Common Mode and Differential Mode Interference

When you are chasing down interference problems in your ham shack, there is one distinction that matters more than almost any other: is the noise traveling as common-mode current or differential-mode current? These two types of current behave differently, travel by different paths, cause different symptoms, and need entirely different cures. Getting this wrong is why so many operators spend money on the wrong fix and still have noise problems afterward.

Think of it this way. Every cable that enters your radio has two conductors. The signal you care about travels as a difference between those two conductors — one conductor carries current one way while the other carries it the opposite way. That is differential mode, and it is how every signal is supposed to travel. But there is a second way current can flow: the same direction on both conductors simultaneously. That is common mode, and it is almost always an unwanted intruder. Understanding this difference is the master key to solving feedline noise problems, shack RFI, and the frustrating situation where your noise level changes when you touch the coax.

The Two Fundamental Current Modes

Every two-conductor cable — coaxial cable, audio cable, USB cable, DC power lead — can carry current in two fundamentally different ways. Understanding this at a deep level is the foundation of all RFI work.

Differential mode is the wanted signal mode. In differential-mode operation, current flows down one conductor and returns on the other in the opposite direction. The two currents are equal in magnitude and opposite in direction. On a coaxial cable, the signal current flows inside the center conductor and the return current flows on the inside surface of the braid. On a balanced line such as ladder line, the current in one wire is equal and exactly opposite to the current in the other wire. Because the currents are equal and opposite, the magnetic fields they produce cancel each other outside the cable. The cable does not radiate, and external fields are not coupled into it. This is exactly the behavior you want from a transmission line.

Common mode is the unwanted intruder mode. In common-mode operation, current flows in the same direction on both conductors simultaneously. On a coaxial cable, common-mode current flows on the outside surface of the braid — not inside the cable where the transmission line mode lives. Because the currents do not cancel, the cable radiates like an antenna, and external fields couple into it equally. A coaxial cable carrying common-mode current is, in every electrical sense, an antenna connected directly to your receiver input or transmitter output.

The critical insight is this: the inside of the coaxial cable and the outside of the braid are two electrically separate surfaces at RF frequencies. Skin effect at HF frequencies means that current on the inside surface of the braid does not see the outside surface and vice versa. So the transmission line mode (differential) and the antenna mode (common) coexist on the same physical cable but do not interact. This separation is why a common-mode choke can block common-mode current without disturbing the transmission line signal — it impedes current on the outside of the braid while leaving the inside untouched.

On a balanced line such as 450-ohm ladder line or open-wire feeders, the same distinction applies but the geometry is different. The differential mode consists of equal and opposite currents in the two wires — current up one wire and down the other. The common mode consists of current flowing in the same direction on both wires. A balanced line is designed to operate in differential mode; any common-mode current is an indication of imbalance in the system.

Differential-mode current (bottom) flows equally and oppositely on both conductors — the desired signal mode. Common-mode current (top) flows in the same direction on both conductors — the antenna mode that causes interference.

View Larger

A critical point that surprises many operators: a normal LC filter treats both modes the same. A low-pass filter on a coaxial feed line will attenuate common-mode and differential-mode currents equally, which means it can make the interference worse on one while helping the other. Only a component that acts differentially on the two modes — specifically, a common-mode choke — can selectively impede common mode while leaving the differential-mode transmission line signal intact.

How Common-Mode Current Arises on Feedlines

Common-mode current on a feedline does not arrive by accident. There are several predictable causes, and understanding each one helps you identify where to apply the cure.

Antenna imbalance at the feedpoint. A dipole antenna is a balanced device — each half of the dipole should be fed with equal and opposite currents. When you feed it with coaxial cable, which is an unbalanced line, you have a potential mismatch. At the feedpoint, the cable shield is connected to one arm of the dipole. If there is no balun to prevent it, RF current from that dipole arm can flow back down the outside of the shield rather than returning through the center conductor. This is the most common cause of common-mode current on a dipole feedline.

The degree of imbalance depends on the antenna geometry. A perfectly symmetric dipole oriented horizontally with the feedline dropping straight down perpendicular to the antenna should have minimal common-mode current because of the natural symmetry. But in practice, antenna installations are rarely perfectly symmetric. The feedline runs off at an angle, one arm is shorter than the other, or nearby structures are not equally placed on both sides. Any asymmetry drives common-mode current onto the feedline.

Off-center fed antennas and end-fed wires. These antennas are inherently unbalanced — the feedpoint is not at a symmetry point of the antenna. An off-center fed dipole (OCFD) has intentional imbalance. An end-fed half-wave (EFHW) antenna presents a very high impedance at the feedpoint and typically relies on the coax shield as a counterpoise. In both cases, significant common-mode current on the feedline is expected, and a common-mode choke is essential.

Feedline running along the antenna. Even with a symmetric dipole and a perfect balun at the feedpoint, common-mode current can be induced further along the feedline if the coax runs parallel to the antenna for any distance. The antenna's near field induces current on the outside of the coax. This is particularly common on small lots where the feedline has to run under or near the antenna before it can drop toward the shack. An additional choke placed where the feedline leaves the proximity of the antenna can cure this problem.

External RF fields. Any strong RF field — from a nearby AM broadcast tower, a neighboring ham's transmitter, or a local RFI source — will induce common-mode current on any conductor in the field, including your feedline. The feedline acts as a receive antenna for these external signals, delivering them directly to your receiver input. This is why you can have a completely quiet feedline on a hill in the country but still pick up RFI if someone near your shack has a switching power supply.

Ground loops between different ground references. If your radio chassis is grounded to the station safety ground, and the antenna is connected to a ground rod at a different part of the building, and those two grounds are at different RF potentials (which they almost certainly are), then a ground loop exists. Common-mode current will flow through the loop in response to any RF in the vicinity. This ground loop current flows through the coax shield, through the radio, and through the ground conductors completing the loop.

Why Common-Mode Current Is Harmful

Common-mode current on a feedline is harmful in two distinct directions: on transmit, it causes RF to radiate from the feedline; on receive, it allows RF to be received on the feedline. Both directions cause serious problems.

On transmit. When your transmitter drives power into the feedline, the desired result is that all of that power arrives at the antenna and is radiated from the antenna according to its design pattern. If common-mode current is present, a portion of that power flows on the outside of the coax shield and is radiated by the feedline. The feedline is physically much closer to the shack than the antenna, so this radiated RF reaches the operating position directly. This is the cause of RF burns on microphone elements, click and distortion in audio from computers and headsets when you key up, interference to nearby electronic devices, and the seatbelt-buckle shock that some operators experience when transmitting.

Common-mode radiation from the feedline also distorts the antenna's radiation pattern. The feedline acts as an additional radiating element with its own gain and pattern, and the combined pattern of antenna plus feedline-as-antenna is unpredictable and often has deep nulls in directions where you want to communicate.

On receive. The feedline acts as a receive antenna for any RF in its vicinity — RFI sources in the shack, the neighbor's switching power supply, a nearby LED lamp, the power line transformer at the end of the street. All of this noise couples onto the outside of the feedline as common-mode current and arrives at the receiver input added to the antenna signal. You cannot tell the receiver which current came from the antenna and which came from the feedline as a noise pickup path; it sees all currents at its input indiscriminately.

This is the mechanism behind the classic symptom: operator disconnects the antenna, and the noise floor barely changes. The noise is not coming through the antenna from outside — it is being injected into the receive chain by the feedline acting as a local noise antenna. Adding a longer feedline can make this worse because there is more cable surface area in the noise environment.

Common-mode problems can be location-specific in a counterintuitive way. An operator living in a very quiet rural area may still have terrible noise levels if the feedline runs near the computer desk, the power supply, or any other local noise source. Moving the feedline even a foot away from the power supply can produce a 10 dB improvement. Adding a choke at the shack entry can produce a 20 or 30 dB improvement.

Measuring and Identifying Common-Mode Current

Before applying any cure, it is worth confirming that you actually have a common-mode problem and locating where along the feedline the current is highest. Several simple tests can confirm a common-mode problem without specialized equipment.

The disconnect test. Disconnect the coax from the radio (or antenna tuner) and terminate it into a 50-ohm dummy load. Then listen to the radio on an indoor antenna or with no antenna at all. If the noise level changes significantly when you connect versus disconnect the coax — even with a dummy load on the far end — you have common-mode current on the outside of the feedline delivering noise to the chassis and receiving system of the radio. The differential-mode (inside the coax) path is terminated into the dummy load and should not change with connection.

The station ground test. On your operating radio with the antenna connected, try lifting the station ground connection — disconnect the chassis strap or ground rod lead briefly. If the noise level changes significantly, you have a ground loop contributing common-mode noise. This test should only be done briefly for diagnostic purposes; operating without a safety ground is not recommended.

RF current probe. A split-core current transformer — often called an RF current probe or sniffer — can clip around the coaxial cable without breaking the connection. It measures the total RF current flowing on the cable. The transmission line mode carries equal and opposite currents that cancel in the probe, so the probe reads only the common-mode current. Clip the probe at several points along the feedline — at the antenna feedpoint, halfway down the feedline, and at the shack entry. The highest reading identifies where the common-mode current enters or is highest. Adding a choke at that location will have the most effect.

SDR-based comparison. If you have an SDR receiver with a small omnidirectional antenna, you can place it in the shack and observe the noise floor. Then connect your main antenna feedline to your radio and observe whether the noise floor on the SDR changes. If it does, your feedline is radiating common-mode noise into the shack environment. Alternatively, compare your noise floor on the main receiver with the feedline connected versus disconnected — a significant change indicates feedline pickup.

A 1:1 current balun installed at the feedpoint presents high impedance to common-mode current on the outside of the coax braid while leaving the differential-mode signal inside the coax completely unaffected.

View Larger

Differential-Mode Interference

Differential-mode interference is conceptually different from common-mode. Where common-mode current flows on the outside of cables as a collective phenomenon involving both conductors equally, differential-mode interference travels between the two conductors as a normal signal would — only it is an unwanted signal rather than a wanted one.

Every cable entering your radio station is a potential differential-mode interference path. The DC power cable running from your 13.8-volt power supply to the radio, the USB cable connecting the computer to the radio for digital modes, the audio cables between the radio and computer sound card, and the control cables for CAT interface — all of these carry differential-mode voltages between their conductors, and any noise on those conductors arrives at the radio as a legitimate-looking differential signal.

Switching power supply noise. The most common source of differential-mode interference in modern shacks is the switching power supply. A switch-mode power supply (SMPS) operates by switching transistors on and off at frequencies from 50 kHz to several hundred kilohertz. This high-frequency switching generates harmonics well into the HF spectrum. The hash appears on the DC output leads as a differential-mode voltage between the positive and negative output conductors. When this hash reaches the radio through the power cable, it enters the radio's internal circuits directly — bypassing the antenna system entirely. This is why adding ferrite chokes to the antenna feedline does nothing to cure power supply hash: the noise is on a completely different path.

Computer noise on USB and audio cables. Computers generate substantial RF noise from the processor clock and its harmonics, the display, the hard drive, and many other sources. This noise couples onto USB cables, audio cables, and control cables through capacitive and inductive coupling inside the computer. It then travels down these cables to your radio as differential-mode interference. You may notice that the noise level on the radio changes when you move the USB cable or change which USB port you use — that is differential-mode noise from the computer changing based on path.

Ground loops and differential mode. A ground loop produces both common-mode and differential-mode noise depending on its geometry. When two pieces of equipment are connected by a signal cable and also connected to earth ground at two different points, any voltage difference between the two ground points (caused by RF currents in the building structure, by 60 Hz currents, or by other interference) drives a differential-mode current through the signal cable. This is why audio hum (60 Hz and its harmonics) often appears in receivers connected to computers — the two chassis are at different potentials and the hum drives a differential-mode current through the audio cable.

The cure for differential-mode interference is different from common-mode: bypass capacitors placed directly between the two conductors (across the signal path), series inductors in both conductors, or LC pi-section filters on the DC power leads. A ferrite choke applied to a cable — wound as a common-mode choke — does not address differential-mode noise because the opposing currents cancel in the core and produce no choking action.

The Four RFI Entry Paths Into a Receiver

When you are systematically chasing interference in your station, it helps to think of the receiver as having exactly four potential interference entry paths. Each path can carry either common-mode or differential-mode interference, and each needs its own specific treatment.

A complete noise solution typically requires addressing all four paths. Curing the antenna port but ignoring the power cable will leave the noise from the power supply entering through a different door. This systematic approach — identify each path, test each path, treat each path — is what separates effective RFI troubleshooting from random component changes that produce unpredictable results.

One path that surprises many operators is the power cable. Because the power cable enters through the rear of the radio next to the antenna connector, it is easy to assume they are the same problem. But they are not. A common-mode choke on the antenna feedline has no effect on differential-mode noise entering through the power connector. You need a separate filter on the DC power leads. The most effective approach is a pi-section LC filter with bypass capacitors on both the input and output, and a series inductor in both the positive and negative leads.

The Balun vs Unun Distinction

No discussion of common-mode interference would be complete without clarifying the often-confused relationship between baluns, ununs, and common-mode chokes. These terms are used loosely in the amateur community, and using the wrong device for your application can leave a common-mode problem completely uncured.

A balun (balanced-to-unbalanced) is a device that connects an unbalanced transmission line (coaxial cable) to a balanced load (a dipole antenna). Its purpose includes two things: providing impedance transformation if needed, and suppressing common-mode current on the feedline. A properly designed 1:1 current balun — also called a choke balun — does both. It presents high impedance to common-mode currents flowing on the outside of the coax while passing the differential-mode signal unchanged inside. This is the most important device for eliminating feedline radiation from dipole antennas.

A voltage balun works differently from a current balun. It forces equal voltages at its output terminals but does not directly impede common-mode current. The 4:1 voltage balun often sold for use with center-fed dipoles provides impedance transformation from 200 ohms to 50 ohms, but it is a poor common-mode choke. Many operators install a 4:1 voltage balun and find their common-mode problems persist — the balun did not address the root cause.

An unun (unbalanced-to-unbalanced) is an impedance transformer for use between two unbalanced circuits. A 9:1 unun is commonly used at the feedpoint of an end-fed half-wave antenna to transform the high antenna feedpoint impedance down to 50 ohms. An unun does not inherently address common-mode current at all. For an EFHW antenna, common-mode current on the feedline is an expected consequence of the antenna's inherent asymmetry, and a separate common-mode choke is needed in addition to the unun if you want to suppress feedline radiation.

The 1:1 current balun, the common-mode choke, and the choke balun are all names for the same device: a coaxial cable wound through a ferrite toroid core (or formed as a coil of coax) so that the ferrite presents high impedance to any current that tries to flow on the outside of the braid, without affecting what is inside the coax. This is the correct device for addressing feedline common-mode problems at the antenna feedpoint.

Prevention and Cure

With a solid understanding of the two modes and their origins, the solutions become logical rather than arbitrary. Each cure targets a specific mechanism.

1:1 current balun (choke) at the feedpoint. This is the single most effective measure for a dipole or vertical antenna. Installed at the antenna feedpoint, it presents high impedance to any common-mode current attempting to flow from the antenna back down the outside of the feedline. For a dipole, a well-designed choke balun at the feedpoint typically reduces feedline common-mode current by 20 to 40 dB. It should be the first device installed on any new feedline.

Additional choke at the shack entry. Even with a choke at the feedpoint, the feedline itself can pick up common-mode current from external sources as it runs from the antenna to the shack. A second choke installed at the point where the feedline enters the building provides a second barrier. It prevents any common-mode current that was induced on the feedline during its run from entering the shack. The two-choke approach — one at the feedpoint, one at the shack entry — is considered best practice for clean operation.

Power supply filtering for differential mode. An LC pi-section filter on the DC power leads addresses differential-mode noise from the switching power supply. Commercial versions are available (often sold as power line filters or RF chokes for DC lines), and effective homebrew versions can be made with a toroid inductor and bypass capacitors. The filter goes on the power supply output, as close to the radio power connector as practical.

Break ground loops with a single-point ground or RF choke. Ground loops driving differential-mode noise through signal cables are best eliminated by ensuring all equipment chassis are at the same RF potential — a single-point star ground. Alternatively, floating one end of a balanced audio cable via an audio isolation transformer breaks the ground loop path. For RF-induced common mode on audio cables, a ferrite choke wound as a common-mode choke on the audio cable is effective.

Star ground topology. In a properly designed station, all equipment is grounded to a single common point called the station ground bus. This eliminates the potential differences between chassis that drive ground loop currents. Every piece of equipment has a short, wide copper strap running directly to the ground bus rather than being connected to the next piece of equipment in a daisy chain. Star grounding is the structural solution that prevents ground loop problems from arising in the first place.