers. If the current exceeds the valley current-limit threshold, the PWM controller is not allowed to initiate a new cycle. The actual peak current is greater than the valley current-limit threshold by an amount equal to the inductor ripple current. Therefore, the exact current-limit characteristic and maximum load capability are a function of the inductor value and battery voltage. When combined with the undervoltage-protection circuit, this current-limit method is effective in almost every circumstance. In forced-PWM mode, the MAX17000A also implements a negative current limit to prevent excessive reverse inductor currents when VOUT is sinking current. The negative current-limit threshold is set to approximately 115% of the positive current limit. See Figure 5. IPEAK INDUCTOR CURRENT ILOAD ILIMIT ( LIR2 ) ILIM(VAL) = ILOAD(MAX) 1- 0 TIME Figure 5. Valley Current-Limit Threshold Point Power-Good Outputs (PGOOD1 and PGOOD2) The MAX17000A features two power-good outputs. PGOOD1 is the open-drain output for a window comparator that continuously monitors the SMPS output. PGOOD1 is actively held low in shutdown and during soft-start and soft-shutdown. After the soft-start terminates, PGOOD1 becomes high impedance as long as the SMPS output voltage is between 115% (typ) and 85% (typ) of the regulation voltage. When the SMPS output voltage exceeds the 115%/85% regulation window, the MAX17000A pulls PGOOD1 low. Any fault condition on the SMPS output forces PGOOD1 and PGOOD2 low and latches off until the fault latch is cleared by toggling SHDN or cycling VCC power below 1V. Detection of an OVP event immediately pulls PGOOD1 low, regardless of the OVP state (OVP enabled or disabled). 20 PGOOD2 is the open-drain output for a window comparator that continuously monitors the VTT output. PGOOD2 is actively held low in standby, shutdown, and during soft-start. PGOOD2 becomes high impedance as long as the VTT output voltage is within 10% of the regulation voltage. When the VTT output exceeds the 10% threshold, the MAX17000A pulls PGOOD2 low. If PGOOD2 remains low for 5ms (typ), the MAX17000A latches off with the soft-shutdown sequence. For logic-level output voltages, connect an external 100k pullup resistor from PGOOD1 and PGOOD2 to VDD. POR, UVLO Power-on reset (POR) occurs when VCC rises above approximately 2V, resetting the fault latch and soft-start circuit and preparing the controller for power-up. When OVP protection is enabled, a rising edge on POR turns on the 16 discharge MOSFET on CSL and VTT. When OVP is disabled, the internal 16 discharge MOSFETs on CSL and VTT also remain off. V CC undervoltage lockout (UVLO) circuitry inhibits switching until VCC reaches 4.1V (typ). When VCC rises above 4.1V, the controller activates the PWM controller and initializes soft-start. When VCC drops below the UVLO threshold (falling edge), the controller stops, DL is pulled low, and the internal 16 discharge MOSFETs on the CSL and VTT outputs are enabled, if OVP is enabled. Soft-Start and Soft-Shutdown Soft-start and soft-shutdown for the MAX17000A PWM block is voltage based. Soft-start begins when SHDN is driven high. During soft-start, the PWM output is ramped up from 0V to the final set voltage in 1.4ms. This reduces inrush current and provides a predictable ramp-up time for power sequencing. The MAX17000A always uses skip mode during startup, regardless of the SKIP and STDBY setting. The SKIP and STDBY controls take effect after soft-start is done. The MAX17000A VTT LDO regulator uses a current-limited soft-start function. When the VTT block is enabled, the internal source and sink current limits are linearly increased from zero to the full-scale limit in 160s. Fullscale current limit is available when the VTT output is in regulation, or after 160s, whichever is earlier. The VTTR reference buffer does not have any soft-start control. Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution SHDN STDBY INT_REF REFOK SMPS_RUNOK 1.4ms 2.8ms 25mV SMPS OUTPUT VTT OUTPUT VTTR OUTPUT VTT CURRENT LIMIT PGOOD1 160s PGOOD2 DL SKIP FPWM VTT 16 FET CSL 16 FET Figure 6. MAX17000A Startup/Shutdown Timing with OVP Enabled Soft-shutdown begins after SHDN goes low, an output undervoltage fault occurs, or a thermal fault occurs. A fault on the SMPS (UV fault for more than 200s (typ)), or fault on the VTT output that persists for more than 5ms (typ), triggers shutdown of the whole IC. During soft-shutdown, the output is ramped down to 0V in 2.8ms, reducing negative inductor currents that can cause negative voltages on the output. At the end of soft-shutdown, DL is driven low. Maxim Integrated When OVP is enabled (OVP = VCC), the internal 16 discharging MOSFETs on CSL and VTT are enabled until startup is triggered again by a rising edge of SHDN. When OVP is disabled (OVP = AGND), the CSL and VTT internal 16 discharging MOSFETs are not enabled in shutdown. Output Fault Protection The MAX17000A provides overvoltage/undervoltage fault protections for the PWM output. Drive OVP to enable and disable fault protection as shown in Table 4. 21 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Table 4. Fault Protection and Shutdown Setting Truth Table OVP OVP Disabled Discharge Disabled (OVP = Low) MODE REACTION/DRIVER STATE Shutdown (SHDN = low) DL immediately pulled low. VTTR tracks the SMPS output during soft-shutdown. CSL and VTT are high impedance at the end of soft-shutdown (16 discharge MOSFETs disabled). Outputs highimpedance in shutdown. SMPS UVP DL immediately pulled low. VTTR tracks the SMPS output during soft-shutdown. CSL and VTT are high impedance at the end of soft-shutdown (16 discharge MOSFETs disabled). SMPS latched fault condition. SMPS OVP (disabled) Controller remains active (normal operation). Note: An OVP detection still pulls PGOOD1 low. PGOOD2 immediately pulled low. VTT < -90% or Soft-shutdown initiated if fault persists for more than 5ms (typ). DH VTT > +110% not used in soft-shutdown. DL low after soft-shutdown completed. VTTR tracks the SMPS output soft-shutdown. OVP Enabled Discharge Enabled (OVP = High) 22 Only PGOOD1 pulled low; fault not latched. VTT latched fault condition if fault persists for more than 5ms (typ). VCC UVLO falling edge DL and DH immediately pulled low. PGOOD1 and PGOOD2 immediately forced low. VTT and VTTR blocks immediately disabled (high impedance, no 16 discharge on outputs). Shutdown (SHDN = low) Soft-shutdown initiated. DL high after soft-shutdown completed. VTTR tracks the SMPS output during soft-shutdown. Internal 16 discharge MOSFETs on CSL and VTT enabled after soft-shutdown. 16 discharge MOSFETs on CSL and VTT enabled in shutdown. SMPS UVP Soft-shutdown initiated. DH not used in soft-shutdown. DL low after soft-shutdown completed. VTTR tracks the SMPS output during soft-shutdown. Internal 16 discharge MOSFETs on CSL and VTT enabled after soft-shutdown. SMPS latched fault condition. SMPS OVP (enabled) DL immediately latched high, DH forced low. PGOOD1 and PGOOD2 immediately forced low. VTT and VTTR blocks immediately shut down. Internal 16 discharge MOSFETs on CSL and VTT enabled. SMPS latched fault condition. PGOOD2 immediately pulled low. Soft-shutdown initiated if fault persists for more than 5ms (typ). DH not used in soft-shutdown. DL low after soft-shutdown completed. VTTR tracks the SMPS output during soft-shutdown. Internal 16 discharge MOSFETs on CSL and VTT enabled after soft-shutdown. VTT latched fault condition if fault persists for more than 5ms (typ). VTT < 90% or VTT > 110% OVP Enabled Discharge Enabled (OVP = High) COMMENT VCC UVLO falling edge DL and DH immediately pulled low. PGOOD1 and PGOOD2 immediately forced low. VTT and VTTR blocks immediately disabled. Internal 16 discharge MOSFETs on CSL and VTT enabled immediately. -- -- Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Table 4. Fault Protection and Shutdown Setting Truth Table (continued) OVP MODE Thermal fault General Shutdown and Fault Conditions REACTION/DRIVER STATE Active-fault condition. VCC UVLO rising edge Activate INT_REF once VCC rises above UVLO, and SHDN = high. Once REFOK is valid (high), initiate the soft-start sequence. DL remains low until switching/soft-start begins. -- VCC POR rising edge DL forced low. -- VCC POR falling edge DL = Don't care. VCC less than 2VT is not sufficient to turn on the MOSFETs. -- SMPS Overvoltage Protection (OVP) If the output voltage of the SMPS rises 115% above its nominal regulation voltage while OVP is enabled (OVP = VCC), the controller sets its overvoltage fault latch, pulls PGOOD1 and PGOOD2 low, and forces DL high. The VTT and VTTR block shut down immediately, and the internal 16 discharge MOSFETs on CSL and VTT are turned on. If the condition that caused the overvoltage persists (such as a shorted high-side MOSFET), the battery fuse blows. Cycle VCC below 1V or toggle SHDN to clear the overvoltage fault latch and restart the controller. OVP is disabled when OVP is connected to AGND (Table 4). PGOOD1 upper threshold remains active at 115% of nominal regulation voltage even when OVP is disabled and the 16 discharge MOSFETs on CSL and VTT are not enabled in shutdown. SMPS Undervoltage Protection (UVP) If the output voltage of the SMPS falls below 85% of its regulation voltage for more than 200s (typ), the controller sets its undervoltage fault latch, pulls PGOOD1 and PGOOD2 low, and begins soft-shutdown pulsing DL. DH remains off during the soft-shutdown sequence initiated by an undervoltage fault. After soft-shutdown has completed, the MAX17000A forces DL and DH low, and enables the internal 16 discharge MOSFETs on CSL and VTT. Cycle VCC below 1V or toggle SHDN to clear the undervoltage fault latch and restart the controller. VTT Overvoltage and Undervoltage Protection If the output voltage of the VTT regulator exceeds 10% of its regulation voltage for more than 5ms (typ), the controller sets its fault latch, pulls PGOOD1 and PGOOD2 low, and begins soft-shutdown pulsing DL. DH remains off during the soft-shutdown sequence initiated by an undervoltage fault. After soft-shutdown has Maxim Integrated COMMENT DL and DH immediately pulled low. PGOOD1 and PGOOD2 immediately forced low. VTT and VTTR blocks immediately disabled (high impedance, no 16 discharge on outputs). completed, the MAX17000A forces DL and DH low, and enables the internal 16 discharge MOSFETs on CSL and VTT. Cycle VCC below 1V or toggle SHDN to clear the undervoltage fault latch and restart the controller. Thermal-Fault Protection The MAX17000A features a thermal-fault protection circuit. When the junction temperature rises above +160C, a thermal sensor activates the fault latch, pulls PGOOD1 and PGOOD2 low, and shuts down using the shutdown sequence. Toggle SHDN or cycle VCC power below VCC POR to reactivate the controller after the junction temperature cools by 15C. Design Procedure Firmly establish the input voltage range and maximum load current before choosing a switching frequency and inductor operating point (ripple-current ratio). The primary design trade-off lies in choosing a good switching frequency and inductor operating point, and the following four factors dictate the rest of the design: * Input Voltage Range: The maximum value (VIN(MAX)) must accommodate the worst-case input supply voltage allowed by the notebook's AC adapter voltage. The minimum value (V IN(MIN) ) must account for the lowest input voltage after drops due to connectors, fuses, and battery selector switches. If there is a choice at all, lower input voltages result in better efficiency. * Maximum Load Current: There are two values to consider. The peak load current (ILOAD(MAX)) determines the instantaneous component stresses and filtering requirements, and thus drives output capacitor selection, inductor saturation rating, and the design of the current-limit circuit. The continuous load current (ILOAD) determines the thermal 23 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution stresses and thus drives the selection of input capacitors, MOSFETs, and other critical heat-contributing components. Most notebook loads generally exhibit ILOAD = ILOAD(MAX) x 80%. * * Switching Frequency: This choice determines the basic trade-off between size and efficiency. The optimal frequency is largely a function of maximum input voltage, due to MOSFET switching losses that are proportional to frequency and VIN2. The optimum frequency is also a moving target, due to rapid improvements in MOSFET technology that are making higher frequencies more practical. Inductor Operating Point: This choice provides trade-offs between size vs. efficiency and transient response vs. output noise. Low inductor values provide better transient response and smaller physical size, but also result in lower efficiency and higher output noise due to increased ripple current. The minimum practical inductor value is one that causes the circuit to operate at the edge of critical conduction (where the inductor current just touches zero with every cycle at maximum load). Inductor values lower than this grant no further size-reduction benefit. The optimum operating point is usually found between 20% and 50% ripple current. Inductor Selection The switching frequency and operating point (% ripple current or LIR) determine the inductor value as follows: VOUT VIN - VOUT L= x fSW x ILOAD(MAX) x LIR VIN Find a low-loss inductor having the lowest possible DC resistance that fits in the allotted dimensions. Ferrite cores are often the best choice, although powdered iron is inexpensive and can work well at 200kHz. The core must be large enough not to saturate at the peak inductor current (IPEAK): LIR IPEAK = ILOAD(MAX) x 1 + 2 Setting the Valley Current Limit The minimum current-limit threshold must be high enough to support the maximum load current when the current limit is at the minimum tolerance value. The valley of the inductor current occurs at ILOAD(MAX) minus half the ripple current; therefore: LIR ILIMIT(LOW) > ILOAD(MAX) x 1 2 where I LIMIT(LOW) equals the minimum current-limit threshold voltage divided by the output sense element (inductor DCR or sense resistor). The valley current limit is fixed at 17mV (min) across the CSH to CSL differential input. Special attention must be made to the tolerance and thermal variation of the on-resistance in the case of DCR sensing. Use the worst-case maximum value for RDCR from the inductor data sheet, and add some margin for the rise in RDCR with temperature. A good general rule is to allow 0.5% additional resistance for each degree Celsius of temperature rise, which must be included in the design margin unless the design includes an NTC thermistor in the DCR network to thermally compensate the current-limit threshold. The current-sense method (Figure 7) and magnitude determine the achievable current-limit accuracy and power loss. The sense resistor can be determined by: RSENSE = VLIMIT/ILIMIT INPUT (VIN) DH NH CIN SENSE RESISTOR L LESL RSENSE CEQREQ = LX MAX17000A DL NL DL REQ CEQ LESL RSENSE COUT PGND1 CSH CSL A) OUTPUT SERIES RESISTOR SENSING Figure 7a. Current-Sense Configurations (Sheet 1 of 2) 24 Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution INPUT (VIN) DH NH CIN INDUCTOR L RDCR RCS = LX MAX17000A DL NL DL R1 PGND1 R2 R2 RDCR R1 + R2 COUT RDCR = CEQ L CEQ [ R11 + R21 ] CSH CSL B) LOSSLESS INDUCTOR SENSING FOR THERMAL COMPENSATION: R2 SHOULD CONSIST OF AN NTC RESISTOR IN SERIES WITH A STANDARD THIN-FILM RESISTOR. Figure 7b. Current-Sense Configurations (Sheet 2 of 2) For the best current-sense accuracy and overcurrent protection, use a 1% tolerance current-sense resistor between the inductor and output as shown in Figure 7a. This configuration constantly monitors the inductor current, allowing accurate current-limit protection. However, the parasitic inductance of the current-sense resistor can cause current-limit inaccuracies, especially when using low-value inductors and current-sense resistors. This parasitic inductance (LESL) can be cancelled by adding an RC circuit across the sense resistor with an equivalent time constant: CEQ x REQ = LESL RSENSE Alternatively, low-cost applications that do not require highly accurate current-limit protection could reduce the overall power dissipation by connecting a series RC circuit across the inductor (Figure 7b) with an equivalent time constant: RCS = R2 x RDCR R1 + R2 and: RDCR = L 1 1 x + CEQ R1 R2 where RCS is the required current-sense resistance, and RDCR is the inductor's series DC resistance. Use the worst-case inductance and RDCR values provided by the inductor manufacturer, adding some margin for the inductance drop over temperature and load. Maxim Integrated MOSFET Gate Drivers (DH, DL) The DH and DL drivers are optimized for driving moderate-sized high-side, and larger low-side power MOSFETs. This is consistent with the low duty factor seen in notebook applications, where a large VIN - VOUT differential exists. The high-side gate driver (DH) sources and sinks 1.2A, and the low-side gate driver (DL) sources 1.0A and sinks 2.4A. This ensures robust gate drive for high-current applications. The DH floating high-side MOSFET driver is powered by an internal boost switch charge pump at BST, while the DL synchronous-rectifier driver is powered directly by the 5V bias supply (VDD). PWM Output Capacitor Selection The output filter capacitor must have low enough effective series resistance (ESR) to meet output ripple and load-transient requirements, yet have high enough ESR to satisfy stability requirements. In core and chipset converters and other applications where the output is subject to large-load transients, the output capacitor's size typically depends on how much ESR is needed to prevent the output from dipping too low under a load transient. Ignoring the sag due to finite capacitance: V (RESR + RPCB ) I STEP LOAD(MAX) In low-power applications, the output capacitor's size often depends on how much ESR is needed to maintain an acceptable level of output ripple voltage. The output ripple voltage of a step-down controller equals the total inductor ripple current multiplied by the output capacitor's ESR. 25 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution The maximum ESR to meet ripple requirements is: VIN x fSW x L RESR x VRIPPLE ( VIN - VOUT ) x VOUT where fSW is the switching frequency. With most chemistries (polymer, tantalum, aluminum, electrolytic), the actual capacitance value required relates to the physical size needed to achieve low ESR and the chemistry limits of the selected capacitor technology. Ceramic capacitors provide low ESR, but the capacitance and voltage rating (after derating) are determined by the capacity needed to prevent VSAG and VSOAR from causing problems during load transients. Generally, once enough capacitance is added to meet the overshoot requirement, undershoot at the rising load edge is no longer a problem. Thus, the output capacitor selection requires carefully balancing capacitor chemistry limitations (capacitance vs. ESR vs. voltage rating) and cost. PWM Output Capacitor Stability Considerations For Quick-PWM controllers, stability is determined by the in-phase feedback ripple relative to the switching frequency, which is typically dominated by the output ESR. The boundary of instability is given by the following equation: fSW 1 2 x REFF x COUT REFF = RESR + ACS x RSENSE where COUT is the total output capacitance, RESR is the total equivalent series resistance of the output capacitors, RSENSE is the effective current-sense resistance (see Figure 7), and ACS is the current-sense gain of 2. For a standard 300kHz application, the effective zero frequency must be well below 95kHz, preferably below 50kHz. With these frequency requirements, standard tantalum and polymer capacitors already commonly used have typical ESR zero frequencies below 50kHz, allowing the stability requirements to be achieved without any additional current-sense compensation. In the standard application circuit (Figure 1), the ESR needed to support a 15mV P-P ripple is 15mV/(10A x 0.3) = 5m. Two 330F, 9m polymer capacitors in parallel provide 4.5m (max) ESR and 1/(2 x 330F x 9m) = 53kHz ESR zero frequency. Ceramic capacitors have a high-ESR zero frequency, but applications with sufficient current-sense compensation can still take advantage of the small size, low ESR, and high reliability of the ceramic chemistry. By the inductor current DCR sensing, applications with 26 ceramic output capacitors can be compensated using either a DC-compensation or AC-compensation method. The DC-coupling requires fewer external compensation capacitors, but this also creates an output load line that depends on the inductor's DCR (parasitic resistance). Alternatively, the current-sense information can be AC-coupled, allowing stability to be dependent only on the inductance value and compensation components and eliminating the DC load line. When only using ceramic output capacitors, output overshoot (VSOAR) typically determines the minimum output capacitance requirement. Their relatively low capacitance value can allow significant output overshoot when stepping from full-load to no-load conditions, unless a small inductor value and high switching frequency are used to minimize the energy transferred from inductor to capacitor during load-step recovery. Unstable operation manifests itself in two related, but distinctly different ways: double pulsing and feedback loop instability. Double pulsing occurs due to noise on the output or because the ESR is so low that there is not enough voltage ramp in the output voltage signal. This "fools" the error comparator into triggering a new cycle immediately after the minimum off-time period has expired. Double pulsing is more annoying than harmful, resulting in nothing worse than increased output ripple. However, it can indicate the possible presence of loop instability due to insufficient ESR. Loop instability can result in oscillations at the output after line or load steps. Such perturbations are usually damped, but can cause the output voltage to rise above or fall below the tolerance limits. The easiest method for checking stability is to apply a very fast zero-to-max load transient and carefully observe the output voltage ripple envelope for overshoot and ringing. It can help to simultaneously monitor the inductor current with an AC current probe. Do not allow more than one cycle of ringing after the initial step-response undervoltage/overshoot. Input Capacitor Selection The input capacitor must meet the ripple current requirement (IRMS) imposed by the switching currents. The IRMS requirements can be determined by the following equation: I IRMS = LOAD VOUT x ( VIN - VOUT ) VIN The worst-case RMS current requirement occurs when operating with VIN = 2VOUT. At this point, the above equation simplifies to: IRMS = 0.5 x ILOAD Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution For most applications, nontantalum chemistries (ceramic, aluminum, or OS-CON) are preferred due to their resistance to inrush surge currents typical of systems with a mechanical switch or connector in series with the input. If the Quick-PWM controller is operated as the second stage of a two-stage power-conversion system, tantalum input capacitors are acceptable. In either configuration, choose an input capacitor that exhibits less than +10C temperature rise at the RMS input current for optimal circuit longevity. MOSFET Selection Most of the following MOSFET guidelines focus on the challenge of obtaining high load-current capability when using high-voltage (> 20V) AC adapters. Lowcurrent applications usually require less attention. The high-side MOSFET (NH) must be able to dissipate the resistive losses plus the switching losses at both V IN(MIN) and V IN(MAX) . Calculate both these sums. Ideally, the losses at VIN(MIN) should be roughly equal to losses at VIN(MAX), with lower losses in between. If the losses at VIN(MIN) are significantly higher than the losses at VIN(MAX), consider increasing the size of NH (reducing RDS(ON) but with higher CGATE). Conversely, if the losses at VIN(MAX) are significantly higher than the losses at VIN(MIN), consider reducing the size of NH (increasing RDS(ON) to lower CGATE). If VIN does not vary over a wide range, the minimum power dissipation occurs where the resistive losses equal the switching losses. Choose a low-side MOSFET that has the lowest possible on-resistance (RDS(ON)), comes in a moderate-sized package (i.e., one or two 8-pin SOs, DPAK, or D2PAK), and is reasonably priced. Make sure that the DL gate driver can supply sufficient current to support the gate charge and the current injected into the parasitic gateto-drain capacitor caused by the high-side MOSFET turning on; otherwise, cross-conduction problems can occur (see the MOSFET Gate Drivers (DH, DL) section). MOSFET Power Dissipation Worst-case conduction losses occur at the duty factor extremes. For the high-side MOSFET (NH), the worstcase power dissipation due to resistance occurs at the minimum input voltage: V 2 PD (NH Resistive) = OUT x (ILOAD ) x RDS(ON) VIN Generally, a small high-side MOSFET is desired to reduce switching losses at high input voltages. However, the RDS(ON) required to stay within package power dissipation often limits how small the MOSFET Maxim Integrated can be. Again, the optimum occurs when the switching losses equal the conduction (RDS(ON)) losses. Highside switching losses do not usually become an issue until the input is greater than approximately 15V. Calculating the power dissipation in high-side MOSFET (NH) due to switching losses is difficult since it must allow for difficult quantifying factors that influence the turn-on and turn-off times. These factors include the internal gate resistance, gate charge, threshold voltage, source inductance, and PCB layout characteristics. The following switching-loss calculation provides only a very rough estimate and is no substitute for breadboard evaluation, preferably including verification using a thermocouple mounted on NH: QG(SW) PD (NH Switching) = VIN(MAX) x ILOAD x fSW IGATE C x VIN2 x fSW + OSS 2 where COSS is the NH MOSFET's output capacitance, QG(SW) is the charge needed to turn on the NH MOSFET, and IGATE is the peak gate-drive source/sink current (2.2A typ). Switching losses in the high-side MOSFET can become an insidious heat problem when maximum AC adapter voltages are applied, due to the squared term in the C x VIN2 x fSW switching-loss equation. If the high-side MOSFET chosen for adequate RDS(ON) at low battery voltages becomes extraordinarily hot when biased from V IN(MAX) , consider choosing another MOSFET with lower parasitic capacitance. For the low-side MOSFET (NL), the worst-case power dissipation always occurs at maximum input voltage: V 2 PD (NL Resistive) = 1- OUT x (ILOAD ) x RDS(ON) VIN(MAX) The worst case for MOSFET power dissipation occurs under heavy overloads that are greater than ILOAD(MAX), but are not quite high enough to exceed the current limit and cause the fault latch to trip. To protect against this possibility, the circuit can be "over designed" to tolerate: I ILOAD = IVALLEY(MAX) + INDUCTOR 2 ILOAD(MAX) x LIR = IVALLEY(MAX) + 2 27 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution where I VALLEY(MAX) is the maximum valley current allowed by the current-limit circuit, including threshold tolerance and on-resistance variation. The MOSFETs must have a good size heatsink to handle the overload power dissipation. Choose a Schottky diode (DL) with a forward voltage low enough to prevent the low-side MOSFET body diode from turning on during the dead time. Select a diode that can handle the load current during the dead times. This diode is optional and can be removed if efficiency is not critical. Setting the PWM Output Voltage Preset Output Voltages The MAX17000A's Dual ModeTM operation allows the selection of common voltages without requiring external components. Connect FB to AGND for a fixed 1.5V output, to V CC for a fixed 1.8V output, or connect FB directly to OUT for a fixed 1.0V output. Adjustable Output Voltage The output voltage can be adjusted from 1.0V to 2.7V using a resistive voltage-divider (Figure 8). The MAX17000A regulates FB to a fixed reference voltage (1.0V). The adjusted output voltage is: VTTI Input Capacitor Stability Considerations The value of the VTTI bypass capacitor is chosen to limit the amount of ripple/noise at VTTI, and the amount of voltage dip during a load transient. Typically, VTTI is connected to the output of the buck regulator, which already has a large bulk capacitor. Nevertheless, a ceramic capacitor of equivalent value to the VTT output capacitor must be used and must be added and placed as close as possible to the VTTI pin. This value must be increased with larger load current, or if the trace from the VTTI pin to the power source is long and has significant impedance. Setting VTT Output Voltage The VTT output stage is powered from the VTTI input. The output voltage is set by the REFIN input. REFIN sets the feedback regulation voltage (VTTR = VTTS = VREFIN) of the MAX17000A. Connect a 0.1V to 2.0V voltage input to set the adjustable output for VTT, VTTS, and VTTR. If REFIN is tied to VCC, the internal CSL/2 divider is used to set VTT voltage; hence, VTT tracks the VCSL voltage and is set to VCSL/2. This feature makes the MAX17000A ideal for memory applications in which the termination supply must track the supply voltage. VTT Output Capacitor Selection R VOUT = VFB x 1 + FBA RFBB A minimum value of 9F is needed to stabilize a 300mA VTT output. This value of capacitance limits the regulator's unity-gain bandwidth frequency to approximately 1.2MHz (typ) to allow adequate phase margin for stability. To keep the capacitor acting as a capacitor within the regulator's bandwidth, it is important that ceramic capacitors with low ESR and ESL be used. where VFB is 1.0V. L1 VOUT LX DL NL COUT D1 PGND1 MAX17000A CSH CSL RFBA FB RFBB Figure 8. Setting VOUT with a Resistive Voltage-Divider Dual Mode is a trademark of Maxim Integrated Products, Inc. 28 Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Since the gain bandwidth is also determined by the transconductance of the output FETs, which increases with load current, the output capacitor might need to be greater than 20F if the load current exceeds 1.5A, but can be smaller than 20F if the maximum load current is less than 1.5A. As a guideline, choose the minimum capacitance and maximum ESR for the output capacitor using the following: COUT _ MIN = 20F x ILOAD 1.5A COUT_MIN needs to be increased by a factor of 2 for low-dropout operation: RESR _ MAX = 5m x 1.5A ILOAD RESR_MAX value is measured at the unity-gain-bandwidth frequency given by approximately: fGBW = I 36 x LOAD COUT 1.5A Once these conditions for stability are met, additional capacitors, including those of electrolytic and tantalum types, can be connected in parallel to the ceramic capacitor (if desired) to further suppress noise or voltage ripple at the output. VTTR Output Capacitor Selection The VTTR buffer is a scaled-down version of the VTT regulator, with much smaller output transconductance. Its compensation capacitor can, therefore, be smaller and its ESR larger than what is required for its larger counterpart. For typical applications requiring load current up to 4mA, a ceramic capacitor with a minimum value of 0.33F is recommended (R ESR < 0.3). Connect this capacitor between VTTR and the analog ground plane. Power Dissipation Power loss in the MAX17000A is the sum of the losses of the PWM block, the VTT LDO block, and the VTTR reference buffer: PD(PWM) = IBIAS x 5V = 40mA x 5V = 0.2 W PD(VTT) = 2 A x 0.9V = 1.8 W PD(VTTR) = 3mA x 0.9V = 2.7mW Maxim Integrated PD(Total) = 2 W The 2W total power dissipation is within the 24-pin TQFN multilayer board power dissipation specification of 2.22W. The typical application does not source or sink continuous high currents. VTT current is typically 100mA to 200mA in the steady state. VTTR is down in the microamp range, though the Intel specification requires 3mA for DDR1 and 1mA for DDR2. True worstcase power dissipation occurs on an output short-circuit condition with worst-case current limit. The MAX17000A does not employ any foldback current limiting, and relies on the internal thermal shutdown for protection. Both the VTT and VTTR output stages are powered from the same VTTI input. Their output voltages are referenced to the same REFIN input. The value of the VTTI bypass capacitor is chosen to limit the amount of ripple/noise at VTTI, or the amount of voltage dip during a load transient. Typically, VTTI is connected to the output of the buck regulator, which already has a large bulk capacitor. Boost Capacitors The boost capacitors (CBST) must be selected large enough to handle the gate-charging requirements of the high-side MOSFETs. Typically, 0.1F ceramic capacitors work well for low-power applications driving medium-sized MOSFETs. However, high-current applications driving large, high-side MOSFETs require boost capacitors larger than 0.1F. For these applications, select the boost capacitors to avoid discharging the capacitor more than 200mV while charging the highside MOSFETs' gates: CBST = QGATE 200mV where QGATE is the total gate charge specified in the high-side MOSFET's data sheet. For example, assume the FDS6612A n-channel MOSFET is used on the high side. According to the manufacturer's data sheet, a single FDS6612A has a maximum gate charge of 13nC (VGS = 5V). Using the above equation, the required boost capacitance would be: CBST = 13nC = 0.065F 200mV Selecting the closest standard value, this example requires a 0.1F ceramic capacitor. 29 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Applications Information PCB Layout Guidelines Careful PCB layout is critical to achieve low switching losses and clean, stable operation. The switching power stage requires particular attention. If possible, mount all the power components on the topside of the board, with their ground terminals flush against one another. Follow these guidelines for good PCB layout: * Keep the high-current paths short, especially at the ground terminals. This practice is essential for stable, jitter-free operation. * Keep the power traces and load connections short. This practice is essential for high efficiency. Using thick copper PCBs (2oz vs. 1oz) can enhance fullload efficiency by 1% or more. Correctly routing PCB traces is a difficult task that must be approached in terms of fractions of centimeters, where a single milliohm of excess trace resistance causes a measurable efficiency penalty. * Minimize current-sensing errors by connecting CSH and CSL directly across the current-sense resistor (RSENSE). * When trade-offs in trace lengths must be made, it is preferable to allow the inductor-charging path to be made longer than the discharge path. For example, it is better to allow some extra distance between the input capacitors and the high-side MOSFET than to allow distance between the inductor and the lowside MOSFET or between the inductor and the output filter capacitor. * Layout Procedure 1) Place the power components first, with ground terminals adjacent (low-side MOSFET source, C IN, COUT, and anode of the low-side Schottky). If possible, make all these connections on the top layer with wide, copper-filled areas. 2) Mount the controller IC adjacent to the low-side MOSFET, preferably on the backside opposite the MOSFETs to keep LX, GND, DH, and the DL gatedrive lines short and wide. The DL and DH gate traces must be short and wide (50 mils to 100 mils wide if the MOSFET is 1in from the controller IC) to keep the driver impedance low and for proper adaptive dead-time sensing. 3) Group the gate-drive components (BST diode and capacitor, VDD bypass capacitor) together near the controller IC. 4) Make the DC-DC controller ground connections as shown in Figures 1 and 9. This diagram can be viewed as having two separate ground planes: power ground, where all the high-power components go; and an analog ground plane for sensitive analog components. The analog ground plane and power ground plane must meet only at a single point directly at the IC. 5) Connect the output power planes directly to the output filter capacitor positive and negative terminals with multiple vias. Place the entire DC-to-DC converter circuit as close as is practical to the load. Table 5 lists the design differences between the MAX17000 and MAX17000A. Route high-speed switching nodes (BST, LX, DH, and DL) away from sensitive analog areas (REFIN, FB, CSH, and CSL). Table 5. MAX17000 vs. MAX17000A Design Differences MAX17000 MAX17000A STDBY = Low turns off VTT and overrides the SKIP setting, forcing the SMPS to enter a low-quiescent current ultra-skip mode. STDBY = Low only turns off VTT rail, and does not affect SMPS operation. 30 Maxim Integrated MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution KELVIN SENSE VIAS UNDER THE INDUCTOR (SEE EVALUATION KIT) POWER STAGE LAYOUT (TOP SIDE OF PCB) OUTPUT CEQ COUT CSL CSH COUT INDUCTOR L1 RNTC R2 R1 POWER GROUND CIN1 KELVIN-SENSE VIAS TO INDUCTOR PAD INPUT INDUCTOR DCR SENSING SMPS CONNECT AGND AND PGND1 TO THE CONTROLLER AT THE EXPOSED PAD CONNECT THE EXPOSED PAD TO ANALOG GROUND VDD BYPASS CAPACITOR VTTI BYPASS CAPACITOR VIA TO POWER GROUND VCC BYPASS CAPACITOR VTT BYPASS CAPACITOR X-RAY VIEW. IC MOUNTED ON BOTTOM SIDE OF PCB. IC LAYOUT Figure 9. PCB Layout Example Package Information Chip Information TRANSISTOR COUNT: 7856 PROCESS: BiCMOS Maxim Integrated For the latest package outline information and land patterns (footprints), go to www.maximintegrated.com/packages. Note that a "+", "#", or "-" in the package code indicates RoHS status only. Package drawings may show a different suffix character, but the drawing pertains to the package regardless of RoHS status. PACKAGE TYPE PACKAGE CODE OUTLINE NO. LAND PATTERN NO. 24 TQFN T2444-4 21-0139 90-0022 31 MAX17000A Complete DDR2 and DDR3 Memory Power-Management Solution Revision History REVISION NUMBER REVISION DATE 0 10/08 Initial release 1 12/08 Modified STDBY pin function -- 5, 11, 12, 13, 17, 23, 2 11/10 Changed RESR_MAX equation 29 3 4/13 Updated Absolute Maximum Ratings 2 DESCRIPTION PAGES CHANGED Maxim Integrated cannot assume responsibility for use of any circuitry other than circuitry entirely embodied in a Maxim Integrated product. No circuit patent licenses are implied. Maxim Integrated reserves the right to change the circuitry and specifications without notice at any time. The parametric values (min and max limits) shown in the Electrical Characteristics table are guaranteed. Other parametric values quoted in this data sheet are provided for guidance. 32 ________________________________Maxim Integrated 160 Rio Robles, San Jose, CA 95134 USA 1-408-601-1000 (c) 2013 Maxim Integrated Products, Inc. Maxim Integrated and the Maxim Integrated logo are trademarks of Maxim Integrated Products, Inc. Mouser Electronics Authorized Distributor Click to View Pricing, Inventory, Delivery & Lifecycle Information: Maxim Integrated: MAX17000AETG+C00 MAX17000AETG+TC00 MAX17000AETG+ MAX17000AETG+T MAX17000AETG+G40 MAX17000AETG+TG40