How to build a radiated emissions pre-check workflow

A radiated emissions pre-check can expose a failing clock harmonic in minutes, but an uncontrolled setup can conceal the same problem just as quickly.

The purpose of an internal radiated emissions workflow is not to recreate a formal chamber on a laboratory bench. It is to produce repeatable engineering evidence, identify dominant sources and determine whether design changes genuinely improve the product before formal testing.

Define what the pre-check must answer

Start with a specific engineering question. “Does it pass?” is usually the wrong question because an informal environment cannot reproduce every feature of a compliant test site. Better questions include:

• Which frequencies dominate the emissions profile?
• Which cable, enclosure seam or circuit region is responsible?
• Does changing the clock configuration reduce a suspect harmonic?
• Does a proposed filter improve emissions without moving the problem elsewhere?
• Is sufficient margin visible to justify booking formal testing?

The applicable product or product-family standard should guide the pre-check. CISPR 11 and CISPR 32, for example, cover different product scopes and must not be treated as interchangeable. Product standards may define operating modes, test distances, detector requirements, limits and frequency ranges. The latest active edition, product scope and any customer-specific test plan must be checked rather than inferred from a previous project.

The IEC information on electromagnetic compatibility provides useful background on the standards framework, but the published standard remains the controlling reference.

Build a stable radiated emissions setup

Repeatability begins with geometry. Record the equipment under test, or EUT, position, antenna orientation, antenna distance, support height, cable routing, peripheral arrangement and bonding configuration. A photograph taken from several angles is often more useful during a later retest than a page of vague notes.

Use a fixed measurement position wherever possible. A screened room, absorber-lined area or suitable test facility provides better ambient control, but useful development comparisons can also be made in a controlled engineering space. Results from an uncontrolled room should not be presented as equivalent to formal compliance data.

The basic measurement chain normally includes an appropriate antenna, low-loss coaxial cable and an EMI receiver or spectrum analyser with suitable frequency coverage and detector capability. A preamplifier may improve system sensitivity, provided that its gain, noise figure and overload behaviour are understood. Adding gain indiscriminately can compress the front end when strong broadcast, mobile or local digital signals are present. The resulting trace may look quieter while being less trustworthy.

Characterise the complete path. Record antenna factors, cable loss, preamplifier gain and any attenuator values used by the measurement system. 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 better comparison between development measurements and subsequent formal testing.

Keep the EUT configuration representative

Exercise the product in the operating mode most likely to maximise emissions. That might involve continuous data transfer, maximum processor activity, active displays, motor operation, loaded power converters or simultaneous use of external interfaces. An idle prototype can produce an attractive trace that bears little resemblance to the fielded product.

Cable layout matters. An I/O lead can become an efficient common-mode radiator, even when the originating noise is generated deep inside the PCB. Fix cable lengths and positions, then document them. Moving a cable between baseline and modified measurements can create a larger amplitude change than the component under investigation.

A repeatable measurement sequence

A useful radiated emissions workflow separates reconnaissance from evidence gathering. Fast scans find problems. Controlled measurements determine whether a design change is real.

Capture the ambient before powering the EUT

Run an ambient scan with the EUT switched off while leaving the measurement chain unchanged. Log persistent broadcast, communications and local equipment signals. Repeat the ambient scan if nearby machinery, lighting or radio activity changes.

Ambient subtraction should not be applied blindly. Two signals at similar frequencies may vary independently, and a spectrum analyser trace does not contain enough phase information for simple amplitude subtraction to produce a defensible EUT result. Instead, compare EUT-on and EUT-off traces, investigate suspicious frequencies and move to a better-controlled environment where ambiguity remains.

Run a fast peak-detector scan

Use peak detection for rapid source discovery because it captures intermittent and modulated disturbances efficiently. Select the resolution bandwidth, video bandwidth, sweep time and frequency span with reference to the applicable method and the analyser’s behaviour. For many CISPR radiated measurements between 30 MHz and 1 GHz, a 120 kHz resolution bandwidth is commonly encountered, but this is not a universal setting. Different frequency ranges and product standards may require different parameters.

Do not use the fast scan as the final compliance measurement. Formal measurements may require CISPR quasi-peak, average or other specified detection, together with prescribed antenna-height searches, azimuth investigation and measurement distances. A conventional spectrum analyser may support useful development work without implementing a compliant CISPR detector accurately.

Log peaks with enough context

Create a peak log rather than relying on screenshots alone. For each suspect, capture:

• Frequency and measured amplitude.
• Detector and resolution bandwidth.
• Antenna type, polarisation and distance.
• EUT operating mode and software version.
• Cable and accessory configuration.
• Known ambient overlap or measurement uncertainty.
• Likely clock, switching or interface relationship.

Correlate peaks with known source frequencies. Harmonic spacing often points towards a crystal oscillator, memory bus, display interface, Ethernet clock or switched-mode converter. Sidebands can reveal modulation by frame rates, load cycling or control loops. Frequency correlation is evidence, not proof. Confirm the coupling path before modifying hardware.

Source hunting without corrupting the baseline

Near-field electric and magnetic probes are useful for locating energetic PCB regions, connector noise and enclosure leakage. They do not directly predict far-field compliance because probe orientation, distance and loading dominate the indicated level. Use them comparatively, keeping the probe position repeatable.

A current probe around an external cable can reveal common-mode current associated with a far-field peak. If the cable current and antenna response change together when a ferrite or bonding change is introduced, the evidence for that coupling path becomes stronger. The current probe is being used diagnostically here, not as a substitute for the radiated emissions antenna measurement.

Controlled interventions are more informative than random component swapping. Change one variable at a time:

• Disable one clock or interface in firmware.
• Apply a temporary ferrite to one cable.
• Improve one enclosure bond.
• Shorten one internal return path.
• Temporarily shield one suspected circuit region.

Record every change. An unnecessarily long bonding strap adds parasitic inductance, so a modification that works at a few megahertz may provide little benefit at several hundred megahertz. Likewise, copper tape across a seam can identify leakage but may not represent a manufacturable fix.

Retest under controlled conditions

Return the EUT and antenna to the documented baseline geometry before comparing results. Use the same instrument settings, operating mode, cables and software build. If the original peak was intermittent, capture sufficient observation time to understand its duty cycle.

Review the whole trace after each change. A filter may reduce one harmonic while increasing common-mode current elsewhere. Spread-spectrum clocking can lower narrowband peak amplitude but distribute energy across a wider band. A single marker reading will miss that trade-off.

Promising changes should then be checked in a more representative environment using suitable calibrated equipment. Accessible EMC pre-compliance testing allows teams to debug prototypes earlier, gather calibrated engineering data and reduce the financial risk of discovering systemic problems during formal testing. Pre-compliance results improve confidence but do not automatically prove compliance.

Typical scenario

Consider an illustrative industrial controller with a switched-mode power supply, Ethernet interface and several external sensor cables. A fast antenna scan shows a family of peaks related to a digital clock. Near-field probing finds strong activity around the processor, but placing a current probe around the longest sensor lead shows correlated common-mode current.

The team must decide whether the dominant mechanism is direct PCB radiation, enclosure leakage or cable radiation driven by poor common-mode control. Changing the PCB immediately could waste a layout cycle if the real fault is the connector bonding or cable return path.

A sensible pre-check would preserve the baseline setup, log the suspect peaks, compare cable currents, vary one cable at a time and test temporary bonding or filtering measures. The preferred modifications would then be retested across the complete scan range before committing them to production.

EMC Hire can support this work through equipment selection, short-term instrument hire, on-site testing, facility access and formal compliance testing where appropriate. Relevant starting points include the radiated emissions measurement system information and practical EMC equipment selection guides.

When to Hire EMC Equipment

Hiring is often more rational than buying when radiated emissions work is tied to a prototype phase, design investigation or short formal test window. Suitable antennas, receivers, preamplifiers and probes represent more than capital expenditure. Ownership also brings calibration scheduling, storage, servicing, connector care and the risk that the equipment will not suit a later programme.

Rental allows a team to match frequency coverage, detector capability, dynamic range and accessories to the current product. It also provides temporary capacity when several projects reach verification together. This reduces the chance of buying a general-purpose analyser that later proves unsuitable for the required CISPR measurements.

Radiated problems can originate from noise conducted onto power or signal cables. Where diagnosis points towards the power port, a separate conducted emissions system using an appropriate LISN may be required. A LISN is used for conducted emissions measurements on relevant power ports. It is not a radiated emissions transducer or a conducted immunity device.

Common EMC Testing Mistakes to Avoid

Changing cable routing between scans

Cable movement changes common-mode radiation efficiency and can falsely attribute improvement to a circuit modification. Mark cable positions and photograph the setup.

Using peak data as proof of compliance

A peak scan is conservative and efficient for investigation, but it does not replace detectors, geometry and procedures required by the applicable standard. Treat it as screening data.

Ignoring receiver overload

Strong ambient signals or excessive preamplifier gain can create intermodulation products, compression or apparent noise-floor changes. Check levels with attenuation and preamplifier bypass tests.

Testing an unrepresentative operating mode

An inactive interface or lightly loaded converter may hide the product’s worst emissions state. Record loads, traffic patterns, peripherals, firmware and power conditions.

Failing to preserve peak records

A screenshot without detector, bandwidth, antenna, distance and EUT configuration cannot support a credible retest. Weak records also undermine the technical file and make later mitigation evidence difficult to defend.

Confusing near-field intensity with far-field risk

A strong local field may couple poorly to an external radiating structure, while a modest PCB source may drive a long cable efficiently. Use correlation and controlled changes before choosing a redesign.

Frequently Asked Questions (FAQs)

How much margin should a pre-check show?

There is no universal allowance. Site correlation, ambient conditions, measurement uncertainty, product variability and setup repeatability all matter. Establish an internal engineering margin from controlled comparisons with formal measurements rather than adopting an arbitrary figure.

Can a spectrum analyser replace an EMI receiver?

It can be effective for source hunting and peak screening if its frequency coverage, dynamic range and overload performance are suitable. Formal measurements may require CISPR detector behaviour and other receiver characteristics that a general-purpose analyser does not provide.

Should horizontal and vertical antenna polarisations be checked?

Yes, where relevant to the intended test method. Cable orientation, PCB currents and enclosure geometry can produce substantially different responses. Recording only one polarisation risks missing the dominant emission.

When should testing move from the bench to a facility?

Move when ambient signals obscure suspects, when reliable distance and antenna positioning are needed, or when the design decision depends on margin to a formal limit. Facility testing is also appropriate before committing expensive tooling or PCB revisions.

Can pre-compliance results support CE or UKCA work?

Calibrated pre-compliance data may support engineering decisions, EMC risk assessments, mitigation records and the technical file. It does not by itself complete every conformity obligation. The manufacturer or responsible economic operator must confirm the applicable legislation, standards, conformity assessment route and Declaration of Conformity requirements.

Plan the next measurement

For help defining a radiated emissions workflow, selecting suitable hire equipment or arranging a controlled retest, contact the EMC Hire engineering team. Support is available for equipment hire, on-site investigation, pre-compliance engineering, formal compliance testing where appropriate and bookings at the EMC Hire test facility.

Call +44 (0)1462 817111 or email sales@emchire.co.uk to discuss the EUT, applicable standard, frequency range, project stage and evidence required.