Good-Ark — discrete power semiconductors: rectifiers, Schottky, TVS, MOSFETs, SiC and IGBT
Request a Quote

MOSFET Gate Ringing: Causes, Scope Clues, and First Fixes

The gate trace shows several oscillations after turn-on. Before ordering a larger resistor or rejecting the MOSFET, shorten the probe connection and…

MOSFET Gate Ringing: Causes, Scope Clues, and First Fixes

The gate trace shows several oscillations after turn-on. Before ordering a larger resistor or rejecting the MOSFET, shorten the probe connection and repeat the measurement. If the displayed ringing changes substantially, part of the problem may be the measurement loop.

For an engineer or buyer reviewing an alternative device, that first distinction matters. A transistor comparison based on different probing, layout, or operating conditions can blame the component for a change introduced elsewhere.

MOSFET gate ringing: where should troubleshooting start?

Start MOSFET gate-ringing diagnosis by verifying the gate-to-source measurement, then inspect gate-loop inductance, driver impedance, source return, and drain-to-gate coupling. Tune resistance only after those checks. Confirm that reduced oscillation still meets switching-loss, timing, and gate-voltage requirements across the operating range; a quieter trace alone does not qualify the circuit.

This sequence separates three possibilities: a probe artifact, an actual gate-loop oscillation, and voltage induced by the surrounding switching stage. More than one can be present at the same time.

First branch: is the waveform measured correctly?

Measure between gate and the relevant source connection at the device. A driver output measured against a distant ground can differ from the actual Vgs because current and inductance affect the return path. Long probe leads can also create their own resonant behavior.

Use the probe arrangement appropriate to the voltage, bandwidth, and common-mode conditions. The Tektronix floating-measurement guide explains why an earth-referenced ground clip cannot be attached indiscriminately to floating power nodes. Never defeat protective earth to make that connection.

Keep the measurement position and setup with each capture. A changed probe arrangement is a changed experiment, not evidence that a new component fixed the circuit.

Goodark AGMN01004R LFPAK5060 MOSFET package
Goodark AGMN01004R LFPAK5060 MOSFET package.

Second branch: identify the resonant loop

Gate capacitances and loop inductance can form an oscillatory network. Driver output resistance, external resistance, and transistor internal gate resistance contribute damping. Capacitances vary with voltage, so a fixed LC model is an approximation rather than a complete transistor model.

For an illustrative ideal model, assume L = 20 nH and C = 2 nF. The resonance estimate 1 / (2π√LC) is approximately 25 MHz. Increasing that assumed inductance fourfold would halve the estimated frequency. These figures are hypothetical and are not extracted from a measured board.

TI’s external gate-resistor guide discusses gate-loop parasitics and damping. Use its method with the actual circuit rather than copying a worked resistor value.

Third branch: inspect the source and power-loop coupling

Common source inductance couples power-current changes into the gate reference. A Kelvin source connection, where provided and properly routed, can help separate the driver return from the high-current path. It must be used according to the selected package and layout guidance.

Drain-voltage transitions can also inject current through gate-drain capacitance. Inspect the off-state gate during the opposite switch’s transition in a half bridge. Do not assume every excursion is the same mechanism as turn-on gate-loop ringing.

The TI gate-driver fundamentals report explains gate charge and switching behavior. The important review is the complete loop and reference, including driver decoupling and the physical return path.

Choose the first change from the evidence

Observation and a useful next check
Observation Next check Tradeoff to retain
Ringing changes with probe lead Improve and validate measurement arrangement Keep comparable captures
Actual gate-loop oscillation Shorten loop and evaluate series damping Transition time and loss
Off-state gate rises during drain transition Review Miller current, sink path and source reference Driver and clamp availability
Clean driver output but poor device Vgs Inspect interconnect and source return Physical layout and reference voltage

A split source/sink driver can allow different turn-on and turn-off paths. The TI UCC27511/UCC27512 document provides one example. Any revised connection needs a fresh check of timing, drive current, and the relevant output-state behavior.

Measure the tradeoff after changing resistance

Increase or decrease one parameter at a time where practical. Observe Vgs peak and undershoot, drain voltage, relevant current, transition times, and temperature. A slower edge can reduce one oscillation while increasing switching overlap and heat.

Compare the same supply, load, board, and temperature conditions. Include startup and representative operating corners before accepting the result. If the chosen value works only at one bench condition, the investigation remains incomplete.

For an apparently damaged part, begin with MOSFET multimeter checks after safe discharge. A static pass does not validate the device’s dynamic switching behavior.

Two qualitative damped oscillations illustrate that gate-loop and probing conditions affect the observed waveform; these are not oscilloscope captures.
Two qualitative damped oscillations illustrate that gate-loop and probing conditions affect the observed waveform; these are not oscilloscope captures.

Avoid substitutions that change several unknowns

A new MOSFET can change gate charge, internal resistance, capacitances, and package inductance. A driver substitution can change source and sink impedance. Either change may require retuning even when the nominal voltage and current ratings appear similar.

Keep exact part identities and source revisions in the comparison. The MOSFET versus BJT switching guide addresses a broader technology choice; it cannot establish identical gate-drive behavior between two MOSFETs.

Keep clamps and system protection in their own review

A gate clamp or active Miller clamp serves a different purpose from a passive bias resistor or series damping resistor. Check its operating state and dynamic behavior. Do not assume a protection symbol removes every gate excursion.

The TVS protection guide covers transient protection, while the overview of PV inverter semiconductor operation provides wider system context. In a power stage, local waveforms, control timing, and fault handling still need their own evidence.

Close the investigation with a reproducible result

Record the original symptom, verified probing method, suspected mechanism, change made, and resulting waveforms and temperatures. Include the final resistor paths, source return, and layout. The result should explain why the revised stage meets its limits, rather than merely reporting that the trace looks smoother.

Keep unresolved observations visible. If a waveform changes with load direction or only appears after warm-up, reproduce that state before closing the review. A component purchase based on an incomplete bench comparison can preserve the original problem while introducing new timing or thermal conditions that have not been evaluated.

FAQ

Is every oscillation on a gate trace real gate ringing?

No. A probe loop or an inappropriate reference can introduce or exaggerate oscillation. Verify a suitably rated gate-to-source measurement at the device and compare the result with a lower-inductance arrangement. Actual gate-loop resonance and switching-stage coupling can also coexist with measurement artifacts. Preserve the probe position and setup before comparing resistor values or deciding that a transistor substitution changed the circuit behavior.

Will a larger gate resistor always improve the design?

It can improve damping in a particular loop, but it also changes drive current, transition time, and switching loss. Excess resistance can increase heat or disrupt timing. First verify the measurement and layout, then evaluate resistance while observing gate limits, drain behavior, current, and temperature. A reduction in visible ringing is useful evidence, but it is not the only condition needed to approve the circuit.

Does a Kelvin source pin eliminate all ringing?

No. A properly used Kelvin connection can reduce the influence of shared source-path inductance on the drive reference, but gate-loop inductance, capacitance, driver behavior, drain coupling, and measurement effects still matter. Follow the selected package and layout guidance, then verify the actual Vgs and switching waveforms. A Kelvin pin left unused or routed poorly does not automatically provide the intended benefit.

Copyright Suzhou Good-Ark Electronics Co., Ltd. All Rights Reserved