I am a bit confused. Let's take a HF transceiver for example.

My understanding is that the output impedance of the transmitter is 50 ohms. The impedance of the coaxial cable feedline is 50 ohms, and if we have a resonant antenna at the operating frequency, the antenna will have an impedance of 50 ohms, so all good, a maximum transfer of power, and no need for an antenna tuner. Great.

Now if we have a non-resonant antenna, the transmitter output impedance will still be 50 ohms, the feedline still 50 ohms, but there is now an impedance mismatch between the end of the coax and the antenna. Shouldn't the antenna tuner be placed there to solve the mismatch?

How is it that an antenna tuner (internal inside a transceiver, or placed at the output of the transceiver) will solve this issue?

Will this not change the output impedance of the transmitter so that now we have an impedance mismatch between the transmitter output an the coaxial cable feedline which is fixed at 50 ohms? Thanks in advance for any clarity on this.


Having a transmatch (antenna tuner) at the feed point of the antenna can be an excellent way to solve the problem of a mismatched antenna; an antenna system with the transmatch at the feed point of the antenna will almost always have less loss than an antenna system with the transmatch at the output of the transmitter, because the feedline operates the most efficiently. Remote transmatches have many disadvantages also:

  • Usually more expensive than a transmatch in the shack
  • Only work for a single antenna
  • Must be weatherproofed
  • Can be difficult to mount
  • Can be difficult to access for service
  • Harder to protect from lightning
  • Usually don't support more power than 100 or 200 W
  • Power must be provided, either from a separate power line or by using devices that send power through the coax
  • Must be remotely controlled somehow; if they sense the frequency of the transmitted RF, then the transmatch may be mis-tuned for a second or two after changing frequencies or bands, possibly forcing the transmitter to reduce power or be damaged

For an antenna tuner in the shack, the transmitter sees an antenna system with a nice 50 Ω input impedance (ideally), so the transmitter is happy. Things are not quite as optimal in the feed line, because standing waves form, which raises voltages at every point in the feed line, causing higher losses. The impedance mismatch between the feed line and the antenna causes reflections (which is what creates the standing waves), but reflections are generally not the problem; we are usually more concerned with losses in the feed line. Often, the losses are negligible or low.

Suppose that a transmitter transmits 100 W, the coax has a loss of 10 W, and a remote transmatch has an insertion loss of 10 W. The power to the antenna is 80 W. Suppose that the transmatch is moved to the shack, and the power line losses increase by 20 W; the power to the antenna is now 60 W. Another 20 W may seem like a lot, but that's only another 1.2 dB of loss, a small fraction of an S-unit. On HF, only dedicated contesters, DXers, and other weak-signal aficionados would notice a loss of 1.2 dB.

To sum up, remote transmatches are a fine solution to some mismatch problems, but not all mismatch problems. A transmatch in the shack keeps the transmitter just as happy. Transmatches in the shack cause more loss, but the loss is often negligible in practical terms. Transmatches in the shack are less expensive, and often quite a bit more practical, than remote transmatches.


The output of the transmitter will always be 50 Ohms, it's what the transmitter sees that's important. If the transmatch is in the shack and tuned properly, the transmitter will see 50 Ohms. When there's an impedance mismatch at the feed point of the antenna, the antenna impedance, what ever it is, shifts as it moves down the transmission line toward the transmitter and repeats every half wavelength of line. If the transmission line is an exact multiple of a half wavelength, the mismatched antenna impedance will be the impedance seen at the transmatch on the transmission line side. If the line is a length other than a half wavelength the impedance will be some other value. Knowing the SWR at the line feed point and the amount of half wavelengths of the line, this value can be calculated on a Smith chart or measured with a vector network type analyzer. Now all the transmatch has to do is transform this impedance to the 50 Ohms the transmitter wants to see. Yes, there will be standing waves on the transmission line but if the line losses are low there's no problem.


In the special case where transmitter output impedance, transmission line characteristic impedance, and antenna impedance are all the same, for example 50 +j0 ohms, then everywhere on the transmission line there is no standing wave and the line impedance (live impedance during operation) will equal the characteristic impedance of the line.

Assuming that transmitter and characteristic impedance of the transmission line are matched, as soon as there is a mismatch at the junction of transmission line and antenna, there will be a standing wave on the line, and this standing wave causes the line impedance at each point along the transmission line to change as a function of the distance from the antenna.

This means that at each point along the line, when there is a standing wave present, the SWR will be the same but the impedance will be different. The SWR is determined by the ratio of transmission line impedance to antenna impedance, the line impedance at each point along the transmission line is determined by the ratio of transmission line impedance to antenna impedance and the distance from the antenna.

An antenna tuner, depending on it's configuration, allows reactance of varying amplitude and phase to be added in parallel or series with the transmission line. When installed at the transmitter end, it allows reactance present at that end of the transmission line to be cancelled out by adding reactance of the same value but opposite polarity. This also allows a non-reactive impedance which isn't 50 ohms eg: 35 + J0 ohms to be transformed to exactly 50 + j0 ohms by adding the appropriate amount of positive and negative reactance in series, or in parallel if the impedance is higher than 50 ohms.

Using an antenna tuner at the transmitter end is obviously much more practical than connecting at the antenna, the disadvantage of doing this is that the tuner doesn't remove the standing wave from a transmission line, and I²R losses will be higher due to the increased current caused by the presence of the standing wave.

In fact the conditions in terms of impedance on a transmission line are determined by what is connected at both ends. This is described as boundary conditions in transmission line theory. If there is a standing wave then there can be reflection also at the transmitter end if transmitter output impedance and line characteristic impedance are different.

If an antenna tuner is adjusted so it's output is anything other than 50 ohms, this will result in current reflected back from a mismatched antenna to be re-reflected at the antenna tuner, resulting in complex multiple reflections which also affect the impedance at every point along the line.

Using an antenna tuner at the antenna end results in no standing wave on the line and is better for this reason only. Often the increased I²R losses are negligible and only need be considered when using high power.

Hope that helps !


It's important to remember that a feedline having a 50-ohm characteristic impedance doesn't "force" any signal appearing at its terminal to be 50 ohms; it just means that if you give it a signal where V/I = 50 ohm, then it will make it to the other end without reflection or transformation, while other impedances will be reflected at its boundaries and transformed along its length.

Will this not change the output impedance of the transmitter so that now we have an impedance mismatch between the transmitter output an the coaxial cable feedline which is fixed at 50 ohms?

Yes, and this is exactly what you want to happen. As you probably know, an impedance mismatch causes a reflection. A well-matched tuner causes just the right mismatch at the transceiver end of the feedline to re-reflect the reflection caused by the mismatch at the antenna end of the feedline.

The feedline itself sees mismatches at both ends, but the antenna impedance, as transformed by the feedline and the tuner, looks like 50 ohms to the transmitter; and the transmitter impedance, as transformed by the tuner and the feedline, looks like a conjugate match to the antenna, so both of them are happy and transferring optimum amounts of power to/from the feedline.

Since the feedline is operating at an SWR > 1:1, the standing-wave current causes additional losses compared to the case where the antenna is perfectly matched (or the case where the tuner is at the antenna end), but the magnitude of those additional losses depends not only on the SWR, but on how much loss the feedline has at the operating frequency to begin with. If the SWR is moderate, the cable run is short, the frequency is low, and/or the cable is low-loss, then the mismatch loss will be manageable.

The fact that the reflections have to travel down and back the feedline (potentially multiple times) before getting absorbed by the antenna does cause some amount of distortion in the form of inter-symbol interference, but for reasonably narrowband signals, and reasonable SWRs, this effect is absolutely negligible.

  • $\begingroup$ I'd think that signals where V/I = 50 Ω (SWR = 1) would still be attenuated by the feedline, just not as much attenuation as if the SWR were > 1. $\endgroup$
    – rclocher3
    Nov 15 at 20:13
  • $\begingroup$ @rclocher3 yes, that's why I said additional feedline losses :) $\endgroup$ Nov 15 at 20:23

Lets simplify the scenario a little by using a resistor at the end of a 50 foot long 50 Ohm impedance feedline instead of an antenna. If the resistor is zero Ohms, then about 100 nS after enabling a DC output, the transmitter will see a short, not 50 Ohms. If the resistor is, say greater than 100M Ohms, then, after 100 nS, the transmitter will see an open, not 50 Ohms. If you put some sort of configurable resistor divider network at the transmitter, then you can make the transmitter see 50 Ohms after about 100 nS of constant voltage output into the 50' feedline, either open or shorted at the end, even though it might see some weird impedance for the first 100 nS or so (while waiting for a reflection).

  • $\begingroup$ I don't think a time-domain example helps a beginner. In fact no-one using narrowband signals should be thinking in the time domain, it's only for high-speed digital people and their eye diagrams. $\endgroup$
    – tomnexus
    Nov 11 at 5:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.