Cement fineness parallel sample testing is a laboratory quality control procedure in which two or more independently split portions of the same cement sample are analyzed for fineness—typically 45 μm or 80 μm sieve residue and Blaine specific surface area—to verify the precision and reliability of the reported results. Conducted in accordance with standards such as GB/T 1345, GB/T 8074, EN 196-6, and ASTM C430, parallel determination reveals operator, apparatus, or sampling errors before data are released. This article explains the sample types, tested parameters, test methods, acceptance criteria, and practical applications of parallel fineness testing in cement and concrete laboratories.
Technical Overview
Fineness is one of the most influential physical characteristics of hydraulic cement. It controls the rate of hydration, early and late strength development, water demand, heat evolution, and long-term durability. A cement that is ground too coarsely hydrates slowly and gains strength sluggishly, while an excessively fine cement raises water demand, grinding energy, and the risk of early-age cracking. In practice, fineness is expressed either as a sieve residue retained on a specified aperture (most commonly 45 μm or 80 μm) or as a specific surface area determined by air permeability, in m²/kg or cm²/g.
Parallel sample testing (duplicate or replicate determination) means that a single homogenized laboratory sample is divided into at least two independent test portions, which are then prepared and tested under repeatability conditions—same operator, same apparatus, same environment, and within a short time interval. The agreement between the parallel results is a direct measure of the precision of the determination. If the difference between the two results falls within the repeatability limit defined by the governing method, the mean value is reported; if not, the cause must be investigated and the determination repeated.
- GB/T 1345 specifies the sieve methods for cement fineness (negative-pressure air-jet sieve, water sieve, and manual dry sieve) using 45 μm and 80 μm square-hole sieves.
- GB/T 8074 and ASTM C204 specify the Blaine air permeability method for specific surface area, while EN 196-6 covers both sieving and permeability fineness determination.
- ASTM C430 specifies the 45 μm (No. 325) sieve method for hydraulic cement, and ASTM C115 provides the turbidimetric alternative.
- GB 175 defines the product-level fineness requirement—for example, a minimum specific surface area of 300 m²/kg for Portland and ordinary Portland cements, or a maximum 80 μm sieve residue of 10 % for most blended cements.
Testing Samples
Ordinary Portland and Blended Cements
The core scope covers all common hydraulic cements: Portland cement and ordinary Portland cement, slag cement, pozzolanic cement, fly ash cement, and composite cement classified under GB 175, CEM I to CEM V under EN 197-1, and Types I to V under ASTM C150 and ASTM C595. Parallel fineness testing is routinely applied to these materials for both factory quality control and third-party conformity assessment.
Supplementary Cementitious Materials and Mineral Additions
Ground granulated blast-furnace slag, fly ash, limestone filler, and natural pozzolana are frequently checked for fineness because their reactivity is strongly particle-size dependent. The same parallel-sample protocol applies when these materials are evaluated against GB/T 18046, GB/T 1596, EN 450, EN 15167, or ASTM C618 requirements.
Special Cements and Intermediate Products
White cement, rapid-hardening cement, masonry cement, cement clinker, and interground clinker–gypsum mixtures at mill discharge can all be characterized with parallel fineness determinations. For these samples, the chosen method and sieve aperture must match the relevant product standard or the agreed technical specification.
Sample Preparation and Splitting
Correct splitting is the foundation of a valid parallel test. The laboratory sample is first homogenized and gently passed through a 0.90 mm sieve to break soft agglomerates and remove foreign particles. It is then divided into two or more nearly equal portions using a riffle splitter or the cone-and-quartering technique. Typical portion masses are approximately 10 g for a 45 μm sieve determination and approximately 25 g for an 80 μm determination under GB/T 1345, about 1 g under ASTM C430, and a density-based calculated mass for the Blaine method. Portions must be stored in airtight containers and tested promptly to prevent moisture pickup and carbonation.
Testing Items
Parallel sample testing for cement fineness typically covers the following measurands and control parameters:
- 45 μm sieve residue, expressed as the mass percentage retained on the sieve after correction.
- 80 μm sieve residue, used in Chinese practice as the alternative conformity indicator.
- 90 μm sieve residue, where EN 196-6 sieving is applied.
- Blaine specific surface area in m²/kg, determined at a compacted bed porosity of 0.500.
- Particle size distribution by laser diffraction (Dv10, Dv50, Dv90, fines fraction, and coarse tail) as an optional complementary characterization.
- Sieve correction coefficient and apparatus calibration status, verified against a certified reference sieve or reference cement.
- Repeatability indicators of the parallel test itself: the absolute difference between the two results and the relative deviation, calculated as |x1 − x2| divided by the mean of x1 and x2, multiplied by 100 %.
Testing Methods
Negative-Pressure Air-Jet Sieve Method
A weighed test portion is placed on a certified 45 μm or 80 μm square-hole sieve enclosed in a sieve seat connected to a vacuum source. Air is drawn through the mesh at a controlled negative pressure—commonly in the range of 4000 to 6000 Pa—for about two minutes, while a rotating brush keeps the mesh clear. The retained material is brushed from the sieve, transferred, and weighed to 0.01 g. The residue percentage is calculated from the retained mass divided by the original mass and multiplied by the sieve correction coefficient.
Water (Wet) Sieve Method
The portion is washed on the mesh with a gentle spray of water until the rinse water runs clear, dispersing agglomerates that dry sieving cannot break down. The residue is dried to constant mass, cooled, and weighed. The water sieve is a robust cross-check against the air-jet sieve and is particularly useful for cements prone to electrostatic clumping.
Manual Dry Sieve Method
Where powered equipment is unavailable, the portion is sieved by hand with a regular tapping and rotating motion until the passing rate becomes negligible, after which the residue is weighed. Because manual results carry higher variability, the parallel-sample comparison is especially important for validating this method.
Blaine Air Permeability Method
A precisely weighed mass of cement, calculated from the cement density to achieve a bed porosity of 0.500, is compacted in a cylindrical cell. The time required for a fixed volume of air to pass through the bed under a defined pressure head is measured and related to specific surface area through an apparatus constant determined with a certified reference cement. Parallel determinations are made on two independently prepared beds, and the results must agree within the tolerance stated in the method—commonly within about 1 to 2 % relative of the mean in routine laboratory practice.
Laser Diffraction Particle Size Analysis
As a complementary technique performed in accordance with ISO 13320, laser diffraction delivers a full particle size distribution rather than a single fineness index. It is valuable for diagnosing grinding circuit problems, such as an excessive coarse tail above 45 μm or over-grinding below 3 μm, and it provides an independent check on sieve and Blaine results.
Parallel Determination Procedure and Acceptance Criteria
Both portions are tested in immediate succession by the same operator using the same calibrated apparatus. The results are compared against the repeatability limit of the method: under GB/T 1345, a common acceptance criterion is that the absolute difference between parallel sieve residue determinations does not exceed 0.5 %. If the limit is met, the arithmetic mean is reported; if it is exceeded, the laboratory must check sample homogeneity, sieve condition, weighing accuracy, and compaction technique, then repeat the determination. For Blaine testing, exceeding the allowed difference triggers preparation of a fresh bed and a third determination, with the outlier assessed and handled according to the laboratory's documented statistical rules.
Applications
Parallel fineness testing supports every stage of the cement and concrete value chain:
- Ball mill and vertical roller mill optimization, including separator or classifier adjustment and specific-energy monitoring in cement plants.
- Factory quality control and statistical process control, where parallel results feed control charts and trend analysis.
- Conformity assessment against GB 175, EN 197-1, ASTM C150, and ASTM C595 for certification, batch release, and regulatory filing.
- Incoming material verification at ready-mixed concrete plants, precast factories, and construction project laboratories.
- Independent testing by accredited third-party laboratories for trade settlement, contract disputes, and import or export inspection.
- Interlaboratory comparison and proficiency testing schemes that benchmark measurement capability across laboratories.
- Research into hydration kinetics, packing density, and the grindability of supplementary cementitious materials.
FAQ
Why is parallel sample testing required for cement fineness?
Fineness results depend on subtle factors such as sieve wear, bed compaction, ambient humidity, and weighing precision. Testing two independently split portions under repeatability conditions quantifies the random error of the determination, confirms that the apparatus and technique are under control, and prevents a single anomalous value from being reported.
What acceptance criterion applies to the difference between parallel results?
The governing method defines the limit. For sieve residue under GB/T 1345, the absolute difference between parallel determinations is commonly required not to exceed 0.5 %. For Blaine specific surface area, the two results must agree within the tolerance set by the applicable method, typically within about 1 to 2 % relative of the mean. The relative deviation formula |x1 − x2| / mean × 100 % is used for the evaluation.
What should be done if parallel results exceed the allowed difference?
The laboratory must stop and investigate: verify sample homogeneity and splitting, inspect the sieve mesh and its correction coefficient, check balance calibration and Blaine cell compaction, and review environmental conditions. The determination is then repeated on fresh portions, and the retest results govern the reported value in accordance with the standard and the laboratory's quality procedures.
How are sieves and Blaine apparatus kept traceable?
Working sieves are verified against a certified reference sieve at defined intervals and after a specified number of tests, and a correction coefficient is applied to each result; the coefficient must remain within the range allowed by the standard, otherwise the sieve is replaced. Blaine apparatus is calibrated with a certified reference cement of known specific surface, with the calibration record maintained for full metrological traceability.
How much sample is needed for a parallel fineness test?
A practical submission of about 100 to 200 g of cement is sufficient. Each individual determination consumes only a small portion—roughly 10 g for a 45 μm sieve test and 25 g for an 80 μm sieve test under GB/T 1345, about 1 g under ASTM C430, and a few grams for a Blaine bed—so the balance of the material serves as retention sample for possible retesting.
How does parallel testing differ from routine duplicate checking?
Parallel testing is a defined precision requirement embedded in the test method itself, performed on every determination under strict repeatability conditions. Duplicate checking, by contrast, is an internal quality assurance step in which a portion is retested at a later time or by another operator to monitor long-term stability. Both practices together form the backbone of reliable fineness reporting in accredited third-party laboratories and factory laboratories alike.