Full Length ArticleMapping K factor variations and its causes in a modern, spark-ignition engine
Introduction
Fuel economy regulations have been consistently driving the development of more efficient spark-ignition (SI) engines [1], [2], [3]. The application of modern technologies, such as high compression ratio, turbocharging, engine downsizing and downspeeding, promotes the tendency of engine knock. Various measures have been implemented to control knock, principally spark timing retard and fuel enrichment at some conditions. However, these measures have negative effects on engine efficiency and exhaust emissions.
The knock resistance of a fuel is characterized by two octane numbers – the Research Octane Number (RON) [4] and the Motor Octane Number (MON) [5]. These are determined with the Cooperative Fuel Research (CFR) engine using standard methods (Table 1). The RON and MON are determined by referencing the knock resistance of a test fuel to that of Primary Reference Fuels (PRFs) at corresponding conditions. PRFs are binary mixtures of iso-octane and n-heptane with the former defining an octane number of 100 and the latter defining an octane number of 0. The octane number of a PRF mixture is defined as the volumetric fraction of iso-octane in the mixture. As shown in Table 1, the RON is tested at an overall lower intake temperature than MON, and fuel vaporization cooling further reduces the mixture temperature past the carburetor. For most practical fuels, the RON is higher than the MON, but for PRFs their RON and MON are defined to be the same. The difference between RON and MON is known as octane sensitivity (S = RON – MON). The origin of S lies in the different autoignition chemistry between PRFs and octane sensitive fuels [6], [7], [8] as well as the different physical properties such as heat of vaporization (HoV) [9], [10].
Historically, the two octane tests were designed to bracket the operating conditions of typical engines in the 1930s, where higher RON and MON generally meant better anti-knock quality. However, modern engines with direct injection (DI), turbocharging and variable valve timing (VVT) have vastly different cylinder conditions from those in the octane rating tests. As a result, the relevance of RON and MON to modern engines has received significant attention. Various studies in gasoline turbocharged direct-injection (GTDI) engines found that fuels of higher RON and lower MON (i.e. higher S) showed higher knock resistance and therefore allowed engines to operate with higher efficiency at high-load conditions [13], [14], [15], [16], [17], [18]. Vehicle tests reported similar trends, where the acceleration performance improved with fuels of high RON and low MON [15], [19], [20]. These findings indicate the relevance of RON and MON to knock resistance in modern SI engines should be re-evaluated.
An alternative metric, Octane Index (OI), was proposed by Kalghatgi in 2001 [20], [21] to characterize a fuel’s actual anti-knock quality in modern engines. As given by Eq. (1), OI is a linear combination of the RON and MON. K is a weighting factor that is primarily dependent on engine design and engine operating conditions. For example, the anti-knock index (AKI) used in North America approximates the OI by evenly weighting the RON and MON to a fuel’s actual anti-knock quality (K = 0.5). The OI concept therefore generally reflects a fuel’s knock resistance relative to its RON and MON ratings. A more commonly used expression for OI is given by Eq. (2), where S is the octane sensitivity. For an octane sensitive fuel (S greater than 0), its OI can be larger than the RON (when K less than 0), or smaller than the MON (when K greater than 1), or in between, depending on the engine operation conditions. For PRFs, the OI always equals the RON and MON because S = 0. Although K is irrelevant in this case, PRFs provide a valuable reference for K determination, as shown later.
In principle, K can be used to quantify the relative significance of the RON and MON to SI engine combustion. For example, when K is greater than 0.5, the MON is more important than the RON, and the higher the MON, the stronger the knock resistance is (at that operating condition in that engine); however, when K is negative, a higher MON causes a lower OI. Indeed, previous studies in modern SI engines showed that K was negative at most high-load conditions and therefore concluded that fuels of lower MON were advantageous [15], [16], [17], [21], [22], [23]. However, in practical driving, engines primarily operate at part loads, and recent studies found that K could be positive at part-load, knock-limited conditions [24], [25]. Therefore, further study is needed to determine K over a wide range of conditions in modern SI engines.
The objective of this work, therefore, is to determine K values in the Octane Index model over the full operation map of a modern GTDI engine, including severe operating conditions that involve high intake air and/or high coolant temperatures. A novel fuel blending system is developed to continuously prepare and supply PRF mixtures of the needed composition in real time to match the anti-knock performance of a 92 RON/83 MON gasoline, from which the K of each condition is determined. The obtained K dataset is used to generate a contour map of K overlaid on the engine operation map. The comprehensive dataset also enables statistical analysis for potential correlations between K and the end gas states, of which fundamental understanding is currently lacking.
Section snippets
Engine setup
The experiments are conducted on a Ford EcoBoost engine connected to a Horiba transient dynamometer. The engine is a 2-litre, 4-cylinder, GTDI engine with a compression ratio of 9.3:1 and twin independent variable cam timing, which is a representative of modern downsized SI engines (Table 2). Engine coolant temperature was controlled by a coolant conditioning system. Intake air temperature was controlled by a water-to-air intercooler. The engine was controlled and monitored by a Motec M142 ECU
Repeatability tests
To ensure the consistency of K measurement, reference experiments were repeated regularly at 1500 RPM, 6.5 bar BMEP and normal temperature conditions. Fig. 4 shows a set of repeatability tests taken over several days of experimentation. The observed confidence interval of 0.025 corresponds to an uncertainty of 1.2% relative to the total measured range of K, from −1.0 to 1.1.
Normal temperature results
The results of K and OI for load sweeps at 1000–3000 RPM at normal temperature conditions are plotted in Fig. 5. The K
Conclusions
This work examined how the K factor defined in the Octane Index model varied across a wide range of operating conditions in a modern, 4-cylinder, 2L, turbocharged, gasoline direct injection engine. K quantifies the relative significance of RON and MON. For example, when K is less than 0.5, knock propensity is better described by RON than MON. The main findings of this paper are summarized as follows.
- •
A novel, real-time and controllable Primary Reference Fuel (PRF) blending system was first
CRediT authorship contribution statement
Zhenbiao Zhou: Methodology, Investigation, Formal analysis, Writing - original draft. Tanmay Kar: Investigation. Yi Yang: Conceptualization, Methodology, Supervision, Writing - review & editing, Funding acquisition. Michael Brear: Conceptualization, Supervision, Writing - review & editing, Funding acquisition. Jingran Xu: Investigation. Joshua Lacey: Supervision, Writing -, review & editing, Funding acquisition. Thomas Leone: Conceptualization, Resources, Writing - review & editing, Funding
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
This research was supported by the Ford Motor Company.
References (46)
- et al.
Chemical kinetics of octane sensitivity in a spark-ignition engine
Combust Flame
(2017) - et al.
Significance of RON and MON to a modern DISI engine
Fuel
(2017) - et al.
The octane numbers of ethanol blended with gasoline and its surrogates
Fuel
(2014) - et al.
The research and motor octane numbers of liquefied petroleum gas (LPG)
Fuel
(2013) - et al.
Formulating gasoline surrogate for emulating octane blending properties with ethanol
Fuel
(2020) - et al.
Pressure and temperature effects on fuels with varying octane sensitivity at high load in SI engines
Combust Flame
(2017) - et al.
Quantifying uncertainty in kinetic simulation of engine autoignition
Combust Flame
(2020) - et al.
Understanding chemistry-specific fuel differences at a constant RON in a boosted SI engine
Fuel
(2018) - et al.
A study on the influence of burning rate on engine knock from empirical data and simulation
Combust Flame
(2015) - et al.
Fuel economy potential of variable compression ratio for light duty vehicles
SAE International
(2017)
The effect of compression ratio, fuel octane rating, and ethanol content on spark-ignition engine efficiency
Environ Sci Technol
IC-engine downsizing and pressure-wave supercharging for fuel economy
SAE International
The chemical origin of fuel octane sensitivity
SAE Technical Paper
Detailed chemistry promotes understanding of octane numbers and gasoline sensitivity
Energy Fuels
Exploring the relationship between octane sensitivity and heat-of-vaporization
SAE Int J Fuels Lubr
The effect of charge cooling on the RON of ethanol/gasoline blends
SAE Int J Fuels Lubr
Understanding the octane appetite of modern vehicles
SAE Int J Fuels Lubr
Octane appetite: the relevance of a lower limit to the MON specification in a downsized, highly boosted DISI engine
SAE Int J Fuels Lubr
Cited by (13)
Prediction of spark ignition engine performance and emissions using RERNN approach
2024, International Journal of Hydrogen EnergyThe significance of octane numbers to hybrid electric vehicles with turbocharged direct injection engines
2023, FuelCitation Excerpt :The effects of different RON and MON on drive cycle fuel efficiency are calculated, and the significance of octane numbers to hybrid vehicles is discussed. The drive cycle analysis is based on vehicle modelling which incorporates the K-map previously determined for a Ford GTDI (EcoBoost™) engine [15] and the method for calculating fuel consumption parameters and RON/MON effects [16]. Details of the vehicle modelling are provided below followed by a summary of the K-map data and the method for fuel efficiency calculation.
Effect of fuel composition and EGR on spark-ignited engine combustion with LPG fueling: Experimental and numerical investigation
2022, FuelCitation Excerpt :Finally, the knock mitigation effectiveness of EGR for LPG fueling is evaluated using RON and MON regimes over different pressure–temperature trajectories. This is useful to design the EGR strategy for LPG fueling over the operating map as those regimes represent engine operating conditions spanning from naturally aspirated or lightly boosted engines (i.e., ‘Beyond MON’ space with high temperature and low pressure) to highly boosted, power-dense engines (i.e., ‘Beyond RON’ space with high pressure and low temperature) [36,73]. For a general understanding of fuel chemical kinetics interaction with engine operation, an example of ignition delay time (IDT) profile of conventional SI engine fuel is presented in Fig. 22, highlighting the possible ignition zones over the pressure–temperature domain.
A comparative study of directly injected, spark ignition engine performance and emissions with natural gas, gasoline and charge dilution
2021, FuelCitation Excerpt :Maximum brake torque (MBT) ignition timing (location of 50% mass fraction burned between 8 and 10 deg aTDCFire) was maintained for both GDI and DI CNG operation so long as trace knock was not encountered. In this work, a “trace-knock” or “borderline-knock” condition, which is the transition between knock-free normal combustion and continuous knocking, was determined by both audible signals from a microphone installed on the engine block and in-cylinder pressure oscillations measured by the data acquisition system [31,32]. For all operating conditions, the coefficient of variation of the net indicated mean effective pressure, COVIMEP (as defined in the Appendix) was maintained below 5%.
The significance of octane numbers to drive cycle fuel efficiency
2021, FuelCitation Excerpt :The K-maps, the vehicle test data, and the methods for calculating the fuel consumption parameters and RON/MON effects are introduced in the following sections. Comprehensive sets of K values were recently determined in a Ford 2.0-L 4-cylinder GTDI engine, which covered nearly all knock-limited conditions and some over-advanced conditions [25]. The contour maps of K are shown in Fig. 1.