Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
IEC 61726: Cable Assemblies, Cables and Connectors — Shielding Attenuation Test Method
Tip: IEC 61726 defines the absorbent clamp method for measuring the shielding attenuation of cable assemblies, cables, and connectors. It is an essential standard for EMC engineers evaluating the electromagnetic shielding effectiveness of interconnected systems across the frequency range of 30 MHz to 1 GHz.
1. Scope and Test Principle
IEC 61726 specifies a test method for measuring the shielding attenuation (also called coupling attenuation) of cable assemblies, coaxial cables, screened cables, and connectors used in telecommunications, data transmission, and RF applications. The standard addresses a fundamental EMC concern: cables and connectors often act as unintended antennas, both radiating electromagnetic energy and picking up external interference. The shielding attenuation quantifies how effectively the cable assembly’s shield reduces this electromagnetic coupling compared to an unscreened reference configuration.
The test method is based on the absorbent clamp technique, originally developed for measuring the screening effectiveness of mains cables and later extended to signal and data cables. The absorbent clamp (also known as the ferrite clamp or absorbing clamp) consists of a set of ferrite toroids that are placed around the cable under test. These ferrites absorb radiated power from the cable, and a built-in current transformer or directional coupler measures the power picked up from the shield surface currents. The clamp is moved along the cable length to find the position of maximum coupling, ensuring that the worst-case shielding performance is captured.
Important: IEC 61726 measures the overall shielding attenuation of the complete cable assembly, not just the cable shield or the connector individually. This is a critical distinction because the assembly’s shielding effectiveness is often limited by the connector-to-cable interface, where shield termination imperfections create leakage paths. A cable with excellent braid shielding can be rendered ineffective by a poorly terminated connector.
The frequency range of IEC 61726 is 30 MHz to 1 GHz, covering the VHF and UHF bands where cable radiation and pick-up are most problematic for EMC compliance. Below 30 MHz, cable dimensions are electrically small relative to the wavelength, and other test methods (such as injection probe or triaxial cell methods per IEC 62153-4 series) are more appropriate. Above 1 GHz, waveguide effects in the test setup become significant, and the absorbent clamp method’s accuracy degrades. The standard provides guidance on extending measurements up to 2 GHz with reduced accuracy for applications where higher frequency performance is critical, such as in aerospace or 5G telecommunications infrastructure.
2. Test Methodology and Measurement Procedure
The IEC 61726 test procedure follows a substitution principle. First, the test setup is calibrated using a reference configuration — typically a known-length unscreened wire or a calibration fixture with defined transmission characteristics. Then, the cable assembly under test (DUT) is installed in the same geometry, and the power coupled through the shield is measured using the absorbent clamp. The shielding attenuation as is calculated as the ratio (in dB) of the power received in the reference configuration to the power received with the shielded cable assembly, at each frequency of interest.
Shielding Attenuation Calculation (IEC 61726):
as(f) = Pref(f) − Pmeasured(f) − C(f) − Lc(f)
Where:
as(f) = shielding attenuation at frequency f (dB)
Pref(f) = reference power level (dBm)
Pmeasured(f) = measured power with DUT (dBm)
C(f) = clamp calibration factor (dB)
Lc(f) = cable loss correction (dB)
Higher values of as indicate better shielding performance.
The test setup consists of a signal generator connected to one end of the cable assembly, with the other end terminated in a matched load (typically 50 Ω for RF cables). The absorbent clamp is placed around the cable and connected to a spectrum analyser or EMI receiver through a calibrated cable. The clamp is moved along the cable, and the maximum received power is recorded at each test frequency. The requirement to find the maximum coupling means that the test must be performed at multiple clamp positions, typically at intervals of λ/4 at the highest test frequency, which corresponds to approximately 7.5 cm steps at 1 GHz.
| Frequency Range | Clamp Step Size | Typical as for Good Shield | Common Artifacts |
|---|---|---|---|
| 30–100 MHz | 75–30 cm | >60 dB | Ambient noise pick-up, cable resonance |
| 100–300 MHz | 30–10 cm | >50 dB | Connector leakage, shield weave effects |
| 300–600 MHz | 10–5 cm | >40 dB | Clamp position sensitivity, mode conversion |
| 600–1000 MHz | 5–3 cm | >30 dB | Waveguide modes, direct radiation to clamp |
Good Practice: When testing multipin or multi-coaxial cable assemblies, IEC 61726 requires that each conductor or coaxial line be tested individually while all others are terminated in matched loads. The shielding attenuation is reported for the worst-case conductor. This approach reveals whether shield current coupling between conductors (transfer impedance effects) degrades the overall assembly performance even when individual shields appear adequate.
The standard specifies stringent requirements for the test environment. Reflections from nearby objects must be minimised, which typically necessitates performing the test in an anechoic chamber or a screened room with RF absorbent material on the walls. The cable under test must be maintained in a straight line, at least 50–100 mm above the floor or any conducting surface, to ensure reproducible test geometry. The absorbent clamp itself must be calibrated regularly, with the calibration traceable to national standards. The clamp’s performance — particularly its absorption factor and current-to-voltage conversion factor — varies with frequency, and the calibration correction factor C(f) must be applied to all measurements.
3. Engineering Design Insights for Optimising Shielding Performance
IEC 61726 testing reveals several practical engineering insights for designing high-performance shielded cable assemblies. The most critical finding from extensive testing across different cable types is that the connector-to-cable shield termination is almost always the limiting factor in overall assembly shielding attenuation. Even a 1–2 mm gap in the 360° shield connection can reduce shielding attenuation by 20–40 dB at frequencies above 100 MHz, effectively negating the benefit of an otherwise excellent cable shield.
The standard highlights the importance of shield coverage and braid design. For braided shields, the optical coverage (percentage of the cable surface covered by braid wires) is a first-order determinant of shielding attenuation. A braid with 85% coverage typically provides shielding attenuation of 50–60 dB at lower frequencies, dropping to 30–40 dB at 1 GHz as the diamond-shaped apertures between braid wires become electrically significant. Combining a braid with a foil shield (foil + braid construction) increases effective coverage to near 100% and dramatically improves high-frequency performance, typically achieving 70–80 dB at lower frequencies and maintaining 50–60 dB at 1 GHz.
| Shield Type | Coverage | as at 30 MHz | as at 300 MHz | as at 1 GHz |
|---|---|---|---|---|
| Single braid (60% coverage) | 60–70% | 40–50 dB | 25–35 dB | 15–25 dB |
| Single braid (85% coverage) | 80–90% | 50–60 dB | 35–45 dB | 25–35 dB |
| Double braid (optimised lay) | 95–98% | 65–75 dB | 50–60 dB | 40–50 dB |
| Foil + braid (combined) | >99% | 70–80 dB | 60–70 dB | 50–60 dB |
| Solid tube (semi-rigid) | 100% | >80 dB | >80 dB | >80 dB |
Critical: One of the most common EMC design errors exposed by IEC 61726 testing is the “pigtail” termination — where the shield of a cable is terminated to a connector shell or chassis ground through a single wire (pigtail) rather than a full 360° shield band. At 100 MHz, a 10 mm pigtail introduces approximately 2Ω of common-mode impedance, which can reduce shielding attenuation by 15–25 dB compared to a 360° contact. For high-frequency applications, pigtail terminations must be completely avoided.
For engineers designing cable assemblies for EMC-critical applications (military, aerospace, medical, or high-speed telecommunications), IEC 61726 provides quantitative guidance for shield termination design. The standard recommends that the shield termination to the connector backshell achieves 360° contact with an impedance discontinuity of less than 1Ω. This is typically achieved through: (1) conductive ferrule compression termination, where the braid is captured between an inner ferrule and a compression sleeve; (2) solder-preform termination, where the braid is soldered directly to the connector body using a pre-formed solder ring; or (3) conductive adhesive termination using silver-loaded epoxy for applications where soldering temperatures are unacceptable.
Another important insight relates to the cable resonance effect. At frequencies where the cable length is an integer multiple of half-wavelengths, standing waves on the shield’s exterior surface create resonant current distributions that significantly increase radiation. IEC 61726’s requirement to move the absorbent clamp along the cable specifically addresses this — the maximum coupling position corresponds to a current anti-node of the resonant standing wave. Engineers can mitigate cable resonances by: (1) adding ferrite chokes at strategic positions along the cable; (2) optimising the cable length to avoid resonance at critical operating frequencies; or (3) using cables with higher shield transfer impedance to damp the resonance. The standard’s clamp-position sweep procedure provides the data needed to evaluate the effectiveness of these mitigation techniques.
Finally, IEC 61726 is increasingly relevant for high-speed digital data cables (USB 3.x, HDMI 2.1, Ethernet up to 25GBASE-T) where the fundamental clock frequencies or harmonics fall within the 30 MHz to 1 GHz test range. For these cables, the shielding attenuation requirements are driven not only by emission limits (FCC Part 15, CISPR 32) but also by immunity requirements in industrial environments with high RF field strengths. The standard provides a common measurement framework for evaluating and comparing the shielding performance of different cable assembly designs, enabling data-driven procurement decisions for EMC-sensitive applications.
Frequently Asked Questions
Q1: What is the difference between shielding attenuation (IEC 61726) and transfer impedance (IEC 62153-4)?
Shielding attenuation (as) measures the total radiated power coupled through the shield over the entire cable assembly, including connector effects. Transfer impedance (ZT) measures the shield’s intrinsic effectiveness per unit length and is more suited to characterising the cable alone. IEC 61726 is preferred for system-level EMC compliance testing, while transfer impedance methods are preferred for cable development and quality control. The two metrics are complementary — high ZT inevitably leads to low as, but poor connectors can make as much worse than ZT alone would suggest.
Q2: Can IEC 61726 be used for testing cables with non-50 Ω characteristic impedance?
Yes, with appropriate terminations. The standard requires that the cable assembly be terminated in its characteristic impedance at both ends. For non-50 Ω cables (e.g., 75 Ω, 93 Ω, 120 Ω differential pairs), the test setup must use impedance-matching networks or baluns at the connection points to the 50 Ω test equipment. The standard provides guidance on the correction factors required for non-50 Ω terminations.
Q3: How does cable length affect the shielding attenuation measurement?
Cable length significantly affects the measurement through standing wave resonances. IEC 61726 addresses this by requiring the absorbent clamp to be moved along the cable to find the maximum coupling position. For a given cable type, measurements on different lengths should ideally produce the same minimum shielding attenuation, provided the clamp covers at least one current anti-node. The standard recommends a minimum cable length of 2 metres for measurements down to 30 MHz, with longer cables needed at lower frequencies.
Q4: Is the IEC 61726 absorbent clamp method applicable to differential pair (balanced) cables?
Yes, but with additional setup. For balanced cables (e.g., Ethernet twisted pairs, automotive LVDS), the test requires a balun (balanced-to-unbalanced transformer) to convert the differential signal to a single-ended 50 Ω signal for the test equipment. The balun’s common-mode rejection ratio (CMRR) must be significantly higher than the expected shielding attenuation to avoid measurement errors. IEC 61726 provides specific guidance on balun selection and verification for balanced cable testing.
