DEBUGMISSION 023 · Measurements · Probe artifactsFoundational

The Scope Invented the Ringing

A 20 MHz clock appears to overshoot badly enough to trigger a board redesign. Before adding a terminator, compare probe configurations and decide which waveform is evidence about the PCB, and which is evidence about the measurement loop.

Requirements

Clock · 20 MHzEdge · 2 nsLogic rail · 3.3 VQualified overshoot · ≤0.50 VProbe loading · ≤15 pF
  1. 01Build a trustworthy measurement
  2. 02Qualify the comparison
Circuit overview

Supplied comparison capture

Clock captures with long ground lead and ground springThe long ground lead capture has large overshoot and ringing. The ground spring capture settles near 3.3 volts with modest overshoot.Same 20 MHz clock node · same acquisition settings18 cm ground lead · 4.75 V peakGround spring · 3.55 V peak5 V2.5 V0 V3.3 V0 ns25 ns50 ns75 ns100 nsCompare the probe loop before judging the board waveform.

The two authored captures use the same 10× mode, local pads and acquisition settings; only the return attachment differs.

Circuit and case data
Circuit and case data
Measurement loop
Oscilloscope probe return geometryA probe tip and its return form a measurement loop. The selected ground and reference arrangement changes the loop area.CLK PADREMOTE RETURN1X PROBE
Displayed peak6.45 VDisplayed overshoot3.15 VProbe capacitance100 pFGround inductance110 nH
Technique note

Compare the supplied configurations at the same node and acquisition settings. Configuration peaks and 110 nH / 5 nH loop-inductance values are authored comparison parameters, not a field-solved probe model. The supplied 10× probe and oscilloscope have been checked for sufficient bandwidth and correct probe compensation for this 2 ns-edge exercise. Real measurements must also verify probe compensation, probe/scope bandwidth, loading, ground safety, and repeatability.

Evidence notebook 0 saved

No evidence saved yet. Measure, inspect the result, then save useful observations.

Current task
Current task

Step 1 of 2

Build a trustworthy measurement

0 evaluated decisions · Draft and evidence saved locally
Technical references