How to simplify synchronous buck-boost converter design?

Author: Timothy Hegarty (System Engineer, Analog Products Group, Texas Instruments Silicon Valley)

One of the most notable variations in power converter specifications is the need to convert a wide range of input voltages to a regulated output voltage. [1] However, this task becomes more challenging when the unregulated input voltage varies between above, below, or equal to the regulated output voltage set point, requiring a buck-boost conversion.

Buck-boost conversion is essential for a wide range of applications, including battery charging, solid-state lighting, industrial computing, and automotive applications.[2]

This article briefly reviews the many factors involved in designing a 4-switch buck-boost converter. In particular, it answers questions about component selection and power calculation, as well as using the Quick Start Calculator tool [3] to streamline and accelerate the converter design process.

Synchronous Buck-Boost Converter Operation

As an effective method to provide both step-up conversion and step-down conversion, a properly designed buck-boost circuit becomes an indispensable device due to its convenience. Let's review the 4-switch (non-inverting) synchronous buck-boost topology shown in Figure 1.

The main advantage of the buck-boost power stage is that the buck, boost, and buck-boost conversion modes can be used as needed to achieve high efficiency over a wide input voltage and load current range. It also provides a positive output voltage compared to the similar single-switch (inverting) buck-boost, as well as lower power losses and higher power density relative to SEPIC, flyback, and cascaded boost-buck topologies.

Figure 1. 4-switch synchronous buck-boost converter power stage.

In Figure 1, four power MOSFETs are arranged as buck and boost legs in an H-bridge configuration, with switch nodes SW1 and SW2 connected by an inductor LF. When the input voltage is higher or lower than the output voltage, respectively, the synchronous buck or boost begins to operate, and the high-side MOSFET of the opposite non-switching leg operates as a conducting device. More importantly, when the input voltage approaches the output voltage, the switching buck or boost leg reaches the desired duty cycle limit, triggering the transition to buck-boost mode of operation. The change in operating mode should be smooth and autonomous, without changing the control configuration. How this is achieved and the possible interdependencies between the power stage and the control mechanism are very important.

Design Flow for Current-Mode Buck-Boost Converters

A complete schematic of a 4-switch synchronous buck-boost converter is shown in Figure 2. This schematic includes components for the power stage, bootstrap circuit for the gate driver, current sensing network, spread spectrum frequency modulation (SSFM) for lower electromagnetic interference (EMI),[5] programmable undervoltage lockout (UVLO), output feedback, and loop compensation.

Figure 2. Schematic diagram of a 4-switch buck-boost converter with a current-mode controller.

A quick-start tool resource [3] provides an analysis and design framework for a 4-switch buck-boost converter. The process proceeds from converter specification to component selection, to performance verification (efficiency, component dissipation, and Bode plots), followed by iterative design if necessary. Using the LM5175 synchronous buck-boost controller as a starting point, let’s review the design process step by step for a 400kHz converter that provides a 12V output at a rated current of 6A from a 6V to 42V input source.

Step 1: Run Tech Specs

The screenshot in Figure 3 shows Step 1, or the user specification entry for input voltage range, output voltage, load current, and switching frequency.

Step 2: Inductor Screening

The inductance depends on the input voltage range and the target peak-to-peak inductor ripple current ratio. Equation 1 sets the target ripple current ratio in the deep boost and deep buck operating points at 30% and 80%, respectively.

There are three main parameters that characterize inductor performance—resistance (DCR), saturation current ( ), and core losses. Inductors with iron powder core material offer outstanding performance at switching frequencies up to 400kHz, making them a mainstream solution in many applications. A noteworthy and desirable characteristic is that the inductance decreases gradually with increasing current. At the same time, inductors with ferrite cores have relatively low core losses, although they prevent the inductance from dropping sharply at the onset of saturation.

Figure 3. Steps 1 through 3 refer to running specifications, inductor screening, and current sensing, respectively. This circuit schematic is automatically assembled based on the entered and calculated component values.

Step 3: Shunt Resistors

The shunt resistor is set based on the relevant threshold for current limit. For example, Equation 2 applies to the LM5175 and specifies an 80mV valley threshold in the buck and a 160mV peak threshold in the boost. The shunt power dissipation peaks at the lowest input voltage when the boost duty cycle is at its maximum. A wide aspect ratio shunt resistor, such as a 1225 package size resistor, helps place the component in the PCB layout[5] close to the source connections of the two low-side MOSFETs.

Next, slope compensation takes the sensed signal and adds a ramp component equal to the inductor ramp-up in buck mode or a ramp component equal to the inductor ramp-down in boost mode. The calculation of the ramp capacitor is given in Equation 3 [4]

Steps 4 and 5: Input and Output Capacitor Screening

In Figure 4, steps 4 and 5 refer to the input and output capacitor values ​​set by the buck and boost modes of operation, respectively. High-density designs increasingly combine several X5R- or X7R-dielectric ceramic components, sometimes with a small electrolytic capacitor for bulk energy storage. Equation 4 uses the peak-to-peak ripple voltage to set the baseline capacitance estimate assuming no equivalent series resistance (ESR) ripple component.

Then, after the capacitor values ​​are chosen, the respective peak-to-peak ripple voltages are back-calculated knowing the ESR.

The input capacitor RMS current (and ripple voltage) is maximum during buck mode at a duty cycle of 50%. On the other hand, the highest output capacitor RMS current occurs during boost mode when the duty cycle is maximum. The expression for the RMS current is

Figure 4. Steps 4 to 7 refer to capacitor selection, compensator design, and Bode plot analysis.

Step 6: Soft-Start, Dithering, and Undervoltage Lockout (UVLO)

Based on the startup time specification, the required soft-start capacitor value is

The next option is to select the dither capacitor value to set the spread spectrum modulation frequency using Equation 8 [5], where Gd is the controller-dependent conductance.

The undervoltage lockout resistors set the rising and falling input voltage thresholds for converter startup and shutdown, respectively. The upper UVLO resistor value is selected to set the hysteresis. Then, if is the upper UVLO comparator threshold, the corresponding lower UVLO resistor value ends up being [4]

Step 7: Loop Compensation

It is important to note that the crossover frequency in boost mode tends to be lower due to the reduced current mode modulator gain (proportional to ). Indeed, a quick inspection of the Bode plot at minimum input voltage clearly shows whether the compensator zero helps achieve sufficient phase near the crossover frequency.

Step 8: Efficiency Prediction

Step 8 shown in Figure 5 provides a graph of efficiency and component power dissipation versus line and load.

Equations 12 and 13 calculate the conduction, switching, and gate drive losses in buck and boost modes, respectively. The corresponding expression for buck-boost mode is a weighted combination of equations 12 and 13, based on the operating point in the buck-boost window, and dividing the frequency by 2.

As expected, inductor copper and core losses, switch dead-time conduction losses, shunt losses, and bias regulator losses also contribute to the efficiency calculation. If the losses are considered overall, a 4-switch buck-boost converter with a 12V regulated output can achieve efficiencies above 96% over a wide range of output currents and input voltages.

Figure 5. Step 8 refers to MOSFET specifications, efficiency graphs, and power loss analysis.

Summarize

Buck-boost converters for industrial and automotive applications have unique power solution requirements. Proving their ease of use, high efficiency, small size, and low overall bill of materials cost, 4-switch synchronous buck-boost converters offer a combination of advantages to meet the key functions required. Given the component interrelationships and functional trade-offs involved, a quick start calculator is a handy tool to speed up and simplify converter design.