Feedpoint Choke vs Choke at the End of the Buried Coax Counterpoise

Why EFHW and end-fed off-center-fed dipoles do not ask the same thing from a common-mode choke

A lot of confusion around end-fed antennas comes from treating every choke as if it were doing the same job. It is not. In both EFHW systems and end-fed off-center-fed dipoles, the outside of the coax shield can either be excluded from the antenna system or intentionally used over a defined length as the practical return conductor. Once you understand which of those two modes you are building, the apparent contradiction around choke placement largely disappears.

A choke mounted directly at the feedpoint is trying to stop outer-surface current right where the antenna is trying to launch it. A choke mounted farther down the coax can be doing something very different: it can be marking the end of an intentional coax-counterpoise section, preventing that current from continuing toward the shack. Those two jobs are not equivalent. That is why a feedpoint choke usually needs much higher Zcm than a choke placed at the correct distance along the coax.

The real question is where the return current is allowed to flow

Every antenna system needs a complete current loop. If current is launched into the visible radiator, current must return through something else. In end-fed and asymmetrical wire systems, that return path is often where the practical problems start. If you do not provide a dedicated local return conductor, the outside of the coax shield will often become part of the return path whether you intended it or not.

That is why what hams often call common-mode current is not always just an accidental defect. In many end-fed installations it is the antenna system trying to close the loop on the outside of the feedline. The real design choice is therefore not simply “where do I put the choke?” The real design choice is whether the coax should be excluded from the return path at the feedpoint, or whether a defined section of coax should intentionally serve as the return conductor and then be stopped at a chosen point.

The coax can be feedline on the inside and counterpoise on the outside

This is the part many operators miss. Coax is not one RF path. It supports at least two practically important ones in this discussion. The wanted differential transmission mode exists between the center conductor and the inner surface of the shield. The unwanted or intentionally used return path exists on the outer surface of the shield. That is why the same cable can deliver power to the antenna and at the same time become part of the antenna system.

Once the outside of the shield is recruited into the antenna, it behaves like a conductor with its own current distribution, its own coupling to ground and nearby objects, and its own electrical length. At that point the coax is no longer just plumbing between antenna and shack. At least the section carrying outer-surface current has become part of the RF structure.

Two choke strategies that look similar but are not

• Feedpoint choke: the choke is placed directly at or immediately behind the transformer or feedpoint. The design goal is to keep the coax out of the antenna system and force the return current into a local counterpoise, short return leg, or some other intentional conductor at the feedpoint.
• Remote choke at the end of the coax counterpoise: the first section of coax is intentionally allowed to act as the return conductor. The choke is then placed where that intentional counterpoise is supposed to end, so the current does not keep traveling toward the station.

These are not two versions of the same idea. They define two different current paths. A feedpoint choke is trying to stop the coax from entering the antenna system at all. A remote choke is allowing the coax to participate over a defined distance and then stopping it there.

Why the feedpoint choke usually needs much more Zcm

A feedpoint choke sits at the source end of the outer-surface current path. It is trying to create high impedance exactly where the antenna would otherwise most readily launch current onto the coax. In simple terms, the common-mode current is set by the available driving voltage and the total impedance in that path: Icm ≈ Vcm / (Zpath + Zcm). Near the feedpoint, the natural impedance seen by that current is often low enough that the added choke impedance must be very large before the coax stops looking attractive as a return path.

Put differently, a feedpoint choke is trying to prevent the current from getting established in the first place. That is hard work. If the choke only provides moderate impedance, the antenna still finds the coax easy enough to use, and significant outer-surface current will remain. This is why feedpoint choking usually demands the highest and broadest Zcm you can realistically achieve.

The requirement becomes even more severe when the intended local return path is weak, too short, badly placed, or absent. Then the antenna still needs to close the loop somewhere, so the common-mode drive against the coax remains strong. In other words, the choke is not just cleaning up a little leftover current. It is being asked to deny the antenna one of its main return paths.

Why a remote choke can often work with less Zcm

A remote choke at the end of an intentional coax-counterpoise section is doing a different job. It is not trying to stop the return current from starting at the feedpoint. It is allowing that current to exist on the first section of coax and only preventing it from continuing beyond a chosen point. In that operating mode, the coax between feedpoint and choke is not an ordinary feedline with an accidental problem on it. It is part of the return conductor by design.

Along that intentional outer-surface conductor, the common-mode current is usually strongest near the feedpoint end and lower farther along the allowed section. By the time you reach the end of the intended counterpoise length, the current that wants to continue toward the shack is often much smaller than the current that existed where the path started. The choke placed there therefore does not need to fight the full feedpoint drive. It only needs enough Zcm to keep the remaining continuation toward the station unattractive.

That is the main reason a remote choke can often work with less Zcm than a feedpoint choke. The laws of RF have not changed. The choke is simply being installed at a different point in the current distribution and is performing a different job.

Practical note: this only makes sense when the coax between feedpoint and choke is being treated as an intentional RF conductor. “Put the choke a bit down the line” is not a design rule by itself.

This is a distributed problem, not just a single impedance number

One reason this topic gets so muddled is that people speak about choke impedance as if the whole question were just one lumped series resistor. It is not. Once the outside of the coax is allowed to carry return current, you have a distributed conductor with capacitance to the environment, coupling to the radiator, coupling to ground, and frequency-dependent current maxima and minima. Moving the choke changes the boundary condition of that conductor, so location matters almost as much as the choke itself.

This also explains why the same choke can look hopeless at the feedpoint and perfectly adequate farther away. In the first case it is being asked to suppress the current at its source. In the second it is being used to define the end of an already established and often calmer section of counterpoise.

Why the usual “current maximum” advice is not the whole story

General choke-placement advice often says that a series choke is most effective where common-mode current is high and the wave impedance on the outside of the line is low. That advice is valid when you are trying to suppress an already unwanted feedline current on a line that is not supposed to be part of the antenna.

But the situation is different when the section between the feedpoint and the choke is intentionally being used as the counterpoise. In that case the choke is not just damping a random current maximum somewhere on the line. It is also defining where the intentional conductor ends and where the quiet feedline begins. The better way to think about it is not “remote chokes always belong at current minima” or “all chokes belong at current maxima,” but rather: what section of coax is supposed to carry return current, and where should that section stop?

The “right distance” is not the coax velocity factor

There is another subtle but important point here. Once the relevant current is on the outside of the shield, the ordinary catalog velocity factor of the coax no longer tells the whole story. The common-mode path is not the 50 Ω transmission line inside the cable. It is the outer shield acting as a single conductor in the external environment.

That means the electrical length of the allowed coax-counterpoise section depends mainly on what surrounds the outside of the cable: air, jacket material, nearby conductors, supports, and especially earth when the cable is buried or laid on ground. For buried coax the soil changes both phase velocity and loss. So the “right distance” to the choke is not simply the physical length multiplied by the coax’s internal VF. It is the electrical length of the outer-surface current path in that actual installation.

Why this applies to both EFHWs and end-fed off-center-fed dipoles

None of this is unique to EFHW antennas. The same return-path physics applies to end-fed off-center-fed dipoles and other asymmetrical wire systems. In all of them, current launched into the main radiator has to return through something. If you make that return path local at the feedpoint, the feedpoint choke has to work hard and the local geometry becomes critical. If you intentionally allow a section of coax shield to become the return conductor, then the remote choke defines where that conductor ends.

So the real distinction is not the marketing label of the antenna. The real distinction is the chosen current-path strategy. Both EFHWs and end-fed off-center-fed dipoles can be operated with a feedpoint choke and a separate local return conductor, or with a defined section of coax shield acting as the counterpoise before a remote choke.

What burying the coax actually changes

When that intentional counterpoise section is buried or laid on the ground, several things happen at once. The outer-surface conductor couples more strongly to the earth, the fields are altered, resonances are damped, and some energy is dissipated in the soil. From an operating perspective this often makes the system calmer: less coupling into the shack, less touch sensitivity, and less dependence on how the rest of the feedline wanders indoors.

But buried does not mean absent. The return current is still there on the outside of the shield. Burying mainly changes the environment of that conductor. It can reduce radiation and make the system more stable, but it can also introduce loss. So a buried coax counterpoise is a control strategy, not a magical way to make outer-surface current disappear.

Practical note: a buried coax counterpoise often feels more forgiving, but “more forgiving” and “more efficient” are not automatically the same thing. Soil can stabilize the system while also adding loss.

Why a feedpoint choke without a real local return often disappoints

If you place a choke directly at the feedpoint but do not provide the antenna with a credible local return conductor, the system still has to close the current loop somehow. It will then try to use whatever is available: a short pigtail, a mast, a gutter, nearby wiring, capacitive coupling around the choke, or a combination of these. That is why a feedpoint choke is not a substitute for return-path design. It only works as intended when the current has somewhere better to go.

This is also why feedpoint choking tends to demand not only more Zcm, but also cleaner geometry. The more you try to confine the return current to a compact local region, the more every centimeter of conductor routing starts to matter.

Why choke construction still matters

Because the feedpoint choke is doing the hardest job, it is usually where a broadband ferrite choke earns its keep. A simple air-wound coax coil may look good on one band and disappointing on another, especially when its reactive behavior starts interacting with the rest of the common-mode path. A well-designed ferrite choke is more predictable and more likely to provide the high, broadband Zcm needed when you are trying to keep the coax out of the antenna at its source.

A remote choke at the end of the intentional counterpoise section may sometimes get away with less Zcm, but it still has to remain effective on the bands of interest and it still must not create an unwanted resonance. Lower required impedance does not mean any choke will do.

Measure current, not folklore

The cleanest way to verify all of this is to measure outer-surface current along the feedline. Probe the coax near the feedpoint, somewhere along the intentional buried section, and again just beyond the choke. If the design is using the first section of coax as the counterpoise, you should expect current on that section and a strong reduction beyond the remote choke. If the design is supposed to exclude the coax at the feedpoint, then current should already be low immediately below the feedpoint choke.

SWR alone does not settle this question. An antenna can show an acceptable match while still placing significant RF current on the outside of the line.

The real takeaway

A choke at the feedpoint and a choke at the end of an intentional buried coax counterpoise are not interchangeable placements. A feedpoint choke is trying to keep the coax out of the antenna at the source end of the common-mode path, so it usually needs much higher Zcm. A remote choke is allowing a defined section of coax shield to do the return-current job and only stopping that current from continuing toward the station, so it can often work with less Zcm when placed at the correct electrical distance.

That is true for EFHW antennas and for end-fed off-center-fed dipoles alike. The important question is not the antenna label. The important question is where you want the return path to exist, and where you want it to stop.

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 technical support and antenna guidance.