Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
CAN CSA IEC CISPR 16-2-3-18: Radiated Disturbance Measurement Methods for EMC Compliance
Technical Overview of CISPR 16-2-3:2018 Requirements for Radiated Emissions Testing
Scope and Application
CAN CSA IEC CISPR 16-2-3-18 is the Canadian adoption of the international standard IEC CISPR 16-2-3:2018, titled Specification for radio disturbance and immunity measuring apparatus and methods – Part 2-3: Methods of measurement of disturbances and immunity – Radiated disturbance measurements. This standard defines the fundamental methods for measuring radiated electromagnetic disturbances from electrical and electronic equipment in the frequency range from 9 kHz to 18 GHz.
The document is a cornerstone of EMC (electromagnetic compatibility) testing and is referenced by numerous product emission standards across industrial, consumer, automotive, and medical sectors. It provides the measurement framework that ensures reproducible, comparable, and traceable results at designated test sites, including open-area test sites (OATS), semi-anechoic chambers (SAC), and fully-anechoic rooms (FAR).
Note: CAN CSA IEC CISPR 16-2-3-18 mirrors the international version without national deviations, making it directly applicable to both domestic and global compliance programs.
The primary users of this standard are EMC test laboratories, product manufacturers, compliance engineers, and regulatory authorities who need to verify that products do not cause unacceptable radio interference.
Technical Requirements
Measurement Instrumentation
The standard specifies the use of CISPR 16-1‑1 compliant measuring receivers along with antennas, preamplifiers, and attenuators. Requirements for receivers include:
- Six‑dB bandwidths corresponding to CISPR bands (e.g., 200 Hz for 9–150 kHz, 9 kHz for 0.15–30 MHz, 120 kHz for 30 MHz–1 GHz).
- Quasi‑peak, peak, and average detectors with defined charging/discharging time constants.
- Overall measurement uncertainty that meets the limits of CISPR 16‑1‑1.
Antenna Types and Calibration
For different frequency ranges, specific antenna types are recommended. Table 1 summarises the typical antennas and their calibration requirements.
| Frequency Range | Antenna Type | Calibration Method |
|---|---|---|
| 9 kHz – 30 MHz | Active loop antenna (rod antenna for magnetic field) | Calibrated antenna factor traceable to national standards |
| 30 MHz – 200 MHz | Biconical antenna | Free‑space antenna factor via standard site method |
| 200 MHz – 1 GHz | Log‑periodic dipole array (LPDA) | Free‑space antenna factor via standard site method or reference antenna |
| 1 GHz – 18 GHz | Horn antenna (double‑ridged or conical log‑spiral) | Gain comparison or three‑antenna method |
Tip: Antenna calibration must be performed at the measurement distance used (typically 3 m, 10 m, or 30 m). Free‑space antenna factors are now preferred to reduce near‑field effects.
Test Site Requirements
CAN CSA IEC CISPR 16-2-3-18 defines three categories of validated test sites:
- OATS: Ground plane with at least 6 m diameter ellipse clearance, free from reflecting objects. Normalised site attenuation (NSA) deviation must be within ±4 dB.
- SAC: Shielded enclosure with absorber lining on all walls and ceiling, but with a reflective ground plane. Must meet same NSA criteria as OATS.
- FAR: Fully absorber‑lined room where the ground plane is also covered with absorbers, achieving free‑space conditions. Validated with site voltage standing wave ratio (SVSWR) method for frequencies above 1 GHz.
Important: For frequencies above 1 GHz, the SVSWR method replaces the traditional NSA method. The mutual coupling between antenna and absorbers must be carefully controlled.
Measurement Distance and Height Scan
Standard measurement distances are 3 m, 10 m, and 30 m. For OATS and SAC, the receiving antenna is scanned in height from 1 m to 4 m (or up to 6 m for certain configurations) to capture the maximum field strength. The turntable rotates the equipment under test (EUT) through 360° while the antenna alternates between horizontal and vertical polarisation. The highest emission level found is recorded.
Implementation Highlights
Measurement Setup Configuration
The standard prescribes a meticulous arrangement of the EUT and associated cables to ensure repeatable results. Key setup parameters include:
- Table height for table‑top equipment: 0.8 m above the ground plane.
- Floor‑standing equipment placed directly on the ground plane.
- Cable lengths standardised and routed in a manner representative of typical installation, with ferrite beads used where necessary to suppress common‑mode currents.
- Peripheral cables, if any, must be included in the configuration.
Best Practice: Pre‑scanning with peak detector at a fast sweep speed to identify critical frequencies, followed by quasi‑peak or average measurement only at those frequencies, significantly reduces total test time.
Use of Alternative Test Sites
When a full OATS or SAC is not available, alternative methods such as the Transverse Electromagnetic (TEM) cell (for small EUTs) or the radiated emission test in a fully anechoic room (FAR) may be used. The standard outlines the equivalence criteria and uncertainty contributions for such substitutions, emphasising the need for validation against reference measurements.
Measurement Uncertainty
An uncertainty budget must be calculated in accordance with CISPR 16‑4‑2. Major uncertainty contributors include:
- Measuring receiver uncertainty (amplitude, frequency response).
- Antenna factor calibration uncertainty.
- Mismatch effects between antenna and receiver.
- Site imperfections (deviation from ideal NSA or SVSWR).
- Cable loss and attenuator uncertainty.
- EUT positioning and ambient electromagnetic noise.
The expanded measurement uncertainty (k = 2) should typically be less than 6 dB for radiated emissions up to 1 GHz. If uncertainty exceeds this, the test laboratory must adjust limits or apply guard bands.
Compliance and Validation Notes
Site Validation Frequency
Both initial and periodic re‑validation of the test site is mandatory. OATS and SAC must demonstrate NSA compliance annually, while FAR needs SVSWR validation to ensure absorber performance remains stable. Any change in absorber configuration, antenna mount, or significant metallic structures warrants a new validation.
Non‑compliance Risk: Using a test site that has not been validated within the required interval may result in rejection of test reports by regulatory bodies and accreditation bodies.
Correlation with Product Standards
CAN CSA IEC CISPR 16-2-3-18 is a horizontal standard – it does not specify emission limits. Instead, product standards (e.g., CISPR 11, CISPR 32, IEC 55035) call up this method and set the appropriate limit lines. When applying CAN CSA IEC CISPR 16-2-3-18, the test laboratory must ensure that:
- The selected measurement distance matches the product standard’s requirement.
- The detector function (peak, quasi‑peak, average) corresponds to the limit type.
- Frequency range and bandwidth are set according to the product specification.
Common Compliance Pitfalls
Many initial tests fail due to:
- Inadequate cable management (unused ports not terminated, cables too long).
- Wrong antenna polarisation used for final measurement.
- EUT not operating in the worst‑case mode (all modes must be exercised).
- Ambient signals misinterpreted as EUT emissions – the standard requires an ambient check without EUT.
Tip: Always perform a pre‑compliance scan with a simple receiver before scheduling a full compliance test. This identifies likely problem frequencies and helps reduce expensive retesting.
Documentation and Reporting
The test report must include a description of the test site, EUT configuration, instrumentation details, calibration certificates, measurement uncertainty budget, and a statement of compliance for each frequency where the limit is exceeded. The standard also requires a photograph of the setup for traceability.
Frequently Asked Questions
Q: Does CAN CSA IEC CISPR 16-2-3-18 apply to both conducted and radiated measurements?
A: No. This part specifically covers radiated disturbance measurements (9 kHz–18 GHz). Conducted disturbance measurements are covered by CISPR 16-2-1, while disturbance power measurements are covered by CISPR 16-2-2.
A: No. This part specifically covers radiated disturbance measurements (9 kHz–18 GHz). Conducted disturbance measurements are covered by CISPR 16-2-1, while disturbance power measurements are covered by CISPR 16-2-2.
Q: Can I use a 3 m test distance for all products?
A: Only if the product standard explicitly permits it. Many product standards specify 10 m or 30 m for compliance. However, CAN CSA IEC CISPR 16-2-3-18 describes methods that are distance‑independent; the limits and distance are always defined by the referencing standard.
A: Only if the product standard explicitly permits it. Many product standards specify 10 m or 30 m for compliance. However, CAN CSA IEC CISPR 16-2-3-18 describes methods that are distance‑independent; the limits and distance are always defined by the referencing standard.
Q: How often must an OATS or SAC be re‑validated?
A: At least once per year, and after any significant structural change. Some accreditation bodies (e.g., A2LA, IAS) may require re‑validation every 12 months. The standard itself recommends checking after relocation or modification of absorbers/ground plane.
A: At least once per year, and after any significant structural change. Some accreditation bodies (e.g., A2LA, IAS) may require re‑validation every 12 months. The standard itself recommends checking after relocation or modification of absorbers/ground plane.
Q: What is the maximum allowed measurement uncertainty under this standard?
A: CISPR 16-4-2 sets a global uncertainty limit of 6 dB for radiated emission measurements below 1 GHz (k = 2). For measurements above 1 GHz, the limit is 5 dB. Laboratories must prepare a budget and demonstrate compliance; otherwise, they must apply a guard band to the limit.
A: CISPR 16-4-2 sets a global uncertainty limit of 6 dB for radiated emission measurements below 1 GHz (k = 2). For measurements above 1 GHz, the limit is 5 dB. Laboratories must prepare a budget and demonstrate compliance; otherwise, they must apply a guard band to the limit.
© 2026 International Standards Press. This article is provided for informational purposes and does not constitute legal or regulatory advice. Always refer to the latest published version of the standard.