Carbon Radiators: What Does a Resistance Measurement Leave Open?

The honest position is open, not evasive. Resistance is one layer in the evidence chain. Identify the specimen, separate tube and joint contributions, map the installed RF current path, and observe gain or efficiency and temperature under declared conditions before choosing a side.

What an End-to-End DC Measurement Establishes

A carbon-fibre-reinforced polymer is not a homogeneous metal. Current travels mainly through conductive fibres and across fibre, ply, coating and metal-contact interfaces. Conductivity depends on fibre direction, layup, resin, wall structure, surface treatment, damage and electrode geometry. A telescoping or segmented radiator adds collars, overlaps and pressure-dependent contacts.

A carefully made DC measurement can answer practical questions:

• Is the complete extended element electrically continuous?
• How much resistance belongs to each segment and each physical joint?
• Does contact resistance change when a section is rotated, flexed or re-extended?
• How strongly does resistance change with temperature, moisture or mechanical cycling?
• Are production specimens consistent within a declared tolerance?

It does not show where RF current is largest, how current transfers through a three-dimensional joint at frequency, or how much accepted power becomes radiation. Those require additional measurements.

Start With a Traceable Four-Wire Measurement

Two-wire resistance measurements include lead, probe and contact resistance. That error can be comparable to the feature being investigated, especially at a collar or plated interface. A four-wire, or Kelvin, connection uses one pair of leads to force current and another pair to sense voltage close to the intended measurement planes.

Record the instrument, range, calibration or verification resistor, force current, polarity, lead placement, contact method and ambient temperature. Reverse the current and average the two readings when thermoelectric offsets matter. Do not press sharp probes into a composite and then assume the punctured surface represents its service contact.

Measure the parts before trusting the total

1. Identify the specimen. Record dimensions, construction, segment order, joint type, coating, connectors, production lot and visible condition.
2. Define the planes. Mark exactly where current is forced and voltage is sensed. Photograph the arrangement.
3. Measure each segment. Use repeatable electrodes that reproduce the intended axial current direction.
4. Measure each joint. Compare readings at declared overlap, contact pressure and orientation.
5. Measure the complete element. Extend and assemble it exactly as used, then compare the total with the sum of the parts.
6. Repeat. Reassemble several times and test more than one specimen. Report spread, not just the best reading.
7. Apply controlled ageing. Repeat after extension cycles, flexing, vibration, wet/dry exposure and thermal cycling appropriate to the intended service.

Why DC Resistance Is Not RF Loss Resistance

At RF, electric and magnetic fields set a distributed current density. In an isotropic metal this leads to familiar skin and proximity effects. In a carbon-fibre composite, longitudinal, transverse and through-thickness conduction can differ greatly, while the feed and collars determine how current enters and spreads through the laminate. A single scalar resistance cannot describe that tensor and contact network.

The installed antenna also has a non-uniform current distribution. A resistance near a current maximum contributes more loss than the same resistance near a current minimum. Joints therefore have to be represented at their physical locations. A statement that carbon necessarily behaves “better at HF” than its DC value, or that current crowding automatically reduces loss, is not a measurement result.

The first expression is useful only when the equivalent loss resistance and reference current describe the same distributed system. The second requires radiated and accepted power at declared reference planes. Neither quantity can be obtained from an end-to-end ohmmeter reading alone.

Build the RF Evidence in Layers

A network analyser is useful for impedance and fixture work, but low SWR does not prove low loss. A lossy radiator can be easy to match and can show broad impedance bandwidth. To estimate efficiency, use a calibrated antenna measurement method or a controlled A/B comparison that keeps geometry, matching, counterpoise, feedline, receiver or transmitter path and surroundings equivalent.

A controlled comparison

1. Use elements with the same intended external geometry and document every unavoidable difference.
2. Keep feedpoint, counterpoise or ground system, matching network, feedline routing and calibration plane unchanged.
3. Measure complex impedance and accepted power, then measure gain or efficiency with uncertainty.
4. Swap element position, instrumentation channel or measurement sequence to expose systematic bias.
5. Repeat across frequency and environmental conditions. Report the distribution of results.

Thermal Testing Establishes the Operating Envelope

RF heating depends on local RMS current and the loss at that location. Temperature then depends on heat spreading, resin and adhesive properties, coating, collar mass, ambient temperature, wind, sun, mounting and time. One surface-temperature reading is not automatically the temperature of a buried contact, and thermal-camera readings require an emissivity appropriate to each surface.

Use a staged test:

1. Define frequency, waveform, forward and accepted power, mismatch, transmit fraction, duration, ambient conditions and cooling.
2. Measure current and voltage where practical, and place electrically safe temperature sensors at predicted current maxima, joints and feed transitions.
3. Begin at low stress. Hold each step long enough to observe the relevant thermal time constant.
4. Stop on unexpected temperature rise, resistance drift, intermittent contact, odour, noise, arcing, delamination or mechanical change.
5. Allow the assembly to cool, then repeat its electrical measurements to detect permanent change.
6. Test mismatch, faults and environmental extremes separately in controlled, protected conditions.

A mode name is not a duty-cycle specification. Carrier or PEP, waveform, transmit fraction, message timing and test duration must be stated. A result for one specimen in still indoor air does not establish an outdoor rating across production, ageing and installations.

How to Report the Result

A useful report includes the raw readings and the boundary of every conclusion:

• specimen identity, dimensions, construction, joint settings and condition;
• instrument models, calibration state, fixtures, reference planes and photographs;
• DC segment, joint and total resistance with temperature and repeatability;
• RF impedance or fixture results versus frequency, including de-embedding;
• installed geometry, ground or counterpoise, matching and feedline details;
• accepted power, gain or efficiency method and measurement uncertainty;
• temperature versus time at each instrumented location; and
• pre-test and post-test resistance, visual inspection and any permanent change.

From that evidence, it is reasonable to state that a tested assembly remained stable within the declared envelope, or that a particular joint dominated resistance and heating. It is not reasonable to publish a universal dB difference, receive-performance verdict or safe wattage merely by scaling one resistance reading to a different radiator.

What Remains Open

The lower-than-expected resistance result deserves follow-up, not a victory lap or a dismissal. It should begin—not end—the investigation. The next useful evidence is repeatability across specimens, frequency-dependent fixture data, the installed current distribution, a controlled gain or efficiency comparison and instrumented temperature versus time.

Until that evidence is available, the balanced conclusion is simple: the worst-case assumption may have been too pessimistic, while a product-specific RF-loss, efficiency or power claim remains unproven. That is not indecision. It is exactly where the measurement boundary currently sits.

Questions, antenna-factor records or height trials to share? Contact RF.Guru.