Diplexers and Common Mode: Test the Hidden Coupling Path
Diplexers and Common Mode: Test the Hidden Coupling Path
A diplexer can provide excellent band separation between correctly terminated ports and still fail to describe every path in the installed station. Its specified filter response concerns the intended port mode. Current on cable exteriors, conversion at imperfect junctions, chassis coupling and receiver overload are different questions.
RF.Guru working definition: Common-mode current is the non-cancelling phasor-sum current in a specified set of conductors, evaluated at a defined cross-section and using a declared current-direction convention. In the intended differential transmission-line mode, the outgoing and return currents are equal and opposite, so their phasor sum is zero. When they do not cancel, the remaining current must close through another reference or return path—such as the outside of a coax shield, a mast, equipment chassis, station wiring, nearby structures, earth, the operator, or distributed coupling through the environment.
This broader working definition is especially useful in practical antenna systems. On transmit, non-cancelling current on the outside of the coax can make the feedline and connected structures part of the radiating antenna system unless that path is intentional, clearly defined and properly controlled—for example by providing the required return path and placing a suitable common-mode choke at the correct boundary.
I like diplexers because they solve a clean frequency-routing problem. I distrust the moment when that clean three-port drawing becomes the entire station model. If unwanted current can travel on the outside of a cable, through a shared chassis or along another bonded conductor, it has found a path that the advertised passband and isolation trace may never have measured.
Short version: a diplexer separates frequency bands for the mode and terminations used in its test. It neither creates nor cures common mode by definition. Measure ordinary port isolation, exterior cable current and receiver linearity separately.
What the Diplexer Specification Actually Describes
A low-pass/high-pass diplexer joins two frequency-selective branches at a common port. In the intended coaxial mode, current leaves on the centre conductor and returns on the shield’s inner surface. A calibrated vector network analyser can characterize insertion loss, return loss and isolation between the connector reference planes.
Those are conditional measurements. Frequency span, calibration plane, port impedance, unused-port terminations, drive level and temperature belong to the result. The Mini-Circuits examples linked below publish separate insertion-loss, return-loss and crossover-isolation data for a declared device and test condition. They do not promise infinite rejection or describe every conductor attached to an installation.
Port isolation is especially easy to overread. It is the measured transfer between two declared ports under specified termination conditions. A mismatched antenna, open auxiliary port, long interconnect or unexpected load can change the network presented to the filters. The resulting station isolation can therefore differ from a catalogue curve without proving that the diplexer itself is defective.
The Exterior of Coax Is Another Conductor
The wanted coaxial mode is largely confined between the centre conductor and the shield’s inner surface. Current can also flow on the shield exterior relative to the surrounding station, building, earth and other conductors. That exterior current does not become harmless merely because it passes a diplexer connector.
The metal enclosure and connector shells may carry the exterior current from one cable to another, divide it among several routes or couple it into nearby wiring. The impedance of those routes varies with frequency, length, geometry, bonding and the surrounding structures. They may be resonant, lossy or strongly coupled; common mode is not a frequency-independent bypass.
This is the hidden path to check. A diplexer’s normal single-ended coaxial S-parameters can be entirely valid while the installed exterior-current network remains outside the calibrated fixture. The right conclusion is not that the diplexer is blind. It is that one measurement model does not contain every station mode.
Mode Conversion Happens at Asymmetry
A diplexer does not automatically generate common-mode current. Conversion requires asymmetry or coupling: an uneven connector transition, incomplete shield termination, a chassis seam, unequal cable routing, an unbalanced load or fields coupling differently into the conductors. The attached antenna and cables can supply that asymmetry even when the filter network is well designed.
Likewise, capacitors or inductors drawn from the signal node to the diplexer enclosure are not proof of a harmful common-mode return. In a coaxial filter they are part of the intended centre-to-shield network. Whether current appears on the outside depends on the complete electromagnetic structure, enclosure currents, cable attachments and environment.
Mixed-mode S-parameters make the distinction explicit for multi-conductor fixtures: differential-to-differential, common-to-common and the two conversion terms are separate responses. A conventional three-port coaxial sweep normally characterizes the intended connector mode. A mode-conversion investigation needs the additional conductor set, fixture and reference definition.
Keep two matrices in mind. The diplexer has a measured port-transfer matrix for its intended mode. The installed station also has a conductor-and-field coupling network. Good engineering measures the one that matches the fault being diagnosed.
Why Cross-Band Symptoms Have Several Causes
An HF transmission appearing in a VHF receiver, or VHF transmission desensitizing an HF path, does not identify one mechanism. Ordinary stop-band leakage may exceed the receiver’s available headroom. Transmitter harmonics can fall directly in the victim passband. Antennas and cables can couple through space. Exterior currents can bridge the physical installation. Nonlinear junctions can generate intermodulation. The receiver or an upstream amplifier can overload and create products internally.
Audio hum, elevated noise, “ghost” signals and pattern changes are symptoms, not common-mode detectors. Start by measuring the frequency and level at declared reference planes. Then test cable current and fields. Finally verify that the receiving chain remains linear at the applied total power.
A useful overload check is to insert known attenuation ahead of the suspected active stage or add an appropriate preselector. A wanted external signal in a linear chain falls predictably with attenuation. Internally generated products can fall faster, rearrange or disappear. Repeat with another receiver or signal path when possible, and never exceed an instrument’s maximum input level.
Measure the Intended Ports First
- Declare the reference planes. Calibrate at the diplexer connectors or de-embed characterized adapters and cables.
- Terminate every unused port correctly. Use the impedance and power handling required by the manufacturer and by the measurement.
- Sweep every relevant path. Record insertion loss, return loss and port-to-port isolation across both passbands, transition regions and important transmitter harmonics.
- Stay inside the linear test range. A low-level VNA sweep does not establish high-power loss, voltage, temperature, duty-cycle or intermodulation behaviour.
- Include the installed loads. Repeat with representative antennas or characterized load networks when their mismatch is part of the question.
Do not key a transmitter into an unverified network. Powered tests require rated terminations, directional couplers or protected samplers, suitable attenuation, interlocks and a thermal plan. Observe the diplexer manufacturer’s frequency, power, load and environmental limits.
Map the Exterior-Current and Field Paths
Clamp a suitable RF current probe around the complete coax cable, not around only the centre conductor. That arrangement responds to net enclosed current and is the practical way to look for the exterior mode. Measure at several positions on the common cable and both branch cables while keeping probe orientation and receiver settings fixed.
Probe transfer impedance, frequency range, aperture loading, cable position and analyser input calibration determine whether the result is quantitative. An uncharacterized clamp can still support relative A/B/A work, but it should not be labelled as an absolute current measurement.
Use small magnetic- and electric-field probes around connectors, enclosure seams, cable entries and nearby equipment to localize coupling. Near-field amplitude depends strongly on distance and orientation, so preserve the geometry. These probes locate mechanisms; they do not by themselves produce a radiated-emissions compliance result.
Change one item at a time, then restore it. Reroute one cable, disconnect one optional low-voltage interconnection according to the equipment instructions, add one measured choke or substitute one terminated load, and verify that the baseline response returns. Never lift protective earth, lightning bonding or another required safety conductor to make an RF symptom disappear.
Fix the Mechanism You Demonstrated
If intended-mode isolation is inadequate, select a diplexer or additional filter whose measured passband, rejection, power, loss and termination requirements cover the actual transmit and receive levels. More filtering is useful only when it is placed before the nonlinear or vulnerable stage and remains stable under the real load.
If exterior common-mode current is demonstrated, place suitable common-mode impedance at the boundary where that current should stop. Choose the completed choke by measured complex impedance over the affected frequencies, cable current, temperature, insulation and environmental limits. The shared port is one candidate location, not a universal answer.
If mode conversion occurs at a connector, shield transition or chassis junction, repair that geometry and bonding path. If direct field coupling dominates, improve physical separation, cable routing, shielding and enclosure continuity. If receiver overload dominates, reduce upstream gain, add preselection or attenuation, and preserve the wanted-signal noise figure only as far as the link budget requires.
“Always choke before you diplex” is memorable, but incomplete. A choke at the wrong boundary can shift a resonance without removing the responsible path. Measure the port network, cable current and receiver state before deciding where any added impedance belongs.
Sources and Engineering Context
- Mini-Circuits — diplexer structure, reflective/absorptive stopbands and port functions
- Mini-Circuits ZDPL-6588-75-F+ datasheet — separate insertion loss, isolation and return loss at declared ports
- Keysight — mixed-mode S-parameters and differential/common-mode conversion terms
- ITU-T K.136 — converted common-mode current and receiver cable-entry control
- CISPR 16-1-2:2014+A1:2017 — current/voltage probes and conducted-disturbance coupling devices
- CISPR 16-2-1:2014+A1:2017 — conducted-disturbance measurement methods
- Rohde & Schwarz EPL manual — input attenuation, preselection and overload/intermodulation control
Practical Conclusion
A diplexer splits frequency bands in the mode its ports were designed and measured to carry. That is valuable, but it does not erase the cable exterior, chassis, bonds, nearby wiring or receiver input from the station.
When the wrong band appears where it should not, I first verify the ordinary port isolation with correct terminations. Then I map exterior current, inspect conversion points and test receiver headroom. Only after those paths are separated do I decide whether the cure is a different filter, a choke, a better shield transition, more separation or more receiver selectivity.
Have a diplexer installation with a repeatable coupling path? Share the port measurements, terminations, current map and A/B/A result through the RF.Guru contact page.
Mini-FAQ
- Does a diplexer block common-mode current? Not by definition. Its ordinary specifications describe transfer in the intended connector mode. Exterior cable current needs its own measurement and, when present, its own controlled boundary.
- Does a diplexer create common mode? Not automatically. Mode conversion requires asymmetry or coupling in the diplexer, connectors, loads, cable routing, chassis or surrounding installation.
- Why might installed isolation differ from the datasheet? Isolation depends on frequency, reference planes, terminations, load mismatch, drive level and the paths included in the test. The installed station can add paths outside the catalogue fixture.
- How should I look for current on the outside of coax? Use a suitable characterized RF current probe around the complete cable and compare several positions with fixed settings. Preserve probe orientation and calibration details.
- Can receiver desense occur even when diplexer isolation is good? Yes. Transmitter harmonics, direct antenna coupling, exterior current, nonlinear junctions and receiver or preamplifier overload can bypass or exceed the intended isolation result.
- Should a choke always go at the shared diplexer port? No. Put common-mode impedance at the demonstrated unwanted-current boundary and verify the result. The shared port is a candidate, not a universal location.