GB/T 17623-2017 Determination of the contents of dissolved gas components in insulating oil by gas chromatography

Summary page. This page is a compiled excerpt/summary of the standard — it is not the full standard text and does not replace it. Verify every value against the official standard and the product certificate before use.

GB/T 17623-2017 specifies the gas chromatographic determination of the component contents of gases dissolved in insulating oil; this page is historical study material on the withdrawn 2017 edition, covering sampling, the degassing routes, calibration, volume correction and precision.

Evidence note: the 2017 edition was withdrawn on 2026-08-01 and wholly replaced by the 2026 edition. This page is historical study material for the designated 27-page original of the 2017 edition, not the full text of the standard. Technical review is pending, and the page is not a basis for controlled operation, automatic calculation or product applicability.

Identity and historical position of the 2017 edition: PDF 1/3; official old-edition detail and the 63 items of the 2026 announcement. This page studies GB/T 17623-2017, "Determination of the contents of dissolved gas components in insulating oil by gas chromatography". The cover states issue on 2017-05-12 and implementation on 2017-12-01, replacing the 1998 edition; the official old-edition record states explicitly that it was withdrawn on 2026-08-01 and wholly replaced by the 2026 edition. This page keeps the old-edition study entry, does not call it current, and does not substitute the 2026 text for the 2017 original.

Gases covered and conditions of the result: PDF 4, clauses 1 and 3. Gas is extracted and separated by gas chromatography to determine hydrogen, methane, ethane, ethylene, acetylene, carbon monoxide, carbon dioxide and components such as oxygen and nitrogen in insulating oil. The concentration of the extracted gas must still be converted to the content in oil through the gas-extraction system; the concentration in the chromatographic gas sample cannot be taken directly as the result in oil. Dissolved gas results are referred to 20 °C and 101.3 kPa and expressed as microlitres of gas per litre of oil, i.e. μL/L. Free gas, gas in oil and the acetone of Annex G in mg/L are measured separately and are not mixed.

Sample collection and preservation: PDF 4, clause 4 and PDF 6, clause 5.4. The original cites the sampling provisions of GB/T 7597 and requires transport and storage to avoid contamination, leakage, light and drying. Whether leakage or inter-phase exchange occurs after sampling affects low-level results; this page does not reconstruct the complete sampling standard or the controlled operations. The tightness of glass syringes must be checked. Original 5.4 fills 40.0 mL of hydrogen into a 100 mL syringe at 20 °C and 101.3 kPa and compares the gas released after 6 h, requiring a decrease of not more than 2.5 %. This is an equipment check, not a product shelf life or a storage guarantee.

The three routes of gas extraction: PDF 4/5, clause 5.1 and PDF 8–11, clause 8.1. Constant-temperature timed shaking extraction is based on headspace partition equilibrium; vacuum degassing uses vacuum and repeated collection; the automatic headspace sampler injects after the closed system reaches the specified equilibrium. The three routes use their own calibration and standardisation relationships, and the coefficients of one device must not be applied directly to another. Original 5.1.2 requires a vacuum degassing rate of not less than 90 % with a variation in the mean degassing rate of not more than 5 %. The Toepler pump schematic and the complete vacuum safety and leak-testing procedures have not been reconstructed.

Chromatographic detectors and calibration: PDF 5/6, clause 5.2 and PDF 11, clause 8.2.1. TCD and FID analyse different species; carbon monoxide and carbon dioxide pass through a conversion path and are then detected by FID; the separation of hydrogen and air components must match the actual column and carrier gas. The original flow chart lists several configurations, and no single configuration is held to suit all components. The standard mixed gas needs component concentrations and evidence of production and verification; the original uses peak area or peak height to obtain the average response. The relative deviation of repeated measurements in gas standardisation should lie within ± 1.5 % of the mean, with at least two repeats. The instrument calibration threshold is not the repeatability of the whole method.

Minimum detectable concentrations of the original: PDF 8, clause 6, Table 3. Table 3 gives the minimum detectable concentration of components in oil; it is not the measured detection limit of every instrument, nor a promise for a particular product or batch. The entry "air" is retained as listed in the original and is not split into nitrogen and oxygen limits on this page. Gas category (as listed) | Minimum detectable concentration / μL·L⁻¹ Hydrogen | 2 Hydrocarbons | 0.1 Carbon monoxide | 5 Carbon dioxide | 10 Air | 50

Manual shaking headspace: time and temperature sequence: PDF 9/10, clause 8.1.1.2.1. The original manual procedure uses a 100 mL syringe with 40.0 mL of oil sample and 5.0 mL of nitrogen or argon. After reaching 50 °C it is shaken for 20 min, left to stand for 10 min, and then cooled below 10 °C before the volume and separation steps continue. The manual 20 min cannot replace the 40 min of Table 4 for automatic headspace. Excluding needle/interface dead volume, maintaining gas-liquid separation and the orientation of the syringe all affect the result. The complete manual transfer, needle handling and safety protection have not been reconstructed, and the card should not be followed as a direct operating instruction.

Automatic headspace conditions as the original example: PDF 7/8, clause 5.8 and Table 4; PDF 10, clause 8.1.1.2.2. The headspace vial has a volume of 20 mL, with 10.0 mL of oil sample actually added. Original Table 4 relates to this automatic sampler system and must match the equipment and method validation; it cannot be extrapolated as a universal procedure for all headspace instruments. Item (original Table 4) | Original condition Sample temperature | 50 °C Inlet line temperature | 150 °C Loop temperature | 150 °C Headspace vial pressure | 70 kPa Nitrogen or helium carrier gas | 12 mL/min Shaking equilibration time at 50 °C | 40 min Pressurisation / pressure equilibration / loop filling | 0.25 min each Injection time | 0.9 min

Volume correction and temperature basis: PDF 11/12, clause 8.3.1.1, equations 3/4/5. Manual headspace equation 3 corrects the equilibrium gas volume: V′g = Vg × 323/(273 + t). Equation 4 corrects the oil sample: V′l = Vl[1 + 0.0008(50 − t)]. Both use 50 °C as the basis and t is the room temperature in °C. Automatic headspace equation 5 is V′g = V − V′l, where V is the vial volume. Vacuum extraction equations 7/8 instead correct to 20 °C and 101.3 kPa; 323 and 293, or the two sample-temperature coefficients, must not be mixed. No calculator is published on this page; all of equations 6–10 and the linkage of percentage recovery inputs require separate validation in controlled software.

Partition coefficients and recovery kept separate: PDF 9/12/13; PDF 18–25, Annexes B/D/E/F. The gas partition coefficient K, the vacuum degassing rate R, the calibration gas concentration and the peak area have different roles. Original Table 5 lists coefficients at 50 °C for domestic mineral insulating oils, and Annex E lists coefficients at 70 °C for imported mineral insulating oils; they cannot be reused directly when oil type, temperature and extraction route do not match. Annex D determines the partition coefficient by the two-equilibration method; two results in the same laboratory must not differ by more than 5 % relative, and results from different laboratories by more than 10 % relative. This is the precision of the coefficient determination and cannot replace the requirements for gas-in-oil results in clause 9. Annex F is used for the second dissolution equilibrium determination where K is not cited, and does not apply to low gas concentrations.

Concentration boundaries of the clause 9 precision: PDF 13/14, clauses 9.1/9.2. The arithmetic mean of two parallel determinations is finally used. When the concentration in oil is greater than 10 μL/L, the difference between the two results must be less than 10 % of the mean; when it is less than 10 μL/L, less than the sum of 15 % of the mean and the minimum detectable concentration of that gas. The original does not specify a branch for exactly 10 μL/L, and ≤ or ≥ is not added here. Relative deviation between laboratories: less than 15 % when above 10 μL/L, and less than 30 % when below 10 μL/L. The original gives no statement of 95 % probability, and 95 % must not be added; this method precision is not the uncertainty of a batch certificate.

Recovery verification and calibration records: PDF 14, clause 9.3 and PDF 18/19, Annex B. The original verifies accuracy by a recovery test with standard oil; the recovery must not be lower than 90 %, and the cause is investigated when it is not met. The preparation of the standard oil, the gas extraction and the calculation of the result must all be recorded; the certified value of the mixed gas cannot be taken directly as the true value of the oil sample after preparation. Annex C for automatic headspace links the response correction factor to the automatic flow, with repeatability of ± 1.5 % and at least two repeats listed separately. The oil-preparation calculation, percentage inputs and software path have not been fully validated, and a passing recovery is not extrapolated into an unbiased result or a promise of product applicability.

Nature of the annexes and the separate acetone analysis: PDF 2/15–27. Annex A (Toepler pump degassing) and Annex D (determination of partition coefficients) are marked normative; B (standard oil recovery), C (automatic sampler calibration), E (coefficients for imported mineral oils), F (second equilibrium) and G (acetone analysis) are informative. The marking of a document's nature cannot replace a controlled work instruction. Annex G is devoted to the analysis of acetone in operating transformer oil, with FID, a DB-WAX column and an independent mg/L external-standard relationship. Its standard stock solution and the 65 °C/30 min vial conditions are not conditions for other gas determinations or for product storage. The figures, manufacturing dimensions, all annex formulas and controlled safety have not been reconstructed.

Source evidence and outstanding acceptance: All 27 physical pages of the designated 2017 original PDF; official old-edition record and new-edition announcement. This study card was compiled from a visual overview of the 27 physical pages of the 2017 original and checks of key single pages such as 8/13/14. A strict mapping that has no OCR of the same year does not mean the original PDF is missing pages; the 2026 text has not been used as a substitute. The original is kept read-only and scans are not published. The complete chromatographic flow path, the cell-by-cell column specification, the Toepler pump assembly, the annex oil-preparation values and software, and all controlled safety and bench procedures still await technical review. Official edition evidence is metadata and cannot serve as independent technical acceptance. GEO citation, actual visits or enquiries have no effect evidence yet.

View public related products