LTC3548-2 Dual Synchronous, Fixed /Adjustable Output, 2.25MHz Step-Down DC/DC Regulator DESCRIPTION FEATURES n n n n n n n n n n n n n n High Efficiency: Up to 95% 1.8V/800mA Fixed Output and 400mA Adjustable Output Voltage Only 40μA Quiescent Current (Both Channels) 2.25MHz Constant-Frequency Operation High Switch Current Limits: 1.2A and 0.7A No Schottky Diodes Required Low RDS(ON) Internal Switches: 0.35Ω Current Mode Operation for Excellent Line and Load Transient Response Short-Circuit Protected Low Dropout Operation: 100% Duty Cycle Ultralow Shutdown Current: IQ < 1μA Power-On Reset Output Externally Synchronizable Oscillator Small, Thermally Enhanced 3mm × 3mm DFN Package APPLICATIONS n n n n n n PDAs/Palmtop PCs Digital Cameras Cellular Phones Portable Media Players PC Cards Wireless and DSL Modems The LTC®3548-2 is a dual, constant-frequency, synchronous step-down DC/DC converter. Intended for low power applications, it operates from 2.5V to 5.5V input voltage range and has a constant 2.25MHz switching frequency, allowing the use of tiny, low cost capacitors and inductors with a profile ≤1.2mm. For channel 1, the output voltage is fixed at 1.8V/800mA; for channel 2, the output voltage is adjustable from 0.6V to VIN Internal synchronous 0.35Ω, 1.2A/0.7A power switches provide high efficiency without the need for external Schottky diodes. A user selectable mode input is provided to allow the user to trade-off noise ripple for low power efficiency. Burst Mode® operation provides high efficiency at light loads, while pulse-skipping mode provides low noise ripple at light loads. To further maximize battery run time, the P-channel MOSFETs are turned on continuously in dropout (100% duty cycle), and both channels draw a total quiescent current of only 40μA. In shutdown, the device draws <1μA. L, LT, LTC, LTM, Linear Technology, the Linear logo and Burst Mode are registered trademarks of Linear Technology Corporation. All other trademarks are the property of their respective owners. Protected by U.S. Patents, including 5481178, 6127815, 6304066, 6498466, 6580258 6611131.. TYPICAL APPLICATION Efficiency and Power Loss vs Load Current VIN 2.8V TO 5.5V 10μF 1000 100 RUN2 VIN MODE/SYNC RUN1 100k 95 POR RESET EFFICIENCY 68pF 887k 4.7μF 2.2μH SW2 SW1 VOUT1 VFB2 280k VOUT1 1.8V 800mA GND 85 POWER LOSS 10 80 75 70 10μF 35482 F01 Figure 1. 2.5V/1.8V at 400mA/800mA Step-Down Regulators 100 65 60 1 VIN = 3.6V 1 VOUT = 1.8V Burst Mode OPERATION CHANNEL 1 NO LOAD ON CHANNEL 2 0.1 10 100 1000 LOAD CURRENT (mA) 35482 F01b POWER LOSS (mW) 4.7μH VOUT2 2.5V 400mA EFFICIENCY (%) 90 LTC3548-2 35482fb 1 LTC3548-2 ABSOLUTE MAXIMUM RATINGS PIN CONFIGURATION (Note 1) TOP VIEW VIN Voltages .................................................– 0.3V to 6V VOUT1, VFB2 Voltages ........................ –0.3V to VIN + 0.3V RUN1, RUN2 Voltages .................................–0.3V to VIN MODE/SYNC Voltage........................ –0.3V to VIN + 0.3V SW1, SW2 Voltages ......................... –0.3V to VIN + 0.3V POR Voltage ................................................. –0.3V to 6V Ambient Operating Temperature Range (Note 2)....................................................– 40°C to 85°C Junction Temperature (Note 5) ............................. 125°C Storage Temperature Range...................– 65°C to 125°C VOUT1 1 RUN1 2 10 VFB2 9 RUN2 11 VIN 3 SW1 4 7 SW2 GND 5 6 MODE/ SYNC 8 POR DD PACKAGE 10-LEAD (3mm × 3mm) PLASTIC DFN TJMAX = 125°C, θJA = 40°C/W, θJC = 3°C/W (4-LAYER BOARD) EXPOSED PAD (PIN 11) IS GND, MUST BE SOLDERED TO PCB GND ORDER INFORMATION LEAD FREE FINISH TAPE AND REEL PART MARKING PACKAGE DESCRIPTION TEMPERATURE RANGE LTC3548EDD-2#PBF LTC3548EDD-2#TRPBF LCDK 10-Lead (3mm × 3mm) Plastic DFN –40°C to 85°C Consult LTC Marketing for parts specified with wider operating temperature ranges. Consult LTC Marketing for information on non-standard lead based finish parts. For more information on lead free part marking, go to: http://www.linear.com/leadfree/ For more information on tape and reel specifications, go to: http://www.linear.com/tapeandreel/ ELECTRICAL CHARACTERISTICS The l denotes the specifications which apply over the full operating temperature range, otherwise specifications are at TA = 25°C. VIN = 3.6V, unless otherwise specified. (Note 2) SYMBOL PARAMETER CONDITIONS MIN VIN Operating Voltage Range l IFB2 Feedback Pin Input Current l VFB2 Feedback Voltage of Channel 2 (Note 3) VOUT1 Output Voltage of Channel 1 (Note 3) ΔVLINEREG Reference Voltage Line Regulation ΔVLOADREG Output Voltage Load Regulation IS TYP 2.5 MAX UNITS 5.5 V 30 nA l 0.588 0.585 0.6 0.6 0.612 0.612 V V l 1.764 1.755 1.8 1.8 1.836 1.836 V V VIN = 2.5V to 5.5V (Note 3) 0.3 0.5 (Note 3) 0.5 Input DC Supply Current (Note 4) Active Mode Sleep Mode Shutdown VOUT1 = 1.5V, VFB2 = 0.5V VOUT1 = 1.9V, VFB2 = 0.63V, MODE/SYNC = 3.6V RUN = 0V, VIN = 5.5V, MODE/SYNC = 0V 700 40 0.1 950 60 1 μA μA μA fOSC Oscillator Frequency VOUT1 = 1.8V, VFB2 = 0.6V 2.25 2.7 MHz fSYNC Synchronization Frequency ILIM Peak Switch Current Limit Channel 1 Peak Switch Current Limit Channel 2 VIN = 3V, Duty Cycle <35% VIN = 3V, VFB2 = 0.5V, Duty Cycle <35% RDS(ON) Top Switch On-Resistance Bottom Switch On-Resistance ISW(LKG) Switch Leakage Current 0°C ≤ TA ≤ 85°C –40°C ≤ TA ≤ 85°C 0°C ≤ TA ≤ 85°C –40°C ≤ TA ≤ 85°C l 1.8 % 2.25 0.95 0.6 %/V MHz 1.2 0.7 1.6 0.9 A A (Note 6) (Note 6) 0.35 0.30 0.45 0.45 Ω Ω VIN = 5V, VRUN = 0V, VFB2 = 0V, VOUT1 = 0V 0.01 1 μA 35482fb 2 LTC3548-2 ELECTRICAL CHARACTERISTICS The l denotes the specifications which apply over the full operating temperature range, otherwise specifications are at TA = 25°C. VIN = 3.6V, unless otherwise specified. (Note 2) SYMBOL PARAMETER VMODE Mode Threshold Low Mode Threshold High POR Power-On Reset Threshold CONDITIONS MIN TYP 0 VIN – 0.5 VOUT1, VFB2 Ramping Down, MODE/SYNC = 0V UNITS 0 .5 VIN V V –8.5 Power-On Reset On-Resistance 100 Power-On Reset Delay % 200 262,144 VRUN RUN Threshold l IRUN RUN Leakage Current l Note 1: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. Exposure to any Absolute Maximum Rating condition for extended periods may affect device reliability and lifetime. Note 2: The LTC3548-2 is guaranteed to meet specified performance from 0°C to 85°C. Specifications over the – 40°C to 85°C operating temperature range are assured by design, characterization and correlation with statistical process controls. SW 5V/DIV VOUT 20mV/DIV 35482 G01 1 1.5 V 0.01 1 μA Load Step SW 5V/DIV IL 200mA/ DIV Cycles TA = 25°C unless otherwise specified. Pulse-Skipping Mode Burst Mode Operation 0.3 Ω Note 3: The LTC3548-2 is tested in a proprietary test mode that connects the output of the error amplifier to an outside servo loop. Note 4: Dynamic supply current is higher due to the internal gate charge being delivered at the switching frequency. Note 5: TJ is calculated from the ambient TA and power dissipation PD according to the following formula: TJ = TA + (PD • θJA). Note 6: The DFN switch on-resistance is guaranteed by correlation to wafer level measurements. TYPICAL PERFORMANCE CHARACTERISTICS VIN = 3.6V 2μs/DIV VOUT = 1.8V ILOAD = 60mA CHANNEL 1; CIRCUIT OF FIGURE 3 MAX VOUT 200mV/DIV IL 200mA/DIV IL 500mA/DIV VOUT 10mV/DIV ILOAD 500mA/DIV VIN = 3.6V 1μs/DIV VOUT = 1.8V ILOAD = 30mA CHANNEL 1; CIRCUIT OF FIGURE 3 35482 G02 VIN = 3.6V 20μs/DIV VOUT = 1.8V ILOAD = 80mA TO 800mA CHANNEL 1; CIRCUIT OF FIGURE 3 35482 G03 35482fb 3 LTC3548-2 TYPICAL PERFORMANCE CHARACTERISTICS TA = 25°C unless otherwise specified. Oscillator Frequency vs Temperature Efficiency vs Input Voltage 2.5 100 Oscillator Frequency vs Input Voltage 10 VIN = 3.6V 8 95 FREQUENCY DEVIATION (%) 2.4 100mA FREQUENCY (MHz) 90 1mA 80 800mA 75 70 60 2 3 4 2.2 2.1 VOUT = 1.8V, CHANNEL 1 Burst Mode OPERATION CIRCUIT OF FIGURE 3 65 2.3 5 6 INPUT VOLTAGE (V) 50 25 75 0 TEMPERATURE (°C) Reference Voltage vs Temperature –2 –4 –6 100 125 2 450 0.605 400 VIN = 2.7V VIN = 4.2V 450 0.595 RDS(ON) (mΩ) 300 250 0.585 –50 –25 200 100 MAIN SWITCH 350 0.590 SYNCHRONOUS SWITCH 2 3 Efficiency vs Load Current 95 1.5 1.0 85 80 PULSE-SKIPPING MODE 70 VOUT ERROR (%) Burst Mode OPERATION 4 VIN (V) 5 6 VIN = 3.6V, VOUT = 1.8V NO LOAD ON OTHER CHANNEL CHANNEL 1; CIRCUIT OF FIGURE 3 1 10 100 LOAD CURRENT (mA) 100 –50 –25 7 1000 35482 G10 MAIN SWITCH SYNCHRONOUS SWITCH 25 50 75 100 125 150 0 JUNCTION TEMPERATURE (°C) 3548 G09 Line Regulation 0.5 VOUT = 1.8V IOUT = 200mA 0.4 0.3 Burst Mode OPERATION 0.5 0 PULSE-SKIPPING MODE –0.5 –1.0 60 250 Load Regulation 2.0 65 300 35482 G08 100 75 350 200 35482 G07 90 VIN = 3.6V 400 150 1 125 6 RDS(ON) vs Junction Temperature 550 500 50 25 75 0 TEMPERATURE (°C) 5 35482 G06 VIN = 3.6V 0.600 4 3 VIN (V) RDS(ON) vs Input Voltage RDS(ON) (mΩ) REFERENCE VOLTAGE (V) 0 –10 500 0.610 EFFICIENCY (%) 2 35482 G05 35482 G04 0.615 4 –8 2.0 –50 –25 VOUT ERROR (%) EFFICIENCY (%) 10mA 85 6 VIN = 3.6V, VOUT = 1.8V NO LOAD ON OTHER CHANNEL CHANNEL 1; CIRCUIT OF FIGURE 3 –1.5 –2.0 1 10 100 LOAD CURRENT (mA) 1000 35482 G11 0.2 0.1 0 –0.1 –0.2 –0.3 –0.4 –0.5 2 3 4 VIN (V) 5 6 35482 G12 35482fb 4 LTC3548-2 TYPICAL PERFORMANCE CHARACTERISTICS Efficiency vs Load Current Efficiency vs Load Current Efficiency vs Load Current 100 100 95 95 95 90 90 VIN = 2.7V 85 VIN = 4.2V 80 75 70 VOUT2 = 1.5V Burst Mode OPERATION NO LOAD ON OTHER CHANNEL CIRCUIT OF FIGURE 3 65 60 1 10 100 LOAD CURRENT (mA) 1000 VIN = 4.2V 85 EFFICIENCY (%) 90 EFFICIENCY (%) VIN = 3.6V 100 VIN = 2.7V VIN = 3.6V EFFICIENCY (%) TA = 25°C unless otherwise specified. 80 75 70 VOUT2 = 2.5V Burst Mode OPERATION NO LOAD ON OTHER CHANNEL CIRCUIT OF FIGURE 3 65 60 1 10 100 LOAD CURRENT (mA) 35482 G13 1000 35482 G14 VIN = 5V 85 80 75 70 VOUT2 = 3.3 V Burst Mode OPERATION NO LOAD ON OTHER CHANNEL CIRCUIT OF FIGURE 3 65 60 1 10 100 LOAD CURRENT (mA) 1000 35482 G15 PIN FUNCTIONS VOUT1 (Pin 1): Output Voltage Feedback for Channel 1. An internal resistive divider divides the output voltage down for comparison to the internal reference voltage. RUN1 (Pin 2): Regulator 1 Enable. Forcing this pin to VIN enables regulator 1, while forcing it to GND causes regulator 1 to shut down. This pin must be driven; do not float. VIN (Pin 3): Input Power Supply. Must be closely decoupled to GND. SW1 (Pin 4): Regulator 1 Switch Node Connection to the Inductor. This pin swings from VIN to GND. GND (Pin 5): Ground. This pin is not connected internally. Connect to PCB ground for optimum shielding. MODE/SYNC (Pin 6): Combination Mode Selection and Oscillator Synchronization. This pin controls the operation of the device. When tied to VIN or GND, Burst Mode operation or pulse-skipping mode is selected, respectively. Do not float this pin. The oscillation frequency can be synchronized to an external oscillator applied to this pin and pulse-skipping mode is automatically selected. SW2 (Pin 7): Regulator 2 Switch Node Connection to the Inductor. This pin swings from VIN to GND. POR (Pin 8): Power-On Reset. This common-drain logic output is pulled to GND when the output voltage falls below –8.5% of regulation and goes high after 117ms when both channels are within regulation. RUN2 (Pin 9): Regulator 2 Enable. Forcing this pin to VIN enables regulator 2, while forcing it to GND causes regulator 2 to shut down. This pin must be driven; do not float. VFB2 (Pin 10): Output Feedback. Receives the feedback voltage from the external resistive divider across the output. Normal voltage for this pin is 0.6V. Exposed Pad (GND) (Pin 11): Ground. Connect to the (–) terminal of COUT, and (–) terminal of CIN. Must be connected to PCB ground for electrical contact and rated thermal performance. 35482fb 5 LTC3548-2 BLOCK DIAGRAM REGULATOR 1 MODE/SYNC 6 BURST CLAMP VIN SLOPE COMP 0.6V EA VOUT1 ITH1 BURST SLEEP – + 5Ω ICOMP + 0.35V – 1 EN – + R1 VFB1 S Q RS LATCH R Q R2 0.55V – UVDET SWITCHING LOGIC AND BLANKING CIRCUIT UV + ANTI SHOOTTHRU 4 SW1 + OVDET – + 0.65V OV IRCMP SHUTDOWN – 11 GND VIN 3 VIN PGOOD1 RUN1 2 RUN2 9 0.6V REF 8 POR OSC OSC POR COUNTER 5 GND PGOOD2 REGULATOR 2 (IDENTICAL TO REGULATOR 1) 0.6V 7 SW2 + EA ITH2 – VFB2 10 0.55V – UVDET UV + + OVDET 0.65V OV – 35482 BD 6 35482fb LTC3548-2 OPERATION The LTC3548-2 uses a constant-frequency, current mode architecture. The operating frequency is set at 2.25MHz and can be synchronized to an external oscillator. Both channels share the same clock and run in-phase. To suit a variety of applications, the selectable MODE pin allows the user to choose between low noise and high efficiency. The output voltage for channel 1 is set by an internal divider returned to the VOUT1 pin. The output voltage for channel 2 is set by an external divider returned to the VFB2 pin. An error amplifier compares the feedback output voltage with a reference voltage of 0.6V and adjusts the peak inductor current accordingly. An undervoltage comparator will pull the POR output low if the output voltage is not above –8.5% of the reference voltage. The POR output will go high after 262,144 clock cycles (about 117ms in pulseskipping mode) of achieving regulation. Main Control Loop During normal operation, the top power switch (P-channel MOSFET) is turned on at the beginning of a clock cycle when the VFB voltage (see the Block Diagram) is below the the reference voltage. The current into the inductor and the load increases until the current limit is reached. The switch turns off and energy stored in the inductor flows through the bottom switch (N-channel MOSFET) into the load until the next clock cycle. The peak inductor current is controlled by the internally compensated ITH voltage, which is the output of the error amplifier. This amplifier compares the VFB to the 0.6V reference. When the load current increases, the VFB voltage decreases slightly below the reference. This decrease causes the error amplifier to increase the ITH voltage until the average inductor current matches the new load current. The main control loop is shut down by pulling the RUN pin to ground. To optimize efficiency, the Burst Mode operation can be selected. When the load is relatively light, the LTC3548-2 automatically switches into Burst Mode operation in which the PMOS switch operates intermittently based on load demand with a fixed peak inductor current. By running cycles periodically, the switching losses which are dominated by the gate charge losses of the power MOSFETs are minimized. The main control loop is interrupted when the output voltage reaches the desired regulated value. A voltage comparator trips when ITH is below 0.35V, shutting off the switch and reducing the power. The output capacitor and the inductor supply the power to the load until ITH exceeds 0.65V, turning on the switch and the main control loop which starts another cycle. For lower ripple noise at low currents, the pulse-skipping mode can be used. In this mode, the LTC3548-2 continues to switch at a constant frequency down to very low currents, where it will begin skipping pulses. The efficiency in pulse-skipping mode can be improved slightly by connecting the SW node to the MODE/SYNC input which reduces the clock frequency by approximately 30%. Dropout Operation When the input supply voltage decreases toward the output voltage, the duty cycle increases to 100% which is the dropout condition. In dropout, the PMOS switch is turned on continuously with the output voltage being equal to the input voltage minus the voltage drops across the internal P-channel MOSFET and the inductor. An important design consideration is that the RDS(ON) of the P-channel switch increases with decreasing input supply voltage (see Typical Performance Characteristics). Therefore, the user should calculate the power dissipation when the LTC3548-2 is used at 100% duty cycle with low input voltage (see Thermal Considerations in the Applications Information section). Low Current Operation By selecting MODE/SYNC (Pin 6), two modes are available to control the operation of the LTC3548-2 at low currents. Both modes automatically switch from continuous operation to the selected mode when the load current is low. Low Supply Operation To prevent unstable operation, the LTC3548-2 incorporates an undervoltage lockout circuit that shuts down the part when the input voltage drops below about 1.65V. 35482fb 7 LTC3548-2 APPLICATIONS INFORMATION A general LTC3548-2 application circuit is shown in Figure 2. External component selection is driven by the load requirement, and begins with the selection of the inductor L. Once the inductor is chosen, CIN and COUT can be selected. VIN 2.5V TO 5.5V CIN RUN2 BM* PS* VIN MODE/SYNC RUN1 R5 POWER-ON RESET POR Inductor Core Selection LTC3548-2 L1 L2 VOUT2 SW2 SW1 C5 COUT2 R1 VOUT1 VOUT1 VFB2 R2 GND COUT1 35482 F02 *MODE/SYNC = 0V: PULSE SKIPPING MODE/SYNC = VIN: Burst Mode OPERATION Figure 2. LTC3548-2 General Schematic Inductor Selection Although the inductor does not influence the operating frequency, the inductor value has a direct effect on ripple current. The inductor ripple current ΔIL decreases with higher inductance and increases with higher VIN or VOUT: ΔIL = The inductor value will also have an effect on Burst Mode operation. The transition from low current operation begins when the peak inductor current falls below a level set by the burst clamp. Lower inductor values result in higher ripple current which causes this to occur at lower load currents. This causes a dip in efficiency in the upper range of low current operation. In Burst Mode operation, lower inductance values will cause the burst frequency to increase. VOUT ⎛ VOUT ⎞ • ⎜ 1– ⎟ fO • L ⎝ VIN ⎠ Accepting larger values of ΔIL allows the use of low inductances, but results in higher output voltage ripple, greater core losses, and lower output current capability. A reasonable starting point for setting ripple current is ΔIL = 0.3 • IOUT(MAX), where IOUT(MAX) is 800mA for channel 1 and 400mA for channel 2. The largest ripple current ΔIL occurs at the maximum input voltage. To guarantee that the ripple current stays below a specified maximum, the inductor value should be chosen according to the following equation: ⎛ ⎞ V V L ≥ OUT • ⎜ 1 – OUT ⎟ fO • Δ IL ⎝ VIN(MAX ) ⎠ Different core materials and shapes will change the size/ current and price/current relationship of an inductor. Toroid or shielded pot cores in ferrite or permalloy materials are small and do not radiate much energy, but generally cost more than powdered iron core inductors with similar electrical characteristics. The choice of which style inductor to use often depends more on the price vs size requirements and any radiated field/EMI requirements than on what the LTC3548-2 requires to operate. Table 1 shows some typical surface mount inductors that work well in LTC3548-2 applications. Table 1. Representative Surface Mount Inductors PART NUMBER VALUE (μH) DCR (Ω MAX) MAX DC CURRENT (A) SIZE W × L × H (mm3) Sumida CDRH3D16 2.2 3.3 4.7 0.075 0.110 0.162 1.20 1.10 0.90 3.8 × 3.8 × 1.8 Sumida CDRH2D11 1.5 2.2 0.068 0.170 0.900 0.780 3.2 × 3.2 × 1.2 Sumida CMD4D11 2.2 3.3 0.116 0.174 0.950 0.770 4.4 × 5.8 × 1.2 Murata LQH32CN 1.0 2.2 0.060 0.097 1.00 0.79 2.5 × 3.2 × 2.0 Toko D312F 2.2 3.3 0.060 0.260 1.08 0.92 2.5 × 3.2 × 2.0 Panasonic ELT5KT 3.3 4.7 0.17 0.20 1.00 0.95 4.5 × 5.4 × 1.2 Input Capacitor (CIN) Selection In continuous mode, the input current of the converter is a square wave with a duty cycle of approximately VOUT/VIN. To prevent large voltage transients, a low equivalent series resistance (ESR) input capacitor sized for the maximum 35482fb 8 LTC3548-2 APPLICATIONS INFORMATION RMS current must be used. The maximum RMS capacitor current is given by: IRMS ≈ IMAX VOUT ( VIN – VOUT ) VIN where the maximum average output current IMAX equals the peak current minus half the peak-to-peak ripple current, IMAX = ILIM – ΔIL/2. This formula has a maximum at VIN = 2VOUT, where IRMS = IOUT/2. This simple worst-case is commonly used to design because even significant deviations do not offer much relief. Note that capacitor manufacturer’s ripple current ratings are often based on only 2000 hours lifetime. This makes it advisable to further derate the capacitor, or choose a capacitor rated at a higher temperature than required. Several capacitors may also be paralleled to meet the size or height requirements of the design. An additional 0.1μF to 1μF ceramic capacitor is also recommended on VIN for high frequency decoupling, when not using an all ceramic capacitor solution. Output Capacitor (COUT) Selection The selection of COUT is driven by the required ESR to minimize voltage ripple and load step transients. Typically, once the ESR requirement is satisfied, the capacitance is adequate for filtering. The output ripple (ΔVOUT) is determined by: ⎛ ⎞ 1 ΔVOUT ≈ ΔIL ⎜ ESR + 8 • fO • COUT ⎟⎠ ⎝ where fO = operating frequency, COUT = output capacitance and ΔIL = ripple current in the inductor. The output ripple is highest at maximum input voltage since ΔIL increases with input voltage. With ΔIL = 0.3 • IOUT(MAX) the output ripple will be less than 100mV at maximum VIN and fO = 2.25MHz with: ESRCOUT < 150mΩ Once the ESR requirements for COUT have been met, the RMS current rating generally far exceeds the IRIPPLE(P-P) requirement, except for an all ceramic solution. In surface mount applications, multiple capacitors may have to be paralleled to meet the capacitance, ESR or RMS current handling requirement of the application. Aluminum electrolytic, special polymer, ceramic and dry tantalum capacitors are all available in surface mount packages. The OS-CON semiconductor dielectric capacitor available from Sanyo has the lowest ESR (size) product of any aluminum electrolytic at a somewhat higher price. Special polymer capacitors, such as Sanyo POSCAP, Panasonic Special Polymer (SP) and Kemet A700, offer very low ESR, but have a lower capacitance density than other types. Tantalum capacitors have the highest capacitance density, but they have a larger ESR and it is critical that the capacitors are surge tested for use in switching power supplies. An excellent choice is the AVX TPS series of surface mount tantalums, available in case heights ranging from 2mm to 4mm. Aluminum electrolytic capacitors have a significantly larger ESR, and are often used in extremely cost-sensitive applications provided that consideration is given to ripple current ratings and long term reliability. Ceramic capacitors have the lowest ESR and cost, but also have the lowest capacitance density, a high voltage and temperature coefficient, and exhibit audible piezoelectric effects. In addition, the high Q of ceramic capacitors along with trace inductance can lead to significant ringing. In most cases, 0.1μF to 1μF of ceramic capacitors should also be placed close to the LTC3548-2 in parallel with the main capacitors for high frequency decoupling. Ceramic Input and Output Capacitors Higher value, lower cost ceramic capacitors are now becoming available in smaller case sizes. These are tempting for switching regulator use because of their very low ESR. Unfortunately, the ESR is so low that it can cause loop stability problems. Solid tantalum capacitor ESR generates a loop zero at 5kHz to 50kHz that is instrumental in giving acceptable loop phase margin. Ceramic capacitors remain capacitive to beyond 300kHz and usually resonate with their ESL before ESR becomes effective. Also, ceramic caps are prone to temperature effects which requires the designer to check loop stability over the operating temperature range. To minimize their large temperature and voltage coefficients, only X5R or X7R ceramic capacitors should be used. A good selection of ceramic capacitors is available from Taiyo Yuden, AVX, Kemet, TDK and Murata. 35482fb 9 LTC3548-2 APPLICATIONS INFORMATION Great care must be taken when using only ceramic input and output capacitors. When a ceramic capacitor is used at the input and the power is being supplied through long wires, such as from a wall adapter, a load step at the output can induce ringing at the VIN pin. At best, this ringing can couple to the output and be mistaken as loop instability. At worst, the ringing at the input can be large enough to damage the part. Since the ESR of a ceramic capacitor is so low, the input and output capacitor must instead fulfill a charge storage requirement. During a load step, the output capacitor must instantaneously supply the current to support the load until the feedback loop raises the switch current enough to support the load. The time required for the feedback loop to respond is dependent on the compensation and the output capacitor size. Typically, 3 to 4 cycles are required to respond to a load step, but only in the first cycle does the output drop linearly. The output droop, VDROOP, is usually about 2 to 3 times the linear drop of the first cycle. Thus, a good place to start is with the output capacitor size of approximately: ΔIOUT COUT ≈ 2.5 fO • VDROOP More capacitance may be required depending on the duty cycle and load step requirements. In most applications, the input capacitor is merely required to supply high frequency bypassing, since the impedance to the supply is very low. A 10μF ceramic capacitor is usually enough for these conditions. Setting the Output Voltage for Channel 2 The LTC3548-2 develops a 0.6V reference voltage between the feedback pin, VFB2, and the ground as shown in Figure 2. The output voltage, VOUT2, is set by a resistive divider according to the following formula: ⎛ R2 ⎞ VOUT2 = 0.6 V ⎜ 1+ ⎟ ⎝ R1⎠ Keeping the current small (<5μA) in these resistors maximizes efficiency, but making them too small may allow stray capacitance to cause noise problems and reduce the phase margin of the error amp loop. To improve the frequency response, a feed forward capacitor CF may also be used. Great care should be taken to route the VOUT1, VFB2 line away from noise sources, such as the inductor or the SW line. Power-On Reset The POR pin is an open-drain output which pulls low when either regulator is out of regulation. When both output voltages are above –8.5% of regulation, a timer is started which releases POR after 218 clock cycles (about 117ms in pulse-skipping mode). This delay can be significantly longer in Burst Mode operation with low load currents, since the clock cycles only occur during a burst and there could be milliseconds of time between bursts. This can be bypassed by tying the POR output to the MODE/SYNC input, to force pulse-skipping mode during a reset. In addition, if the output voltage faults during Burst Mode sleep, POR could have a slight delay for an undervoltage output condition. This can be avoided by using pulse-skipping mode instead. When either channel is shut down, the POR output is pulled low, since one or both of the channels are not in regulation. Mode Selection and Frequency Synchronization The MODE/SYNC pin is a multipurpose pin which provides mode selection and frequency synchronization. Connecting this pin to VIN enables Burst Mode operation, which provides the best low current efficiency at the cost of a higher output voltage ripple. Connecting this pin to ground selects pulse-skipping mode, which provides the lowest output ripple, at the cost of low current efficiency. The LTC3548-2 can also be synchronized to an external 2.25MHz clock signal by the MODE/SYNC pin. During synchronization, the mode is set to pulse-skipping and the top switch turn-on is synchronized to the rising edge of the external clock. 35482fb 10 LTC3548-2 APPLICATIONS INFORMATION Checking Transient Response Efficiency Considerations The regulator loop response can be checked by looking at the load transient response. Switching regulators take several cycles to respond to a step in load current. When a load step occurs, VOUT immediately shifts by an amount equal to ΔILOAD • ESR, where ESR is the effective series resistance of COUT. ΔILOAD also begins to charge or discharge COUT, generating a feedback error signal used by the regulator to return VOUT to its steady-state value. During this recovery time, VOUT can be monitored for overshoot or ringing that would indicate a stability problem. The percent efficiency of a switching regulator is equal to the output power divided by the input power times 100%. It is often useful to analyze individual losses to determine what is limiting the efficiency and which change would produce the most improvement. Percent efficiency can be expressed as: The initial output voltage step may not be within the bandwidth of the feedback loop, so the standard secondorder overshoot/DC ratio cannot be used to determine phase margin. In addition, a feed forward capacitor, CF, can be added to improve the high frequency response, as shown in Figure 2. Capacitor CF provides phase lead by creating a high frequency zero with R2, which improves the phase margin. The output voltage settling behavior is related to the stability of the closed-loop system and will demonstrate the actual overall supply performance. For a detailed explanation of optimizing the compensation components, including a review of control loop theory, refer to Application Note 76. In some applications, a more severe transient can be caused by switching loads with large (>1μF) load input capacitors. The discharged load input capacitors are effectively put in parallel with COUT, causing a rapid drop in VOUT. No regulator can deliver enough current to prevent this problem, if the switch connecting the load has low resistance and is driven quickly. The solution is to limit the turn-on speed of the load switch driver. A Hot Swap™ controller is designed specifically for this purpose and usually incorporates current limiting, short-circuit protection, and soft-starting. Hot Swap is a trademark of Linear Technology Corporation. % Efficiency = 100% – (L1 + L2 + L3 + ...) where L1, L2, etc. are the individual losses as a percentage of input power. Although all dissipative elements in the circuit produce losses, four main sources usually account for most of the losses in LTC3548-2 circuits: 1) VIN quiescent current, 2) switching losses, 3) I2R losses, 4) other losses. 1. The VIN current is the DC supply current given in the Electrical Characteristics which excludes MOSFET driver and control currents. VIN current results in a small (<0.1%) loss that increases with VIN, even at no load. 2. The switching current is the sum of the MOSFET driver and control currents. The MOSFET driver current results from switching the gate capacitance of the power MOSFETs. Each time a MOSFET gate is switched from low to high to low again, a packet of charge dQ moves from VIN to ground. The resulting dQ/dt is a current out of VIN that is typically much larger than the DC bias current. In continuous mode, IGATECHG = fO(QT + QB), where QT and QB are the gate charges of the internal top and bottom MOSFET switches. The gate charge losses are proportional to VIN and thus their effects will be more pronounced at higher supply voltages. 3. I2R losses are calculated from the DC resistances of the internal switches, RSW, and external inductor, RL. In continuous mode, the average output current flows through inductor L, but is chopped between the internal top and bottom switches. Thus, the series resistance looking into the SW pin is a function of both top and bottom MOSFET RDS(ON) and the duty cycle (D) as follows: RSW = (RDS(ON)TOP)(D) + (RDS(ON)BOT)(1 – D) 35482fb 11 LTC3548-2 APPLICATIONS INFORMATION The RDS(ON) for both the top and bottom MOSFETs can be obtained from the Typical Performance Characteristics curves. Thus, to obtain I2R losses: I2R losses = IOUT 2(RSW + RL) 4. Other hidden losses such as copper trace and internal battery resistances can account for additional efficiency degradations in portable systems. It is very important to include these system level losses in the design of a system. The internal battery and fuse resistance losses can be minimized by making sure that CIN has adequate charge storage and very low ESR at the switching frequency. Other losses including diode conduction losses during dead-time and inductor core losses generally account for less than 2% total additional loss. Thermal Considerations In a majority of applications, the LTC3548-2 does not dissipate much heat due to its high efficiency. However, in applications where the LTC3548-2 is running at high ambient temperature with low supply voltage and high duty cycles, such as in dropout, the heat dissipated may exceed the maximum junction temperature of the part. If the junction temperature reaches approximately 150°C, both power switches will turn off and the SW node will become high impedance. To prevent the LTC3548-2 from exceeding the maximum junction temperature, the user will need to do some thermal analysis. The goal of the thermal analysis is to determine whether the power dissipated exceeds the maximum junction temperature of the part. The temperature rise is given by: TRISE = PD • θJA where PD is the power dissipated by the regulator and θJA is the thermal resistance from the junction of the die to the ambient temperature. The junction temperature, TJ, is given by: TJ = TRISE + TAMBIENT As an example, consider the case when the LTC3548-2 is at an input voltage of 2.7V with a load current of 400mA and 800mA and an ambient temperature of 70°C. From the Typical Performance Characteristics graph of Switch Resistance, the RDS(ON) resistance of the main switch is 0.425Ω. Therefore, power dissipated by each channel is: PD = I2 • RDS(ON) = 272mW and 68mW The DFN package junction-to-ambient thermal resistance, θJA, is 40°C/W. Therefore, the junction temperature of the regulator operating in a 70°C ambient temperature is approximately: TJ = (0.272 + 0.068) • 40 + 70 = 83.6°C which is below the absolute maximum junction temperature of 125°C. Design Example As a design example, consider using the LTC3548-2 in an portable application with a Li-Ion battery. The battery provides a VIN = 2.8V to 4.2V. The load requires a maximum of 800mA in active mode and 2mA in standby mode. The output voltage is VOUT = 1.8V. Since the load still needs power in standby, Burst Mode operation is selected for good low load efficiency. First, calculate the inductor value for about 30% ripple current at maximum VIN: L≥ ⎛ 1.8 V ⎞ 1.8 V • ⎜ 1– = 1.9μH 2.25MHz • 240mA ⎝ 4.2V ⎟⎠ Choosing a vendor’s closest inductor value of 2.2μH, results in a maximum ripple current of: ΔIL = 1.8 V ⎛ 1.8 V ⎞ • ⎜ 1− = 208mA 2.25MHz • 2.2μ ⎝ 4.2V ⎟⎠ For cost reasons, a ceramic capacitor will be used. COUT selection is then based on load step droop instead of ESR requirements. For a 5% output droop: COUT ≈ 2.5 800mA = 9.9μF 2.25MHz •(5% • 1.8 V) A good standard value is 10μF. Since the output impedance of a Li-Ion battery is very low, CIN is typically 10μF. Following the same procedure for VOUT2 = 2.5V, the inductor value is derived as 4.7μH and the output capacitor value is 4.7μF. 35482fb 12 LTC3548-2 APPLICATIONS INFORMATION The output voltage, VOUT2, can now be programmed by choosing the values of R1 and R2. To maintain high efficiency, the current in these resistors should be kept small. Choosing 2μA with the 0.6V feedback voltage makes R1 approximately 280k. A close standard 1% resistor is 280k, and R2 is then 887k. The POR pin is a common drain output and requires a pullup resistor. A 100k resistor is used for adequate speed. Figure 3 shows the complete schematic for this design example. Board Layout Considerations When laying out the printed circuit board, the following checklist should be used to ensure proper operation of the LTC3548-2. These items are also illustrated graphically in the layout diagram of Figure 4. Check the following in your layout: 1. Does the capacitor CIN connect to the power VIN (Pin 3) and GND (Exposed Pad) as close as possible? This capacitor provides the AC current to the internal power MOSFETs and their drivers. VIN 2.5V* TO 5.5V RUN2 VIN L2 4.7μH RUN1 4. Keep sensitive components away from the SW pins. The input capacitor CIN and the resistors R1 to R2 should be routed away from the SW traces and the inductors. 5. A ground plane is preferred, but if not available, keep the signal and power grounds segregated with small signal components returning to the GND pin at one point and should not share the high current path of CIN or COUT. 6. Flood all unused areas on all layers with copper. Flooding with copper will reduce the temperature rise of power components. These copper areas should be connected to VIN or GND. R5 100k RUN2 POWER-ON RESET POR L1 2.2μH LTC3548-2 SW1 SW2 C5 68pF C3 4.7μF 3. The resistor divider, R1 and R2, must be connected between the (+) plate of COUT and a ground sense line terminated near GND (Exposed Pad). The feedback signals VOUT1, VFB2 should be routed away from noisy components and traces, such as the SW line (Pins 4 and 7), and its trace should be minimized. VIN C1 10μF MODE/SYNC VOUT2 2.5V* 400mA 2. Are the COUT and L1 closely connected? The (–) plate of COUT returns current to GND and the (–) plate of CIN. R2 887k R1 280k CIN GND RUN1 POR LTC3548-2 VOUT1 1.8V 800mA VOUT2 L1 L2 SW2 C2 10μF VOUT1 SW1 C5 VOUT1 VFB2 VIN MODE/SYNC COUT2 VOUT1 VFB2 R2 GND R1 COUT1 35482 F03 C1, C2, C3: TAIYO YUDEN JMK212BJ106MG C3: TAIYO YUDEN JMK212BJ475MG L1: MURATA LQH32CN2R2M11 L2: MURATA LQH32CN4R7M23 *VOUT CONNECTED TO VIN FOR VIN b 2.8V (DROPOUT) Figure 3. LTC3548-2 Typical Application 35482 F04 BOLD LINES INDICATE HIGH CURRENT PATHS Figure 4. LTC3548-2 Layout Diagram (See Board Layout Checklist) 35482fb 13 LTC3548-2 PACKAGE DESCRIPTION Low Ripple Buck Regulators Using Ceramic Capacitors 100 C1 10μF RUN2 VIN RUN1 R5 100k POR LTC3548-2 L2 10μH SW2 L1 4.7μH SW1 C5 68pF C3 10μF R2 887k R1 280k VOUT1 1.8V 800mA 90 VOUT1 VFB2 MODE/SYNC 85 VOUT1 = 1.8V 80 75 70 C2 10μF GND VOUT2 = 2.5V 95 POWER-ON RESET EFFICIENCY (%) VIN 2.5V TO 5.5V VOUT2 2.5V 400mA Efficiency vs Load Current 65 VIN = 3.3V PULSE-SKIPPING MODE 35482 TA01 60 C1, C2, C3: TDK C2012X5R0J106M L1: SUMIDA CDRH2D18/HP-4R7NC L2: SUMIDA CDRH2D18/HP-100NC 10 1mm Profile Core and I/O Supplies 100 C1* 10μF RUN2 VIN MODE/SYNC RUN1 SW2 R5 100k L1 2.2μH SW1 C5 68pF 95 POWER-ON RESET POR LTC3548-2 L2 4.7μH VOUT2 3.3V 400mA 1000 35482 TA01b Efficiency vs Load Current VOUT1 1.8V 800mA VOUT2 = 3.3V 90 EFFICIENCY (%) VIN 3.6V TO 5.5V 100 LOAD CURRENT (mA) VOUT1 = 1.8V 85 80 75 70 C3 4.7μF R2 887k VOUT1 VFB2 R1 196k GND C2 10μF 35482 TA02a C1, C2: MURATA GRM219R60J106KE19 C3: MURATA GRM219R60J475KE19 L1: COILTRONICS LPO3310-222MX L2: COILTRONICS LPO3310-472MX *IF C1 IS GREATER THAN 3" FROM POWER SOURCE, ADDITIONAL CAPACITANCE MAY BE REQUIRED. 65 VIN = 5V Burst Mode OPERATION 60 1 10 100 LOAD CURRENT (mA) 1000 35482 TA02b 35482fb 14 LTC3548-2 PACKAGE DESCRIPTION DD Package 10-Lead Plastic DFN (3mm × 3mm) (Reference LTC DWG # 05-08-1699 Rev B) 0.70 p0.05 3.55 p0.05 1.65 p0.05 2.15 p0.05 (2 SIDES) PACKAGE OUTLINE 0.25 p 0.05 0.50 BSC 2.38 p0.05 (2 SIDES) RECOMMENDED SOLDER PAD PITCH AND DIMENSIONS 3.00 p0.10 (4 SIDES) R = 0.125 TYP 6 0.40 p 0.10 10 1.65 p 0.10 (2 SIDES) PIN 1 TOP MARK (SEE NOTE 6) (DD) DFN REV B 0309 5 0.200 REF 1 0.25 p 0.05 0.50 BSC 0.75 p0.05 0.00 – 0.05 2.38 p0.10 (2 SIDES) BOTTOM VIEW—EXPOSED PAD NOTE: 1. DRAWING TO BE MADE A JEDEC PACKAGE OUTLINE M0-229 VARIATION OF (WEED-2). CHECK THE LTC WEBSITE DATA SHEET FOR CURRENT STATUS OF VARIATION ASSIGNMENT 2. DRAWING NOT TO SCALE 3. ALL DIMENSIONS ARE IN MILLIMETERS 4. DIMENSIONS OF EXPOSED PAD ON BOTTOM OF PACKAGE DO NOT INCLUDE MOLD FLASH. MOLD FLASH, IF PRESENT, SHALL NOT EXCEED 0.15mm ON ANY SIDE 5. EXPOSED PAD SHALL BE SOLDER PLATED 6. SHADED AREA IS ONLY A REFERENCE FOR PIN 1 LOCATION ON THE TOP AND BOTTOM OF PACKAGE 35482fb Information furnished by Linear Technology Corporation is believed to be accurate and reliable. However, no responsibility is assumed for its use. Linear Technology Corporation makes no representation that the interconnection of its circuits as described herein will not infringe on existing patent rights. 15 LTC3548-2 TYPICAL APPLICATION 2mm Height Lithium-Ion Single Inductor Buck-Boost Regulator and a Buck Regulator VIN 2.8V TO 4.2V C1 10μF RUN2 VIN MODE/SYNC L2 15μH D1 VOUT2 3.3V 100mA POWER-ON RESET POR L1 2.2μH LTC3548-2 SW2 VOUT1 1.8V 800mA SW1 C5 22pF M1 + R5 100k RUN1 C6 22μF R4 887k C3 4.7μF VOUT1 VFB2 R3 196k GND C1, C2: TAIYO YUDEN JMK316BJ106ML C3: MURATA GRM21BR60J475KA11B C6: KEMET C1206C226K9PAC D1: PHILIPS PMEG2010 L1: MURATA LQH32CN2R2M33 L2: TOKO A914BYW-150M (D52LC SERIES) M1: SILICONIX Si2302DS C2 10μF 35482 TA03a Efficiency vs Load Current Efficiency vs Load Current 90 100 95 80 2.8V 4.2V 3.6V 2.8V 70 60 50 EFFICIENCY (%) EFFICIENCY (%) 90 85 4.2V 3.6V 80 75 70 40 VOUT = 3.3V Burst Mode OPERATION NO LOAD ON OTHER CHANNEL 30 1 10 100 LOAD CURRENT (mA) VOUT = 1.8V Burst Mode OPERATION NO LOAD ON OTHER CHANNEL 65 1000 60 35482 TA03b 1 10 100 LOAD CURRENT (mA) 1000 3548 TA03c RELATED PARTS PART NUMBER DESCRIPTION COMMENTS LTC1878 600mA (IOUT), 550kHz, Synchronous Step-Down DC/DC Converter 95% Efficiency, VIN: 2.7V to 6V, VOUT(MIN) = 0.8V, IQ = 10μA, ISD < 1μA, MSOP-8 Package LT1940 Dual Output 1.4A (IOUT), Constant 1.1MHz, High Efficiency Step-Down DC/DC Converter VIN: 3V to 25V, VOUT(MIN) = 1.2V, IQ = 2.5mA, ISD = < 1μA, TSSOP-16E Package LTC3405/LTC3405A 300mA (IOUT), 1.5MHz, Synchronous Step-Down DC/DC Converters 96% Efficiency, VIN: 2.5V to 5.5V, VOUT(MIN) = 0.8V, IQ = 20μA, ISD < 1μA, ThinSOT Package LTC3406/LTC3406B 600mA (IOUT), 1.5MHz, Synchronous Step-Down DC/DC Converters 96% Efficiency, VIN: 2.5V to 5.5V, VOUT(MIN) = 0.6V, IQ = 20μA, ISD < 1μA, ThinSOT Package LT3407/LT3407-2 600mA/1.5MHz, 800mA/2.25MHz Dual Synchronous Step-Down DC/DC Converter 96% Efficiency, VIN: 2.5V to 5.5V, VOUT(MIN) = 0.6V, IQ = 40μA, ISD < 1μA, MSE, DFN Package LTC3410/LTC3410B 300mA, 2.25MHz Synchronous Step-Down DC/DC Converters 95% Efficiency, VIN: 2.5V to 5.5V, VOUT(MIN) = 0.8V, IQ = 26μA, ISD < 1μA, SC70 Package LTC3411 1.25A (IOUT), 4MHz, Synchronous Step-Down DC/DC Converter 95% Efficiency, VIN: 2.5V to 5.5V, VOUT(MIN) = 0.8V, IQ = 60μA, ISD < 1μA, MSOP-10 Package LTC3412/LTC3412A 2.5A (IOUT), 4MHz, Synchronous Step-Down DC/DC Converter 95% Efficiency, VIN: 2.5V to 5.5V, VOUT(MIN) = 0.8V, IQ = 60μA, ISD < 1μA, TSSOP-16E Package LTC3414 4A (IOUT), 4MHz, Synchronous Step-Down DC/DC Converter 95% Efficiency, VIN: 2.25V to 5.5V, VOUT(MIN) = 0.8V, IQ = 64μA, ISD < 1μA, TSSOP-28E Package LTC3548/LTC3548-1 800mA/400mA (IOUT), 2.25MHz, Dual Synchronous Step-Down DC/DC Converter 95% Efficiency, VIN: 2.5V to 5.5V, VOUT(MIN) = 0.6V, IQ = 40μA, ISD < 1μA, MSE, DFN-10 Packages 35482fb 16 Linear Technology Corporation LT 0709 REV B • PRINTED IN USA 1630 McCarthy Blvd., Milpitas, CA 95035-7417 (408) 432-1900 ● FAX: (408) 434-0507 ● www.linear.com © LINEAR TECHNOLOGY CORPORATION 2006