Coax unbalanced by definition?

Coax Isn’t “Unbalanced Because Ground”

A follow-up to “50 Ω Coax — Balanced at Its Design Impedance, Unbalanced When It’s Not.”

Terminology note: In this article, “common-mode” is used in the common ham-radio sense as a catch-all for unwanted current on the outside of the coax shield caused by imbalance / mode conversion. In strict EMC terminology, common-mode is typically defined and measured with respect to a reference structure (often a chassis/reference plane) and frequently involves currents on multiple conductors relative to that reference. The practical issue we’re fixing here is better described as stray/unintended shield return current that recruits the environment as part of its closure path.

Coax is constantly described as unbalanced, usually with a quick explanation like “the shield is grounded, so it’s unbalanced by definition.” That sounds tidy, but it smuggles in an assumption that isn’t actually part of what coax is.

The better mental model is simple: balanced vs unbalanced is about symmetry and modes, not about whether a conductor is nicknamed “ground” on a schematic. When coax misbehaves in ham stations, the culprit is almost always common-mode current on the outside of the shield, not the existence (or non-existence) of “RF ground.”

Coax impedance is defined between two conductors

A coaxial cable has a characteristic impedance (50 Ω, 75 Ω, etc.). That impedance is defined between its two conductors:

• The center conductor
• The inside surface of the shield

Not to Earth. Not to a ground rod. Not to “the universe.” If you take a spool of coax floating in free space (no chassis bond, no station ground, nothing), it’s still approximately 50 Ω between its two terminals. The cable did not change personality because you removed a ground strap.

What people usually mean by “coax is unbalanced because the shield is ground” is this: most coax-fed equipment presents a single-ended port where the shield is used as the local reference, and that reference is often bonded to chassis and safety earth somewhere in the system. That’s a property of the interface and installation, not a fundamental property of the transmission line.

RF ground is not magic, but it isn’t meaningless

It’s true that dirt under your feet is often a poor RF conductor, and a literal “infinite RF sink” doesn’t exist. But it does not follow that “RF ground” is nonsense.

At RF, “ground” is best understood as a reference conductor or conductor network that provides a return path with some impedance (frequency-dependent, geometry-dependent, and installation-dependent).

• Earth is not automatically “RF ground.”
• A chassis, counterpoise, radial field, ground plane, tower, or large bonded metal surface can be an excellent RF reference.

Not magical does not mean not useful.

Balanced vs unbalanced is about symmetry and modes

A practical RF definition that survives real installations:

• Balanced behavior is when the desired differential mode has equal-and-opposite currents, and the geometry/environment does not strongly encourage conversion into common-mode.
• Unbalanced behavior is when the system intentionally or accidentally makes common-mode easy to excite (often because one conductor is tied into a larger conducting “world” like chassis, wiring, mast, tower, etc.).

Notice what’s missing: an absolute, mystical “ground.” The question is not “is the shield ground?” The question is “did we keep the wanted mode the wanted mode?”

The intended TEM mode inside coax is electrically differential

In the TEM mode coax is designed to carry, current on the center conductor is returned by an equal-and-opposite current on the inside surface of the shield. The fields are largely confined between those two conductors.

So if your definition of unbalanced is “one conductor is ground,” you’ve already lost the plot, because coax’s normal operation is not “signal to ground,” it’s “signal between two conductors.”

The real origin of the “unbalanced coax” story: the shield has an outside surface

Here’s the crucial RF fact: at RF, the shield has two distinct surfaces that can carry different currents. Due to skin effect, those currents are largely independent.

• Inside surface of the shield: participates in the intended TEM transmission line with the center conductor.
• Outside surface of the shield: a separate conductor relative to the environment, free to carry common-mode current.

That’s why you can have a perfectly valid differential signal inside the coax and still have RF current on the outside that causes all the classics: RF in the shack, hot mic cases, audio buzz, receive noise pickup, distorted antenna patterns, and “mystery SWR” changes because the feedline became part of the radiating structure.

Bonding the shield to chassis doesn’t “break” the TEM mode

Yes, bonding makes the equipment port single-ended in a practical sense: the shield becomes the local reference for the station hardware. But that does not mean the internal TEM transmission mode is suddenly “unbalanced.”

A voltage is always between two points. When we say “the signal is on the center conductor,” we’re using shorthand for: the signal is between center conductor and shield. Bonding the shield for safety, shielding, and EMC can be very beneficial and still leaves the internal coax mode behaving exactly as intended.

What actually creates the station problem: mode conversion into outside-of-shield current

The “unbalanced problem” happens when a transition or installation asymmetry converts some differential energy into common-mode, for example:

• Feeding a balanced antenna directly with coax without an effective current choke at the feedpoint
• Unequal coupling of antenna halves to the environment (nearby metal, mast, tower, gutters, wiring)
• Coax routing that “drops away” from one side of a dipole or runs close to metal on one side more than the other
• Tuner or station layouts that recruit feedline and house wiring as part of the return path

The key point stays clean: Unbalance is not a property of “having a shield tied to ground.” Unbalance is what happens when the installation drives common-mode on the outside of the shield.

How mismatch fits in without breaking causality

Mismatch does not inherently create common-mode. But mismatch often makes common-mode problems easier to trigger and easier to notice:

• Standing-wave peaks increase local E-field stress around connectors, tuners, and feedpoint hardware
• Asymmetries couple more strongly when fields are larger near parts of the system
• The outside-of-shield path gets more opportunity to be “recruited” as part of the antenna system

A good match reduces the opportunity for surprises; it doesn’t “philosophically balance” the cable.

Why the shield can be a return without “RF ground”

Return current does not require Earth. It requires a closed path. In differential TEM mode, the closed path is: center conductor → load → inner surface of shield → back to source.

Common-mode uses a different loop: outside of shield → coupling to the environment (house wiring, mast, tower, radials, Earth, operator, etc.) → back to the station/antenna system somehow. That’s the loop that feels “messy” because the environment is now part of the circuit and its impedance is complicated.

Practical rule that stays true in every argument

Coax is a differential line internally.
It only behaves “unbalanced” in the ham-problem sense when you allow (or accidentally drive) current on the outside of the shield.

Best practice follows directly:

• Place a current choke at the point where coax meets a structure that can excite common-mode (balanced antennas, end-fed arrangements, asymmetrical layouts).
• If needed, add a second choke at the station entry to keep any remaining common-mode off equipment and house wiring.

This approach doesn’t require debating whether “RF ground exists.” It just fixes the mode you don’t want.

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 help and feedback.