EMI/EMC Academy ~30 min Interactive

Measurement Technique & the Diagnostic Toolkit

Good technique beats good equipment. The same probe in trained hands localizes a source in minutes; in untrained hands it produces noise. This module is the how: probe motion, current-probe use, and the delta method that underpins all debugging.

Module progress0%
Learning objectives. Use H- vs E-field probes correctly; scan for hotspots and read relative amplitude; measure cable current with a clamp; run a disciplined before/after (delta) measurement; and recall which detector (Peak/Quasi-Peak/Average) applies where.

In the near field, the electric and magnetic fields are not in the fixed 377 Ω ratio of a plane wave, so you probe them separately. An H-field loop couples to current — hold it flat over a trace so the changing magnetic field threads the loop. An E-field stub couples to voltage — bring its tip near a hot node, heatsink, or connector pin.

PCB (ground plane / traces) high-di/dt trace H-field loop flat over current hot node (V) E-field stub tip near voltage
Orientation is everything: the loop must face the current, the stub must approach the voltage.
Common mistake: chasing absolute numbers with a near-field probe. Coupling depends on distance and angle, so a 6 dB change from moving your hand means nothing. Keep the probe fixed and change the board.

Pick what you're hunting and the recommended probe/technique appears.

Because bench numbers are relative, the unit of progress is the delta. The discipline: capture a reference trace, use max-hold so intermittent peaks are caught, change exactly one thing, and overlay the new trace on the reference. The difference at the peak of interest is your result.

  1. Freeze the setup — probe position, cable dress, EUT mode all fixed.
  2. Capture reference (save trace / screenshot with markers).
  3. Apply one change (add ferrite, short a stitch, fit a choke).
  4. Re-measure without moving anything else; read Δ at the marker.
  5. Keep or revert, then move to the next hypothesis.
Change one variable at a time. If you add a choke and reroute a cable at once, a 10 dB improvement tells you nothing about which change did it.

Limits are usually specified with Quasi-Peak and Average detectors. On the bench, debug with Peak (fastest, always ≥ QP) — if Peak is under the limit, QP certainly is. Only switch to QP/Average near the end when you need a realistic margin estimate. This is covered in depth in the RE measurement module.

Detector Reads Use on bench
Peak Highest instantaneous level Primary debug — fast, conservative
Quasi-Peak Weighted by pulse repetition Late-stage margin check
Average Mean level Narrowband/CW near limit
Key takeaways. Probe by field type, read relative not absolute, hunt cables with a current clamp, and prove every fix with a single-variable delta. Next we learn to read what the spectrum is telling us.
← Previous Next: Reading the Spectrum →