Common-Mode Choke Placement: Follow the Installed Current
Common-Mode Choke Placement: Follow the Installed Current
A feedpoint, a counterpoise boundary, the building entry and the equipment end are all possible choke locations. None is universally correct. First define which conductors are meant to carry antenna current, then measure where unwanted current continues.
“Where does the choke go on this antenna?” sounds like a simple question. The honest answer is another question: where should the intended RF return path end? A choke is a boundary in a complete current loop. Dipole, EFHW, OCF, vertical and loop are useful geometry names, but none reveals every current on the feedline, mast, radials, counterpoise, control cables and station wiring after installation.
Joeri’s practical rule: keep impedance transformation and common-mode suppression visible as separate jobs. Select the transformer from the measured complex load. Place and specify the choke from the installed exterior-current path. If one assembly claims to do both, verify both functions.
What Current Is the Choke Suppressing?
In the wanted coaxial transmission-line mode, current on the centre conductor is accompanied by equal-and-opposite current on the inside of the shield. A different current can flow on the outside of the shield. Roy Lewallen, W7EL, calls this an imbalance current in Baluns: What They Do and How They Do It. That exterior current can radiate, receive local fields, change the installed pattern and connect the antenna to the station through an unintended route.
A choke around the complete cable inserts a complex impedance into that exterior-current loop while ideally leaving the wanted differential mode nearly unchanged:
ZCM(f) = RCM(f) + jXCM(f)
Both terms affect current. The resistive term dissipates real power as heat; the reactive term stores and returns energy and can participate in a new resonance with the conductors on either side.
“High impedance” is therefore incomplete unless frequency, fixture, common-mode source and load, voltage, current and temperature are stated. The Fair-Rite technical catalogue shows that ferrite impedance is frequency-dependent and contains both resistive and reactive components. Winding capacitance, lead length and the installed cable network further change the finished choke.
Decide Which Conductors Belong to the Antenna
Before moving ferrite, draw both sides of the RF source. One side may be an obvious wire or element. The other may be a second wire, a radial field, a counterpoise, a mast, a deliberate coax-exterior section, capacitance to nearby structures or some combination. Current does not stop because the drawing labels one terminal “ground.”
| Installed condition | Boundary to investigate | What still has to be measured |
|---|---|---|
| The radiator and its deliberate return structure are complete at the feedpoint | The feedpoint is the first choke candidate if the coax exterior should not participate. | Exterior current immediately below the feedpoint and farther down the route, on every band. |
| A defined section of coax exterior is an intentional return branch | The candidate boundary is the far end of that deliberate section, not automatically the transformer. | Current along the section, current beyond the choke, tuning, loss and installed field behaviour. |
| The coax exterior is carrying current accidentally | First create or repair the intended return structure; then test a choke at its boundary. | Whether the change moved the current to the mast, wiring or another parallel conductor. |
| The antenna boundary is controlled but current reappears downstream | A second location may be useful where coupling creates a different exterior-current loop. | Multi-position current before and after each change; receive and transmit behaviour separately. |
This makes the word counterpoise concrete. If a coax-exterior length is meant to carry current, it is an antenna conductor. The choke defines where that branch should end. If that length is not meant to radiate or receive, do not retroactively call accidental feedline current a designed counterpoise.
Why Fixed Wavelength Fractions Fail as Placement Rules
A fraction such as 0.05 wavelength can describe one proposed geometry, but it cannot choose a universal choke position. The current on an installed exterior path depends on frequency, conductor length, route, height, capacitance to surroundings, mast and support coupling, transformer capacitance, radials or counterpoise, equipment bonds and the impedances at both ends.
Multiband operation makes a single fraction even less decisive. A physical length represents different electrical lengths on every band, and the exterior-mode velocity is not automatically the published velocity factor for the coax’s internal differential mode. A length that lies near a current minimum on one band may lie near a current maximum on another.
Use a proposed length as a model input, then verify it with current measurements at several positions. Do not publish a 0.05-to-0.25-wavelength range as if every value inside it were safe, quiet or efficient.
Placement Questions for Common Antenna Families
The table below is deliberately a question map, not a set of recipes:
| Antenna family | Current-path question | Useful first test |
|---|---|---|
| Centre-fed dipole, folded dipole, Yagi, quad or balanced loop | Do the two intended sides remain sufficiently balanced after feedline routing, boom or mast coupling and nearby structures are included? | Compare exterior current just below the feedpoint and farther down the coax, with a characterised feedpoint choke inserted and removed. |
| Off-centre-fed dipole | What differential transformation is required, and which conductor provides the common-mode return created by the installed asymmetry? | Measure complex feedpoint load and exterior coax current separately; do not ask transformer ratio to prove choking. |
| EFHW or other end-fed wire | Is the return a separate wire, a deliberate coax-exterior section, mast or an uncontrolled station path? | Map exterior current from the transformer through and beyond the intended return boundary on every operating band. |
| Ground-mounted or elevated vertical and inverted-L | Do radials or counterpoise carry the intended return, and are mast, coax, control cable or soil adding parallel branches? | Compare radial or counterpoise currents and scan the complete feedline exterior above and below the first candidate choke. |
| Receive loop, probe or active antenna | Is the cable carrying local noise to the antenna, converting common mode to differential mode or disturbing the element pattern? | Freeze receiver settings and compare wanted signal, noise and cable current with an A/B/A choke or routing change. |
A choke at a feedpoint is often a sensible first candidate when the intended radiator and return are already complete there. It is not guaranteed to solve every downstream path. A feedline can be excited again by fields, can couple to a mast or control cable, or can bypass a finite choke through another conductor.
Keep Transformation and Choking Separate
Many practical amateur installations do not preserve textbook symmetry after cable routing and surroundings are included. My default in an intentionally unbalanced system is therefore explicit: if the measured complex load calls for an impedance transformation, use a suitable UNUN for that differential job and a separately characterised choke at the intended exterior-current boundary.
That is not a universal 4:1 ratio, and it does not make every antenna unbalanced by definition. A genuinely balanced installed load can be the right place for a suitable current balun. An integrated transformer-plus-choke network can also work when its differential transfer, balance, common-mode impedance, voltage, current and temperature are all demonstrated over the actual load and frequency range.
Physical proximity does not prove integration. An UNUN beside a choke remains two functions unless the combined current paths and stresses have been measured. Likewise, a component called a balun does not automatically supply the needed impedance ratio, current balance and common-mode suppression for every load.
When a Second Choke Is Useful
A second choke at a building entry or near equipment can address a second coupling path. It can reduce exterior current picked up along a noisy cable route, limit current entering cabinets or isolate a cable segment excited after the antenna boundary. It should be added because a measurement identifies that boundary—not because every station needs a particular count.
An entry choke does not replace shield bonding, protective earth, equipotential bonding or lightning and surge protection. IEC TR 61000-5-1:2023 treats earthing, bonding, cables, shielding, filters, isolation and surge protection as coordinated but distinct EMC measures. Safety conductors remain in place during RF experiments.
A choke near the transceiver can also reveal rather than solve a poor antenna boundary: if substantial current reaches the final metres of cable, the equipment-end choke may become the highest-voltage or hottest part of the common-mode circuit. Measure what it changes upstream and downstream before calling the installation fixed.
Map Current at More Than One Point
One clamp reading cannot describe a standing-wave current distribution. Mark repeatable positions along the coax, mast and parallel cables. Use the same current probe orientation and receiver or transmitter settings for each scan. Record frequency, power, mode, duty cycle, cable route, antenna configuration and weather or soil state when relevant.
- Freeze the intended geometry. Document radiator, return conductors, coax route, mast, radials, counterpoise, bonds and attached cables.
- Mark the reference planes. State where impedance, SWR, insertion loss and current are measured.
- Scan the baseline. Measure exterior current at several positions on each band before adding ferrite.
- Choose one boundary. Install one choke whose complex impedance is known across the required frequencies.
- Repeat the full scan. Look for current reduction, redistribution or a new maximum rather than measuring only beside the choke.
- Restore and repeat. Use A/B/A trials to catch cable movement, propagation, AGC, warming and source variation.
- Check the desired result. Record tuning, differential loss, pattern or field evidence, receive SNR and station symptoms as applicable.
- Verify operating stress. Test realistic power and duty cycle after thermal equilibrium and inspect cable, connectors, winding and ferrite temperature.
For receiving tests, compare wanted signal and noise with fixed bandwidth, gain and attenuation. A lower noise floor alone is ambiguous because the choke may also have changed the wanted antenna response. For transmitting tests, use equal accepted power when comparing field results and keep people outside the assessed RF-exposure region.
A Choke Is a Component With Limits
Small-signal common-mode impedance is necessary evidence, not a complete power rating. In operation, ferrite loss creates heat; winding and connector loss add heat; common-mode voltage appears across the choke; and high differential standing-wave voltage or current can stress the coax even when common-mode current is modest.
A valid installation record therefore includes:
- complex common-mode impedance versus frequency with fixture and reference plane;
- differential insertion and return loss with representative complex loads;
- the cable, ferrite material, winding geometry, connectors and enclosure actually used;
- installed exterior-current maps before and after the choke;
- operating voltage, current, waveform, power and duty cycle; and
- temperature after equilibrium plus repeat measurement after thermal cycling.
If that evidence is missing, the honest claim is that the choke is a candidate at a chosen current boundary. It is not yet proof of broadband suppression, QRO survival, a quieter receiver or a corrected radiation pattern.
Primary and Authoritative Sources
- Roy Lewallen, W7EL — Baluns: What They Do and How They Do It: original current analysis and experiments on feedline imbalance, exterior current and balun behaviour.
- Fair-Rite Products — 17th-edition technical catalogue: manufacturer data and equivalent-circuit treatment of ferrite’s frequency-dependent resistive and reactive impedance.
- IEC TR 61000-5-1:2023: current general EMC installation and mitigation guidance covering coordinated earthing, bonding, cables, shielding, filtering, isolation and surge-protection measures.
- IEC 60364-5-54: authoritative separation of earthing arrangements, protective conductors and protective bonding for electrical-installation safety.
Joeri’s Bottom Line
I do not place chokes from an antenna-name chart. I decide which conductors should carry RF, choose the impedance transformer from the measured differential load, and map the current that remains on every exterior path. The first choke goes where the deliberate return branch should end. Any additional choke must earn its place by controlling another measured path.
This approach keeps both worlds covered. A suitable UNUN plus a separate choke makes transformation and common-mode control explicit in a normally unbalanced installation. A current balun remains valid when the installed load is genuinely balanced, and an integrated network remains valid when both jobs are proven. The current map—not the label or a fixed wavelength fraction—decides.
Mini-FAQ
- Does a dipole always need its choke at the feedpoint? The feedpoint is the first candidate when the two intended radiator sides are complete there, but installed asymmetry and downstream coupling still require exterior-current measurements.
- Why can an end-fed antenna need the choke away from the transformer? A deliberate section of coax exterior may form the return branch. The choke belongs where that intended branch should end, as verified on every band.
- Is 0.05 wavelength a reliable universal choke distance? No. It can describe one proposed geometry, but installed current depends on frequency, routing, surroundings, return structure, transformer and the impedances at both ends.
- Should an off-centre-fed dipole use a choke at the transformer? Not from the antenna name alone. Measure the differential load and the exterior-current path separately, then place the common-mode boundary where the installed system requires it.
- Can a station-entry choke replace bonding or lightning protection? No. It may control a measured RF path, but protective earth, bonding, entry shielding and lightning or surge protection are separate safety functions.
- How do I know whether a second choke helps? Scan exterior current at several positions before and after one controlled change, restore the baseline, and verify current, SNR or field result plus temperature on every operating band.