Why Coax Length Matters Less at VHF and Above—Except When It Does
Why Coax Length Matters Less at VHF and Above—Except When It Does
At VHF and UHF, random feed-line length is usually a poor tuning control. That does not repeal transmission-line physics: mismatch still transforms, phase still accumulates and cable loss becomes expensive.
On HF, amateurs regularly cut coax until the tuner or radio displays a prettier number. At VHF and above, that habit becomes less attractive because cable attenuation, connector quality and installation repeatability consume the margin quickly. But “length matters less” is an operating rule, not a new law of nature.
What Length Actually Changes
For a uniform transmission line with characteristic impedance Z0, propagation constant γ, physical length l and load ZL, the input impedance is:
Zin = Z0(ZL + Z0 tanh γl) / (Z0 + ZL tanh γl)
For an ideal lossless line, γ becomes jβ and tanh γl becomes j tan βl. A mismatched load therefore presents different complex impedances at different distances along the line. This remains true at HF, VHF, UHF and microwave frequencies.
There is one important special case. If the load equals the line impedance, its reflection coefficient is zero and a uniform line presents Z0 at every point. “Resonant,” however, does not necessarily mean 50 Ω. An antenna can be resonant with a resistance far from 50 Ω, and a nominal 50 Ω antenna can move away from that value after mounting, weather, nearby metal or a different measurement plane.
The useful distinction: changing cable length can change the impedance displayed at the station without changing the antenna feedpoint. A better station reading does not prove that the antenna radiates more efficiently.
Why the Trick Becomes Less Useful
Many purpose-built VHF and UHF antennas are designed for a reasonably narrow operating range and an impedance close to the cable system. When the installed return loss is already good, the reflection coefficient is small, so moving the reference plane around a low-loss line produces only a modest impedance excursion. There is little reason to cut cable as a matching network.
The line is also electrically long very quickly. At 145 MHz, a free-space quarter wavelength is about 0.517 m. The physical quarter wavelength inside coax is that value multiplied by the cable’s velocity factor: about 0.34 m for solid-polyethylene coax with a velocity factor near 0.66, or about 0.43 m for a foam cable specified near 0.84. A practical mast run may contain many electrical half-waves, and a small change in connector, routing or dielectric state can move the phase appreciably.
That makes “trim until the SWR is low” fragile. It uses the feed line as an undocumented transformer, moves the symptom to the station plane and can stop working after the cable or antenna changes. If a deliberate matching section is needed, design it from measured complex impedances, the actual velocity factor and the intended bandwidth.
Loss Can Hide a Mismatch, Not Repair It
A load reflection returning through a lossy line is attenuated on the outward and return journeys. The station can therefore display a better return loss than exists at the antenna. That does not mean the mismatch disappeared. Some forward power and reflected power were dissipated in the cable, connectors and transitions before the instrument saw them.
At higher frequencies, conductor and dielectric loss generally increase. The actual figure belongs to a named cable, temperature, frequency, connector set and installation—not to the label “RG-213” or “low loss” alone. For example, Belden’s current MRG2132 data gives nominal attenuation values for that specific construction, while Times Microwave publishes a different table and an 84% velocity factor for LMR-400. Those manufacturer values are starting points; water ingress, tight bends, poor terminations and ageing can make the installed assembly worse.
Convert every cable-table number to the actual one-way run and include connectors. A 3 dB one-way loss delivers only about half the incident power to a well-matched antenna and also raises the receive-system noise contribution ahead of any masthead amplifier. With mismatch, calculate or measure accepted power and cable heating rather than subtracting a single matched-line loss number.
Length Still Matters Wherever Phase Matters
There are plenty of VHF-and-above systems where electrical length is not incidental:
- Phased arrays and power dividers: relative phase and amplitude at each element depend on the complete cable assemblies, connectors and temperature.
- Quarter-wave transformers and stubs: the intended impedance or filter action comes directly from electrical length and characteristic impedance.
- Duplexer, cavity and filter interconnects: changing line length can rotate the impedance presented to a resonator and alter the tuned system.
- Masthead LNAs and receive converters: line loss before the first low-noise gain stage directly degrades system sensitivity; stability can also depend on source and load phase.
- High-power amplifiers: a reflected wave rotated by the feed line can present different voltage and current stress at the amplifier connector even when the load VSWR magnitude is unchanged.
- Time, delay and ranging systems: propagation delay and phase stability are the measurement, not a side effect.
In these cases, specify electrical delay or phase at frequency, not merely tape-measure length. Use the cable assembly’s measured S-parameters where the phase and loss budget justify it.
The Reference Plane Decides What the Number Means
An antenna analyser connected in the shack measures the network at the shack end of the feed line. It does not magically report the bare antenna feedpoint. A calibrated VNA can place its reference plane at the cable input, and port extension or de-embedding can move that plane only when delay, loss and characteristic impedance are known well enough.
Keysight’s calibration guidance explicitly models transmission-line offset delay, offset loss and characteristic impedance. Its field cable-and-antenna guidance separates insertion loss, return loss and distance-to-fault. That separation matters: a good station-end S11 trace can coexist with excessive insertion loss, and a low-loss cable can coexist with a bad antenna or connector.
| Question | Useful measurement | Declared plane |
|---|---|---|
| Is the antenna matched? | Complex S11 or impedance over the operating band | Antenna feedpoint, after an appropriate calibration or de-embedding |
| How much does the assembly lose? | S21 insertion loss of cable plus connectors | Between the two assembly connector mating planes |
| Where is a discontinuity? | Time-domain or distance-to-fault response | Calibrated cable input with correct velocity factor |
| Does it preserve array phase? | Complex S21 magnitude and phase versus frequency and temperature | Matched reference planes for every branch |
| Is power safely accepted? | Forward/reflected power at a stated plane plus temperature | At the source and, when practical, at the load |
A Practical VHF Feed-Line Workflow
- Define the operating job. Record frequency span, transmit power and waveform, receive noise target, run length, connectors, weather and whether phase is critical.
- Choose from a real datasheet. Use attenuation, velocity factor, power/voltage limits, bend radius, shielding and environmental rating for the exact cable construction.
- Measure the complete cable assembly. Calibrate at the instrument ports, terminate the far end correctly, record S21 loss and phase, then inspect S11 and distance-to-fault for connectors or damage.
- Measure the installed antenna at its intended plane. When safe and practical, calibrate at the feedpoint. Otherwise characterize the line and de-embed it with an uncertainty statement.
- Test at operating conditions. Check connector and cable temperature, amplifier stability, receive noise, weather sensitivity and mechanical movement. Repeat the baseline after any change.
- Cut only for a reason. Service slack and weather routing are valid reasons. A designed phase line, stub or transformer is another. A prettier unexplained shack-end SWR is not.
Measurement and RF safety: never connect an analyser or VNA to an energized transmitter path. Discharge antennas safely, respect instrument input limits, use rated loads and couplers, maintain weather sealing and bonding, and verify RF-exposure and connector-temperature limits under the real operating waveform.
Bottom line: at VHF and above, coax length is often less useful as an accidental matching control and more important as loss, phase and reference-plane error. Build the antenna to the intended impedance, choose the cable from measured requirements and verify the complete installed path.
Primary manufacturer and measurement references
Mini-FAQ
- Does coax length stop transforming impedance at VHF? No. Any mismatched load is transformed along the line. The effect is small only when the reflection is small, and loss can mask the returning reflection.
- Does antenna resonance guarantee a 50 Ω feedpoint? No. Resonance means the reactance is zero at the stated plane; the resistive part can still differ substantially from 50 Ω.
- Is a quarter wavelength at 145 MHz about 50 cm of coax? About 0.517 m is the free-space value. The physical length inside coax is shorter by the cable’s velocity factor.
- Can cable loss make the SWR display look better? Yes. The reflected wave is attenuated on its return, but that is lost signal rather than a repaired antenna match.
- When must VHF cable lengths be controlled? Control electrical length for phased arrays, stubs, transformers, filter interconnects, delay systems and any circuit whose source or load phase affects performance or stability.
- What should I measure before replacing a cable? Record complete-assembly insertion loss, return loss, phase or delay when relevant, distance-to-fault, connector condition and the installed antenna impedance at a declared reference plane.