Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
IEC 61672-2:2013 โ Electroacoustics: Sound Level Meters โ Pattern Evaluation Tests
💡 Standard Framework: IEC 61672 is the internationally accepted standard family governing the performance, testing, and calibration of sound level meters. Part 2 specifically addresses pattern evaluation — the type-approval testing that establishes whether a sound level meter design meets the performance requirements of Part 1.
1. Scope and Classification System
IEC 61672-2:2013 specifies the detailed test procedures and acceptance criteria for pattern evaluation of sound level meters. The standard establishes two performance classes: Class 1 (precision) and Class 2 (general purpose), with Class 1 having tighter tolerance limits suitable for laboratory and regulatory noise measurements, while Class 2 is appropriate for industrial hygiene and environmental noise screening. The pattern evaluation tests cover all aspects of sound level meter performance including frequency response, time-weighting characteristics, level linearity, detector accuracy, and environmental stability.
This edition (2013) includes the correction (Cor. 1:2014) addressing clarifications in test signal specifications and tolerance interpretations, particularly for infrasonic and ultrasonic frequency ranges.
⚠ Important Distinction: Pattern evaluation (Part 2) is a type-approval of the instrument design, performed once per model. This is distinct from periodic verification (IEC 61672-3) which is the routine calibration performed annually during the instrument’s service life.
2. Key Test Procedures
2.1 Frequency Weighting Tests
The frequency weighting test verifies that the A-weighting and C-weighting filters conform to the specified attenuation characteristics relative to the reference frequency of 1 kHz. The test applies sinusoidal signals at one-third octave intervals from 10 Hz to 20 kHz (extended range) and measures the deviation from the ideal weighting curve.
Class 1 Tolerance Example (A-weighting):
±0.7 dB @ 125 Hz — 1 kHz
±1.0 dB @ 63 Hz — 4 kHz
±1.5 dB @ 20 Hz — 8 kHz
±3.0 dB @ 10 Hz — 16 kHz
±0.7 dB @ 125 Hz — 1 kHz
±1.0 dB @ 63 Hz — 4 kHz
±1.5 dB @ 20 Hz — 8 kHz
±3.0 dB @ 10 Hz — 16 kHz
2.2 Level Linearity and Detector Accuracy
The level linearity test measures the sound level meter’s response accuracy across its full dynamic range, typically from the noise floor (around 15 dB SPL for Class 1) to the maximum input level (typically 140 dB SPL). The detector accuracy test specifically evaluates the RMS detector’s response to signals of varying crest factor (from pure sine waves at 3 dB crest factor to impulsive signals with crest factors exceeding 20 dB).
| Test Parameter | Class 1 Tolerance | Class 2 Tolerance | Test Frequency / Condition |
|---|---|---|---|
| Frequency response (A-weighting) | ±0.7 dB | ±1.0 dB | 125 Hz — 1 kHz |
| Frequency response (C-weighting) | ±0.7 dB | ±1.0 dB | 125 Hz — 1 kHz |
| Level linearity | ±0.4 dB | ±0.6 dB | Reference level range |
| RMS detector accuracy | ±0.5 dB | ±1.0 dB | Crest factor = 5 |
| Time weighting F (Fast) | ±0.5 dB | ±0.5 dB | 125 ms exponential |
| Time weighting S (Slow) | ±0.5 dB | ±0.5 dB | 1 s exponential |
| Self-generated noise | ≤ 20 dB(A) | ≤ 25 dB(A) | Microphone preamp noise floor |
| Electrostatic discharge immunity | ±1.0 dB deviation | ±2.0 dB deviation | 4 kV contact discharge |
3. Engineering Design Insights for SLM Manufacture and Use
From an engineering perspective, several lessons emerge from the IEC 61672-2 pattern evaluation requirements:
- Microphone preamplifier design: Meeting the self-generated noise requirement of ≤20 dB(A) for Class 1 instruments demands extremely low-noise JFET or FET input stages with noise figures below 1 dB. The microphone cartridge capacitance (typically 12-30 pF) directly affects the noise floor through the capacitance-divider effect.
- Environmental stability: The standard requires that the sound level meter maintain accuracy within tolerances across the temperature range -10°C to +50°C and 25% to 90% relative humidity. This necessitates temperature-compensated reference oscillators, humidity-sealed electret microphone capsules, and careful thermal management of the analog processing chain.
- Digital versus analog implementation: Modern Class 1 instruments increasingly use digital signal processing with 24-bit sigma-delta converters operating at 48-192 kHz sample rates. The pattern evaluation tests reveal that while DSP offers superior stability, the anti-aliasing filter design is critical — inadequate filtering can produce aliasing artifacts that violate the frequency response tolerances above 10 kHz.
✅ Practical Advice: When selecting a sound level meter for a specific application, note that the pattern evaluation certificate (supplied with each new instrument model) only verifies the design. Individual instrument conformity must be verified through the individual verification tests of IEC 61672-1, Clause 9, which includes calibration at 1 kHz and a basic functional check. Always request the pattern evaluation certificate from the manufacturer.
4. Application Example: Environmental Noise Monitoring
For environmental noise monitoring per ISO 1996-2, a Class 1 sound level meter is required. The pattern evaluation must demonstrate compliance with all applicable tests. During a typical 24-hour environmental survey, the sound level meter measures LAeq,T (equivalent continuous A-weighted sound level), LAFmax (maximum A-weighted Fast time weighting), and LAF90 (percentile level exceeded for 90% of the measurement period). The instrument’s long-term stability — tested during pattern evaluation via the 1-hour drift test — ensures that measurement uncertainty remains within acceptable limits throughout the survey duration.
❓ Q1: What is the difference between pattern evaluation (Part 2) and periodic verification (Part 3)?
A: Pattern evaluation is a one-time type test that validates the instrument design meets the standard. Periodic verification (Part 3) is the annual calibration that confirms the individual instrument remains within tolerance during its service life. Pattern evaluation is performed by an independent testing laboratory; periodic verification can be performed by accredited calibration laboratories.
❓ Q2: Can a Class 2 sound level meter be used for regulatory noise measurements?
A: This depends on the specific regulation. Occupational noise exposure measurements (e.g., OSHA, EU Directive 2003/10/EC) typically require Class 2 or better. Environmental noise measurements for compliance with ISO 1996-2 generally require Class 1. Always check the applicable regulation.
❓ Q3: How often should a sound level meter be recalibrated?
A: The standard requires periodic verification at intervals not exceeding 2 years (recommended annually). Additionally, field calibration should be checked before and after each measurement session using an acoustic calibrator meeting IEC 60942 Class 1 or Class 2 requirements.
❓ Q4: What is the significance of crest factor handling in the pattern evaluation?
A: Crest factor (peak-to-RMS ratio) is critical for measuring impulsive noises such as gunshots, impact hammers, or explosive events. A sound level meter that meets the crest factor requirement at 20 dB (Class 1) can accurately measure signals where the peak is 10× higher than the RMS value — essential for accurate impulse noise measurement.
