How to read an EMC test report

A pass statement can be misleading if the wrong operating mode, cable arrangement or limit was used. The result only has meaning when the measured data and test configuration are read together.

An EMC test report should provide enough evidence for a technically competent reviewer to understand what was tested, how it was tested and whether the conclusions follow from the recorded results. Reading only the front-page verdict misses most of the engineering information.

Start with the report scope, not the result

First confirm the identity of the equipment under test, usually abbreviated to EUT. Model numbers, hardware revisions, firmware versions, power supplies, accessories and representative peripheral equipment should be stated. A result from one build configuration cannot automatically be assigned to another.

This becomes particularly significant when a product family shares an enclosure but uses different processors, switch-mode converters, display interfaces or radio modules. The highest-risk variant may not be the most fully featured model. A lower-cost version with a different power supply can produce a worse conducted emissions result.

Check whether the report covers a complete product, a subassembly or an engineering prototype. Also identify any exclusions. A statement such as “USB interface not exercised” can materially restrict the value of the evidence if that port is available during normal use.

The report should identify the standards and editions applied. Generic EMC standards are not always the right starting point. Product and product-family standards may define different limits, immunity levels, port applicability, configurations and performance criteria. The manufacturer remains responsible for confirming the relevant legislation, standard editions, product scope and conformity assessment route.

Understand what the limits actually apply to

An emissions table usually contains frequency, measured level, applicable limit and margin. Those columns must use the correct physical quantity. Conducted disturbance voltage is commonly reported in dBµV, while radiated electric-field strength is commonly expressed in dBµV/m. Treating these as interchangeable indicates a reporting or interpretation error.

Conducted emissions measurements on relevant power ports commonly use a line impedance stabilisation network, or LISN. Many standards assess frequencies from 150 kHz to 30 MHz, although that range is not universal. Radiated emissions commonly begin at 30 MHz, with the upper frequency determined by the applicable requirements and characteristics of the EUT.

The limit line may vary with frequency. Do not compare a measurement against a single value copied from another part of the band. Check whether the result relates to Class A, Class B, a specific installation environment or another product classification. Applying a more relaxed industrial limit to equipment intended for a residential environment can produce a paper pass that does not support the intended application.

Detector selection changes the result

Peak, quasi-peak and average detectors are not interchangeable. Peak scanning is often used to find candidate emissions efficiently, while final measurements may require quasi-peak or average detection according to the applicable standard. A peak result below a quasi-peak limit can be useful evidence because a subsequent quasi-peak value would not normally exceed the peak reading, provided the measurement chain and settings are valid. The reverse assumption is unsafe.

Resolution bandwidth also matters. CISPR measurement bandwidths depend on the frequency range and method. If an analyser bandwidth is too narrow, it may under-report broadband disturbances. If it is too wide, the noise floor and measured amplitude can increase. The report should identify receiver settings or reference a controlled procedure that defines them.

Margin is more informative than pass or fail

Margin is normally the difference between the applicable limit and measured level. Reports may express it as a positive margin for a pass or as a negative number when the limit is exceeded, but some reporting systems use the opposite convention. Read the column definition before interpreting the sign.

A result 0.5 dB below a limit and one 15 dB below it are both technically passes under a simple binary decision rule. They do not carry the same engineering risk. Small margins can be consumed by production tolerance, cable positioning, component substitutions, ambient conditions or differences between test sites.

Near-limit results deserve closer inspection:

• Was the final detector used, or is the value from a preliminary peak scan?
• Has transducer correction, cable loss, preamplifier gain and LISN or antenna factor been applied correctly?
• Was the EUT orientation and cable placement varied to maximise emissions?
• Is the reported margin smaller than, similar to or comfortably greater than the stated measurement uncertainty?
• Could normal production variation change the emission mechanism?

Margin should not be converted into an unsupported claim about manufacturing yield. It is better treated as a risk indicator that helps decide whether further investigation or design improvement is justified.

Measurement uncertainty and decision rules

Measurement uncertainty describes the range associated with a reported measurement, based on contributions from equipment, transducers, calibration, site performance, repeatability and other factors. It is not an allowance that engineers may simply subtract from a failing result.

The report should state how uncertainty was handled when making the conformity decision. Different standards, accreditation arrangements, contractual specifications or laboratory decision rules may treat uncertainty differently. A statement of compliance without an identified decision rule can be ambiguous, particularly where a result lies close to the limit.

For an accredited test, review whether the reported activity falls inside the laboratory's accredited scope rather than assuming that every service offered by an accredited organisation is covered. UKAS information on laboratory accreditation provides useful context on ISO/IEC 17025 and laboratory competence.

EMC Hire uses test equipment with calibration traceable through an appropriate ISO/IEC 17025 accredited calibration provider where calibration is relevant to the measurement. Suitable traceability supports repeatability, confidence in recorded data and comparison between development measurements and later formal testing. It does not, by itself, make a test accredited or prove product compliance.

The setup photographs are engineering evidence

A technically useful EMC test report documents the physical setup, not merely the instrument list. Photographs should show EUT orientation, cable routing, support equipment, ground plane relationships, antenna polarisation where relevant and the position of auxiliary equipment.

Cables are often efficient common-mode radiators. Moving a cable from the ground plane edge, changing its height or coiling excess length can shift resonances and alter a radiated emissions plot. A photograph that hides the cable bundle behind the EUT makes exact reproduction difficult.

For conducted emissions, confirm which power port was connected through the LISN, how auxiliary equipment was powered and whether unused LISN ports were terminated correctly. For conducted RF immunity under IEC 61000-4-6 where called up by the applicable product standard, the report should identify the coupling device, such as a suitable CDN, and the relevant port configuration. A BCI probe belongs to applicable current-injection immunity methods and should not be treated as a universal replacement for a CDN.

Radiated immunity reports should document antenna polarisation, field calibration approach, frequency range, dwell or exposure conditions, modulation and applied field strength in V/m. Those values cannot be inferred from a radiated emissions plot in dBµV/m because emissions and immunity are separate test phenomena.

Operating modes and performance criteria

The EUT must be exercised in a mode that exposes credible worst-case EMC behaviour. “Powered and operational” is not enough. A useful report describes processor load, data traffic, motor speed, converter operating point, radio activity, display content and representative port loading where these affect the product.

During immunity testing, the monitored functions and pass or fail criteria need to be defined before testing. Product standards may permit temporary degradation under particular performance criteria, while loss of stored data or an unsafe state may remain unacceptable. A narrative comment such as “minor display disturbance observed” cannot be evaluated without the agreed performance criterion and recovery behaviour.

Software monitoring also needs scrutiny. A product that continues responding to a simple heartbeat may still have corrupted measurements, lost packets or entered a degraded control mode. Recorded telemetry and independent monitoring generally provide stronger evidence than visual observation alone.

Measured data versus narrative comments

Tables, plots and instrument records contain measured data. Engineer comments provide context. Both are useful, but they serve different purposes.

A note stating “emission attributed to the external power adaptor” is a diagnosis, not a measurement. Look for supporting evidence such as a frequency correlation, an alternative supply comparison or current-probe investigation. Similarly, “no degradation observed” is stronger when the report identifies what was monitored, the monitoring resolution and the acceptance criterion.

Failure notes should record the frequency or test condition, observed EUT behaviour, affected function, recovery method and any configuration changes. If a ferrite, filter or firmware adjustment was introduced during the session, results before and after the change must remain distinguishable. Otherwise, the final report can accidentally imply that the original production configuration passed.

Typical scenario

Consider an illustrative mains-powered controller undergoing external testing for conducted and radiated emissions. The report states that it passed, but the conducted emissions margin is less than 2 dB at several frequencies. Photographs show the communication cable disconnected even though the interface is normally active in service.

The engineering team must establish whether the applied product standard and limits are correct, whether the disconnected interface invalidates the intended operating mode, and whether final quasi-peak or average measurements were made where required. They should also review measurement uncertainty and determine whether the narrow margin creates an unacceptable production risk.

Recreating the setup early may reveal that connecting and exercising the communication cable raises common-mode emissions. Finding this before technical documentation is finalised leaves time to investigate bonding, filtering and cable current rather than redesigning during a formal programme.

EMC Hire can support that work through equipment hire, pre-compliance engineering, on-site testing or access to an EMC laboratory for hire. Where appropriate, formal compliance testing can then produce controlled evidence for the technical file, Declaration of Conformity and wider compliance assessment. Testing alone does not complete every CE or UKCA obligation, and the manufacturer or responsible economic operator must confirm the applicable route.

When to Hire EMC Equipment

Hiring is technically sensible when the measurement requirement is short-lived, project-specific or outside the organisation's normal capability. Purchasing a receiver, LISN, CDN, transient generator or calibrated RF system for one development cycle ties up capital and creates continuing storage, servicing and calibration obligations.

Equipment selection still needs care. Frequency coverage, power rating, coupling path, software options and accessories must match the test method. Hiring an analyser without the appropriate transducer, limiter or preamplifier can leave a team with an incomplete setup and potentially exposed front-end hardware.

Rental also allows capability to scale during project peaks without buying equipment that may be unsuitable for later programmes. EMC Hire can help define a suitable configuration for a planned test window, while its EMC facility overview and test facility hire options provide alternatives when a controlled environment is more practical than an in-house setup.

For focused debugging, a short EMC facility session with engineering support can help distinguish a genuine product emission from an unsuitable bench arrangement. Accessible pre-compliance work can gather calibrated engineering data and reduce redesign risk, but it should not be presented as automatic proof of compliance.

Common EMC Testing Mistakes to Avoid

Reviewing only the summary page

A pass statement can conceal excluded ports, untested modes or deviations. Without checking scope and configuration, the evidence may not represent the product being placed on the market.

Ignoring detector and bandwidth settings

An attractive plot produced with the wrong detector or resolution bandwidth may understate disturbances or make comparison with the limit invalid. Confirm that final measurements use the settings required by the applicable method.

Treating setup photographs as decoration

Missing cable positions, ground-plane details and auxiliary equipment arrangements prevent reliable reproduction. This weakens failure diagnosis and any later comparison between laboratories.

Assuming a small margin is robust

A narrow pass may reverse with production tolerance or a more representative cable arrangement. Compare the margin with uncertainty and known configuration variability before deciding that no engineering action is needed.

Accepting vague immunity observations

“No issue noted” is not a defined performance assessment. Without monitored functions, criteria and recovery records, a temporary reset or data error may be missed.

Combining modified and unmodified results

If suppression components are added during testing, the report must identify the exact configuration associated with each result. Poor change control leaves an unsuitable evidence trail and may cause production to omit the components that achieved the pass.

Frequently Asked Questions (FAQs)

Does an EMC test report prove that a product complies?

Not by itself. It can provide significant evidence, but the manufacturer must assess applicable legislation, product standards, variants, intended environment, risk, production controls and documentation. Some products follow a self-certification route, while contractual or sector requirements may call for testing by an appropriately accredited laboratory.

How much margin to the limit is enough?

There is no universal design margin. Consider measurement uncertainty, production variation, cable variability, component tolerance and differences between sites. A technically informed risk assessment is more useful than applying one arbitrary dB target to every product.

Should pre-scan plots appear in the final report?

They can be useful for showing the emissions profile, provided they are clearly identified. Compliance conclusions should rely on the measurements and detectors required by the applicable standard, not solely on a fast peak scan.

What should a good failure note contain?

It should identify the test condition, frequency or disturbance level, affected function, observed behaviour, performance criterion, recovery method, EUT configuration and any modification made. That information allows the failure to be reproduced and investigated.

Can an old report support a revised product?

Possibly, but only after a documented engineering assessment. Changes to clocks, enclosure bonding, PCB layout, firmware activity, power supplies, cables or peripherals may alter EMC performance. The assessment should justify whether existing evidence remains representative and whether gap testing is needed.

What calibration evidence should I expect?

The report should identify relevant measuring equipment and its calibration status, normally through controlled equipment records or calibration references. Calibration should be current, suitable for the quantities measured and traceable through an appropriate calibration chain. Calibration alone does not correct an unsuitable setup or test method.

Turning the report into defensible evidence

A well-read EMC test report is more than a pass certificate. It informs the technical file, EMC risk assessment, mitigation records, supplier discussions and market-entry decisions. It also exposes gaps before they become expensive redesign work.

If a report contains uncertain margins, incomplete setup information or failure notes that need investigation, consult the EMC Hire engineering team. EMC Hire can provide equipment hire quotations, pre-compliance support, on-site testing, facility bookings and formal compliance testing where appropriate.

Call +44 (0)1462 817111 or email sales@emchire.co.uk to discuss the product, applicable test programme and evidence required.