IEC 62150: Fibre Optic Active Components — Test and Measurement Procedures

The reliability of optical communication networks depends on the ability to accurately and reproducibly measure the performance of active optical components. IEC 62150 provides the standardised test and measurement procedures that make this possible. While IEC 62148 defines the package interface (the “what connects”) and IEC 62149 defines the performance limits (the “how good”), IEC 62150 defines the measurement methods (the “how to measure”) — specifying everything from optical power metre calibration to modulation bandwidth characterisation and eye diagram analysis. Without these standardised procedures, performance data from different manufacturers or test labs would be incomparable, making system design and procurement impossible.

💡 Series Structure: IEC 62150 is divided into parts, with Part 1 (General and guidance) providing the overarching principles, and subsequent parts addressing specific measurement techniques for transmitters, receivers, and specialised components such as VCSELs and coherent modulators.

1. Optical Power Measurement — The Foundation

All optical component testing begins with accurate optical power measurement. IEC 62150-1 establishes the reference methods for:

Measurement	Method	Uncertainty Target	Calibration Reference
Average optical power	Ge/InGaAs photodiode with integrating sphere	±0.5 dB (absolute)	NIST/PTB primary standard
Peak optical power (pulsed)	Sampling oscilloscope + calibrated photodiode	±1.0 dB	Internal calibration source
Optical power density	Speckle-corrected integrating sphere	±0.7 dB	Calibrated source at same wavelength

A critical requirement in IEC 62150-1 is the use of an integrating sphere for all absolute power measurements of multi-mode and high-power devices. Direct illumination of a photodiode yields errors >1 dB due to beam position sensitivity, polarisation dependence, and mode distribution effects. The standard specifies sphere diameter (typically 25 mm for ≤10 mW, 50 mm for higher powers), port geometry (90° detector port, 0° input port), and baffle placement to prevent direct line-of-sight between input and detector.

⚠️ Common Error Source: When measuring sub-µW receiver sensitivity levels, the integrating sphere’s own thermal noise floor becomes significant. IEC 62150 specifies lock-in amplification or synchronous detection techniques for signals below −30 dBm. The measurement bandwidth must be matched to the data rate — typically 0.75 × bit rate for NRZ modulation.

2. Spectral Characterisation

2.1 Wavelength and Spectral Width

IEC 62150-2 specifies the measurement of centre wavelength and spectral width using an optical spectrum analyser (OSA). Key procedural requirements include:

Resolution bandwidth: 0.05 nm for DFB lasers, 0.5 nm for FP lasers, 1.0 nm for LEDs and VCSELs
Video bandwidth: ≤ 0.1 × resolution bandwidth to suppress noise
Peak search algorithm: Quadratic interpolation across three highest points for sub-resolution accuracy
Side-mode suppression ratio (SMSR): Measured as the power difference between the main mode and the highest side mode, with a typical minimum of 30 dB for DFB lasers

2.2 Chirp and Transient Wavelength Behaviour

For directly modulated lasers, wavelength chirp during bit transitions causes dispersion-induced signal degradation. IEC 62150-3 defines the time-resolved chirp measurement using either a fibre Bragg grating edge filter or a delayed self-heterodyne method. The standard specifies that chirp is characterised as the RMS wavelength deviation during the first 200 ps of a “0” to “1” transition, with acceptable limits depending on the link dispersion budget.

3. Modulation Bandwidth and Eye Diagram Testing

3.1 Small-Signal Modulation Response

The -3 dB modulation bandwidth is the fundamental metric for laser and modulator speed. IEC 62150-4 specifies measurement using a vector network analyser (VNA) with an optical receiver calibrated for flat frequency response. The test setup injects a small-signal swept-frequency sine wave (typically −10 dBm electrical) onto the laser bias, and the optical response is detected and compared to the electrical stimulus. Key specifications:

Parameter	Requirement
VNA frequency range	DC to 3× the expected bandwidth
RF power level	−10 dBm ± 2 dB (small-signal linear regime)
Optical receiver bandwidth	≥ 1.5× the measured bandwidth
Calibration	Full two-port SOLT calibration at the electrical reference plane
Low-frequency reference	Response at 50 MHz normalised to 0 dB

3.2 Eye Diagram Analysis

IEC 62150-5 covers eye diagram testing using a digital sampling oscilloscope (DSO) with optical plug-in. The standard specifies:

Sampling rate: ≥ 10× the bit rate for NRZ, ≥ 40× for PAM4
Number of samples per eye: ≥ 200,000 for mask testing, ≥ 50,000 for routine characterisation
Filtering: 4th-order Bessel-Thomson filter at 0.75 × bit rate for NRZ
Eye mask margins: Voltage and time margin relative to the IEEE or ITU-T defined mask

✅ Engineering Insight — PAM4 Testing: For PAM4 (4-level pulse amplitude modulation), IEC 62150-5 introduces the transmitter and distortion eye closure quaternary (TDECQ) metric. The measurement compares the actual PAM4 eye diagram to an ideal reference eye, computing a penalty in dB. A TDECQ < 3.4 dB (for 400GBASE-LR8) is typical. The standard stresses that the reference equaliser used for TDECQ must be specified — a 5-tap T-spaced FFE is standard for 400G-LR8.

4. Receiver Sensitivity and BER Testing

IEC 62150-6 specifies bit-error-ratio (BER) measurement procedures for optical receivers. The critical aspect is stressed-eye testing — the receiver must meet its BER specification (typically 10⁻¹²) under conditions that simulate the worst-case optical signal it would encounter after transmission through a dispersive, noisy fibre link.

The stressed eye is generated using a combination of:

Vertical eye closure (VEC): achieved by attenuating the signal and adding controlled noise
Jitter: sinusoidal jitter at 10–100 MHz with amplitude up to 0.3 UI peak-to-peak
Dispersion: chromatic dispersion emulated by adding fibre sections or a CD emulator
Polarisation rotation: for coherent receivers, polarisation state scrambling at < 1 kHz

5. FAQ

Q1: Why can’t we use the same test method for all types of optical transceivers?
Different modulation formats (NRZ vs. PAM4), data rates (1 Gbps to 800 Gbps), and transmission distances (100 m to 120 km) require different measurement bandwidths, noise floors, and reference conditions. IEC 62150 provides a modular structure — the general principles in Part 1 are extended by device-specific parts that add the required precision and test conditions for each application.

Q2: How often should test equipment be recalibrated?
IEC 62150 recommends: optical power metres — every 12 months (primary traceable calibration) with monthly in-house verification using a reference source; OSAs — every 24 months with wavelength verification before each use; sampling oscilloscopes — every 12 months with daily amplitude and timing verification; VNAs — every 12 months with daily check using an electronic calibration module.

Q3: What is the largest source of measurement uncertainty in eye diagram testing?
Jitter in the clock recovery unit (CRU) is the dominant contributor. IEC 62150-5 specifies that the CRU jitter must be < 1% UI for data rates ≤ 28 Gbps and < 2% UI for higher rates. For PAM4 testing, the vertical noise uncertainty of the oscilloscope (typically ±2% of full scale) adds approximately 0.2–0.5 dB of TDECQ uncertainty.

Q4: How does IEC 62150 handle the measurement of coherent optical transceivers?
Coherent transceiver testing is specified in IEC 62150-7 and IEC 62150-8. These parts add test procedures for: local oscillator (LO) linewidth measurement using self-heterodyne delayed interferometry; I/Q modulator bias control accuracy; transmitter constellation error vector magnitude (EVM); and receiver front-end frequency response including the 90° hybrid phase error.