EMC testing checklist before compliance testing

A laboratory booking can be consumed by a missing cable, an undocumented firmware build or an EUT that cannot sustain its worst-case operating mode. The measurement equipment may be ready, but the product is not.

This EMC testing checklist is intended for hardware engineers and project managers preparing for pre-compliance testing, an internal spectrum analyser investigation or a formal compliance programme. It focuses on the details that determine whether the resulting data is repeatable, diagnostically useful and suitable for inclusion in a defensible engineering evidence trail.

Define the compliance route before selecting equipment

Start with the product, its intended environment and its ports. Do not begin by hiring a spectrum analyser and deciding later what to measure. The applicable legislation, product standard or product-family standard determines the emissions limits, immunity requirements, port applicability, operating modes and test configuration.

For UK and European market access, the manufacturer or responsible economic operator must identify the applicable requirements and conformity assessment route. Some products may follow a self-certification route, but testing alone does not complete every obligation. The technical file, risk assessment, Declaration of Conformity and supporting records must also be addressed where applicable. The official UK guidance on the Electromagnetic Compatibility Regulations provides a regulatory starting point, although product-specific advice may still be required.

Generic EMC standards are not automatically suitable for every product. Product and product-family standards can specify different limits, performance criteria and configurations. IEC 61800-3, for example, addresses adjustable-speed electrical power drive systems, while CISPR 11 and CISPR 32 cover emissions within defined product scopes. The IEC 61000-4-x series provides basic immunity test methods, not general emissions limits.

Check the latest active edition, amendments, frequency ranges, test levels, limits, equipment arrangements and documentation requirements. Customer contracts may impose additional tests, particularly in defence, automotive and aerospace programmes.

Build a test matrix from ports and functions

A useful test plan maps each physical port to the applicable test method. List AC and DC power ports, signal and control lines, wired network interfaces, antenna ports, accessible enclosures and any long cables that could behave as coupling paths.

Keep emissions and immunity separate. A LISN provides a defined impedance and measurement port for conducted emissions testing on relevant power lines. It is not a conducted immunity device. Conducted RF immunity under IEC 61000-4-6 commonly uses a CDN or another coupling method permitted by the applicable standard. BCI probes belong to applicable current-injection immunity procedures used in automotive, military, aerospace and certain product-specific methods. A CDN and BCI probe are not interchangeable simply because both inject RF energy.

The initial matrix should identify:

• The applicable product or product-family standard and edition.
• Each EUT port and whether it is exercised, terminated or monitored.
• The required emissions and immunity methods.
• Frequency ranges, limits, test levels and performance criteria.
• Required support equipment, loads, simulators and ancillary devices.
• Operating modes and dwell times.
• Pass or fail observations that can be measured objectively and the appropriate performance criteria.

Without this matrix, teams frequently arrive with the wrong coupling network, insufficient amplifier coverage or no practical way to exercise a data port. The result is usually a shortened test programme or data that does not represent the intended product configuration.

Freeze and identify the equipment under test

Record exactly what is being tested. A model name is not enough when two samples contain different PCB revisions, filters or firmware.

• Product model, serial number and hardware revision.
• PCB assembly and revision identifiers.
• Firmware, FPGA, software and bootloader versions.
• Power supply type, rating, cable and manufacturer part number.
• Fitted modules, options, antennas and peripherals.
• Any engineering modifications, temporary connections or filters.
• Known deviations from the proposed production build.

Photograph internal and external configurations where permitted. Mark temporary modifications on the circuit diagram and test record. An undocumented ferrite added during debugging can turn a useful result into false confidence because the tested article no longer matches the production definition.

Choose operating modes that expose the real risk

An idle EUT usually produces an attractive emissions plot. It may also conceal the switching activity, bus traffic and cable currents that dominate normal operation.

Identify operating modes that maximise processor activity, display updates, motor commutation, power conversion, radio transmission, memory access and external data traffic. For immunity testing, choose functions that are sensitive to disturbance and can be monitored continuously. A product that silently resets and recovers may appear operational unless reset counters, communications logs or output tolerances are observed.

Automate mode cycling where possible, but make the cycle long enough for the receiver sweep or immunity dwell to capture each state. Record command scripts and settings. If an engineer manually changes modes during a scan without time stamps, correlating a disturbance with a particular function becomes guesswork.

Control cabling, loads and support equipment

Cables are part of the EMC configuration. Their type, length, shielding, termination and physical routing affect common-mode current and radiation efficiency. Replacing a specified screened cable with a convenient unscreened lead can create a failure that the production system would never experience. Tightly coiling excess cable can produce the opposite problem by reducing its effectiveness as a radiating structure.

Prepare a cable schedule with part numbers, lengths, connector types and shield termination details. Include photographs or a dimensioned sketch of the intended routing. Support equipment should be placed and filtered so that its emissions are distinguishable from those of the EUT.

Use representative loads. A power converter tested at a light resistive load may not enter the control mode associated with its highest disturbance. Motors, solenoids and long communications cables can also change both the source and coupling path. If the actual load is impractical in the laboratory, agree a technically defensible simulator before the booking.

Prepare for conducted emissions measurements

Conducted emissions measurements are commonly made from 150 kHz to 30 MHz on relevant power ports, but that range must not be assumed for every product standard. Confirm the required LISN type, supply rating, phase configuration, impedance network and protective arrangements.

Check that the EUT current, inrush and supply voltage are within the LISN and laboratory supply ratings. Establish how protective earth will be connected. An improvised earth lead can alter the RF return path, while an unnecessarily long bonding connection adds inductance and reduces high-frequency effectiveness.

A spectrum analyser can support development work, but its input must be protected against transients and overload. Suitable attenuation, limiting and DC blocking may be required depending on the LISN measurement port and instrument. Receiver settings, resolution bandwidth and peak, quasi-peak or average detection must follow the applicable method. A fast peak scan is useful for finding problems, but it cannot automatically replace final measurements using the specified detector and bandwidth.

Further setup considerations are covered in EMC Hire's conducted emissions system information.

Prepare for radiated emissions measurements

Radiated emissions commonly begin at 30 MHz, but the upper frequency, antenna arrangement, test distance and measurement environment depend on the product and standard. Confirm whether the planned system covers the full required range and whether its antennas, cables, preamplifiers and receiver have suitable characterised performance.

Preserve test-distance discipline. Moving the antenna or EUT to improve access changes the geometry and can invalidate comparison with an earlier scan. Cable placement also matters. A small change in the route of an I/O cable can shift resonances and change the measured field strength by enough to misdirect a debugging exercise.

Before scanning, check the analyser reference level, attenuation, preamplifier state and overload indication. Front-end compression can produce understated amplitudes, harmonics or apparently inconsistent results. An ambient scan with the EUT switched off helps identify broadcast, mobile radio and local digital sources, but it must use the same relevant setup and sufficient observation time.

See the radiated emissions system information and EMC equipment selection guides when defining the measurement chain.

Plan immunity monitoring before applying stress

An immunity test needs a defined performance criterion and an observation method. “The unit still works” is rarely adequate. Set acceptable limits for output error, communications loss, display corruption, resets, actuator movement and recovery time.

Monitoring equipment can create an unintended coupling path. Long oscilloscope leads, USB cables or unearthed instrumentation may alter the EUT response or expose support equipment to the applied disturbance. Fibre-optic links, isolated sensors or shielded monitoring arrangements can reduce that risk, provided they remain representative of the test plan.

Review safety and failure behaviour before applying ESD, surge, electrical fast transient, conducted RF or radiated RF immunity. The selected immunity test and level must come from the applicable product requirements. Do not apply a convenient generic level without considering the EUT ports, installation environment and manufacturer safety information.

Capture evidence that another engineer can reproduce

Save more than screenshots. A plot without configuration data is weak evidence because there is no reliable way to reconstruct the measurement.

• Instrument models, serial numbers and calibration status.
• Software versions and saved instrument state files.
• Detector, bandwidth, attenuation, sweep and preamplifier settings.
• Transducer factors, cable losses and correction files.
• EUT mode, load, supply conditions and environmental observations.
• Photographs showing cable routing, antenna position and bonding.
• Raw traces, final plots, marker data and ambient observations.
• Modifications made, their sequence and the result after each change.

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 evidence available to engineering teams, corporate stakeholders and regulatory reviewers. Calibration does not correct a poor test setup, so configuration control remains equally important.

Typical scenario

Consider an illustrative industrial controller with an external AC power supply, Ethernet, sensor cables and a switched motor output. The development team wants a quick spectrum analyser scan before booking formal testing.

The likely programme includes conducted emissions on the relevant power port using a suitable LISN and radiated emissions using an appropriate antenna system. Immunity requirements may include separate ESD, radiated RF, transient and conducted RF tests if called up by the applicable product standard. Each method needs its own equipment and configuration.

Before testing, the team must decide which motor load is representative, how Ethernet traffic will be sustained, which cable lengths match the intended installation and how resets will be detected. Selecting a low-current LISN, an antenna with inadequate frequency coverage or an analyser without sufficient overload protection could delay the work or damage equipment. Testing the controller while idle could miss emissions produced during motor switching and network traffic.

Early EMC pre-compliance testing allows the team to isolate dominant sources while PCB, enclosure and filter changes are still practical. EMC Hire can support equipment selection, hire, accessible test setups, on-site investigation, facility bookings and formal compliance testing where appropriate. Pre-compliance results provide engineering evidence and reduce uncertainty, but do not by themselves prove compliance.

When to Hire EMC Equipment

Hiring is technically sensible when demand is short-term, irregular or linked to a defined development milestone. It allows engineers to select a LISN, receiver, spectrum analyser, antenna system, probe or immunity generator for the actual method rather than forcing a previously purchased instrument into an unsuitable role.

Rental can avoid capital expenditure and the long-term costs of storage, servicing and calibration. It also helps teams scale capability during project peaks or run parallel investigations before a laboratory date. This reduces the risk of buying equipment whose frequency coverage, power rating or accessories do not suit later programmes.

The hire specification still needs engineering review. Provide the intended standard, frequency range, EUT power requirements, port types and test objective. EMC Hire can then help identify an appropriate configuration for the defined test window rather than supplying isolated instruments that do not form a workable measurement chain.

Common EMC Testing Mistakes to Avoid

Treating the first spectrum plot as a compliance result

A development scan may use peak detection, reduced dwell or a non-compliant environment. Presenting it as formal evidence can create false confidence. Record its limitations and use the specified detector, bandwidth, distance and site arrangement for the relevant final measurement.

Changing cable positions without recording them

Cable movement changes common-mode current distribution and radiation. If every debugging change includes a different cable route, the engineer cannot tell whether improvement came from the circuit modification or geometry.

Using the wrong coupling device

A LISN cannot substitute for a CDN, and a BCI probe cannot automatically replace either. The wrong device applies or measures a different physical quantity and impedance, producing invalid or non-comparable data.

Ignoring analyser overload

Strong out-of-band energy can compress the front end even when the displayed trace looks reasonable. Check attenuation, preselection, preamplifier state and overload indicators before trusting amplitudes.

Testing an unrepresentative operating mode

Idle processors and lightly loaded converters may suppress the mechanisms responsible for the highest emissions. Immunity monitoring that misses resets or packet loss can also produce a false pass assessment.

Leaving the evidence trail until later

Reconstructing cable positions, firmware versions and receiver settings from memory is unreliable. Missing configuration records weaken repeatability and may make the data unsuitable for the technical file or formal test planning.

Frequently Asked Questions (FAQs)

Can a spectrum analyser replace an EMI receiver?

It can be useful for pre-compliance investigation if its frequency coverage, dynamic range, detectors, bandwidths and overload performance suit the task. Formal measurements must meet the applicable standard's instrumentation and detector requirements, so suitability must be confirmed rather than assumed.

How many product samples should we prepare?

At least one fully functional, configuration-controlled EUT is needed, with spares where failure or destructive stress is plausible. Additional samples can protect the schedule, but differences between samples must be documented.

Should firmware be frozen before pre-compliance testing?

Use an identified build that exercises representative worst-case functions. Firmware may still change afterwards, but each build used for testing should be archived with its settings and release notes so later results can be compared properly.

Does passing pre-compliance mean the product complies?

No. Pre-compliance improves confidence and supports debugging, but setup limitations, measurement uncertainty and incomplete test coverage may remain. Formal testing and the wider conformity assessment process must follow the applicable requirements.

What should be sent to the laboratory before booking?

Send the proposed standards, product description, port list, power requirements, dimensions, operating modes, cable schedule, support equipment list and any previous plots. Early review often identifies missing loads, unsuitable samples or unrealistic test durations.

Can EMC Hire support defence, automotive or aerospace projects?

EMC Hire can provide pre-compliance equipment and engineering support for these sectors. Final testing may need an appropriately accredited laboratory depending on contractual, regulatory or programme requirements.

Plan the test window with defensible inputs

A well-prepared EMC testing checklist protects laboratory time and improves the value of every measurement. If you need help selecting equipment, reviewing a test setup, arranging on-site testing, booking space at the EMC Hire test facility or planning pre-compliance and formal compliance work, speak with the EMC Hire engineering team.

Call +44 (0)1462 817111 or email sales@emchire.co.uk to discuss the EUT, applicable methods and required test window, or to request an equipment hire quotation.