Measuring Diesel Emissions with a Split Exhaust Configuration 2001-01-1949

West Virginia University evaluated diesel oxidation catalysts (DOC) and lean-NO_X catalysts as part of Diesel Emissions Control-Sulfur Effects (DECSE) project. In order to perform thermal aging of the DOC and lean-NO_X catalysts simultaneously and economically, each catalyst was sized to accommodate half of the engine exhaust flow. Simultaneous catalyst aging was then achieved by splitting the engine exhaust into two streams such that approximately half of the total exhaust flowed through the DOC and half through the lean-NO_X catalyst. This necessitated splitting the engine exhaust into two streams during emissions measurements. Throttling valves installed in each branch of the split exhaust were adjusted so that approximately half the engine exhaust passed though the active catalyst under evaluation and into a full flow dilution tunnel for emissions measurement. The remaining exhaust flowed through a passive catalyst of identical dimensions and containing a similar ceramic substrate, but with no washcoat of active catalyst material, and was vented from the test facility. Prior to the test cycle, the throttling valves were adjusted at a prescribed steady-state operating point such that the flows in the two branches of the exhaust system were the same while maintaining the specified engine backpressure. The fraction of exhaust entering the measurement system was calculated based on a comparison of measured CO₂ concentrations taken during a full flow exhaust run and compared to the partial flow exhaust run. Emissions sampled from the partial exhaust stream were then corrected based on these measured CO₂ concentrations to reflect the original full flow levels. Instantaneous split ratios were difficult to quantify during transient engine operation, since any variance in time alignment of the CO₂ analyzer output trace between the full flow and successive split flow runs would cause substantial errors in the calculation of the split ratio. However, integrated engine-out emissions measured in the split exhaust streams during transient operation generally agreed well with emissions measured using the full exhaust stream. Measured emissions in the partial exhaust configuration remained within 5% of a 50/50 split except at idle or low power operation. When these results from a split-flow test were corrected for full flow, measured emissions were within 11% of the full exhaust case for transient tests and 13% for steady state tests. An average difference between full flow and 50/50 corrected test runs was 4% for transient runs and 7% for steady state runs.