Measurement of RON Requirements for Turbocharged SI Engines: One Step to the Octane on Demand Concept

Knock phenomena in Spark Ignition (SI) engines (especially for turbocharged engine) is limiting both for engine global efficiency at high load and maximum performance. Several parameters can change the occurrence of knock and might be classified into two different categories: common engine tuning parameters such as Spark Advance (SA), Variable Valve Timing (VVT) position, Start Of Injection (SOI), dilution of the fuel/air mixture if available on the one hand, and hard to change parameters such as compression ratio (CR), fuel octane number (RON: Research Octane Number), etc. on the other hand.
The aim of the current research program is to use the octane number as a tuning parameter and to improve the engine efficiency and its CO2 emissions. The idea is to keep the engine operating on the entire map without occurrence of knock by adapting its RON feed in order to preserve its cycle efficiency (optimum Spark Advance). One major step in reaching this goal is to first quantify the octane quality needed to keep the best efficiency (optimal spark advance (SA) without engine knock) for each operating point (OP) of the entire engine map.
On an up-to-date turbocharged SI engine (1.6L Gasoline Direct Injection (GDI)), tests have been performed at test bench. RON was widely varied from 71 to 111, using surrogate fuels (TRF: Toluene Reference Fuels which are mixtures of n-heptane and toluene in varying proportions). This supplementary degree of freedom (RON value of the fuel) provides further opportunity for a new compression ratio optimization. In this context, besides RON variation tests, a wide three-step variation of compression ratio (CR): 7.5:1, 10.5:1 (stock CR) and 12:1 has been performed. To be representative of the real engine behavior under real driving conditions, the major part of the engine map has been tested at the test bench: from very low load to full load and from 1000 rpm to 5000 rpm.
Stock compression ratio (10.5:1) results show the improvement of combustion phasing (less retarded) with RON increase. Lowering RON is also very interesting. The lowest RON value (71) may be used with the optimal combustion phasing on a significant part of the engine map for the three compression ratios. Obviously, at low CR this area is larger than at higher CR. In other words, low octane fuel (i.e. RON 71) can be suitable for some parts of driving cycles, even for high CR. In case of a dual fuel (Octane on Demand) engine, a significant fraction of such a low octane fuel can be anticipated.
Furthermore, this large RON variation (from 71 to 111), especially for CR 10.5:1 and 12:1, reveals the non-linearity of RON effect on combustion phasing (i.e. knock occurrence). In fact, for low octane values (<97) the anti-knock behavior is lower than at higher RON value.
The overall comparison of RON and CR variation results allows to clearly show the efficiency and CO2 emissions benefits of the Octane on Demand concept.