Patrick Kirchen

A skeletal kinetic mechanism for PRF combustion in HCCI engines

Abstract A single zone thermodynamic model, coupled to a kinetic mechanism, is developed and is capable of predicting the ignition timing of Primary Reference Fuels (PRFs) in a Homogeneous Charge Compression Ignition (HCCI) engine. A new combination of kinetic mechanisms is used, which includes 120 reactions and 58 species for both ignition and high temperature reactions. The model is validated using a step by step methodology. The validation compares ignition delays predicted by the model with published measurements from a rapid compression machine, shock tube as well as the cylinder pressure histories taken from two different experimental HCCI engines for various operating conditions. The model is able to qualitatively predict the effect of different parameters such as gas temperature, gas pressure, equivalence ratio and octane number on the HCCI ignition delay.

Exhaust-Stream and In-Cylinder Measurements and Analysis of the Soot Emissions From a Common Rail Diesel Engine Using Two Fuels

The operation and emissions of a four cylinder, passenger car common-rail diesel engine operating with two different fuels was investigated on the basis of exhaust-stream and in-cylinder soot measurements, as well as a thermodynamic analysis of the combustion process. The two fuels considered were a standard diesel fuel and a synthetic diesel (fuel two) with a lower aromatic content, evaporation temperature, and cetane number than the standard diesel. The exhaust-stream soot emissions, measured using a filter smoke number system, as well as a photo-acoustic soot sensor (AVL Micro Soot Sensor), were lower with the second fuel throughout the entire engine operating map. To elucidate the cause of the reduced exhaust-stream soot emissions, the in-cylinder soot temperature and the KL factor (proportional to concentration) were measured using miniature, three-color pyrometers mounted in the glow plug bores. Using the maximum KL factor value to quantify the soot formation process, it was seen that for all operating points, less soot was formed in the combustion chamber using the second fuel. The oxidation of the soot, however, was not strongly influenced by the fuel, as the relative oxidized soot fraction was not significantly different for the two fuels. The reduced soot formation of fuel two was attributed to the lower aromatic content of the fuel. The soot cloud temperatures for operation with the two fuels were not seen differ significantly. Similar correlations between the cylinder-out soot emissions, characterized using the pyrometers, and the exhaust-stream soot emissions were seen for both fuels. The combustion process itself was only seen to differ between the two fuels to a much lesser degree than the soot formation process. The predominant differences were seen as higher maximum fuel conversion rates during premixed combustion at several operating points, when fuel two was used. This was attributed to the lower evaporation temperatures and longer ignition delays (characterized by the lower cetane number) leading to larger premixed combustion fractions.

A Phenomenological Mean Value Soot Model for Transient Engine Operation

A means of characterizing the raw soot emissions from modern common-rail diesel engines is required so that their control and the regeneration of an associated particulate filter can be optimized, though no suitable sensor currently exists. Thus, under the framework of a FVV research project, a fast soot model to calculate the raw soot emissions from a common-rail diesel engine operating under both steady state and transient operating conditions was developed and validated at the Aerothermochemistry and Combustion Systems Laboratory (LAV) at the ETH Zurich.

Characterization of reaction zone growth in an optically accessible heavy-duty diesel/methane dual-fuel engine

The performance of dual-fuel engines in terms of fuel conversion efficiency and unburned hydrocarbon emission is strongly influenced by the turbulent flame propagation through the premixed natural gas. To improve dual-fuel engine design and provide validation data for numerical models, the fuel conversion process must be better characterized, specifically the reaction zone growth rate. In this work, high-speed imaging of OH*-chemiluminescence is performed in an optically accessible 2 L engine operated with port-injected CH4 and direct-injected diesel for different diesel and CH4 fueling rates and pilot injection pressures (Ppilot). The cumulative histogram time series is introduced for directly comparing high-speed optical data of dual-fuel combustion with simultaneously measured apparent heat release rate. The cumulative histogram time series diagram is also used to evaluate a “global” reaction zone speed, SRZ,g, based on OH*-chemiluminescence images. The SRZ,g calculation normalizes the reaction zone area growth rate by the perimeter of the reaction zone to determine the velocity scale, while a “local” reaction zone speed, SRZ,l, is based on the local displacement of the reaction zone boundary per unit time. From the distribution of SRZ,l for each image frame, a previously proposed metric for determining the transition from pilot autoignition based on apparent heat release rate was validated and used to evaluate a single mean flame propagation speed, S ¯ FP . Using these metrics, it was noted that increasing ϕCH4 from 0.40 to 0.69 results in an increase in S ¯ FP from 4 to 8 m/s and 8 to 14 m/s for pilot injection pressures of 300 and 1300 bar, respectively. The spatial distribution of SRZ,l also indicates that autoignition of the pilot jets is not simultaneous (arising from asymmetric injector geometry) and leads to an overlap of the autoignition and flame propagation processes. This is not considered in the conventional conceptual model of dual-fuel combustion and impacts calculation of S ¯ FP for the small diesel injections commonly used for dual-fuel engines.

Phänomenologisches Mittelwertmodell für Ruß in transientem Motorbetrieb

Um die Steuerung und Regelung eines Dieselmotors und die Regeneration seines Ruspartikelfilters zu optimieren, mussen die Rusrohemissionen an Serienmotoren quantifiziert werden, wofur aber zurzeit kein geeigneter Sensor existiert. Im Rahmen eines FVV-Forschungsvorhabens wurde anhand von stationaren und transienten Rus emissionsmessungen ein schnelles Mittelwert-Rusmodell sowohl fur stationaren als auch transienten Betrieb eines Common-Rail-Dieselmotors am Laboratorium fur Aerothermochemie und Verbrennungssysteme (LAV) der ETH Zurich entwickelt und validiert.