This report gives the results of two cross-testing experiments on the Los Angeles (LA) test - a "European" experiment involving 28 laboratories from 13 countries, and a "French" experiment involving 33 French laboratories. (Four of the French laboratories are included in both experiments.) The three materials that were used in these experiments were used throughout the four-year programme of cross-testing experiments involving tests of the mechanical properties of aggregates.
The results of the experiments show that the reproducibility of the Los Angeles test is adequate for it to be used to check the compliance of aggregates with specifications. The results also show that it should be possible to improve its reproducibility. As this will be beneficial for aggregates that give results close to specification limits, it is recommended that the cross-testing experiments should be followed-up, in those laboratories that gave the highest and lowest average results, by comparisons of equipment and procedures, and by experiments to identify the causes of laboratory biasses. The results of these investigations should be collated and made available to all users of the test.
It is concluded that the repeatability of the test is such that the results of duplicate tests (tests on two test portions from the same laboratory sample) should rarely differ by more than 1.0 (LA units). However, some laboratories reported results of duplicate tests whose ranges were much larger than this, some much larger. Such results will undermine confidence in the test method, so it is recommended that all laboratories should do duplicate tests regularly, and use the results to check that their repeatability is satisfactory. It will be of interest to see if this degree of repeatability can be achieved with heterogeneous materials, such as recycled aggregates produced from construction and demolition waste.
The results of the "European" experiment allow the precision of the Los Angeles test to be as:
r1 = 0.06 X (LA units) and R1 = 0.17 X (LA units)
where "X" represents the average of the test results.