Advanced science.  Applied technology.


Reconciling Octane Rating and Autoignition, 03-R8820

Principal Investigators
Vickey Kalaskar
Robert Legg
Adam Van Horn
Julian Wallace
Inclusive Dates 
01/01/18 - Current


When octane ratings were first codified nearly 100 years ago, it represented engines, fuels, and operating conditions of the time. Since then, both the fuels and the engines in the market have changed significantly. Octane numbers have steadily increased and today many market fuels include high octane blend components, typically alcohols and ethers, to comply with renewable fuels legislation. Engine technology has undergone an equally dramatic evolution: higher compression ratios, forced induction, fuel injection, better materials, addition of aftertreatment, and huge advances in engine controls, to name a few. The industry at large, realizing the disconnect between the knock-limited operation in a Cooperative Fuel Research (CFR) octane rating engine and the engines in the market, is tackling the problem by one of two broad approaches. The first is to adapt the octane rating to be more representative of modern engines. The second approach is simply to use the octane numbers as measured and attempt to correlate Research and Motor Octane Numbers (RON and MON respectively) to the knock-limited performance of modern engines.


This project attempted to reconcile knock in different engines and the underlying autoignition by studying fuels in both the CFR engine and the SwRI developed Single Cylinder Research Engine (SCRE), representing a modern direct injection, turbocharged SI engine. 13 different fuels were selected for this study, which included primary reference and toluene standardization fuels, as well as three full boiling range gasolines with a selection of 0, 10 and 20 % ethanol content. Both engines utilized high speed data acquisition to capture cylinder pressure, exhaust oxygen and a variety of temperatures. The ultimate goal of the project was to develop a unifying autoignition model that is able to describe and predict knock in both modern engines and the CFR engine, and to additionally employ big data capabilities at SwRI to parameterize the model for individual engine cycles, rather than the traditional approach of utilizing a single median or representative pressure trace.


All test fuels were prepared and relevant properties determined, including laminar burning velocity. CFR engine tests were completed at standard knock intensity under both RON and MON test conditions, as well as with a stoichiometric mixture and under borderline knock and knock-free conditions. SCRE test points covered the full range of speeds and loads and included both knock-free and knocking conditions. For knocking conditions, additional tests were completed to asses each fuel’s sensitivity to varying ignition timing and air-fuel ratios. Fuel consumption measurements for the different fuels were used to predict fuel economy values over various standard emission drive cycles using a vehicle model. The analyses revealed that lower ethanol content and higher fuel density correlated with improved predicted fuel economy and that the impact of higher RON depended on the drive cycle, providing some benefit for the higher loaded US06 and WLTC drive cycles. These fuel economy analyses results were accepted for publication at the 2019 SAE World Congress (WCX19). Data assimilation and autoignition modelling is ongoing, as is the big data analyses.