Why dB Attenuation Specs on Ham Chokes Are a Mess
Why dB Attenuation Specs on Ham Chokes Are a Mess
“40 dB” can be a valid measurement, a useful calculation or empty marketing. Until the quantity, circuit model and test conditions are stated, you do not know which.
RF.Guru working definition: Common-mode current is the non-cancelling phasor-sum current in a specified set of conductors, evaluated at a defined cross-section and using a declared current-direction convention. In the intended differential transmission-line mode, the outgoing and return currents are equal and opposite, so their phasor sum is zero. When they do not cancel, the remaining current must close through another reference or return path—such as the outside of a coax shield, a mast, equipment chassis, station wiring, nearby structures, earth, the operator, or distributed coupling through the environment.
This broader working definition is especially useful in practical antenna systems. On transmit, non-cancelling current on the outside of the coax can make the feedline and connected structures part of the radiating antenna system unless that path is intentional, clearly defined and properly controlled—for example by providing the required return path and placing a suitable common-mode choke at the correct boundary.
A decibel is a ratio, not a physical property. A common-mode choke has a measurable complex impedance, ZCM(f). The current reduction after installation depends on that impedance and on the rest of the common-mode circuit. A bare claim such as “60 dB choke” quietly hides the most important information.
The short version: ask “dB of what, measured how, between which reference planes, with what source and load, in which mode, and at what frequency?” If those answers are missing, the number cannot be transferred safely to your station.
Four Different Numbers Commonly Called “Attenuation”
Ham discussions often place several unlike quantities on the same chart. They may all use dB, but they do not answer the same question.
| Label | What it describes | What must be specified |
|---|---|---|
| S21 | Forward travelling-wave transmission through a particular two-port fixture | Reference impedance, calibration plane, fixture topology and mode |
| Insertion loss | The change produced by inserting the DUT into a defined matched or mismatched system | Source, load, reference condition and sign convention |
| Current suppression | The ratio of common-mode current before and after installation | Position, geometry, common-mode source and complete return path |
| Common-mode rejection | Usually a system or balanced-receiver property, not automatically a choke property | Definition of wanted and unwanted modes, ports and terminations |
For voltage or current ratios, the magnitude ratio is expressed with 20 log10. For power ratios, it is 10 log10:
AI = 20 log10 |Ibefore / Iafter|
AP = 10 log10 (Pbefore / Pafter)
A passive VNA trace is often shown as a negative number, for example S21 = −34 dB. A data sheet may report the same magnitude as 34 dB of insertion loss. That sign change is only a reporting convention; it does not create another 34 dB.
What the Choke Itself Contributes
For common mode on a coaxial feedline, the wanted centre-conductor/inside-shield transmission-line current is not the target. The choke is intended to add impedance to the external mode that includes the outside of the shield and its return through the antenna, mast, station wiring, earth and surrounding electromagnetic field.
ZCM(f) = RCM(f) + jXCM(f)
|ZCM| = √(RCM² + XCM²)
This complex impedance versus frequency is the most useful starting description of a choke. It is more portable than an unexplained dB figure, but it is not completely fixture-free or a promise of installed suppression. Calibration, lead geometry, winding capacitance, enclosure coupling and the measurement’s usable impedance range still matter.
Nor is “mostly resistive” a universal requirement. A resistive component damps resonances and often gives predictable broadband suppression, but it dissipates power when common-mode current flows. A sufficiently large reactive impedance can also reduce current while dissipating less. The best balance depends on bandwidth, existing common-mode current, voltage stress and thermal design.
Why the Same Choke Produces Different dB Results
Represent the external common-mode circuit by a source voltage VCM and the impedance of everything already in the loop, Zpath. Before adding the choke:
Ibefore = VCM / Zpath
Iafter = VCM / (Zpath + ZCM)
AI = 20 log10 |(Zpath + ZCM) / Zpath|
That last expression is an installed-system estimate, not a nameplate rating. Zpath is complex and changes with frequency, feedline length, antenna geometry, mast bonding, station wiring and nearby objects. A reactive choke can even form a new resonance with that path. Therefore, a larger |ZCM| usually helps when it dominates the loop, but “twice the choke impedance equals twice the isolation” is not a general rule.
Do not substitute feedpoint SWR for Zpath. The differential impedance seen between the centre conductor and inside of the shield is not the external common-mode loop impedance. A perfect 50 + j0 Ω differential match can coexist with large outside-shield current, and a badly mismatched differential load can coexist with very little common mode.
What a 50 Ω S21 Test Actually Says
A 50 Ω VNA is not a “fantasy world.” It provides a defined, repeatable reference environment. The mistake is claiming that its result is automatically the current reduction in an unknown antenna installation.
In an ideal port-1-to-port-2 series fixture with equal reference impedances Z0, a series choke impedance gives:
S21 = 2Z0 / (2Z0 + ZCM)
ZCM = 2Z0(1 − S21) / S21
The two 50 Ω terms are the VNA’s source and receiving terminations. With Z0 = 50 Ω and a purely resistive 5 kΩ choke, |S21| is about 0.0196, or −34.2 dB. Calling that “34.2 dB insertion loss in this fixture” is reasonable. Calling it “34.2 dB common-mode reduction in every station” is not.
Now place the same ideal 5 kΩ choke in a simplified common-mode loop whose existing impedance is one 50 Ω resistance. The calculated current suppression is about 40.1 dB. The difference is approximately 6 dB because the circuit model changed from two 50 Ω VNA terminations to one 50 Ω loop impedance. Nothing was secretly added to the choke.
Where Y21 Fits—and What It Does Not Do
The Y21 approach starts with the full complex two-port S-parameter matrix, converts it to admittance parameters and treats the DUT fixture as a π network. For an ideal series branch between the ports:
Y21 = −1 / Zseries
ZCM = −1 / Y21
This can be very useful because shunt capacitance from either DUT terminal to the fixture reference is represented separately from the series branch. It is an impedance-extraction method; it does not “measure extra dB” and it does not contain an inherent 6 dB correction.
Y21 is also not magic. Direct port-to-port leakage, mutual fixture coupling, imperfect calibration, insufficient receiver dynamic range and an invalid lumped π model can corrupt the result. A proper conversion needs all four complex S-parameters—or an equivalent rigorously assembled reciprocal data set—not merely an S21 magnitude trace.
Important qualification: Y21 is a strong, accessible method for many ham choke fixtures, especially when shunt parasitics matter. It is not the only scientifically valid way to measure impedance. One-port reflection, series-through conversion, impedance analysers and calibrated bridge methods can also be valid when the chosen method suits the impedance range and the fixture is controlled.
A Number Is Only as Good as Its Fixture
For a useful published result, the report should identify:
- Quantity: S21, positive insertion loss, extracted ZCM, or measured current ratio.
- Mode and topology: how the fixture excites the external common mode while leaving the differential path out of the measurement.
- Reference impedance: normally 50 Ω for a VNA, plus the source/load model used for any derived dB curve.
- Calibration plane: whether SOLT/TRL calibration or fixture compensation reaches the DUT terminals.
- Complex data: RCM, XCM and |ZCM| versus frequency—not only a peak magnitude.
- Fixture limits: residual shunt capacitance, lead inductance, direct coupling, leakage and dynamic range.
- DUT construction: core material and count, turns, cable, winding layout and enclosure.
- Power conditions: small-signal VNA data do not establish high-power thermal or voltage performance.
The last point matters at QRO. Ferrite properties, cable loss and temperature can change under stress. Common-mode dissipation is associated with ICM,rms²RCM, while high common-mode voltage can stress winding insulation even when current is modest. “Handles 1.5 kW” needs frequency, duty cycle, SWR, common-mode excitation and thermal conditions.
How to Compare Chokes Without Fooling Yourself
- Start with complex ZCM(f). Compare R and X over every band you intend to use.
- Check the measurement method. Confirm fixture topology, calibration and the method’s usable impedance range.
- Read any dB curve as a defined model. Look for the stated source, load and equation.
- Check differential performance separately. A coax choke should pass the wanted mode with acceptable return loss and insertion loss.
- Check power and voltage separately. Small-signal impedance is necessary information, not a complete power rating.
- Verify the installation. Measure common-mode current at several positions along the feedline before and after fitting the choke.
The honest label is not “40 dB choke.” It is closer to: “ZCM = R + jX over this frequency range, measured with this calibrated fixture; the plotted dB value is the result of this explicitly stated circuit model.”
Bottom Line
dB specifications are not inherently bad. Undefined dB specifications are. S21 is a legitimate two-port measurement, Y21 can extract the series branch of a suitable model, and a current probe can show what happened in the real installation. Trouble begins when those three answers are presented as though they were the same property.
For buying or building a choke, use measured complex common-mode impedance as the starting point, then evaluate it in the actual common-mode circuit. For proving that the station improved, measure the current in the station. One number cannot do both jobs.
Mini-FAQ
- Is a “40 dB choke” necessarily dishonest? No. The figure may be valid for a stated fixture or circuit model. It becomes misleading when those conditions are omitted.
- Is 50 Ω testing wrong? No. It is a repeatable measurement reference, not a universal model of the installed common-mode path.
- Does Y21 add or lose 6 dB? No. Y21 extracts an admittance parameter. A 6 dB difference appears when different source/load models are used to calculate attenuation.
- Is higher |ZCM| always better? Not unconditionally. The complex system impedance, bandwidth, resonance, voltage and heating all matter.
- Should the choke be mostly resistive? Often useful for broadband damping, but sufficient reactive impedance can also suppress current. Examine R and X, not a slogan.
- What proves performance in my station? Common-mode current measurements at several feedline positions, made before and after the change with the geometry held constant.