Testing of industrial products - Aggregates for construction

Determination of the Particle Density of Fillers - Summary

This report gives the results of a cross-testing experiment on the test method used to determine the particle density of filler. 20 laboratories took part from 13 countries and three materials were used.

The reproducibility of the determination of the particle density of fillers is surprisingly high for such a simple test method, involving only a number of weighings that should be carried out to an accuracy of 0.01 g . Sensitivity analysis reveals that the determination of particle density is very sensitive to variations in the results of determinations of the density of the liquid used in the test. It would be of interest to carry out an experiment to establish the reproducibility of the determination of the density of the liquid. If it is as large as R = 0.004 Mg/m3 then this source of variation alone would cause the reproducibility of the determination of the particle density of filler to be R1 = 0.100 Mg/m3. The determination of particle density is sensitive to the determination of liquid density because the specimen volume (Vt = 4.5 mL, approximately) is small compared with the volume of the pyknometer (V = 50 mL, approximately). The sensitivity could be reduced substantially by increasing the size of the test specimen, and this should improve the reproducibility of the determination of particle density. If this is not practical, then to improve the reproducibility of the test method it will be necessary to carry out the weighings to a higher degree of accuracy than is currently required.

It is recommended that replicate determinations of the liquid density should be made, and that the use of pyknometers with volumes larger than 50 mL should be discouraged. With regard to the determination of the pyknometer volume, the Standard requires the weighings to be made to 0.01 g, so that it is justifiable to record the volume of the pyknometer to the nearest 0.01 mL . This is satisfactory according to sensitivity analysis. (It should be noted that Annex A in the draft CEN method requires the pyknometer volume to be recorded to the nearest 0.001 m3, which is obviously a mistake.) It may be acceptable to increase the tolerance on the temperature of the water bath from 0.1 C to 0.5 C.

Several laboratories reported data that gave rise to between-test-result ranges or between-specimen standard deviations that were outliers. The test method requires three determinations to be carried out to obtain a test result. It is therefore recommended that whenever they do a test, laboratories should calculate the range of the three determinations and compare it with the critical range of Wc = 0.025 Mg/m3 established in this experiment. Also, it should be noted that the data as reported contained many inconsistencies between the measured and calculated values. These inconsistencies could have arisen because of calculation errors, transcription errors, or rounding.

The reproducibility and repeatability of the determination of the particle density of filler may be summarised as:

r1 = 0.015 Mg/m3

and

R1 = 0.106 Mg/m3