Quarter-Wave Transformer and Stub Matching
A quarter-wavelength of transmission line has a strange and useful property: it inverts and transforms impedance, so a section of cable with the right characteristic impedance can match two different real impedances together with no coils, no capacitors, and nothing to tune except a saw and a splice. The same quarter-wave length, left open or shorted at the far end instead of terminated in a load, becomes a matching or grounding stub — the trick behind the J-pole's feed and behind common-mode chokes at a stub-matched feedpoint. This guide covers both uses, a worked length and impedance example, a calculator, and troubleshooting.
Why a quarter wavelength of line is special
Along any transmission line, the impedance you measure looking into it depends on both the load at the far end and how far back from that load you are, measured in wavelengths. At exactly a quarter wavelength back from a purely resistive load, that relationship simplifies to a clean, useful formula: the impedance looking into the line equals the line's own characteristic impedance squared, divided by the load resistance. Pick a cable with the right characteristic impedance and a quarter-wave section transforms one real impedance into another, entirely through geometry — no reactive components required.
Building the Q-section in practice
The impedance a quarter-wave transformer needs is rarely a value you can buy off the shelf — matching 25Ω to 50Ω calls for about 35Ω, a cable impedance that basically doesn't exist as a standard product. The practical workaround is combining standard-impedance cables: two runs of 75Ω coax connected in parallel present 37.5Ω, close enough for most matching purposes, and is the classic way hams build Q-sections for phased vertical arrays. Where the target value is closer to a standard cable (50Ω or 75Ω) directly, a single run works fine with no combining needed.
Stub matching — the other quarter-wave trick
A quarter-wave section left unterminated at the far end, instead of connected to a load, behaves very differently depending on whether that far end is shorted or open: a shorted quarter-wave stub looks like an open circuit (very high impedance) at its input, and an open quarter-wave stub looks like a short circuit (very low impedance) at its input. This is exactly the mechanism behind a J-pole's matching stub — the shorted parallel-rod section transforms the high impedance at the end of the half-wave radiator down to 50Ω at a tap point along the stub, while also providing a DC path to ground for static drainage.
- Shorted stub: reflects as an open circuit at resonance — used where you need to support or ground an element without loading the RF at that point.
- Open stub: reflects as a short circuit at resonance — less common in ham construction since an unterminated open end is more exposed to moisture and detuning.
Bandwidth and power handling
Both the Q-section transformer and the matching stub are narrowband by nature — the quarter-wave condition is only exact at one frequency, and the match degrades as you move away from it, usable across roughly 10-15% of the center frequency before SWR climbs past a comfortable working range. On the plus side, because the "component" is just a length of coax rather than a lumped inductor or capacitor, a quarter-wave transformer or stub has essentially no power-handling limit beyond the cable's own voltage and current rating — there's no small capacitor to arc or inductor to overheat.
| Method | Elements Needed | Bandwidth | Best For |
|---|---|---|---|
| L-Network | 1 series + 1 shunt (L and C) | Narrow, single frequency, field-tunable | General-purpose matching of any real feedpoint resistance, at the antenna or in the shack |
| Quarter-Wave Transformer (Q-Section) | 1 transmission-line section, ~λ/4 long | Narrow (roughly 10–15% of center frequency) | Matching two known real impedances with no lumped components; phased and stacked arrays |
| Stub Match (open/shorted λ/4 stub) | 1 shorted or open transmission-line stub | Narrow | Canceling reactance or providing a DC ground path at a specific point on the line, e.g. J-pole feeds |
| Series Match | 1 series L or C | Narrow | Canceling small residual reactance when the resistance is already close to Z0 |
Quarter-Wave Transformer / Stub Length Calculator
Materials for a quarter-wave transformer or matching stub
A quarter-wave coax transformer section spliced inline in a phased vertical array's feedline, with the splice weatherproofed near the antenna base.
Building a Quarter-Wave Transformer or Stub
Cut long, trim to resonance — velocity factor varies enough between cable batches that the calculated length is a starting point, not a final cut.
Confirm the impedances you're matching
Measure or confirm the actual load resistance you need to transform (e.g. a phased array's combined feedpoint, or a single antenna's resonant resistance) and the impedance of the main line you're matching it to. The calculator above only handles purely resistive loads — if there's significant reactance, resonate the antenna or array first.
Determine the transformer impedance and select cable
Use the calculator to find the required transformer characteristic impedance. If it doesn't match a standard cable, plan a parallel or series combination of standard-impedance cables that gets you close — two 75Ω runs in parallel for 37.5Ω is the most common combination in ham construction.
Cut the section long and prepare the ends
Cut each cable run several inches longer than the calculated physical length so you have room to trim. Strip and prepare connectors on both ends, and if paralleling two cables, keep both runs the same length so they stay in phase with each other.
Splice the section inline
Connect the transformer section between the load and the main line using connectors or a direct solder splice, and weatherproof every joint with self-fusing tape followed by heat-shrink. Label the section clearly — an unlabeled matching section looks identical to ordinary feedline and is easy to accidentally cut or replace during future maintenance.
Sweep and trim to resonance
Sweep SWR looking into the main line. If the SWR minimum sits above your design frequency, the section is electrically too short (trim is not possible — you'll need to remake it slightly longer); if it sits below, the section is electrically too long and can be shortened in small increments, re-sweeping after each trim, until the minimum lands on target.
| Symptom | Most likely cause | Diagnosis | Fix |
|---|---|---|---|
| SWR minimum is offset from the design frequency | Actual cable velocity factor differs from the value used in the length calculation | Compare how far off the dip is — a dip below the design frequency means the section is electrically too long, above means too short | Trim a too-long section in small increments; remake a too-short section slightly longer since coax can't be lengthened |
| SWR never reaches a clean minimum, just a broad shallow dip | Paralleled cables aren't the same electrical length, so they're not staying in phase with each other | Physically re-measure both cable runs and confirm they were cut to matching lengths, not just similar ones | Re-cut the shorter run's counterpart or add a small trim to equalize both lengths |
| Good match right after installation, degraded weeks later | Moisture has entered a splice, changing its effective impedance and loss | Inspect every splice for cracked tape, discoloration, or visible moisture | Re-do the weatherproofing with fresh self-fusing tape and heat-shrink, sealing every seam completely |
| Match is fine at the design frequency but unusable just outside the band | Normal narrowband behavior of a quarter-wave section, not a fault | Confirm the SWR curve is a clean, narrow dip centered on the design frequency rather than erratic or absent | This is expected; if wider coverage is needed, an L-network or remote ATU may be a better fit than a fixed quarter-wave section |
| Someone (possibly future you) accidentally shortened or replaced the matching section during other feedline work | Section wasn't clearly labeled as a critical-length matching component | Check whether SWR degraded immediately after unrelated feedline maintenance | Remake the section to the correct length and label it clearly on both ends before closing everything back up |
Can I use any coax for a quarter-wave transformer, or does it have to be a specific impedance?
The characteristic impedance of the cable is exactly what sets the transformation ratio — it's not optional. If the exact value your calculation calls for isn't available as a standard product, combine standard-impedance cables (most commonly two 75Ω runs in parallel for about 37.5Ω) to get close enough for a good match.
Why does a J-pole use a shorted stub instead of a plain quarter-wave transformer?
The J-pole's feed problem is different from a simple two-impedance match — it needs to transform the very high impedance at the end of a half-wave radiator down to 50Ω at a selectable tap point, while also giving the antenna a DC ground path. A shorted quarter-wave stub does both jobs at once: it reflects as an open circuit at its far (shorted) end while providing exactly that ground path and a range of impedances along its length for the feedpoint tap.
How accurate does the velocity factor I use need to be?
Close enough matters — a few percent of velocity factor error shifts the section's electrical length by roughly the same percentage, which can move the SWR dip noticeably off frequency. The printed spec on a cable's datasheet is a good starting point, but measuring the actual reel with a NanoVNA's time-domain function, if available, removes that uncertainty entirely.
Is a quarter-wave transformer lossier than an L-network?
Generally no — a well-built quarter-wave section is just a length of coax, so its loss is simply the cable's normal per-foot loss at that frequency, with no lossy inductor core or marginal capacitor in the signal path. An L-network with a low-Q inductor can actually be lossier in practice, even though both methods are capable of a low-loss match when built well.
Do I need to worry about power handling with a quarter-wave section?
Only to the extent that any coax run does — check the cable's voltage and current ratings against your power level and expected SWR during use, same as you would for ordinary feedline. There's no small lumped component here to be the weak link the way a capacitor is in an L-network.