Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
IEC 61207: Gas Analyzers — Expression of Performance
✅ Standard at a Glance
IEC 61207 is a multi-part international standard that establishes uniform methods for expressing the performance of gas analyzers. Developed by IEC Technical Committee 65 (Industrial-process measurement, control and automation), this standard provides a common terminology and specification framework across all gas analyzer technologies, including paramagnetic, zirconia, thermal conductivity, NDIR (non-dispersive infrared), chemiluminescence, flame ionization, and electrochemical types. The standard ensures that performance specifications from different manufacturers can be meaningfully compared and that users can select the appropriate analyzer technology for their specific industrial process application.
IEC 61207 is a multi-part international standard that establishes uniform methods for expressing the performance of gas analyzers. Developed by IEC Technical Committee 65 (Industrial-process measurement, control and automation), this standard provides a common terminology and specification framework across all gas analyzer technologies, including paramagnetic, zirconia, thermal conductivity, NDIR (non-dispersive infrared), chemiluminescence, flame ionization, and electrochemical types. The standard ensures that performance specifications from different manufacturers can be meaningfully compared and that users can select the appropriate analyzer technology for their specific industrial process application.
🔌 1. Standardized Performance Terminology
1.1 The Importance of Uniform Performance Expression
Before IEC 61207, gas analyzer manufacturers used widely varying definitions for terms like “accuracy,” “drift,” and “response time,” making it nearly impossible for users to compare products from different vendors. IEC 61207 establishes a rigorous, unambiguous terminology that covers all aspects of gas analyzer performance. The standard defines performance characteristics in three categories:
- Static characteristics: Accuracy, precision, linearity, sensitivity, detection limit, drift (zero and span), and calibration curve characteristics.
- Dynamic characteristics: Response time (rise time, fall time, delay time, total response time), warm-up time, and recovery time after overload.
- Influence quantities: Effects of ambient temperature, pressure, flow rate, sample humidity, electrical supply variations, vibration, and interfering gases.
💡 Engineering Insight
The most commonly mis-specified gas analyzer parameter is “accuracy.” Some manufacturers quote accuracy as a percentage of the reading (% of reading), others as a percentage of full scale (% FS), and still others as a percentage of the measurement range plus a constant. IEC 61207 requires that accuracy always be stated as the maximum error under reference conditions, expressed either as an absolute value (e.g., ±2 ppm CO) or as a percentage of the measured value (% of reading). To fully capture both high-range and low-range performance, the standard recommends the composite form: ±1% of reading or ±1 ppm, whichever is greater. This prevents the misleading situation where a manufacturer quotes “±1% FS” on a 0-1000 ppm analyzer, which would imply ±10 ppm error at any reading — but at 10 ppm, this represents a 100% error at the measurement value, even though the specification appears tight.
The most commonly mis-specified gas analyzer parameter is “accuracy.” Some manufacturers quote accuracy as a percentage of the reading (% of reading), others as a percentage of full scale (% FS), and still others as a percentage of the measurement range plus a constant. IEC 61207 requires that accuracy always be stated as the maximum error under reference conditions, expressed either as an absolute value (e.g., ±2 ppm CO) or as a percentage of the measured value (% of reading). To fully capture both high-range and low-range performance, the standard recommends the composite form: ±1% of reading or ±1 ppm, whichever is greater. This prevents the misleading situation where a manufacturer quotes “±1% FS” on a 0-1000 ppm analyzer, which would imply ±10 ppm error at any reading — but at 10 ppm, this represents a 100% error at the measurement value, even though the specification appears tight.
1.2 Performance Parameters and Definitions
| Parameter | IEC 61207 Definition | Typical Expression | Measurement Method |
|---|---|---|---|
| Linear error (linearity deviation) | The maximum deviation between the measured value and a best-fit straight line (calibration curve) over the specified measurement range | ≤ ±1% of span | Measure at 5-10 equally spaced gas concentrations covering 0-100% of range; calculate deviation from linear regression |
| Zero drift | The change in the output signal for a zero gas over a specified time period under specified conditions | ≤ ±1% of span per 7 days | Supply zero gas continuously; record output at 24 h intervals for 7 days; report maximum deviation |
| Span drift | The change in the output signal for a span gas over a specified time period under specified conditions | ≤ ±2% of span per 7 days | Supply span gas continuously; record output at 24 h intervals for 7 days; report maximum deviation |
| Repeatability | The closeness of agreement between successive measurements of the same gas sample under the same conditions | ≤ ±0.5% of span | 10 consecutive measurements of the same gas sample; calculate 2σ (twice standard deviation) |
| Response time (t90) | The time from the instant a step change in gas concentration is applied at the analyzer inlet to the moment the output reaches 90% of the final steady-state value | ≤ 30 s (typical for extractive analyzers) | Step change from zero gas to span gas; measure time to reach 90% of final value |
| Detection limit | The minimum concentration of the measured component that can be distinguished from zero with a specified confidence level (typically 95%) | ≤ 1 ppm | Measure zero gas repeatedly (n ≥ 20); detection limit = 3σ of zero signal / sensitivity |
| Cross-sensitivity | The response of the analyzer to a gas component other than the one being measured, expressed as equivalent concentration of the measured component | ≤ 1% (e.g., CO2 cross-sensitivity on an NDIR CO analyzer) | Expose analyzer to 100% of the interfering gas (or the maximum expected concentration); record indicated concentration of measured component |
💡 2. Technology-Specific Considerations
2.1 Overview of Gas Analyzer Technologies
IEC 61207 has multiple sub-parts, each addressing a specific measurement technology or application. The key sub-parts include:
| IEC 61207 Part | Technology | Measured Gases | Principle | Typical Applications |
|---|---|---|---|---|
| IEC 61207-1 | General guidelines | All gases | Framework terminology, definitions, and test conditions applicable to all analyzer types | Foundation for all other parts |
| IEC 61207-2 | Paramagnetic oxygen analyzers | O2 | Oxygen’s paramagnetic property (unpaired electrons) creates a magnetic susceptibility ≈ 100x greater than most other gases; the “magnetic wind” or “magnetic pressure” effect is measured | Flue gas monitoring, inert gas blanketing, combustion control, safety (O2 deficiency) |
| IEC 61207-3 | Zirconia oxygen analyzers | O2 | Nernst potential across a stabilized zirconia electrolyte at high temperature (700-800 °C) is proportional to the logarithm of the oxygen partial pressure ratio | Boiler combustion control, kiln atmosphere monitoring, steel furnace optimization |
| IEC 61207-6 | NDIR analyzers | CO, CO2, NO, SO2, CH4, N2O, hydrocarbons | Molecular absorption of infrared radiation at specific wavelengths (2.5-25 µm); double-beam or gas filter correlation design | Emissions monitoring (CEMS), process control, biogas quality, automotive exhaust |
| IEC 61207-7 | Tunable diode laser (TDLAS) | O2, H2O, NH3, H2S, HCl, HF, CH4, CO, CO2 | Single-mode diode laser tuned across a specific absorption line; direct absorption or wavelength modulation spectroscopy (WMS) | In-situ process monitoring, CEMS, natural gas pipeline quality, hazardous area monitoring |
| IEC 61207-8 | Thermal conductivity (TCD) | H2, He, CH4 (in binary or pseudo-binary mixtures) | Measurement of gas thermal conductivity changes; Wheatstone bridge with reference cell | Hydrogen purity monitoring, ammonia synthesis, biogas methane content |
⚠️ Critical Application Note: Paramagnetic O2 vs. Zirconia O2
While both technologies measure oxygen, their performance characteristics differ fundamentally. Paramagnetic analyzers measure O2 concentration directly (typically 0-1% to 0-100% range) with ±0.1% absolute accuracy but have a response time of 7-30 seconds (limited by the gas transport path). Zirconia analyzers measure O2 logarithmically — they excel at low O2 concentrations (e.g., 0-10% range) with fast response (< 3 s) but lose sensitivity at high O2 concentrations. For combustion control where the O2 set-point is typically 2-5%, zirconia probes provide superior response speed, enabling closed-loop trim control of the air/fuel ratio. For safety applications (O2 deficiency monitoring in confined spaces), paramagnetic analyzers provide higher absolute accuracy across the full 0-25% range without the high-temperature safety concern of zirconia probes (which require 700+ °C operation).
While both technologies measure oxygen, their performance characteristics differ fundamentally. Paramagnetic analyzers measure O2 concentration directly (typically 0-1% to 0-100% range) with ±0.1% absolute accuracy but have a response time of 7-30 seconds (limited by the gas transport path). Zirconia analyzers measure O2 logarithmically — they excel at low O2 concentrations (e.g., 0-10% range) with fast response (< 3 s) but lose sensitivity at high O2 concentrations. For combustion control where the O2 set-point is typically 2-5%, zirconia probes provide superior response speed, enabling closed-loop trim control of the air/fuel ratio. For safety applications (O2 deficiency monitoring in confined spaces), paramagnetic analyzers provide higher absolute accuracy across the full 0-25% range without the high-temperature safety concern of zirconia probes (which require 700+ °C operation).
2.2 Influence of Sample Conditioning
IEC 61207 emphasizes that the overall measurement system performance depends as much on the sample conditioning system as on the analyzer itself. The standard identifies critical sample conditioning parameters:
- Pressure and flow rate control: Most gas analyzers are sensitive to sample pressure and flow rate. A pressure change of 1 mbar can cause a 0.1% reading error in NDIR analyzers. The standard specifies that the sample pressure must be regulated to within ±0.5% of the calibration pressure for accurate measurements.
- Dew point control: Condensation in the analyzer measurement cell causes severe errors (scattering of IR radiation in NDIR, condensation on the zirconia sensor). The sample must be conditioned to a dew point at least 5 °C below the minimum ambient temperature of the analyzer location.
- Particulate filtration: Particles > 2 µm must be removed to prevent scattering errors in optical analyzers and abrasion of sensitive components.
- Chemical scrubbers: If the sample contains gases that would damage the analyzer or cause interference (e.g., acidic gases like HF, HCl in an NDIR analyzer), appropriate scrubbers must be installed upstream.
🔬 3. Testing, Calibration and Validation
3.1 Calibration Gas Requirements
IEC 61207 specifies strict requirements for calibration gases. The standard requires that calibration gas mixtures be traceable to national or international standards (e.g., NIST or IRMM certified reference materials). The uncertainty of the calibration gas concentration must be at least 3 times better than the specified accuracy of the analyzer under test. For example, if the analyzer has a specified accuracy of ±1% of reading, the calibration gas must have an uncertainty of ≤ ±0.3% of the certified value. The standard also specifies the minimum number of calibration points: at least 5 equally spaced points across the measurement range for linearity determination, plus a zero and span check before each test series.
3.2 Type Testing vs. Routine Testing
IEC 61207 distinguishes between two levels of performance testing:
- Type testing (performance qualification): A comprehensive evaluation performed on a representative unit of a given analyzer model to establish the full performance envelope. Type tests include all influence quantity tests (temperature, pressure, flow, voltage, vibration), long-term drift tests (7-30 days), and cross-sensitivity tests against all likely interfering gases. Type test results are generally valid for the entire production series, unless design changes are made.
- Routine testing (acceptance testing): A reduced set of tests performed on each individual analyzer before delivery and after installation. Routine tests typically include accuracy verification at 3 concentrations (zero, mid-span, span), response time measurement, and zero/span drift over 24 hours.
✅ Practical Calibration Strategy for Process Analyzers
For continuous process gas analyzers in industrial service, IEC 61207 recommends the following calibration strategy: (1) Auto-calibration using integrated zero and span gas valves at least once every 24 hours. (2) Manual multi-point calibration (5 points) once per month, or whenever auto-calibration shows zero drift > 2% of span or span drift > 3% of span between consecutive auto-calibrations. (3) Full performance validation (including response time and cross-sensitivity checks) every 6 months. (4) After any major maintenance event (replacement of the IR source, detector, or sample cell cleaning), a full type-test-equivalent recalibration should be performed. This strategy balances measurement integrity with maintenance cost — excessive calibration wastes calibration gas and increases analyzer downtime, while insufficient calibration risks undetected measurement errors that can cause process upsets or emissions non-compliance.
For continuous process gas analyzers in industrial service, IEC 61207 recommends the following calibration strategy: (1) Auto-calibration using integrated zero and span gas valves at least once every 24 hours. (2) Manual multi-point calibration (5 points) once per month, or whenever auto-calibration shows zero drift > 2% of span or span drift > 3% of span between consecutive auto-calibrations. (3) Full performance validation (including response time and cross-sensitivity checks) every 6 months. (4) After any major maintenance event (replacement of the IR source, detector, or sample cell cleaning), a full type-test-equivalent recalibration should be performed. This strategy balances measurement integrity with maintenance cost — excessive calibration wastes calibration gas and increases analyzer downtime, while insufficient calibration risks undetected measurement errors that can cause process upsets or emissions non-compliance.
❓ Frequently Asked Questions
Q1: How does “cross-sensitivity” differ from “interference” as defined by IEC 61207?
A: IEC 61207 defines cross-sensitivity as the quantifiable response of the analyzer to a specific interfering gas, expressed as an equivalent concentration of the measured component. For example, an NDIR CO analyzer may have a cross-sensitivity of 0.5% to CO2, meaning that 10% CO2 produces a reading equivalent to 0.05% CO. Interference is a broader term encompassing all sources of error not related to the measurand, including cross-sensitivity, but also temperature effects, pressure effects, electrical interference, and sample matrix effects. The standard requires that cross-sensitivities to all likely interfering gases be documented in the type test report, while interference from environmental influences (temperature, pressure) is specified separately as “influence quantities.”
Q2: Can IEC 61207 performance specifications be compared directly across different gas analyzer technologies?
A: Generally yes, because IEC 61207 provides a unified framework for expressing performance. However, the test conditions must be carefully compared. For example, drift specifications measured under controlled laboratory conditions (constant temperature 23 °C ± 2 °C) may be very different from field performance. Some manufacturers specify drift “under reference conditions” while others specify “under normal operating conditions.” IEC 61207 requires that the test conditions be stated alongside every performance specification. When comparing two analyzers, ensure that the test conditions (temperature range, gas pressure, flow rate, calibration frequency, and time period) are equivalent. The standard also notes that performance specifications measured with zero-grade calibration gases may not be achievable with real process gases containing trace contaminants.
Q3: What is the significance of the “response time” specification for in-situ (non-extractive) gas analyzers like TDLAS?
A: For in-situ analyzers (where the measurement is made directly in the process duct or stack), the response time specification includes both the analyzer’s electronic response and the time required for the gas composition at the measurement location to change. IEC 61207-7 (TDLAS) specifies that the response time must be stated as the t90 value measured by injecting a known gas concentration into the measurement path (e.g., using a calibrated gas cell inserted into the beam path) or by rapidly changing the gas concentration in a test cell. For TDLAS analyzers, t90 is typically 1-10 seconds, significantly faster than extractive analyzers (which add 5-60 seconds of transport delay through the sampling line). This speed advantage makes TDLAS the preferred technology for closed-loop process control applications where rapid composition changes must be detected, such as combustion optimization or chemical reactor feed control.
Q4: How does sample pressure variation affect the performance of different gas analyzer technologies?
A: The pressure sensitivity varies significantly across technologies: Paramagnetic O2: Linear pressure dependence (1% pressure change = 1% reading change). Requires pressure regulation to ±0.5% for 0.5% accuracy. Zirconia O2: Measures the oxygen partial pressure ratio, so absolute pressure changes affect both the sample and the reference equally when a reference gas is used. However, the Nernst equation has a logarithmic pressure dependence, making zirconia less sensitive to pressure than paramagnetic. NDIR: Pressure broadening of absorption lines and changes in the number of absorbing molecules per unit volume affect the measurement. A 1% pressure change typically causes 0.8-1.2% reading change. TDLAS: The absorption line shape (Voigt profile) depends on pressure. At low pressures (< 100 mbar), the lines are narrow (Doppler-broadened), and at high pressures (> 1 bar), they are broad (collision-broadened). TDLAS analyzers can compensate for pressure using the line shape itself (the line width reveals the pressure), making them inherently pressure-compensated in suitable implementations. IEC 61207 requires that the pressure dependence be documented for each analyzer type and that the sample pressure range over which the accuracy specification is maintained be clearly stated.
