The reliability of automotive batteries is an important factor, especially for the OEM market. To ensure consistent reliability all automotive batteries are subjected to electrical load tests prior to shipment. These tests are performed in accordance with industry accepted standards. It is essential that the energy discharged from the battery during test be minimized to avoid the cost and delay associated with recharge. When defining the ideal high rate discharge test process one must consider these basic requirements. Immediately after formation, and preferably, after post cleaning, batteries are discharged at high current for a few seconds. The voltage under load at the end of test is determined and compared to nominal values resulting in a pass/fail decision. This paper discusses the effectiveness of the present high rate discharge test evaluation criteria in terms of it's ability to accurately discriminate between acceptable and unacceptable batteries. Variations in materials and manufacturing processes affect open-circuit voltage (OCV) and constant current voltage (CCV) measurements leading to erroneous pass/fail determinations. The paper will illustrate how these variations can be recognized and statistically evaluated. A supplemental evaluation criterion is presented which significantly improves the reliability of test results. This criterion is the gradient DeltaV of the discharge curve. Evaluation of test data using this criterion yields information that directly correlates to the failure mode of the rejected battery, including irregularities in pasting and assembly. This paper concludes that the implementation of DeltaV criterion significantly improves the failure detection of batteries subjected to production line high-rate discharge tests. Using the DeltaV criterion to evaluate the test results is a more accurate determinant of an acceptable battery than the traditional criteria and results in fewer customer returns. (C) 2001 Elsevier Science B.V. All rights reserved.