Conducted Susceptibility
How Interference Couples Onto Cables
Conducted susceptibility addresses the reality that most real-world interference never has to penetrate a metal enclosure. It rides in on the wires. Power feeds, control harnesses, and antenna runs act as efficient conduits that carry external RF energy straight to printed circuit boards and integrated circuits inside. Because cables behave electrically short relative to a wavelength below roughly 30 MHz, conduction is the dominant entry mechanism across the low and mid HF range, which is exactly why the standardized CS tests concentrate their energy there.
The injected energy is almost always common-mode, meaning equal current flows on every conductor in a bundle and returns through the chassis or ground reference. A victim circuit only upsets when that common-mode current converts to a differential voltage across a sensitive node, through finite common-mode rejection, ground impedance, or unbalanced loading. This is the key insight for designers: raising the conducted susceptibility threshold is mostly about preventing common-mode to differential conversion, not about brute-force attenuation alone.
The pass or fail criterion is the susceptibility threshold, the lowest interference level that produces a defined indication of upset such as a bit error, a relay chatter, or a measurement excursion beyond tolerance. Qualification compares that threshold against the platform limit curve. Aircraft and shipboard equipment see the harshest limits, with CS114 currents reaching 109 dBuA, while a benign commercial bench instrument may only need to survive a few volts of injected RF under IEC 61000-4-6.
Governing Relationships
Margin (dB) = Vthreshold (dB) − Vlimit (dB) → pass when ≥ 0 (typically ≥ 6 dB)
CS114 injected current from probe power:
Icm ≈ √(Pfwd / 50) (probe calibrated in a 50 Ω jig)
Current limit in dBµA:
IdBµA = 20 × log10(IµA) → 109 dBµA ≈ 0.28 A
Common-mode to differential conversion:
Vdiff ≈ Icm × Zimbalance / CMRR
Where Pfwd = forward power into the probe, Icm = common-mode cable current, Zimbalance = unbalanced source/load impedance, and CMRR = circuit common-mode rejection ratio. Example: 1 W into a CS114 probe produces ≈ 141 mA (≈ 103 dBµA).
Standard Test Methods Compared
| Standard / Test | Coupling Method | Frequency Range | Typical Level | Primary Use |
|---|---|---|---|---|
| MIL-STD-461 CS101 | Series voltage on power leads | 30 Hz to 150 kHz | 1 to 5 V rms | Power-line ripple immunity |
| MIL-STD-461 CS114 | Bulk current injection probe | 10 kHz to 200 MHz | 77 to 109 dBµA | Cable-bundle RF immunity |
| MIL-STD-461 CS115 | Cable impulse (BCI) | 30 ns pulse, 30 pps | 5 A peak (default) | Fast transient / ESD-class |
| MIL-STD-461 CS116 | Damped sinusoid | 10 kHz to 100 MHz | 5 to 10 A peak | Ringing transient immunity |
| IEC 61000-4-6 | CDN / EM clamp | 150 kHz to 80 MHz | 1, 3, or 10 V rms | Commercial conducted RF |
| ISO 11452-4 | BCI (substitution) | 1 MHz to 400 MHz | 30 to 200 mA | Automotive harness immunity |
Frequently Asked Questions
What is the difference between conducted susceptibility and radiated susceptibility?
Conducted susceptibility tests interference that reaches a device through its cables, injected as a voltage or current on the leads, while radiated susceptibility tests interference arriving as a field that illuminates the enclosure. They overlap in frequency: CS114 runs 10 kHz to 200 MHz, and RS103 covers 2 MHz to 18 GHz (extendable to 40 GHz by procurement). Below about 30 MHz cables are electrically short and conduction dominates; above that, cables act as antennas. A complete design needs both cable filtering and enclosure shielding.
How does CS114 bulk current injection work and what levels are typical?
CS114 clamps a current probe around a cable bundle to induce a calibrated RF current without cutting the harness. The probe is first calibrated in a 50 Ω jig to find the forward power for the target current, then that power drives the equipment cable. Limits run from about 77 dBµA on benign platforms to 109 dBµA (≈ 0.28 A) on severe ones, often with 6 dB of qualification margin, swept from 10 kHz to 200 MHz. Because it couples common-mode current, results depend heavily on shield termination quality.
What design techniques improve conducted susceptibility immunity?
Attack the coupling path first. Feedthrough filters, common-mode chokes, and ferrite beads roll off injected energy, with pi filters giving 40 to 60 dB of power-lead attenuation across the band. A 360 degree shield termination at the connector backshell keeps shield current off the conductors, and a low-impedance chassis bond holds the reference quiet. Differential signaling with strong common-mode rejection, transient protection, and good IC decoupling raise the upset threshold above the limit by 6 dB or more.