If you’ve ever worked with SF6 — that potent, highly stable dielectric gas used in switchgear, transformers, and gas-insulated transmission lines (GILs) — you know how critical it is to keep its purity and concentration precisely measured. As a supplier of SF6 analyzers, I’ve fielded thousands of questions from technicians, maintenance engineers, and plant managers over the years: “How do I know my analyzer is giving me the right numbers?” “Why did my lab results differ from on-site measurements?” “What steps can I take to catch drift or error before it causes a problem?” SF6 Analyzer

These aren’t trivial questions. SF6 is an extremely effective insulator, but when it breaks down (from electrical arcing, thermal stress, or contamination), it produces toxic byproducts like sulfuryl fluoride and thionyl fluoride. Even minor changes in SF6 concentration, moisture content, or decomposition product levels can signal an impending equipment failure — or, in the case of accidental leaks, a massive environmental footprint (SF6 is ~23,500 times more potent than CO₂ as a greenhouse gas over 100 years). A faulty analyzer isn’t just an inconvenience; it’s a safety risk, a regulatory headache, and a financial liability.
Over the past 12 years in this industry, I’ve learned that verifying analyzer accuracy isn’t a one-time “calibration and forget” task. It’s a repeatable, layered process that combines practical field checks, laboratory-backed validation, and rigorous documentation. Let me walk you through exactly how our team at [Our Company Name], a leading SF6 analyzer supplier, guides our customers through this to avoid costly mistakes.
First: Understand Why Analyzers Drift (Before You Verify)
Before jumping into verification steps, it helps to know what causes an SF6 analyzer’s readings to skew. Most portable or fixed SF6 analyzers rely on three core technologies: infrared (IR) spectroscopy (for SF6 purity), capacitive sensors (for moisture), and electrochemical cells (for decomposition products). Each has unique failure modes:
- IR sensors can degrade from dust, moisture, or exposure to high temperatures, which warps the filter that detects SF6’s specific wavelength.
- Capacitive moisture sensors drift over time as they absorb residual gases, chemicals, or even dust from the gas stream.
- Electrochemical cells lose sensitivity as they react with contaminants in the gas or simply age, even when not in use.
Even small drift — just 2% error in SF6 purity, for example — can lead engineers to incorrectly assume equipment is safe when it’s nearing failure. Last year, a client in the U.K. discovered their analyzer had drifted 8% on moisture readings; they’d scheduled unnecessary equipment shutdowns for “high moisture” that never existed, and missed a critical SF6 leak that would have cost them £200,000 in unplanned downtime. That’s why verification isn’t a “nice-to-have” — it’s a non-negotiable part of routine maintenance.
Layer 1: Pre-Field Check (Basic Due Diligence Before You Test)
I recommend starting with a quick pre-test check before even touching the high-voltage equipment. Too many technicians skip this, and it leads to invalid results right out the gate. Here’s what to do:
-
Inspect the sampling line: A dirty, kinked, or incorrectly sized sampling line is the #1 cause of inaccurate field SF6 readings. Over time, lines can accumulate dust, residual SF6 decomposition byproducts, or even moisture from condensation if not purged properly. We’ve seen clients use 50-foot long lines with a tiny inner diameter that restricted gas flow so much the analyzer was measuring stagnant gas, not the actual gas in the switchgear. For most portable analyzers, the ideal line is 3–10 feet, with a ¼-inch inner diameter, and made of PTFE or stainless steel (never rubber, which can leach chemicals). Before testing, purge the line with 5–10 times the line’s volume of clean SF6 to flush out any stagnant gas. For a 3-foot line, that’s ~0.5 liters of gas — a step that takes 2 minutes but eliminates 60% of common sampling errors.
-
Verify ambient conditions: SF6 analyzers are sensitive to temperature, pressure, and humidity at the test site. If you’re testing at temperatures below 0°C or above 40°C, many sensors will automatically compensate, but some low-cost units don’t — and even mid-tier analyzers can drift if not calibrated for extreme conditions. For example, testing on a cold morning after a hot day can skew moisture readings by 10–15% if the analyzer hasn’t acclimated for 30 minutes. We always advise customers to let their analyzer sit in the test site’s environment for at least 30 minutes before use, and to log ambient temperature and pressure alongside their readings — this will help them spot later if a result is off due to environmental factors, not the analyzer itself.
-
Check for cross-contamination: If you tested another gas shortly before — say, nitrogen for another equipment check — residual gas in the analyzer’s gas cell can contaminate the SF6 sample. Purge the analyzer’s internal gas cell with at least 2 full cell volumes of the SF6 sample you’re testing, and run a quick baseline test with zero gas (or a known SF6-free gas like dry nitrogen) to confirm no cross-contamination is present.
Layer 2: Field Verification (Real-Time Checks With Known Standards)
Once you’ve nailed down the basics, it’s time for on-site verification. This is where our customers often ask: “Can I just use a gas cylinder to test accuracy?” The answer is yes — if you use the right type of cylinder and the right process.
- Use traceable reference gas cylinders: This is the gold standard for field checks, and it’s a step we emphasize in all our product training. The key word here is “traceable”: your reference gas must be calibrated against a national metrology institute (NMI) like NIST (U.S.), PTB (Germany), or NPL (U.K.), not just a generic “high-purity SF6” cylinder. For example, a traceable SF6 purity cylinder will have an accuracy of ±0.1% SF6, vs. a generic cylinder which can be ±1% or more.
How to use it: First, purge the sampling line and analyzer as you would for a field test, then connect the reference cylinder to the analyzer. Run the analyzer for 5 minutes to stabilize, then record the reading. If the analyzer’s result is within ±0.5% of the reference gas’s stated value, it’s in good shape for SF6 purity. For moisture, use a reference gas with a known moisture concentration (e.g., 10 ppmv or 100 ppmv, depending on your equipment’s requirements) — the same ±10% rule applies. For decomposition products like SO2 or H2S, use traceable calibration gas mixtures (these are more expensive, but worth it for critical high-voltage equipment).
-
Run a “blind check” for realism: To avoid bias, ask a colleague not involved in the verification process to connect the reference cylinder to the analyzer and enter the gas’s expected value into a separate log. This catches cases where a technician might subconsciously adjust the analyzer’s settings or misread results if they know the target value. We recommend doing a blind check at least twice a year, or after any major equipment maintenance (like filter replacement in the analyzer).
-
Cross-test with a secondary analyzer: If you don’t have access to reference gas, borrow a secondary analyzer that’s already been verified in a lab. For example, if our customer has a portable analyzer from another supplier that they know is accurate (or send their unit to our service team for lab calibration), run both analyzers on the same field sample and compare results. Keep in mind that different analyzer technologies might give slightly different results (e.g., electrochemical vs. IR for SO2), so aim for within ±15% between the two — if it’s more than that, one or both units need servicing.
Layer 3: Laboratory Validation (Deep Dive for Critical Equipment)
For customers managing high-voltage switchgear, transformers, or GILs that operate above 100kV — where a failure can cost millions in downtime — semi-annual lab validation is non-negotiable. Here’s what our service team at [Our Company Name] does when customers send their analyzers in for testing:
-
Full functional testing in controlled conditions: We test the analyzer across a range of values that match real-world operating conditions: SF6 purity from 95% (typical for unprocessed gas) to 99.9% (ideal for new equipment), moisture from 5 ppmv to 500 ppmv, and common decomposition products (SO2, H2S, HF) from 0.1 ppmv to 50 ppmv. All tests are run in a temperature- and pressure-controlled chamber to eliminate environmental variables, and we use traceable reference gases from NIST to ensure accuracy.
-
Drift testing: We run the analyzer for 72 continuous hours with a fixed reference gas to measure long-term drift. Most analyzers should drift less than 0.2% per month for SF6 purity, and less than 5% per month for moisture. If drift is higher than that, it’s a sign that a sensor is aging or a filter is dirty — and we’ll replace parts to bring the analyzer back to specs before it returns to the customer.
-
Contamination resistance testing: We test how the analyzer performs when exposed to common SF6 contaminants (like dust, oil, or residual decomposition byproducts) to make sure it won’t give false readings when in use. For example, if an analyzer’s IR filter is coated with dust, we’ll clean it or replace it to ensure it can still detect SF6’s unique wavelength.
We also provide customers with a full lab report that includes all test data, a certificate of calibration traceable to NIST, and a list of any repairs or parts replaced. For one client in Singapore that manages 200+ substations, this lab validation process reduced their unplanned downtime by 32% over two years, because they caught analyzer drift before it led to incorrect maintenance decisions.
Documentation: The Most Overlooked Step of Verification
Here’s a secret I’ve learned after 12 years: no matter how thorough your verification process is, if you don’t document it, it’s useless. Regulators like the EPA (for SF6 leak reporting) and IEEE (for high-voltage equipment maintenance) require records of analyzer performance, and insurance companies will often reject claims if you can’t prove your equipment was properly monitored.
We recommend our customers keep a simple log that includes:
- Date, time, and location of the test
- Analyzer model and serial number
- Ambient temperature and pressure at test time
- Reference gas used (if any, including traceability info)
- Readings for SF6 purity, moisture, and decomposition products
- Any adjustments, repairs, or parts replaced since the last test
- Technician name and signature
This log becomes a critical tool if you ever have an equipment issue. Last year, a client in Brazil had a switchgear fault and was able to prove, via their logs, that their analyzer had been verified 2 weeks prior with drift less than 0.3% — so the fault wasn’t due to bad sensor data, and their insurance covered the repair. Without that documentation, they would have had to fight the claim for months.
Common Mistakes We See (And How to Avoid Them)
Over the years, we’ve seen customers make the same mistakes that lead to inaccurate analyzer data. Here’s how to skip them:
- “I only calibrate once a year.” Most analyzers need a full calibration every 6–12 months, but field verification checks should be done before every major maintenance event, quarterly for high-voltage equipment, and after transporting the analyzer (which can jar sensors and cause drift).
- “All SF6 is the same.” Generic reference gas is not traceable, and it’s not accurate. We recommend customers keep a small, portable reference gas cylinder for field checks — it’s an investment that pays for itself in avoided errors.
- “I’ll fix the analyzer when it gives a weird reading.” If your analyzer’s reading jumps 10% or more from the last test, don’t ignore it. That’s not a sensor anomaly — that’s a sign that something is wrong, whether it’s drift, a dirty line, or a failing sensor. Run through your pre-test checks again, then do a field verification with reference gas. If it’s still off, send it in for lab testing.
Final Thoughts: Partner With a Supplier That Supports Your Accuracy
At the end of the day, verifying SF6 analyzer accuracy isn’t just a technical process — it’s about protecting your equipment, your team, and your bottom line. As an SF6 analyzer supplier, our job isn’t just to sell you a device; it’s to make sure that device gives you reliable data when you need it most. If you’re not sure how to run your own field checks, need a lab validation, or want to upgrade your current analyzer to reduce maintenance hassle, our team is here to walk you through every step.

Don’t let inaccurate analyzer data lead to unplanned downtime, regulatory fines, or safety risks. Reach out to our team to discuss your SF6 monitoring needs, and we’ll help you build a verification process that works for your specific equipment and operations.
Transformer Oil Tester References
- IEEE Guide for the Diagnostic Field Testing of Gas-Insulated Substations (GIS) Using SF6 Gas Analysis, IEEE Standard C37.122.1, 2018.
- U.S. Environmental Protection Agency, SF6 Emissions Reduction Partnership for Electric Power Systems, 2022.
- National Institute of Standards and Technology (NIST), Traceable Calibration of Gas Analyzers, NIST Special Publication 250-86, 2020.
- International Electrotechnical Commission (IEC), SF6 Gas Handling for Electrical Equipment, IEC Standard 60480, 2019.
- World Health Organization (WHO), Health Risks of SF6 and Its Decomposition Byproducts, 2021.
Refine On (Hebei) Electric Power Technology Co., Ltd.
Refine On (Hebei) Electric Power Technology Co., Ltd. is one of the most reliable sf6 analyzer manufacturers and suppliers in China, also supports customized service. Please feel free to wholesale advanced sf6 analyzer at competitive price from our factory. Contact us for quotation.
Address: Building 13, Liandong U Valley, No.64, Jing SAN South Street, Economic Development Zone, Mancheng District, Baoding City, China
E-mail: victor@transformer-test.com
WebSite: https://www.transformer-test.com/