R
Member
When a prototype antenna comes back from the fab with a feed-point impedance nowhere near 50 ohms, a stub tuner is often the fastest way to find out whether the design is salvageable before ordering another spin. Here is what I have learned using both open and shorted stubs on the bench.
1. The basic idea
A stub is a length of transmission line connected in shunt (or series) at the feed. An open stub looks capacitive below a quarter wavelength and inductive above it; a shorted stub does the opposite. By choosing the length, you add a controlled susceptance that cancels the reactive part of the antenna impedance.
2. Open vs. shorted, in practice
Open stubs are easier to fabricate — no via to ground, no soldering a short. But an open end radiates a little and is sensitive to nearby objects; I have seen a hand waved near an open stub move S11 by 0.5 dB. Shorted stubs are more stable and repeatable, at the cost of a via or a soldered short, and the short itself has parasitic inductance at higher frequencies.
3. My bench procedure
Measure the antenna with a VNA, note the impedance at the target frequency, then compute the stub length analytically (Smith chart or a quick calculator) as a starting point. Then cut the stub 10% long and trim in small steps while watching S11 live — the analytic answer is never exact because of the connector and the launch.
4. What to watch out for
A stub fixes one frequency. If the antenna needs 10% bandwidth, a single stub will not save it — that is a design problem, not a matching problem. Also check the stub's effect on efficiency: a stub with high current can add loss, especially on lossy substrates.
When do you reach for a stub tuner, and when do you just respin the board?
1. The basic idea
A stub is a length of transmission line connected in shunt (or series) at the feed. An open stub looks capacitive below a quarter wavelength and inductive above it; a shorted stub does the opposite. By choosing the length, you add a controlled susceptance that cancels the reactive part of the antenna impedance.
2. Open vs. shorted, in practice
Open stubs are easier to fabricate — no via to ground, no soldering a short. But an open end radiates a little and is sensitive to nearby objects; I have seen a hand waved near an open stub move S11 by 0.5 dB. Shorted stubs are more stable and repeatable, at the cost of a via or a soldered short, and the short itself has parasitic inductance at higher frequencies.
3. My bench procedure
Measure the antenna with a VNA, note the impedance at the target frequency, then compute the stub length analytically (Smith chart or a quick calculator) as a starting point. Then cut the stub 10% long and trim in small steps while watching S11 live — the analytic answer is never exact because of the connector and the launch.
4. What to watch out for
A stub fixes one frequency. If the antenna needs 10% bandwidth, a single stub will not save it — that is a design problem, not a matching problem. Also check the stub's effect on efficiency: a stub with high current can add loss, especially on lossy substrates.
When do you reach for a stub tuner, and when do you just respin the board?
