Cement Mortar Strength Comparison Testing: Principles, Procedures, and Statistical Evaluation

Cement mortar strength comparison testing is a systematic interlaboratory or intra-laboratory experiment designed to verify the accuracy, repeatability, and reproducibility of compressive and flexural strength determinations for cement. Conducted in accordance with ISO 679, EN 196-1, and GB/T 17671, the experiment compares strength results obtained from identically prepared mortar prisms to evaluate laboratory competence and testing quality. This article presents a complete technical guide covering sample requirements, testing items, methodologies, statistical evaluation criteria, and practical applications.

Technical Overview

Cement mortar strength comparison testing refers to a structured comparative experiment in which the compressive and flexural strength of cement mortar specimens is determined under strictly standardized conditions, and the results are compared against reference values, parallel specimens, or the outcomes reported by other laboratories. It is a fundamental quality assurance tool used to confirm that strength measurements are accurate, consistent, and technically valid.

The core experimental procedure follows ISO 679, EN 196-1, and the Chinese national standard GB/T 17671 (Test Method of Cement Mortar Strength — ISO Method). Mortar prisms measuring 40 mm × 40 mm × 160 mm are prepared from a fixed proportion of cement, ISO standard sand, and water, cured under controlled temperature and humidity conditions, and loaded to failure at defined test ages. When the comparison is organized between different laboratories, the design and statistical evaluation of the experiment follow ISO/IEC 17043 (conformity assessment — proficiency testing) and ISO 13528 (statistical methods for proficiency testing), while the technical competence of participating laboratories is governed by ISO/IEC 17025.

The primary objectives of a cement mortar strength comparison experiment include:

• Verifying the consistency of strength results among laboratories, operators, or pieces of equipment within the same laboratory.
• Identifying systematic deviations caused by instrumentation, environmental conditions, or procedural errors.
• Providing objective evidence of technical competence for accreditation and internal quality control purposes.
• Validating that a laboratory reproduces the precision limits for repeatability and reproducibility stated in the test standards.

Testing Samples

Cement Sample

The cement used in the comparison experiment should originate from a single, homogeneous production batch, typically a common Portland cement conforming to GB 175 or EN 197-1. The sample is thoroughly mixed, divided into identical sealed portions using a sample splitter or quartering method, and distributed to all participating parties. Airtight packaging and protection from moisture are essential, because premature hydration or carbonation can significantly distort strength results. A reference cement with certified characteristic values may also be used as the assigned-value material.

ISO Standard Sand

The mortar aggregate must be ISO (CEN) standard sand, a naturally graded silica sand with particle sizes between approximately 0.08 mm and 2.0 mm, supplied in pre-weighed bags of 1350 g. Each bag corresponds to one mortar batch for three prisms. Using standard sand from the same production lot across all participants eliminates aggregate variability as a source of deviation, which is a prerequisite for a meaningful comparison.

Mixing Water

Each batch requires 225 g of potable or distilled water at a constant water-to-cement ratio of 0.50 by mass. The water temperature should be 20 °C ± 2 °C at the time of mixing. Distilled water is preferred in comparison experiments to remove any influence of dissolved salts or impurities.

Mortar Prism Specimens

The test specimens are prisms of 40 mm × 40 mm × 160 mm. For each test age, one batch of mortar produces three prisms, yielding three flexural strength values and six compressive strength values (two half-prisms from each prism). A typical full comparison covering 3-day and 28-day ages therefore requires at least six prisms per participant, with additional prisms prepared as spares.

Testing Items

The comparison experiment evaluates the following strength parameters and statistical indicators:

• Flexural strength at 3 days and 28 days, expressed in MPa.
• Compressive strength at 3 days and 28 days, expressed in MPa.
• Optional early or intermediate ages, such as 24 hours or 7 days, when specified by the comparison protocol.
• Arithmetic means or medians of the individual strength values, calculated according to the result-processing rules of the applicable standard.
• Standard deviation and coefficient of variation within each laboratory (repeatability) and across laboratories (reproducibility).
• Statistical comparison metrics, including z-scores, En numbers, and deviations from the assigned value, used to judge the acceptability of each participant's results.

For routine comparison purposes, the 28-day compressive strength is the most heavily weighted indicator, because it represents the principal strength classification parameter for commercial cements.

Testing Methods

Specimen Preparation

Each mortar batch consists of 450 g of cement, 1350 g of ISO standard sand, and 225 g of water. The materials are mixed in a planetary-type mixer following a timed program: water is placed in the bowl, cement is added and mixed at low speed, sand is introduced steadily during continued low-speed mixing, followed by high-speed mixing, a 90-second pause during which mortar adhering to the bowl and blade is scraped into the mix, and a final high-speed mixing stage. The complete mixing cycle lasts approximately four minutes.

The fresh mortar is cast into 40 mm × 40 mm × 160 mm steel molds in two approximately equal layers, each layer compacted on a jolting apparatus delivering 60 shocks in 60 seconds. The surface is struck off level, and the molds are immediately identified with laboratory codes and specimen numbers to preserve traceability throughout the comparison.

Curing Conditions

Filled molds are stored in a moist cabinet or curing room at 20 °C ± 1 °C with relative humidity of at least 90%. Specimens are demolded between 20 and 24 hours after casting, or earlier if 24-hour strength is required, and then submerged in a water bath at 20 °C ± 1 °C until the test age. Curing water must be changed at defined intervals and must not become acidic or contaminated. Strict adherence to temperature and humidity tolerances is critical, because a deviation of even 1-2 °C during curing can shift 28-day strength by a measurable margin.

Flexural Strength Determination

At the specified test age (72 h ± 45 min for 3-day strength; 28 d ± 8 h for 28-day strength), each prism is removed from the water, wiped to a saturated surface-dry condition, and tested within the time limits defined by the standard. Flexural strength is measured by three-point bending on a compression-flexure machine with a support span of 100 mm, applying a loading rate of 50 N/s ± 10 N/s until fracture. The flexural strength is calculated as the maximum load multiplied by the support span and bending factor, divided by the cube of the prism width.

Compressive Strength Determination

The two half-prisms recovered from each flexural failure are immediately tested in compression using 40 mm × 40 mm auxiliary platens, with a loading rate of 2400 N/s ± 200 N/s. The compressive strength equals the maximum force divided by the loaded cross-sectional area of 1600 mm². Result processing follows the applicable standard: ISO 679 and EN 196-1 calculate the flexural strength as the median of three values and the compressive strength as the median of six values, while GB/T 17671 applies arithmetic means with defined outlier rejection limits.

Statistical Evaluation of Comparison Results

After all participants report their results, the coordinating body evaluates the data using the statistical framework of ISO/IEC 17043 and ISO 13528:

• The assigned value is established from the certified value of a reference cement or from the robust consensus mean of all participant results.
• Each laboratory receives a z-score, calculated as the difference between its result and the assigned value divided by the standard deviation for proficiency assessment.
• A result is judged satisfactory when the absolute z-score is ≤ 2, questionable when between 2 and 3, and unsatisfactory when ≥ 3.
• For paired comparisons against a reference laboratory, the En number may be used, with results considered acceptable when the absolute En value is ≤ 1.
• Within-laboratory and between-laboratory precision are compared against the repeatability and reproducibility limits published in ISO 679, EN 196-1, and GB/T 17671.

Laboratories with unsatisfactory results are expected to investigate root causes, implement corrective actions, and participate in a follow-up round to demonstrate restored performance.

Applications

Cement mortar strength comparison experiments serve a wide range of technical and regulatory purposes:

• Proficiency testing programs and interlaboratory comparisons that support the accreditation of laboratories and accredited third-party laboratories under ISO/IEC 17025.
• Internal quality control in cement plants for monitoring mill performance, clinker quality, gypsum dosage, and strength-grade conformity (for example, 42.5 or 52.5 grades).
• Verification of newly installed or repaired testing equipment, including planetary mixers, jolting tables, flexural fixtures, and compression machines, before routine use.
• Operator training and qualification, ensuring that different technicians achieve statistically consistent results on identical materials.
• Research and development of blended cements, supplementary cementitious materials, and chemical admixtures, where reliable baseline strength data are essential.
• Technical arbitration and dispute resolution between cement producers and purchasers when reported strength values conflict.
• Calibration interval reviews and measurement uncertainty validation for mechanical testing systems.

FAQ

What is the difference between a comparison experiment and routine strength testing?

Routine testing determines whether a particular cement meets its specified strength class. A comparison experiment goes further: identical material is tested by multiple operators, machines, or laboratories under the same standard, and the results are statistically compared to verify that the testing process itself is accurate and consistent, independent of the cement being evaluated.

Which standards govern cement mortar strength comparison testing?

The specimen preparation and mechanical testing procedures are defined by ISO 679, EN 196-1, and GB/T 17671. The design and statistical evaluation of the comparison are governed by ISO/IEC 17043 and ISO 13528, while laboratory competence requirements are specified in ISO/IEC 17025. ASTM C109 describes an alternative cube-based method used mainly in North America, and comparisons should only be made between results produced by the same method.

How many specimens are required for one comparison?

One mortar batch produces three prisms, giving three flexural and six compressive determinations at a single age. A complete comparison covering both 3-day and 28-day strengths requires at least two batches, or six prisms, per participant, with spare specimens recommended to cover handling damage or invalid tests.

What defines a satisfactory result in the comparison?

A laboratory's result is satisfactory when its absolute z-score is no greater than 2, meaning its deviation from the assigned value is within two standard deviations of the proficiency assessment. Results between 2 and 3 indicate a warning condition, and results of 3 or more require root-cause investigation and corrective action.

What are the most common sources of discrepancy between laboratories?

Frequent causes include inaccurate loading rates on testing machines, improper calibration of platens and force sensors, curing water temperature outside 20 °C ± 1 °C, delayed testing beyond the permitted age window, non-uniform compaction on a malfunctioning jolting table, moisture contamination of the cement sample, and inconsistent specimen positioning in the flexural or compressive fixtures.

How often should comparison experiments be performed?

Interlaboratory comparisons are typically conducted at least once per year as part of external quality assurance, while internal comparisons between operators or machines are recommended quarterly or whenever equipment is relocated, overhauled, or recalibrated. Higher frequency is advisable for laboratories supporting cement certification or commercial dispute resolution.