Access Official Technical PDF

Share Summary:LinkedInEmail

In industrial and embedded equipment, point-of-load power conversion must fit within electrical, thermal, and board-layout constraints. The MIC2168 synchronous buck controller is designed to support compact step-down converters using external N-channel MOSFETs. It operates from a 3V to 14.5V input supply, regulates output voltages down to 0.8V, and switches at a nominal 1 MHz. Its control and protection features can simplify a power-stage design, but component selection and layout remain central to reliable operation.

How the MIC2168 synchronous buck controller regulates output

The MIC2168 uses voltage-mode control. A feedback divider scales the output voltage and feeds it to the controller’s error amplifier, where it is compared with a 0.8V reference. Changes in the output caused by input variation or a changing load shift the error-amplifier output. The PWM comparator uses that signal alongside an oscillator ramp to adjust the high-side MOSFET on-time.

For example, when a load step pulls the output voltage down, the feedback signal falls relative to the reference. The control loop responds by increasing the high-side switch’s on-time, transferring more energy through the inductor to the output. As the output recovers, the loop adjusts switching duty to maintain regulation.

The IC’s nominal 1 MHz switching frequency allows designers to consider a compact power stage, while the external inductor, MOSFETs, capacitors, and compensation network determine much of the converter’s behavior. The specified maximum duty cycle is 90%, and the stated minimum on-time is 30ns typical, with a 60ns maximum under the specified condition. These limits should be checked against the intended input and output operating points.

Current limiting without a separate sense resistor

Rather than inserting a dedicated current-sense resistor in the main power path, the MIC2168 detects voltage across the high-side MOSFET during its on-time. This approach can reduce component count and avoid the additional conduction loss associated with a series shunt. The controller’s current-limit input uses an external resistor to establish the limit, and the datasheet gives a typical internal current-sink value of 200µA for this setting.

MOSFET on-resistance changes with temperature, so this sensing method is less precise than a separate low-value sense resistor. The device uses a positive temperature coefficient intended to track the MOSFET’s increasing RDS(ON), but designers still need to account for device variation and operating temperature. The datasheet recommends adding a 50% margin to the load current when calculating the current-limit resistor to reduce the risk of false limiting as the MOSFET heats.

For dependable sensing, the current-limit resistor should connect directly to the drain of the high-side MOSFET, with the switch-sense resistor connected at its source as shown in the recommended circuit. A 0.1µF capacitor in parallel with the current-limit resistor is also recommended to filter switching noise. These connections help keep the sensed signal representative of MOSFET voltage drop rather than unwanted board-level interference.

Soft-start and fault response

At startup, the internal soft-start sequence controls the rise of the output voltage. This limits input surge current by gradually increasing the converter’s duty cycle rather than applying full power immediately. The COMP/EN pin has three roles: it provides the connection for external loop compensation, participates in soft-start, and can disable the IC when pulled low. The datasheet specifies a typical shutdown current of 50µA when this pin is held at 0V.

Protection behavior depends on the type of fault. For a hard short, when the feedback voltage is below 0.67V, the controller discharges the compensation capacitor, resets its digital counter, and shuts off the high-side gate drive before restarting the soft-start cycle. This repeated startup behavior is called hiccup-mode protection. Overvoltage and undervoltage feedback thresholds are specified at approximately 3% above and below the reference, respectively.

Choosing the power-stage components

MOSFETs and gate drive

The controller drives external N-channel MOSFETs in a synchronous buck arrangement. Its internal regulator supplies a nominal 5V gate-drive rail when the input is above 5V; below 5V, the regulator operates in dropout. Accordingly, MOSFET selection must reflect the available gate voltage: the datasheet calls for low-threshold devices suitable for a 2.5V gate drive at low input voltages, and logic-level devices specified at a 4.5V gate drive when the input is above 5V.

Compare MOSFET voltage rating, on-resistance, and total gate charge. Higher on-resistance increases conduction losses, while gate charge contributes to the energy required to switch the devices. Because MOSFET on-resistance rises with temperature, calculations should use a realistic hot operating value rather than relying only on a room-temperature specification. The datasheet also recommends allowing a 20% margin above the input voltage when selecting MOSFET drain-to-source voltage ratings to accommodate switching spikes.

Inductor and capacitors

Inductor selection involves balancing ripple current, component size, cost, and losses. Higher ripple increases current stress and can raise dissipation in the inductor and MOSFETs; reducing ripple generally requires a larger inductance. As a design starting point, the datasheet recommends setting peak-to-peak inductor ripple to about 20% of maximum output current. Core loss and winding resistance also matter: winding resistance affects efficiency at higher load, while core losses may be more significant at lower output current.

Output capacitor ESR influences both output ripple and feedback-loop behavior. The power-stage LC network introduces a two-pole response, while capacitor ESR adds a zero that affects the phase response. Compensation components at COMP/EN must be selected with the output inductor and capacitors as a system; the datasheet illustrates how lower ESR can require more phase boost from the error amplifier. Input capacitors, meanwhile, must meet voltage and ripple-current requirements and withstand startup inrush conditions appropriate to their technology.

Layout is part of the control design

Fast switching edges make physical connections important. The datasheet recommends placing the current-limit resistor close to its MOSFET connections, keeping feedback resistors near FB, and routing the feedback trace away from the switch node. Compensation components should sit close to COMP/EN and return directly to GND. Place input bulk capacitance near the high-side MOSFET drain, and position the recommended ceramic bypass capacitors close to VIN and VDD.

The VSW connection should run directly to the high-side MOSFET source through a 4Ω to 10Ω resistor to filter the switch node. Keep high-current paths short and use suitable copper connections between the MOSFETs, input capacitor, output capacitor, and ground. These measures support cleaner sensing and reduce the likelihood that switching noise will disturb control signals.

Practical implications for power-system design

The MIC2168 combines voltage-mode PWM control, external synchronous MOSFET drive, MOSFET-based current sensing, soft-start, and fault protection in a 10-lead MSOP package. For engineers integrating it into a point-of-load supply, the main design task is coordinating the control loop with real component characteristics: MOSFET gate drive and temperature, inductor ripple and losses, capacitor ESR, current-limit margin, and PCB parasitics. Applying the datasheet’s component-selection and layout guidance provides a foundation for a compact converter, while final designs still need validation across their intended operating conditions.

Access Official Technical PDF

About The Author