Internal combustion engines performance, fuel economy and emissions

Internal Combustion Engines: Performance, Fuel Economy and Emissions Tai ngay!!! Ban co the xoa dong chu nay!!! Combustion Engines and Fuels Group Organising Committee: Prof Paul Shayler (Chair) University of Nottingham Dr Frank Atzler Continental Automotive Prof Choongsik Bae KAIST Hugh Blaxill Mahle Powertrain Brian Cooper Jaguar Land Rover Prof Colin Garner Loughborough University Dr Roy Horrocks Ford Motor Company Dr Mike Richardson Jaguar Land Rover Dr Martin Twigg Consultant Dr Matthias Wellers AVL Powertrain Steve Whelan Clean Air Power Prof Hua Zhao Brunel University The Committee would like to thank the following supporters: Automobile Division Internal Combustion Engines: Performance, Fuel Economy and Emissions 27–28 NOVEMBER 2013 IMECHE, LONDON Oxford Cambridge Philadelphia New Delhi Published by Woodhead Publishing Limited 80 High Street, Sawston, Cambridge CB22 3HJ, UK www.woodheadpublishing.com www.woodheadpublishingonline.com Woodhead Publishing, 1518 Walnut Street, Suite 1100, Philadelphia, PA 19102-3406, USA Woodhead Publishing India Private Limited, G-2, Vardaan House, 7/28 Ansari Road, Daryaganj, New Delhi – 110002, India www.woodheadpublishingindia.com First published 2013, Woodhead Publishing Limited © The author(s) and/or their employer(s) unless otherwise stated, 2013 The authors have asserted their moral rights This book contains information obtained from authentic and highly regarded sources Reprinted material is quoted with permission, and sources are indicated Reasonable efforts have been made to publish reliable data and information, but the authors and the publisher cannot assume responsibility for the validity of all materials Neither the authors nor the publisher, nor anyone else associated with this publication, shall be liable for any loss, damage or liability directly or indirectly caused or alleged to be caused by this book Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming and recording, or by any information storage or retrieval system, without permission in writing from Woodhead Publishing Limited The consent of Woodhead Publishing Limited does not extend to copying for general distribution, for promotion, for creating new works, or for resale Specific permission must be obtained in writing from Woodhead Publishing Limited for such copying Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Control Number: 2013954934 ISBN 978 78242 183 (print) ISBN 978 78242 184 (online) Produced from electronic copy supplied by authors Printed in the UK and USA Printed in the UK by 4edge Ltd, Hockley, Essex Ultra boost for economy: realizing a 60% downsized engine concept J W G Turner, A Popplewell, S Richardson Powertrain Research, Jaguar Land Rover Ltd, UK A G J Lewis, S Akehurst, C J Brace Department of Mechanical Engineering, University of Bath, UK S W Bredda GE Precision Engineering, UK ABSTRACT The paper discusses Ultra Boost for Economy, a collaborative project part-funded by the Technology Strategy Board, the UK’s innovation agency ‘Ultraboost’ combines industry- and academia-wide expertise to demonstrate that it is possible to reduce engine capacity by 60% and still achieve the torque curve of a large naturally-aspirated engine, while encompassing the attributes necessary to employ such a concept in premium vehicles In addition to achieving the torque curve of the Jaguar Land Rover 5.0 litre V8 engine, the main project target was to show that such a downsized engine could in itself provide a viable route to a 35% reduction in vehicle tailpipe CO2, with the target drive cycle being the New European Drive Cycle In order to this vehicle modelling was employed to set part load operating points representative of a target vehicle and to provide weighting factors for these points The engine was sized by using the fuel consumption improvement targets while a series of specification steps, designed to ensure that the required full-load performance and driveability could be achieved, was followed The intake port in particular was the subject of much effort, and data is presented showing its performance versus a current stateof-the-art production design The use of a test-cell-based charging system, while the engine-mounted charging system was being developed and characterized in parallel, is discussed This approach allowed development of the base engine and combustion system without the complicating effects of the charging system performance coming into play Finally, data is presented comparing the performance of the engine in this guise with that when the engine-driven turbocharger was used, showing that the peak torque and power targets have already been met ABBREVIATIONS ATDC BDC BMEP BTDC CAHU After top dead centre Bottom dead centre Brake mean effective pressure Before top dead centre Combustion air handling unit _ © The author(s) and/or their employer(s), 2013 CPS DCVCP DF DI EAT EGR IEM IMEP JLR MOP NA NEDC pbrake pind PCP SI TDC UB VSwept WCEM mech Cam profile switching Dual continuously-variable camshaft phasing Downsizing factor Direct injection Exhaust after treatment Exhaust gas recirculation Integrated exhaust manifold Indicated mean effective pressure Jaguar Land Rover Maximum opening point Naturally-aspirated New European Drive Cycle Brake mean effective pressure Gross indicated mean effective pressure Peak cylinder pressure Spark-ignition Top dead centre Ultraboost (Ultra Boost for Economy) Swept volume Water-cooled exhaust manifold Mechanical efficiency INTRODUCTION 1.1 Spark-ignition engine downsizing Spark-ignition (SI) engine downsizing is now established as a ‘megatrend’ in the automotive industry, providing as it does an affordable solution to the twin issues of reducing tailpipe CO2 emissions and improving fuel economy while providing improved driveability from gasoline engines The ‘downsizing factor’ is here defined to be DF  VSweptNA  VSweptDownsized VSweptNA , Eqn where DF is the downsizing factor, VSweptNA is the swept volume of a naturallyaspirated engine of a given power output and VSweptDownsized is the swept volume of a similarly-powerful downsized alternative To the OEM the attractions of a downsizing strategy include that gasoline engine technology is very cost-effective to produce versus diesel engines (especially when the costs of the exhaust after treatment (EAT) system are included), that there are still significant efficiency gains to be made due to the losses associated with the 4stroke Otto cycle, and that pursuing the technology does not entail investing in completely new production facilities (as would be required by a quantum shift to electric or fuel-cell vehicles, for example) The advantages of downsizing a 4-stroke spark-ignition (SI) engine stem chiefly from shifting the operating points used in the engine map for any given flywheel torque, so that the throttle is wider-open to the benefit of reduced pumping losses At the same time, the mechanical efficiency increases, this being defined as mech  where pbrake , pind Eqn mech is the mechanical efficiency, pbrake is the brake mean effective pressure (BMEP) and pind is the gross indicated mean effective pressure (IMEP) [1] Thermal losses also improve and, in the case of downsizing and ‘decylindering’ from a Vee-configuration engine to an in-line one, crevice volume losses can be markedly reduced and there are potentially significant bill of materials (BOM) and manufacturing cost savings, too These savings can help to offset the additive technologies required to recover the power output, because some means of increasing specific output has to be provided to retain installed power in a vehicle This is normally done by pressure charging the engine, with turbocharging generally being favoured because it allows some exhaust gas energy recovery There are significant synergies with other commonplace technologies such as direct injection (DI) and camshaft phasing devices, too [2] To date production downsized engines have generally been configured with a DF in the region of approximately 40%, with one research engine shown with this value at 50% [3] Consequently the Ultraboost project was formed with the major tasks of specifying, designing, building and operating an engine with a minimum of 60% downsizing factor Through the results obtained it was intended to establish whether 60% is a practical limit for the approach or whether there would be benefit in further downsizing, and that such a downsized engine could in itself provide a route to a 35% reduction in vehicle tailpipe CO2 (importantly, without the use of hybridization other than a Stop/Start system) Consequently a primary aim of the project was to achieve the power and torque curves of the Jaguar Land Rover 5.0 litre AJ133 naturally-aspirated V8 engine with a pressure-charged engine of approximately 2.0 litre capacity These curves are reproduced in Figure 1, together with the associated BMEP values required from the downsized engine at peak torque, peak power and 1000 rpm The CO2 emissions and fuel consumption improvement was to be demonstrated by using dynamometer measurements and vehicle modelling, with the target drive cycle being the New European Drive Cycle (NEDC) 515 Nm at 3500 rpm 283 kW / 380 bhp at 6500 rpm 415 Nm at 6500 rpm 400 Nm at 1000 rpm 26.1 bar 25.1 bar Corrected Power / [kW] Corrected Torque / [Nm] 32.4 bar Engine Speed / [rpm] Fig 1: Target power and torque curves and selected associated BMEPs for a 2.0 litre engine 1.2 Ultraboost project partners The Ultraboost project comprised eight partners, Jaguar Land Rover (JLR), GE Precision Engineering, Lotus Engineering, CD-adapco, Shell, the University of Bath, Imperial College London and the University of Leeds It started in September 2010 with a duration of three years JLR is the lead partner, with responsibility for engine build, general procurement, engine-mounted charging system integration and project management GE Precision provided engine design and machining capabilities as well as background knowledge on the design of high-specific-output racing engines Lotus Engineering provided a dedicated engine management system (EMS), 1-D modelling and knowhow on pressure-charged engines, and support for engine testing All engine testing was to be conducted at the University of Bath, where dedicated boosting and cooled exhaust gas recirculation (EGR) rigs were used for initial testing of the demonstrator engine CD-adapco supported the design process with steady-state and transient CFD analysis primarily in order to support intake port design, which is discussed in detail below Shell provided test fuels and autoignition know-how Imperial College specified the charging system components, with support from both JLR and Lotus, and tested them in order accurately to characterize them so that the 1-D model was as robust as possible Finally, the University of Leeds developed their autoignition model to assist with the 1-D modelling process This project structure was reviewed in an earlier publication [4], where some of the background detail to the establishment of the projects targets was also discussed 1.3 Phases of the Ultraboost project The project was split into several parts In Phase 1, a production JLR 5.0 litre AJ133 V8 engine was commissioned on the test bed at the University of Bath using the Denso engine management system (EMS) then used for production This was then replaced by the Lotus EMS, which was demonstrated to be capable of controlling the engine and giving exactly the same performance at full and part load, including matching the steady-state fuel consumption of the production engine and Denso EMS combination to ≤ 0.5% This phase therefore set the fuel consumption benchmarks for the project’s downsized engine design and proved the capability of the Lotus EMS when controlling a direct-injection engine with many high-technology features, including multiple-injection strategies In parallel with the Phase engine test work, Phase specified, designed and procured the core Ultraboost engine (known as UB100) To this the pooled knowledge of all the parties was used, resulting in a current industry best-practice high-BMEP engine with some additional novel features The Phase test programme utilized a test bed combustion air handling unit (CAHU) and a speciallydesigned EGR pump rig It was primarily intended to prove out the efficacy of the newly-developed combustion system The testing portion of this phase also permitted fuel testing to be undertaken without the complicating effects of an engine-driven charging system, although this important subsystem would also be specified, modeled, procured and validated in a parallel work stream within this phase Phase was intended to comprise any necessary redesign of the UB100 engine coupled with mounting the engine-driven charging system The engine was then to be known as UB200 The present paper discusses some of the engine-specific technologies configured and tested in Phase 2; the results of the fuels testing and of the Phase engine will be reported separately in later publications Ultimately, the level of achievement of the project targets will be demonstrated by a combination of direct measurement (power, torque, driveability etc.) and modelling (by the application of gathered minimap fuel consumption data to a vehicle performance model, this being necessary since the baseline AJ133 engine is no longer fitted to the target vehicle) ENGINE DESIGN 2.1 Derivation of engine swept volume At the start of the project the actual swept volume was unconstrained In order to establish this parameter, vehicle modelling was employed to set part load operating points representative of the target vehicle and to provide weighting factors for these points The engine swept volume was then determined by using the fuel consumption improvement targets and a series of specification steps designed to ensure that the required full-load performance and driveability could be achieved; these were informed by previous work undertaken by JLR [5] The engine was then designed in conjunction with 1-D modelling which helped to combine the various technology packages of the project These included an advanced charging system (discussed in a previous paper [6]) and a valvetrain system with the necessary variability to deliver target performance The modelling also helped to determine the flow characteristics required of the intake port Ultimately this had stretch targets set for it to ensure the necessary charge motion for fuel mixing and to help suppress knock, and was subjected to a full transient CFD analysis This is discussed later In Phases and of the project the 1-D model was also used to guide testing, primarily to set intake and exhaust system boundary conditions to make them representative of what could be expected of the real charging system It was also used to calculate the extra torque that the core engine would have to produce for the results to be representative of the combined engine and charging system It was also used to help to explain trends in the results 2.2 General engine specification From this preliminary work the engine was specified as shown in Table The undersquare nature of the engine is readily apparent; this helps to shorten the flame travel to the benefit of knock and to reduce thermal losses It also possibly benefits preignition, the causes of which are believed to include oil being ejected from the piston top land, and reducing the bore diameter directly reduces the top land area [7,8] Effectively, the engine is one bank1 of a heavily-modified AJ133 V8, with a new bore and stroke, a flat-plane crankshaft and attendant firing order This approach was taken because the bearings and scantlings of the AJ133 engine would easily be capable of handling the performance A CAD image of the UB100 engine, fitted with the original log-type exhaust manifold, is shown in Figure The engine management system was configured to be capable of controlling the many functions on the engine as detailed in Table and ultimately also the selected charging system components, including the supercharger clutch and bypass system [6] The engine has been designed to withstand a peak cylinder pressure (PCP) of 130 bar, with known further countermeasures should it be considered advantageous to increase this to a higher level (for instance, when investigating high-octane fuels) The active bank is the A Bank (on the right-hand side of the engine) The aluminium alloy piston itself is safe to a PCP of 145 bar for the sort of duty cycle a research engine is typically used for Table 1: Ultraboost UB100 engine specification General architecture 4-cylinder in-line with valves per cylinder and double overhead camshafts Construction All-aluminium AJ133 cylinder block converted to single-bank operation on the A Bank (right-hand side) Siamesed liner pack to facilitate reduced bore diameter Dedicated cylinder head Bore 83 mm Stroke 92 mm Swept volume 1991 cc Firing order 1-3-4-2 Combustion system Pent-roof combustion chamber with asymmetric central direct injection and spark plug High-tumble intake ports Auxiliary port-fuel injection Possible second spark plug position in an under-intake-port location Compression ratio 9.0:1 Valve gear Chain-driven double overhead camshafts with fast-acting dual continuously-variable camshaft phasers (DCVCP) Cam profile switching (CPS) tappets on inlet and exhaust Fig 2: CAD images of assembled UB100 engine, as originally tested with a log-type exhaust manifold; note coolant bypass pipe for the absent B Bank cylinder head 2.3 Intake port design and flow-rig performance In order to achieve the necessary air motion and mixture preparation in DISI engines there has been a general evolution of high-tumble intake ports; this has only been made possible by the simultaneous adoption of pressure charging to overcome the flow loss generally associated with this move It is worth noting that under-port placement of the injector had a symbiotic relationship with this evolution emissions performance by shortening service replacement schedule for high cost injectors i.e the customer pays, is likely not a convincing long-term marketing approach to a prudent, quality-focused VM Periodic cleaning of injector deposits could be explored; even using lasers in the combustion chamber has been suggested to help mitigate this issue [24] PN prevention based stage emissions compliance strategies will need to overcome this issue Retarded Spark to Suppress PM Simply, by delaying the spark, the mixture preparation period is increased enabling droplet evaporation Use at high speed coupled with a move to early injection could be important as injection quantity is large (i.e stretching the mix-time from both ends) Retarding spark however is wasteful regards CO2 and usually avoided, except during cold start where it is employed to enable fast-catalyst light-off Secondary-Air This means late cycle secondary oxygen injection in the combustion chamber, when the temperature is high enough to allow some soot oxidation, but not high enough to promote NOx production (explored for diesel) [25] Limitations would include cost of the implementation of the air introduction medium e.g solenoid and additional control complexity Conceivably combustion systems that utilise truly dynamic valve trains (solenoid, pneumatic etc.) could potentially use valve eventing in combination with forced induction to provide this secondary air Increase Fuel Temperature, Oil Film Minimisation, Induction Filtration and Oil Ventilation Increasing fuel temperature can improve the likelihood of fuel vaporisation on injection and reduces the quantity of fuel which can impinge upon the piston top and walls As well as potential safety implications, increased tip deposit formation would need to be overcome Efficient induction filtration and prevention of oil vapour and thin oil can also help keep down particulate formed in the DISI (use of modern thin oils is important) Cold Ambient Conditions Cold temperatures represent a challenge to air quality and PN compliance, as temperature decreases, the viscosity of the lube oil raises and this exacerbates the effort required to pump the oil through the engine Subsequently we see high mechanical losses, increased fuel consumption and an increase in particle formation Cold ambient temperatures represent a specific challenge to all formation-based PN compliance combustion concepts Cold start at low ambient for ethanol containing fuels such as European market fuel E10, can be particularly difficult due to vaporisation issues and will need proving Consequently, gasoline particulate filters (GPFs) may represent the only robust solution for particle filtration at low ambient temperatures This is because GPF filtration efficiency does not degrade at cold ambient temperatures [26] METHODS TO REMOVE PARTICULATE (AFTERTREATMENT) The three-way catalyst (TWC) has been shown to have a limited effect on particulate emission reduction; the TWC tends to reduce the quantity of small particles that are thought to be SOF in nature [27] The effectiveness of a physical filter is dependent upon both the particle geometry and filter geometry [28] 239 There are several methods by which particulate can be physically filtered; figure depicts the mechanisms, detailed explanation can be found in [12] Mechanisms b and f are thought to have low significance regards vehicle particulate Figure Transport and deposition mechanisms: (a) interception (b) sedimentation (c) inertial impaction (d) diffusion (e) thermophoresis (f) electrostatic attraction, adapted from [12] The filtration requirements for soot reduction (mass and number) from DISI exhaust gas are similar to DDI, but could be optimised for DISI due to the altered soot spectrum, (the particle peak emission for DISI is shifted towards smaller particles compared to DDI, meaning that a diffusion biased filtration system would see increased filtration efficiency, and interception will become a lesser factor vs DDI filtration) [29, 30, 31, 32] Gasoline Particulate Filters GPFs are physical filtration devices that can remove soot (solid PM) from gasoline engines exhaust; there are several types of filtration device, it is believed only three types have seen major use to-date in automotive DDI applications Figure Wall-flow Monolith Filter [33] Wall-Flow Ceramic Monolith Filter The most commonly used filter type for DDI emissions compliance is the wall flow ceramic monolith DPF (figure 7) Exhaust gas containing particulate (arrows) enters the filter from the left The cells of the filter are plugged at the downstream end of 240 the filter so the exhaust cannot exit the cell directly The exhaust gas must instead pass through the porous walls of the filter channels The solid particulate now becomes trapped within the porous walls, and the inlet channel surface on the upstream side of the filter PM will gradually fill the DPF, so it has to be removed, or it will cause blockage This is achieved by oxidation of the soot in a regeneration process There are several available routes to achieve this, either actively (by raising local temperatures and possibly oxygen) and passively (at the natural exhaust temperatures) Active regeneration requires periodic input of energy, resulting in reduced FE While passive regeneration does not Gasoline exhaust gas temperatures are significantly higher than equivalent diesel conditions This property coupled with potential frequent use of highly oxygenated exhaust provided by deceleration-fuel shut-off (DFSO) or valve-event related, such as blow-through, enable increased chance of continuous regeneration during normal operation for a DISI vs DDI Natural high exhaust gas temperatures for the DISI therefore reduce the need for either a fuel/durability costly regeneration strategy, or for expensive high platinum based catalysts (akin to those employed in some diesel systems) which promote NO + 1/2O2 ↔ NO2 i.e production of a powerful oxidant to lower the temperature at which soot oxidation occurs (O2 >500°C vs NO2>250°C) Research [34, 35] has highlighted that gasoline wall flow filter filtration efficiency is at lower levels than comparative diesel systems Continuous regeneration of the GPF and hence low soot-cake filtration in gasoline exhaust is likely a contributor GPFs can be coated with a 3-way catalyst washcoat at an on-cost which is increased vs flow-through substrates due to coating complexity washcoat quantity and high comparative scrapage rate Adding a washcoat to a GPF will have to be assessed for gain in filtration efficiency and effect on soot oxidation temperatures for regeneration (basically, is the on-cost and extra exhaust backpressure (EBP) worth the gains in regeneration enable and filtration efficiency) [36, 37] Washcoating of a filter can be considered an open-ended subject in terms of complex cost-vs benefit studies, applications will need to be considered on a case by case basis There are many tools for the catalyst designer that have been developed for flow-through and within diesel filter technology implementation An evident example would be coating the filter with rare earth oxide metals such as ceria to momentarily store oxygen and release it later, i.e potentially an aid to regeneration for low oxygen conditions Washcoat based hydrocarbon (HC) storage and release components like zeolites might also be worth considering for low-temperature underbody filters (e.g severe tier emissions with lightoff issues, where a GPF is implemented for high-soot applications, gives potential to store HC during cold-start Figure Low backpressure wall(as this is not really possible at the flow GPF, printed courtesy of close-coupled TWC due to short life of Sumitomo Chemical the zeolites under intense thermal aging in the close-coupled position)) Studies have revealed success in the removal of DISI particulate for the wall-flow GPF concept to proposed S6(c) PN limits Filtration efficiency has been shown to be 241 linked with exhaust temperature and slightly lower than DDI Pre-test preparation of the filter has also been shown to be important (an improved explanation for this phenomenon should be sought) [38] New wall-flow-through concepts are under development, designed for the GPF e.g chiral inlet to outlet channel to enable gas flow and soot storage see figure Control limitations/Challenges to Implement GPFs It is possible to overload a wall-flow GPF with particulate, filters have a limit that depends on the structure and material limitations (soot mass limit (SML)) For a robust GPF system all operational conditions and driving styles need to be covered during GPF operation Extreme levels of soot in a GPF can be a problem, as when the soot burns the exotherm can be extreme, causing the filter to crack or potentially melt or burn Exceeding thermal limitations will likely mean emissions non-compliance and a melted or burnt filter VMs will need to ensure that emissions compliance is maintained out to FUL (defined as 160k km in Europe) Control systems will need to be designed and implemented to protect the GPF and maintain operation out to FUL Sensors and complex control algorithms may need to be developed to implement certain GPF concepts (though it is anticipated that the challenge should be reduced vs the task for DDI at stages and 5) Applications will likely not all require the same amount of control e.g low PN emitters vs high, cold location of filter vs hot-location, or indeed fundamental exhausted heat will all play a role Specific conditions thought to be control challenges: low-temperature running, repeated short journeys and extended idle Basically conditions where temperature and or oxygen are deficient, and continuous regeneration of the filter might be hampered There are however some, (improved vs DDI) tools available via calibration and control that could be utilised to overcome limitations e.g periodic use of spark retard, changes to idle set-point to help raise temperatures and potentially valve overlap in a dynamic valve-train to enable oxygen Increased EBP with GPF Introduction Filters represent an obstruction to flow within the exhaust system, this trait is unavoidable for physical filtration (cure) based routes to particulate emissions compliance EBP is a hindrance to flow, most especially at high power-output, where flow hindrance can have a direct impact on the peak-achievable power target for a power-plant EBP can have direct negative FE impact FE impact of filter implementation may not translate directly to low medium exhaust flow rate conditions of an NEDC based FE figure (due to limited operation at higher exhaust mass air-flow (MAF) conditions of the NEDC) However, under real driving conditions e.g autobahn and highway cruise, where exhaust MAF is high, a system with GPF implemented could have significant impact upon real world FE and hence climate CO2 It is probably not the intent of S6 particulate legislation to increase CO2, as S6 legislation attempts to make account of real-driving emissions (RDE at S6 (c) vs previous legislation based solely on NEDC) EBP could be reduced by optimising the design of the filter; there are simple intuitive design options available The simplest are based on cross-sectional open frontal area (OFA) which are also true for flow through substrates e.g increase bore of the particulate filter, (enabled aftertreatment package is paramount) OFA positive designs like hex structure which increase the effective hydraulic diameter per channel (figure is an asymmetric hex), decrease cell density and wall thickness (math is outlined in [39]) Secondly there are ways to reduce frictional losses within the wall as the gas passes through, e.g extend or shorten the filter (package dependant) As cell wall length increases there is a factor increasing EBP due to axial friction In the case of a low porosity wall, the friction due to gas 242 passing through the y-axis can be dominant and can severely impact the filters ability to flow (cause high EBP); in this case lengthening the filter may yield EBP benefit, as there is extended porous area through which the gas can pass (a coated filter is likely to be less porous) In the case of a highly porous GPF a short length filter may see improved EBP due to the dominance of axial friction e.g a highly porous thin-wall non-coated design whose inception was likely directed specifically towards gasoline particulate filtration A tertiary friction parameter is also seen affected by mean pore size By reducing the wall thickness, the wall loss friction factor is reduced Primary filter function factors, as spelt out for DPF [40] remain true for GPF, therefore, EBP optimisation is a compromise, as moving towards low EBPs can reduce filter function Re-calling that the filter has to provide sufficient filtration efficiency to meet the legislation and have sufficient mechanical and thermal integrity to survive use out to FUL Ash and Poison Capacity of a GPF The filter size is important regards durability, as the filter needs to be sized with sufficient volume to store non-combustible ash material and be robust to poisons, to maintain all filter function out to FUL mileage These parameters can be complex and difficult to determine for VMs without extensive testing and will either need significant study or costly safety margin (e.g over-engineer by over-sizing the GPF) for robustness VMs have the advantage of recent DDI development work for stage and as a development resource Other options include costly customer services of the filter (employed by some VMs for DDI at S5) using de-mountable filters at mileage intervals The filter can either be replaced or re-conditioned to removed poisons and ash Metallic Fleece (Partial Flow Filter) Sometimes coined the soot capacitor, partial flow filters (PFFs) are the second most successful to-date in terms of serial production for diesel applications, they are particularly popular for retro-fit Metallic filter manufacturers include; Emitec, DCL International Inc and Ecocat Oy Low filtration efficiency (45% PN reduction on NEDC [41] means that metallic partial flow filters can only see potential for low soot producing applications, or when used in combination with a soot prevention based strategy which can push low soot generation i.e a prevention cure hybrid Soot filtration in this filter type is based on obstructions to flow forcing gas-borne particulate via diffusion towards a sintered porous metallic gauze/fleece Particulate intercepts with the fleece and is held until the local conditions become sufficient to facilitate soot oxidation The key aspect of this filter type is the open (but tortuous) path inlet to outlet This feature makes the PFF design robust to overload, i.e flow is possible even when the filter is saturated with soot, this feature means there is only a limited increase in EBP between an unloaded and loaded filter This advantage could be employed to overcome strange/unusual operational conditions where PM might not continuously regenerate e.g cold conditions, continuous cold short journeys or extended cold Figure Partial-flow-filter from Emitec (left); idle (potentially reducing structure and principle of coated filter the control and sensor-set operation of a partial-flow-filter (right), required to implement) adapted from [42, 43, 44] 243 Figure shows a prominent PFF design from Emitec The filtration mechanisms used in a PFF is thought to be particularly efficient for small diffusion impacted particulate, interestingly this size range is thought to be prevalent for the DISI engine i.e filtration efficiency is likely increased vs diesel for comparative PM emissions for equivalent length filters [45, 46, 47, 48] One limiting factor could be that soot oxidation rate for a PFF has been shown to be dependent on the filter soot loading, meaning an improved soot oxidation is permitted when the filter is heavily loaded with soot Potentially for an oxidationinhibited application this could be an Achilles due to the lower DISI soot mass emission vs DDI [49, 50, 51] PFFs might be more appropriate for applications where large PN reduction is not required and EBP effect is of large importance e.g high power variants The wall thickness and hence OFA restriction of a metallic filter is reduced vs ceramic monolith Cost and filtration efficiency are likely the key drawbacks Additionally there may be difficulty as regards emissions effectiveness at the PM saturation point (when the gauze/fleece has reached capacity, soot will pass through the filter) Emissions compliance at PM saturation point (extreme low occurrence event for most drivers e.g low ambient temperature extended idle) however, these conditions could be considered safety or damage limitation critical and therefore beyond the scope of reasonable legislation Metal Foam Particulate Filters These designs have seen relatively niche use to-date Downstream flow uniformity is increased vs equivalent monolith wall flow (this is thought to be of benefit at high exhaust mass air-flow vs wall-flow [52] for systems with downstream aftertreatment, which is not so likely for DISI applications) This improvement in flow uniformity likely has an EBP cost for equivalent sized filters, foam vs wall-flow monolith Figure 10 SEM pictures of metal foams with various pore sizes (a) 450microns, (b) 800microns (c) Radial-flow concept, (d) coiling of foam sheets around a porous tube, (e) Coiled filter without external shell Adapted from [53] 244 For the design detailed in figure 10c, a longer filter gives lower EBP; there may be a possibility to implement this for designs within the package of existing underbody resonators i.e as a resonator substitute, 300-400mm resonators being typical in gasoline applications to suppress standing waves One advantage of this technology is that increased length filters are possible; ceramic wall-flow designs have a limit on (D:L) ratio, this might help if the vehicle package places a severe restriction on the contour of the aftertreatment e.g package restriction due to additional features including hybridisation, crash, interior ergonomics etc One potential negative is potential soot blow-off when hydrodynamic forces exceed the particle adhesive force to the foam, (move to RDE will mean an increased MAF operational window is required and therefore possibly increased blow-off risk) 4-Way Wall-Flow Catalytic Converter A 4-way catalytic converter would be an all-in-one catalyst and particulate filter performing normal 3-way function i.e HC and CO oxidation and NOx reduction, combined with elemental C oxidation (the 4th way) Conceivably a way converter could be based on all of the GPF substrates previously outlined The filter would be coated with washcoat containing platinum group metals (PGM) and rare earth oxide (REO) in a similar manner to the TWC, though the coating method will need aligning with the filter type e.g flow-through monolith channel coating might cause excessive EBPs so a coating that targets deposition within the porous structure of the wall may be preferred Coaters have various methods that can be used to hone the desired physical fixation of washcoat/components and the physical position of active material within a substrate structure, these methods include: impregnation (primary method for auto-catalyst), incipient wetness, electrostatic adsorption, ion exchange; Heck and Farrauto [54] give a useful guide to these processes 4-way likely represents a new exploratory challenge to washcoat manufacturers Potentially a 4-way wall-flow could give benefit regards EBP compared to series TWC + GPF; however, since oxidation and reductions performed over a catalytic converter are reliant on gas-active site physical interaction, a 4-way wall-flow solution may require extreme PGM loadings to give equivalent efficiencies to those of the flow-through TWC The 4-way converter might see exploration for applications that have severe package and weight restriction limitations for e.g premium/performance leaders with reduced cost control expectation Challenges that will need to be overcome; difficulty in achieving PGM utilisation efficiency vs high efficiencies realised by flow-through catalysts, resilience to aging in hot closecoupled position (wall-flow vs flow-through) may increase PGM required for the aged condition (bearing in mind that the catalyst has to perform out to FUL) Prevention vs Cure Table contrasts some of the strategies that could be used to achieve stage particulate legislation, the column inputs are subjective and open to interpretation It should also be noted that many of the prevention techniques really describe a part of a complex system and it is thought likely that prevention techniques would more realistically be combined to yield a holistic particulate prevention route VMs will likely use elements as required to achieve incremental particulate legislation standards Each VM will have a unique DI combustion system and control mechanism, the scale of the PN compliance challenge will also vary by VM Successful VMs likely have clear brand DNA expectations that will push development in a particular direction The time available before legislative introduction is finite and may limit 245 the options available It is therefore believed that VMs will take various paths to PN compliance at stage (i.e there is no clear one size fits all solution) Table Prevention vs cure pros and cons Particulate Particulate Compliance Compliance NEDC RDE Time Fuel Ambient Cold COST to Economy Conditions Implement (CO2) Peak Control Power Complexity LOW PARTICULATE EMITTING VEHICLE Chamber Re-design Prevention High Injection Pressure OOSI Injection Strategy PFDI Retarded Spark Secondary-Air Cure Wall-flow Monolith Partial Flow Filter Metal Foam Filter 4-Way Catalytic Converter ×       × × × ? × ?  -         + ++ + + LOW HIGH HIGH HIGH HIGH HIGH LOW MED HIGH MED V.LOW V.LOW HIGH HIGH MED MED MED MED MED MED HIGH MED -0 0 0 0 0 - - - - 0 0 0 0 - - - - HIGH PARTICULATE EMITTING VEHICLE × × × ? Secondary-Air ×       Wall-flow Monolith   Partial Flow Filter ×   × ×  Chamber re-design Prevention Injection Pressure OOSI Injection Strategy PFDI Cure Retarded Spark Metal Foam Filter 4-Way Catalytic Converter × ? ? + ++ + + LOW HIGH HIGH HIGH HIGH HIGH LOW MED HIGH MED V.LOW V.LOW HIGH HIGH MED MED MED MED MED MED HIGH MED CONCLUSIONS DISI powered vehicles represent an important part of many VMs portfolios, due to thermal efficiency and potential regarding power-to-weight and enhancement of powertrain downsizing DISI engines create a relatively high (vs PFI) amount of 246 particulate, problem areas include high speed running, short time period, under high load, high fuel quantity, with high in-cylinder temperature New PN Stage European legislation will be applied to the DISI architecture and it will be introduced in two stages, a limit on particle count at 6.0 × 1012 in 2014, moving to 6.0 × 1011 in 2017 VMs will need to make changes to meet this new legislation, the process will be complex and expensive Prevention High temperature and oxygen are the two most important factors to prevent and reduce particulate Strategy and methods that promote this can help mitigate particulate formation Physical fundamental design parameters such as; combustion chamber design, injector design, injector tips (and coatings) and injection pressure can be optimised to reduce particulate formation Injections can be split and optimised to fully utilise the combustion chamber (chase the piston without touching the piston top and cylinder walls) across the full speed load engine map Injector deposits are anticipated to be a challenge which will require resolution for compliance out to FUL Spark retard could be utilised, though is not efficient New dynamic valvetrains that could enable post particulate inception removal techniques, (secondary air) could be employed (but also bring new complexity and new poorly known control challenges) Cold ambient temperatures represent a specific challenge to all formation based PN compliance combustion concepts which will be difficult to solve It is believed that multiple prevention techniques combined with hardware upgrades would need to be explored simultaneously to achieve robust particulate compliance at S6 Cure Gasoline particulate filters (GPFs) are physical filtration devices that can remove soot (solid PM) from a gasoline engines emission, there are several types The wallflow monolith is a proven technology (S4 and S5 DDI) for engine out particulate at an order of magnitude above the emission of a DISI DISI exhaust gas temperatures are significantly higher than DDI, when coupled with potential frequent use of high oxygen conditions provided by DFSO or valve event related, such as blow-through, increased chance of continuous regeneration during normal operation is enabled for a DISI vs DDI PFFs might be a good option for low soot emitting vehicles, they may also require less control provision vs wall-flow GPF Metallic foam particulate filters may be suited to low-to-moderate particulate emitting DISI applications as a direct resonator replacement A 4-way catalytic converter is an all-in-one catalyst and particulate filter 4-way could give benefit regards EBP compared to series TWC + GPF, but is expected to come at a cost due to relative high PGM efficiency of a flow-through catalyst vs wall flow filter 4-way may be of interest to VMs with rigorous package, weight and performance requirements The challenge to introduce filter technology to the DISI architecture is expected to be achievable, but at attribute penalty (peak power, cycle and most especially offcycle CO2) attribute detriment can be reduced by increasing the volume of aftertreatment package and moving to higher cost internal components Surprisingly the old adage ‘prevention is better than cure’ may actually not ring true when applied to robust DISI particulate compliance at stage 247 REFERENCES 10 11 12 13 14 15 16 17 18 19 20 21 248 EC, 2012 Regulation No 459/2012 This is the regulation for stage PN, in stages (b) at 6.0×1012 and (c) at 6.0 × 1011 Review: Epidemiological Basis for Particulate Air Pollution Health Standards C Arden Pope, III Aerosol Science and Technology 32, 4-14 (2000) Air pollution and health (review): Bert Brunekreef Stephen T Holgate Lancet 360: 1233–42 (2002) Public-health impact of outdoor and traffic-related air pollution: a European assessment N Künzli, R Kaiser, S Medina, M Studnicka, O Chanel, P Filliger, M Herry, F Horak Jr, V Puybonnieux-Texier, P Quénel, J Schneider, R Seethaler, J-C Vergnaud, H Sommer Lancet 356, 795–801 (2000) Health effects of particulate air pollution: A review of epidemiological evidence, Regina Rücker, Alexandra Schneider, Susanne Breitner, Josef Cyrys, and Annette Peters, Inhalation Toxicology, 23(10), 555–592 (2011) Cost effectiveness of particulate filter installation on Direct Injection Gasoline vehicles: Athanasios Mamakos, Nikolaus Steininger, Giorgio Martini, Panagiota Dilara, Yannis Drossinos Atmospheric Environment 77, 16-23 (2013) Gasoline direct injection spray simulation: Rossella Rotondi, Gino Bella, International Journal of Thermal Sciences 45, 168–179 (2006) French Patent, Brevet D’invention (1902) No 322.904 The History of Scania Vehicles 1891-1991: ISBN 91-7886-074-1 Automotive spark-ignited direct-injection gasoline engines: F Zhao, M.-C Laia, D.L Harrington, Progress in Energy and Combustion Science 25, 437– 562 (1999) Exhaust emissions control by the Ford Programmed Combustion process— PROCO: Simko A, Choma MA, Repko LL, SAE Technical Paper, No 720052, (1972) Particulate Emissions from Vehicles: Peter Eastwood, ISBN: 9780470724552 ENGINES AND NANOPARTICLES: A REVIEW David B Kittelson, J Aerosol Sci 29, No 5/6, 575-588, (1998) Comprehensive particle characterization of modern gasoline and diesel passenger cars at low ambient temperatures: Urs Mathis, Martin Mohr and Anna-Maria Forss, Atmospheric Environment 39, 107–117 (2005) State of California Air Resources Board, Preliminary Discussion Paper – Proposed Amendments to California’s Low-Emission Vehicle Regulations – Particulate Matter Mass, Ultrafine Solid Particle Number, and Black Carbon Emissions: CARB, Sacramento, CA, (2010) Leidenfrost, J G De Aquae Communis Nonnullis Qualitatibus Tractatus Duisburg, (1756) Stabilization of Leidenfrost vapour layer by textured superhydrophobic surfaces: Ivan U Vakarelski, Neelesh A Patankar, Jeremy O Marston, Derek Y C Chan, and Sigurdur T Thoroddsen, Nature 489, 274–277 (2012) Automotive Gasoline Direct-Injection Engines: Fuquan Zhao, David L Harrington and Ming-Chia Lai, (2002) ISBN: 0-7680-0882-4 Piezo direct injector accessed August 2013 http://forums.tdiclub.com/showthread.php?t=347581andpage=5 Diffusion Combustion Phenomena in GDi Engines caused by Injection Process: Axel Berndorfer, Stephan Breuer, Walter Piock and Paul Von Bacho, SAE 201301-0261 Internal and near nozzle flow characteristics in an enlarged model of an outwards opening pintle-type gasoline injector: J M Nouri, E Abo-Serie, A Marchi, N Mitroglou and C Arcoumanis Third International Conference on Optical and Laser Diagnostics IOP Publishing Journal of Physics: Conference Series 85 012035 (2007) 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Strategies for Particle Emissions Reduction from GDI Engines: Youngjae Kim, Yongha Kim, SangYoul Jun, Kang Hun Lee, SeungHyun Rew and Donghyeon Lee Simsoo Park, SAE 2013-01-1556 Diffusion Combustion Phenomena in GDi Engines caused by Injection Process: Axel Berndorfer, Stephan Breuer, Walter Piock and Paul Von Bacho, SAE 201301-0261 Laser cleaning of optical windows in internal combustion engines: Helmut Ranner, Pratush Kumar Tewari, Heinrich Kofler, Maximilian Lackner, Ernst Wintner, Avinash Kumar Agarwal, Franz Winter, Opt Eng 46(10), 104301 (2007) Modelling the Effects of Late Cycle Oxygen Enrichment on Diesel Engine Combustion and Emissions: D K Mather, D E Foster, R B Poola, D E Longman, A Chanda and T J Vachon, SAE 2002-01-1158 Impact of Ambient Temperature on Gaseous and Particle Emissions from a Direct Injection Gasoline Vehicle and its Implications on Particle Filtration: Tak W Chan and Eric Meloche, Joseph Kubsh and Rasto Brezny, Deborah Rosenblatt and Greg Rideout, SAE 2013-01-0527 The Effect of a Three-Way Catalytic Converter on Particulate Matter from a Gasoline Direct-Injection Engine During Cold- Start: Ian Whelan, David Timoney, William Smith and Stephen Samuel, SAE 2013-01-1305 Examination of the oxidation behavior of biodiesel soot: J Song, M Alam, A.L Boehman, U Kim, Combust Flame 146, 589–604 (2006) Soot structure in SIDI engines: T.L Barone, J.M.E Storey, A.D Youngquist, J.P Szybist, Atmos Environ 49, 268–274 (2012) TEM analysis of volatile nanoparticles from particle trap equipped diesel and direct-injection spark-ignition vehicles: U Mathis, R Kaegi, M Mohr, R Zenobi, Atmos Environ 38, 4347–4355 (2004) Physical and chemical characterization of SIDI engine particulates: Chethan K Gaddam, Randy L Vander Wal Combustion and Flame 160, 2517–2528 (2013) A 2-D computational model describing the flow and filtration characteristics of a ceramic diesel particulate trap: Opris, Cornelius N.; Johnson, John H., SAE 980545 Wall-flow Monolith Particulate filter Accessed September 13 http://www.autodiagnos.com Evaluation of a Gasoline Particulate Filter to Reduce Particle Emissions from a Gasoline Direct Injection Vehicle: Tak W Chan and Eric Meloche, Joseph Kubsh, Deborah Rosenblatt, Rasto Brezny and Greg Rideout, SAE 2012-011727 Experimental Evaluation of a Wall-Flow Filter for Gasoline Engine Particulate Emission Control: Rubino, L, Crane, R I., Shrimpton, J S., Arcoumanis, C, SAE 2001-24-0072 Application of Catalyzed Gasoline Particulate Filters to GDI Vehicles: Joerg Michael Richter, Raoul Klingmann, Stephanie Spiess and Ka-Fai Wong, SAE 2012-01-1244 Simultaneous Reduction of PM, HC, CO and NOx Emissions from a GDI Engine: Reggie Zhan, Scott T Eakle and Phillip Weber, SAE 2010-01-0365 Design and Durability of Standard and Advanced Ceramic Substrates: Gulati, S T., SAE 2001-26-0011 Advanced Ceramic Wall Flow Filter for Reduction of Particulate Number Emission of Direct Injection Gasoline Engines: Yoshitaka Ito, Takehide Shimoda, Takashi Aoki, Yukinari Shibagaki, Kazuya Yuuki, Hirofumi, Sakamoto and Claus Vogt, Tasuku Matsumoto, Wolfgang Heuss, Philipp Kattouah, Mikio Makino and Kyohei Kato, SAE 2013-01-0836 Diesel particulate trap regeneration techniques: Wade, W., White, J., and Florek, J., SAE 810118 (1981) 249 41 42 43 44 45 46 47 48 49 50 51 52 53 54 250 The Effect of a Particle Oxidation Catalyst (POC (registered)) on Particle Emissions of a GDI Car During Transient Engine Operation: Happonen, Matti; Matilainen, Pekka; Kanniainen, Kauko; Kinnunen, Toni; Karjalainen, Panu; Heikkila, Juha; Ronkko, Topi; Keskinen, Jorma; Lahde, Tero; Malinen, Antti; Pirjola, Liisa; Tzamkiozis, Theodoros, SAE 2013-01-0839 The PM-Metalit: A PM control technology for Tier Off-Highway Applications: Jan Kramer, Ulrich Pfahl Claus Bruestle, Robert Diewald and Gary Hunter, SAE 2009-01-2838 PM Metalit® - A Continuously Regenerating Partial Flow Particulate Filter Concept and Experience with Korean Retrofit Programme: K.V.R.Babu, Chris Dias, Shivraj Waje Emitec Emission Control Technologies India Pvt Ltd Alfred Reck Emitec GmbH Kim Wonsik Emitec Korea, SAE 2008-28-0008 A General 3D Model to Analyze Particle Transport into a Partial-FlowParticulate-Filter: S Cordiner, G De Simone and V Mulone, SAE 2010-010881 Metal Supported Flow-Through Particulate Trap; a Non-Blocking Solution: Rolf Brück, Peter Hirth, Meike Reizig, Peter Treiber and Jürgen Breuer, SAE 200101-1950 Particle Oxidation Catalyst (POC®) - From Diesel To GDI - Studies on Particulate Number and Mass Efficiency: Kinnunen, Toni; Matilainen, Pekka; Scheder, Daniel; Czika, Werner; Waters, David; Russ, Gerald, SAE 2012-010845 The Effect of a Particle Oxidation Catalyst (POC®) on Particle Emissions of a GDI Car during Transient Engine Operation: Happonen, Matti; Matilainen, Pekka; Kanniainen, Kauko; Kinnunen, Toni; Karjalainen, Panu; Heikkila, Juha; Ronkko, Topi; Keskinen, Jorma; Lahde, Tero; Malinen, Antti; Pirjola, Liisa; Tzamkiozis, Theodoros, SAE 2013-01-0839 Development of Partial Filter Technology for HDD Retrofit: Todd Jacobs, Sougato Chatterjee, Ray Conway and Andy Walker Johnson Matthey ECT Jan Kramer and Klaus Mueller-Haas, SAE 2006-01-0213 Partial-Flow Deep-bed-Filter for Ultra-low Emission Concepts: Rolf Bruck, Roman Konieczny, Andreas Scheeder and Peter Treiber http://www.emitec.com/fileadmin/user_upload/Bibliothek/Vortraege/070125_ complete_Paper_final%20version.pdf accessed Sept (2013) Soot trapping and continuously oxidizing behavior by flow-through Metallic PM filter: Okawara Seji, Tsuji Shinji, Inoue Mikio, Itatsu Toshiro, Nohara Tetsuo, Komatsu Kazunari, SAE 2005-08-0492 Metal Supported Flow-Through Particulate Trap; a Non-Blocking Solution: Rolf Brück, Peter Hirth, Meike Reizig, Peter Treiber, Jürgen Breuer, SAE 2001-011950 Ceramic foam substrates for automotive catalyst applications: fluid mechanic analysis Panayotis Dimopoulos, Eggenschwiler Dimitrios, N Tsinoglou, Jacqueline Seyfert, Christian Bach, Ulrich Vogt Michal Gorbar, Exp Fluids 47, 209–222 (2009) Filtration and Regeneration Performance of a Catalyzed Metal Foam Particulate Filter: G C Koltsakis, D K Katsaounis and Z C Samaras Aristotle University Thessaloniki D Naumann, Sh Saberi and A Böhm, SAE 2006-01-1524 Catalytic Air Pollution Control Commercial Technology: Ronald M Heck, Robert J Farrauto with Suresh T Gulati Third edition (2009) ISBN 978-0-470-275030 AUTHOR INDEX Akehurst, S Akroyd, J 137 Anbari Attar, M 159 Bae, C 179 Banton, K M 137 Bassett, M D 79 Bhave, A N 137 Bosteels, D 55 Bounos, K 67 Brace, C J Bredda, S W Brogan, M 231 Cracknell, R 149 Davy, M H 33 Dunn, M E 123 Eastwood, P 231 Efthymiou, P 33 Favre, C 55 Fennell,D 193 Garner, C P 33 Hall, J 79 Hargrave, G K 33 Harris, J 33 Hayden, D 193 Head, B 149 Herfatmanesh, M R 159 Hillman, M D 137 Holderbaum, B 205 Ioakimidis, C 67 Johansson, B 179 Jung, J 179 Körfer, Th 205 Kraft, M 137 Leach, F 193 Lewis, A G J Luebbers, M 149 May, J 55 McGhee, M J 95 McTaggart-Cowan, G P 123 Morgan, N M 137 Oh, H 179 Ojapah, M M 19 Pegg, I 95 Popplewell, A Prakash, A 149 Remmert, S 149 Richardson, D 33, 193 Richardson, S Riehl, R C R 137 Rimmer, J E T 33 Rounce, P 231 Saunders, J 123 Savvidis, D 67 Schnorbus, Th 205 Shayler, P J 95 Smallbone, A J 137 Sochacki, B 67 Stone, R 193 Turner, J W G Twigg, M V 219 Warth, M 79 Whelan, S 111 Wicks, N 193 Wittka, Th 205 Wong, H C 111 Wu, Y 149 Zammit, J P 95 Zhang, Y 19, 169 Zhao, H 19, 159, 169