Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
๐จ IEC 60591 Sampling & Analysis of Gas from Transformers โ Free Gases
Edition: IEC 60591:1978 | Status: Published International Standard (free gas sampling; for comprehensive DGA see IEC 60567 & IEC 60599)
📋 Standard Overview
IEC 60591 specifies standard methods for extracting, collecting, and analyzing free gases from transformers and other oil-filled electrical equipment. “Free gases” refer to gases that have accumulated in the headspace above the oil surface in oil-filled equipment, such as the gas collection chamber of the Buchholz relay or the gas pocket at the top of the tank, as distinct from “dissolved gases” absorbed within the insulating oil. The presence of free gases is typically a direct indicator of a severe internal fault—such as arcing or intense overheating causing rapid oil decomposition with gas generation rates exceeding the oil’s dissolution capacity—and constitutes a more urgent alarm signal than dissolved gas analysis (DGA).
The standard covers standardized operating procedures for gas sampling from Buchholz relay gas collection chambers, tank tops, and other gas collection points; technical requirements for sampling apparatus such as gas sampling syringes, cylinders, and gas sampling bags; and gas chromatography analysis conditions and parameters. IEC 60591 focuses on “gases that have already escaped from the oil,” and together with dissolved gas analysis (IEC 60567 and IEC 60599) forms a dual-layer protection system for condition monitoring and fault diagnosis of oil-filled electrical equipment.
🔬 Sampling Methodology
The quality of free gas sampling directly determines the reliability of analytical results. Key sampling requirements stipulated by the standard are as follows:
| Sampling Aspect | Technical Requirement | Key Point |
|---|---|---|
| Sampling Container | Gas-tight glass syringe (with 3-way valve), stainless steel gas cylinder, multi-layer composite gas bag (e.g., Tedlar bag) | Prohibit rubber or plastic seals — significant permeability to H₂ and CO₂ |
| Sampling Point Selection | Buchholz relay top sampling valve, tank top vent valve, bushing turret sampling port | Purge dead volume gas before sampling (≥3× line volume); record oil temp, load, ambient temp |
| Sampling Operation | “Oil displacement method”: fill and displace gas in the sampling line with the transformer’s own insulating oil to prevent air ingress | Never use compressed air or nitrogen for purging — may dilute or contaminate the sample |
| Sample Preservation & Transport | Analyze within 24 hours; store away from light at 10°C–30°C; secure syringe plunger with rubber band during transport | H₂ and CO₂ may diffuse out through certain polymer walls; glass syringes preferred over plastic |
| Analytical Instrument | Gas chromatograph (GC) with TCD + FID + methanizer | Simultaneous detection of 9 gases: H₂, O₂, N₂, CH₄, CO, CO₂, C₂H₂, C₂H₄, C₂H₆ |
Free gas analysis results are typically reported in volume percent (Vol%). Unlike DGA results, the compositional ratios of free gases directly reflect fault severity and nature. For instance, detection of high acetylene concentrations (C₂H₂ >1%) in free gas almost certainly indicates an arcing fault within the transformer. If air components (O₂ + N₂) account for over 90% of the free gas, it may merely originate from air ingress due to inadequate sealing rather than internal fault gas generation.
🏭 Synergistic Diagnosis with Dissolved Gas Analysis
Free gas analysis and dissolved gas analysis (DGA) constitute two complementary dimensions of transformer fault diagnosis. Dissolved gas concentrations in oil are influenced by multiple factors—gas solubility in oil, oil temperature, gas partitioning equilibrium between oil and gas phases (Ostwald coefficient)—and represent a time-accumulated quantity. Free gases, by contrast, are the direct overflow when the fault gas generation rate exceeds the oil’s dissolution capacity, exhibiting instantaneous and eruptive characteristics that make them particularly sensitive to rapidly developing severe faults such as turn-to-turn arcing.
In engineering practice, free gas results must be evaluated alongside dissolved gas results: ① If dissolved gas concentrations are steadily rising but no free gas has appeared, the fault is developing slowly, and planned outages can be scheduled; ② If combustible gases appear in the free gas, immediate component analysis should be performed and emergency measures taken based on fault characteristics; ③ If the Buchholz relay has tripped and collected free gas contains high concentrations of combustible gases (H₂ + C₂H₂ etc.), immediate de-energization and incident investigation are generally required. Deviation of the O₂/N₂ ratio in free gas from the atmospheric ratio (~0.27) can indicate whether air ingress or dilution by internally generated gas has occurred—a critical criterion for distinguishing genuine internal faults from sealing system deficiencies.
⚠️ Engineering Design Insight: The greatest pitfall in free gas sampling is air contamination. Even trace air ingress severely dilutes the sample, introduces spurious O₂ and N₂ readings, and may alter actual gas compositions through oxidative reactions (e.g., H₂ oxidized to H₂O by O₂). High-quality gas sampling operations must obey the following iron rules: ① Purge dead volume residual gas from the sampling line first (≥3× line volume); ② Use the transformer’s own oil as the filling and displacement medium for the sampling line, never any gas; ③ The hermeticity of sampling syringes or cylinders must be pre-verified via helium leak testing; ④ Retain at least one aliquot of each sample until analytical results are confirmed before discarding. At the automated monitoring level, while online DGA monitors (e.g., Hydran, Trafotest) enable real-time continuous monitoring, periodic offline laboratory GC analysis remains essential as the calibration benchmark.
🔑 Bottom Line: IEC 60591 provides the operational specification for standardized sampling and analysis of free gases from transformers. Correct gas sampling procedures are the first line of defense for obtaining reliable DGA results—deviations introduced by improper sampling can lead to fault misdiagnosis or missed alarms, with consequences potentially extending to substation- or grid-level major incidents. This standard’s core value lies in safeguarding the authenticity and reproducibility of fault diagnostic data through rigorous control of every step from sampling to detection.