LTC1871X Wide Input Range, No RSENSE™ Current Mode Boost, Flyback and SEPIC Controller FEATURES DESCRIPTION High Efficiency (No Sense Resistor Required) nn Wide Input Voltage Range: 2.5V to 36V nn Current Mode Control Provides Excellent Transient Response nn High Maximum Duty Cycle (92% Typ) nn ±2% RUN Pin Threshold with 100mV Hysteresis nn ±2% Internal Voltage Reference nn Micropower Shutdown: I = 10μA Q nn Programmable Operating Frequency (65kHz to 900kHz) with One External Resistor nn Synchronizable to an External Clock Up to 1.3 × f OSC nn User-Controlled Pulse Skip or Burst Mode® Operation nn Internal 5.2V Low Dropout Voltage Regulator nn Output Overvoltage Protection nn Capable of Operating with a Sense Resistor for High Output Voltage Applications nn Small 10-Lead MSOP Package LTC®1871X is a wide input range, current mode, boost, flyback or SEPIC controller that drives an N-channel power MOSFET. LTC1871X is rated to 175°C junction temperature and is 100% tested at 175°C. Intended for low to medium power applications, it eliminates the need for a current sense resistor by utilizing the power MOSFET’s on-resistance, thereby maximizing efficiency. nn APPLICATIONS nn nn Telecom Power Supplies Portable Electronic Equipment The IC’s operating frequency can be set with an external resistor over a 65kHz to 900kHz range, and can be synchronized to an external clock using the MODE/SYNC pin. Burst Mode operation at light loads, a low minimum operating supply voltage of 2.5V and a low shutdown quiescent current of 10μA make the LTC1871X ideally suited for battery-operated systems. For applications requiring constant frequency operation, Burst Mode operation can be defeated using the MODE/ SYNC pin. Higher output voltage boost, SEPIC and flyback applications are possible with the LTC1871X by connecting the SENSE pin to a resistor in the source of the power MOSFET. The LTC1871X is available in the 10-lead MSOP package. L, LT, LTC, LTM, Linear Technology, the Linear logo and Burst Mode are registered trademarks and No RSENSE is a trademark of Linear Technology Corporation. All other trademarks are the property of their respective owners. TYPICAL APPLICATION FB Voltage vs Temperature 1.250 100µF 6.3V RUN ITH MODE/SYNC 4.7µF 10nF 80.6k 1.240 SENSE LTC1871X V INTVCC 33.2k 1.245 10µH IN GATE 110k FB FREQ GND 12.4k 1871x F01 VOUT 12V 2A 47µF 25V ×8 GND Figure 1. High Efficiency 5V Input, 12V Output Boost Converter (Bootstrapped) FB VOLTAGE (V) VIN 5V 1.235 1.230 1.225 1.220 1.215 1.210 –50 –25 0 25 50 75 100 125 150 175 200 TEMPERATURE (°C) 1871x F01b 1871xf For more information www.linear.com/LTC1871X 1 LTC1871X ABSOLUTE MAXIMUM RATINGS PIN CONFIGURATION (Note 1) VIN Voltage................................................. – 0.3V to 36V INTVCC Voltage............................................. –0.3V to 7V INTVCC Output Current...........................................50mA GATE Voltage............................. –0.3V to VINTVCC + 0.3V ITH, FB Voltages......................................... –0.3V to 2.7V RUN, MODE/SYNC Voltages......................... –0.3V to 7V FREQ Voltage............................................. –0.3V to 1.5V SENSE Pin Voltage...................................... –0.3V to 36V Operating Junction Temperature Range (Notes 2, 3) LTC1871X................................................ –40°C to 175°C Storage Temperature Range................... –65°C to 175°C Lead Temperature (Soldering, 10 sec).................... 300°C ORDER INFORMATION TOP VIEW RUN ITH FB FREQ MODE/SYNC 1 2 3 4 5 10 9 8 7 6 SENSE VIN INTVCC GATE GND MS PACKAGE 10-LEAD PLASTIC MSOP TJMAX = 175°C, θJA = 120°C/W http://www.linear.com/product/LTC1871X#orderinfo LEAD FREE FINISH TAPE AND REEL PART MARKING PACKAGE DESCRIPTION TEMPERATURE RANGE LTC1871XMS#PBF LTC1871XMS#TRPBF LTGZM 10-Lead Plastic MSOP –40°Cto 175°C Consult LTC Marketing for parts specified with wider operating temperature ranges. 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/. Some packages are available in 500 unit reels through designated sales channels with #TRMPBF suffix. 1871xf 2 For more information www.linear.com/LTC1871X LTC1871X ELECTRICAL CHARACTERISTICS The l denotes the specifications which apply over the specified operating junction temperature range, otherwise specifications are at TA = 25°C (Note 2). VIN = VINTVCC = 5V, VRUN = 1.5V, RFREQ = 80k, VMODE/SYNC = 0V, unless otherwise specified. SYMBOL PARAMETER CONDITIONS MIN TYP MAX UNITS Main Control Loop VIN(MIN) Minimum Input Voltage (Note 2) IQ Input Voltage Supply Current (Note 4) Continuous Mode VMODE/SYNC = 5V, VFB = 1.4V, VITH = 0.75V VMODE/SYNC = 5V, VFB = 1.4V, VITH = 0.75V, (Note 2) l 2.5 V 2.5 V l 550 1000 μA 550 1200 μA μA Burst Mode Operation, No Load VMODE/SYNC = 0V, VITH = 0.2V (Note 5) 250 500 VMODE/SYNC = 0V, VITH = 0.2V (Note 5), (Note 2) l 250 650 μA Shutdown Mode VRUN = 0V 10 20 μA 10 65 μA VRUN = 0V, (Note 2) VRUN+ Rising RUN Input Threshold Voltage VRUN – Falling RUN Input Threshold Voltage VRUN(HYST) RUN Pin Input Threshold Hysteresis l 1.348 1.223 (Note 2) l 1.179 50 (Note 2) l 1.248 V 1.273 1.465 100 V V 150 mV 400 mV 1 60 nA 1.230 1.242 V 35 IRUN RUN Input Current VFB Feedback Voltage IFB FB Pin Input Current VITH = 0.2V (Note 5) ∆VFB ∆VIN Line Regulation 3.5V ≤ VIN ≤ 30V ∆VFB ∆VITH Load Regulation ∆VFB(OV) ∆FB Pin,Overvoltage Lockout VFB(OV) – VFB(NOM) in Percent gm Error Amplifier Transconductance ITH Pin Load = ±5μA (Note 5) 650 μmho VITH(BURST) Burst Mode Operation ITH Pin Voltage Falling ITH Voltage (Note 5) 0.3 V VSENSE(MAX) Maximum Current Sense Input Threshold VITH = 0.2V (Note 5) VITH = 0.2V (Note 5), (Note 2) 3.5V ≤ VIN ≤ 30V, (Note 2) 1.218 l l VMODE/SYNC = 0V, VITH = 0.5V to 0.9V (Note 5) VMODE/SYNC = 0V, VITH = 0.5V to 0.9V (Note 5) (Note 2) l Duty Cycle < 20% Duty Cycle < 20%, (Note 2) 1.205 V 60 nA 0.002 0.02 %/V 0.002 0.02 %/V –1 –0.1 % –1 –0.1 % 2.5 6 115 l 1.255 18 150 100 10 % 185 mV 200 mV ISENSE(ON) SENSE Pin Current (GATE High) VSENSE = 0V 35 50 μA ISENSE(OFF) SENSE Pin Current (GATE Low) VSENSE = 30V 0.1 5 μA Oscillator Frequency RFREQ = 80k 250 300 350 kHz 240 300 Oscillator fOSC RFREQ = 80k, (Note 2) l Oscillator Frequency Range 50 (Note 2) DMAX l Maximum Duty Cycle (Note 2) l 65 375 kHz 1000 kHz 900 kHz 87 92 97 % 87 92 97 % 1871xf For more information www.linear.com/LTC1871X 3 LTC1871X ELECTRICAL CHARACTERISTICS The l denotes the specifications which apply over the specified operating junction temperature range, otherwise specifications are at TA = 25°C (Note 2). VIN = VINTVCC = 5V, VRUN = 1.5V, RFREQ = 80k, VMODE/SYNC = 0V, unless otherwise specified. SYMBOL PARAMETER CONDITIONS fSYNC/fOSC Recommended Maximum Synchronized Frequency Ratio fOSC = 300kHz (Note 6) tSYNC(MIN) MODE/SYNC Minimum Input Pulse Width VSYNC = 0V to 5V 25 tSYNC(MAX) MODE/SYNC Maximum Input Pulse Width VSYNC = 0V to 5V 0.8/fOSC VIL(MODE) Low Level MODE/SYNC Input Voltage VIH(MODE) MIN fOSC = 300kHz (Note 6), (Note 2) l (Note 2) l (Note 2) l High Level MODE/SYNC Input Voltage RMODE/SYNC MODE/SYNC Input Pull-Down Resistance VFREQ Nominal FREQ Pin Voltage TYP MAX 1.25 1.30 1.25 1.30 UNITS ns ns 0.3 V 0.3 V 1.2 V 1.2 V 50 kΩ 0.62 V Low Dropout Regulator VINTVCC INTVCC Regulator Output Voltage VIN = 7.5V 5.2 5.4 V ∆VINTVCC ∆VIN1 INTVCC Regulator Line Regulation 7.5V ≤ VIN ≤ 15V 5.0 8 25 mV ∆VINTVCC ∆VIN2 INTVCC Regulator Line Regulation 15V ≤ VIN ≤ 30V 70 200 mV VLDO(LOAD) INTVCC Load Regulation 0 ≤ IINTVCC ≤ 20mA, VIN = 7.5V VDROPOUT INTVCC Regulator Dropout Voltage IINTVCC –2 –0.2 % VIN = 5V, INTVCC Load = 20mA 280 mV Bootstrap Mode INTVCC Supply Current in Shutdown RUN = 0V, SENSE = 5V 10 20 μA tr GATE Driver Output Rise Time CL = 3300pF (Note 7) 17 100 ns tf GATE Driver Output Fall Time CL = 3300pF (Note 7) 8 100 ns GATE Driver 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 LTC1871X is guaranteed over the full –40°C to 175°C operating junction temperature range. High junction temperatures degrade operating lifetimes. Operating lifetime is derated at junction temperatures greater than 125°C. Note 3: TJ is calculated from the ambient temperature TA and power dissipation PD according to the following formula: TJ = TA + (PD • 120°C/W) Note 4: The dynamic input supply current is higher due to power MOSFET gate charging (QG • fOSC). See Applications Information. Note 5: The LTC1871X is tested in a feedback loop which servos VFB to the reference voltage with the ITH pin forced to the midpoint of its voltage range (0.3V ≤ VITH ≤ 1.2V, midpoint = 0.75V). Note 6: In a synchronized application, the internal slope compensation gain is increased by 25%. Synchronizing to a significantly higher ratio will reduce the effective amount of slope compensation, which could result in subharmonic oscillation for duty cycles greater than 50%. Note 7: Rise and fall times are measured at 10% and 90% levels. 1871xf 4 For more information www.linear.com/LTC1871X LTC1871X TYPICAL PERFORMANCE CHARACTERISTICS FB Voltage vs Temperature FB Voltage Line Regulation 1.250 FB Pin Current vs Temperature 150 1.231 135 1.245 120 1.235 1.230 1.225 FB PIN CURRENT (nA) FB VOLTAGE (V) FB VOLTAGE (V) 1.240 1.230 1.220 105 90 75 60 45 30 1.215 15 1.210 –50 –25 0 1.229 25 50 75 100 125 150 175 200 TEMPERATURE (°C) 5 0 1871x G01 10 15 20 VIN (V) 25 30 0 –50 –25 0 35 25 50 75 100 125 150 175 200 TEMPERATURE (°C) 1871x G03 1871x G02 Shutdown Mode IQ vs Temperature Shutdown Mode IQ vs VIN 10 30 20 VIN (V) 28 24 20 16 12 8 400 300 200 100 4 0 –50 –25 0 40 Burst Mode IQ vs VIN 500 32 0 25 50 75 100 125 150 175 200 TEMPERATURE (°C) 0 10 20 VIN (V) 1871x G05 18 Dynamic IQ vs Frequency 60 CL = 3300pF IQ(TOT) = 550µA + Qg • f 16 40 Gate Drive Rise and Fall Time vs CL 50 14 12 40 10 8 6 RISE TIME 30 20 FALL TIME 4 10 2 0 30 1871x G06 1871x G04 TIME (ns) 0 36 Burst Mode IQ (µA) SHUTDOWN MODE IQ CURRENT (µA) 10 IQ (mA) SHUTDOWN MODE IQ (µA) 20 0 600 40 30 0 200 400 600 800 FREQUENCY (kHz) 1000 1200 0 0 2000 1871x G07 4000 6000 8000 CL (pF) 10000 12000 1871x G08 1871xf For more information www.linear.com/LTC1871X 5 LTC1871X TYPICAL PERFORMANCE CHARACTERISTICS RUN Thresholds vs VIN RUN Thresholds vs Temperature RT vs Frequency 1.60 1.5 1000 1.4 1.3 1.50 1.45 RT (kΩ) RUN THRESHOLDS (V) RUN THRESHOLDS (V) 1.55 1.40 1.35 100 1.30 1.25 1.2 0 10 30 20 VIN (V) 1.20 –50 –25 0 40 10 25 50 75 100 125 150 175 200 TEMPERATURE (°C) 1871x G10 0 100 200 300 400 500 600 700 800 900 1000 FREQUENCY (kHz) 1871x G11 1871x G09 Maximum Sense Threshold vs Temperature Frequency vs Temperature 330 SENSE Pin Current vs Temperature 160 35 325 MAX SENSE THRESHOLD (mV) GATE FREQUENCY (kHz) 315 310 305 300 295 290 285 SENSE PIN CURRENT (µA) 34 320 155 150 145 32 31 30 29 28 27 26 280 275 –50 –25 0 33 25 50 75 100 125 150 175 200 TEMPERATURE (°C) 140 –50 –25 0 25 –50 –25 0 25 50 75 100 125 150 175 200 TEMPERATURE (°C) 1871x G12 25 50 75 100 125 150 175 200 TEMPERATURE (°C) 1871x G13 INTVCC Load Regulation 1871x G14 INTVCC Line Regulation 5.4 VIN = 7.5V INTVCC VOLTAGE (V) INTVCC VOLTAGE (V) 5.2 5.1 5.0 0 10 20 30 40 50 60 INTVCC LOAD (mA) 70 80 5.3 5.2 5.1 0 5 1871x G15 10 15 20 25 VIN (V) 30 35 40 1871x G16 1871xf 6 For more information www.linear.com/LTC1871X LTC1871X PIN FUNCTIONS RUN (Pin 1): The RUN pin provides the user with an accurate means for sensing the input voltage and programming the start-up threshold for the converter. The falling RUN pin threshold is nominally 1.248V and the comparator has 100mV of hysteresis for noise immunity. When the RUN pin is below this input threshold, the IC is shut down and the VIN supply current is kept to a low value (typ 10μA). The Absolute Maximum Rating for the voltage on this pin is 7V. operating frequency to an external clock. If the MODE/ SYNC pin is connected to ground, Burst Mode operation is enabled. If the MODE/SYNC pin is connected to INTVCC, or if an external logic-level synchronization signal is applied to this input, Burst Mode operation is disabled and the IC operates in a continuous mode. ITH (Pin 2): Error Amplifier Compensation Pin. The current comparator input threshold increases with this control voltage. Nominal voltage range for this pin is 0V to 1.40V. INTVCC (Pin 8): The Internal 5.20V Regulator Output. The gate driver and control circuits are powered from this voltage. Decouple this pin locally to the IC ground with a minimum of 4.7μF low ESR tantalum or ceramic capacitor. FB (Pin3): Receives the feedback voltage from the external resistor divider across the output. Nominal voltage for this pin in regulation is 1.230V. FREQ (Pin 4): A resistor from the FREQ pin to ground programs the operating frequency of the chip. The nominal voltage at the FREQ pin is 0.6V. MODE/SYNC (Pin 5): This input controls the operating mode of the converter and allows for synchronizing the GND (Pin 6): Ground Pin. GATE (Pin 7): Gate Driver Output. VIN (Pin 9): Main Supply Pin. Must be closely decoupled to ground. SENSE (Pin 10): The Current Sense Input for the Control Loop. Connect this pin to the drain of the power MOSFET for VDS sensing and highest efficiency. Alternatively, the SENSE pin may be connected to a resistor in the source of the power MOSFET. Internal leading edge blanking is provided for both sensing methods. 1871xf For more information www.linear.com/LTC1871X 7 LTC1871X BLOCK DIAGRAM RUN 1 BIAS AND START-UP CONTROL SLOPE COMPENSATION C2 1.248V IOSC FREQ 4 V-TO-I VIN OSC 9 0.6V MODE/SYNC 5 INTVCC 50k PWM LATCH 85mV 0V + 1.230V S GATE Q GND R 0.30V BURST COMPARATOR CURRENT COMPARATOR SENSE 10 FB 3 7 LOGIC C1 gm 1.230V ILOOP ITH 2 V-TO-I RLOOP INTVCC 5.2V 8 LDO 1.230V 1.230V BIAS VREF GND UV 2.00V SLOPE TO STARTUP CONTROL 6 VIN 1871x BD 1871xf 8 For more information www.linear.com/LTC1871X LTC1871X OPERATION L VIN D VOUT VIN + SENSE VSW COUT GATE GND GND 2a. SENSE Pin Connection for Maximum Efficiency (VSW < 36V) L VIN VIN D VOUT VSW GATE SENSE GND GND + COUT RS 1871x F02 2b. SENSE Pin Connection for Precise Control of Peak Current or for VSW > 36V Figure 2. Using the SENSE Pin On the LTC1871 The nominal operating frequency of the LTC1871X is programmed using a resistor from the FREQ pin to ground and can be controlled over a 65kHz to 900kHz range. In addition, the internal oscillator can be synchronized to an external clock applied to the MODE/SYNC pin and can be locked to a frequency between 100% and 130% of its nominal value. When the MODE/SYNC pin is left open, it is pulled low by an internal 50k resistor and Burst Mode operation is enabled. If this pin is taken above 2V or an external clock is applied, Burst Mode operation is disabled and the IC operates in continuous mode. With no load (or an extremely light load), the controller will skip pulses in order to maintain regulation and prevent excessive output ripple. The RUN pin controls whether the IC is enabled or is in a low current shutdown state. A micropower 1.248V reference and comparator C2 allow the user to program the supply voltage at which the IC turns on and off (comparator C2 has 100mV of hysteresis for noise immunity). With the RUN pin below 1.248V, the chip is off and the input supply current is typically only 10µA. An overvoltage comparator OV senses when the FB pin exceeds the reference voltage by 6.5% and provides a reset pulse to the main RS latch. Because this RS latch is reset-dominant, the power MOSFET is actively held off for the duration of an output overvoltage condition. The LTC1871X can be used either by sensing the voltage drop across the power MOSFET or by connecting the SENSE pin to a conventional shunt resistor in the source of the power MOSFET, as shown in Figure 2. Sensing the voltage across the power MOSFET maximizes converter efficiency and minimizes the component count, but limits the output voltage to the maximum rating for this pin (36V). By connecting the SENSE pin to a resistor in the source of the power MOSFET, the user is able to program output voltages significantly greater than 36V. Programming the Operating Mode For applications where maximizing the efficiency at very light loads (e.g., <100µA) is a high priority, the current in the output divider could be decreased to a few microamps and Burst Mode operation should be applied (i.e., the MODE/SYNC pin should be connected to ground). In applications where fixed frequency operation is more critical than low current efficiency, or where the lowest output ripple is desired, pulse-skip mode operation should be used and the MODE/SYNC pin should be connected to the INTVCC pin. This allows discontinuous conduction mode (DCM) operation down to near the limit defined by the chip’s minimum on-time (about 175ns). Below this output current level, the converter will begin to skip cycles in order to maintain output regulation. Figures 3 and 4 show the light load switching waveforms for Burst Mode and pulse-skip mode operation for the converter in Figure 1. Burst Mode Operation Burst Mode operation is selected by leaving the MODE/ SYNC pin unconnected or by connecting it to ground. In normal operation, the range on the ITH pin corresponding to no load to full load is 0.30V to 1.2V. In Burst Mode operation, if the error amplifier EA drives the ITH voltage below 0.525V, the buffered ITH input to the current comparator C1 will be clamped at 0.525V (which corresponds to 25% of maximum load current). The inductor current peak is then held at approximately 30mV divided by the power 1871xf For more information www.linear.com/LTC1871X 9 LTC1871X OPERATION MOSFET RDS(ON). If the ITH pin drops below 0.30V, the Burst Mode comparator B1 will turn off the power MOSFET and scale back the quiescent current of the IC to 250µA (sleep mode). In this condition, the load current will be supplied by the output capacitor until the ITH voltage rises above the 50mV hysteresis of the burst comparator. At light loads, short bursts of switching (where the average inductor current is 20% of its maximum value) followed by long periods of sleep will be observed, thereby greatly improving converter efficiency. Oscilloscope waveforms illustrating Burst Mode operation are shown in Figure 3. Pulse-Skip Mode Operation With the MODE/SYNC pin tied to a DC voltage above 2V, Burst Mode operation is disabled. The internal, 0.525V buffered ITH burst clamp is removed, allowing the ITH pin to directly control the current comparator from no load to full load. With no load, the ITH pin is driven below 0.30V, the power MOSFET is turned off and sleep mode is invoked. Oscilloscope waveforms illustrating this mode of operation are shown in Figure 4. VIN = 3.3V VOUT = 5V IOUT = 500mA MODE/SYNC = 0V (Burst Mode OPERATION) VOUT 50mV/DIV When an external clock signal drives the MODE/SYNC pin at a rate faster than the chip’s internal oscillator, the oscillator will synchronize to it. In this synchronized mode, Burst Mode operation is disabled. The constant frequency associated with synchronized operation provides a more controlled noise spectrum from the converter, at the expense of overall system efficiency of light loads. When the oscillator’s internal logic circuitry detects a synchronizing signal on the MODE/SYNC pin, the internal oscillator ramp is terminated early and the slope compensation is increased by approximately 30%. As a result, in applications requiring synchronization, it is recommended that the nominal operating frequency of the IC be programmed to be about 75% of the external clock frequency. Attempting to synchronize to too high an external frequency (above 1.3fO) can result in inadequate slope compensation and possible subharmonic oscillation (or jitter). The external clock signal must exceed 2V for at least 25ns, and should have a maximum duty cycle of 80%, as shown in Figure 5. The MOSFET turn on will synchronize to the rising edge of the external clock signal. MODE/ SYNC 2V TO 7V tMIN = 25ns 0.8T T T = 1/fO IL 5A/DIV GATE 10µs/DIV Figure 3. LTC1871X Burst Mode Operation (MODE/SYNC = 0V) at Low Output Current VIN = 3.3V VOUT = 5V IOUT = 500mA D = 40% 1871x F03 IL 1871x F05 MODE/SYNC = INTVCC (PULSE-SKIP MODE) Figure 5. MODE/SYNC Clock Input and Switching Waveforms for Synchronized Operation VOUT 50mV/DIV IL 5A/DIV 2µs/DIV 1871x F04 Figure 4. LTC1871X Low Output Current Operation with Burst Mode Operation Disabled (MODE/SYNC = INTVCC) 1871xf 10 For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION Programming the Operating Frequency INTVCC Regulator Bypassing and Operation The choice of operating frequency and inductor value is a tradeoff between efficiency and component size. Low frequency operation improves efficiency by reducing MOSFET and diode switching losses. However, lower frequency operation requires more inductance for a given amount of load current. An internal, P-channel low dropout voltage regulator produces the 5.2V supply which powers the gate driver and logic circuitry within the LTC1871X, as shown in Figure 7. The INTVCC regulator can supply up to 50mA and must be bypassed to ground immediately adjacent to the IC pins with a minimum of 4.7µF tantalum or ceramic capacitor. Good bypassing is necessary to supply the high transient currents required by the MOSFET gate driver. The LTC1871X uses a constant frequency architecture that can be programmed over a 65kHz to 900kHz range with a single external resistor from the FREQ pin to ground, as shown in Figure 1. The nominal voltage on the FREQ pin is 0.6V, and the current that flows into the FREQ pin is used to charge and discharge an internal oscillator capacitor. A graph for selecting the value of RT for a given operating frequency is shown in Figure 6. RT (kΩ) 1000 100 10 For input voltages that don’t exceed 7V (the absolute maximum rating for this pin), the internal low dropout regulator in the LTC1871X is redundant and the INTVCC pin can be shorted directly to the VIN pin. With the INTVCC pin shorted to VIN, however, the divider that programs the regulated INTVCC voltage will draw 10µA of current from the input supply, even in shutdown mode. For applications that require the lowest shutdown mode input supply current, do not connect the INTVCC pin to VIN. Regardless of whether the INTVCC pin is shorted to VIN or not, it is always necessary to have the driver circuitry bypassed with a 4.7µF tantalum or low ESR ceramic capacitor to ground immediately adjacent to the INTVCC and GND pins. In an actual application, most of the IC supply current is used to drive the gate capacitance of the power MOSFET. As a result, high input voltage applications in which a large power MOSFET is being driven at high frequencies can cause the LTC1871X to exceed its maximum junc- 0 100 200 300 400 500 600 700 800 900 1000 FREQUENCY (kHz) 1871x F06 Figure 6. Timing Resistor (RT) Value INPUT SUPPLY 2.5V TO 30V VIN 1.230V – P-CH + CIN R2 R1 5.2V INTVCC + LOGIC DRIVER CVCC 4.7µF GATE M1 GND 1871x F07 GND PLACE AS CLOSE AS POSSIBLE TO DEVICE PINS Figure 7. Bypassing the LDO Regulator and Gate Driver Supply 1871xf For more information www.linear.com/LTC1871X 11 LTC1871X APPLICATIONS INFORMATION tion temperature rating. The junction temperature can be estimated using the following equations: IQ(TOT) ≈ IQ + f • QG PIC = VIN • (IQ + f • QG) TJ = TA + PIC • RTH(JA) The total quiescent current IQ(TOT) consists of the static supply current (IQ) and the current required to charge and discharge the gate of the power MOSFET. The 10-pin MSOP package has a thermal resistance of RTH(JA) = 120°C/W. As an example, consider a power supply with VIN = 5V and VO = 12V at IO = 1A. The switching frequency is 500kHz, and the maximum ambient temperature is 70°C. The power MOSFET chosen is the IRF7805, which has a maximum RDS(ON) of 11mΩ (at room temperature) and a maximum total gate charge of 37nC (the temperature coefficient of the gate charge is low). IQ(TOT) = 600µA + 37nC • 500kHz = 19.1mA PIC = 5V • 19.1mA = 95mW TJ = 70°C + 120°C/W • 95mW = 81.4°C This demonstrates how significant the gate charge current can be when compared to the static quiescent current in the IC. To prevent the maximum junction temperature from being exceeded, the input supply current must be checked when operating in a continuous mode at high VIN. A tradeoff between the operating frequency and the size of the power MOSFET may need to be made in order to maintain a reliable IC junction temperature. Prior to lowering the operating frequency, however, be sure to check with power MOSFET manufacturers for their latest-and-greatest low QG, low RDS(ON) devices. Power MOSFET manufacturing technologies are continually improving, with newer and better performance devices being introduced almost yearly. Output Voltage Programming The output voltage is set by a resistor divider according to the following formula: ⎛ R2 ⎞ VO = 1.230V • ⎜ 1+ ⎟ ⎝ R1⎠ The external resistor divider is connected to the output as shown in Figure 1, allowing remote voltage sensing. The resistors R1 and R2 are typically chosen so that the error caused by the current flowing into the FB pin during normal operation is less than 1% (this translates to a maximum value of R1 of about 250k). Programming Turn-On and Turn-Off Thresholds with the RUN Pin The LTC1871X contains an independent, micropower voltage reference and comparator detection circuit that remains active even when the device is shut down, as shown in Figure 8. This allows users to accurately program an input voltage at which the converter will turn on and off. The falling threshold voltage on the RUN pin is equal to the internal reference voltage of 1.248V. The comparator has 100mV of hysteresis to increase noise immunity. The turn-on and turn-off input voltage thresholds are programmed using a resistor divider according to the following formulas: ⎛ R2 ⎞ VIN(OFF) = 1.248V • ⎜ 1+ ⎟ ⎝ R1⎠ ⎛ R2 ⎞ VIN(ON) = 1.348V • ⎜ 1+ ⎟ ⎝ R1⎠ The resistor R1 is typically chosen to be less than 1M. For applications where the RUN pin is only to be used as a logic input, the user should be aware of the 7V Absolute Maximum Rating for this pin! The RUN pin can be connected to the input voltage through an external 1M resistor, as shown in Figure 8c, for “always on” operation. Application Circuits A basic LTC1871X application circuit is shown in Figure 1. The circuit in Figure 1 was tested at 175°C to verify circuit operation. External component selection is driven by the characteristics of the load and the input supply. The first topology to be analyzed will be the boost converter, followed by SEPIC (single ended primary inductance converter). 1871xf 12 For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION VIN + R2 RUN + RUN COMPARATOR BIAS AND START-UP CONTROL 6V INPUT SUPPLY – OPTIONAL FILTER CAPACITOR R1 1.248V µPOWER REFERENCE GND – 1871x F8a Figure 8a. Programming the Turn-On and Turn-Off Thresholds Using the RUN Pin VIN + RUN COMPARATOR RUN EXTERNAL LOGIC CONTROL + R2 1M RUN + 6V INPUT SUPPLY – 6V 1.248V – 1871x F08b RUN COMPARATOR GND – 1.248V 1871x F08c Figure 8b. On/Off Control Using External Logic Figure 8c. External Pull-Up Resistor On RUN Pin for “Always On” Operation Boost Converter: Duty Cycle Considerations Boost Converter: The Peak and Average Input Currents For a boost converter operating in a continuous conduction mode (CCM), the duty cycle of the main switch is: The control circuit in the LTC1871X is measuring the input current (either by using the RDS(ON) of the power MOSFET or by using a sense resistor in the MOSFET source), so the output current needs to be reflected back to the input in order to dimension the power MOSFET properly. Based on the fact that, ideally, the output power is equal to the input power, the maximum average input current is: ⎛V +V –V ⎞ D = ⎜ O D IN ⎟ ⎝ VO + VD ⎠ where VD is the forward voltage of the boost diode. For converters where the input voltage is close to the output voltage, the duty cycle is low and for converters that develop a high output voltage from a low voltage input supply, the duty cycle is high. The maximum output voltage for a boost converter operating in CCM is: VO(MAX) = VIN(MIN) (1– DMAX ) IIN(MAX) = IO(MAX) 1– DMAX The peak input current is: ⎛ χ ⎞ IO(MAX) IIN(PEAK) = ⎜ 1+ ⎟ • ⎝ 2 ⎠ 1– DMAX – VD The maximum duty cycle capability of the LTC1871X is typically 92%. This allows the user to obtain high output voltages from low input supply voltages. The maximum duty cycle, DMAX, should be calculated at minimum VIN. 1871xf For more information www.linear.com/LTC1871X 13 LTC1871X APPLICATIONS INFORMATION Boost Converter: Ripple Current ∆IL and the ‘χ’ Factor The constant ‘χ’ in the equation above represents the percentage peak-to-peak ripple current in the inductor, relative to its maximum value. For example, if 30% ripple current is chosen, then χ = 0.30, and the peak current is 15% greater than the average. For a current mode boost regulator operating in CCM, slope compensation must be added for duty cycles above 50% in order to avoid subharmonic oscillation. For the LTC1871X, this ramp compensation is internal. Having an internally fixed ramp compensation waveform, however, does place some constraints on the value of the inductor and the operating frequency. If too large an inductor is used, the resulting current ramp (∆IL) will be small relative to the internal ramp compensation (at duty cycles above 50%), and the converter operation will approach voltage mode (ramp compensation reduces the gain of the current loop). If too small an inductor is used, but the converter is still operating in CCM (near critical conduction mode), the internal ramp compensation may be inadequate to prevent subharmonic oscillation. To ensure good current mode gain and avoid subharmonic oscillation, it is recommended that the ripple current in the inductor fall in the range of 20% to 40% of the maximum average current. For example, if the maximum average input current is 1A, choose a ∆IL between 0.2A and 0.4A, and a value ‘χ’ between 0.2 and 0.4. Boost Converter: Inductor Selection Given an operating input voltage range, and having chosen the operating frequency and ripple current in the inductor, the inductor value can be determined using the following equation: L= VIN(MIN) ΔIL • f The minimum required saturation current of the inductor can be expressed as a function of the duty cycle and the load current, as follows: ⎛ χ ⎞ IO(MAX) IL(SAT) ≥ ⎜ 1+ ⎟ • ⎝ 2 ⎠ 1– DMAX The saturation current rating for the inductor should be checked at the minimum input voltage (which results in the highest inductor current) and maximum output current. Boost Converter: Operating in Discontinuous Mode Discontinuous mode operation occurs when the load current is low enough to allow the inductor current to run out during the off-time of the switch, as shown in Figure 9. Once the inductor current is near zero, the switch and diode capacitances resonate with the inductance to form damped ringing at 1MHz to 10MHz. If the off-time is long enough, the drain voltage will settle to the input voltage. Depending on the input voltage and the residual energy in the inductor, this ringing can cause the drain of the power MOSFET to go below ground where it is clamped by the body diode. This ringing is not harmful to the IC and it has not been shown to contribute significantly to EMI. Any attempt to damp it with a snubber will degrade the efficiency. VIN = 3.3V IOUT = 200mA VOUT = 5V MOSFET DRAIN VOLTAGE 2V/DIV • DMAX where: ΔIL = χ • current is limited only by the input supply capability. For applications requiring a step-up converter that is shortcircuit protected, please refer to the applications section covering SEPIC converters. INDUCTOR CURRENT 2A/DIV IO(MAX) 1– DMAX 2µs/DIV Remember that boost converters are not short-circuit protected. Under a shorted output condition, the inductor 1871x F09 Figure 9. Discontinuous Mode Waveforms 1871xf 14 For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION Once the value for L is known, the type of inductor must be selected. High efficiency converters generally cannot afford the core loss found in low cost powdered iron cores, forcing the use of more expensive ferrite, molypermalloy or Kool Mµ® cores. Actual core loss is independent of core size for a fixed inductor value, but is very dependent on the inductance selected. As inductance increases, core losses go down. Unfortunately, increased inductance requires more turns of wire and therefore, copper losses will increase. Generally, there is a tradeoff between core losses and copper losses that needs to be balanced. Ferrite designs have very low core losses and are preferred at high switching frequencies, so design goals can concentrate on copper losses and preventing saturation. Ferrite core material saturates “hard,” meaning that the inductance collapses rapidly when the peak design current is exceeded. This results in an abrupt increase in inductor ripple current and consequently, output voltage ripple. Do not allow the core to saturate! Molypermalloy (from Magnetics, Inc.) is a very good, low cost core material for toroids, but is more expensive than ferrite. A reasonable compromise from the same manufacturer is Kool Mµ. Boost Converter: Power MOSFET Selection The power MOSFET serves two purposes in the LTC1871X: it represents the main switching element in the power path, and its RDS(ON) represents the current sensing element for the control loop. Important parameters for the power MOSFET include the drain-to-source breakdown voltage (BVDSS), the threshold voltage (VGS(TH)), the onresistance (RDS(ON)) versus gate-to-source voltage, the gate-to-source and gate-to-drain charges (QGS and QGD, respectively), the maximum drain current (ID(MAX)) and the MOSFET’s thermal resistances (RTH(JC) and RTH(JA)). The gate drive voltage is set by the 5.2V INTVCC low drop regulator. Consequently, logic-level threshold MOSFETs should be used in most LTC1871X applications. If low input voltage operation is expected (e.g., supplying power from a lithium-ion battery or a 3.3V logic supply), then sublogic-level threshold MOSFETs should be used. Pay close attention to the BVDSS specifications for the MOSFETs relative to the maximum actual switch voltage in the application. Many logic-level devices are limited to 30V or less, and the switch node can ring during the turn-off of the MOSFET due to layout parasitics. Check the switching waveforms of the MOSFET directly across the drain and source terminals using the actual PC board layout (not just on a lab breadboard!) for excessive ringing. During the switch on-time, the control circuit limits the maximum voltage drop across the power MOSFET to about 150mV (at low duty cycle). The peak inductor current is therefore limited to 150mV/RDS(ON). The relationship between the maximum load current, duty cycle and the RDS(ON) of the power MOSFET is: RDS(ON) ≤ VSENSE(MAX) • 1– DMAX ⎛ χ⎞ ⎜⎝ 1+ 2 ⎟⎠ •IO(MAX) • ρT The VSENSE(MAX) term is typically 150mV at low duty cycle, and is reduced to about 100mV at a duty cycle of 92% due to slope compensation, as shown in Figure 10. The ρT term accounts for the temperature coefficient of the RDS(ON) of the MOSFET, which is typically 0.4%/°C. MAXIMUM CURRENT SENSE VOLTAGE (mV) Boost Converter: Inductor Core Selection 200 150 100 50 0 0 0.2 0.5 0.4 DUTY CYCLE 0.8 1.0 1871 F10 Figure 10. Maximum SENSE Threshold Voltage vs Duty Cycle 1871xf For more information www.linear.com/LTC1871X 15 LTC1871X APPLICATIONS INFORMATION Another method of choosing which power MOSFET to use is to check what the maximum output current is for a given RDS(ON), since MOSFET on-resistances are available in discrete values. 1– DMAX IO(MAX) = VSENSE(MAX) • ⎛ χ⎞ ⎜⎝ 1+ 2 ⎟⎠ • RDS(ON) • ρT It is worth noting that the 1 – DMAX relationship between IO(MAX) and RDS(ON) can cause boost converters with a wide input range to experience a dramatic range of maximum input and output current. This should be taken into consideration in applications where it is important to limit the maximum current drawn from the input supply. Calculating Power MOSFET Switching and Conduction Losses and Junction Temperatures In order to calculate the junction temperature of the power MOSFET, the power dissipated by the device must be known. This power dissipation is a function of the duty cycle, the load current and the junction temperature itself (due to the positive temperature coefficient of its RDS(ON)). As a result, some iterative calculation is normally required to determine a reasonably accurate value. Since the controller is using the MOSFET as both a switching and a sensing element, care should be taken to ensure that the converter is capable of delivering the required load current over all operating conditions (line voltage and temperature), and for the worst-case specifications for VSENSE(MAX) and the RDS(ON) of the MOSFET listed in the manufacturer’s data sheet. The power dissipated by the MOSFET in a boost converter is: 2 ⎛ IO(MAX) ⎞ PFET = ⎜ ⎟ • RDS(ON) • DMAX • ρT ⎝ 1– DMAX ⎠ IO(MAX) +k • VO1.85 • •C •f (1– DMAX ) RSS The first term in the equation above represents the I2R losses in the device, and the second term, the switching losses. The constant, k = 1.7, is an empirical factor inversely related to the gate drive current and has the dimension of 1/current. From a known power dissipated in the power MOSFET, its junction temperature can be obtained using the following formula: TJ = TA + PFET • RTH(JA) The RTH(JA) to be used in this equation normally includes the RTH(JC) for the device plus the thermal resistance from the case to the ambient temperature (RTH(CA)). This value of TJ can then be compared to the original, assumed value used in the iterative calculation process. Boost Converter: Output Diode Selection To maximize efficiency, a fast switching diode with low forward drop and low reverse leakage is desired. The output diode in a boost converter conducts current during the switch off-time. The peak reverse voltage that the diode must withstand is equal to the regulator output voltage. The average forward current in normal operation is equal to the output current, and the peak current is equal to the peak inductor current. ⎛ χ ⎞ IO(MAX) ID(PEAK) =IL(PEAK) = ⎜ 1+ ⎟ • ⎝ 2 ⎠ 1– DMAX The power dissipated by the diode is: PD = IO(MAX) • VD and the diode junction temperature is: TJ = TA + PD • RTH(JA) The RTH(JA) to be used in this equation normally includes the RTH(JC) for the device plus the thermal resistance from the board to the ambient temperature in the enclosure. Remember to keep the diode lead lengths short and to observe proper switch-node layout (see Board Layout Checklist) to avoid excessive ringing and increased dissipation. 1871xf 16 For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION Boost Converter: Output Capacitor Selection Contributions of ESR (equivalent series resistance), ESL (equivalent series inductance) and the bulk capacitance must be considered when choosing the correct component for a given output ripple voltage. The effects of these three parameters (ESR, ESL and bulk C) on the output voltage ripple waveform are illustrated in Figure 11e for a typical boost converter. The choice of component(s) begins with the maximum acceptable ripple voltage (expressed as a percentage of the output voltage), and how this ripple should be divided between the ESR step and the charging/discharging ∆V. For the purpose of simplicity we will choose 2% for the maximum output ripple, to be divided equally between the ESR step and the charging/discharging ∆V. This percentage ripple will change, depending on the requirements of the application, and the equations provided below can easily be modified. For a 1% contribution to the total ripple voltage, the ESR of the output capacitor can be determined using the following equation: ESRCOUT ≤ 0.01• VO IIN(PEAK) where: ⎛ χ ⎞ IO(MAX) IIN(PEAK)= ⎜ 1+ ⎟ • ⎝ 2 ⎠ 1– DMAX For the bulk C component, which also contributes 1% to the total ripple: COUT ≥ IO(MAX) 0.01• VO • f For many designs it is possible to choose a single capacitor type that satisfies both the ESR and bulk C requirements for the design. In certain demanding applications, however, the ripple voltage can be improved significantly by connecting two or more types of capacitors in parallel. For example, using a low ESR ceramic capacitor can minimize the ESR step, while an electrolytic capacitor can be used to supply the required bulk C. Once the output capacitor ESR and bulk capacitance have been determined, the overall ripple voltage waveform should be verified on a dedicated PC board (see Board Layout section for more information on component placement). Lab breadboards generally suffer from excessive series inductance (due to inter-component wiring), and these parasitics can make the switching waveforms look significantly worse than they would be on a properly designed PC board. The output capacitor in a boost regulator experiences high RMS ripple currents, as shown in Figure 11. The RMS output capacitor ripple current is: IRMS(COUT) ≈IO(MAX) • VO – VIN(MIN) VIN(MIN) Note that the ripple current ratings from capacitor manufacturers are often based on only 2000 hours of life. This makes it advisable to further derate the capacitor or to choose a capacitor rated at a higher temperature than required. Several capacitors may also be placed in parallel to meet size or height requirements in the design. Manufacturers such as Nichicon, United Chemicon and Sanyo should be considered for high performance throughhole capacitors. The OS-CON semiconductor dielectric capacitor available from Sanyo has the lowest product of ESR and size of any aluminum electrolytic, at a somewhat higher price. In surface mount applications, multiple capacitors may have to be placed in parallel in order to meet the ESR or RMS current handling requirements of the application. Aluminum electrolytic and dry tantalum capacitors are both available in surface mount packages. In the case of tantalum, it is critical that the capacitors have been surge tested for use in switching power supplies. An excellent choice is AVX TPS series of surface mount tantalum. Also, ceramic capacitors are now available with extremely low ESR, ESL and high ripple current ratings. 1871xf For more information www.linear.com/LTC1871X 17 LTC1871X APPLICATIONS INFORMATION L VIN D SW Please note that the input capacitor can see a very high surge current when a battery is suddenly connected to the input of the converter and solid tantalum capacitors can fail catastrophically under these conditions. Be sure to specify surge-tested capacitors! VOUT COUT RL 11a. Circuit Diagram Burst Mode Operation and Considerations IIN IL The choice of MOSFET RDS(ON) and inductor value also determines the load current at which the LTC1871X enters Burst Mode operation. When bursting, the controller clamps the peak inductor current to approximately: 11b. Inductor and Input Currents IBURST(PEAK) = ISW tON 11c. Switch Current ID tOFF which represents about 20% of the maximum 150mV SENSE pin voltage. The corresponding average current depends upon the amount of ripple current. Lower inductor values (higher ∆IL) will reduce the load current at which Burst Mode operations begins, since it is the peak current that is being clamped. IO 11d. Diode and Output Currents ΔVCOUT VOUT (AC) RINGING DUE TO TOTAL INDUCTANCE (BOARD + CAP) ΔVESR 1871x F11 11e. Output Voltage Ripple Waveform Figure 11. Switching Waveforms for a Boost Converter Boost Converter: Input Capacitor Selection The input capacitor of a boost converter is less critical than the output capacitor, due to the fact that the inductor is in series with the input and the input current waveform is continuous (see Figure 11b). The input voltage source impedance determines the size of the input capacitor, which is typically in the range of 10µF to 100µF. A low ESR capacitor is recommended, although it is not as critical as for the output capacitor. The RMS input capacitor ripple current for a boost converter is: IRMS(CIN) = 0.3 • 18 VIN(MIN) L•f • DMAX 30mV RDS(ON) The output voltage ripple can increase during Burst Mode operation if ∆IL is substantially less than IBURST. This can occur if the input voltage is very low or if a very large inductor is chosen. At high duty cycles, a skipped cycle causes the inductor current to quickly decay to zero. However, because ∆IL is small, it takes multiple cycles for the current to ramp back up to IBURST(PEAK). During this inductor charging interval, the output capacitor must supply the load current and a significant droop in the output voltage can occur. Generally, it is a good idea to choose a value of inductor ∆IL between 25% and 40% of IIN(MAX). The alternative is to either increase the value of the output capacitor or disable Burst Mode operation using the MODE/SYNC pin. Burst Mode operation can be defeated by connecting the MODE/SYNC pin to a high logic-level voltage (either with a control input or by connecting this pin to INTVCC). In this mode, the burst clamp is removed, and the chip can operate at constant frequency from continuous conduction mode (CCM) at full load, down into deep discontinuous conduction mode (DCM) at light load. Prior to skipping pulses at very light load (i.e., < 5% of full load), the controller will operate with a minimum switch on-time in DCM. 1871xf For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION Table 1. Recommended Component Manufacturers VENDOR COMPONENTS AVX BH Electronics TELEPHONE WEB ADDRESS Capacitors (207) 282-5111 avxcorp.com Inductors, Transformers (952) 894-9590 bhelectronics.com Coilcraft Inductors (847) 639-6400 coilcraft.com Coiltronics Inductors (407) 241-7876 coiltronics.com Diodes (805) 446-4800 diodes.com MOSFETs (408) 822-2126 fairchildsemi.com Diodes (516) 847-3000 generalsemiconductor.com MOSFETs, Diodes (310) 322-3331 irf.com Diodes, Inc Fairchild General Semiconductor International Rectifier IRC Kemet Sense Resistors (361) 992-7900 irctt.com Tantalum Capacitors (408) 986-0424 kemet.com Toroid Cores (800) 245-3984 mag-inc.com Microsemi Magnetics Inc Diodes (617) 926-0404 microsemi.com Murata-Erie Inductors, Capacitors (770) 436-1300 murata.co.jp Capacitors (847) 843-7500 nichicon.com Nichicon On Semiconductor Diodes (602) 244-6600 onsemi.com Panasonic Capacitors (714) 373-7334 panasonic.com Sanyo Capacitors (619) 661-6835 sanyo.co.jp Sumida Inductors (847) 956-0667 sumida.com Taiyo Yuden Capacitors (408) 573-4150 t-yuden.com Capacitors, Inductors (562) 596-1212 component.tdk.com Thermalloy Heat Sinks (972) 243-4321 aavidthermalloy.com Tokin Capacitors (408) 432-8020 nec-tokinamerica.com Toko Inductors (847) 699-3430 tokoam.com United Chemicon Capacitors (847) 696-2000 chemi-com.com Vishay/Dale Resistors (605) 665-9301 vishay.com TDK Vishay/Siliconix MOSFETs (800) 554-5565 vishay.com Vishay/Sprague Capacitors (207) 324-4140 vishay.com Small-Signal Discretes (631) 543-7100 zetex.com Zetex Pulse skipping prevents a loss of control of the output at very light loads and reduces output voltage ripple. Efficiency Considerations: How Much Does VDS Sensing Help? The efficiency of a switching regulator is equal to the output power divided by the input power (×100%). Percent efficiency can be expressed as: % Efficiency = 100% – (L1 + L2 + L3 + …), where L1, L2, etc. are the individual loss components as a percentage of the input power. It is often useful to analyze individual losses to determine what is limiting the efficiency and which change would produce the most improvement. Although all dissipative elements in the circuit produce losses, four main sources usually account for the majority of the losses in LTC1871X application circuits: 1. The supply current into VIN. The VIN current is the sum of the DC supply current IQ (given in the Electrical Characteristics) and the MOSFET driver and control currents. The DC supply current into the VIN pin is typically about 550µA and represents a small power loss (much less than 1%) that increases with VIN. The driver current results from switching the gate capacitance of the power MOSFET; this current is typically much larger than the DC current. Each time the MOSFET is switched on and 1871xf For more information www.linear.com/LTC1871X 19 LTC1871X APPLICATIONS INFORMATION then off, a packet of gate charge QG is transferred from INTVCC to ground. The resulting dQ/dt is a current that must be supplied to the INTVCC capacitor through the VIN pin by an external supply. If the IC is operating in CCM: IQ(TOT) ≈ IQ = f • QG PIC = VIN • (IQ + f • QG) 2. Power MOSFET switching and conduction losses. The technique of using the voltage drop across the power MOSFET to close the current feedback loop was chosen because of the increased efficiency that results from not having a sense resistor. The losses in the power MOSFET are equal to: 2 ⎛ IO(MAX) ⎞ PFET = ⎜ ⎟ • RDS(ON) • DMAX • ρT ⎝ 1– DMAX ⎠ IO(MAX) +k • VO1.85 • •C •f (1– DMAX ) RSS The I2R power savings that result from not having a discrete sense resistor can be calculated almost by inspection. 2 ⎛ IO(MAX) ⎞ • RSENSE • DMAX PR(SENSE) = ⎜ 1– DMAX ⎟⎠ ⎝ To understand the magnitude of the improvement with this VDS sensing technique, consider the 3.3V input, 5V output power supply shown in Figure 1. The maximum load current is 7A (10A peak) and the duty cycle is 39%. Assuming a ripple current of 40%, the peak inductor current is 13.8A and the average is 11.5A. With a maximum sense voltage of about 140mV, the sense resistor value would be 10mΩ, and the power dissipated in this resistor would be 514mW at maximum output current. Assuming an efficiency of 90%, this sense resistor power dissipation represents 1.3% of the overall input power. In other words, for this application, the use of VDS sensing would increase the efficiency by approximately 1.3%. For more details regarding the various terms in these equations, please refer to the section Boost Converter: Power MOSFET Selection. 20 3. The losses in the inductor are simply the DC input current squared times the winding resistance. Expressing this loss as a function of the output current yields: 2 ⎛ IO(MAX) ⎞ • RW PR(WINDING) = ⎜ 1– DMAX ⎟⎠ ⎝ 4. Losses in the boost diode. The power dissipation in the boost diode is: PDIODE = IO(MAX) • VD The boost diode can be a major source of power loss in a boost converter. For the 3.3V input, 5V output at 7A example given above, a Schottky diode with a 0.4V forward voltage would dissipate 2.8W, which represents 7% of the input power. Diode losses can become significant at low output voltages where the forward voltage is a significant percentage of the output voltage. 5. Other losses, including CIN and CO ESR dissipation and inductor core losses, generally account for less than 2% of the total additional loss. Checking Transient Response The regulator loop response can be verified by looking at the load transient response. Switching regulators generally take several cycles to respond to an instantaneous step in resistive load current. When the load step occurs, VO immediately shifts by an amount equal to (∆ILOAD)(ESR), and then CO begins to charge or discharge (depending on the direction of the load step) as shown in Figure 12. The regulator feedback loop acts on the resulting error amp output signal to return VO to its steady-state value. During this recovery time, VO can be monitored for overshoot or ringing that would indicate a stability problem. IOUT 2V/DIV VIN = 3.3V VOUT = 5V MODE/SYNC = INTVCC (PULSE-SKIP MODE) VOUT (AC) 100mV/DIV 100µs/DIV 1871x F12 Figure 12. Load Transient Response for a 3.3V Input, 5V Output Boost Converter Application, 0.7A to 7A Step For more information www.linear.com/LTC1871X 1871xf LTC1871X APPLICATIONS INFORMATION A second, more severe transient can occur when connecting loads with large (> 1µF) supply bypass capacitors. The discharged bypass capacitors are effectively put in parallel with CO, causing a nearly instantaneous drop in VO. No regulator can deliver enough current to prevent this problem if the load switch resistance is low and it is driven quickly. The only solution is to limit the rise time of the switch drive in order to limit the inrush current di/ dt to the load. can be used. Because the duty cycle is 39%, the maximum SENSE pin threshold voltage is reduced from its low duty cycle typical value of 150mV to approximately 140mV. Assuming a MOSFET junction temperature of 125°C, the room temperature MOSFET RDS(ON) should be less than: 1– DMAX ⎛ χ⎞ ⎜⎝ 1+ 2 ⎟⎠ •IO(MAX) • ρT 1– 0.39 = 0.140V • = 6.8mΩ ⎛ 0.4 ⎞ 1+ • 7A • 1.5 ⎜⎝ 2 ⎟⎠ RDS(ON) ≤ VSENSE(MAX) • Boost Converter Design Example The design example given here will be for the circuit shown in Figure 1. The input voltage is 3.3V, and the output is 5V at a maximum load current of 7A (10A peak). 1. The duty cycle is: ⎛ V + V – V ⎞ 5 + 0.4 – 3.3 = 38.9% D = ⎜ O D IN ⎟ = VO + VD ⎠ 5 + 0.4 ⎝ 2. Pulse-skip operation is chosen so the MODE/SYNC pin is shorted to INTVCC. 3. The operating frequency is chosen to be 300kHz to reduce the size of the inductor. From Figure 5, the resistor from the FREQ pin to ground is 80k. 4. An inductor ripple current of 40% of the maximum load current is chosen, so the peak input current (which is also the minimum saturation current) is: The MOSFET used was the Fairchild FDS7760A, which has a maximum RDS(ON) of 8mΩ at 4.5V VGS, a BVDSS of greater than 30V, and a gate charge of 37nC at 5V VGS. 6. The diode for this design must handle a maximum DC output current of 10A and be rated for a minimum reverse voltage of VOUT, or 5V. A 25A, 15V diode from On Semiconductor (MBRB2515L) was chosen for its high power dissipation capability. 7. The output capacitor usually consists of a high valued bulk C connected in parallel with a lower valued, low ESR ceramic. Based on a maximum output ripple voltage of 1%, or 50mV, the bulk C needs to be greater than: COUT ≥ IO(MAX) 7 ⎛ χ⎞ IIN(PEAK) = ⎜1+ ⎟ • = 1.2 • = 13.8A ⎝ 2 ⎠ 1– DMAX 1– 0.39 The inductor ripple current is: ΔIL = χ • IO(MAX) 1– DMAX = 0.4 • 7 = 4.6A 1– 0.39 And so the inductor value is: L= VIN(MIN) ΔIL • f • DMAX = IOUT(MAX) 0.01• VOUT • f = 7A = 466µF 0.01• 5V • 300kHz The RMS ripple current rating for this capacitor needs to exceed: IRMS(COUT) ≥IO(MAX) • 3.3V • 0.39 = 0.93µH 4.6A • 300kHz The component chosen is a 1µH inductor made by Sumida (part number CEP125-H 1ROMH) which has a saturation current of greater than 20A. 5. With the input voltage to the IC bootstrapped to the output of the power supply (5V), a logic-level MOSFET 7A • VO – VIN(MIN) VIN(MIN) = 5V – 3.3V = 5A 3.3V To satisfy this high RMS current demand, four 150µF Panasonic capacitors (EEFUEOJ151R) are required. In parallel with these bulk capacitors, two 22µF, low ESR (X5R) Taiyo Yuden ceramic capacitors (JMK325BJ226MM) are added for HF noise reduction. For more information www.linear.com/LTC1871X 1871xf 21 LTC1871X APPLICATIONS INFORMATION Check the output ripple with a single oscilloscope probe connected directly across the output capacitor terminals, where the HF switching currents flow. 8. The choice of an input capacitor for a boost converter depends on the impedance of the source supply and the amount of input ripple the converter will safely tolerate. For this particular design and lab setup a 100µF Sanyo Poscap (6TPC 100M), in parallel with two 22µF Taiyo Yuden ceramic capacitors (JMK325BJ226MM) is required (the input and return lead lengths are kept to a few inches, but the peak input current is close to 20A!). As with the output node, check the input ripple with a single oscilloscope probe connected across the input capacitor terminals. PC Board Layout Checklist 1. In order to minimize switching noise and improve output load regulation, the GND pin of the LTC1871X should be connected directly to 1) the negative terminal of the INTVCC decoupling capacitor, 2) the negative terminal of the output decoupling capacitors, 3) the source of the power MOSFET or the bottom terminal of the sense resistor, 4) the negative terminal of the input capacitor and 5) at least one via to the ground plane immediately adjacent to Pin 6. The ground trace on the top layer of the PC board should be as wide and short as possible to minimize series resistance and inductance. 2. Beware of ground loops in multiple layer PC boards. Try to maintain one central ground node on the board and use the input capacitor to avoid excess input ripple for high output current power supplies. If the ground plane is to be used for high DC currents, choose a path away from the small-signal components. 3. Place the CVCC capacitor immediately adjacent to the INTVCC and GND pins on the IC package. This capacitor carries high di/dt MOSFET gate drive currents. A low ESR and ESL 4.7µF ceramic capacitor works well here. 4. The high di/dt loop from the bottom terminal of the output capacitor, through the power MOSFET, through the boost diode and back through the output capacitors should be kept as tight as possible to reduce inductive ringing. Excess inductance can cause increased stress on the power MOSFET and increase HF noise on the output. If low ESR ceramic capacitors are used on the output to reduce output noise, place these capacitors close to the boost diode in order to keep the series inductance to a minimum. 5. Check the stress on the power MOSFET by measuring its drain-to-source voltage directly across the device terminals (reference the ground of a single scope probe directly to the source pad on the PC board). Beware of inductive ringing which can exceed the maximum specified voltage rating of the MOSFET. If this ringing cannot be avoided and exceeds the maximum rating of the device, either choose a higher voltage device or specify an avalanche-rated power MOSFET. Not all MOSFETs are created equal (some are more equal than others). 6. Place the small-signal components away from high frequency switching nodes. In the layout shown in Figure 13, all of the small-signal components have been placed on one side of the IC and all of the power components have been placed on the other. This also allows the use of a pseudo-Kelvin connection for the signal ground, where high di/dt gate driver currents flow out of the IC ground pin in one direction (to the bottom plate of the INTVCC decoupling capacitor) and small-signal currents flow in the other direction. 7. If a sense resistor is used in the source of the power MOSFET, minimize the capacitance between the SENSE pin trace and any high frequency switching nodes. The LTC1871X contains an internal leading edge blanking time of approximately 180ns, which should be adequate for most applications. 8. For optimum load regulation and true remote sensing, the top of the output resistor divider should connect independently to the top of the output capacitor (Kelvin connection), staying away from any high dV/dt traces. Place the divider resistors near the LTC1871X in order to keep the high impedance FB node short. 9. For applications with multiple switching power converters connected to the same input supply, make sure that the input filter capacitor for the LTC1871X is not shared with other converters. AC input current 1871xf 22 For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION VIN L1 JUMPER R3 RC R4 CC J1 CIN PIN 1 R2 LTC1871X R1 RT CVCC PSEUDO-KELVIN SIGNAL GROUND CONNECTION SWITCH NODE IS ALSO THE HEAT SPREADER FOR L1, M1, D1 M1 COUT COUT D1 VIAS TO GROUND PLANE VOUT TRUE REMOTE OUTPUT SENSING BULK C LOW ESR CERAMIC 1871x F13 Figure 13. LTC1871X Boost Converter Suggested Layout VIN R3 CC R1 R2 R4 1 RC 2 3 4 RT 5 RUN L1 SENSE VIN ITH LTC1871X FB FREQ INTVCC GATE MODE/ SYNC GND 10 J1 SWITCH NODE 9 D1 8 7 6 CVCC M1 + CIN GND + PSEUDO-KELVIN GROUND CONNECTION COUT BOLD LINES INDICATE HIGH CURRENT PATHS 1871x F14 VOUT Figure 14. LTC1871X Boost Converter Layout Diagram 1871xf For more information www.linear.com/LTC1871X 23 LTC1871X APPLICATIONS INFORMATION from another converter could cause substantial input voltage ripple, and this could interfere with the operation of the LTC1871X. A few inches of PC trace or wire (L ≈ 100nH) between the CIN of the LTC1871X and the actual source VIN should be sufficient to prevent current sharing problems. SEPIC Converter Applications The LTC1871X is also well suited to SEPIC (single-ended primary inductance converter) converter applications. The SEPIC converter shown in Figure 15 uses two inductors. The advantage of the SEPIC converter is the input voltage may be higher or lower than the output voltage, and the output is short-circuit protected. VIN C1 L1 D1 + + • SW L2 COUT RL • 15a. SEPIC Topology VIN • + VIN RL • 15b. Current Flow During Switch On-Time VIN D1 + + • VOUT + VIN RL • For a SEPIC converter operating in a continuous conduction mode (CCM), the duty cycle of the main switch is: ⎛ VO + VD ⎞ D= ⎜ ⎝ VIN + VO + VD ⎟⎠ where VD is the forward voltage of the diode. For converters where the input voltage is close to the output voltage the duty cycle is near 50%. The maximum output voltage for a SEPIC converter is: DMAX 1 – VD 1– DMAX 1– DMAX The maximum duty cycle of the LTC1871X is typically 92%. SEPIC Converter: The Peak and Average Input Currents VOUT + + SEPIC Converter: Duty Cycle Considerations VO(MAX) = ( VIN + VD ) VOUT + and size. All of the SEPIC applications information that follows assumes L1 = L2 = L. 1871x F15 15c. Current Flow During Switch Off-Time Figures 15. SEPIC Topology and Current Flow The first inductor, L1, together with the main switch, resembles a boost converter. The second inductor, L2, together with the output diode D1, resembles a flyback or buck-boost converter. The two inductors L1 and L2 can be independent but can also be wound on the same core since identical voltages are applied to L1 and L2 throughout the switching cycle. By making L1 = L2 and winding them on the same core the input ripple is reduced along with cost The control circuit in the LTC1871X is measuring the input current (either using the RDS(ON) of the power MOSFET or by means of a sense resistor in the MOSFET source), so the output current needs to be reflected back to the input in order to dimension the power MOSFET properly. Based on the fact that, ideally, the output power is equal to the input power, the maximum input current for a SEPIC converter is: D IIN(MAX) =IO(MAX) • MAX 1– DMAX The peak input current is: D ⎛ χ⎞ IIN(PEAK) = ⎜ 1+ ⎟ •IO(MAX) • MAX ⎝ 2⎠ 1– DMAX The maximum duty cycle, DMAX, should be calculated at minimum VIN. The constant ‘χ’ represents the fraction of ripple current in the inductor relative to its maximum value. For example, if 30% ripple current is chosen, then χ = 0.30 and the peak current is 15% greater than the average. 1871xf 24 For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION It is worth noting here that SEPIC converters that operate at high duty cycles (i.e., that develop a high output voltage from a low input voltage) can have very high input currents, relative to the output current. Be sure to check that the maximum load current will not overload the input supply. SEPIC Converter: Inductor Selection For most SEPIC applications the equal inductor values will fall in the range of 10µH to 100µH. Higher values will reduce the input ripple voltage and reduce the core loss. Lower inductor values are chosen to reduce physical size and improve transient response. Like the boost converter, the input current of the SEPIC converter is calculated at full load current and minimum input voltage. The peak inductor current can be significantly higher than the output current, especially with smaller inductors and lighter loads. The following formulas assume CCM operation and calculate the maximum peak inductor currents at minimum VIN: V +V ⎛ χ⎞ IL1(PEAK) = ⎜ 1+ ⎟ •IO(MAX) • O D ⎝ 2⎠ VIN(MIN) VIN(MIN) + VD ⎛ χ⎞ IL2(PEAK) = ⎜ 1+ ⎟ •IO(MAX) • ⎝ 2⎠ VIN(MIN) The ripple current in the inductor is typically 20% to 40% (i.e., a range of ‘χ’ from 0.20 to 0.40) of the maximum average input current occurring at VIN(MIN) and IO(MAX) and ∆IL1 = ∆IL2. Expressing this ripple current as a function of the output current results in the following equations for calculating the inductor value: L= VIN(MIN) ΔIL • f • DMAX where: ΔIL = χ •IO(MAX) • DMAX 1– DMAX By making L1 = L2 and winding them on the same core, the value of inductance in the equation above is replace by 2L due to mutual inductance. Doing this maintains the same ripple current and energy storage in the inductors. For example, a Coiltronix CTX10-4 is a 10µH inductor with two windings. With the windings in parallel, 10µH inductance is obtained with a current rating of 4A (the number of turns hasn’t changed, but the wire diameter has doubled). Splitting the two windings creates two 10µH inductors with a current rating of 2A each. Therefore, substituting 2L yields the following equation for coupled inductors: L1= L2 = VIN(MIN) 2 • ΔIL • f • DMAX Specify the maximum inductor current to safely handle IL(PK) specified in the equation above. The saturation current rating for the inductor should be checked at the minimum input voltage (which results in the highest inductor current) and maximum output current. SEPIC Converter: Power MOSFET Selection The power MOSFET serves two purposes in the LTC1871X: it represents the main switching element in the power path, and its RDS(ON) represents the current sensing element for the control loop. Important parameters for the power MOSFET include the drain-to-source breakdown voltage (BVDSS), the threshold voltage (VGS(TH)), the on-resistance (RDS(ON)) versus gate-to-source voltage, the gate-to-source and gate-to-drain charges (QGS and QGD, respectively), the maximum drain current (ID(MAX)) and the MOSFET’s thermal resistances (RTH(JC) and RTH(JA)). The gate drive voltage is set by the 5.2V INTVCC low dropout regulator. Consequently, logic-level threshold MOSFETs should be used in most LTC1871X applications. If low input voltage operation is expected (e.g., supplying power from a lithium-ion battery), then sublogic-level threshold MOSFETs should be used. The maximum voltage that the MOSFET switch must sustain during the off-time in a SEPIC converter is equal to the sum of the input and output voltages (VO + VIN). As a result, careful attention must be paid to the BVDSS specifications for the MOSFETs relative to the maximum actual switch voltage in the application. Many logic-level devices are limited to 30V or less. Check the switching waveforms directly across the drain and source terminals of the power MOSFET to ensure the VDS remains below the maximum rating for the device. 1871xf For more information www.linear.com/LTC1871X 25 LTC1871X APPLICATIONS INFORMATION During the MOSFET’s on-time, the control circuit limits the maximum voltage drop across the power MOSFET to about 150mV (at low duty cycle). The peak inductor current is therefore limited to 150mV/RDS(ON). The relationship between the maximum load current, duty cycle and the RDS(ON) of the power MOSFET is: RDS(ON) ≤ VSENSE(MAX) IO(MAX) • 1 1 • ⎛ χ⎞ ⎛V +V ⎞ ⎜⎝ 1+ 2 ⎟⎠ • ρT ⎜ O D ⎟ + 1 ⎝ VIN(MIN) ⎠ The VSENSE(MAX) term is typically 150mV at low duty cycle and is reduced to about 100mV at a duty cycle of 92% due to slope compensation, as shown in Figure 8. The constant ‘χ’ in the denominator represents the ripple current in the inductors relative to their maximum current. For example, if 30% ripple current is chosen, then χ = 0.30. The ρT term accounts for the temperature coefficient of the RDS(ON) of the MOSFET, which is typically 0.4%/°C. Figure 9 illustrates the variation of normalized RDS(ON) over temperature for a typical power MOSFET. Another method of choosing which power MOSFET to use is to check what the maximum output current is for a given RDS(ON) since MOSFET on-resistances are available in discrete values. IO(MAX) ≤ VSENSE(MAX) RDS(ON) 1 1 • • ⎛ χ⎞ ⎛V +V ⎞ ⎜⎝ 1+ 2 ⎟⎠ • ρT ⎜ O D ⎟ + 1 ⎝ VIN(MIN) ⎠ Calculating Power MOSFET Switching and Conduction Losses and Junction Temperatures In order to calculate the junction temperature of the power MOSFET, the power dissipated by the device must be known. This power dissipation is a function of the duty cycle, the load current and the junction temperature itself. As a result, some iterative calculation is normally required to determine a reasonably accurate value. Since the controller is using the MOSFET as both a switching and a sensing element, care should be taken to ensure that the converter is capable of delivering the required load current over all operating conditions (load, line and temperature) and for the worstcase specifications for VSENSE(MAX) and the RDS(ON) of the MOSFET listed in the manufacturer’s data sheet. The power dissipated by the MOSFET in a SEPIC converter is: 2 ⎛ ⎞ D PFET = ⎜ IO(MAX) • MAX ⎟ • RDS(ON) • DMAX • ρT 1– DMAX ⎠ ⎝ 1.85 D + k • VIN(MIN) + VO •IO(MAX) • MAX • CRSS • f 1– DMAX ( ) The first term in the equation above represents the I2R losses in the device and the second term, the switching losses. The constant k = 1.7 is an empirical factor inversely related to the gate drive current and has the dimension of 1/current. From a known power dissipated in the power MOSFET, its junction temperature can be obtained using the following formula: TJ = TA + PFET •RTH(JA) The RTH(JA) to be used in this equation normally includes the RTH(JC) for the device plus the thermal resistance from the board to the ambient temperature in the enclosure. This value of TJ can then be used to check the original assumption for the junction temperature in the iterative calculation process. SEPIC Converter: Output Diode Selection To maximize efficiency, a fast-switching diode with low forward drop and low reverse leakage is desired. The output diode in a SEPIC converter conducts current during the switch off-time. The peak reverse voltage that the diode must withstand is equal to VIN(MAX) + VO. The average forward current in normal operation is equal to the output current, and the peak current is equal to: ⎛V +V ⎞ ⎛ χ⎞ ID(PEAK) = ⎜ 1+ ⎟ •IO(MAX) • ⎜ O D + 1⎟ ⎝ 2⎠ ⎝ VIN(MIN) ⎠ The power dissipated by the diode is: PD = IO(MAX) • VD and the diode junction temperature is: TJ = TA + PD • RTH(JA) The RTH(JA) to be used in this equation normally includes the RTH(JC) for the device plus the thermal resistance from the board to the ambient temperature in the enclosure. 1871xf 26 For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION SEPIC Converter: Output Capacitor Selection Because of the improved performance of today’s electrolytic, tantalum and ceramic capacitors, engineers need to consider the contributions of ESR (equivalent series resistance), ESL (equivalent series inductance) and the bulk capacitance when choosing the correct component for a given output ripple voltage. The effects of these three parameters (ESR, ESL, and bulk C) on the output voltage ripple waveform are illustrated in Figure 16 for a typical coupled-inductor SEPIC converter. IL1 IIN SW ON SW OFF IO 16b. Output Inductor Current IIN IC1 IO 16c. DC Coupling Capacitor Current ID1 IO 16d. Diode Current VOUT (AC) ΔVCOUT ΔVESR 1871x F16 For a 1% contribution to the total ripple voltage, the ESR of the output capacitor can be determined using the following equation: ESRCOUT ≤ 16a. Input Inductor Current IL2 The choice of component(s) begins with the maximum acceptable ripple voltage (expressed as a percentage of the output voltage), and how this ripple should be divided between the ESR step and the charging/discharging ∆V. For the purpose of simplicity we will choose 2% for the maximum output ripple, to be divided equally between the ESR step and the charging/discharging ∆V. This percentage ripple will change, depending on the requirements of the application, and the equations provided below can easily be modified. RINGING DUE TO TOTAL INDUCTANCE (BOARD + CAP) 16e. Output Ripple Voltage Figure 16. SEPIC Converter Switching Waveforms 0.01• VO ID(PEAK) where: ⎛V +V ⎞ ⎛ χ⎞ ID(PEAK) = ⎜ 1+ ⎟ •IO(MAX) • ⎜ O D + 1⎟ ⎝ 2⎠ ⎝ VIN(MIN) ⎠ For the bulk C component, which also contributes 1% to the total ripple: COUT ≥ IO(MAX) 0.01• VO • f For many designs it is possible to choose a single capacitor type that satisfies both the ESR and bulk C requirements for the design. In certain demanding applications, however, the ripple voltage can be improved significantly by connecting two or more types of capacitors in parallel. For example, using a low ESR ceramic capacitor can minimize the ESR step, while an electrolytic or tantalum capacitor can be used to supply the required bulk C. Once the output capacitor ESR and bulk capacitance have been determined, the overall ripple voltage waveform should be verified on a dedicated PC board (see Board Layout section for more information on component placement). Lab breadboards generally suffer from excessive series inductance (due to inter-component wiring), and these parasitics can make the switching waveforms look significantly worse than they would be on a properly designed PC board. 1871xf For more information www.linear.com/LTC1871X 27 LTC1871X APPLICATIONS INFORMATION The output capacitor in a SEPIC regulator experiences high RMS ripple currents, as shown in Figure 16. The RMS output capacitor ripple current is: IRMS(COUT) =IO(MAX) • VO VIN(MIN) Note that the ripple current ratings from capacitor manufacturers are often based on only 2000 hours of life. This makes it advisable to further derate the capacitor or to choose a capacitor rated at a higher temperature than required. Several capacitors may also be placed in parallel to meet size or height requirements in the design. Manufacturers such as Nichicon, United Chemicon and Sanyo should be considered for high performance throughhole capacitors. The OS-CON semiconductor dielectric capacitor available from Sanyo has the lowest product of ESR and size of any aluminum electrolytic, at a somewhat higher price. In surface mount applications, multiple capacitors may have to be placed in parallel in order to meet the ESR or RMS current handling requirements of the application. Aluminum electrolytic and dry tantalum capacitors are both available in surface mount packages. In the case of tantalum, it is critical that the capacitors have been surge tested for use in switching power supplies. An excellent choice is AVX TPS series of surface mount tantalum. Also, ceramic capacitors are now available with extremely low ESR, ESL and high ripple current ratings. SEPIC Converter: Input Capacitor Selection The input capacitor of a SEPIC converter is less critical than the output capacitor due to the fact that an inductor is in series with the input and the input current waveform is triangular in shape. The input voltage source impedance determines the size of the input capacitor which is typically in the range of 10µF to 100µF. A low ESR capacitor is recommended, although it is not as critical as for the output capacitor. The RMS input capacitor ripple current for a SEPIC converter is: 1 IRMS(CIN) = • ΔIL 12 Please note that the input capacitor can see a very high surge current when a battery is suddenly connected to the input of the converter and solid tantalum capacitors can fail catastrophically under these conditions. Be sure to specify surge-tested capacitors! SEPIC Converter: Selecting the DC Coupling Capacitor The coupling capacitor C1 in Figure 15 sees nearly a rectangular current waveform as shown in Figure 16. During the switch off-time the current through C1 is IO(VO/VIN) while approximately – IO flows during the on-time. This current waveform creates a triangular ripple voltage on C1: ΔVC1(P−P) = IO(MAX) C1• f • VO VIN + VO + VD The maximum voltage on C1 is then: VC1(MAX) = VIN + ΔVC1(P−P) 2 which is typically close to VIN(MAX). The ripple current through C1 is: IRMS(C1) =IO(MAX) • VO + VD VIN(MIN) The value chosen for the DC coupling capacitor normally starts with the minimum value that will satisfy 1) the RMS current requirement and 2) the peak voltage requirement (typically close to VIN). Low ESR ceramic and tantalum capacitors work well here. SEPIC Converter Design Example The input voltage is 5V to 15V and the output is 12V at a maximum load current of 1.5A (2A peak). 1871xf 28 For more information www.linear.com/LTC1871X LTC1871X APPLICATIONS INFORMATION 1. The duty cycle range is: ⎛ VO + VD ⎞ D= ⎜ = 45.5% to 71.4% VIN + VO + VD ⎟⎠ ⎝ 2. The operating mode chosen is pulse skipping, so the MODE/SYNC pin is shorted to INTVCC. 3. The operating frequency is chosen to be 300kHz to reduce the size of the inductors; the resistor from the FREQ pin to ground is 80k. 4. An inductor ripple current of 40% is chosen, so the peak input current (which is also the minimum saturation current) is: V +V ⎛ χ⎞ IL1(PEAK) = ⎜ 1+ ⎟ •IO(MAX) • O D ⎝ 2⎠ VIN(MIN) 12 + 0.5 ⎛ 0.4 ⎞ = ⎜ 1+ = 4.5A • 1.5 • ⎟ ⎝ 2 ⎠ 5 The inductor ripple current is: DMAX 1– DMAX 0.714 = 0.4 • 1.5 • = 1.5A 1– 0.714 ΔIL = χ •IO(MAX) • And so the inductor value is: L= VIN(MIN) 2 • ΔIL • f • DMAX = 5 • 0.714 = 4µH 2 • 1.5 • 300k The component chosen is a BH Electronics BH5101007, which has a saturation current of 8A. 5. With an minimum input voltage of 5V, only logic-level power MOSFETs should be considered. Because the maximum duty cycle is 71.4%, the maximum SENSE pin threshold voltage is reduced from its low duty cycle typical value of 150mV to approximately 120mV. Assuming a MOSFET junction temperature of 125°C, the room temperature MOSFET RDS(ON) should be less than: VSENSE(MAX) 1 1 RDS(ON) ≤ • • IO(MAX) ⎛ χ⎞ ⎛V +V ⎞ ⎜⎝ 1+ 2 ⎟⎠ • ρT ⎜ O D ⎟ + 1 ⎝ VIN(MIN) ⎠ 0.12 1 1 = • • = 12.7mΩ 12.5 1.5 1.2 • 1.5 ⎛ ⎞ ⎜⎝ 5 ⎟⎠ + 1 For a SEPIC converter, the switch BVDSS rating must be greater than VIN(MAX) + VO, or 27V. This comes close to an IRF7811W, which is rated to 30V, and has a maximum room temperature RDS(ON) of 12mΩ at VGS = 4.5V. 6. The diode for this design must handle a maximum DC output current of 2A and be rated for a minimum reverse voltage of VIN + VOUT, or 27V. A 3A, 40V diode from International Rectifier (30BQ040) is chosen for its small size, relatively low forward drop and acceptable reverse leakage at high temp. 7. The output capacitor usually consists of a high valued bulk C connected in parallel with a lower valued, low ESR ceramic. Based on a maximum output ripple voltage of 1%, or 120mV, the bulk C needs to be greater than: IOUT(MAX) COUT ≥ = 0.01• VOUT • f 1.5A = 41µF 0.01• 12V • 300kHz The RMS ripple current rating for this capacitor needs to exceed: IRMS(COUT) ≥IO(MAX) • 1.5A • VO VIN(MIN) = 12V = 2.3A 5V 1871xf For more information www.linear.com/LTC1871X 29 LTC1871X APPLICATIONS INFORMATION To satisfy this high RMS current demand, two 47µF Kemet capacitors (T495X476K020AS) are required. As a result, the output ripple voltage is a low 50mV to 60mV. In parallel with these tantalums, two 10µF, low ESR (X5R) Taiyo Yuden ceramic capacitors (TMK432BJ106MM) are added for HF noise reduction. Check the output ripple with a single oscilloscope probe connected directly across the output capacitor terminals, where the HF switching currents flow. 8. The choice of an input capacitor for a SEPIC converter depends on the impedance of the source supply and the amount of input ripple the converter will safely tolerate. For this particular design and lab setup, a single 47µF Kemet tantalum capacitor (T495X476K020AS) is adequate. As with the output node, check the input ripple with a single oscilloscope probe connected across the input capacitor terminals. If any HF switching noise is observed it is a good idea to decouple the input with a low ESR, X5R ceramic capacitor as close to the VIN and GND pins as possible. 9. The DC coupling capacitor in a SEPIC converter is chosen based on its RMS current requirement and must be rated for a minimum voltage of VIN plus the AC ripple voltage. Start with the minimum value which satisfies the RMS current requirement and then check the ripple voltage to ensure that it doesn’t exceed the DC rating. IRMS(CI) ≥IO(MAX) • = 1.5A • VO + VD VIN(MIN) 12V + 0.5V = 2.4A 5V For this design a single 10µF, low ESR (X5R) Taiyo Yuden ceramic capacitor (TMK432BJ106MM) is adequate. 1871xf 30 For more information www.linear.com/LTC1871X LTC1871X PACKAGE DESCRIPTION Please refer to http://www.linear.com/product/LTC1871X#packaging for the most recent package drawings. MS Package 10-Lead Plastic MSOP (Reference LTC DWG # 05-08-1661 Rev F) 0.889 ±0.127 (.035 ±.005) 5.10 (.201) MIN 3.20 – 3.45 (.126 – .136) 3.00 ±0.102 (.118 ±.004) (NOTE 3) 0.50 0.305 ±0.038 (.0197) (.0120 ±.0015) BSC TYP RECOMMENDED SOLDER PAD LAYOUT 0.254 (.010) 3.00 ±0.102 (.118 ±.004) (NOTE 4) 4.90 ±0.152 (.193 ±.006) DETAIL “A” 0.497 ±0.076 (.0196 ±.003) REF 10 9 8 7 6 0° – 6° TYP GAUGE PLANE 1 2 3 4 5 0.53 ±0.152 (.021 ±.006) DETAIL “A” 0.18 (.007) SEATING PLANE 1.10 (.043) MAX 0.17 – 0.27 (.007 – .011) TYP 0.50 (.0197) NOTE: BSC 1. DIMENSIONS IN MILLIMETER/(INCH) 2. DRAWING NOT TO SCALE 3. DIMENSION DOES NOT INCLUDE MOLD FLASH, PROTRUSIONS OR GATE BURRS. MOLD FLASH, PROTRUSIONS OR GATE BURRS SHALL NOT EXCEED 0.152mm (.006") PER SIDE 4. DIMENSION DOES NOT INCLUDE INTERLEAD FLASH OR PROTRUSIONS. INTERLEAD FLASH OR PROTRUSIONS SHALL NOT EXCEED 0.152mm (.006") PER SIDE 5. LEAD COPLANARITY (BOTTOM OF LEADS AFTER FORMING) SHALL BE 0.102mm (.004") MAX 0.86 (.034) REF 0.1016 ±0.0508 (.004 ±.002) MSOP (MS) 0213 REV F 1871xf 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 itsinformation circuits as described herein will not infringe on existing patent rights. For more www.linear.com/LTC1871X 31 LTC1871X TYPICAL APPLICATION High Efficiency 5V Input, 12V Output Boost Converter (Bootstrapped) VIN 5V 100µF 6.3V L1 10µH RUN ITH INTVCC 33.2k IN GATE MODE/SYNC 4.7µF 10nF D1 SENSE LTC1871X V M1 FREQ 80.6k 110k FB 12.4k GND 1871x TA01 VOUT 12V 2A 47µF 25V ×8 GND L1: AGP2923 D1: VBT1045BP M1: IPC80N04S4 RELATED PARTS PART NUMBER DESCRIPTION COMMENTS LT®1619 Current Mode PWM Controller 300kHz Fixed Frequency, Boost, SEPIC, Flyback Topology LTC1624 Current Mode DC/DC Controller SO-8; 300kHz Operating Frequency; Buck, Boost, SEPIC Design; VIN Up to 36V LTC1700 No RSENSE Synchronous Step-Up Controller Up to 95% Efficiency, Operation as Low as 0.9V Input LTC1871-7 Wide Input Range Controller No RSENSE, 7V Gate Drive, Current Mode Control LTC1872 SOT-23 Boost Controller Delivers Up to 5A, 550kHz Fixed Frequency, Current Mode LT1930 1.2MHz, SOT-23 Boost Converter Up to 34V Output, 2.6V ≤ VIN ≤ 16V, Miniature Design LT1931 Inverting 1.2MHz, SOT-23 Converter Positive-to-Negative DC/DC Conversion, Miniature Design LTC3401/LTC3402 1A/2A 3MHz Synchronous Boost Converters Up to 97% Efficiency, Very Small Solution, 0.5V ≤ VIN ≤ 5V LTC3704 Positive-to-Negative DC/DC Controller No RSENSE, Current Mode Control, 50kHz to 1MHz LT3782 2-Phase Step-Up DC/DC Controller 6V ≤ VIN ≤ 40V; 4A Gate Drive, 150kHz to 500kHz 1871xf 32 Linear Technology Corporation 1630 McCarthy Blvd., Milpitas, CA 95035-7417 For more information www.linear.com/LTC1871X (408) 432-1900 ● FAX: (408) 434-0507 ● www.linear.com/LTC1871X LT 0117 • PRINTED IN USA LINEAR TECHNOLOGY CORPORATION 2017