Translating an ISO 17892-4:2016 soil particle-size distribution report is not simply a matter of changing “sand,” “silt,” and “clay” into another language. Engineers searching for soil gradation translation, sieve-analysis report terminology, hydrometer particle-size results, pipette sedimentation data, percent passing, and geotechnical laboratory certificates need a bilingual account of what was measured and how the distribution was obtained. The resulting words may influence earthworks specifications, drainage assessments, material classification, foundation investigations, and conversations between laboratories and designers.
A reliable ISO 17892-4 translation preserves the sample identifier, preparation history, particle-size boundaries, mass basis, sieve openings, sedimentation method, percentage calculations, reporting precision, and uncertainty or limitations. A soil grading curve is an estimate for the tested specimen, not an automatic description of an entire site or a promise that the material will behave in one particular way. One mistranslated sieve opening or one reversed “retained” and “passing” column can change the apparent material classification.
The official ISO catalogue lists ISO 17892-4:2016 as a published, confirmed standard for geotechnical soil particle-size distribution, with sieving and sedimentation methods, including hydrometer and pipette approaches. This article explains the language and interpretation of reports that cite it. It is educational guidance, not the copyrighted laboratory procedure, a certification service, a substitute for a geotechnical engineer, or permission to invent results that a laboratory did not issue.
1. Establish the question that the grading report answers
A particle-size distribution describes how much of a specimen belongs to different particle-size ranges. This sounds simple until someone uses the result to answer a design question that was never tested. A grading curve can support classification and comparison, but it does not by itself measure shear strength, settlement, mineralogy, permeability, plasticity, or whether a slope will remain stable. Those matters require additional evidence and professional interpretation. The translator should keep “particle-size distribution” as a measured characteristic rather than convert it into a judgement such as “good foundation soil.”
For a fictional borehole sample, the source report may show a broad mixture of coarse and fine material. The receiving engineer may eventually use the data alongside groundwater observations and other laboratory measurements. A precise translation labels the distribution and leaves the design conclusion to that engineer. This avoids two familiar problems: hiding technical distinctions from non-specialists and giving the numbers an authority they never possessed.
2. The full reference identifies the method, not the test date
“ISO 17892-4:2016” contains the standard number, part number, and edition year. It does not mean the sample was collected in 2016. A certificate issued in October 2026 can cite the 2016 method if that is what the laboratory actually used. The official ISO record describes this edition as published and confirmed, so changing its date to make the document seem modern would corrupt the reference. Preserve other references, such as a client specification, separately instead of blending their numbers into the method name.
Different dates have different jobs. A collection date describes when the soil left the ground; receipt identifies when the laboratory accepted it; preparation and analysis dates document the work; and the issue date records approval of the final report. In a multilingual package, each should retain its original label and applicable time zone. Where a revision was issued, the revised report identifier is distinct from the ISO edition.
3. Sample identity is the anchor for every figure
A geotechnical project may generate dozens of samples with similar descriptions. The borehole identifier, sampling depth, sample number, client reference, test portion, and laboratory accession number are therefore not decorative metadata. A single lost leading zero can attach a curve to the wrong location. Where a depth interval is recorded, preserve both endpoints and the unit. “Five metres” cannot replace “5.00–5.50 m” because it discards the interval that identifies the material.
Imagine two samples labelled BH-07/S12 and BH-07/S21. They share a borehole but not a depth or material history. If a target-language editor swaps their curve captions because the labels appear similar, a later discussion could treat a near-surface layer as a deeper stratum. Bilingual documents should preserve original identifiers exactly, with any translation of descriptive words placed beside, not inside, those codes.
4. Soil specimen is not the same thing as the whole ground layer
A laboratory typically works with a defined specimen from a collected sample. That specimen represents only the material submitted and prepared under the laboratory’s instructions. It might not capture boulders excluded during sampling, abrupt layering, local lenses, or the full geological variability between sampling points. Translating “specimen” as “entire soil layer” enlarges the claim. Likewise, a report’s “representative portion” is a sampling description, not a guarantee of perfect representation.
When communicating the result to a contractor, say that the distribution was measured for the identified laboratory specimen. If a geotechnical report later generalises across a stratum, the interpretive statement belongs in that separate report. Keeping the two statements apart allows a future reader to examine how site-wide conclusions were supported. It is an important form of accountability, not pedantry.
5. Distinguish grading from geotechnical classification
“Grading” commonly concerns size distribution. “Classification” combines the distribution with a named system and sometimes other attributes, such as plasticity. A translator must not assume that the same percentages generate the same group symbol under every national or contractual classification scheme. Size names and boundary conventions can differ. A laboratory may report a distribution without assigning a final class, and that absence must remain visible in the target version.
A useful bilingual legend identifies which classification system, if any, the source uses. If the data table labels a fraction “fine particles,” resist turning it into a locally defined “clay” fraction simply because the local language uses that everyday word for very small particles. Particle size and mineralogical clay are not interchangeable. The receiving engineer should be able to reclassify from documented data rather than from a translator’s unrecorded assumption.
6. Sieving and sedimentation measure different parts of the distribution
The ISO public description allows sieving, sedimentation, or a combination, depending on soil composition. Coarser material is commonly separated by sieves with defined openings. Fine material can require sedimentation analysis to estimate an equivalent particle size from settling behaviour. The two routes do not literally observe the same physical property in the same manner. A translated report should therefore distinguish “sieve analysis” from “hydrometer analysis” and “pipette analysis,” even where all three contribute to one final grading curve.
When two datasets are joined, the laboratory’s correction and combination choices matter. A translated graph must not imply that every plotted point came directly from a sieve. Labelling the entire curve “sieve result” can conceal the method used for the fine tail, where interpretation is often most sensitive. Instead use a combined-method label with the source’s stated boundaries and make any laboratory-specific method notes accessible.
7. Sieve opening is a dimension, not a sieve serial number
A sieve has a nominal aperture or opening size, often expressed in millimetres or micrometres. A mesh designation may be a separate naming convention; its meaning cannot always be inferred through a universal multiplication. The translator should preserve the actual opening value and unit when supplied. “0.063 mm” and “63 µm” describe the same length, while “63 mm” is a thousand times larger. That difference changes the location of a boundary on the curve.
If a report includes sieve numbers and aperture values, keep both columns rather than translating a number into a guessed opening. A plain-language gloss such as “particles passing the stated aperture” may help readers unfamiliar with sieve terminology. But the source designation remains the reference for equipment traceability. A cross-language reviewer should be able to match each table row to the laboratory equipment record.
8. Retained percentage and passing percentage point in opposite directions
The fraction retained on a sieve is material remaining above its openings after the test process. The fraction passing is the material that moves through. Cumulative retained and cumulative passing use related but opposite perspectives. A common calculation is percent passing equals 100 minus cumulative percent retained, after accounting for the report’s chosen mass basis and rounding. In translation, the original column heading is critical because identical-looking numbers can describe different quantities.
For a simple illustration, suppose 28% of the original test mass is cumulatively retained at a specified aperture. The corresponding passing percentage on a consistent basis would be 72%. If a translator copies 28 into a column labelled “percent finer,” the curve is materially changed. That error may go unnoticed because each individual number still falls between zero and one hundred. The safeguard is to check the relationships, not just the spelling.
9. Mass basis is needed to understand every percentage
Percentages depend on the denominator. A laboratory can record total dry mass, a prepared test portion, a coarse fraction, or a subdivided fraction used for a particular analysis. The report should make clear which basis underlies the numbers. If a sieve table and a sedimentation table use different intermediate bases, someone must apply the laboratory’s documented combination method before comparing them. The translator should never conceal an apparent difference by rewriting values.
Take a fictional test in which a 400 g dry portion contains 100 g retained above a chosen sieve. On that simple basis, the retained fraction is 25%. If the table instead reports 100 g out of a separately prepared 250 g coarse portion, the same mass corresponds to 40%. Neither percentage is meaningful without knowing the denominator. Keep phrases such as “of total dry mass” and “of tested fraction” where they occur.
10. Drying, washing and preparation affect what can be counted
Soil particles may be bound into aggregates or coated with finer material. Preparation can therefore change what a sieve separates. A source report may note drying, soaking, washing, disaggregation, or removal of visible foreign matter. Each term has a physical meaning and should not be simplified into “sample cleaned” unless the source actually says that. Washing fines through a sieve, for example, is not equivalent to washing all soil away as waste.
If a client is comparing two curves from different laboratories, the preparation notes help explain possible discrepancies. The translated report should describe those notes neutrally, not resolve the disagreement without the raw records. Where preparation details are absent, mark them as not stated. Filling the gap with an assumed routine can make a method comparison look more controlled than it was.
11. Hydrometer readings describe suspension behaviour
A hydrometer is an instrument that can be used in a sedimentation test. Its reading relates to the condition of the soil suspension as particles settle, with interpretation depending on calibration and procedural corrections. A translated heading should not call it a “moisture meter,” “water content reading,” or “particle counter.” Those labels suggest entirely different measurements. The laboratory converts observations into reported size-distribution information under its method.
Temperature, elapsed time, density-related corrections, and preparation conditions may appear in the source worksheet. Keep each variable with its unit and time reference. If the report cites a calculated equivalent diameter, it is an inference from sedimentation behaviour, not a direct measurement of the longest dimension of every irregular grain. This distinction matters when someone tries to compare the result with microscope images or laser diffraction measurements.
12. Pipette analysis is not merely another name for hydrometer analysis
The pipette method is a sedimentation approach in which the analysis of withdrawn suspension portions supports the size distribution estimate. Hydrometer and pipette approaches share a general settling principle but differ in measurement workflow and source records. Translating one as the other can misstate the laboratory’s method. If a certificate reports pipette analysis, preserve that specific wording and any sampling depth or time details actually documented.
People often shorten method names in conversations, particularly when translating a long laboratory report for a purchasing team. In technical tables, however, abbreviations require clear definitions. A line headed “sedimentation” may be acceptable as a broad family label, but method-specific records need the exact instrument or procedure. Avoid adding steps from an online description when the source gives no indication that those steps occurred.
13. Equivalent particle diameter has a model behind it
Fine-particle sedimentation calculations generally describe an equivalent size under a physical model rather than a three-dimensional scan of individual soil grains. The model uses settling behaviour and assumptions about the medium and particles. Non-spherical grains, flocculation, and variations in particle density can influence interpretation. A strong translation retains terms such as “equivalent diameter” instead of promising a directly observed geometrical dimension.
When a report compares a hydrometer-derived fine fraction with a sieved coarse fraction, the results may fit onto one curve because the methods are used in a defined laboratory framework. That continuity should not be interpreted as proof that every individual fine particle has the same shape or density. A future reviewer may need to understand these assumptions to evaluate unusual samples or differences between methods.
14. Soil density is not the same as bulk density
Particle density concerns the solid material itself. Bulk density concerns a volume containing solids and voids, with dry or wet variants depending on context. Translating one as the other can damage sedimentation calculations and site assessments. The method worksheet might use a particle-density assumption or a measured value, while another project document lists field dry density. These are separate numbers with separate uses, even if both display mass-per-volume units.
A bilingual glossary should bind each density term to its specimen and condition. If a laboratory uses a default value rather than a measured one, retain the qualification. The translator need not determine whether the assumption is suitable; that is a technical review question. What matters is that a reader can see which quantity was used and whether it came from measurement or an assumption.
15. Flocculation and dispersion should not be treated as decorative language
Fine particles can form clusters or remain separately suspended depending on the material and preparation. A report may refer to dispersion, dispersant, or flocculation as part of explaining the test condition. Those terms are not interchangeable with ordinary “mixing.” A mixture may be stirred without being adequately dispersed for the intended analytical meaning, and a visibly clear supernatant is not sufficient evidence of a particular fine-particle distribution.
When a source mentions unusual aggregation or incomplete dispersion, translating it away may make the curve appear more reliable than the analyst believed. Preserve cautionary observations and place them near the results they qualify. If the translator cannot resolve a chemical term, use a bilingual term note rather than an approximate colloquial substitute. The best technical wording helps the receiver understand what was observed and what remains uncertain.
16. Temperature and time belong to their measurements
Hydrometer or pipette worksheets may include elapsed settlement time, suspension temperature, sampling time, and sometimes corrections. A record showing “20 °C” is not an instruction to translate a temperature into room-temperature prose. Likewise, “two minutes after agitation” is an elapsed time, not the time of day. Numerical fields need attached labels so that calendar times, durations, and instrument observations cannot be confused.
A practical review compares the original table row by row. Does each time retain its original unit? Are decimal commas interpreted as decimal separators rather than thousands separators? Do negative signs, correction symbols, and superscripts survive? These small questions are often more important than fluency of the surrounding prose because they determine whether a second analyst can reproduce the meaning of the worksheet.
17. Soil fractions are defined by named size boundaries
The words gravel, sand, silt, and clay are familiar, but their size boundaries can depend on the reporting convention or classification system. Translators should preserve boundaries stated in the source instead of substituting those from a different country’s preferred scheme. The particle-size distribution is the data; fraction names are labels applied according to a defined system. Replacing the label without its boundary can produce an internally inconsistent report.
For example, a report may divide a curve at a stated aperture and assign percentages above and below it. A target-language editor might recognise the local word “fine sand” but not its formal size range. The safe solution is to retain the numerical boundary and, where needed, add a note identifying the source convention. The receiving engineer can then compare classifications with full awareness of their different definitions.
18. Percent finer is not percent clay mineral
A sedimentation curve may show a fraction smaller than a specified nominal diameter. That result is about size under the chosen method. It does not establish the mineral composition of the material. Clay minerals can contribute to soil behaviour, but size classification and mineral identification are distinct questions. Translating a fine-size fraction as a mineralogical diagnosis can mislead readers of foundation, environmental, or agricultural reports.
If the document needs to communicate both, use separate headings such as “particle-size fraction” and “mineral identification.” Where the source contains only the former, do not add the latter. A bilingual footnote can explain that “clay-sized” refers to a size class. The note is a clarification, not a new laboratory result, and should be marked as such in sensitive engineering documentation.
19. D10, D30 and D60 have precise graphical meanings
A grading curve may report characteristic diameters associated with particular cumulative percentages passing. Under a usual percent-finer convention, D10 denotes the particle diameter at 10% passing, D30 at 30%, and D60 at 60%. These are positions on a cumulative distribution, not three separate tests. Preserve the D notation and the curve convention, including any interpolation method stated by the laboratory.
A common mistranslation makes D60 “60% larger than average” or “average of sixty particles.” Neither describes the value. If the report calculates coefficients from characteristic diameters, its meaning depends on the correct reading of the curve. A reader should be able to locate the corresponding percentage on the axis and trace to the reported size without needing to trust a translated marketing sentence.
20. Uniformity coefficient and curvature coefficient require their inputs
Some reports derive the uniformity coefficient from D60 divided by D10 and a curvature coefficient from a combination of D30, D10, and D60. These are dimensionless quantities under their customary definitions, provided the same length units are used. The translator should preserve the exact formula or named calculation shown in the report rather than replacing it with words such as “uniform quality” or “good curvature,” which sound evaluative.
If D10 is not established reliably, a coefficient based on D10 may also be unavailable or qualified. Do not calculate an attractive value from a guess or change “not determined” into zero. An unavailable characteristic diameter is a data condition, whereas a numerical zero can make the formula nonsensical. The responsible translation keeps the calculation status as visible as the result.
21. Semi-log plots need the correct axis labels
Soil grading curves often plot particle size on a logarithmic horizontal axis and cumulative percentage on a linear vertical axis. A logarithmic axis has nonuniform-looking numeric spacing, but each tick corresponds to a specific size. If someone rebuilds the graph in a target-language report using an ordinary linear axis, the shape can change markedly even with correct labels. Translations should preserve the graphic or recreate it only from verified numerical data.
The direction of the percentage axis matters too. A curve labelled “percentage passing” cannot simply be relabelled “percentage retained” while keeping the same plotted line. Check legends, units, axis direction, and any series that distinguish tests or samples. When a graph image is too low-resolution to recover, request original data instead of reading approximate values from pixels and presenting them as exact.
22. Balance checks reveal impossible translations
The total retained mass, recovered fines, and reported percentages should follow the laboratory’s stated accounting rules. A table with passing percentages that rise while sieve aperture decreases may be suspicious under a conventional cumulative grading arrangement. However, a translator should not quietly modify the original data to repair a perceived inconsistency. The right response is to mark the issue for laboratory verification while preserving the source.
A simple quality-control worksheet can recompute selected percentages from raw masses. It should also record rounding and any mass loss reported during washing or transfer. This is a verification exercise, not permission to issue new certified figures. If numbers fail to reconcile, the translation may reproduce them with a clear query so the laboratory can decide whether a formal correction is necessary.
23. Censored, missing and below-range values are distinct
Some laboratories may omit a fine tail because the test did not extend that far, mark a result as below reporting range, or label a measurement invalid. These states are not synonymous. “Not measured” does not mean the soil contains no fine particles. “Below limit” does not mean an exact zero. “Not applicable” may reflect the report format rather than a physical property. Translate the reason where it is supplied.
The same caution applies to blanks and dashes in a table. A dash might represent no observation, unavailable value, or formatting convention. Ask what it means before turning it into a quantity. Preserving uncertainty often feels less polished than filling every cell, but the blank is part of the evidence. It tells the receiving analyst where the record stops.
24. The laboratory’s conclusion is not a design approval
A geotechnical laboratory may provide descriptive test results without deciding whether a soil can be used in an embankment, filter layer, road base, or foundation. Suitability decisions depend on additional specifications and environmental conditions. Therefore a translation should not turn “test completed” into “material approved.” Even where an acceptance note is present, identify the criterion and the authority applying it.
For a hypothetical fill specification, a contractor could compare the curve with an allowed grading envelope. The translation may show the envelope and measured curve, but if the source does not give an acceptance decision, the translator should not manufacture one. The engineer or designated approving party remains responsible for interpreting compliance. This boundary helps international teams work together without transferring technical authority to a language editor.
25. Scenario: a reversed column changes the soil story
Consider a fictitious report with 82% passing a particular sieve and 18% cumulatively retained. In the target-language table, a well-intentioned editor switches the headings while leaving the numerals untouched. The new document appears to describe predominantly coarser material at that aperture. A contractor working from the translation may be led towards the wrong material comparison despite the source laboratory having provided correct data.
The repair is not simply to replace a word. The reviewer should verify the arithmetic, restore the proper axis labels in any associated graph, check whether the executive summary repeats the wrong inference, and record the corrected version. This illustrates why a bilingual numerical report requires systematic field checking after normal language editing. Fluent sentences alone cannot secure technical meaning.
26. Scenario: the fine fraction is reported as mineral clay
A fictional sedimentation report identifies a percentage of grains below a stated equivalent diameter. The translation says that the soil contains the same percentage of clay minerals. A designer could understandably read this as a mineralogical composition result. Yet sedimentation did not establish which minerals are present. The statement adds an untested claim and may cause incorrect expectations about swelling or chemical behaviour.
A more accurate version names the size fraction and leaves mineralogical questions open. If there is a separate X-ray diffraction or other mineral study, cite its report identifier rather than merging both into one claim. This distinction allows useful cross-disciplinary synthesis while respecting the evidence collected by each method. Translation becomes a bridge between datasets, not an invention of one.
27. Scenario: metres and millimetres are exchanged
A geotechnical schedule lists sample depth in metres and sieve openings in millimetres. During formatting, a column header is duplicated and the sieve values acquire the depth unit. The numbers may still look plausible to a hurried reader because both tables contain decimals. This sort of unit error is dangerous precisely because no foreign-language word looks obviously wrong; the document simply describes a different physical scale.
The prevention is to audit units as pairs with their values. Preserve “0.063 mm” together, rather than storing the number in one sheet and the unit in another where it can drift. Cross-check every figure caption against the axis units. When conversions are needed, retain the source quantity and an independently checked converted value so the trail remains inspectable.
28. Scenario: two boreholes are combined in one figure
A project summary contains two grading curves for neighbouring boreholes and a legend that uses similar shades. After localisation, the legend is reordered alphabetically, but the plotted series keep their original colours. The target-language report now reverses which borehole has the finer soil. A discussion of drainage or excavation behaviour may proceed from the wrong curve, despite all the numbers surviving.
The safest response is to preserve series identity with sample codes and line styles, not colour alone. Verify each series against the raw data and caption. If a graph is rebuilt for accessibility, use a table or alternative text that names the correct sample and describes the plotted quantities. Translation quality includes graphical relationships whenever those relationships communicate the test result.
29. Scenario: a report shows only the coarse tail
A hypothetical contractor receives a sieve-only result for a soil with substantial fines. The purchasing summary describes it as the complete gradation. The summary may omit that the fine tail was not resolved by sedimentation. Without that qualification, reviewers could think the small-particle fraction was measured more precisely than it was. The translation must preserve the scope of the actual laboratory work.
A sound wording is “sieve-analysis results for the stated test portion,” followed by any source note on the unresolved fines. If a combined distribution is required for a particular specification, the technical reviewer can request additional analysis. The translator should not interpolate a missing tail from an unrelated sample simply to produce a visually complete plot.
30. Build a bilingual data dictionary before editing the prose
The most useful glossary for a soil laboratory document is short and anchored to real fields: sample, specimen, dry mass, aperture, cumulative retained, percent passing, hydrometer, pipette, sedimentation, equivalent diameter, grading curve, and test limitation. For each entry, store the source label, approved target equivalent, units where applicable, and one authentic example from the document. This reduces inconsistent translations across tables, captions, and certificates.
Do not try to turn this field glossary into a competing general vocabulary hub. It is a document-level tool for maintaining meaning in one technical task. Broader language-learning concepts can be learned elsewhere. During final review, use the glossary to search for competing terms and verify whether differences are intentional. This simple step catches a surprising number of errors in large multilingual report bundles.
31. A practical two-person translation and engineering review
One person can translate the prose, but a second reviewer should check the technical relationships. The language reviewer examines terminology, grammar, qualifiers, and whether descriptions remain readable. The technical reviewer independently checks units, curve labels, sample IDs, mass bases, and calculations against the source. Both reviewers should log queries that require the issuing laboratory rather than improvise answers. Separate roles help prevent a fluent but incorrect document.
A useful acceptance checklist includes every source table, every figure, every paragraph containing “shall,” “may,” or “not,” and every statement claiming suitability. Verify that appendices and footnotes have not been lost during layout. Record what version was reviewed, who approved the target language, and which source was authoritative. A translation can be concise and accessible without leaving its evidence untraceable.
32. Explain results to a non-specialist without reducing accuracy
A parent, buyer, project manager, or contractor may not need every method detail in a first conversation. A helpful explanation is that the lab estimated how the specimen’s solid material was distributed across different grain sizes. The curve shows the proportion finer than selected sizes, with methods and preparation conditions determining how those estimates were obtained. This explanation is accessible, yet it does not promise strength, suitability, or site-wide uniformity.
The best summary then points readers to the sample identity, grading plot, method notes, and any stated limitations. If someone asks “Is the ground safe?” the correct answer cannot be obtained from particle-size information alone. Make that boundary explicit rather than treating the question as naive. This kind of respectful explanation supports better decisions because it tells the reader both what the report knows and what further evidence would be needed.
33. Review exercises for translators and students
Exercise one: a row gives 26% cumulative retained at an aperture. Explain why the matching percent passing is 74% on the same mass basis, and identify which words in a bilingual table would distinguish those two numbers. Exercise two: a plotted D10 is unavailable because the curve does not reach ten percent passing. Explain why inserting zero would damage a uniformity-coefficient calculation rather than help complete the table.
Exercise three: two reports refer to the same sample code but have different preparation methods. Explain what evidence would be needed before calling their curves contradictory. Exercise four: a target-language summary calls a size fraction “clay mineral.” Rewrite it as an explicitly size-based observation. The purpose of these exercises is not to memorise a whole testing standard; it is to practise keeping observations, calculations, assumptions, and engineering decisions in separate, clearly named places.
Frequently asked questions
Does a soil grading report tell me whether a building foundation is safe?
No. It describes particle-size distribution for a tested specimen. Foundation safety depends on many additional measurements, geological information, loading conditions, design requirements, and professional assessment.
Can “percent retained” be translated as “percent passing”?
No. They have opposite cumulative interpretations. Confirm the underlying mass basis and the numerical relationship before editing either heading.
Is the hydrometer method a water-content test?
No. In this setting it is a sedimentation-based way to estimate fine-particle size distribution, not a measurement of the soil’s natural water content.
Do clay-sized particles prove the presence of clay minerals?
No. Size classification and mineral identification are separate kinds of evidence.
Can I replace a missing D10 with zero?
No. A missing or unresolvable characteristic diameter remains unavailable unless the issuing laboratory supplies an authorised determination.
Can I describe a sieve-only analysis as a complete combined grading test?
Only if the source supports that description. Preserve exactly which method was used and whether any size range remains unresolved.
Must the standard edition year match the laboratory’s issue date?
No. The standard edition, test date, and report issue date are separate facts.
What should a translator do when the table and graph disagree?
Preserve the source records, document the discrepancy, and request clarification from the issuing laboratory. Do not silently invent corrected values.
Sources and educational links
The official ISO 17892-4:2016 catalogue documents the standard’s scope, publication, and confirmation status. The established Master Art of Translation provides the broader translation architecture, while the protected Vocabulary Learning Hub and How English Works support vocabulary learning and nuanced English interpretation. This page is deliberately limited to the translation of soil particle-size test evidence.
The lasting test of a good translation is whether an independent geotechnical reader can identify the specimen, reconstruct the meaning of every percentage and curve, see which methods were used, and recognise what the laboratory did not measure. Preserving those relationships is far more valuable than polishing a report until every technical distinction disappears.