ISL8112 ® Data Sheet August 10, 2010 High Light-Load Efficiency, Dual-Output, Main Power Supply Controllers ISL8112 is a dual-output Synchronous Buck controller with 2A integrated driver. It features high light load efficiency which is especially preferred in systems concerned with high efficiency in wide load range, like the battery powered system. ISL8112 includes two constant on-time PWM controllers. Either of the two outputs can operate in output fixed mode or adjustable mode. In fixed mode, one output can be 5V or 3.3V and the other can output 1.5V or 1.05V. In output adjustable mode, one output can be 0.7V to 5.5V, and the other output can range from 0V to 2.5V (sensing output voltage directly) or up to 5V (using resistor divider voltage for voltage sensing). This device also features a linear regulator providing 3.3V/5V, or adjustable from 0.7V to 4.5V via LDOREF. The linear regulator provides up to 100mA output current with automatic linear-regulator bootstrapping to the BYP input. When in switch over, the LDO output can source up to 200mA. ISL8112 includes on-board power-up sequencing, the powergood (PGOOD_) outputs, digital soft-start, and internal softstop output discharge that prevents negative voltages on shutdown. ISL8112 is implemented with constant on-time PWM control scheme which need no sense resistors and provides 100ns response to load transients while maintaining a relatively constant switching frequency. The unique ultrasonic pulseskipping mode maintains the switching frequency above 25kHz, eliminating undesired audible noises in low frequency operation at light load. Other features include pulse skipping which maximizes efficiency in light-load applications, and fixed-frequency PWM mode which reduces RF interference in sensitive applications. Ordering Information PART NUMBER (Note) ISL8112IRZ* PART MARKING TEMP. RANGE (°C) FN6396.1 Features • Wide Input Voltage Range 5.5V to 25V • Constant ON-TIME Control with 100ns Load-Step Response • Dual Fixed Outputs of 1.05V (3.3V) and 1.5V (5.0V), or Adjustable Outputs of 0.7V to 5.5V (SMPS1) and 0V to 2.5V/5V (SMPS2), ±1.5% Accuracy • Adjustable Switching Frequency: 400/500kHz, 300/400kHz, 200/300kHz • Very High Light Load Efficiency (Skip Mode) • 5mW Quiescent Power Dissipation • ±1.5% (LDO): 100mA, 200mA (Switch Over) • 3.3V Reference Voltage ±2.0%: 5mA • 2.0V Reference Voltage ±1.0%: 50µA • Temperature Compensated rDS(ON) Current Sensing • Programmable Current Limit with Foldback Capability • Selectable PWM, Skip or Ultrasonic Mode • Independent PGOOD1 and PGOOD2 Comparators • Soft-Start with Pre-Biased Output and Soft-Stop • 1.7ms Digital Soft-Start and Independent Shutdown • Independent ENABLE • Thermal Shutdown • Extremely Low Components Count • Pb-Free Available (RoHS Compliant) Applications • Power Supply for Telecom/Datacom and POL PACKAGE ISL8112 IRZ -40 to +100 32 Ld QFN (Pb-free) PKG. DWG. # L32.5x5B *Add “-T” suffix for tape and reel. Please refer to TB347 for details on reel specifications. • System Requiring High Efficiency in Wide Load Range • Compact Design with Minimum Components Count • PDAs and Mobile Communication Devices • 3- and 4-Cell Li+ Battery-Powered Devices • DDR1, DDR2, and DDR3 Applications NOTE: These Intersil Pb-free plastic packaged products employ special Pb-free material sets, molding compounds/die attach materials, and 100% matte tin plate plus anneal (e3 termination finish, which is RoHS compliant and compatible with both SnPb and Pb-free soldering operations). Intersil Pb-free products are MSL classified at Pb-free peak reflow temperatures that meet or exceed the Pb-free requirements of IPC/JEDEC J STD-020. 1 CAUTION: These devices are sensitive to electrostatic discharge; follow proper IC Handling Procedures. 1-888-INTERSIL or 1-888-468-3774 | Intersil (and design) is a registered trademark of Intersil Americas Inc. Copyright Intersil Americas Inc. 2006, 2010. All Rights Reserved All other trademarks mentioned are the property of their respective owners. ISL8112 Pinout 2 OUT2REF ILIM2 VSEN2 MODE PGOOD2 EN2 UG2 PH2 ISL8112 (32 LD 5X5 QFN) TOP VIEW 32 31 30 29 28 27 26 25 VCC 3 22 PGND EN_LDO 4 21 GND VREF2 5 20 NC VIN 6 19 PVCC LDO 7 18 LG1 LDOREF 8 17 BOOT1 9 10 11 12 13 14 15 16 PH1 LG2 UG1 23 EN1 2 PGOOD1 FS ILIM1 BOOT2 FB1 24 VSEN1 1 BYP VREF1 FN6396.1 August 10, 2010 ISL8112 Absolute Voltage Ratings Thermal Information VIN, EN_LDO to GND . . . . . . . . . . . . . . . . . . . . . . . . . -0.3V to +27V BOOT_ to GND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.3V to +33V BOOT_ to PH_ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.3V to +6V VCC, EN_, MODE, FS, PVCC, PGOOD_ to GND . . . . . . . . . . . . . . . . . . . . . . . -0.3V to +6V LDO, FB1, OUT2REF, LDOREF to GND . . . . -0.3V to (VCC+0.3V) VSEN_, VREF2, VREF1 to GND . . . . . . . . . . . -0.3V to (VCC+0.3V UG_ to PH_ . . . . . . . . . . . . . . . . . . . . . . . . . -0.3V to (PVCC + 0.3V) ILIM_ to GND . . . . . . . . . . . . . . . . . . . . . . . . . -0.3V to (VCC + 0.3V) LG_, BYP to GND . . . . . . . . . . . . . . . . . . . . -0.3V to (PVCC + 0.3V) PGND to GND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.3V to + 0.3V LDO, VREF1, VREF2 Short Circuit to GND . . . . . . . . . . Continuous VCC Short Circuit to GND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1s LDO Current (Internal Regulator) Continuous . . . . . . . . . . . . 100mA LDO Current (Switched Over to VSEN1) Continuous . . . . . +200mA Thermal Resistance (Typical, Note 1) θJA (°C/W) θJC (°C/W) 32 Ld QFN (Notes 1, 2) . . . . . . . . . . . . 32 3.0 Operating Temperature Range . . . . . . . . . . . . . . . .-40°C to +100°C Junction Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +150°C Storage Temperature Range . . . . . . . . . . . . . . . . . .-65°C to +150°C Pb-Free Reflow Profile. . . . . . . . . . . . . . . . . . . . . . . . .see link below http://www.intersil.com/pbfree/Pb-FreeReflow.asp CAUTION: Do not operate at or near the maximum ratings listed for extended periods of time. Exposure to such conditions may adversely impact product reliability and result in failures not covered by warranty. NOTE: 1. θJA is measured in free air with the component mounted on a high effective thermal conductivity test board with “direct attach” features. See Tech Brief TB379. 2. For θJC, the “case temp” location is the center of the exposed metal pad on the package underside. Electrical Specifications Circuit of Figure 17, and Figure 18, no load on LDO, VSEN1, VSEN2, VREF2, and VREF1, VIN = 12V, EN2 = EN1 = VCC, VBYP = 5V, PVCC = 5V, VEN_LDO = 5V, TA = -40°C to +100°C, unless otherwise noted. Typical values are at TA = +25°C. PARAMETER CONDITIONS MIN (Note 3) TYP MAX (Note 3) UNITS 25 V MAIN SMPS CONTROLLERS VIN Input Voltage Range LDO in regulation 5.5 V 3.3V Output Voltage in Fixed Mode VIN = 5.5V to 25V, OUT2REF > (VCC - 1V), MODE = 5V 3.285 3.330 3.375 V 1.05V Output Voltage in Fixed Mode VIN = 5.5V to 25V, 3.0 < OUT2REF < (VCC - 1.1V), MODE = 5V 1.038 1.05 1.062 V 1.5V Output Voltage in Fixed Mode VIN= 5.5V to 25V, FB1 = VCC, MODE = 5V 1.482 1.500 1.518 V 5V Output Voltage in Fixed Mode VIN= 5.5V to 25V, FB1 = GND, MODE = 5V 4.975 5.050 5.125 V FB1 in Output Adjustable Mode VIN = 5.5V to 25V 0.693 0.700 0.707 V OUT2REF in Output Adjustable Mode VIN = 5.5V to 25V 0.7 2.50 V SMPS1 Output Voltage Adjust Range SMPS1 0.70 5.50 V SMPS2 Output Voltage Adjust Range SMPS2 0.50 2.50 V SMPS2 Output Voltage Accuracy (Referred for OUT2REF) OUT2REF = 0.7V to 2.5V, MODE = VCC -1.0 1.0 % DC Load Regulation Either SMPS, MODE = VCC, 0A to 5A -0.1 % Either SMPS, MODE = VREF1, 0A to 5A -1.7 % VIN = LDO, VSEN1 < 4.43V 5.5 4.5 Either SMPS, MODE = GND, 0A to 5A Line Regulation Either SMPS, 6V < VIN < 24V Current-Limit Current Source Temperature = +25°C ILIM_ Adjustment Range 4.75 -1.5 % 0.005 %/V 5 0.2 Current-Limit Threshold (Positive, Default) 3 ILIM_ = VCC, GND - PH_ (No temperature compensation) 93 100 5.25 µA 2 V 107 mV FN6396.1 August 10, 2010 ISL8112 Electrical Specifications Circuit of Figure 17, and Figure 18, no load on LDO, VSEN1, VSEN2, VREF2, and VREF1, VIN = 12V, EN2 = EN1 = VCC, VBYP = 5V, PVCC = 5V, VEN_LDO = 5V, TA = -40°C to +100°C, unless otherwise noted. Typical values are at TA = +25°C. (Continued) PARAMETER MIN (Note 3) TYP MAX (Note 3) UNITS 40 50 60 mV VILIM_ = 1V 93 100 107 mV VILIM_ = 2V 185 200 215 mV CONDITIONS Current-Limit Threshold (Positive, Adjustable) GND - PH_ VILIM_ = 0.5V Zero-Current Threshold MODE = GND, VREF1, or OPEN, GND - PH_ Current-Limit Threshold (Negative, Default) MODE = VCC, GND - PH_ Soft-Start Ramp Time Zero to full limit Operating Frequency (VFS = GND), MODE = VCC On-Time Pulse Width -120 mV 1.7 ms 400 kHz SMPS 2 500 kHz (VFS = VREF1 or OPEN), MODE = VCC SMPS 1 400 kHz SMPS 2 300 kHz (VFS = VCC), MODE = VCC SMPS 1 200 kHz SMPS 2 300 kHz VSEN1 = 5.00V 0.895 1.052 1.209 µs VSEN2 = 3.33V 0.475 0.555 0.635 µs VFS = VREF1 or OPEN (400kHz/300kHz) VSEN1 = 5.05V 0.895 1.052 1.209 µs VSEN2 = 3.33V 0.833 0.925 1.017 µs VFS = VCC (200kHz/300kHz) VSEN1 = 5.05V 1.895 2.105 2.315 µs VSEN2 = 3.33V 0.833 0.925 1.017 µs 200 300 400 ns Minimum Off-Time VFS = GND VFS = VREF1 or OPEN VFS = VCC Ultrasonic SKIP Operating Frequency mV SMPS 1 VFS = GND (400kHz/500kHz) Maximum Duty Cycle 3 VSEN1 = 5.05V 88 % VSEN2 = 3.33V 85 % VSEN1 = 5.05V 88 % VSEN2 = 3.33V 91 % VSEN1 = 5.05V 94 % VSEN2 = 3.33V 91 % 25 37 kHz MODE = VREF1 or OPEN INTERNAL REGULATOR AND REFERENCE LDO Output Voltage BYP = GND, 5.5V < VIN < 25V, LDOREF < 0.3V, 0 < ILDO < 100mA 4.925 5.000 5.075 V LDO Output Voltage BYP = GND, 5.5V < VIN < 25V, LDOREF > (VCC-1V), 0 < ILDO < 100mA 3.250 3.300 3.350 V LDO Output in Adjustable Mode VIN = 5.5V to 25V, VLDO = 2 x VLDOREF 4.5 V VIN = 5.5V to 25V, VLDOREF = 0.35V to 0.5V ±2 % VIN = 5.5V to 25V, VLDOREF = 0.5V to 2.25V ±1.5 % 2.25 V 100 mA LDO Output Accuracy in Adjustable Mode LDOREF Input Range VLDO = 2 x VLDOREF LDO Output Current BYP = GND, VIN = 5.5V to 25V (Note 4) 0.7 0.35 LDO Output Current During Switch Over BYP = 5V, VIN = 5.5V to 25V, LDOREF < 0.3V 200 mA LDO Output Current During Switch Over to 3.3V BYP = 3.3V, VIN = 5.5V to 25V, LDOREF > (VCC-1V) 100 mA LDO Short-Circuit Current LDO = GND, BYP = GND 200 400 mA Undervoltage-Lockout Fault Threshold Rising edge of PVCC Falling edge of PVCC 4.35 4.05 4.5 V 4 3.9 FN6396.1 August 10, 2010 ISL8112 Electrical Specifications Circuit of Figure 17, and Figure 18, no load on LDO, VSEN1, VSEN2, VREF2, and VREF1, VIN = 12V, EN2 = EN1 = VCC, VBYP = 5V, PVCC = 5V, VEN_LDO = 5V, TA = -40°C to +100°C, unless otherwise noted. Typical values are at TA = +25°C. (Continued) MIN (Note 3) TYP MAX (Note 3) UNITS Rising edge at BYP regulation point LDOREF = GND 4.53 4.68 4.83 V LDO 3.3V Bootstrap Switch Threshold to BYP Rising edge at BYP regulation point LDOREF = VCC 3.0 3.1 3.2 V PARAMETER CONDITIONS LDO 5V Bootstrap Switch Threshold to BYP LDO 5V Bootstrap Switch Equivalent Resistance LDO to BYP, BYP = 5V, LDOREF > (VCC-1V) (Note 4) 0.7 1.5 Ω LDO 3.3V Bootstrap Switch Equivalent Resistance LDO to BYP, BYP = 3.3V, LDOREF < 0.3V (Note 4) 1.5 3.0 Ω VREF2 Output Voltage No external load, VCC > 4.5V 3.235 3.300 3.365 V No external load, VCC < 4.0V 3.220 3.300 3.380 V VREF2 Load Regulation 0 < ILOAD < 5mA 10 VREF2 Current Limit VREF2 = GND 10 17 mA VREF1 Output Voltage No external load 2.000 2.020 V 1.980 mV VREF1 Load Regulation 0 < ILOAD < 50µA VREF1 Sink Current VREF1 in regulation 10 mV VIN Operating Supply Current Both SMPSs on, FB1 = MODE = GND, OUT2REF = VCC VSEN1 = BYP = 5.3V, VSEN2 = 3.5V 25 50 µA VIN Standby Supply Current VIN = 5.5V to 25V, both SMPSs off, EN_LDO = VCC 180 250 µA VIN Shutdown Supply Current VIN = 4.5V to 25V, EN1=EN2=EN_LDO=0V 20 30 µA Quiescent Power Consumption Both SMPSs on, FB1 = MODE = GND, OUT2REF = VCC, VSEN1 = BYP = 5.3V, VSEN2 = 3.5V 5 7 mW % 10 µA FAULT DETECTION Overvoltage Trip Threshold FB1 with respect to nominal regulation point +8 +11 +14 OUT2REF with respect to nominal regulation point +12 +16 +20 Overvoltage Fault Propagation Delay FB1 or OUT2REF delay with 50mV overdrive PGOOD_ Threshold FB1 or OUT2REF with respect to nominal output, falling edge, typical hysteresis = 1% PGOOD_ Propagation Delay Falling edge, 50mV overdrive PGOOD_ Output Low Voltage ISINK = 4mA PGOOD_ Leakage Current High state, forced to 5.5V % 10 -12 -9 µs -6 % 10 Thermal-Shutdown Threshold µs 0.2 V 1 µA +150 °C Output Undervoltage Shutdown Threshold FB1 or OUT2REF with respect to nominal output voltage 65 70 75 % Output Undervoltage Shutdown Blanking Time From EN_ signal 10 20 30 ms 0.3 V INPUTS AND OUTPUTS FB1 Input Voltage Low level High level OUT2REF Input Voltage 5 VCC-1.0 V VSEN2 Dynamic Range, VSEN2= VOUT2REF 0.5 2.50 V Fixed VSEN2 = 1.05V 3.0 VCC1.1 V Fixed VSEN2 = 3.3V VCC-1.0 V FN6396.1 August 10, 2010 ISL8112 Electrical Specifications Circuit of Figure 17, and Figure 18, no load on LDO, VSEN1, VSEN2, VREF2, and VREF1, VIN = 12V, EN2 = EN1 = VCC, VBYP = 5V, PVCC = 5V, VEN_LDO = 5V, TA = -40°C to +100°C, unless otherwise noted. Typical values are at TA = +25°C. (Continued) PARAMETER CONDITIONS LDOREF Input Voltage MIN (Note 3) TYP Fixed LDO = 5V VSEN2 Dynamic Range, VLDO = 2 x VLDOREF Fixed LDO = 3.3V MODE Input Voltage 0.35 Input Leakage Current V Float level (ULTRASONIC SKIP) 1.7 High level (PWM) 2.4 Float level 1.7 High level 2.4 V V 0.8 V 2.3 V V 0.8 V 2.3 V V Clear fault level/SMPS off level EN_LDO Input Voltage 0.30 2.25 Low level EN1, EN2 Input Voltage UNITS VCC-1.0 Low level (SKIP) FS Input Voltage MAX (Note 3) 0.8 V 2.3 V Delay start level 1.7 SMPS on level 2.4 Rising edge 1.2 1.6 2.0 Falling edge 0.94 1.00 V V 1.06 V -1 +1 µA -0.1 +0.1 µA -1 +1 µA VFB1 = 0V or 5V -0.2 +0.2 µA VREFIN = 0V or 2.5V -0.2 +0.2 µA VLDOREF = 0V or 2.75V -0.2 +0.2 µA VFS = 0 or 5V VEN_ = VEN_LDO = 0V or 5V VMODE = 0V or 5V INTERNAL BOOT DIODE VD Forward Voltage PVCC - VBOOT, IF = 10mA IBOOT_LEAKAGE Leakage Current VBOOT = 30V, PH = 25V, PVCC = 5V 0.65 0.8 V 500 nA MOSFET DRIVERS UG_ Gate-Driver Sink/Source Current UG1, UG2 forced to 2V 2 A LG_ Gate-Driver Source Current LG1 (source), LG2 (source), forced to 2V 1.7 A LG_ Gate-Driver Sink Current LG1 (sink), LG2 (sink), forced to 2V 3.3 A UG_ Gate-Driver On-Resistance BST_ - PH_ forced to 5V (Note 4) 1.5 4.0 Ω LG_ Gate-Driver On-Resistance LG_, high state (pull-up) (Note 4) 2.2 5.0 Ω 0.6 1.5 Ω LG_ Rising 15 20 35 ns UG_ Rising 20 30 50 ns 25 40 Ω LG_, low state (pull-down) (Note 4) Dead Time VSEN1, VSEN2 Discharge On Resistance NOTES: 3. Parameters with MIN and/or MAX limits are 100% tested at +25°C, unless otherwise specified. Temperature limits established by characterization and are not production tested. 4. Limits established by characterization and are not production tested. 6 FN6396.1 August 10, 2010 ISL8112 Pin Descriptions PIN NAME FUNCTION 1 VREF1 2V Reference Output. Bypass to GND with a 0.1µF (min) capacitor. VREF1 can source up to 50μA for external loads. Loading VREF1 degrades FB and output accuracy according to the VREF1 load-regulation error. 2 FS 3 VCC 4 EN_LDO 5 VREF2 6 VIN Power-Supply Input. VIN is used for the constant-on-time PWM on-time one-shot circuits. VIN is also used to power the linear regulators. The linear regulators are powered by SMPS1 if VSEN1 is set greater than 4.78V and BYP is tied to VSEN1. Connect VIN to the battery input and bypass with a 1µF capacitor. 7 LDO Linear-Regulator Output. LDO can provide a total of 100mA external loads. The LDO regulate at 5V If LDOREF is connected to GND. When the LDO is set at 5V and BYP is within 5V switch over threshold, the internal regulator shuts down and the LDO output pin connects to BYP through a 0.7Ω switch. The LDO regulate at 3.3V if LDOREF is connected to VCC. When the LDO is set at 3.3V and BYP is within 3.3V switch over threshold, the internal regulator shuts down and the LDO output pin connects to BYP through a 1.5Ω switch. Bypass LDO output with a minimum of 4.7µF ceramic. 8 LDOREF LDO Reference Input. Connect LDOREF to GND for fixed 5V operation. Connect LDOREF to VCC for fixed 3.3V operation. LDOREF can be used to program LDO output voltage from 0.7V to 4.5V. LDO output is two times the voltage of LDOREF. There is no switch over in adjustable mode. 9 BYP BYP is the switch over source voltage for the LDO when LDOREF connected to GND or VCC. Connect BYP to 5V if LDOREF is tied to GND. Connect BYP to 3.3V if LDOREF is tied to VCC. The BYP is also controlled by EN_LDO. When LDOREFIN is tied to GND, the BYP is not switched over to LDO until SMPS1 finished soft-starting. 10 VSEN1 SMPS1 Output Voltage-Sense Input. Connect to the SMPS1 output. VSEN1 is an input to the Constant on-time-PWM on-time one-shot circuit. It also serves as the SMPS1 feedback input in fixed-voltage mode. 11 FB1 12 ILIM1 SMPS1 Current-Limit Adjustment. The GND-PH1 current-limit threshold is 1/10th the voltage seen at ILIM1 over a 0.2V to 2V range. There is an internal 5µA current source from VCC to ILIM1. Connect ILIM1 to VREF1 for a fixed 200mV threshold. The logic current limit threshold is default to 100mV value if ILIM1 is higher than VCC - 1V. 13 PGOOD1 SMPS1 Power-Good Open-Drain Output. PGOOD1 is low when the SMPS1 output voltage is more than 10% below the normal regulation point or during soft-start. PGOOD1 is high impedance when the output is in regulation and the softstart circuit has terminated. PGOOD1 is low in shutdown. 14 EN1 SMPS1 Enable Input. The SMPS1 is enabled if EN1 is greater than the logic high level and disabled if EN1 is less than the logic low level. If EN1 is connected to VREF1, the SMPS1 starts after the SMPS2 reaches regulation (delay start). Drive EN1 below 0.8V to clear fault level and reset the fault latches. 15 UG1 High-Side MOSFET Floating Gate-Driver Output for SMPS1. UG1 swings between PH1 and BOOT1. 16 PH1 Inductor Connection for SMPS1. PH1 is the internal lower supply rail for the UG1 high-side gate driver. PH1 is the current-sense input for the SMPS1. 17 BOOT1 18 LG1 19 PVCC 20 NC 21 GND 22 PGND Frequency Select Input. Connect to GND for 400kHz/500kHz operation. Connect to VREF1 (or leave OPEN) for 400kHz/300kHz operation. Connect to VCC for 200kHz/300kHz operation (5V/3.3V SMPS switching frequencies, respectively). Analog Supply Voltage for PWM Core. Bypass to GND with a 1µF ceramic capacitor. LDO Enable Input. The LDO is enabled if EN_LDO is within logic high level and VIN is higher than POR threshold. The LDO is disabled if EN_LDO is less than the logic low level. 3.3V Reference Output. VREF2 can source up to 5mA for external loads. Bypass to GND with a 0.01µF capacitor if loaded. Leave open if there is no load. SMPS1 Feedback Input. Connect FB1 to GND for fixed 5V operation. Connect FB1 to VCC for fixed 1.5V operation Connect FB1 to a resistive voltage-divider from VSEN1 to GND to adjust the output from 0.7V to 5.5V. Boost Flying Capacitor Connection for SMPS1. Connect to an external capacitor according to the typical application circuits (Figure 17 and Figure 18). See “MOSFET Gate Drivers (UG_, LG_)” on page 19. SMPS1 Synchronous-Rectifier Gate-Drive Output. LG1 swings between GND and PVCC. PVCC is the supply voltage for the low-side MOSFET driver LG_. Connect a 5V power source to the PVCC pin (bypass with 1µF MLCC capacitor to PGND if necessary). There is internal 10Ω PFET connecting PVCC to VCC. Make sure that both VCC and PVCC are bypassed with 1µF MLCC capacitors. No connection pin. Externally connect it to ground. Analog Ground for both SMPS_ and LDO. Connect externally to the underside of the exposed pad. Power Ground for SMPS_ controller. Connect PGND externally to the underside of the exposed pad. 7 FN6396.1 August 10, 2010 ISL8112 Pin Descriptions (Continued) PIN NAME 23 LG2 24 BOOT2 25 PH2 Inductor Connection for SMPS2. PH2 is the internal lower supply rail for the UG2 high-side gate driver. PH2 is the current-sense input for the SMPS2. 26 UG2 High-Side MOSFET Floating Gate-Driver Output for SMPS2. UG1 swings between PH2 and BOOT2. 27 EN2 SMPS2 Enable Input. The SMPS2 is enabled if EN2 is greater than the logic high level and disabled if EN2 is less than the logic low level. If EN2 is connected to VREF1, the SMPS2 starts after the SMPS1 reaches regulation (delay start). Drive EN2 below 0.8V to clear fault level and reset the fault latches. 28 PGOOD2 SMP2 Power-Good Open-Drain Output. PGOOD2 is low when the SMPS2 output voltage is more than 10% below the normal regulation point or during soft-start. PGOOD2 is high impedance when the output is in regulation and the softstart circuit has terminated. PGOOD2 is low in shutdown. 29 MODE Low-Noise Mode Control. Connect MODE to GND for normal Idle-Mode (pulse-skipping) operation or to VCC for PWM mode (fixed frequency). Connect to VREF1 or leave floating for ultrasonic skip mode operation. 30 VSEN2 SMPS2 Output Voltage-Sense Input. Connect to the SMPS2 output. VSEN2 is an input to the Constant on-time-PWM on-time one-shot circuit. It also serves as the SMPS2 feedback input in fixed-voltage mode. 31 ILIM2 SMPS2 Current-Limit Adjustment. The GND-PH1 current-limit threshold is 1/10th the voltage seen at ILIM2 over a 0.2V to 2V range. There is an internal 5µA current source from VCC to ILIM2. Connect ILIM2 to VREF1 for a fixed 200mV. The logic current limit threshold is default to 100mV value if ILIM2 is higher than VCC - 1V. 32 FUNCTION SMPS2 Synchronous-Rectifier Gate-Drive Output. LG2 swings between GND and PVCC. Boost Flying Capacitor Connection for SMPS2. Connect to an external capacitor according to the typical application circuits (Figure 17 and Figure 18). See “MOSFET Gate Drivers (UG_, LG_)” on page 19. OUT2REF Output voltage control for SMPS2. Connect OUT2REF to VCC for fixed 3.3V. Connect OUT2REF to VREF2 for fixed 1.05V. OUT2REF can be used to program SMPS2 output. VSEN2 equals OUT2REF from 0.5V to 2.50V. SMPS2 output voltage is 0V if OUT2REF < 0.5V. Typical Performance Curves 12 VIN ULTRA SKIP MODE 25 VIN SKIP MODE 25 VIN PWM MODE 25 VIN ULTRA SKIP MODE 1.0 1.0 0.9 0.9 0.8 0.8 0.7 0.7 0.6 0.6 EFFICIENCY EFFICIENCY 7 VIN SKIP MODE 7 VIN PWM MODE 7 VIN ULTRA SKIP MODE 12 VIN SKIP MODE 12 VIN PWM MODE Circuit of Figure 17 and Figure 18, no load on LDO, VSEN1, VSEN2, VREF2, and VREF1, VIN = 12V, EN2 = EN1 = VCC, VBYP = 5V, PVCC = 5V, VEN_LDO = 5V, TA = -40°C to +100°C, unless otherwise noted. Typical values are at TA = +25°C. 0.5 0.4 0.3 0.1 1.000 10.000 FIGURE 1. VOUT2 = 1.05V EFFICIENCY vs LOAD (300kHz) 8 1.000 0.3 0.2 0.100 OUTPUT LOAD (A) 0.010 0.4 0.1 0.010 12 VIN ULTRA SKIP MODE 25 VIN SKIP MODE 25 VIN PWM MODE 25 VIN ULTRA SKIP MODE 0.5 0.2 0 0.001 7 VIN SKIP MODE 7 VIN PWM MODE 7 VIN ULTRA SKIP MODE 12 VIN SKIP MODE 12 VIN PWM MODE 0 0.001 0.100 OUTPUT LOAD (A) 10.000 FIGURE 2. VOUT1 = 1.5V EFFICIENCY vs LOAD (200kHz) FN6396.1 August 10, 2010 ISL8112 Typical Performance Curves 0.9 0.8 0.8 0.7 0.7 0.6 0.5 0.4 0.3 0.5 0.4 0.3 0.2 0.1 0.1 0.010 0.100 OUTPUT LOAD (A) 1.000 0 0.001 10.000 0.010 1.000 10.000 FIGURE 4. VOUT1 = 5V EFFICIENCY vs LOAD (400kHz) 300 50 45 250 40 35 RIPPLE (mV) 200 PWM 150 100 ULTRA-SKIP 50 30 PWM 25 ULTRA-SKIP 20 15 10 0.010 0.100 SKIP 5 SKIP 0 0.001 1.000 0 0.001 10.000 0.010 OUTPUT LOAD (A) 0.100 1.000 10.000 OUTPUT LOAD (A) FIGURE 5. VOUT2 = 1.05V FREQUENCY vs LOAD FIGURE 6. VOUT2 = 1.05V RIPPLE vs LOAD 250 50 45 PWM 200 40 PWM 35 RIPPLE (mV) FREQUENCY (kHz) 0.100 OUTPUT LOAD (A) FIGURE 3. VOUT2 = 3.3V EFFICIENCY vs LOAD (500kHz) FREQUENCY (kHz) 0.6 0.2 0 0.001 12 VIN ULTRA SKIP MODE 25 VIN SKIP MODE 25 VIN PWM MODE 25 VIN ULTRA SKIP MODE 7 VIN SKIP MODE 7 VIN PWM MODE 7 VIN ULTRA SKIP MODE 12 VIN SKIP MODE 12 VIN PWM MODE 1.0 0.9 EFFICIENCY EFFICIENCY 12 VIN ULTRA SKIP MODE 25 VIN SKIP MODE 25 VIN PWM MODE 25 VIN ULTRA SKIP MODE 7 VIN SKIP MODE 7 VIN PWM MODE 7 VIN ULTRA SKIP MODE 12 VIN SKIP MODE 12 VIN PWM MODE 1.0 Circuit of Figure 17 and Figure 18, no load on LDO, VSEN1, VSEN2, VREF2, and VREF1, VIN = 12V, EN2 = EN1 = VCC, VBYP = 5V, PVCC = 5V, VEN_LDO = 5V, TA = -40°C to +100°C, unless otherwise noted. Typical values are at TA = +25°C. (Continued) 150 100 ULTRA-SKIP 50 30 25 SKIP 20 ULTRA-SKIP 15 10 SKIP 0 0.001 5 0.010 0.100 1.000 OUTPUT LOAD (A) FIGURE 7. VOUT1 = 1.5V FREQUENCY vs LOAD 9 10.000 0 0.001 0.010 0.100 OUTPUT LOAD (A) 1.000 10.000 FIGURE 8. VOUT1 = 1.5V RIPPLE vs LOAD FN6396.1 August 10, 2010 ISL8112 Typical Performance Curves Circuit of Figure 17 and Figure 18, no load on LDO, VSEN1, VSEN2, VREF2, and VREF1, VIN = 12V, EN2 = EN1 = VCC, VBYP = 5V, PVCC = 5V, VEN_LDO = 5V, TA = -40°C to +100°C, unless otherwise noted. Typical values are at TA = +25°C. (Continued) 14 600 PWM PWM 12 10 400 RIPPLE (mV) FREQUENCY (kHz) 500 300 200 ULTRA-SKIP 100 ULTRA-SKIP 4 0.010 0.100 OUTPUT LOAD (A) 1.000 0 0.001 10.000 450 40 400 35 250 200 150 25 10.000 PWM ULTRA-SKIP 20 SKIP 15 5 SKIP 0 0.001 0.010 0.100 OUTPUT LOAD (A) 1.000 0 0.001 10.000 FIGURE 11. VOUT1 = 5V FREQUENCY vs LOAD 0.010 0.100 OUTPUT LOAD (A) 1.000 10.000 FIGURE 12. VOUT1 = 5V RIPPLE vs LOAD 5.04 3.35 5.02 BYP = 0V 5.00 3.30 OUTPUT VOLTAGE (V) OUTPUT VOLTAGE (V) 1.000 10 ULTRA-SKIP 50 4.98 4.96 4.94 4.92 4.90 BYP = 5V 4.88 3.25 BYP = 0V 3.20 3.15 BYP = 3.3V 3.10 3.05 4.86 4.84 0 0.100 OUTPUT LOAD (A) 30 PWM RIPPLE (mV) FREQUENCY (kHz) 350 300 0.010 FIGURE 10. VOUT2 = 3.3V RIPPLE vs LOAD FIGURE 9. VOUT2 = 3.3V FREQUENCY vs LOAD 100 SKIP 6 2 SKIP 0 0.001 8 50 100 OUTPUT LOAD (mA) 150 FIGURE 13. LDO OUTPUT 5V vs LOAD 10 200 3.00 0 50 100 OUTPUT LOAD (mA) 150 200 FIGURE 14. LDO OUTPUT 3.3V vs LOAD FN6396.1 August 10, 2010 ISL8112 Circuit of Figure 17 and Figure 18, no load on LDO, VSEN1, VSEN2, VREF2, and VREF1, VIN = 12V, EN2 = EN1 = VCC, VBYP = 5V, PVCC = 5V, VEN_LDO = 5V, TA = -40°C to +100°C, unless otherwise noted. Typical values are at TA = +25°C. (Continued) 177.5 26.5 177.0 26.0 176.5 25.5 INPUT CURRENT (µA) INPUT CURRENT (µA) Typical Performance Curves 176.0 175.5 175.0 174.5 174.0 173.5 173.0 7 25.0 24.5 24.0 23.5 23.0 22.5 9 11 13 15 17 19 INPUT VOLTAGE (V) 21 23 FIGURE 15. STANDBY INPUT CURRENT vs VIN (EN = EN2 = 0, EN_LDO = VCC) Typical Application Circuits The typical application circuits are shown in Figures 17, 18 and 19. In Figure 17, the power supply system generates 1.25V/5A and dynamic voltage/10A. Figure 18 shows system having1.5V/5A and 1.05V/5A output. The input supply range is 5.5V to 25V. Figure 19 shows system having1.2V/15A and 2.5V/5A output. The input supply range is 5.5V to 25V and 4.5V to 5.5V respectively. Detailed Description The ISL8112 dual-buck, BiCMOS, switch-mode powersupply controller generates logic supply voltages for notebook computers. The ISL8112 is designed primarily for battery-powered applications where high efficiency and lowquiescent supply current are critical. The ISL8112 provides a pin-selectable switching frequency, allowing operation for 200kHz/300kHz, 400kHz/300kHz, or 400kHz/500kHz on the SMPSs. Light-load efficiency is enhanced by automatic Idle-Mode operation, a variable-frequency pulse-skipping mode that reduces transition and gate-charge losses. Each step-down, power-switching circuit consists of two n-channel MOSFETs, a rectifier, and an LC output filter. The output voltage is the average AC voltage at the switching node, which is regulated by changing the duty cycle of the MOSFET switches. The gate-drive signal to the n-channel high-side MOSFET must exceed the battery voltage, and is provided by a flying-capacitor boost circuit that uses a 100nF capacitor connected to BOOT_. Both SMPS1 and SMPS2 PWM controllers consist of a triple-Mode feedback network and multiplexer, a multi-input PWM comparator, high-side and low-side gate drivers and logic. In addition, SMPS2 can also use OUT2REF to track its output from 0.5V to 2.50V. The ISL8112 contains faultprotection circuits that monitor the main PWM outputs for 11 22.0 7 25 9 11 13 15 17 19 INPUT VOLTAGE (V) 21 23 25 FIGURE 16. SHUTDOWN INPUT CURRENT vs VIN (EN = EN2 = EN_LDO = 0) undervoltage and overvoltage conditions. A power-on sequence block controls the power-up timing of the main PWMs and monitors the outputs for undervoltage faults. The ISL8112 includes an adjustable low drop-out linear regulator. The bias generator blocks include the linear regulator, 3.3V precision reference, 2V precision reference and automatic bootstrap switch over circuit. The synchronous-switch gate drivers are directly powered from PVCC, while the high-side switch gate drivers are indirectly powered from PVCC through an external capacitor and an internal Schottky diode boost circuit. An automatic bootstrap circuit turns off the LDO linear regulator and powers the device from BYP if LDOREF is set to GND or VCC. See Table 1. TABLE 1. LDO OUTPUT VOLTAGE TABLE LDO VOLTAGE CONDITIONS COMMENT VOLTAGE at BYP LDOREF < 0.3V, BYP > 4.63V Internal LDO is disabled. VOLTAGE at BYP LDOREF > VCC - 1V, BYP > 3V Internal LDO is disabled. 5V LDOREF < 0.3V, BYP < 4.63V Internal LDO is active. 3.3V LDOREF > VCC - 1V, BYP < 3V Internal LDO is active. 2 x LDOREF 0.35V <LDOREF < 2.25V Internal LDO is active. FREE-RUNNING, CONSTANT ON-TIME PWM CONTROLLER WITH INPUT FEED-FORWARD The constant on-time PWM control architecture is a pseudo-fixed-frequency, constant on-time, current-mode type with voltage feed forward. The constant on-time PWM control architecture relies on the output ripple voltage to FN6396.1 August 10, 2010 ISL8112 provide the PWM ramp signal; thus the output filter capacitor's ESR acts as a current-feedback resistor. The high-side switch on-time is determined by a one-shot whose period is inversely proportional to input voltage and directly proportional to output voltage. Another one-shot sets a minimum off-time (300ns typ). The on-time one-shot triggers when the following conditions are met: the error comparator's output is high, the synchronous rectifier current is below the current-limit threshold, and the minimum off time one-shot has timed out. where: • VDROP1 is the sum of the parasitic voltage drops in the inductor discharge path, including synchronous rectifier, inductor, and PC board resistances • VDROP2 is the sum of the parasitic voltage drops in the charging path, including high-side switch, inductor, and PC board resistances • tON is the on-time calculated by the ISL8112. TABLE 2. APPROXIMATE K-FACTOR ERRORS ON-TIME ONE-SHOT (FS) Each PWM core includes a one-shot that sets the high-side switch on-time for each controller. Each fast, low-jitter, adjustable one-shot includes circuitry that varies the on-time in response to battery and output voltage. 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. This algorithm results in a nearly constant switching frequency despite the lack of a fixed-frequency clock generator. The benefit of a constant switching frequency is that the frequency can be selected to avoid noise-sensitive frequency regions: K ( V OUT + I LOAD ⋅ r DSON ( LOWERQ ) ) t ON = -----------------------------------------------------------------------------------------------------V IN (EQ. 1) SMPS APPROXIMATE SWITCHING K-FACTOR FREQUENCY K-FACTOR ERROR (%) (kHz) (µs) (FS = GND, VREF1, or OPEN), VSEN1 400 2.5 ±10 (FS = GND), VSEN2 500 2.0 ±10 (FS = VCC), VSEN1 200 5.0 ±10 (FS = VCC, VREF1, or OPEN), VSEN2 300 3.3 ±10 See Table 2 for approximate K- factors. Switching frequency increases as a function of load current due to the increasing drop across the synchronous rectifier, which causes a faster inductor-current discharge ramp. On-times translate only roughly to switching frequencies. The on-times guaranteed in the Electrical Characteristics are influenced by switching delays in the external high-side power MOSFET. Also, the dead-time effect increases the effective on-time, reducing the switching frequency. It occurs only in PWM mode (MODE = VCC) and during dynamic output voltage transitions when the inductor current reverses at light or negative load currents. With reversed inductor current, the inductor's EMF causes PH_ to go high earlier than normal, extending the on-time by a period equal to the UG-rising dead time. For loads above the critical conduction point, the actual switching frequency is: V OUT + V DROP1 f = ------------------------------------------------------t ON ( V IN + V DROP2 ) (EQ. 2) 12 FN6396.1 August 10, 2010 ISL8112 VIN: 5.5V to 25V 5V C5 1µF C8 1µF PVCC VCC VIN C10 10µF C1 10µF 10 BOOT2 Q3a C9 0.1µF Q3b C11 330µF 9mΩ 6.3V GND LDOREF BOOT1 SI4816BDY OUT1 – PCI-e L1: 3.3µH 1.25V/5A NC LDO UG1 UG2 PH1 PH2 LG1 LG2 VCC BYP R3 200kΩ R2 10kΩ AGND OUT2REF VCC NC VCC FS 2 BITS DAC + R5 200kΩ VREF2 MODE EN_LDO C2 2 x 330µF 4mΩ 6.3V OUT2REF: DYNAMIC 0 TO 2.5V OUT2REF TIED TO VREF2 = 1.05V OUT2REF TIED TO VCC = 3.3V ILIM2 ILIM1 GND VCC EN2 ISL8112 FB1 FB1 TIED TO GND = 5V FB1 TIED TO VCC = 1.5V Q2 IRF7832 VSEN2 EN1 5V OUT2-GFX L2: 2.2µH TRACK OUT2REF/10A PGND VSEN1 R1 7.87kΩ C4 0.22µF Q1 IRF7821 VREF1 C3 OPEN C7 0.1µF VCC + DROOP + VCC R4 200kΩ R6 200kΩ PGOOD1 PGOOD2 PAD FREQUENCY-DEPENDENT COMPONENTS 1.25V/1.05V SMPS SWITCHING FREQUENCY FS = VCC 200kHz/300kHz L1 3.3µH L2 2.7µH C2 2 x 330µF C11 330µF FIGURE 17. ISL8112 TYPICAL DYNAMIC GFX APPLICATION CIRCUIT 13 FN6396.1 August 10, 2010 ISL8112 VIN: 5.5V to 25V 5V C5 1µF LDOREF TIED TO GND = 5V LDOREF TIED TO VCC = 3.3V LDO C8 1µF PVCC VCC LDO VIN C10 10µF SI4816BDY VCC LDOREF BOOT1 C1 10 10µF BOOT2 Q3a OUT1 1.5V/5A C11 33µF 9mΩ 6.3V Q3b PH1 PH2 LG1 LG2 Q1b SI4816BDY ISL8112 OUT2REF AGND R5 200kΩ MODE VREF2 EN_LDO VREF1 ON OFF NC VCC FS OUT2REF: DYNAMIC 0 TO 2.5V OUT2REF TIED TO VREF2 = 1.05V VREF2 OUT2REF TIED TO VCC = 3.3V ILIM2 ILIM1 VCC C2 330µF 4mΩ 6.3V VCC EN2 FB1 R3 200kΩ OUT2 L2: 2.2µF 1.05V/5A VSEN2 EN1 BYP C4 0.22µF PGND VSEN1 VCC 3.3V VCC FB1 TIED TO GND = 5V FB1 TIED TO VCC = 1.5V Q1a UG2 UG1 C9 0.1µF L1: 3.3µH C6 4.7µF F C3 0.01µF C7 0.1µF VCC R4 200kΩ VCC R6 200kΩ PGOOD1 PGOOD2 PAD FREQUENCY-DEPENDENT COMPONENTS 1.5V/1.05V SMPS SWITCHING FREQUENCY FS = VCC 200kHz/300kHz L1 3.3µH L2 2.7µH C2 330µF C11 330µF FIGURE 18. ISL8112 TYPICAL SYSTEM REGULATOR APPLICATION CIRCUIT 14 FN6396.1 August 10, 2010 ISL8112 VIN: 4.5V to 5.5V R7 Ω 1O C5 1µF C8 1µF PVCC VCC VIN LDO NC GND LDOREF LDOREF C1 C10 BOOT1 10 µ F Q3a SI4816BDY 2.5V/5A C9 0.1µF L2: 1.5µH Q3b C11 330µF Ω 9mO 6.3V UG1 UG2 PH1 PH2 LG1 LG2 R1 Ω 110kO R3 200k Ω R2 Ω 43kO ISL8112 1.2V/15A Q2 C2 3 x 330µF Ω 4mO 6.3V IRF7832 OUT2REF ILIM1 ILIM2 GND EN_LDO VCC NC VREF1 REF VCC EN2 AGND R8 Ω 73kO OUT2REF: DYNAMIC 0 TO 2.5V OUT2REF tied to VREF2 VREF3=1.05V OUT2REF tied to VCC=3.3V R5 Ω 225kO VREF2 MODE VCC L1: 1.5µH 0.22µF VSEN2 FB1 FB1 tied to VCC=1.5V GND=5V FB1 tied to VCC=1.5V IRF7821 PGND EN1 BYP GND Q1 C4 VSEN1 VCC 10 µ F BOOT2 VREF1 R9 Ω 110kO C3 OPEN C7 0.1µF VCC VCC R4 Ω 225kO R6 Ω 225kO PGOOD1 FS PGOOD2 PAD FREQUENCY-DEPENDENT COMPONENTS 1.2V/2.5V SMPS SWITCHING FREQUENCY FS = GND 400kHz/500kHz L1 1.5µH L2 1.5µH C2 3X330µF C11 330µF FIGURE 19. ISL8112 TYPICAL SYSTEM REGULATOR APPLICATION CIRCUIT 15 FN6396.1 August 10, 2010 ISL8112 FS MODE BOOT2 BOOT1 UG2 UG1 PH2 PH1 PVCC PVCC LG1 SMPS1 SMPS2 SYNCH. SYNCH. PWM BUCK PWM BUCK CONTROLLER CONTROLLER GND ILIM1 EN1 FB1 PGOOD1 VSEN1 PGND ILIM2 EN2 OUT2REF PGOOD2 VSEN2 VSEN2 VSEN1 BYP LG2 PGOOD2 + - SW THRES. PGOOD1 LDO LDO VCC LDO VCC INTERNAL LOGIC LDOREF M1 VIN 10Ω PVCC EN_LDO POWER-ON POWER-ON VREF2 SEQUENCE SQUENCE EN1 VREF2 CLEAR FAULT CLEAR FAULT LATCH LATCH EN2 THERMAL THERMAL SHUTDOWN SHUTDOWN VREF1 VREF1 FIGURE 20. DETAIL FUNCTIONAL DIAGRAM ISL8112 16 FN6396.1 August 10, 2010 ISL8112 FS Min. tOFF Q TRIG ONE SHOT VIN VSEN_ + TO UG_DRIVER Q R Q S Q Q OUT2REF (SMPS2) VREF COMP SLOPE COMP + + + ++ ILIM_ + 5µA VCC BOOT_ BOOT UV DETECT + TO LG_ DRIVER Â S + PH_ VSEN_ Q S Q + R Q Q MODE PGOOD_ + OV_LATCH_ FB_ 1.1VREF 0.7VREF UV_LATCH_ + 0.9VREF + FB DECODER FAULT FAULT LATCH LATCH LOGIC 20ms BLANKING FIGURE 21. PWM CONTROLLER (ONE SIDE ONLY) Automatic Pulse-Skipping Switch Over (Idle Mode) K ⋅ V OUT V IN – V OUT I LOAD ( SKIP ) = ------------------------ -------------------------------2⋅L V IN = VIN-VOUT L IPEAK ILOAD= IPEAK /2 (EQ. 3) where K is the on-time scale factor (see “On-Time One-Shot (FS)” on page 12). The load-current level at which PFM/PWM crossover occurs, ILOAD(SKIP), is equal to half the peak-to-peak ripple current, which is a function of the inductor value (Figure 22). For example, in the ISL8112 typical application circuit with VOUT1 = 5V, VIN = 12V, L = 7.6µH, and K = 5µs, switch over to pulse-skipping operation occurs at ILOAD = 0.96A or about on-fifth full load. The crossover point occurs at an even lower value if a swinging (soft-saturation) inductor is used. 17 t INDUCTOR CURRENT In Idle Mode (MODE = GND), an inherent automatic switch over to PFM takes place at light loads. This switch over is affected by a comparator that truncates the low-side switch on-time at the inductor current's zero crossing. This mechanism causes the threshold between pulse-skipping PFM and non-skipping PWM operation to coincide with the boundary between continuous and discontinuous inductor-current operation (also known as the critical conduction point): ΔII 0 ON-TIME TIME FIGURE 22. ULTRASONIC CURRENT WAVEFORMS The switching waveforms may appear noisy and asynchronous when light loading causes pulse-skipping operation, but this is a normal operating condition that results in high light-load efficiency. Trade-offs in PFM noise vs. light-load efficiency are made by varying the inductor value. Generally, low inductor values produce a broader efficiency vs. load curve, while higher values result in higher full-load efficiency (assuming that the coil resistance remains fixed) and less output voltage ripple. Penalties for using FN6396.1 August 10, 2010 ISL8112 higher inductor values include larger physical size and degraded load-transient response (especially at low input-voltage levels). DC output accuracy specifications refer to the trip level of the error comparator. When the inductor is in continuous conduction, the output voltage has a DC regulation higher than the trip level by 50% of the ripple. In discontinuous conduction (MODE = GND, light load), the output voltage has a DC regulation higher than the trip level by approximately 1.0% due to slope compensation. 40µs (MAX) INDUCTOR CURRENT ZERO-CROSSING DETECTION DETECTION 0A FB<REG.POINT FB<REG.POINT Forced-PWM Mode The low-noise, forced-PWM (MODE = VCC) mode disables the zero-crossing comparator, which controls the low-side switch on-time. Disabling the zero-crossing detector causes the low-side, gate-drive waveform to become the complement of the high-side, gate-drive waveform. The inductor current reverses at light loads as the PWM loop strives to maintain a duty ratio of VOUT/VIN. The benefit of forced-PWM mode is to keep the switching frequency fairly constant, but it comes at a cost: the no-load battery current can be 10mA to 50mA, depending on switching frequency and the external MOSFETs. Reference and Linear Regulators (VREF2, VREF1, and LDO) Forced-PWM mode is most useful for reducing audio-frequency noise, improving load-transient response, providing sink-current capability for dynamic output voltage adjustment, and improving the cross-regulation of multiple-output applications that use a flyback transformer or coupled inductor. The 2V reference (VREF1) is accurate to ±1% over temperature, also making VREF1 useful as a precision system reference. Bypass VREF1 to GND with a 0.1µF (min) capacitor. VREF1 can supply up to 50µA for external loads. Enhanced Ultrasonic Mode (25kHz (min) Pulse Skipping) Leaving MODE unconnected or connecting MODE to VREF1 activates a unique pulse-skipping mode with a minimum switching frequency of 25kHz. This ultrasonic pulse-skipping mode eliminates audio-frequency modulation that would otherwise be present when a lightly loaded controller automatically skips pulses. In ultrasonic mode, the controller automatically transitions to fixed-frequency PWM operation when the load reaches the same critical conduction point (ILOAD(SKIP)). An ultrasonic pulse occurs when the controller detects that no switching has occurred within the last 20µs. Once triggered, the ultrasonic controller pulls LG high, turning on the low-side MOSFET to induce a negative inductor current. After FB drops below the regulation point, the controller turns off the low-side MOSFET (LG pulled low) and triggers a constant on-time (UG driven high). When the on-time has expired, the controller re-enables the low-side MOSFET until the controller detects that the inductor current dropped below the zero-crossing threshold. Starting with a LG pulse greatly reduces the peak output voltage when compared to starting with a UG pulse, as long as VFB < VREF, LG is off and UG is on, similar to pure SKIP mode. 18 ON-TIME (TON ON) FIGURE 23. ULTRASONIC CURRENT WAVEFORMS The 3.3V reference (VREF2) is accurate to ±1.5% over temperature, making VREF2 useful as a precision system reference. VREF2 can supply up to 5mA for external loads. Bypass VREF2 to GND with a 0.01µF capacitor. Leave open if there is no load. An internal regulator produces a fixed 5V (LDOREF < 0.2V) or 3.3V (LDOREF > VCC - 1V). In an adjustable mode, the LDO output can be set from 0.7V to 4.5V. The LDO output voltage is equal to two times the LDOREF voltage. The LDO regulator can supply up to 100mA for external loads. Bypass LDO with a minimum 4.7µF ceramic capacitor. When the LDOREF < 0.2V and BYP voltage is 5V, the LDO bootstrapswitch over to an internal 0.7Ω p-channel MOSFET switch connects BYP to LDO pin while simultaneously shutting down the internal linear regulator. These actions bootstrap the device, powering the loads from the BYP input voltages, rather than through internal linear regulators from the battery. Similarly, when the BYP = 3.3V and LDOREF = VCC, the LDO bootstrap-switch over to an internal 1.5Ω P-Channel MOSFET switch connects BYP to LDO pin while simultaneously shutting down the internal linear regulator. No switch over action in adjustable mode. Current-Limit Circuit (ILIM_) with rDS(ON) Temperature Compensation The current-limit circuit employs a "valley" current-sensing algorithm. The ISL8112 uses the on-resistance of the synchronous rectifier as a current-sensing element. If the magnitude of the current-sense signal at PH_ is above the current-limit threshold, the PWM is not allowed to initiate a new cycle. The actual peak current is greater than the FN6396.1 August 10, 2010 ISL8112 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 current-limit threshold, inductor value and input and output voltage. INDUCTOR CURRENT I PEAK I LOAD ΔI I LIMIT I LOAD(MAX) A negative current limit prevents excessive reverse inductor currents when VOUT sinks current. The negative current-limit threshold is set to approximately 120% of the positive current limit and therefore tracks the positive current limit when ILIM_ is adjusted. The current-limit threshold is adjusted with an external resistor for ISL8112 at ILIM_. The current-limit threshold adjustment range is from 20mV to 200mV. In the adjustable mode, the current-limit threshold voltage is 1/10th the voltage at ILIM_. The voltage at ILIM pin is the product of 5µA * RILIM. The threshold defaults to 100mV when ILIM_ is connected to VCC. The logic threshold for switch-over to the 100mV default value is approximately VCC - 1V. The PC board layout guidelines should be carefully observed to ensure that noise and DC errors do not corrupt the current-sense signals at PH_. I LIM( VAL ) = ILOAD - Δ I 2 MOSFET Gate Drivers (UG_, LG_) TIME FIGURE 24. “VALLEY” CURRENT LIMIT THRESHOLD POINT For lower power dissipation, the ISL8112 uses the on-resistance of the synchronous rectifier as the current-sense element. Use the worst-case maximum value for rDS(ON) from the MOSFET data sheet. Add some margin for the rise in rDS(ON) with temperature. A good general rule is to allow 0.5% additional resistance for each °C of temperature rise. The ISL8112 controller has a built-in 5µA current source as shown in Figure 25. Place the hottest power MOSEFTs as close to the IC as possible for best thermal coupling. The current limit varies with the onresistance of the synchronous rectifier. When combined with the undervoltage-protection circuit, this current-limit method is effective in almost every circumstance. ILIM_ The UG_ and LG_ gate drivers sink 2.0A and 3.3A respectively of gate drive, ensuring robust gate drive for high-current applications. The UG_ floating high-side MOSFET drivers are powered by diode-capacitor charge pumps at BOOT_. The LG_ synchronous-rectifier drivers are powered by PVCC. The internal pull-down transistors that drive LG_ low have a 0.6Ω typical on-resistance. These low on-resistance pull-down transistors prevent LG_ from being pulled up during the fast rise time of the inductor nodes due to capacitive coupling from the drain to the gate of the low-side synchronous-rectifier MOSFETs. However, for high-current applications, some combinations of high- and low-side MOSFETs may cause excessive gate-drain coupling, which leads to poor efficiency and EMI-producing shoot-through currents. Adding a 4.7Ω resistor in series with BOOT_ increases the turn-on time of the high-side MOSFETs at the expense of efficiency, without degrading the turn-off time (Figure 26). + PVCC 5V 5µA BOOT_ 4.7Ω + R ILIM VILIM 9R VCC TO CURRENT LIMIT LOGIC UG_ R C BOOT VIN Q1 OUT_ PH_ FIGURE 25. CURRENT LIMIT BLOCK DIAGRAM ISL8112 FIGURE 26. REDUCING THE SWITCHING-NODE RISE TIME 19 FN6396.1 August 10, 2010 ISL8112 Adaptive dead-time circuits monitor the LG_ and UG_ drivers and prevent either FET from turning on until the other is fully off. This algorithm allows operation without shootthrough with a wide range of MOSFETs, minimizing delays and maintaining efficiency. There must be low-resistance, low-inductance paths from the gate drivers to the MOSFET gates for the adaptive dead-time circuit to work properly. Otherwise, the sense circuitry interprets the MOSFET gate as "off" when there is actually charge left on the gate. Use very short, wide traces measuring 10 to 20 squares (50 mils to 100 mils wide if the MOSFET is 1” from the device). Boost-Supply Capacitor Selection (Buck) The boost capacitor should be 0.1µF to 4.7µF, depending on the input and output voltages, external components, and PC board layout. The boost capacitance should be as large as possible to prevent it from charging to excessive voltage, but small enough to adequately charge during the minimum low-side MOSFET conduction time, which happens at maximum operating duty cycle (this occurs at minimum input voltage). The minimum gate to source voltage (VGS(MIN)) is determined by: C BOOT V GS ( MIN ) = PVCC ⋅ --------------------------------------C BOOT + C GS (EQ. 4) where: • PVCC is 5V • CGS is the gate capacitance of the high-side MOSFET POR, UVLO, and Internal Digital Soft-Start Power-on reset (POR) occurs when VIN rises above approximately 3V. UVLO occurs when PVCC drops below approximately 4V. The VIN POR reset the LDO control. The UVLO resets the undervoltage, overvoltage, and thermalshutdown fault latches. PVCC undervoltage lockout (UVLO) circuitry inhibits switching when PVCC is below 4V. LG_ is low during UVLO. The output voltages begin to ramp up once PVCC exceeds its 4V UVLO and VREF1 is in regulation. The internal digital soft-start timer begins to ramp up the maximum-allowed current limit during start-up. The 1.7ms ramp occurs in five steps of positive current limit and the step size is 20%, 40%, 60%, 80% and 100%. Power-Good Output (PGOOD_) The PGOOD_ comparator continuously monitors both output voltages for undervoltage conditions. PGOOD_ is actively held low in shutdown, standby, and soft-start. PGOOD1 releases and digital soft-start terminates when VSEN1 reach the error-comparator threshold. PGOOD1 goes low if VOUT1 output turns off or is 10% below its nominal regulation point. PGOOD1 is a true open-drain output. Likewise, PGOOD2 is used to monitor VSEN2. 20 Fault Protection The ISL8112 provides overvoltage/undervoltage fault protection in the buck controllers. Once activated, the controller continuously monitors the output for undervoltage and overvoltage fault conditions. • Out-of-bound Condition When the output voltage is 5% above the set voltage, the out-of-bound condition activates. LG turns on until output reaches within regulation. Once the output is within regulation, the controller will operate as normal. It is the "first line of defense" before OVP. • Overvoltage Protection When VSEN1 is 11% (16% for VSEN2) above the set voltage, the overvoltage fault protection activates. This latches on the synchronous rectifier MOSFET with 100% duty cycle, rapidly discharging the output capacitor until the negative current limit is achieved. Once negative current limit is met, UG is turned on for a minimum ontime, followed by another LG pulse until negative current limit. This effectively regulates the discharge current at the negative current limit in an effort to prevent excessively large negative currents that cause potentially damaging negative voltages on the load. Once an overvoltage fault condition is set, it can only be reset by toggling SHDN#, EN_, or cycling PVCC(UVLO). • Undervoltage Protection When the output voltage drops below 70% of its regulation voltage for at least 100µs, the controller sets the fault latch and begins the discharge mode (see the Shutdown and Output Discharge section). UVP is ignored for at least 20ms (typical), after start-up or after a rising edge on EN_. Toggle EN_ or cycle PVCC (UVLO) to clear the undervoltage fault latch and restart the controller. UVP only applies to the buck outputs. • Thermal Protection The ISL8112 has thermal shutdown to protect the devices from overheating. Thermal shutdown occurs when the die temperature exceeds +150°C. All internal circuitry shuts down during thermal shutdown. The ISL8112 may trigger thermal shutdown if LDO_ is not bootstrapped from VSEN_ while applying a high input voltage on VIN and drawing the maximum current (including short circuit) from LDO_. Even if LDO_ is bootstrapped from VSEN_, overloading the LDO_ causes large power dissipation on the bootstrap switches, which may result in thermal shutdown. Cycling EN_, EN_LDO, or PVCC(UVLO) ends the thermal-shutdown state. FN6396.1 August 10, 2010 ISL8112 Discharge Mode (Soft-Stop) shutdown mode activates, the 3.3V VREF2 remain on. Both SMPS outputs are discharged to 0V through a 25Ω switch. When a transition to standby or shutdown mode occurs, or the output is discharged to GND through an internal 25Ω switch, the reference remains active to provide an accurate threshold and to provide overvoltage protection. Power-Up Sequencing and On/Off Controls (EN_) EN1 and EN2 control SMPS power-up sequencing. EN1 or EN2 rising above 2.4V enables the respective outputs. EN1 or EN2 falling below 1.6V disables the respective outputs. When the output undervoltage fault latch is set, both channels are discharged to GND through the internal 25Ω switches. Connecting EN1 or EN2 to VREF1 will force its outputs off while the other output is below regulation. The sequenced SMPS will start once the other SMPS reaches regulation. The second SMPS remains on until the first SMPS turns off, the device shuts down, a fault occurs or PVCC goes into undervoltage lockout. Both supplies begin their power-down sequence immediately when the first supply turns off. Driving EN_ below 0.8V clears the overvoltage, undervoltage and thermal fault latches. Shutdown Mode The ISL8112 SMPS1, SMPS2 and LDO have independent enabling control. Drive EN1, EN2 and EN_LDO below the precise input falling-edge trip level to place the ISL8112 in its low-power shutdown state. The ISL8112 consumes only 20µA of quiescent current while in shutdown. When TABLE 3. OPERATING-MODE TRUTH TABLE MODE CONDITION COMMENT Power-Up PVCC < UVLO threshold. Transitions to discharge mode after a PVCC UVLO and after VREF1 becomes valid. LDO, VREF2, and VREF1 remain active. Run EN_LDO = high, EN1 or EN2 enabled. Normal operation Overvoltage Protection Either output > 111% (VSEN1) or 116% (VSEN2) of nominal level. LG_ is forced high. LDO, VREF2 and VREF1 active. Exited by a PVCC UVLO, VCC POR, or by toggling EN1 or EN2. Undervoltage Protection Either output < 70% of nominal after 20ms time-out expires and output is enabled. Both the internal 25Ω switches turn on. LDO, VREF2 and VREF1 are active. Exited by a PVCC UVLO, or by toggling EN1 or EN2. Discharge Either SMPS output is still high in either standby mode or shutdown mode Discharge switch (25Ω) connects VSEN_ to GND. One output may still run while the other is in discharge mode. Activates when PVCC is in UVLO, or transition to UVLO, standby, or shutdown has begun. LDO, VREF2 and VREF1 active. Standby EN1, EN2 < startup threshold, EN_LDO= High LDO, VREF2 and VREF1 active. Shutdown EN1, EN2, EN_LDO = low Discharge switch (25Ω) connects VSEN_ to PGND. All circuitry off except VREF2. Thermal Shutdown TJ > +150°C All circuitry off. Exited by PVCC UVLO or cycling EN_. VREF2 remain active. TABLE 4. SHUTDOWN AND STANDBY CONTROL LOGIS VEN_LDO VEN1 (V) VEN2 (V) LDO SMPS1 SMPS2 Low Low Low Off Off Off “>2.5” → High Low Low On Off Off “>2.5” → High High High On On On “>2.5” → High High Low On On Off “>2.5” → High Low High On Off On “>2.5” → High High VREF1 On On On (after SMPS1 is up) “>2.5” → High VREF1 High On On (after SMPS2 is up) On 21 FN6396.1 August 10, 2010 ISL8112 Adjustable-Output Feedback (Dual-Mode FB) Connect FB1 to GND to enable the fixed 5V or tie FB1 to VCC to set the fixed 1.5V output. Connect a resistive voltage-divider at FB1 between output and GND to adjust the respective output voltage between 0.7V and 5.5V (Figure 27). Choose R2 to be approximately 10k and solve for R1 using Equation 5. ⎛ V OUT1 ⎞ R1 = R2 ⋅ ⎜ ------------------- – 1⎟ ⎝ V FB1 ⎠ (EQ. 5) VIN Q3 ISL88732 ISL88733 ISL8112 ISL6236 ISL88734 ISL88734 OUT1 LGATE_ LGATE1 LG1 Q4 OUT1 VOUT_ VOUT_ VSEN1 FB1 FB_ FB1 R2 FIGURE 27. SETTING VOUT1 WITH A RESISTOR DIVIDER Likewise, connect OUT2REF to VCC to enable the fixed 3.3V or tie OUT2REF to VREF2 to set the fixed 1.05V output. Set OUT2REF from 0 to 2.50V for SMPS2 tracking mode (Figure 28). VIN Q1 ISL88732 ISL88733 ISL6236 ISL8112 ISL88734 OUT2 LG2 LGATE_ LGATE2 Q2 VSEN2 VOUT_ OUT2 OUT2REF FB_ REFIN2 VR R3 R4 FIGURE 28. SETTING VOUT2 WITH A VOLTAGE DIVIDER FOR TRACKING 22 where: • VR = 2V nominal (if tied to VREF1) or • VR = 3.3V nominal (if tied to VREF2) Design Procedure 1. Input Voltage Range. The maximum value (VIN(MAX)) must accommodate the maximum AC adapter voltage. The minimum value (VIN(MIN)) must account for the lowest input voltage after drops due to connectors, fuses and battery selector switches. Lower input voltages result in better efficiency. 2. Maximum Load Current. The peak load current (ILOAD(MAX)) determines the instantaneous component stress 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 stress and drives the selection of input capacitors, MOSFETs and other critical heat-contributing components. R1 UG2 UGATE_ UGATE2 (EQ. 6) OUT2 Establish the input voltage range and maximum load current before choosing an inductor and its associated ripple-current ratio (LIR). The following four factors dictate the rest of the design: where VFB1 = 0.7V nominal. UG1 UGATE_ UGATE1 VR R3 = R4 ⋅ ⎛ ------------------- – 1⎞ ⎝V ⎠ 3. Switching Frequency. This choice determines the basic trade-off between size and efficiency. The optimal frequency is largely a function of maximum input voltage and MOSFET switching losses. 4. Inductor Ripple Current Ratio (LIR). LIR is the ratio of the peak-peak ripple current to the average inductor current. Size and efficiency trade-offs must be considered when setting the inductor ripple current ratio. Low inductor values cause large ripple currents, resulting in the smallest size, but poor efficiency and high output noise. The minimum practical inductor value is one that causes the circuit to operate at 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 ISL8112 pulse-skipping algorithm (MODE = GND) initiates skip mode at the critical conduction point, so the inductor's operating point also determines the load current at which PWM/PFM switch over occurs. The optimum point is usually found between 20% and 50% ripple current. Inductor Selection The switching frequency (on-time) and operating point (% ripple or LIR) determine the inductor value as follows: V OUT_ ( V IN + V OUT_ ) L = --------------------------------------------------------------------V IN ⋅ f ⋅ LIR ⋅ I LOAD ( MAX ) (EQ. 7) FN6396.1 August 10, 2010 ISL8112 Output Capacitor Selection Example: ILOAD(MAX) = 5A, VIN = 12V, VOUT2 = 5V, f = 200kHz, 35% ripple current or LIR = 0.35: 5V ( 12V – 5V ) L = ----------------------------------------------------------------- = 8.3μH 12V ⋅ 200kHz ⋅ 0.35 ⋅ 5A (EQ. 8) Find a low-loss inductor having the lowest possible DC resistance that fits in the allotted dimensions. Ferrite cores are often the best choice. The core must be large enough not to saturate at the peak inductor current (IPEAK): IPEAK = I LOAD ( MAX ) + [ ( LIR ⁄ 2 ) ⋅ I LOAD ( MAX ) ] (EQ. 9) The inductor ripple current also impacts transient response performance, especially at low VIN - VSEN_ differences. Low inductor values allow the inductor current to slew faster, replenishing charge removed from the output filter capacitors by a sudden load step. The peak amplitude of the output transient (VSAG) is also a function of the maximum duty factor, which can be calculated from the on-time and minimum off-time: ⎞⎞ ⎛ ⎛ V OUT_ 2 ( ΔI LOAD ( MAX ) ) ⋅ L ⎜ K ⎜ ------------------- + t OFF ( MIN )⎟ ⎟ V ⎠⎠ ⎝ ⎝ IN VSAG = ---------------------------------------------------------------------------------------------------------------------------– V V ⎛ IN OUT⎞ 2 ⋅ C OUT ⋅ V OUT K ⎜ --------------------------------⎟ - t V IN ⎝ ⎠ OFF ( MIN ) (EQ. 10) where minimum off-time = 0.35µs (max) and K is from Table 2. Determining the Current Limit The output filter capacitor must have low enough equivalent series resistance (ESR) to meet output ripple and load-transient requirements, yet have high enough ESR to satisfy stability requirements. The output capacitance must also be high enough to absorb the inductor energy while transitioning from full-load to no-load conditions without tripping the overvoltage fault latch. In applications where the output is subject to large load transients, the output capacitor's size 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 DIP R SER ≤ ---------------------------------I LOAD ( MAX ) (EQ. 14) where VDIP is the maximum-tolerable transient voltage drop. In non-CPU applications, the output capacitor's size depends on how much ESR is needed to maintain an acceptable level of output voltage ripple: VP – P R ESR ≤ ----------------------------------------------L IR ⋅ I LOAD ( MAX ) (EQ. 15) where VP-P is the peak-to-peak output voltage ripple. The actual capacitance value required relates to the physical size needed to achieve low ESR, as well as to the chemistry of the capacitor technology. Thus, the capacitor is usually selected by ESR and voltage rating rather than by capacitance value (this is true of tantalum, OS-CON, and other electrolytic-type capacitors). The minimum current-limit threshold must be great 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 of the ripple current; therefore: When using low-capacity filter capacitors such as polymer types, capacitor size is usually determined by the capacity required to prevent VSAG and VSOAR from tripping the undervoltage and overvoltage fault latches during load transients in ultrasonic mode. I LIMIT ( LOW ) > I LOAD ( MAX ) – [ ( LIR ⁄ 2 ) ⋅ I LOAD ( MAX ) ] For low input-to-output voltage differentials (VIN/ VOUT < 2), additional output capacitance is required to maintain stability and good efficiency in ultrasonic mode. The amount of overshoot due to stored inductor energy can be calculated as: (EQ. 11) where: ILIMIT(LOW) = minimum current-limit threshold voltage divided by the rDS(ON) of Q2/Q4. Use the worst-case maximum value for rDS(ON) from the MOSFET Q2/Q4 data sheet and add some margin for the rise in rDS(ON) with temperature. A good general rule is to allow 0.2% additional resistance for each °C of temperature rise. Examining the 5A circuit example with a maximum rDS(ON) = 5mΩ at room temperature. At +125°C reveals the following: I LIMIT ( LOW ) = ( 25mV ) ⁄ ( ( 5mΩ × 1.2 ) > 5A – ( 0.35 ⁄ 2 )5A ) (EQ. 12) 4.17A > 4.12A (EQ. 13) 4.17A is greater than the valley current of 4.12A, so the circuit can easily deliver the full-rated 5A using the 30mV nominal current-limit threshold voltage. 23 2 I PEAK ⋅ L V SOAR = -----------------------------------------------2 ⋅ C OUT ⋅ V OUT_ (EQ. 16) where IPEAK is the peak inductor current. Input Capacitor Selection The input capacitors must meet the input-ripple-current (IRMS) requirement imposed by the switching current. The ISL8112 dual switching regulator operates at different frequencies. This interleaves the current pulses drawn by the two switches and reduces the overlap time where they add together. The input RMS current is much smaller in comparison than with both SMPSs operating in phase. The input RMS current varies with load and the input voltage. The maximum input capacitor RMS current for a single SMPS is given by: FN6396.1 August 10, 2010 ISL8112 ⎛ V OUT ( V IN – V OUT_ )⎞ I RMS ≈ I LOAD ⎜ ------------------------------------------------------------⎟ V IN ⎝ ⎠ (EQ. 17) When V IN = 2 ⋅ V OUT_ ( D = 50% ) , IRMS has maximum current of I LOAD ⁄ 2 . The ESR of the input-capacitor is important for determining capacitor power dissipation. All the power (IRMS2 x ESR) heats up the capacitor and reduces efficiency. Nontantalum chemistries (ceramic or OS-CON) are preferred due to their low ESR and resilience to power-up surge currents. Choose input capacitors that exhibit less than +10°C temperature rise at the RMS input current for optimal circuit longevity. Place the drains of the high-side switches close to each other to share common input bypass capacitors. adequate rDS(ON) at low battery voltages if it becomes extraordinarily hot when subjected to VIN(MAX). Calculating the power dissipation in NH (Q1/Q3) due to switching losses is difficult since it must allow for 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 PC board layout characteristics. The following switching-loss calculation provides only a very rough estimate and is no substitute for bench evaluation, preferably including verification using a thermocouple mounted on NH (Q1/Q3): 2 ⎛ C RSS ⋅ f SW ⋅ I LOAD⎞ PD ( Q H Switching ) = ( V IN ( MAX ) ) ⎜ -----------------------------------------------------⎟ I GATE ⎝ ⎠ (EQ. 19) Power MOSFET Selection Most of the following MOSFET guidelines focus on the challenge of obtaining high load-current capability (>5A) when using high-voltage (>20V) AC adapters. Low-current applications usually require less attention. Choose a high-side MOSFET (Q1/Q3) that has conduction losses equal to the switching losses at the typical battery voltage for maximum efficiency. Ensure that the conduction losses at the minimum input voltage do not exceed the package thermal limits or violate the overall thermal budget. Ensure that conduction losses plus switching losses at the maximum input voltage do not exceed the package ratings or violate the overall thermal budget. Choose a synchronous rectifier (Q2/Q4) with the lowest possible rDS(ON). Ensure the gate is not pulled up by the high-side switch turning on due to parasitic drain-to-gate capacitance, causing cross-conduction problems. Switching losses are not an issue for the synchronous rectifier in the buck topology since it is a zero-voltage switched device when using the buck topology. MOSFET Power Dissipation Worst-case conduction losses occur at the duty-factor extremes. For the high-side MOSFET, the worst-case power dissipation (PD) due to the MOSFET's rDS(ON) occurs at the minimum battery voltage: ⎛ V OUT_ ⎞ 2 PD ( Q H Resistance ) = ⎜ ------------------------⎟ ( I LOAD ) ⋅ r DS ( ON ) ⎝ V IN ( MIN )⎠ (EQ. 18) Generally, a small high-side MOSFET reduces switching losses at high input voltage. However, the rDS(ON) required to stay within package power-dissipation limits often limits how small the MOSFET can be. The optimum situation occurs when the switching (AC) losses equal the conduction (rDS(ON)) losses. Switching losses in the high-side MOSFET can become an insidious heat problem when maximum battery voltage is applied, due to the squared term in the CV2f switching-loss equation. Reconsider the high-side MOSFET chosen for 24 where CRSS is the reverse transfer capacitance of QH (Q1/Q3) and IGATE is the peak gate-drive source/sink current. For the synchronous rectifier, the worst-case power dissipation always occurs at maximum battery voltage: V OUT ⎞ ⎛ 2 PD ( Q L ) = ⎜ 1 – --------------------------⎟ I LOAD ⋅ r DS ( ON ) V IN ( MAX )⎠ ⎝ (EQ. 20) The absolute 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, "overdesign" the circuit to tolerate: I LOAD = I LIMIT ( HIGH ) + ( ( LIR ) ⁄ 2 ) ⋅ I LOAD ( MAX ) (EQ. 21) where ILIMIT(HIGH) is the maximum valley current allowed by the current-limit circuit, including threshold tolerance and resistance variation. Rectifier Selection Current circulates from ground to the junction of both MOSFETs and the inductor when the high-side switch is off. As a consequence, the polarity of the switching node is negative with respect to ground. This voltage is approximately -0.7V (a diode drop) at both transition edges while both switches are off (dead time). The drop is I L ⋅ r DS ( ON ) when the low-side switch conducts. The rectifier is a clamp across the synchronous rectifier that catches the negative inductor swing during the dead time between turning the high-side MOSFET off and the synchronous rectifier on. The MOSFETs incorporate a high-speed silicon body diode as an adequate clamp diode if efficiency is not of primary importance. Place a Schottky diode in parallel with the body diode to reduce the forward voltage drop and prevent the Q2/Q4 MOSFET body diodes from turning on during the dead time. Typically, the external diode improves the efficiency by 1% to 2%. Use a Schottky diode with a DC current rating equal to one-third of the load FN6396.1 August 10, 2010 ISL8112 current. For example, use an MBR0530 (500mA-rated) type for loads up to 1.5A, a 1N5817 type for loads up to 3A, or a 1N5821 type for loads up to 10A. The rectifier's rated reverse breakdown voltage must be at least equal to the maximum input voltage, preferably with a 20% derating factor. Applications Information Dropout Performance The output voltage-adjust range for continuous-conduction operation is restricted by the nonadjustable 350ns (max) minimum off-time one-shot. Use the slower 5V SMPS for the higher of the two output voltages for best dropout performance in adjustable feedback mode. The duty-factor limit must be calculated using worst-case values for on-times and off-times, when working with low input voltages. Manufacturing tolerances and internal propagation delays introduce an error to the FS K-factor. Also, keep in mind that transient-response performance of buck regulators operated close to dropout is poor, and bulk output capacitance must often be added (see Equation 10 on page 23). The absolute point of dropout occurs when the inductor current ramps down during the minimum off-time (ΔIDOWN) as much as it ramps up during the on-time (ΔIUP). The ratio h = ΔIUP/ΔIDOWN indicates the ability to slew the inductor current higher in response to increased load, and must always be greater than 1. As h approaches 1, the absolute minimum dropout point, the inductor current is less able to increase during each switching cycle and VSAG greatly increases unless additional output capacitance is used. A reasonable minimum value for h is 1.5, but this can be adjusted up or down to allow trade-offs between VSAG, output capacitance and minimum operating voltage. For a given value of h, the minimum operating voltage can be calculated as: ( V OUT_ + V DROP ) V IN ( MIN ) = --------------------------------------------------- + V DROP2 – V DROP1 t OFF ( MIN ) ⋅ h 1 – ⎛ ------------------------------------⎞ ⎝ ⎠ K (EQ. 22) where VDROP1 and VDROP2 are the parasitic voltage drops in the discharge and charge paths (see “On-Time One-Shot (FS)” on page 12), tOFF(MIN) is from the “Electrical Specifications” table on page 4 and K is taken from Table 2. The absolute minimum input voltage is calculated with h = 1. Operating frequency must be reduced or h must be increased and output capacitance added to obtain an acceptable VSAG if calculated VIN(MIN) is greater than the required minimum input voltage. Calculate VSAG to be sure of adequate transient response if operation near dropout is anticipated. Dropout Design Example: ISL8112: With VOUT2 = 5V, fsw = 400kHz, K = 2.25µs, tOFF(MIN) = 350ns, VDROP1 = VDROP2 = 100mV, and h = 1.5, the minimum VIN is: 25 ( 5V + 0.1V ) V IN ( MIN ) = ---------------------------------------------- + 0.1V – 0.1V = 6.65V 0.35μs ⋅ 1.5 1 – ⎛ -------------------------------⎞ ⎝ 2.25μs ⎠ (EQ. 23) Calculating with h = 1 yields: ( 5V + 0.1V ) V IN ( MIN ) = ----------------------------------------- + 0.1V – 0.1V = 6.04V 0.35μs ⋅ 1 1 – ⎛ --------------------------⎞ ⎝ 2.25μs ⎠ (EQ. 24) Therefore, VIN must be greater than 6.65V. A practical input voltage with reasonable output capacitance would be 7.5V. PC Board Layout Guidelines Careful PC board layout is critical to achieve minimal switching losses and clean, stable operation. This is especially true when multiple converters are on the same PC board where one circuit can affect the other. Refer to the ISL8112 Evaluation Kit data sheet for a specific layout example. Mount all of the power components on the top side of the board with their ground terminals flush against one another, if possible. Follow these guidelines for good PC board layout: • Isolate the power components on the top side from the sensitive analog components on the bottom side with a ground shield. Use a separate PGND plane under the VSEN1 and VSEN2 sides (called PGND1 and PGND2). Avoid the introduction of AC currents into the PGND1 and PGND2 ground planes. Run the power plane ground currents on the top side only, if possible. • Use a star ground connection on the power plane to minimize the crosstalk between VSEN1 and VSEN2. • 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 PC boards (2oz vs. 1oz) can enhance full-load efficiency by 1% or more. Correctly routing PC board traces must be approached in terms of fractions of centimeters, where a single mΩ of excess trace resistance causes a measurable efficiency penalty. • PH_ (ISL8112) and GND connections to the synchronous rectifiers for current limiting must be made using Kelvinsense connections to guarantee the current-limit accuracy with 8-pin SO MOSFETs. This is best done by routing power to the MOSFETs from outside using the top copper layer, while connecting PH_ traces inside (underneath) the MOSFETs. • 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 synchronous rectifier or between the inductor and the output filter capacitor. FN6396.1 August 10, 2010 ISL8112 • Ensure that the VSEN_ connection to COUT_ is short and direct. However, in some cases it may be desirable to deliberately introduce some trace length between the VSEN_ connector node and the output filter capacitor (see the Stability Considerations section). • Route high-speed switching nodes (BOOT_, UG_, PH_, and LG_) away from sensitive analog areas (VREF1, ILIM_, and FB_). Use PGND1 and PGND2 as an EMI shield to keep radiated switching noise away from the IC's feedback divider and analog bypass capacitors. • Make all pin-strap control input connections (MODE, ILIM_, etc.) to GND or VCC of the device. Layout Procedure Place the power components first with ground terminals adjacent (Q2/Q4 source, CIN_, COUT_). If possible, make all these connections on the top layer with wide, copper-filled areas. Mount the controller IC adjacent to the synchronous rectifier MOSFETs close to the hottest spot, preferably on the back side in order to keep UG_, GND, and the LG_ gate drive lines short and wide. The LG_ gate trace must be short and wide, measuring 50 mils to 100 mils wide if the MOSFET is 1” from the controller device. On the board's top side (power planes), make a star ground to minimize crosstalk between the two sides. The top-side star ground is a star connection of the input capacitors and synchronous rectifiers. Keep the resistance low between the star ground and the source of the synchronous rectifiers for accurate current limit. Connect the top-side star ground (used for MOSFET, input, and output capacitors) to the small island with a single short, wide connection (preferably just a via). Create PGND islands on the layer just below the top-side layer (refer to the ISL8112 EV kit for an example) to act as an EMI shield if multiple layers are available (highly recommended). Connect each of these individually to the star ground via, which connects the top side to the PGND plane. Add one more solid ground plane under the device to act as an additional shield, and also connect the solid ground plane to the star ground via. Connect the output power planes (VCORE and system ground planes) directly to the output filter capacitor positive and negative terminals with multiple vias. Group the gate-drive components (BOOT_ capacitor, VIN bypass capacitor) together near the controller device. Make the DC/DC controller ground connections as follows: 1. Near the device, create a small analog ground plane. 2. Connect the small analog ground plane to GND and use the plane for the ground connection for the VREF1 and VCC bypass capacitors, FB dividers and ILIM resistors (if any). 3. Create another small ground island for PGND and use the plane for the VIN bypass capacitor, placed very close to the device. 4. Connect the GND and PGND planes together at the metal tab under device. All Intersil U.S. products are manufactured, assembled and tested utilizing ISO9000 quality systems. Intersil Corporation’s quality certifications can be viewed at www.intersil.com/design/quality Intersil products are sold by description only. Intersil Corporation reserves the right to make changes in circuit design, software and/or specifications at any time without notice. Accordingly, the reader is cautioned to verify that data sheets are current before placing orders. Information furnished by Intersil is believed to be accurate and reliable. However, no responsibility is assumed by Intersil or its subsidiaries for its use; nor for any infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Intersil or its subsidiaries. For information regarding Intersil Corporation and its products, see www.intersil.com 26 FN6396.1 August 10, 2010 ISL8112 Package Outline Drawing L32.5x5B 32 LEAD QUAD FLAT NO-LEAD PLASTIC PACKAGE Rev 3, 5/10 4X 3.5 5.00 28X 0.50 A B 6 PIN 1 INDEX AREA 6 PIN #1 INDEX AREA 32 25 1 5.00 24 3 .30 ± 0 . 15 17 (4X) 8 0.15 9 16 TOP VIEW 0.10 M C A B + 0.07 32X 0.40 ± 0.10 4 32X 0.23 - 0.05 BOTTOM VIEW SEE DETAIL "X" 0.10 C 0 . 90 ± 0.1 C BASE PLANE SEATING PLANE 0.08 C ( 4. 80 TYP ) ( ( 28X 0 . 5 ) SIDE VIEW 3. 30 ) (32X 0 . 23 ) C 0 . 2 REF 5 ( 32X 0 . 60) 0 . 00 MIN. 0 . 05 MAX. DETAIL "X" TYPICAL RECOMMENDED LAND PATTERN NOTES: 1. Dimensions are in millimeters. Dimensions in ( ) for Reference Only. 2. Dimensioning and tolerancing conform to AMSE Y14.5m-1994. 3. Unless otherwise specified, tolerance : Decimal ± 0.05 4. Dimension applies to the metallized terminal and is measured between 0.15mm and 0.30mm from the terminal tip. 5. Tiebar shown (if present) is a non-functional feature. 6. The configuration of the pin #1 identifier is optional, but must be located within the zone indicated. The pin #1 identifier may be either a mold or mark feature. 27 FN6396.1 August 10, 2010