Insulating liquid and gas
SF6 gas quality testing examines the sulfur hexafluoride inside a gas-insulated switch for moisture, purity, decomposition products and oil contamination. The gas is both the insulation and the arc-quenching medium, so its condition tells the owner whether the compartment was filled and sealed correctly and whether it is aging in service.
SF6 gas quality testing is the sampling and analysis of the sulfur hexafluoride gas held inside a gas-insulated medium-voltage switch, switchgear compartment or breaker. The gas serves two jobs at once: it is the dielectric medium that keeps energized parts apart in a compact enclosure, and it is the medium that cools and quenches the arc when the device opens. The test checks several properties of the gas. Moisture content, usually judged as a dew point or a water concentration, shows whether water has entered or was left behind at filling. Purity shows whether air or another gas has diluted the SF6. Decomposition products, including sulfur dioxide and other reactive byproducts, show that the gas has been broken down by arcing or partial discharge. Mineral-oil contamination shows whether compressor or handling oil has reached the compartment. In the published acceptance and maintenance specifications the gas tests appear as an optional item for medium-voltage SF6 switches, usually alongside the other electrical tests on the switch, and they are carried out with field or laboratory instruments designed for SF6.
The gas is invisible, sealed and expected to last for decades, so its condition cannot be seen by inspection. Moisture is the first concern. Water vapor in SF6 can condense on insulating surfaces at low temperature and reduce flashover strength, and under arcing it contributes to corrosive byproducts. Decomposition products are the second. Arcs and partial discharge split the gas into fragments that react with traces of water and with internal materials to form acidic compounds, which attack seals, contacts and insulators and are also hazardous to personnel who open the compartment. Air or oil contamination weakens the dielectric and interrupting ability of the gas. A gas problem may exist even when the gas density gauge reads normal, because density only says how much gas is present and not what it contains. Testing at acceptance proves the factory or field fill was clean and dry, and testing in maintenance catches slow leaks of moisture, early signs of internal arcing, and aging, before they lead to a flashover or a failed interruption in service.
Each property is measured by a method matched to what is being sought. Moisture is measured with a hygrometer that responds to water vapor in a flowing sample, using a chilled-mirror, capacitive or similar sensor, and the result is reported as a dew point or as a concentration. Because the dew point depends on pressure, the reading is related to the gas pressure and the reference conditions used for reporting. Purity is measured with an instrument that responds to the composition of the gas, for example by thermal conductivity or an infrared principle, and it reports the share of the gas that is SF6. Decomposition products are detected with chemical or electrochemical sensors, or with detector tubes, that respond to sulfur dioxide and related compounds. Oil contamination is checked by instruments or by observation of oil traces in the sampled gas. Each method assumes that the sample is representative and that the sampling line is dry and clean, since moisture clings to tubing and a short exposure to room air can affect a delicate reading. The gas is also treated as a controlled material, so the measured gas is recovered and not released.
The switch is made safe under the owner's procedures: de-energized, isolated, grounded and verified, with the compartment to be sampled identified from the drawings. The technician is trained in SF6 handling, with appropriate ventilation and protective equipment, because decomposition products can be toxic and the gas can displace oxygen in confined spaces. The instrument is connected to the compartment's fill or test valve with clean, dry, compatible hose, and the line is purged so room air does not reach the sensors. Gas is drawn slowly at a rate suited to the instrument, without letting the compartment pressure fall below what the manufacturer allows, and the exhaust is routed to a recovery system and not released to the atmosphere. Moisture readings are allowed to stabilize, since a dry-down curve at the start of the test is normal and the final value is what counts. Readings are taken for each compartment of a multi-compartment switch. The temperature and the compartment pressure are noted, and the instrument's calibration or verification status is confirmed. When finished, the connections are removed, the valve is checked for leaks, and gas that was used is accounted for.
The record holds the moisture result as a dew point or concentration with the temperature and pressure it applies to, the purity result, the result for decomposition products such as sulfur dioxide, and any observation of oil contamination. It identifies the compartment tested, the instrument and its calibration status, the sampling arrangement and the time allowed for stabilization. The gas density or pressure gauge reading and the ambient conditions are noted, together with the as-found state of the switch, whether it was newly filled, recently arc-exposed or long in service, and any gas added or recovered. These details let a later reader compare results fairly and tell a change in the gas from a change in the method.
No single reading decides acceptability, and this page does not give one. Results are compared with published limits that differ by property. For moisture the limit depends on the gas pressure and on the application, and often differs between a new fill and gas in service. For purity the criterion is a minimum share of SF6, and for decomposition products the criterion is a low ceiling, with higher concentrations calling for investigation or gas reclamation. The gas test sections of ANSI/NETA ATS-2025 and ANSI/NETA MTS-2023 list the test, and the acceptance tables in ANSI/NETA ATS-2025 give limits for the properties tested. A reading is also compared with the manufacturer's data for the equipment and with earlier results from the same compartment, since a steady drift matters even when the value remains inside a limit. Several properties together tell a story: a moisture rise with decomposition products suggests internal arcing, while moisture alone suggests a seal or fill problem. For the values, refer to your own licensed copies of the standards and the manufacturer's instructions.
At acceptance the test confirms the gas supplied and installed is clean and dry, and it is often run some time after filling so that moisture in the compartment has equilibrated with the gas. In maintenance, it is a condition check, repeated on a schedule or after a fault operation, a suspected leak or a reported gas alarm. Equipment class changes the details. Medium-voltage gas-insulated switches and switchgear have sealed compartments that are rarely opened, so gas testing is one of the few ways to see inside. Breakers that interrupt in SF6 are exposed to more arcing and generate more decomposition products, so the decomposition check carries more weight. Equipment with several gas compartments is sampled compartment by compartment because each has its own gas history. Some designs use a sealed-for-life pressure system with no valve for field sampling, so testing is done only where the manufacturer provides a way to do it. Local environmental rules on handling and recovering the gas apply to every case.
The equipment classes and standard sections that call for this test, and the UltraDb forms that carry it.
A density gauge only shows how much gas is in the compartment, not what the gas contains. Moisture, air, oil and decomposition products can all be present while the gauge looks fine, which is why a gas quality test adds information the gauge cannot supply.
Arcing and partial discharge break down some of the gas, and the fragments react to form compounds such as sulfur dioxide. Finding them suggests that interrupting duty or internal discharge has occurred, and they are also hazardous when a compartment is opened, so they are handled with care.
Water vapor can condense on insulating surfaces in cold conditions and reduce the flashover strength, and under arcing it helps form corrosive acids. The acceptable level depends on pressure and application, so check the published limits for your equipment.
No. SF6 is a potent greenhouse gas and a regulated material, so test exhaust is routed to a recovery system and gas is reclaimed according to the owner's procedures and local rules. Instrument and hose connections are leak-checked afterwards.
UltraDb references ASTM, ANSI, IEEE, NETA, NFPA, NEMA, FAA, TIA, and IEC standards descriptively, to identify the published methods its forms are built from. UltraDb and Verlecta are not affiliated with, endorsed by, certified by, or licensed by NETA, ASTM, ANSI, IEEE, NFPA, NEMA, FAA, TIA, IEC, or Megger. Standard names and section numbers are the property of their respective organizations. Always test to the edition your contract specifies. Where a standard is sold through the ANSI webstore, Verlecta may earn a commission on a purchase made through our link, at no extra cost to you. A commission never decides which standards we cite or how a form is built.