How to validate a rail EMC test setup

A rail EMC result can move several decibels without any change to the equipment under test. Cable geometry, bonding, auxiliary equipment and operating mode are often the real variables.

Validating a rail EMC test setup means proving that the configuration represents the intended installation, follows the applicable test plan and can be reproduced by another competent laboratory. A photograph and a statement that the test followed EN 50121 are not enough.

Start with the applicable rail EMC requirements

The EN 50121 series addresses electromagnetic compatibility within the railway environment, but its different parts cover different equipment and locations. Requirements for apparatus installed on rolling stock are not automatically the same as those for signalling and telecommunications equipment or fixed power installations.

Begin by identifying the product function, installation location, enclosure classification, power interface, signal ports and intended electromagnetic environment. The applicable product or contractual specification may call up emissions and immunity methods while adding installation-specific configurations, performance criteria or test levels.

Rail programmes may also reference EN 50155 for electronic equipment used on rolling stock, customer specifications or project-specific EMC control plans. Those documents can affect the operating state, supply conditions and acceptance criteria even when the underlying measurement method comes from another standard.

Always verify the latest active editions, amendments, national adoption, test levels, frequency ranges, limits and documentation requirements. The British Standards Institution and CEN and CENELEC provide authoritative information on standards development and publication. A general overview of how product, product-family and basic test standards interact is available in EMC Hire's guide to EMC standards explained.

Define the configuration before entering the laboratory

The EUT configuration should be frozen before formal measurements begin. That does not mean every possible operating condition must be tested. It means the engineering team has identified and justified the modes most likely to produce the highest emissions or lowest immunity margin.

Record the hardware revision, firmware version, enabled interfaces, software load, power source, peripheral equipment and cable set. For modular equipment, state which cards are fitted and which slots are unused. Open ports should be terminated or left open according to the representative installation and applicable test plan, not laboratory convenience.

Rail equipment often changes electrical behaviour with operating state. A traction-related controller may have quiet standby periods, high switching activity during simulated demand and intermittent communications bursts. Measuring only the easiest stable state can produce a clean plot that says little about the foreseeable worst case.

Where normal operation depends on external controllers, simulators or loads, decide whether they belong inside or outside the measurement environment. Auxiliary equipment can radiate, conduct disturbances onto shared cables or alter the EUT's common-mode impedance. Its contribution must be controlled and, where possible, characterised.

Control cable routing, bonding and reference planes

Cables are part of the EMC test article. Their length, height, separation, termination and relationship to the reference plane directly affect common-mode current and antenna efficiency.

Use the routing specified by the applicable method and test plan. Record excess cable treatment, support material, bend radius, connector orientation and the distance between power and signal harnesses. Avoid tightly coiling spare cable unless the method specifically requires it. A coil changes inductance and coupling, potentially suppressing one resonance while creating another.

Bonding deserves equal attention. Paint, anodising, oxide layers and loose fixings can turn an apparently solid chassis connection into a frequency-dependent impedance. Measure or otherwise verify the intended bond, then document the bonding hardware and contact preparation. An unnecessarily long earth strap adds inductance. At higher frequencies, that can divert common-mode current through signal screens or laboratory instrumentation and change both emissions and immunity behaviour.

The ground plane or reference plane must have suitable dimensions, construction and bonding for the method in use. Adding an unplanned bench mat, insulating sheet or timber support changes capacitance to the plane. Even a small height change can influence cable coupling, particularly where long harnesses dominate the response.

Distance is a controlled measurement quantity

Radiated measurements depend on defined geometry. Antenna distance should be measured from the reference point required by the applicable method, not estimated from a convenient wall marker. EUT boundary, antenna reference point and permitted scanning volume need to be unambiguous.

Also record antenna height, polarisation, turntable angle and the position of large auxiliary items. A test distance error changes the measured field relationship and can invalidate comparison with the specified limit. Moving a support computer or load cabinet can alter reflections and cable currents even if the nominal EUT-to-antenna distance remains correct.

For conducted measurements, physical placement still matters. A LISN is used for conducted emissions measurements on relevant power ports. Its bonding, connection length and relationship to the reference plane affect RF impedance and measured disturbance voltage in dBµV. It must not be treated as a generic power filter or used as a conducted immunity coupling device.

Conducted RF immunity commonly uses a suitable coupling and decoupling network, such as a CDN, where required by the selected method. BCI probes apply current injection in applicable rail, automotive, military, aerospace or product-specific procedures. A CDN and BCI probe are not interchangeable simply because both introduce RF energy onto cables.

Validate instrumentation and the complete signal chain

An in-date calibration label is only the starting point. Confirm that the receiver, analyser, antenna, LISN, CDN, injection probe, amplifier, directional coupler and monitoring equipment cover the required ranges and levels. Cable loss, transducer factors, attenuators and preamplifier gain must be represented correctly in the measurement system.

For emissions work, receiver detector and resolution bandwidth settings must follow the applicable requirement. Peak, quasi-peak and average detectors serve different purposes. Conducted emissions are commonly assessed over 150 kHz to 30 MHz and radiated emissions commonly start at 30 MHz, but those ranges are not universal. The relevant rail standard, product specification and test plan govern the actual range and method.

Immunity validation should confirm the delivered stress at the calibration plane or injection path defined by the test method. Radiated immunity levels are expressed as electric-field strength in V/m. Conducted current injection is not. Mixing these quantities in a test sheet is a warning that the setup or reporting template has not been properly reviewed.

EMC Hire uses test equipment with calibration traceable through an appropriate ISO/IEC 17025 accredited calibration provider. Suitable traceable calibration supports repeatability, confidence in recorded data and comparison between development and formal testing. The accreditation applies to the calibration provider or activity, not to the physical instrument itself.

Use logging that explains what the plots cannot

Automated data capture is useful, but it rarely records enough about the EUT. Laboratory logging should correlate measured disturbances or immunity responses with operating state, command sequence, communication activity and supply conditions.

A defensible record normally includes:

• EUT identity, serial number, hardware and firmware revisions
• Port allocation, cable type, length, screening and termination
• Operating modes and dwell periods
• Power supply voltage, frequency and source arrangement
• Bonding points, support height and reference-plane configuration
• Antenna, probe, CDN or LISN identity and position
• Receiver settings, detector, bandwidth and correction data
• Ambient scans, photographs and annotated layout drawings
• Observed degradations, recovery behaviour and operator actions
• Deviations from the approved test plan and their engineering justification

Synchronised timestamps help when intermittent events occur. Without them, a burst in the emissions trace may be blamed on the EUT even though it coincided with an auxiliary supply switching state. Likewise, an immunity upset cannot be investigated properly if the applied frequency, dwell time and EUT telemetry were recorded on unrelated clocks.

Typical scenario

Consider an illustrative rolling-stock control unit with Ethernet, discrete I/O, a screened vehicle-bus connection and a nominal DC power input. The unit operates correctly on the bench but has not yet been exercised in a representative rail EMC test setup.

The test team must decide which ports are active, what cable lengths represent installation practice, how screens are bonded and which operating sequence creates the highest processing and communications activity. It must also determine whether the DC source, network simulator and loads could affect the measurement.

If the first formal setup uses short cables and direct screen bonds while the vehicle installation uses longer harnesses and pigtail terminations, the laboratory result may be optimistic. Recreating the installed geometry late in the programme can expose common-mode resonances that require enclosure, connector or filtering changes.

An early pre-compliance investigation allows current probes, near-field probes and representative emissions measurements to identify dominant paths before the design is frozen. Immunity checks can then explore susceptible interfaces using the method appropriate to the project. These results improve engineering confidence but do not automatically prove conformity.

EMC Hire can support equipment selection, pre-compliance engineering, formal compliance testing where appropriate, on-site testing and access to test facilities. For self-certification routes, calibrated engineering data may contribute to the technical file, EMC risk assessment, mitigation evidence and Declaration of Conformity process. The manufacturer remains responsible for identifying the applicable legislation, standards and conformity assessment route. Further context is available in the guides to commercial CE marking support and CE marking responsibilities.

When to Hire EMC Equipment

Hiring makes technical and financial sense when the requirement is specialised, intermittent or tied to a defined project window. A compliance laboratory may need an additional receiver, LISN, CDN, amplifier or monitoring instrument to run parallel programmes without buying equipment that could be unsuitable for the next contract.

Rental can also bridge a calibration or repair period and support temporary on-site work. It avoids long-term storage, servicing and calibration overheads while allowing the selected equipment to match the required frequency range, power, impedance, connector arrangement and test method.

Selection still needs engineering review. Hiring a high-power amplifier is of little use if its gain, mismatch tolerance or harmonic performance does not support the required immunity chain. Similarly, a current probe with inadequate transfer impedance or aperture can compromise diagnostic measurements.

EMC Hire can help define a suitable package for the test window, including practical setup considerations. This approach covers short-term peaks without unnecessary capital expenditure and reduces the risk of purchasing equipment around one programme's configuration.

Common EMC Testing Mistakes to Avoid

Treating installation drawings as test instructions

Vehicle drawings may omit cable height, excess-length treatment and screen termination detail. Translating them directly into a laboratory setup leaves variables uncontrolled, so another test team may not reproduce the result.

Changing cable routing between scans

Moving a cable to reach a connector can change common-mode current and radiated coupling. If movement is unavoidable, photograph and measure the revised position, then repeat the affected scan.

Using the wrong coupling device

A LISN supports conducted emissions measurement on relevant power ports. A CDN supports conducted RF immunity where specified, while a BCI probe supports applicable current-injection methods. Substitution without method review produces a different coupling path and potentially invalid data.

Ignoring ambient signals

Rail radiated emissions work can encounter broadcast, mobile and local industrial signals. An ambient scan and controlled EUT on-off comparison help prevent external signals being reported as product emissions.

Recording only the passing mode

Logging a nominal operating state while omitting communications loading, switching sequences or fault recovery can create false confidence. Test modes should be technically justified and traceable to foreseeable service operation.

Applying generic immunity criteria

Performance criteria should come from the applicable product requirement and agreed test plan. A momentary telemetry error, reset or latched output may have very different significance in rail service. For background on generic immunity standards and their limits of applicability, see EN 61000-6-1 and EN 61000-6-2 generic immunity.

Frequently Asked Questions (FAQs)

How many photographs are needed to validate the setup?

Use enough images to reconstruct every significant geometry. Include wide views, each EUT face, cable terminations, bonding points, auxiliary equipment, antenna or coupling-device position and measurement scales. Annotated drawings should supplement photographs where dimensions are not visually clear.

Should cable positions be measured or simply marked on the bench?

Measure and record the dimensions that affect the test method. Bench markings are useful during a session but may disappear or be misinterpreted. A dimensioned layout provides a stronger long-term record.

Can pre-compliance data be placed in the technical file?

Calibrated pre-compliance data can provide useful engineering and mitigation evidence where appropriate, but its limitations must be stated. It does not automatically replace formal testing or satisfy every contractual or regulatory requirement.

How should configuration changes during testing be handled?

Assign each material change a configuration identifier. Record why it was made, which results preceded it and which tests were repeated. Combining plots from different configurations without clear labelling destroys traceability.

Does a repeatable setup guarantee a compliant product?

No. Repeatability shows that controlled conditions produce consistent results. Compliance depends on the applicable requirements, representative configuration, valid methods, limits, performance criteria and the wider conformity assessment process.

When might an accredited laboratory be required?

Rail customer contracts, procurement rules or programme assurance plans may require testing by an appropriately accredited laboratory. Confirm this before booking work. EMC Hire can provide pre-compliance support and compliance testing for suitable self-certification applications, without implying that every programme follows that route.

Plan the setup before the test slot

A short setup review before equipment arrives can prevent days of disputed measurements. EMC Hire's engineering team can discuss rail EMC test setup configuration, equipment hire, pre-compliance work, formal compliance testing where appropriate, on-site testing and test-facility bookings.

To review a forthcoming programme or request an equipment hire quotation, call +44 (0)1462 817111 or email sales@emchire.co.uk.