Breaking Strength Test Data Processing: Statistical Evaluation, Standards, and Reporting Practices

Breaking strength test data processing transforms raw tensile force and elongation measurements into statistically reliable, standards-compliant results for textiles, geotextiles, and industrial materials. This technical guide explains the complete data-processing workflow—from specimen validity screening and outlier detection to the calculation of mean, standard deviation, coefficient of variation, confidence intervals, and rounding rules—based on ISO, ASTM, EN, and GB standards. It serves as a practical reference for quality engineers and technicians working in accredited third-party laboratories and in-house testing facilities.

Technical Overview

Breaking strength testing, also referred to as breaking force or tensile testing, determines the maximum force a specimen can withstand before rupture, together with its elongation at that force. However, the measurement itself is only half of the process: breaking strength test data processing is the disciplined statistical treatment of raw force–elongation records that converts them into defensible, comparable, and reportable values. Proper data processing ensures that results from different laboratories, machines, and batches can be compared meaningfully and that pass/fail decisions against product specifications are technically sound.

The principal standards governing breaking force determination and its data treatment include:

• ISO 13934-1 / EN ISO 13934-1 / GB/T 3923.1 — determination of maximum force and elongation at maximum force using the strip method
• ISO 13934-2 / EN ISO 13934-2 / GB/T 3923.2 — determination of maximum force using the grab method
• ASTM D5034 — breaking strength and elongation of textile fabrics (grab method)
• ASTM D5035 — breaking strength and elongation of textile fabrics (strip method)
• ISO 13935-1 / ISO 13935-2 / ASTM D1683 — tensile properties of sewn seams and seam slippage
• ISO 9073-3 — tensile strength of nonwoven fabrics
• ISO 10319 / ASTM D4595 / GB/T 15788 — wide-width tensile testing of geosynthetics and geotextiles
• ISO 5725 — accuracy (trueness and precision) of measurement methods and results
• GB/T 4883 — statistical treatment of outlying observations in normal samples
• GB/T 8170 / ASTM E29 — rules for rounding off numerical values and significant figures

A typical data-processing sequence comprises raw data capture, validity screening of individual specimens, unit harmonization, statistical computation, outlier detection, rounding of the final reported value, and evaluation of measurement uncertainty. Each step is examined below.

Testing Samples

Woven and Knitted Fabrics

Apparel, home textile, and technical woven fabrics are the most common samples. Specimens are cut in both the warp (lengthwise) and weft (crosswise) directions, taken no closer than 50 mm to the fabric edge and oriented so that no two specimens share the same set of yarns. Because yarn-level variation directly influences the coefficient of variation, representative sampling is the first step of sound data processing.

Nonwoven Fabrics

Spunbond, meltblown, needle-punched, and hydroentangled materials are tested to ISO 9073-3 or equivalent strip procedures. Nonwovens typically exhibit higher scatter than woven fabrics, so validity screening and statistical verification of the mean are especially important.

Geotextiles and Geosynthetics

Wide-width strip specimens (commonly 200 mm effective width) are tested to ISO 10319, ASTM D4595, or GB/T 15788 for separation, filtration, and reinforcement applications. Results are commonly expressed as breaking strength per unit width in kN/m, which requires an additional normalization step during data processing.

Narrow Fabrics, Webbing, and Straps

Webbing, tapes, seat belts, lifting slings, and hook-and-loop fasteners are tested at full width. Data processing must account for the actual specimen width when comparing against width-normalized specifications.

Sewn Seams

Seam specimens prepared with defined stitch type, stitch density, and seam allowance are tested to ISO 13935-1, ISO 13935-2, or ASTM D1683 to obtain seam breaking force and seam efficiency, expressed as the ratio of seam strength to fabric strength.

Yarns, Threads, and Technical Cords

For linear materials, breaking tenacity in cN/tex is derived by dividing the breaking force by the linear density, an essential normalization step in the data-processing routine.

Testing Items

The parameters generated by breaking strength testing, and subsequently processed and reported, include:

• Breaking force (maximum force) — the highest force recorded before rupture, expressed in newtons (N), or in kN/m when normalized per unit width
• Breaking elongation — extension at maximum force, expressed as a percentage of the initial gauge length
• Breaking tenacity — force normalized by linear density, expressed in cN/tex or N/tex
• Work to rupture — the area under the force–elongation curve, expressed in joules (J)
• Seam breaking force and seam efficiency — for sewn seam evaluations
• Statistical descriptors — arithmetic mean, sample standard deviation, coefficient of variation (CV%), minimum and maximum individual values, 95% confidence limits, and the number of valid and invalid specimens

Reported results must clearly state the number of valid determinations, the test direction, the method standard, the conditioning atmosphere, and any deviations, since these contextual data are required for the statistical interpretation of the mean.

Testing Methods

Strip Test Method

Per ISO 13934-1, GB/T 3923.1, and ASTM D5035, specimens of full fabric width—raveled or cut to a defined effective width (typically 50 mm)—are gripped across their full width and extended to rupture at a crosshead speed assigned from the elongation class specified by the standard. A constant gauge length (commonly 200 mm under ISO procedures and 75 mm under ASTM procedures) and a defined pretension are applied so that elongation measurements start from a reproducible zero point. At least five valid specimens per principal direction are normally required.

Grab Test Method

Per ISO 13934-2, GB/T 3923.2, and ASTM D5034, a wider specimen (commonly 100 mm) is clamped only at its central region so that the jaws grip a defined central strip. The grab method is preferred for high-strength, coated, felted, heavily napped, or glazed fabrics where raveling is impractical, and it better reflects the contribution of yarn interaction to fabric strength.

Data Validation and Specimen Screening

Before any statistics are computed, individual force–elongation curves are reviewed. An individual result is discarded and a replacement specimen tested when:

• the specimen slips visibly in the jaws or slips partially before rupture;
• rupture occurs within 5 mm of the jaw line and the recorded force is below the observed minimum of the valid set;
• the specimen was damaged during cutting, mounting, or raveling;
• the curve shows instrumental artifacts such as force spikes or a zero shift.

Screening is documented, and testing continues until the minimum number of valid results required by the method is obtained. Discarded results are recorded in the raw data but excluded from the statistical calculation.

Statistical Processing

For the set of n valid individual breaking force values, the arithmetic mean is calculated as the sum of all values divided by n. The sample standard deviation is calculated using n − 1 in the denominator, and the coefficient of variation is obtained as the standard deviation divided by the mean, multiplied by 100%. A 95% confidence interval for the true mean is constructed as the mean plus or minus the Student's t value for n − 1 degrees of freedom multiplied by the standard deviation divided by the square root of n. The same computations are applied to elongation values. Modern testing systems perform these calculations automatically, but the analyst remains responsible for verifying that only valid results entered the calculation.

Outlier Detection

When one individual result appears inconsistent with the rest of the data set, a statistical outlier test is applied rather than subjective judgment. The Grubbs test, as specified in GB/T 4883, computes the ratio of the absolute difference between the suspect value and the mean to the sample standard deviation; if this statistic exceeds the critical value for the sample size at the 5% significance level (approximately 1.715 for a two-sided test with five results), the suspect value is flagged. The Dixon Q test is an accepted alternative for very small samples. A flagged value should be removed only after investigation confirms a technical cause—such as a jaw break or a fabrication defect—and the justification must be recorded. Values must never be deleted solely to reduce the coefficient of variation.

Rounding and Expression of Results

All intermediate calculations are carried out with full precision; rounding is applied only once, to the final reported value, following GB/T 8170 or ASTM E29. Common practice for fabric testing is to report the mean breaking force to the nearest 1 N (or as otherwise specified by the governing method) and mean breaking elongation to the precision stated in the standard. When unit conversion is required, 1 kgf equals 9.807 N and 1 lbf equals 4.448 N; conversions are performed before rounding. Width-normalized results (kN/m) and tenacity values (cN/tex) are derived using the measured specimen width or certified linear density, each with its own significant-digit discipline.

Measurement Uncertainty Evaluation

A complete data-processing report includes the expanded measurement uncertainty. Principal uncertainty components include force indication error of the load cell (traceable to ISO 7500-1 calibration), gauge-length and crosshead-speed settings, specimen width and linear density measurement, clamping and jaw-slip effects, conditioning atmosphere, sampling heterogeneity, and operator repeatability. Components are combined as the root sum of squares and multiplied by a coverage factor of k = 2 for approximately 95% confidence. Depending on material homogeneity, the relative expanded uncertainty of fabric breaking force is typically reported in the range of a few percent, and it must be considered when results are compared against specification limits.

Applications

Reliable breaking strength data processing supports a wide range of industrial and regulatory activities:

• Apparel and home textile quality control — verifying fabric strength against buyer specifications and national product standards before shipment
• Technical and industrial textiles — qualifying filter fabrics, conveyor belt fabrics, protective clothing materials, and coated substrates
• Geosynthetics engineering — confirming wide-width tensile strength for geotextiles used in road, railway, landfill, and hydraulic construction projects
• Medical textile validation — supporting strength verification of bandages, surgical gowns, and single-use protective materials
• Automotive and transport interiors — evaluating seat belt webbing, trim fabrics, and airbag materials
• Research and development — comparing yarn structures, fabric constructions, finishes, and weaving parameters using statistically significant data sets
• Certification and compliance — providing defensible test reports for product certification, tender submissions, and dispute resolution

FAQ

How many valid breaking strength results are required for a report?

Most strip and grab methods require a minimum of five valid specimens for each test direction (warp and weft, or machine and cross direction). If specimens are invalidated by jaw breaks or slippage, additional specimens are tested until the minimum number of valid results is reached, and the invalidation events are documented in the raw data.

When must an individual breaking force result be discarded?

A result is discarded when the specimen slips in the jaws, when rupture occurs within 5 mm of the jaw line and the force is below the minimum of the valid results, or when the specimen was damaged during preparation. Discarding is a validity decision based on physical evidence, not a statistical convenience.

What is an acceptable coefficient of variation?

There is no universal limit; acceptance criteria are set by the product specification or contract. As a practical guide, well-prepared woven fabric tests often yield a CV of 10% or less, while nonwovens and heterogeneous materials may naturally show higher scatter. A high CV signals the need to review sampling, specimen preparation, and clamping before questioning the material itself.

Should statistical outliers always be removed from the data set?

No. An outlier identified by the Grubbs or Dixon test should be removed only after a technical root cause is confirmed and recorded. If no cause is found, the value is retained, because it may represent genuine material variability.

What is the difference between breaking force and breaking strength?

Breaking force is the absolute maximum force in newtons sustained by the specimen as tested. Breaking strength is a normalized quantity—force per unit width (kN/m) or tenacity (cN/tex)—that allows comparison between materials of different widths or linear densities. The normalization step is performed during data processing before rounding.

How are final results rounded and reported?

Intermediate values are kept at full precision, and rounding is applied once to the final reported mean, following GB/T 8170 or ASTM E29 and the precision stated in the governing method. The report must state the standard used, the test direction, the number of valid results, the mean, the coefficient of variation, and any deviations that could affect interpretation.