GB/T 382-2017 Standard test method for smoke point of kerosene and aviation turbine fuel

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.

Evidence note: unofficial method study card for the historical 2017 edition; the edition has been withdrawn, and this is neither the full text of the standard nor an operating procedure. Technical review is pending and product applicability is unconfirmed; the original figures have not been reconstructed, and the precision conditions and blank dimensional tolerances are retained.

1. Method identity and version boundaries: Main PDF pages 1 and 5-7; both official status sites checked 2026-10-06. This card covers GB/T 382-2017, Standard test method for smoke point of kerosene and aviation turbine fuel; the original records issue on 2017-07-31 and implementation on 2018-02-01. The foreword states that it replaces the 1983 edition and adds automatic apparatus, quality control samples, and precision and bias for manual and automatic methods. This card does not mix conditions from other edition years into the 2017 edition. Both the National Standards Full-Text Public System and the National Standards Information Public Service Platform mark this edition as withdrawn; the latter records withdrawal on 2026-05-01 and full replacement by GB/T 382-2025. Only the version status is stated here; no technical equivalence of the two editions is inferred, and this does not replace the controlled document or a review of the applicability to an actual sample.

2. Scope, definition and significance: Main PDF pages 6-7, introduction and clauses 1 and 3. This edition covers manual and automatic smoke point determination of kerosene and aviation turbine fuel. The smoke point is expressed in millimetres and is the maximum height of a non-smoking flame under the specified lamp and conditions; it is not the flash point, the combustion temperature or a commercial acceptance limit of the fuel. The introduction uses the smoke point to characterize the relative smoking tendency of a diffusion flame; a high smoke point generally indicates a low smoking tendency, and the relationship with radiative heat transfer is discussed. This does not constitute a commitment by this card regarding a specific fuel, engine life or product performance.

3. Principle and the difference between manual and automatic: Main PDF pages 7-8, clauses 4 and 5. The sample is burned in a closed wick lamp, and the maximum smokeless flame height is determined after calibration with pure hydrocarbon mixtures of known smoke point. Manual observation reads to 0.5 mm; automatic observation reads to 0.1 mm. The 0.05 mm minimum resolution of the camera measurement is a different quantity and must not be confused with the rounding of the reported result. The automatic system identifies the flame using a camera, software, candle displacement and a database of atmospheric pressure/calibration values. Clause 5.3 of this edition specifies the automatic method for arbitration in case of dispute; this is not a basis for considering any automatic instrument to comply with the standard.

4. Apparatus and wick: Main PDF pages 8 and 18-20, clause 5 and Annex A. The apparatus comprises a manual or automatic smoke point lamp, a barometer accurate to ±0.5 kPa, a pipette and a 25 mL graduated cylinder. The wick is a solid cotton yarn round braid: 17 face yarns/3 plies of 66 count, 9 core yarns/4 plies of 100 count, 2 plies of 40 count weft, with 6 weft threads per cm. Annex A covers the top of the wick guide tube level with the zero of the scale, the white mark on the black glass, the 2 mm dividing strip, the 50 mm range with 1 mm divisions, numbering every 10 mm and long lines every 5 mm; the lift travel is at least 10 mm and smooth, the curved window reduces multiple images, and the reservoir connection must not leak oil. The assembly shown is not redrawn in this card, and the dimension table cannot be used alone for manufacture or for judging equipment conformity.

5. Reagents, standard fuels and quality control: Main PDF pages 8-9, clause 6 and Table 1. This edition lists analytical-grade toluene, anhydrous methanol, n-heptane, and 2,2,4-trimethylpentane (isooctane) of minimum mass purity 99.75%. The standard mixtures are prepared with a calibrated burette or pipette; the table below restates the facts of Table 1, where the smoke point corresponds to an atmospheric pressure of 101.3 kPa and the composition is a volume fraction, and it must not be treated as the value of an unknown product or a batch certificate. The quality control sample should be a stable petroleum product representative of the characteristics of the samples under test, used to check the state of the system. The specific procurement identity, packaging and certificate are not confirmed in this card, and no product relationship is added. | Standard smoke point mm (101.3 kPa) | Toluene % (volume) | Isooctane % (volume) | | --- | --- | --- | | 14.7 | 40 | 60 | | 20.2 | 25 | 75 | | 22.7 | 20 | 80 | | 25.8 | 15 | 85 | | 30.2 | 10 | 90 | | 35.4 | 5 | 95 | | 42.8 | 0 | 100 |

6. Sampling and wick treatment: Main PDF pages 9 and 22, clauses 7 and 8.3 and Annex C. Sampling in the original refers to GB/T 4756; the sample is tested as soon as possible after sampling, without artificial heating, and is brought to a room temperature of 20 °C ± 5 °C; cloudy samples or foreign matter are filtered through qualitative filter paper. No shelf life applicable to all samples is given, and none is added by this card. New and reused wicks must be extracted with equal volumes of toluene and anhydrous methanol, for at least 25 cycles of filling and emptying. They are first partly dried in a fume cupboard, or in a forced-ventilation, explosion-proof oven; after drying at 105 °C ± 5 °C for 30 min they are placed in a desiccator. Commercially pre-extracted wicks must be shown to comply with this treatment, and are still extracted under this clause before use. Annex C permits a Soxhlet or similar apparatus, with the time required for 25 cycles calculated from the measured time of two cycles; this substitution of time for counting must be based on an actual cycle measurement, not on an arbitrary fixed extraction duration. Safety conditions are still implemented in accordance with the controlled documents.

7. Instrument preparation and calibration: Main PDF pages 9-10, clauses 8.1-8.2 and 9. The manual lamp is placed vertically in a draught-free environment, with the air holes clean and correctly dimensioned; where a draught shield is needed, non-asbestos flame-retardant material is used, with the top of the shield at least 150 mm above the chimney and the inside dark. For manual calibration, two standard mixtures whose smoke points bracket the sample are preferred; where bracketing is not possible, the two closest are used. Recalibration is required when the instrument or operator changes or when the atmospheric pressure changes by more than 0.7 kPa; each operator verifies daily with the control sample, and calibrates when the control limit is exceeded or with a new instrument. The automatic system stores calibration values for 7 mixtures at different atmospheric pressures together with the measured pressure; a corresponding record within ±0.7 kPa of the observed pressure should exist, and only when the record is missing is recalibration with the 7 mixtures required. The manual rule that a pressure change requires recalibration must not be applied to automatic instruments across the board. Automatic QC verification is at intervals not exceeding 7 days or when the instrument changes, with recalibration beyond the limit; a change of operator alone does not require recalibration.

8. Study of the oil reservoir and wick assembly: Main PDF pages 11-12, clause 10.1 and Figures 3-5. This edition takes at least 125 mm of dry wick, which is fitted after wetting with the sample while avoiding twisting; a newly treated wick is used for dispute/arbitration, with the burning end pre-wetted. About 20 mL, and at least 10 mL, of sample at room temperature is placed in a clean, dry reservoir, avoiding the air holes. When the wick trimmer assembly is not used, the wick head is cut flat and projects 6 mm from the wick tube; the protective coating on the blade must first be removed. When the assembly is used, the illustrated sequence is: holder, piercing with pointed pliers, fitting to the reservoir, trimming flat, removing the holder. Figure 4 of the main scan is very dark, and Figures 4/5 have been cross-checked against a 20-page scan of the same edition; the original images are not published, and this paragraph is not a complete assembly instruction.

9. Manual observation and re-testing: Main PDF pages 12-14, clause 10.2 and Figures 6-7. After ignition the wick burns with a flame of about 10 mm for 5 min, and is then raised until smoking begins and slowly lowered. The original figures distinguish an excessively high pointed flame, flame B where the point has just disappeared, and an excessively low round flame; flame B is read to 0.5 mm. Parallax is eliminated by making the two readings, through the flame and in the mirror image, agree. Independent judgement training without the original figures still requires the controlled standard and equipment training. Three consecutive observations are made; if the change exceeds 1.0 mm, a new sample and a new wick are required. The removed reservoir is rinsed with n-heptane and blown dry. The 1.0 mm limit on the difference between the three observations must not be treated as the repeatability r of clause 12 or as a sample acceptance criterion.

10. Automatic observation and calculation of results: Main PDF pages 14-15, clauses 10.3 and 11; main PDF page 10, equation (1). The automatic route enters the sample data and the pressure at the time, raises and ignites the reservoir, burns at about 10 mm for 5 min, then raises until smoking begins and slowly lowers; the software locates flame B, reads to 0.1 mm, records, lowers and extinguishes. If the change between the 3 observations exceeds 1.0 mm, a new sample and wick are indicated; cleaning is as for the manual method. Instrument-specific actions require the manufacturer's instructions and cannot be replaced by this paragraph. Equation (1): f = [(As/Ad) + (Bs/Bd)] / 2. As and Bs are the standard smoke points of the two mixtures, Ad and Bd the measured smoke points, all in mm. The OCR of this equation was defective and has been restored against main PDF page 10 and another scan page 8. Where the sample exactly equals the determined value of one mixture, the original text requires the second mixture to be chosen slightly higher; other choices retain the conditions of clause 9. Equation (2): smoke point = L × f. L is the mean of 3 independent readings, accurate to 0.1 mm; f is accurate to 0.01; the final smoke point is accurate to 0.1 mm. The automatic method follows the same calculation rule, and the 0.5 mm reading must not be confused with the 0.1 mm of the final report.

11. Precision, bias and reporting: Main PDF pages 15-17, clauses 12-13, equations (3)-(7) and Table 2. The precision comes from 15 blind samples, 11 manual and 13 automatic instruments and different laboratories, at the 95% confidence level. Manual r = 0.06840(X+16), R = 0.09363(X+16); automatic r = 0.02231X, R = 0.01651(X+30). X is the mean of the two results compared, in mm; r is the limit for the difference between repeat determinations by the same operator/equipment and R the limit for independent results from different laboratories; the two are not interchangeable. The table below gives the typical values of original Table 2, with the printed values retained after checking each column against the scan; the four equations have been recomputed to two decimal places and agree with the 28 r/R values in the table. The 15-42 values are the example smoke points listed in that table, and the listed values alone cannot automatically widen or narrow the scope of the method. The original states that the smoke point is defined by this method and distinguishes the manual/automatic relative bias: equation (7) (A+16) = (M+16)/1.016, where A is the automatic and M the manual result. Clause 12.4.3 applies no bias correction to automatic results, and the relationship is not used to rewrite the report on the card's own initiative. Both routes report to 0.1 mm and retain the identity of the manual or automatic method used. | X mm | Manual r mm | Manual R mm | Automatic r mm | Automatic R mm | | --- | --- | --- | --- | --- | | 15 | 2.12 | 2.90 | 0.33 | 0.74 | | 20 | 2.46 | 3.37 | 0.45 | 0.83 | | 25 | 2.80 | 3.84 | 0.56 | 0.91 | | 30 | 3.15 | 4.31 | 0.67 | 0.99 | | 35 | 3.49 | 4.78 | 0.78 | 1.07 | | 40 | 3.83 | 5.24 | 0.89 | 1.16 | | 42 | 3.97 | 5.43 | 0.94 | 1.19 |

12. Annex dimensional facts and drawing boundaries: Main PDF pages 18-20, Table A.1 and Figures A.1-A.2 of Annex A. The table below merges Table A.1 with its continuation into a factual rearrangement, in mm; the accompanying fit requirements, the ellipsis and the blank tolerances of the original table are retained and are not guessed as 0 or ±0.05. The OCR misaligned the 21.25 internal diameter of the reservoir body with the following tolerances, and both scans of the same edition confirm that this row has no numerical tolerance. Figure A.2 marks other dimension lines, 22, 60, 90 and 109; the 60 in the figure must not be taken directly as covering the 82.0 overall length of the wick tube in the table. The dimensional datum and installation relationships of the drawing are not reconstructed in this card; the table only aids comparison with the original and is not a machining drawing, an acceptance checklist for the complete instrument, or the full text of the standard. | Original part / item | Dimension or fit requirement mm | Tolerance mm | | --- | --- | --- | | Holder C internal diameter | 23.8 | ±0.05 | | Wick guide tube D internal diameter | 6.0 | ±0.02 | | Diameter of the 20 air inlets E | 2.9 | ±0.05 | | Platform F external diameter | 35.0 | ±0.05 | | Diameter of the 20 air inlet holes in the platform | 3.5 | ±0.05 | | Lamp body G internal diameter | 81.0 | ±1.0 | | Lamp body G internal depth | 81.0 | ±1.0 | | Chimney H internal diameter | 40.0 | ±1.0 | | Height from chimney top to lamp body centre | 130 | ±1.0 | | Reservoir body internal diameter | 21.25 | No numerical value given (blank in the original table) [source unclear] | | Reservoir body external diameter | Slight sliding fit in the holder | No numerical value given | | Reservoir length (excluding cap) | 109.0 | ±0.05 | | Cap thread | φ9.5 thread, 1.0 pitch | No numerical value given | | Wick tube A internal diameter | 4.7 | ±0.05 | | Wick tube A external diameter | Tight fit with the wick guide tube | No numerical value given | | Wick tube A length | 82.0 | ±0.05 | | Air tube B internal diameter | 3.5 | ±0.05 | | Air tube B length | 90.0 | ±0.05 |

13. Safety and reference dependencies: Main PDF pages 7 and 21-22, clause 2, Annexes B-C. The references include GB/T 4016-1983 terminology, GB/T 4756 sampling and NB/SH/T 0843 statistical evaluation. This round verified only the reference identities on the 2017 pages; the full text of the referenced standards was not reproduced and the current controlled references were not confirmed on behalf of the laboratory. Annex B notes the flammability/vapour hazards of toluene, isooctane, methanol and n-heptane respectively, with methanol additionally carrying a risk of fatal poisoning or blindness; flame operation, extraction and drying require ventilation, fire and explosion protection equipment and personal protection. This card is only a hazard study note and cannot replace the actual SDS, the safety assessment or a complete operating procedure.

14. Verification record, missing items and questions: Page sequence and item-by-item verification evidence for the 24-page main scan and the separate 20-page scan. The 24-page main PDF contains printed body pages 1-16, the foreword/introduction, the contents and blank/copyright pages; the separate 20-page scan carries the same number and year, with its body corresponding to main PDF 7-22 and the other PDF 5-20. The 20 pages alone do not show that 4 body pages are missing; no original PDF or OCR has been merged, overwritten or altered. The 16 body pages of the main scan have been read visually, and the other scan was used to cross-confirm the equations, Figures 4/5, Table 2 and the annex table; no claim is made that all scanned pixels or characters agree verbatim. This card is an original study compilation in 14 sections, retaining the equations and 3 factual tables; it does not publish scans or watermarks or copy the whole standard. Original Figures 1-9 and A.1-A.2 are not redrawn, the wording of the original is not fully transcribed clause by clause, and the interpretation of the figures and the installation relationships must return to the controlled standard; the absence of signs of missing pages is not the same as a verbatim collation against the official controlled copy. Technical review is pending and independent Hermes/Grok acceptance remains outstanding. Can it be operated as a current procedure? No, the edition has been withdrawn and this card is not the full text of the standard. Can it replace the new edition or confirm product suitability or a certificate? No. Does a high smoke point mean all performance requirements are met? No. The blank tolerances and drawing dependencies of the annex are retained as on this page; no OCR value has been altered or number guessed.

View public related products