Why automotive harness length changes test results

A harness shortened to make the bench tidier can move a resonance, change injected current and turn a repeatable automotive EMC result into misleading data.

The cable is not merely an interconnection. At EMC frequencies, the harness, ground plane, loads, enclosures and return paths form a distributed electrical structure whose behaviour depends on geometry.

Why cable length changes the electrical system

At low frequency, engineers can often treat a cable as a lumped resistance with some inductance and capacitance. That approximation deteriorates as the cable length becomes a meaningful fraction of the wavelength. Voltage and current are then no longer uniform along the harness. Reflections, standing waves and position-dependent coupling begin to dominate.

Free-space wavelength is given by λ = c/f, but propagation along a real harness is slower and more complicated. Wire insulation, spacing, twisting, shielding, proximity to a conductive plane and coupling between conductors all affect propagation velocity. The electrically significant length is also not limited to the visible cable. Connector tails, printed circuit board tracks, enclosure bonds, load wiring and return conductors can all become part of the RF path.

A nominal quarter-wave relationship is useful for diagnosis, but it is not a reliable pass or fail calculation. A harness does not behave like an ideal transmission line with a single velocity factor and two well-defined terminations. Branches, unterminated conductors and nonlinear loads create multiple resonant modes.

Common-mode resonance usually causes the larger surprise

Differential signals travel out and back through an intended conductor pair. Common-mode current returns through stray capacitance, shielding, mounting hardware, the vehicle structure or the laboratory ground plane. That return path may be physically broad and poorly controlled.

Changing cable length alters the inductance and capacitance in this common-mode loop. It can move a current maximum towards an electronic control unit connector, place a current null beneath an injection probe or increase the RF voltage developed across an imperfect enclosure bond. The resulting change can be many times larger than expected from a simple cable-loss calculation.

Harness length also affects emissions. A longer common-mode conductor can become a more efficient antenna at some frequencies, but longer does not automatically mean worse. Moving away from one resonance may reduce radiation while approaching another increases it. That is why an arbitrary short harness can produce reassuring pre-compliance data that does not represent the vehicle installation or the formal component test configuration.

Harness geometry is part of the test configuration

Length cannot be separated from layout. Harness height above the reference plane, lateral route, bends, branch positions, connector orientation and distance from chamber absorbers or antenna structures all alter coupling.

A few centimetres of movement may have little effect at lower frequencies yet become significant where standing-wave current varies rapidly with position. Coiling surplus cable is particularly damaging. A coil introduces mutual inductance and inter-turn capacitance, producing a compact resonant component rather than a longer representative harness. Folding a harness back on itself can also create cancellation or enhanced coupling depending on current direction.

Shield termination deserves equal attention. A circumferential connector bond and a long pigtail are not electrically equivalent. Pigtail inductance raises shield-transfer impedance at higher frequencies, allowing greater common-mode voltage between the shield and enclosure. If the harness length is changed at the same time, the engineer can no longer tell whether a result moved because of resonance or shield termination.

Termination impedance determines where current flows

An automotive load simulator, peripheral module or artificial network may present a controlled impedance only over part of the frequency range. Outside that range, connector capacitance, wiring inductance and component protection networks become influential. Powered and unpowered states can differ substantially.

Electronic control unit inputs containing transient protection, common-mode filtering or switched semiconductor junctions may also be nonlinear. RF immunity can therefore depend on modulation, injected level and operating state. A changed harness length modifies the voltage presented to those nonlinear elements, sometimes exposing a susceptibility that was absent with the previous layout.

Injection results depend on probe position

Bulk current injection uses a current injection probe to couple RF energy onto the applicable harness in methods specified by automotive, military, aerospace or product-specific procedures. It is a conducted immunity technique. The probe is not interchangeable with a coupling and decoupling network used for IEC 61000-4-6 testing.

In an automotive harness EMC test, the current produced by a given forward power depends on the injection probe, calibration fixture and cable impedance presented during the substitution or closed-loop method defined by the applicable procedure. Once installed around the real harness, the probe sees an impedance that varies with frequency, cable length and termination.

Probe placement matters because current is position-dependent on a resonant harness. Moving the probe may place it nearer a current node or antinode. Cable between the probe and equipment under test also develops RF voltage, so changing that distance can alter the stress applied at the connector even when monitored current appears similar.

Monitor-probe position, orientation and spacing from the injection probe must remain controlled. Excessive coupling between probes can make the monitoring channel respond directly to the injection field rather than only to harness current. That produces false confidence in the applied stress.

ISO 11452 contains component immunity methods, with different parts addressing different coupling environments. ISO 11452-4 covers harness excitation methods including bulk current injection. ISO 11452-2 addresses absorber-lined shielded enclosure radiated immunity. These methods should not be treated as equivalent simply because both expose an automotive component to RF energy.

Radiated immunity and emissions are equally geometry-sensitive

During radiated immunity testing, the harness can be the dominant receiving structure. The incident electric field induces common-mode voltage and current according to harness orientation, polarisation, height, termination and electrical length. A change in route can therefore alter susceptibility without any modification to the electronic control unit.

Radiated emissions behave in the reciprocal direction. Common-mode current driven onto the harness by internal switching noise can radiate efficiently when the cable geometry supports it. The relevant product or customer standard, such as CISPR 25 where applicable, defines the measurement method and limits. A spectrum analyser sweep with an improvised cable arrangement may identify noise sources, but it is not automatically representative compliance evidence.

Conducted transient immunity under applicable parts of ISO 7637 presents another mechanism. Pulse delivery, source impedance, coupling arrangement and cable configuration influence the waveform reaching the equipment under test. The harness should not be changed casually to solve overshoot or ringing. Doing so can conceal a setup problem or create a configuration that departs from the approved test plan.

Always verify the latest active edition of the relevant standard, product-specific requirements, test levels, frequency ranges, harness configuration, equipment arrangement, limits, performance criteria and documentation rules. Customer specifications frequently add conditions beyond the base ISO method.

Typical scenario

Consider an illustrative electronic control unit undergoing pre-compliance bulk current injection testing. The first bench uses a conveniently shortened harness. A later test uses the specified production-representative length routed over the reference plane, and a functional disturbance appears over a narrow frequency band.

The tempting diagnosis is poor repeatability from the amplifier or injection probe. A more useful investigation would compare the controlled variables: total cable length, branch geometry, probe-to-connector distance, load impedances, bonding, power supply arrangement and operating mode. Measuring harness current at selected positions may show that the longer assembly has moved a current maximum towards the equipment connector.

The team must then decide whether the disturbance represents a genuine design weakness, an unrepresentative setup or both. Repeating the test with documented layouts and controlled length increments can separate a broad immunity problem from a narrow harness resonance. Current-probe measurements and near-field investigation around the connector may identify whether RF is entering through I/O lines, the supply network or enclosure seams.

Early work is cheaper than discovering the same mechanism during a formal programme. Filters, connector grounding and printed circuit board return paths can still be changed without disrupting validated tooling or production documentation.

EMC Hire can support this work through automotive EMC equipment hire and testing support, suitable test system selection, pre-compliance investigation, on-site testing and access to test facilities. Details of available approaches can be discussed through the automotive test systems information service. Where formal testing is appropriate, the objective is to build a controlled and defensible evidence trail rather than simply repeat an unexplained failure.

Controlling repeatability without hiding real variation

A repeatable test needs a defined harness configuration. Record more than nominal cable length. Photographs should show routing, branch points, bends, connector orientation, support materials, ground-plane position and probe locations. Marking the bench or ground plane helps restore positions after diagnostic changes.

Useful records normally include:

• Harness part number, revision, total length and branch lengths.
• Wire type, twist, shielding and shield termination method.
• Height and lateral route relative to the reference plane.
• Injection and monitoring probe positions measured from a defined connector datum.
• Load simulator configuration, peripheral equipment and termination states.
• Equipment under test software, operating mode and monitored functions.
• Amplifier, directional coupler, probes, receiver and calibration identifiers.
• Forward power, monitored current and limiting behaviour where the method requires them.

Use RF interconnects with adequate power handling, shielding and frequency performance. Damaged connectors or unstable cable assemblies can create frequency-selective loss that resembles a harness resonance. EMC Hire can provide guidance on RF and microwave cables and connectors and suitable RF coaxial cable assemblies for measurement and injection systems.

Where calibration is relevant, EMC Hire uses test equipment with calibration traceable through an appropriate ISO/IEC 17025 accredited calibration provider. Suitable traceable calibration supports measurement accuracy, repeatability and comparison between development and formal testing. It also strengthens the recorded evidence available to engineering teams, corporate stakeholders and regulatory reviewers.

When to Hire EMC Equipment

Automotive EMC programmes often need high-power RF amplifiers, injection probes, current monitoring probes, directional couplers, power meters and transient generators for a limited development window. Purchasing a system for irregular demand ties up capital and transfers calibration, servicing, storage and obsolescence risk to the engineering team.

Hiring is technically sensible when the required frequency range, power or coupling device changes between programmes. It also allows a team to scale capability during prototype peaks without buying equipment that may be unsuitable for the next customer specification.

Equipment selection still needs care. An amplifier must provide usable power across the required range with allowance for system losses and mismatch. Probes need appropriate frequency coverage and current or power capability. Coaxial cables, attenuators and couplers must withstand the intended stress. Hiring the wrong nominally similar item can cause amplifier levelling, distorted waveforms or inadequate injection before the requested test level is reached.

For teams without a suitable ground plane, chamber or instrumentation chain, booking a facility or arranging on-site support may be more controlled than assembling an improvised bench. Accessible pre-compliance testing allows prototypes to be debugged early and can provide calibrated engineering data supporting technical documentation and self-certification decisions where legally and technically appropriate. It does not by itself prove compliance.

Common EMC Testing Mistakes to Avoid

Removing excess cable by coiling it

A coil changes inductance, capacitance and magnetic coupling. Results may become repeatable for the coil but irrelevant to the specified harness layout.

Moving the injection probe without updating the record

Probe position changes coupling into a standing-wave current distribution. A shifted susceptibility threshold may then be blamed on the equipment under test rather than the changed injection geometry.

Ignoring the RF return path

Loose enclosure bonds, long ground straps and inconsistent ground-plane connections alter common-mode impedance. At higher frequencies, strap inductance can dominate its DC resistance and make two apparently grounded setups electrically different.

Using an unrepresentative operating mode

A static diagnostic state may not exercise the communication, actuator or sensor paths most susceptible to RF. Passing that mode creates false confidence and leaves the vulnerable state untested.

Changing several variables together

Shortening the harness while adding ferrite, changing shield bonding and rerouting the supply may remove a failure, but it destroys diagnostic traceability. Change one controlled variable at a time unless the purpose is an explicitly documented configuration comparison.

Failing to capture limiting conditions

During BCI testing, the system may reach a forward-power limit before achieving the target monitored current, depending on the prescribed method. Recording only a pass or fail loses the information needed to reproduce the exposure and understand whether the system was power-limited.

Frequently Asked Questions (FAQs)

Can a shorter harness always improve automotive EMC?

No. Shortening may reduce antenna efficiency at one frequency while moving a resonance into a more sensitive band at another. It can also move current maxima towards the equipment connector. Test against the defined harness configuration rather than assuming shorter is better.

How much cable-length variation is acceptable?

There is no universal tolerance. Use the applicable standard, customer specification and approved test plan. If the documentation permits a range, assess whether that range crosses a resonance relevant to the product. A controlled sweep of representative lengths can be useful during development.

Why does BCI current change when the same forward power is applied?

The real harness presents a frequency-dependent impedance to the injection probe. Cable length, termination, routing and nearby conductive structures affect that impedance, so equal forward power does not guarantee equal harness current.

Should harness resonance be damped during pre-compliance testing?

Only if the damping method represents the intended vehicle installation or is being evaluated as a design mitigation. Adding ferrite solely to stabilise the bench may hide a real susceptibility and produce non-representative data.

Can pre-compliance results support a formal automotive programme?

They can support design decisions, test planning, risk assessment and mitigation evidence. Final testing may need an appropriately accredited laboratory where contractual, regulatory or programme requirements demand it. The responsible manufacturer or programme authority must confirm the applicable route and documentation.

What is the best way to distinguish harness resonance from an ECU defect?

Repeat the exposure with controlled changes to cable length or probe position while keeping terminations, operating mode and bonding fixed. Correlate monitored harness current with functional behaviour and use local current or near-field measurements around the connector. A response that tracks geometry strongly suggests a coupling-path resonance, although the ECU may still require improved immunity.

Plan the harness before booking the final test

Automotive harness EMC repeatability comes from controlling the physical RF structure, not merely repeating instrument settings. Define the harness, layout, return paths, operating modes and probe positions before formal testing, then preserve enough evidence to reconstruct them.

EMC Hire can help review the setup, select equipment, provide a hire quotation, arrange on-site testing, support pre-compliance investigation or discuss formal compliance testing where appropriate. Test facility space can also be booked for controlled development work. Contact the engineering team on +44 (0)1462 817111 or email sales@emchire.co.uk.