A switch-mode converter circuit uses a controlled power semiconductor switching technique along with an inductor, transformer, or capacitor as an energy-storage element to transfer dc power from its input to its output. In a basic switch-mode converter, a dc-to-pulse-width converter IC accepts a dc input and produces square waves applied to a power semiconductor switch (Fig. 1). The switched output from the power semiconductor is then rectified and filtered to provide a dc output. This circuit becomes a dc-dc voltage regulator by taking a portion of the dc output and feeding it back to the controller IC to make the circuit maintain a constant output voltage.
The major contributor to the efficiency of the switch-mode converter is the power semiconductor switch. Because they have a relatively high switching speed, MOSFETs are the power semiconductor switch of choice in power supplies. In the typical switch-mode power supply, the MOSFET switch applies power to a load when a control signal tells it to do so. The control signal also tells it to turn off.
Ideally, the power MOSFET switch should turn on and off in zero time. It should have an infinite impedance when turned off so zero current flows to the load. Additionally, it should have zero impedance when turned on so the on-state voltage is zero. Also ideally, the switch input should consume zero power when the control signal is applied. However, these idealistic characteristics are unachievable.
In the real world, actual power MOSFETs don’t meet these ideal characteristics. For example, Figure 2a shows a control signal applied to an ideal power MOSFET switch whose output exhibits zero transition time when turning on and off (Fig. 2b). When the MOSFET is off (not conducting current), power dissipation is very low because current is very low. When the MOSFET is on (conducting maximum current), power dissipation is low because its conducting resistance is low.
In contrast, an actual power MOSFET switch exhibits some delay when turning on and off (Fig. 2c). Therefore, most of the power dissipation occurs when the switch goes through the linear region between on and off, which depends on switching speed. A small amount of power also is dissipated when the MOSFET is on and when it is off.
Ideally, you would like the MOSFET to switch at as high a frequency as possible so the output filter can use small inductors and capacitors to provide very low output ripple. Unfortunately, the MOSFET’s power losses begin to increase at some frequency where it affects the power supply’s efficiency. Therefore, there is a practical limit on the MOSFET’s switching frequency, which depends on the characteristics of the specific device.
One disadvantage of the switch-mode converter is switching noise and output voltage ripple caused by the switching process. Specific control techniques and careful component selection can minimize these noises.
Isolated vs. Non-Isolated
In terms of their response to a dc input, there are two types of dc-dc converters: isolated and non-isolated, which depends on whether there is a direct dc path from the input to the output. An isolated converter employs a transformer to provide isolation between the input and output voltage. In the non-isolated converter, there is a dc path from input to output.
For most applications, non-isolated converters are appropriate. However, some applications require isolation between the input and output voltages, which requires a switching transformer. An advantage of the transformer-based converter is that it has the ability to easily produce multiple output voltages, whereas the inductor-based converter provides only one output.
The only other possible choice as a power source is the linear low-dropout regulator (LDO). Low dropout refers to the difference between the input and output voltages that allows the IC to regulate the output voltage. That is, the LDO device regulates the output voltage until its input and output approach each other at the dropout voltage. Ideally, the dropout voltage should be as low as possible to minimize power dissipation and maximize efficiency.
LDO voltage regulators are linear devices with the topology shown in Figure 3. The main components are the power semiconductor and a differential error amplifier. One input of the differential amplifier monitors a percentage of the output as determined by the resistor ratio of R1 and R2. The second input to the differential amplifier is from a stable voltage reference (VREF). If the output voltage tends to rise relative to VREF, the drive to the power semiconductor changes to maintain a constant output voltage.
These ICs have simpler circuits than their switch-mode cousins and produce less noise (no switching) but are limited by their current-handling and power dissipation capability. Some LDO ICs are specified as low as 50 mA whereas others can handle up to about 1 A. Linear regulators only step down, and their output-to-input efficiency can be on the order of 50%. In contrast, switch-mode converters can step up, step down, and invert at efficiencies in the 80% to 95% range.
A switch-mode converter varies its dc output current in response to load changes. One widely used approach, pulse-width modulation (PWM), controls the power switch output power by varying its on and off times (Fig. 4). The ratio of on time to the switching period time is the duty cycle. The higher the duty cycle, the higher the power output from the power semiconductor switch.
To generate the PWM signal, the error amp accepts the output voltage feedback and a stable voltage reference to produce an output related to the difference of the two inputs. The PWM comparator compares the error amp’s output voltage with the ramp (sawtooth) from the oscillator, producing a modulated pulse width. The comparator output is applied to the switching logic, whose output goes to the output driver for the power-supply circuit. The switching logic provides the capability to enable or disable the PWM signal applied to the power-supply circuit.
Most PWM controller ICs provide current-limiting protection by sensing the output current. If the current sense input exceeds a specific threshold, it terminates the present cycle (cycle-by-cycle current limit). PWM circuits are available as standalone ICs as well as integrated within dc-dc switch-mode converter ICs with internal power switches.
The TPS40210 and TPS40211current-mode PWM controllers from Texas Instruments are wide-input voltage (4.5 V to 52 V), non-synchronous boost controller ICs (Fig. 5). They suit topologies that require a grounded source N-channel FET including boost, flyback, SEPIC, and various LED driver applications.
Features include programmable soft start, overcurrent protection with automatic retry, and programmable oscillator frequency. Their fixed-frequency current-mode control provides improved transient response and simplified loop compensation. The main difference between the two parts is the reference voltage to which its error amplifier regulates the FB pin. Additional features include:
• Internal slope compensation
• Integrated low-side driver
• Programmable closed-loop soft start
• External synchronization capabilities
• Reference 700-mV (TPS40210), 263-mV (TPS40211)
• Low-current disable function
• Hysteretic converter
The basic hysteretic regulator shown in Figure 6 consists of a comparator with input hysteresis that compares the output feedback voltage with a reference voltage. When the feedback voltage exceeds the reference voltage, the comparator output goes low, turning off the buck switch MOSFET. The switch remains off until the feedback voltage falls below the reference hysteresis voltage. Then, the comparator output goes high, turning on the switch and allowing the output voltage to rise again.
There is no voltage-error amplifier in the hysteretic converter, so its response to any change in the load current or the input voltage is virtually instantaneous. Therefore, the hysteretic regulator represents the fastest possible dc-dc converter control technique. A disadvantage of the conventional hysteretic regulator is that its frequency varies proportionally with its output capacitor’s equivalent series resistance (ESR). Since the initial value is often poorly controlled, and the ESR of electrolytic capacitors also changes with temperature and age, practical ESR variations can easily lead to a frequency variation on the order of one to three. Synchronous Rectifiers
A synchronous rectifier provides output rectification in a switch-mode power supply using two transistor switches to emulate the operation of a diode rectifier that allows current to pass in one direction and not in the other (Fig. 7). Transistors in the synchronous rectifier are usually MOSFETs that are driven 180° out of phase—that is, one turns on and the other turns off at the same time.
Because it is more efficient, the MOSFET-based synchronous rectifier has replaced Schottky diodes in many applications. Plus, the MOSFETs usually dissipate less power than the Schottky diode. From a circuit standpoint, a synchronous rectifier is more complicated and usually costs more than the Schottky rectifier. But at the system level, the synchronous rectifier competes favorably on cost because of thermal and packaging considerations. Thus, a cost comparison must be application-dependent.
There must be a slight delay between the operation of the two switches so both switches are never on at the same time, which is called “shoot-through ” or “cross-conduction.” If both switches turn on at the same time, there would be a short circuit across the input source, which could destroy the circuit. There are three ways to prevent this problem:
• Insert a “dead” time between turn-on and turn-off
• Employ a shoot-through prevention circuit
• Ensure that the potential shoot-through is kept very short and the MOSFETs can absorb any possible overload
MOSFET gate charge is an important parameter because the amount of current required to fully enhance the MOSFET channel directly influences circuit efficiency. To optimize efficiency, most synchronous rectifiers employ externally controlled gate-drive ICs. The external gate drivers must supply a current capable of charging the MOSFET’s input capacitance. This charging current determines the charging rate of the MOSFET’s input capacitance, which controls turn-on, turn-off, and propagation delay times. Synchronous rectifiers are usually integrated within power-supply ICs.
Multiphase Converter ICs
The trend toward higher-current, lower-voltage microprocessors has created a need to supply 100 A or more at voltages in the neighborhood of 1 V. The multiphase converter answers this need. Multiphase converters employ two or more identical, interleaved converters connected so their output is a summation of the outputs of the cells (Fig. 8).
To understand the advantages of the multiphase converter, we must first look at the shortcomings of single-phase converters relative to supplying high current and low voltage. With a conventional single-phase converter, the output ripple and dynamic response improve with increased operating frequency. In addition, the physical size and value of the output inductor and capacitor also reduce at higher frequencies. Unfortunately, after the frequency reaches a certain limit, converter switching losses increase and lower the converter’s efficiency. This forces a design tradeoff between operating frequency and efficiency.
To overcome these single-phase frequency limitations, the multiphase cells operate at a common frequency, but are phase shifted so conversion switching occurs at regular intervals controlled by a common control chip. The control chip staggers the switching time of each converter so the phase angle between each converter switching is 360º/n, where n is the number of converter phases. The outputs of the converters are paralleled so the effective output ripple frequency is n × f, where f is the operating frequency of each converter. This provides better dynamic performance and significantly less decoupling capacitance than a single-phase system.
Current sharing among the cells is necessary so one cell doesn’t hog too much current. Ideally, each multiphase cell should consume the same amount of current. To achieve equal current sharing, the output current for each cell must be monitored and controlled.
The multiphase approach also offers packaging advantages. Each converter delivers 1/n of the total output power, reducing the physical size and value of the magnetics employed in each phase. Also, the power semiconductors in each phase only need to handle 1/n of the total power. This spreads the internal power dissipation over multiple power devices, eliminating the concentrated heat sources and possibly the need for a heatsink. Additional advantages include:
• Reduced rms current in the output filter capacitor, which allows the use of smaller and less expensive types
• Higher total power capability
• Increased equivalent frequency without increased switching losses, which permits the use of smaller equivalent inductances that shorten load transient time
Multiphase converters also have some disadvantages that should be considered when choosing the number of phases:
• The need for more switches and output inductors than in a single-phase design, which leads to a higher system cost than a single-phase solution, at least below a certain power level
• More complex control
• The possibility of uneven current sharing among the phases
• Added circuit layout complexity
As current requirements increase, so does the need for increasing the number of phases in the converter. ICs providing just two phases may not be adequate because of their limited output current range. An optimum design requires tradeoffs between the number of phases, current per phase, switching frequency, cost, size, and efficiency. Higher output current and lower voltage require tighter output voltage regulation.
To evaluate multiphase design decisions, review the approaches relative to available ICs. One approach is to use a multiphase IC with an integrated PWM controller and MOSFET drivers, in addition to external MOSFETs. With this technique, heat and noise generated by the on-chip gate drivers affect controller performance. It also is impractical to cascade most of these types of chips for additional phases. Accurate current sharing is difficult. And, three phases appear to be the limit
Another approach is to employ a separate multiphase controller with an integral PWM circuit and external gate drivers and MOSFETs. With this method, the PWM controller is isolated from the heat and noise of the gate drivers. Because the current-sense signal is routed to the controller, current sharing is more complex. There are additional controller-to-driver delays because of the separated ICs as well.
Yet another approach is to use a dual-phase controller with integrated gate drivers and built-in synchronization and current sharing. This method only allows an even number of phases. Although it simplifies the design, it may result in unused or redundant silicon, pins, and external components. Finally, driver heat and noise generated on-chip can degrade controller performance.
Therefore, existing topologies may not provide the freedom required in selecting the number of phases. The ideal solution is a scalable topology that enables the easy addition or removal of any multiphase cell without sacrificing performance. This scalable approach must be able to share current equally among the distributed phase cells. Such a technique also minimizes parasitics and eases board layout.
Intersil’s ISL6313 two-phase buck PWM multiphase controller, which can be used to power microprocessors, integrates power MOSFET drivers into the controller IC (Fig. 9). It includes advanced control loop features for optimal transient response to load apply and removal. Also, it provides accurate, fully differential, continuous dc resistance (DCR) current sensing for load line programming and channel current balance. Its Active Pulse Positioning (APP) Modulation and Adaptive Phase Alignment (APA) allow quicker initial response to high di/dt load transients, reducing the number of output bulk capacitors, also reducing cost.
The ISL6313’s user programmable current-sense resistors only require a single external resistor to set their values. No external current-sense resistors are required. Furthermore, the device’s dynamic voltage identification (VID) compensation pin lets designers add optimizing compensation for well-controlled dynamic VID response.
Protection features include overvoltage, undervoltage, and overcurrent. Furthermore, the ISL6313 includes protection against an open circuit on the remote sensing inputs. Combined, these features protect the associated microprocessor and power system. Additional features include:
• Two-phase or one-phase operation with internal drivers
• ±0.5% system accuracy overtemperature
• Adjustable reference-voltage offset
• Accurate load line programming
• Precision channel current balancing
• Variable gate-drive bias: 5 to 12 V
• Intel VR11 mode of operation
• AMD mode of operation
• Microprocessor voltage identification inputs
• Dynamic VID technology
• Dynamic VID compensation
• Selectable operation frequency up to 1.5 MHz per phase
• Lead-free (RoHS compliance)