California Air Resources Board

The study documented by this report is an addition to previously reported studies (Off-Road Low NOx Demonstration Stage 1, Off-Road Low NOx Demonstration Stage 2, and Off-Road Low NOx Demonstration Stage 3) to demonstrate the feasibility of technologies to reach tailpipe oxides of nitrogen emission reductions of up to 90% from the current Tier 4 final standards, and particulate matter emission reductions of at least 75% from the current Tier 4 final standards. This additional work was aimed at expanding the scope of the original technology demonstration to encompass field duty cycles from real-world applications, in addition to the regulatory test cycles that are typically used for emission certification and compliance testing.

This study leveraged the Package 2 Tier 5 demonstration engine-aftertreatment system that was developed under the California Air Resources Board (CARB) Contract 20MSC005. A number of field duty cycles were developed in partnership with CARB staff, based on application data that was originally recorded in the field for various off-road applications. Each data set represented a full shift-day of operation for a given machine, starting from initial cold-start. The field data sets were translated into normalized duty cycles that could be run on an engine dynamometer, allowing for evaluation of the performance of the Tier 5 demonstration engine over real-world duty cycles. These duty cycles were examined using the full useful life emission control system, which was aged to the equivalent of 8,000 hours in the previous program. Both nominal or low power child ratings of the engine were examined during this effort, and evaluations were conducted using both the base Tier 5 package and an alternate packaging arrangement. Testing was conducted both with and without the use of cylinder deactivation in the form of Dynamic Skip Fire. The Tier 5 demonstration generally performed well under these conditions, showing significant margins to proposed emission thresholds using the proposed bin-moving average window data analysis methods.