Noise Isn’t “E or H”… It’s Coupling

When your waterfall fills with hash, your audio squeals on transmit, your USB devices glitch, or your neighbor’s speakers “talk back” on your CQ, it’s tempting to argue: “It’s E-field!” ... “No, it’s H-field!”

But for practical troubleshooting, the better mental model is this: interference only happens when three things line up ...

• Source ... something generating energy you don’t want (SMPS, LED drivers, solar inverters, Ethernet gear, arcing hardware, your own transmitter, etc.)
• Coupling path ... how that energy gets from the source to somewhere else (this is the whole game)
• Victim ... something that responds to it (receiver front end, audio chain, PC/USB, a neighbor’s device, even your own station wiring)

No path, no problem ... even if the source is loud. (That’s why “E vs H” arguments often stall: they describe a mechanism, not the root cause.)

E-field vs H-field is really “capacitive vs inductive coupling”

In the near field (close to the source compared with wavelength), it can be useful to separate coupling into two familiar buckets.

E-field coupling ... capacitive coupling (driven by dV/dt)

Two conductors separated by air (or insulation) form a tiny capacitance. If one conductor has a fast-changing voltage, a displacement current flows through that capacitance into the other.

Ham example: a switch-mode supply has a noisy “hot node” with sharp edges ... it capacitively injects RF current into nearby wiring ... that wiring becomes an antenna and radiates into your receiver.

How you reduce E-field coupling:

• Reduce capacitance ... increase distance, reduce parallel area
• Reduce dv/dt ... better supply design, snubbers, slower edges
• Use electrostatic shielding ... a grounded conductive barrier in the right place

H-field coupling ... inductive coupling (driven by dI/dt)

Any loop carrying current creates a magnetic field. If there’s another loop nearby, changing flux through that loop induces a voltage.

Ham example: transmit current (or a noisy current loop inside a device) magnetically couples into a nearby audio cable loop ... you hear a raspy buzz that changes with RF power.

How you reduce H-field coupling:

• Reduce loop area ... twist pairs, keep send/return tightly together
• Reduce di/dt where possible
• Re-route for geometry ... distance and orientation are huge
• At HF, “magnetic shielding” is usually not the win ... geometry is

Near field vs far field ... why “E or H” often misleads in a shack

Very close to a source, E and H can behave “separately” depending on the structure. Farther out, they travel together as an electromagnetic wave.

That’s why the better question is usually: “What’s the coupling path?” ... not “Is it E or H?”

The coupling paths that matter in a ham station

Almost every RFI/EMI problem in amateur radio is dominated by one (or more) of these paths:

• Conducted coupling ... noise traveling on wires (AC mains, DC leads, control cables, USB/audio/Ethernet). Often it becomes radiated noise once a cable acts like an antenna.
• Common-impedance coupling ... shared “ground” straps, chassis paths, DC negatives. Noise current creates a voltage on that shared impedance, and the victim experiences it as “noise on ground.”
• Capacitive coupling ... high-voltage or fast-edge nodes injecting current into nearby conductors.
• Inductive coupling ... current loops coupling into other loops (layout and routing problems).
• Radiated coupling ... far-field pickup, plus “accidental antennas” like house wiring and long leads.

Common-mode is the ham-radio supervillain

If there’s one idea that explains most station chaos, it’s the difference between differential mode (DM) and common mode (CM).

• DM ... equal and opposite currents on a pair (what you want in coax or twisted pair)
• CM ... same-direction current on both conductors relative to some outside reference (what makes cables radiate and pick up noise)

Common-mode current on the outside of coax, mic leads, USB cables, power leads, or Ethernet turns them into antennas that both: radiate (causing interference) and receive (importing noise).

The big ham insight: a lot of “RF in the shack” isn’t your antenna “radiating too much” ... it’s your station wiring becoming part of the antenna system because CM current found a path. That’s coupling.

Typical ham problems, translated into coupling language

“My HF noise floor is S7 at home, but S1 in the field”

• Source ... local electronics (SMPS, LEDs, networking gear, solar inverter)
• Path ... conducted noise onto house wiring + radiation from wiring ... and/or direct radiation
• Victim ... your antenna system and receiver front end

Coupling-first fixes:

• Identify the source ... breaker tests, portable RX “sniffer,” and controlled on/off experiments
• Reduce conducted paths ... better supplies, line filtering where appropriate
• Reduce pickup ... feedline CM control (good choking at the right places), bonding, routing

“When I transmit on 40m my computer reboots / my audio glitches”

• Source ... your transmitter
• Path ... RF coupling onto USB/audio/power cables (often CM), plus shared-impedance returns
• Victim ... the computer or interface

Coupling-first fixes:

• Add CM impedance where the cable enters the victim ... ferrites, multiple turns, correct placement
• Reduce shared-impedance returns ... bond smartly, shorten “mystery” ground paths, keep RF off DC negative loops
• Suppress CM at the antenna/feedline first ... it’s often the biggest lever

“My neighbor’s speakers/TV get my voice”

• Source ... your transmitter (and sometimes harmonics/IM if something is wrong)
• Path ... CM currents and unintended radiation ... sometimes conducted on mains
• Victim ... consumer gear with weak filtering or nonlinear inputs

Coupling-first fixes:

• Make sure your station is clean ... harmonics under control, appropriate filtering if needed
• Kill common-mode at the source ... feedline choking, bonding, routing
• If needed and with permission ... add suppression at the victim (ferrites on speaker leads, etc.)

A coupling-first troubleshooting checklist

You don’t need fancy gear ... you need controlled experiments that force a change and reveal the path.

• Name the victim ... what exactly misbehaves? Receiver noise floor, audio chain, USB, SWR readings, neighbor gear?
• Decide RX noise vs TX RFI ... RX noise means you’re the victim; TX RFI means you’re the source.
• Force a change to expose the path ... unplug one cable, move a lead 30 cm, add a choke, change band, change power.
• Break the path ... choke CM currents, reduce loop area, add entry filtering, improve bonding, re-route cables, use galvanic decoupling where it matters.

The punchline

Saying “noise isn’t E or H” doesn’t mean fields aren’t real. It means fields don’t create interference by themselves.

Interference happens when your station (or your house wiring) accidentally provides a coupling path ... often a common-mode path ... from the source to something sensitive.

Interested in more technical content? Subscribe to our updates for deep-dive RF articles and lab notes.

Questions or experiences to share? Feel free to contact RF.Guru for RF support and practical troubleshooting.