Enhancement of the Sequence IIIG by the Study of Oil Consumption

The Sequence IIIG is a newly developed 100 hour test used to evaluate the performance of crankcase engine oils in the areas of high temperature viscosity increase, wear, deposits, pumpability, and ring sticking for the North American GF-4 standard. Data from the ASTM Precision Matrix, completed in the spring of 2003, along with early reference data from the Lubricant Test Monitoring System (LTMS) showed unexpected test results for selected oils and indicated that percent viscosity increase and pumpability were highly correlated with oil consumption. This correlation led to an intensive study of the factors that influence oil consumption and an attempt to compensate for non-oil related oil consumption through a model based adjustment of the results. The study and scrutiny of the IIIG data has led to more uniform oil consumption in the test and improved test precision, and has eliminated the need for a correction equation based on non-oil related oil consumption. While the adoption of a bias correction should not routinely substitute for quality root cause analysis, it can be a temporary practical solution that leads to reduced variability of a test, and therefore to better performance evaluation of the oil. The objective of this paper is to document the statistical methodology and the progress made during the IIIG development with respect to the application of a bias correction to reduce test variability. In the future, if similar problems arise, the paper may be used as a guide and facilitator for discussion towards possible expedient and practical solutions.