Internationaler Motorenkongress 2020 [1. Aufl.] 9783658304997, 9783658305000

Citation preview

Johannes Liebl · Christian Beidl Wolfgang Maus Hrsg.

Internationaler Motorenkongress 2020 Proceedings

Proceedings

Ein stetig steigender Fundus an Informationen ist heute notwendig, um die immer komplexer werdende Technik heutiger Kraftfahrzeuge zu verstehen. Funktio­ nen, Arbeitsweise, Komponenten und Systeme entwickeln sich rasant. In immer schnelleren Zyklen verbreitet sich aktuelles Wissen gerade aus Konferenzen, Tagungen und Symposien in die Fachwelt. Den raschen Zugriff auf diese Infor­ mationen bietet diese Reihe Proceedings, die sich zur Aufgabe gestellt hat, das zum Verständnis topaktueller Technik rund um das Automobil erforderliche spe­ zielle Wissen in der Systematik aus Konferenzen und Tagungen zusammen zu stellen und als Buch in Springer.com wie auch elektronisch in Springer Link und Springer Professional bereit zu stellen. Die Reihe wendet sich an Fahrzeug- und Motoreningenieure sowie Studierende, die aktuelles Fachwissen im Zusammen­ hang mit Fragestellungen ihres Arbeitsfeldes suchen. Professoren und Dozenten an Universitäten und Hochschulen mit Schwerpunkt Kraftfahrzeug- und Moto­ rentechnik finden hier die Zusammenstellung von Veranstaltungen, die sie selber nicht besuchen konnten. Gutachtern, Forschern und Entwicklungsingenieuren in der Automobil- und Zulieferindustrie sowie Dienstleistern können die Procee­ dings wertvolle Antworten auf topaktuelle Fragen geben. Today, a steadily growing store of information is called for in order to understand the increasingly complex technologies used in modern automobiles. Functions, modes of operation, components and systems are rapidly evolving, while at the same time the latest expertise is disseminated directly from conferences, congres­ ses and symposia to the professional world in ever-faster cycles. This series of proceedings offers rapid access to this information, gathering the specific know­ ledge needed to keep up with cutting-edge advances in automotive technologies, employing the same systematic approach used at conferences and congresses and presenting it in print (available at Springer.com) and electronic (at Springer Link and Springer Professional) formats. The series addresses the needs of automotive engineers, motor design engineers and students looking for the latest expertise in connection with key questions in their field, while professors and instructors working in the areas of automotive and motor design engineering will also find summaries of industry events they weren’t able to attend. The proceedings also offer valuable answers to the topical questions that concern assessors, researchers and developmental engineers in the automotive and supplier industry, as well as service providers.

Weitere Bände in der Reihe http://www.springer.com/series/13360

Johannes Liebl · Christian Beidl · Wolfgang Maus (Hrsg.)

Internationaler Motorenkongress 2020

Hrsg. Johannes Liebl Moosburg a.d.Isar, Deutschland Wolfgang Maus WM Engineering & Consulting Bergisch Gladbach, Deutschland

Christian Beidl Combustion Engines & Powertrain Systems Technische Universität, Institute for Internal Darmstadt, Deutschland

ISSN 2198-7440  (electronic) ISSN 2198-7432 Proceedings ISBN 978-3-658-30500-0  (eBook) ISBN 978-3-658-30499-7 https://doi.org/10.1007/978-3-658-30500-0 Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen National­ bibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar. © Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2020 Das Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung des Verlags. Das gilt insbesondere für Vervielfältigungen, Bearbeitungen, Übersetzungen, Mikroverfilmungen und die Einspeicherung und Verarbeitung in elektronischen Systemen. Die Wiedergabe von allgemein beschreibenden Bezeichnungen, Marken, Unternehmensnamen etc. in diesem Werk bedeutet nicht, dass diese frei durch jedermann benutzt werden dürfen. Die Berechtigung zur Benutzung unterliegt, auch ohne gesonderten Hinweis hierzu, den Regeln des Markenrechts. Die Rechte des jeweiligen Zeicheninhabers sind zu beachten. Der Verlag, die Autoren und die Herausgeber gehen davon aus, dass die Angaben und Informa­ tionen in diesem Werk zum Zeitpunkt der Veröffentlichung vollständig und korrekt sind. Weder der Verlag, noch die Autoren oder die Herausgeber übernehmen, ausdrücklich oder implizit, Gewähr für den Inhalt des Werkes, etwaige Fehler oder Äußerungen. Der Verlag bleibt im Hinblick auf geografische Zuordnungen und Gebietsbezeichnungen in veröffentlichten Karten und Institutionsadressen neutral. Springer Vieweg ist ein Imprint der eingetragenen Gesellschaft Springer Fachmedien Wiesbaden GmbH und ist ein Teil von Springer Nature. Die Anschrift der Gesellschaft ist: Abraham-Lincoln-Str. 46, 65189 Wiesbaden, Germany

Vorwort

Klimaschutzziele und Vorgaben zur Luftqualität verändern unsere Mobili­ tät. Europa hat sich auf eine weitere Senkung der CO2-Grenzwerte geeinigt. Die Elektrifizierung der Antriebe ist ein Lösungsweg, der einen Teil unserer Mobilitäts- und Transportbedürfnisse abdecken kann. Nationale und inter­ nationale Märkte erfordern jedoch ein mehrgleisiges Vorgehen, das den Ver­ brennungsmotor nicht vernachlässigen darf. CO2-neutrale Kraftstoffe sind dafür unverzichtbare Bausteine. Der Motorenkongress führt auch 2020 die seit Jahren konsequente Betrachtung des Gesamtsystems aus Verbrennungsmotoren und innovativen Kraftstoffen fort. Effiziente Pkw- und Nfz-Motoren sind im System­ verbund mit neuen Kraftstoffen die Schlüsseltechnologie für eine CO2-neutrale individuelle Mobilität mit niedrigen Emissionen. Sichern Sie sich Ihren Wissensvorsprung und profitieren Sie! Es erwarten Sie internationale Referenten, hochkarätige Vorträge und Dis­ • kussionsrunden • Sie haben jederzeit die Möglichkeit, zwischen allen Vorträgen zu wechseln Nutzen Sie den Kongress zum Netzwerken – Der Abend der Motoren• Community bietet interessante Gespräche in angenehmer Atmosphäre Eine begleitende Fachausstellung informiert über innovative Produkte und • Dienstleistungen im Bereich Verbrennungsmotorenentwicklung

V

VI

Vorwort

Wir freuen uns auf Ihre Teilnahme! Im Namen der Programmbeiräte Dr. Johannes Liebl Wissenschaftlicher Leiter des Kongresses, Herausgeber ATZ | MTZ | ATZelektronik

Editorial

Climate protection targets and air quality specifications are changing our mobility. Europe has agreed on a further reduction in CO2 limits. The electrification of powertrains is one solution that can fulfill part of our mobility and transportation requirements. However, national and international markets require a multi-track approach that must not neglect the internal combustion engine. CO2-neutral fuels are indispensable components of this approach. Once again in 2020, the International Engine Congress will continue its systematic examination of the complete system of internal combustion engines and innovative fuels on which it has consistently focused for many years. Efficient passenger car and commercial vehicle engines combined with new fuels are the key technology for CO2-neutral individual mobility with low emissions. Stay abreast of current trends and benefit from a lead in knowledge! • You can expect international speakers as well as top-level presentations and panel discussions • You can switch between lectures at any time, if you prefer to hear about another topic • The congress is a great opportunity to “network“ – the evening event for the engine community offers stimulating discussions in a pleasant, relaxing atmosphere The trade exhibition, held in parallel, provides ample information about • innovative products and services in the field of combustion engine development

VII

VIII

Editorial

We look forward to your participation. On behalf of the program advisory boards Dr. Johannes Liebl Scientific Director of the Congress, Editor-in-Charge ATZ | MTZ | ATZelektronik

Inhaltsverzeichnis

Results of a patent analysis and a market study to assess future concepts of hybrid vehicles Prof. Dr.-Ing. Wilhelm Hannibal, Niklas Haverkamp und Prof. Dr.-Ing. Peter Eilts TwinRex – Dedicated Hybrid Engine for a serial-parallel powertrain with excellent cost-value index Matthias Thewes, Adrian Schloßhauer, Oguz Budak, Jörg Seibel, Reiner Wohlberg, Andreas Müller, Johannes Moritz Maiterth, Markus Eisenbarth, Ruben Keizer, Farouk Odeim, Michael Kauth, Rene Savelsberg, Georg Birmes, Andreas Balazs, Tolga Uhlmann, Johannes Scharf, Norbert Alt und Andreas Sehr Future diesel powertrain in LCV and SUV – electrified, modular platform with focus on emission, efficiency and cost (Zukünftige Dieselantriebe in LCV und SUV – Elektrifizierte, modulare Plattform mit Fokus auf Emission, Effizienz und Kosten) Dr. Wolfgang Schöffmann, Michael Howlett, Bernhard Enzi, Stefan Krapf, Christoph Sams, Hannes Wancura, Michael Weißbäck und Dr. Helfried Sorger MPI valves for use in large engine applications – challenges in the development and derived benefits for operation Peter Christiner, Claudia Hengstberger, Markus Schmitzberger und Michael Köhler Direction of gas vehicle development in Japan Akiyoshi Kishi Managing cryogenic fuels on heavy-duty HPDI vehicles Adrian Post, David Mumford, Robbi McDonald und Gage Garner

IX

X

Inhaltsverzeichnis

Systemic development approach for optimizing piston ring design to reduce particulate raw emissions Thomas Bastuck, Dipl.-Ing. Richard Mittler und Dipl.-Ing. Steffen Hoppe The ring catalyst – an innovative, ultracompact solution for EU7 Rolf Brück und Thomas Härig Variable valve actuation as an efficient measure for Off-Highway (OHW) drive systems Dr. Georg Töpfer, Adrian Troeger, Justus Himstedt und Stefan Steichele CO2 reduction by minimizing friction at the pcu in conflict of targets with increased oil aerosol formation Magnus Lukas Lorenz und Prof. Thomas Koch HD diesel engine – exhaust gas temperature management and advanced exhaust gas aftertreatment technology for ultra-low NOx emission legislation Jonas Edvardsson, Klaus Hadl, Eric Hein, Georg Kraus, Hannes Noll, Christina Schwarz, Stefanie Tamm und Helmut Theissl Processes for the production of OME fuels Jakob Burger und Hans Hasse Fahrplan zu einer OME-Spezifikation (Roadmap to an OME specification) Dr. Thomas Wilharm, Dr. Hendrik Stein und Innokentij Bogatykh The second generation electrically driven compressor – more power for more possibilities Dr.-Ing. Hermann Breitbach, Dr.-Ing. Ralf Christmann, Dipl.-Ing. H. Gabriel und Dipl.-Ing. Dietmar Metz Design of electrified turbomachinery for use in modern industrial hybrid powertrains Shinri Szymko, Owen Creese-Smith, Gael de Crevoisier, Michela Mascherin, Richard Goodyear und Henry Carr Hydrogen to deal with intermittency of renewable electricity generation Martin Rothbart, Jürge Rechberger, David Reichholf und Richard Schauperl Hydrogen as a fuel – Shell’s view on providing hydrogen for ­heavy-duty mobility applications Paul Karzel, Jason Munster und Andreas Kolbeck

Inhaltsverzeichnis

XI

50 % brake thermal efficiency – the realization of a vision Rolf Heinrich Dreisbach, Martin Wieser, Franz Hofer, Helmut Theissl, Hans Felix Seitz, Kurt Schmidleitner, Heinz-Georg Flesch, Andreas Horvath, Jürgen Gelter, Wolfgang Gruber und Martin Piffl Audi Denkwerkstatt: Which learnings could be transfered from agile start-ups to powertrain world? Dr. Matthias Brendel Passenger Car Emissions and Consumption in Real Driving Conditions from the Point of View of Automobile Clubs Dr. techn. Thomas Hametner, Lone Otto und Dr.-Ing. Reinhard Kolke Interaction and influence of HVO-based fuels on diesel combustion Daniel Erforth, Paul Lagaly und Prof. Thomas Koch 48 V hybrid system based on a switched reluctance motor for city busses Markus Lampalze und Michael Lechner Electrified efficiency – diesel hybrid powertrain concepts for light commercial vehicles Joschka Schaub, Martin Pieper, Stefan Klopstein, Matthias Übbing, Pascal Knappe, Paul Muthyala und Thorsten Schmidt Renewable drop-in fuels as an immediate measure to reduce CO2 emissions of heavy-duty applications Jaykumar Yadav, Vikram Betgeri, Barbara Graziano, Avnish Dhongde, Benedikt Heuser, Markus Schönen und Nina Sittinger Sustained CO2 reduction in vehicle traffic with renewable fuels Dipl. Ing. Karl Dums, Dipl. Ing. Hans-Peter Deeg, Dipl. Ing. Marcos Remedios Marques, Dr. Andre Casal Kulzer und Dipl. Ing. Dietmar Schwarzenthal Life cycle assessment as a tool for analyzing the CO2 footprint of passenger cars with different powertrains Philipp Weber, Jens Buchgeister, Olaf Toedter und Thomas Koch Synthetic, regenerative fuels (reFuels) as enabler for climate neutral mobility and transport Alexander Mokros, Philipp Demel, Friedemar Knost, Markus Münz und Christian Beidl

XII

Inhaltsverzeichnis

Synthesis of Oxymethylene Ether (OME) – a 2nd generation E-fuel Andreas Peter, Harald Scherer, Eberhard Jacob und Ingo Krossing Environmental assessment of OME3-5 synfuel production via the power-toliquid pathway Christoph Hank, Lukas Lazar, Franz Kaspar Mantei, Mohamed Ouda, Robin J. White, Tom Smolinka, Achim Schaadt, Christopher Hebling und Hans-Martin Henning Potential analysis and virtual development of SI engines operated with DMC+ Cornelius Wagner, Dr.-Ing. Michael Grill, Dr.-Ing. Mahir-Tim Keskin, Dr.-Ing. Liming Cai und Prof. Dr.-Ing. Heinz Pitsch

Autorenverzeichnis

Norbert Alt  FEV Europe GmbH, Aachen, Deutschland Andreas Balazs  FEV Europe GmbH, Aachen, Deutschland Thomas Bastuck  Federal-Mogul Burscheid GmbH, Burscheid, Deutschland Christian Beidl  Combustion Engines & Powertrain Systems, Technische Universität, Institute for Internal, Darmstadt, Deutschland Vikram Betgeri  RWTH Aachen University, Aachen, Deutschland Georg Birmes  FEV Europe GmbH, Aachen, Deutschland Innokentij Bogatykh  Hochschule Augsburg, Augsburg, Deutschland Dr.-Ing. Hermann Breitbach  BorgWarner Turbo Systems, Kirchheimbolanden, ­Deutschland Dr. Matthias Brendel  Audi Denkwerkstatt, Berlin, Deutschland Rolf Brück  Vitesco Technologies Emitec GmbH, Lohmar, Deutschland Jens Buchgeister  Karlsruhe Institute of Technology, Karlsruhe, Deutschland Oguz Budak  FEV Europe GmbH, Aachen, Deutschland Jakob Burger  Technical University of Munich, Straubing, Deutschland Dr.-Ing. Liming Cai  ITV RWTH Aachen, Aachen, Deutschland Henry Carr  Cummins Turbo Technologies, Huddersfield, UK Peter Christiner  Robert Bosch AG, Linz, Österreich

XIII

XIV

Autorenverzeichnis

Dr.-Ing. Ralf Christmann  BorgWarner Turbo Systems, Kirchheimbolanden, Deutschland Owen Creese-Smith  Bowman Power Group Ltd, Southhampton, UK Gael de Crevoisier  Bowman Power Group Ltd, Southhampton, UK Dipl.-Ing. Hans-Peter Deeg  Dr. Ing. h.c. F. Porsche AG, Stuttgart, Deutschland Philipp Demel  TU Darmstadt, Darmstadt, Deutschland Avnish Dhongde  FEV Europe GmbH, Aachen, Deutschland Rolf Heinrich Dreisbach  Graz, Österreich Dipl.-Ing. Karl Dums  Dr. Ing. h.c. F. Porsche AG, Stuttgart, Deutschland Jonas Edvardsson  Johnson Matthey AB, Västra Frölunda, Schweden Prof. Dr.-Ing. Peter Eilts  Technical University Braunschweig, Braunschweig, Deutschland Markus Eisenbarth  RWTH Aachen University, Aachen, Deutschland Bernhard Enzi  AVL List GmbH, Graz, Österreich Daniel Erforth  Karlsruhe Institute of Technology, Karlsruhe, Deutschland Heinz-Georg Flesch  Graz, Österreich Dipl.-Ing. H. Gabriel  BorgWarner Turbo Systems, Kirchheimbolanden, Deutschland Gage Garner  Westport Fuel Systems, Vancouver, Kanada Jürgen Gelter  Graz, Österreich Richard Goodyear  Cummins Turbo Technologies, Huddersfield, UK Barbara Graziano  FEV Europe GmbH, Aachen, Deutschland Dipl.-Ing. Michael Grill  FKFS, Stuttgart, Deutschland Wolfgang Gruber  Graz, Österreich Klaus Hadl  AVL List GmbH, Graz, Österreich Dr. techn. Thomas Hametner  ÖAMTC, Wien, Österreich

Autorenverzeichnis

Christoph Hank  Fraunhofer Institute for Solar Energy Systems ISE, Freiburg i. Breisgau, Deutschland Prof. Dr.-Ing. Wilhelm Hannibal  South Westphalia University of Applied Sciences, Iserlohn, Deutschland Thomas Härig  Vitesco Technologies Emitec GmbH, Lohmar, Deutschland Hans Hasse  OME Technologies GmbH, Kaiserslautern, Deutschland Niklas Haverkamp  Trademark and Design Attorney, Iserlohn, Deutschland Christopher Hebling  Fraunhofer Institute for Solar Energy Systems ISE, ­Freiburg i. Breisgau, Deutschland Claudia Hengstberger  Robert Bosch AG, Linz, Österreich Hans-Martin Henning  Fraunhofer Institute for Solar Energy Systems ISE, ­Freiburg i. Breisgau, Deutschland Benedikt Heuser  FEV Europe GmbH, Aachen, Deutschland Justus Himstedt  Mahle GmbH, Stuttgart, Deutschland Franz Hofer  Graz, Österreich Dipl.-Ing. Steffen Hoppe  Federal-Mogul Holding Deutschland GmbH, ­Wiesbaden, Deutschland Andreas Horvath  Graz, Österreich Michael Howlett  AVL List GmbH, Graz, Österreich Eberhard Jacob  Emissionskonzepte Motoren, Bodman-Ludwigshafen, Deutschland Paul Karzel  Shell Deutschland Oil GmbH, Hamburg, Deutschland Michael Kauth  FEV Europe GmbH, Aachen, Deutschland Ruben Keizer  FEV Europe GmbH, Aachen, Deutschland Anton Keller  TCS, Vernier, Schweiz Akiyoshi Kishi  Isuzu Motors Limited, Kanagawa, japan Stefan Klopstein  FEV Europe GmbH, Aachen, Deutschland Pascal Knappe  RWTH Aachen, Aachen, Deutschland

XV

XVI

Autorenverzeichnis

Friedemar Knost  TU Darmstadt, Darmstadt, Deutschland Prof. Thomas Koch  Karlsruhe Institute of Technology, Karlsruhe, Deutschland Michael Köhler  Robert Bosch AG, Linz, Österreich Andreas Kolbeck  Shell Global Solutions Deutschland, Hamburg, Deutschland Dr.-Ing. Reinhard Kolke  ADAC e.V., München, Deutschland Stefan Krapf  AVL List GmbH, Graz, Österreich Georg Kraus  Tenneco GmbH, Edenkoben, Deutschland Ingo Krossing  Albert-Ludwigs-University Freiburg, Freiburg, Deutschland Dr. Andre Casal Kulzer  Dr. Ing. h.c. F. Porsche AG, Stuttgart, Deutschland Paul Lagaly  Karlsruhe Institute of Technology, Karlsruhe, Deutschland Markus Lampalzer  MAN Truck & Bus SE, Nürnberg, Deutschland Lukas Lazar  Fraunhofer Institute for Solar Energy Systems ISE, Freiburg i. Breisgau, Deutschland Michael Lechner  MAN Truck & Bus SE, München, Deutschland Magnus Lukas Lorenz  Daimler Truck AG, Stuttgart, Deutschland Robbi McDonald  Westport Fuel Systems, Vancouver, Kanada Dr.-Ing. Keskin Mahir-Tim  FKFS, Stuttgart, Deutschland Johannes Moritz Maiterth  RWTH Aachen University, Aachen, Deutschland Franz Kaspar Mantei  Fraunhofer Institute for Solar Energy Systems ISE, ­Freiburg i. Breisgau, Deutschland Michaela Mascherin  Bowman Power Group Ltd, Southhampton, UK Dr.-Ing. Dietmar Metz  BorgWarner Turbo Systems, Kirchheimbolanden, Deutschland Dr.-Ing. Richard Mittler  Federal-Mogul Burscheid GmbH, Burscheid, ­Deutschland Alexander Mokros  TU Darmstadt, Darmstadt, Deutschland Andreas Müller  FEV Europe GmbH, Aachen, Deutschland

Autorenverzeichnis

XVII

David Mumford  Westport Fuel Systems, Vancouver, Kanada Jason Munster  Shell Exploratoin and Production Company, Cambridge, USA Markus Münz  TU Darmstadt, Darmstadt, Deutschland Paul Muthyala  RWTH Aachen, Aachen, Deutschland Hannes Noll  AVL List GmbH, Graz, Österreich Farouk Odeim  FEV Europe GmbH, Aachen, Deutschland Lone Otto  FDM, Lyngby, Dänemark Mohamed Ouda  Fraunhofer Institute for Solar Energy Systems ISE, Freiburg i. Breisgau, Deutschland Andreas Peter  Albert-Ludwigs-University Freiburg, Freiburg, Deutschland Martin Piffl  Graz, Österreich Martin Pieper  FEV Europe GmbH, Aachen, Deutschland Prof. Dr.-Ing. Heinz Pitsch  ITV RWTH Aachen, Aachen, Deutschland Adrian Post  Westport Fuel Systems, Vancouver, Kanada Jürgen Rechberger  AVL List GmbH, Graz, Österreich David Reichholf  AVL List GmbH, Graz, Österreich Dipl.-Ing. Marcos Remedios Marques  Dr. Ing. h.c. F. Porsche AG, Stuttgart, Deutschland Martin Rothbart  AVL List GmbH, Graz, Österreich Christoph Sams  AVL List GmbH, Graz, Österreich Rene Savelsberg  FEV Europe GmbH, Aachen, Deutschland Achim Schaadt  Fraunhofer Institute for Solar Energy Systems ISE, Freiburg i. Breisgau, Deutschland Johannes Scharf  FEV Europe GmbH, Aachen, Deutschland Joschka Schaub  FEV Europe GmbH, Aachen, Deutschland Richard Schauperl  AVL List GmbH, Graz, Österreich Harald Scherer  Albert-Ludwigs-University Freiburg, Freiburg, Deutschland

XVIII

Autorenverzeichnis

Adrian Schloßhauer  FEV Europe GmbH, Aachen, Deutschland Kurt Schmidleitner  Graz, Österreich Thorsten Schmidt  Volkswagen AG, Wolfsburg, Deutschland Markus Schmitzberger  Robert Bosch AG, Linz, Österreich Dr. Wolfgang Schöffmann  AVL List GmbH, Graz, Österreich Markus Schönen  FEV Europe GmbH, Aachen, Deutschland Christina Schwarz  AVL List GmbH, Graz, Österreich Dipl.-Ing. Dietmar Schwarzenthal  Dr. Ing. h.c. F. Porsche AG, Stuttgart, Deutschland Andreas Sehr  FEV Europe GmbH, Aachen, Deutschland Jörg Seibel  FEV Europe GmbH, Aachen, Deutschland Hans Felix Seitz  Graz, Österreich Nina Sittinger  OWI Oel-Waerme-Institute gGmbH, Herzogenrath, Deutschland Tom Smolinka  Fraunhofer Institute for Solar Energy Systems ISE, Freiburg i. Breisgau, Deutschland Dr. Helfried Sorger  AVL List GmbH, Graz, Österreich Stefan Steichele  Mahle GmbH, Stuttgart, Deutschland Dr. Hendrik Stein  Hochschule Augsburg, Augsburg, Deutschland Shinri Szymko  Bowman Power Group Ltd, Southhampton, UK Stefanie Tamm  Johnson Matthey AB, Västra Frölunda, Schweden Helmut Theissl  AVL List GmbH, Graz, Österreich Matthias Thewes  FEV Europe GmbH, Aachen, Deutschland Olaf Toedter  Karlsruhe Institute of Technology, Karlsruhe, Deutschland Dr. Georg Töpfer  Deutz AG, Köln, Deutschland Adrian Troeger  Deutz AG, Köln, Deutschland Matthias Übbing  FEV Europe GmbH, Aachen, Deutschland

Autorenverzeichnis

Tolga Uhlmann  FEV Europe GmbH, Aachen, Deutschland Cornelius Wagner  FKFS, Stuttgart, Deutschland Hannes Wancura  AVL List GmbH, Graz, Österreich Philipp Weber  Karlsruhe Institute of Technology, Karlsruhe, Deutschland Michael Weißbäck  AVL List GmbH, Graz, Österreich Robin J. White  Fraunhofer Institute for Solar Energy Systems ISE, Freiburg i. Breisgau, Deutschland Martin Wieser  Graz, Österreich Thomas Wilharm  ASG Analytik-Service Gesellschaft mbH, Neusäss, Deutschland Reiner Wohlberg  FEV Europe GmbH, Aachen, Deutschland Jaykumar Yadav  RWTH Aachen University, Aachen, Deutschland

XIX

Results of a patent analysis and a market study to assess future concepts of hybrid vehicles Prof. Dr.-Ing. Wilhelm Hannibal1, Niklas Haverkamp M.Sc.2 and Prof. Dr.-Ing. Peter Eilts3 1

South Westphalia University of Applied Sciences, Frauenstuhlweg 31, 58644 Iserlohn, Germany 2 Patent attorney Prof. Dr. rer. nat. Jens Haverkamp, European Patent, Trademark and Design Attorney, Gartenstraße 61, 58636 Iserlohn, Germany 3 Institute of Internal Combustion Engine, Technical University Braunschweig, Hermann-Blenk-Straße 42, 38108 Braunschweig, Germany

Abstract. Due to changing requirements of automotive customers, society and politics, alternatives to the well-known pure internal combustion engine drives are needed as propulsion concepts. One possibility is the hybridization of an engine together with at least one electric motor. By different interconnections of these drive units, a wide variety of propulsion topologies can be provided. To estimate future concepts, developments reflected in patent applications are used. Basis is an overall analysis of patent applications filed with the European Patent Office. By extracting cumulations of filings in individual categories, trends of developments can be derived. Applicants based in Europe mainly file a wide range of different topologies of parallel hybrid concepts. Primarily, these are based on further development of the known components. In particular, P1-, and, more recently, P2- and P12-hybrids are frequently found. In contrast, the majority of Applicants based in Asia are focused on power-split hybrids. Toyota, the largest Applicant for patents in the field of hybrid vehicles, is focused on continuously variable transmissions. Other Applicants from Asia are focused on parallel-serial hybrids. In the US, due to a smaller number of applications no profound statements can be made. For a vehicle with a retail price of approx. € 20,000, a hybrid powertrain concept is presented based on a simplified three-cylinder gasoline engine in order to achieve the ambitious cost target. Keywords: Patent Analysis, Hybrid, Powertrain.

1

Introduction

In the automotive industry, a change is noticeable. Due to new legislative and social requirements, the industry is forced to provide alternatives to the conventional engine drive concepts. A promising alternative is the electrification of the powertrain, to which success vehicle manufacturers as well as infrastructure technologies have to deal with

© Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2020 J. Liebl et al. (Hrsg.), Internationaler Motorenkongress 2020, Proceedings, https://doi.org/10.1007/978-3-658-30500-0_1

2

a large number of problems. In an intermediate phase, hybrid technology is therefore an obvious alternative, even if two energy sources have to be used. Current studies by well-known analysts predict that by 2050, 60 % of the vehicles will be hybrid (see Fig. 1) [1] to [11]. In contrast, at the same time, the purely electric drive will be around 20 % only. For this reason, hybrid drive technology must be examined intensively in the medium term. In particular, the greatest advantage is the existing infrastructure for distributing energy – the huge network of petrol stations. The installation of charging stations, however, is neither spread to the same extent nor as powerful. Still, the decision which topology is to be followed has a great impact on the alignment of the product line, the costs and ultimately on the success or failure of a vehicle. Latest figures from the German Federal Motor Transport Authority can confirm the analysts' estimate, at least with regard to the trend [12]. Fig. 2 shows the new registrations of passenger cars over the years, split in purely engine-powered vehicles, hybrid vehicles, electric vehicles, gas-powered and other vehicles, with the hybrid, gas and electric vehicles also shown enlarged for the sake of clarity. The number of hybrid vehicles is increasing rapidly. Around 83,000 new vehicles were registered in 2017 – almost twice as many as in the previous year and eight times as much as in 2010. In contrast, only 24,000 new electric cars were registered in 2017. It can be deduced that hybrid vehicles are much better accepted by customers than electric vehicles. However, the absolute amount of hybrid vehicles is still much smaller than purely engine powered vehicles.

Fig. 1. Trends of different types of drive concepts, from [1–11]

3

Fig. 2. New registrations in Germany, stacked

Since costumers are usually bound to a certain budget, a vehicle ready for large series has to be compact and cheap, e.g. at a retail price of € 20,000. Aiming for such a drive concept is difficult because of the multiple interconnections between the two required powertrains (engine and motor). For the automobile manufacturer, it is difficult to choose the most reasonable solution from the wide variety of concepts. To form a basis for such a choice, it is helpful to take a detailed look at the innovations in the field of hybrid vehicles made by the industry. A profound approach can be an overall analysis of patent applications.

2

Patent study on topologies of hybrid vehicles

In a research project, investigations were carried out to determine whether there are technical trends and innovations in the area of hybrid drive concepts. An intensive study of patent applications tries to answer this question, obtaining an overview of the different topologies possible. 2.1

Statistical analysis of patent applications

Hundreds of millions of patent documents are available worldwide in various patent databases. The database of the European Patent Office alone – Espacenet – contains 110 million documents (as of 2019). It is impossible to view such a number of documents in a reasonable and adequate time. For this reason, first, the patent documents relevant to the technology field of hybrid drive systems must be filtered out.

4

The investigations were limited to the international patent classification class B60K6/00, concerning hybrid vehicles and the patent documents of the European Patent Office. Around 300,000 patent applications are received by the European Patent Office each year. Experience shows that all important projects attempt to obtain a European property right. The cost of obtaining a European patent is significantly higher than that of one national patent, so one can conclude that an application for European patent has a certain relevance. By analyzing European Patents only, "experiments" are left out of the investigation. Furthermore, European patent applications are usually based on a priority application. A priority application is a first filing application, usually in the home-country of the Applicant. Within twelve months, the same invention can be filed within almost each country worldwide, maintaining the initial filing date with respect to the examined state of the art. National patent offices try to communicate a first examination report within nine to eleven months from the filing date. By this, the Applicant can decide, whether he is aiming for a wider spreading of the patent in other countries. If the project is important, a spreading is typically the next step. Therefore, European patents are valuable for answering the question, which concepts seem to be actually implemented.

Fig. 3. Total number of European patent applications and categorized applications in the database over priority date

From Fig. 3 one can derive that the applications concerning hybrid vehicles have skyrocketed. As there is a delay between the filing date and the publication date, the number of publications shown in the graphic is not complete for the past few years – one can assume that the number is still increasing. Furthermore, in Fig. 4 the active Applicants regarding the hybrid vehicles are listed. It can be clearly seen that Toyota – also having sold the most hybrid vehicles on the market – has submitted the most patent applications to the European Patent Office.

5

Fig. 4. Number of patent applications per manufacturer by early 2019

6

2.2

Structure of hybrid vehicles

Topology-related categorization. There are different ways to categorize hybrid vehicles. Aspects of interest can be the hybridization degree, different operating modes, etc. However, for categorizing those hybrid vehicles being implemented, a topology-related categorization is required. Such a categorization is helpful to give the engineer a certain structure, which can be developed in detail. There are basically three basic topologies: the parallel hybrid, the serial hybrid and the power-split hybrid, see Fig. 5. This categorization shows possible basic types, only. In the analyzed patent applications, sometimes several of these basic types are linked to one another to combine different advantages and operation modes. Parallel hybrid. In the case of the parallel hybrid, the internal combustion engine and the electric motor drive the powertrain in parallel, i.e. both driving machines have a mechanical connection to the drive. A distinction is made with regard to the arrangement of the electric motor compared to the main transmission in:  arrangement of the electric motor before the main transmission,  in the main transmission and  after the main transmission. Depending on the arrangement, the efficiency of the electric motor for driving the hybrid vehicle and thus the electrification of the powertrain is different. With regard to some parallel hybrids, a naming introduced by Daimler has prevailed in industry [13]. This is composed of a P for "Parallel" and a code number from 1 to 4 together, which is incorporated at this point. Serial hybrid. The serial hybrid drive is characterized by a series connection of the internal combustion engine and the electric motor, without a mechanical coupling inbetween. In the classic variant, electrical energy generated by a generator driven by an internal combustion engine is made available to one or more electric motors driving the vehicle. In addition, a battery is available as a buffer, which temporarily stores excess energy generated by the generator. The advantage of this topology is the free choice of the operating point of the internal combustion engine, so that the goals of low emissions and the highest possible degree of effectiveness can be achieved. By using a strong electric motor, a large part of the braking energy can be recuperated. On the other hand, the conversion losses from fuel to output are disadvantageous when the vehicle is driven solely on the base of the engine. Power-split hybrid. A power-split hybrid provides a powertrain that has an additional variation in operating modes and connection options due to a division to an electrical and mechanical path. In literature, there is often no clear distinction between the generic term of the power-split hybrid in general and the power-split by means of a planetary gear, see for example [14] versus [13]. Basically, based on [13], when using a planetary gearset to provide a power-split hybrid, we speak of a 1- or 2-mode hybrid, depending

7

on the further set up. Furthermore, power-split hybrids can be provided by connecting the engine and the electric motor by a clutch. In another embodiment, an axle split can be modified by internally connecting the engine with the electric motor.

Fig. 5. Structure diagram of the topology-related hybrid categories

In Fig. 6 there is an overview of possible examples of hybrid concepts based on the principle of sketches, which also makes it clear that it is difficult to understand technically this topic in every detail.

8

Fig. 6. Examples of different hybrid concepts

2.3

Topology analysis of patent applications for hybrid drives

In the research project, European patent documents are managed in a specially developed database, being categorized according the aforementioned structure. By the cutoff date for this analysis – May 15, 2019 – 3,652 applications had been published by the European Patent Office that were assigned the above-mentioned IPC class (B60K6/00). 2,920 of these could be categorized as part of the evaluation (see Fig. 3). The remaining part is half related to other vehicles such as boats, planes or construction machinery, half could not be meaningfully assigned. To view the different matters of these applications, technology charts have been established. In such a technology chart, the number of applications per priority year is plotted against the individual categories. To investigate the developments in different countries, the claimed priority can be used. Normally the priority is claimed from a country where the invention had been carried out. This is leading at least indirectly to the Applicant. Germany priority country. 537 applications of the analyzed patent applications claim a priority in Germany. In the associated technology chart (Fig. 7), it can be clearly recognized that most of the applications filed concern the parallel hybrid. In the field of the serial hybrid, there is less but still continuously activity during the entire analyzed period. For the power-split hybrid, only a few applications have been submitted. Looking more closely at the parallel hybrid, clusters can be recognized. While P1 hybrids were developed in the early days, a tendency towards more complex P2 or P12 hybrids can be seen in younger days. The reason for this may be that a P1 hybrid can be developed much more cost-effectively, since only a few adjustments to the entire powertrain are necessary. This also leads to a few electrical operating modes only:

9

purely electric driving, for example, is characterized by an extremely low level of efficiency due to the internal friction of the engine. To overcome this disadvantage, P2 and P12 hybrids suggest to decouple the electric motor from the engine to increase the efficiency. Amazingly, other parallel topologies are much less represented. Only recently, some applications have been registered. The serial hybrid – more as part of schematic diagrams – was already thought of before 2000, but was afterwards neglected compared to the parallel hybrid. There has been a slight increase since 2006. A topology of a transmission being primarily arranged between the driving electric motor and the output has been used throughout the whole analyzed period. Power-split hybrids are only occasionally pursued in Germany. Only the 2-mode technology has been pursued continuously to a small extend, namely since 2003 with approximately one application per year. The concept of parallel-serial hybrids, especially the Combined Axle, was also developed from around 2008/2010 on.

10

Fig. 7. Technology card, filtered according to German priority applications (A = Axle Drive, E = Electric Motor, G = Transmission, ICE = Internal Combustion Engine)

As in younger days, the tendency to implement hybrid vehicles being more complex to provide purer electric driving modes, it can be postulated that in the future the electric motor may be arranged closer at the output, so that P3 topologies or the axle split will become more relevant. In both topologies, a purely electric drive of the vehicle is efficiently possible. At the same time, the engine ensures powerful propulsion on longer distances. The serial hybrid on the other hand is somehow dropped although purely electric driving is easily implemented with such a hybrid vehicle.

11

Japan priority country. Fig. 8 shows a technology chart for patent applications claiming a priority in Japan. The 1179 evaluated applications are distributed completely different than applications associated with Germany. The focus is on the power-split hybrid, especially concerning hybrids based on the 1-mode concept and the parallel-serial concept. These concepts have been filed continuously, the 1-mode concept even since 1995. In addition, several applications regarding the parallel hybrid are recognizable. The trend from the P1 to the P2 and in particular to the P12 hybrid can also be derived here.

Fig. 8. Technology chart, filtered according to Japanese priority applications (A = Axle Drive, E = Electric Motor, G = Transmission, ICE = Internal Combustion Engine)

12

With regard to the power-split hybrid, the 1-mode transmission is of extraordinary presence, almost exclusively in a configuration in which the engine is connected to the planet carrier, the generator to the sun gear and the output to the ring gear. This is due to the "synergy drive" by Toyota; the majority of these applications are submitted by Toyota itself or the suppliers associated. Between 2009 and 2013, this category seems to decrease. This is, however, not the case: Toyota no longer submitted applications concerning this topology via national Japanese applications, but directly at the European Patent Office. This is why these applications are not shown in this chart. In total, one can derive that no decrease concerning the 1-mode transmission has been taken place. In addition, the parallel-serial hybrid, as Combined Axle (since 2000) and as classic parallel-serial hybrid (since 2006), can also be observed increasingly. Both developments continue until now. With regard to the parallel hybrid, the analyzed period can be subdivided, as to when applications have been increasingly submitted: 1995 to 2003, and 2009 to the present. The P1 hybrid was filed in the first sub period. Since then, this topology has only been suggested occasionally. Instead, the P12 hybrid has been applied for since 2005 (small amount), and from 2010 the P2 hybrid. A change concerning parallel hybrids can thus be recognized in the direction of complex technologies to allow more operating points. In the second sub period, transmissions comprising the electric motor (“E in G”) are also disclosed. In this case, typically, for the electric motor only a small number of gears is provided. More recently, e.g. since 2010, the P3 hybrid has also been increasingly disclosed, so that the electric motor is arranged as close to the output as possible. Typically, only one electric motor is installed, being arranged between differential and main transmission. Concerning the serial hybrid, it has been developed since 2000, increasingly since 2006/2007. Classic serial hybrids have been filed only, namely that the electric motor is arranged before the differential. However, there is no major trend compared to the power-split hybrids and the parallel hybrids. To summarize, Applicants based in Japan focus on power-split hybrids: both, the 1mode concept and parallel-serial hybrids, in particular the Combined Axle, are pursued. The continuity of the number of applications is remarkable. Even when the activities have been dropped for a limited time, they are never completely abating. As a short-term trend-statement, Applicants based in Japan will primarily pursue power-split hybrids, Toyota in particular the 1-mode transmission, other Applicants also the other topologies. With regard to the parallel hybrid, the P2 or P12 hybrid will continue to be of interest. United States priority country. Fig. 9 shows a technology chart of those patent applications claiming priority in the United States. Due to the low total number of applications, the statements derived are not as clear as for applications with priority in Germany or in Japan. Only in 2007, a series of applications relating to the 2-mode concept has been filed 67 times. In Fig. 9 it should be pointed out that this point – marked in orange – is represented only as a quarter of its area as it should actually be compared to other points. Parallel-serial topologies were occasionally developed from 1998 to 2011, but in different, special configurations, so that one cannot derive a clear trend.

13

Regarding the parallel hybrid, primarily P1 and P2 hybrids have been applied for. P12 hybrids are only a minority. More recently, there has been a shift to the P2 hybrid, from 2010 to 2015. The topology in which the electric motor is arranged in the transmission (“E in G”) was also of interest from 2004 to 2011. The topology of a P3 hybrid was developed from 2000-2012. The serial hybrid was generally developed between 2001 and 2012, in particular in a configuration in which a transmission is arranged between the electric motor and the output.

Fig. 9. Technology chart, filtered by applications that claim American priority (A = Axle Drive, E = Electric Motor, G = Transmission, ICE = Internal Combustion Engine)

14

In principle, it can be stated that a trend-like development in the USA is not easily recognizable on the basis of this observation. Different concepts were registered at different times. However, the large number of registrations for the 2-mode transmission in 2007 is striking. Basically, there is a trend towards more complex systems.

3

Scenario of an inexpensive hybrid vehicle concept

Especially, the requirements of the automobile customer must be considered, since it is the customer who ultimately makes an automobile succeed or not, as it is him who decides buying a particular automobile. Various studies have shown that buying behavior is significantly influenced by costs, emotions (driving fun) and the use of a vehicle [18]. 3.1

Criteria for buying a car

A customer will buy a new vehicle, if it suits his financial possibilities and desire. Typically, customers consider the acquisition costs as well as the actual maintenance costs only. The resale value is only of minor importance [18]. In the case of a hybrid vehicle – the same applies to an electric vehicle – the battery price influences as the major costs the total price. Battery prices are currently just under 200 €/kWh [20]. This is an enormous decrease compared to 2010, when it was around 1000 €/kWh [19]. Fig. 10 shows the result of a survey analyzing the most important criteria for buying a car in Germany (from 2014) [18]. The decisive factor is reliability, closely followed by the price-performance ratio. In this context also low fuel consumption can be seen. Criteria, such as styling, design, interior, are only of secondary relevance. From this survey it can be deduced that easy understandable and useable concepts, indicating a high price-performance ratio, have a high chance to succeed.

Fig. 10. Criteria for buying a car in Germany, survey from 2014, according to [18]

15

3.2

Daily route

The simplest benefit analysis for a customer is the evaluation of his daily driving performance. This can provide information on the necessary range of a vehicle, Fig. 11. 70 % of private owners drive 60 km a day only, 85 % not more than 100 km. Commercial owners have higher requirements: 70 % drive less than 100 km a day [19]. Such a consideration is of enormous importance for the definition of the operating strategy and the capacity of the battery required for a hybrid vehicle. For example, if, e. g. by law, in city traffic driving purely electrically is required, the concept must provide an electric range of approximately 60 to 90 km. On the other hand, daily use varies greatly. Already 35 % of private owners do not drive more than 20 km a day. Therefore, it makes sense to provide different customer-dependent battery sizes.

Fig. 11. Daily total mileage of German drivers, according to [19]

3.3

Retail price of hybrid vehicles

In 2019, the average price for a new car in Germany was just over € 30,000. Hybrid vehicles only start with a standard configuration at around € 20,000. However, with such a hybrid vehicle, pure electric driving for several kilometer is not possible due to the very light electrification of the powertrain.

16

Fig. 12. Hybrid car benchmark: retail price over performance; bubble size: number of new registrations

3.4

Target Groups

Expanding global markets are another challenge for automobile manufacturers because of different certification and ecological requirements. Especially China and India are growing markets, of which one must take note. Also, the need for mobility varies in these different markets. Tab. 1 lists region-specific traffic profiles for Europe, Japan and the USA [20, 21]. Table 1. Traffic profiles in Europe, Japan and the USA, according to [20] Europa Balanced mix of traffic freeway and urban profile traffic; Dynamic driving profile Status quo Medium-sized vehicles with petrol or diesel engines

Japan City traffic (stop & go, short driving distances); hardly higher speeds Smaller vehicles with small, fuel-efficient petrol engines

USA Significant proportion of highway journeys, some stop & go in mega cities Mostly uniform traffic flow, moderate speeds Large vehicles with highcapacity and high-torque petrol engines

While vehicles in Europe travel a considerable part of their running time at relatively high speeds on motorways, "stop & go"-traffic can be observed more frequently in Japan and the USA, particularly in the urban agglomerations due to the high traffic volume. Especially for this (slow to medium) traffic profile, hybrid vehicles are advantageous. In future, the same will apply to Europe due to growing traffic density in the mega-cities [20]. As well politics is aiming at an environment friendly strategy which can lead to the requirement to drive purely electric in cities.

17

3.5

Fuel consumption

The consumption of fuel and energy in general play a crucial role in the operation of a vehicle, not because of maintenance costs only, but also because fuel consumption is directly related to CO2 and other emissions. An analysis of vehicles registered in Germany between 1985 and 2012 shows that the average consumption related to the NEDC has decreased by 25 to 35 % during this period. Contrary to this downward trend and although the influence of the vehicle mass on fuel consumption is considerable, the mass of vehicles increased by 25 to 50 % over the same period [19]. 3.6

Scenario for an inexpensive hybrid vehicle

Assuming that pure electric driving may be required in cities in the future, hybrid vehicles will have to carry correspondingly large batteries in order to achieve an electric range of approx. 60 to 90 km. Furthermore, if only one vehicle in family is possible, a four-seater vehicle is required. In view of these main requirements, it is apparent that it will be a challenge to remain in an adequate budget. To achieve a cost target of € 20,000, the entire vehicle must be redesigned, resulting in cancelling of technical features. It remains to be seen whether electrically operated seats, electrical mirror adjustments or other add-on features are dispensable. The vehicle and internal combustion engine key data for this scenario are listed below: Vehicle Key Data: Retail Price: Concept: Purely Electrical Range: Power Electric Motor: Electrical Capacity: Driving Mode: Tank: Maximum Speed: Production Quantity:

approx. € 20,000 Four-seater Vehicle with innovative Design on a new Modular Hybrid Platform approx. 60 to 90 km approx. 70 kW approx. 20 kWh Scenario for approx. 70 to 80 % purely Electric Driving approx. 30 l 130 km/h more than 500,000 per Year

Key Data of the Internal Combustion Engine: Manufacturing Costs, with Control Unit, Cooling System, Exhaust System; Suction System: Annual Number of Pieces: Engine/Type: Power: Speed Range: Design: Valve Train Technology: Mixture Formation: Charge Exchange: Others:

less than € 1,500 more than 500,000 a Year approx. 1.3 to 1.5 l 3-Cylinder Gasoline Engine 30–35 kW approx. 2,500 to 3,500 rpm e.g. Monoblock Construction (Gray Cast Iron) 2-Valve without Variable Valve Control; Pushrod Engine (OHV) Manifold Injection naturally aspirated no other Variabilities; everything reduced to a Minimum

18

It makes sense to develop a new vehicle based on an innovative hybrid vehicle.

Fig. 13. Serial hybrid: 1-motor concept (left), serial-parallel hybrid (right) (Explanation: 1 = Engine, 2 = Tank, 3 = Generator, 4 = E-Motor, 5 = Battery, 6 = Clutch, 7 = Differential, 8 = Output)

A serial hybrid similar to the drive concept of the BMW i3 Range Extender is proposed for a powertrain (see Fig. 13 left). However, the internal combustion engine is designed as a simplified three-cylinder in order to achieve a smoother running. A serial-parallel concept with one electric motor is also conceivable. In this case, a switchable clutch is provided between the electric motor and the generator, which – when actuated at higher speeds – let the engine drive the vehicle directly, ensuring a high efficiency (see Fig. 13 right). It is important to check to what extent such a solution is still within the cost target. With this vehicle concept, energy can be loaded easily by refueling, resulting in a range far beyond pure electric driving concepts. It is still even more advantageous, if this concept is combined with alternative or synthetic fuels. By this, the aim to drive CO2-neutral gets even closer. This leads as well to an improvement in a cradle-to-cradle perspective. This concept is also suitable for small vans, normally having a huge amount of stop & go in their traffic profile. By using the developed platform as well for this type of vehicles, the production quantity can be raised even more.

4

Summary/Outlook

Within the framework of the presented investigations, an overview of the known hybrid topologies is given to evaluate the potential of the connection between the internal combustion engine and the electric motor. The evaluation is based on patent applications filed with the European Patent Office, which have been categorized according to a structure developed thereof. For evaluation purposes, the number of applications related to the different topologies and the individual years were considered. The result of these investigations is that Applicants based in Japan, in particular Toyota, have the highest filing activity in the field of hybrid drive technology. German, US and French Applicants are far behind. Suppliers are also strongly represented among European applicants, especially with older applications. Concerning the topologies, in Germany parallel hybrids are mostly filed, e.g. P1, P2, P12 hybrids, making up the largest share. Recently, there has also been a move towards more complex hybrids such as the P3 hybrid and Axle Split. Sporadically, power-split

19

concepts have also been filed, in particular the 2-mode system. In Japan, on the other hand, the power-split hybrid was emphasized. The largest number of property rights concerns the 1-mode system from Toyota. Other Asian OEMs such as Nissan, Mitsubishi and Honda have also developed power-split hybrids, primarily parallel-serial hybrids. The parallel hybrid, on the other hand, has only been filed concerning P1 and P2 hybrids. In the United States, General Motors has filed a large number of applications for the 2-mode system. However, no significant trend can be observed in the American region. For a vehicle with a retail price of approx. € 20,000, a serial hybrid drive concept is presented, in which a redefinition of components is proposed, especially concerning the engine in order to achieve the ambitious cost target. With a compact battery of approx. 20 kWh resulting in an electric range of 60 to 90 km, most of the customer's daily routes can be driven purely electrically. The further success of hybrid vehicles will also depend on the availability of alternative or synthetic fuels. If one considers the cradle-to-cradle situation of the proposed vehicle concept compared to a pure BEV vehicle with a larger battery and the dependency on an overall charging-infrastructure, it is questionable, if the BEV vehicle is the better choice with regard to its CO2 footprint. This considers as well the enormous investment costs for manufacturing a battery for long range and for the charging-infrastructure. The extent to which a return of invest is present, is questionable. The discussion about the further field of application of the internal combustion engine remains exciting.

References 1. Adolf, J.; Balzer, C.; Joedicke, A.; Schabla, E.; Wilbrand, K.; Rommerskirchen, S.; Anders, N.; Maur, A.; Ehrentraut, O.; Krämer, L.; Straßburg, S.: Shell PKW-Szenarien bis 2040. Hamburg, Shell Deutschland, 2014. 2. Ernst, C.-S.; Olschewski, I.; Neumann, N. R.; Harter, C.; Eckstein, L.: Supplier Strategies 2025: Winning Technologies for the CO2 Challenge. In: 24th Aachen Colloquium Automobile and Engine Technology. 2015, S. 747–768 3. Brokate, J.; Özdemir, E. D.; Kugler, U.: Der Pkw-Markt bis 2040: Was das Auto von morgen antreibt: Szenario-Analyse im Auftrag des Mineralölwirtschaftsverbandes. Stuttgart, Deutsches Zentrum für Luft- und Raumfahrt e. V., 2013 4. Koehler, C.; Stahl, T.: Fast Charging Profitability: How Automotive Suppliers can Capitalise on E-Mobility Today. In: 27th Aachen Colloquium Automobile and Engine Technology. 2018, S. 253–262 5. Lüdiger, T.; Wittler, M.: The Electrification of the Powertrain and its Impact on the Machinery Industry and Component Suppliers. In: 27th Aachen Colloquium Automobile and Engine Technology. 2018, S. 1081–1088 6. Bernhart, W.; Ernst, C.-S.; Yoon, M.; Pieper, G.: Challenges and Opportunities in Lithium-Ion Battery Supply. In: 27th Aachen Colloquium Automobile and Engine Technology. 2018, S. 263–280 7. Schulmeister, U.; Eppler, S.; Christ, A.: Roadmap to a de-fossilized powertrain. In: Bargende, M.; Reuss, H.-C.; Wiedemann, J. (Hrsg.): 17. Internationales Stuttgarter Symposium. Wiesbaden. Springer Fachmedien, 2017, S. 279–291

20 8. N.N.: Deutsches Zentrum für Luft- und Raumfahrt e.V.; Wuppertal Institut für Klima Umwelt Energie GmbH: Begleitforschung zu Technologien, Perspektiven und Ökobilanzen der Elektromobilität. Stuttgart, Deutsches Zentrum für Luft- und Raumfahrt e. V., 2015 9. N. N.: Energy technology perspectives: Scenarios & strategies to 2050. Paris, International Energy Agency, 2012 10. Scharf, J.; Ogretewalla, J.: Ottomotoren für Hybridantriebe: Hochtechnologie oder LowCost Aggregate. In: Internationales Wiener Motorensymposium. 2017 11. Plümer, D.: Light Vehicle Prognose 2050. Schlegel und Partner. 2016. URL: www.schlegelundpartner.com/de/news/light-vehicles-forecast-2050/u/1154/?cpage=6 (abgerufen 05/2018) 12. N.N.: Monatliche Neuzulassungen. Daten des Kraftfahrbundesamtes. https://www.kba.de/DE/Statistik/Fahrzeuge/Neuzulassungen/MonatlicheNeuzulassungen/ monatl_neuzulassungen_node.html Daten Internet KBA, (abgerufen 01/2020) 13. Hofmann, P.: Hybridfahrzeuge: Ein alternatives Antriebssystem für die Zukunft. 2. Auflage. Wien, Springer, 2014 14. Reif, K.: Konventioneller Antriebsstrang und Hybridantriebe: Mit Brennstoffzellen und alternativen Kraftstoffen. 1. Auflage. Wiesbaden, Vieweg Teubner, 2010 15. Reif, K.: Kraftfahrzeug-Hybridantriebe: Grundlagen, Komponenten, Systeme, Anwendungen. Wiesbaden, Vieweg Teubner, 2012 16. N. N.: Informationen rund um hybride Autos. Hybrid Autos. 2008. URL: www.hybridautos.info/ (abgerufen 08/2017) 17. Tschöke, H.: Die Elektrifizierung des Antriebsstrangs: Basiswissen. Wiesbaden, Springer Vieweg, 2015 18. Pischinger, S.; Seiffert, U.: Vieweg Handbuch Kraftfahrzeugtechnik. 8. Auflage. Wiesbaden, Springer Vieweg, 2016 19. Schramm, D.; Koppers, M.: Das Automobil im Jahr 2025: Vielfalt der Antriebstechnik. Wiesbaden, Springer Vieweg, 2014 20. Frei, B.: Regelung eines elektromechanischen Getriebes für Hybridfahrzeuge. 1. Auflage. Aachen, Shaker, 2006 21. Balasubramanian, B.: Entwicklungsprozesse für Kraftfahrzeuge unter den Einflüssen von Globalisierung und Lokalisierung. In: Schindler, V.; Sievers, I. (Hrsg.): Forschung für das Auto von morgen: aus Tradition entsteht Zukunft. Berlin. Springer, 2008, S. 349–362

TwinRex – Dedicated Hybrid Engine for a serial-parallel powertrain with excellent cost-value index Matthias Thewes1, Adrian Schloßhauer 1, Oguz Budak1, Jörg Seibel1, Reiner Wohlberg1, Andreas Müller 1, Johannes Moritz Maiterth 2, Markus Eisenbarth2, Ruben Keizer 1, Farouk Odeim1, Michael Kauth1, Rene Savelsberg1, Georg Birmes1, Andreas Balazs1, Tolga Uhlmann1, Johannes Scharf1, Norbert Alt1 and Andreas Sehr 1 2

1 FEV Europe GmbH, Neuenhofstr. 181, 52078 Aachen, Germany Institute for Combustion Engines (VKA), RWTH Aachen University, Forckenbeckstraße 4, 52074 Aachen, Germany

Abstract. Environmental-friendliness is in the focus of customers and has become a dominant driver for powertrain electrification. FEV has developed the TwinRex hybrid system to match lowest CO2-targets and real life fuel efficiency at affordable costs. A simplified transmission enables full electric driving as well as series and parallel hybrid operation modes. The internal combustion engine (ICE) is dedicated to these hybrid modes. ICE-Technology investment focuses on high efficiency in relevant operation zones, while simplifying technology is sufficient for operation zones of lower interest in this highly electrified environment. Two general engine concepts, a turbocharged and a naturally aspirated version are compared in detail. The turbocharged 1.5 l 3-cylinder engine version achieves a peak efficiency of 43 % at stoichiometric combustion. The naturally aspirated engine version achieves 42.5 % efficiency with a bigger cylinder displacement. Due to the higher torque of the turbocharged engine, the electric machines, inverters and the battery can become smaller, making the hybrid powertrain with the turbocharged engine not only more efficient, but also cheaper. Insight is also given on how such a turbocharged 3-cylinder engine can achieve best NVH performance and how the TwinRex system supports achievement of the expected Post-Euro 6 emission legislation. Keywords: High efficiency, Hybrid, TwinRex

1

Introduction

The automotive industry is facing major changes. The public perception of the automobile is changing and thus not only the legal requirements demand sustainable powertrains. However, today's sales numbers of battery electric vehicles (BEV) are strongly inhibited by high costs, a weak battery charging infrastructure and still relatively short driving ranges. Even in 2018, less than 1.5 % of all vehicles sold worldwide were primarily electrically driven. At the same time, to comply with the EU legislation adopted © Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2020 J. Liebl et al. (Hrsg.), Internationaler Motorenkongress 2020, Proceedings, https://doi.org/10.1007/978-3-658-30500-0_2

2

in 2019, car fleet CO2 emission reductions of 37.5 % are required for 2030 in relation to the 2021 starting point, see Fig. 1. Against this background, hybridized powertrains will have a realistic market chance to significantly reduce CO2 emissions in case hurdles for an increase in battery electric vehicles market shares continue to exist throughout the next years. Another important aspect in the promotion of electrified powertrains has always been governmental subsidies. Germany for example is promoting BEV and plug-in hybrid vehicles (PHEV) through direct incentives and a new regulation on company car taxation [1]. In contrast, the latest changes in the Chinese incentive system are likely to lead to a shift to normal hybrid electric vehicles (HEV).

Fig. 1. CO2 emission legislation

High cost remain a key challenge not only for BEV but for any kind of electrified powertrain. Hence, FEV has developed a cost efficient hybrid powertrain system, the TwinRex concept. In previous published work [2], the focus was on the layout of the entire TwinRex powertrain. In this paper, the internal combustion engine will be in the focus. After an introduction to the TwinRex hybrid system concept, this paper aims to identify the ideal dedicated combustion engine for this concept.

2

The TwinRex hybrid system concept

The TwinRex concept can be realized in two different variants, see Fig. 2. The first variant is characterized by the fact that it has only one direct gear between the input shaft and the output shaft. This automatically increases the proportion and importance of the serial operating mode. The second variant has an additional gear stage that can be selected using simple switching elements and electrical actuators. This automatically affects the design of the hybrid components and the hybrid operating strategy. In addition to the available second gear stage, the selection of the operating mode is simpler, since a considerably higher starting torque is also available in electrical operation due to the lower gear ratio. The parallel hybrid mode can be selected at much lower vehicle

3

speeds. This leads directly to the fact that serial operation loses importance. In most situations, a battery charge-sustaining condition can be fulfilled by a combination of electric driving at low speeds and parallel hybrid operation starting from the minimum possible velocity. The electric machines of the two gear hybrid system can be significantly smaller and thus are more cost-effective. Furthermore, the drivetrain can also be operated more efficiently. In order to reduce the costs for the mechanical components, the electric machines are used for speed synchronization. A disadvantage of this approach is the lack of power-shift capability. This, however, can be addressed in almost all driving situations with predictive strategies based on simple camera systems.

Fig. 2. Cost Comparison of different parallel hybrid transmissions, left: market benchmark, center: 1-speed TwinRex, right: 2-speed TwinRex

3

Dedicated hybrid engines

3.1

Naturally aspirated engine

The focus for any dedicated engine is on achievement of highest thermal efficiency while reaching the target performance. To fulfill future emission regulations and avoid expensive solutions for lean exhaust aftertreatment, the engine will be operated in the complete engine map with a stoichiometric relative air/fuel-ratio (without full load enrichment). The limiting part for stoichiometric operation is the three-way catalyst which has to face not only high inlet temperatures but also exothermic reaction energy from converting a stoichiometric mixture even at rated power. Besides classical exhaust gas cooling via a cooled integrated exhaust manifold, a high stroke-bore-ratio of 1.35 has been chosen for an improved combustion with short burn duration and high combustion efficiency which lowers the exhaust gas temperatures due to an early center of combustion. In addition, a high combustion stability is required since the engine is equipped with external cooled exhaust gas recirculation (EGR). The external EGR is used for de-throttling at lower loads as well as for knocking

4

tendency and exhaust gas temperature reduction at higher loads. This requires an optimization of the combustion chamber and the injector spray pattern to increase charge motion, homogenize mixture formation and finally achieve a high combustion quality. The use of virtual development methods is becoming increasingly important in order to shorten the development time and reduce the costs of expensive prototype engines and vehicles. For this purpose, FEV uses the scatter band and 3D-CFD-supported "Charge Motion Design" process. For example, scatter bands are used to pre-optimize the injector spray pattern considering engine parameters such as bore, stroke, etc. Preoptimized designs are checked by CFD simulations and iteratively optimized to improve mixture homogeneity and fuel wall wetting. The detailed 3D-CFD investigations are afterwards used, to model the combustion for the 1D-gas exchange simulations. The FEV 1D/3D combustion model [3] allows a combined approach using the detailed 3D in-cylinder optimization results and use them for the 1D overall engine layout. In the following, investigations of the most important engine design parameters which were are essential for the layout of the internal combustion engine for the intended application are presented. A comparison of a port fuel injection (PFI) concept and a gasoline direct injection (GDI) system is shown in Fig. 3 at a sweet spot operating point. GDI PFI

Efficiency

D = 0.5 %-points

12

13

14

15 16 17 Compression ratio / 1

18

19

Fig. 3. Efficiency potential of GDI vs. PFI at sweet spot operating point

The effect of direct injection is depending on the compression ratio since the knocking tendency is influenced by the injection system. At the same time, the additional friction for the drive of the high pressure fuel pump of the GDI engine needs to be considered. These effects compensate each other at compression ratio of 13 in the shown example. At a compression ratio of about 16, the efficiency with direct injection is approx. 0.4%-points higher than with PFI. The engine efficiency is increasing with higher com-

5

pression ratio due to the increased thermal efficiency until the increasing knocking tendency overcompensates this effect. The optimum compression ratio at this base operation point is at about 15.5. Such high compression ratios limit the maximum power of the engine (see Fig. 4, left). However, because of the moderate rated power target for the intended application, even the highest compression ratio may be chosen. Furthermore, the intake and exhaust system can be optimized to reach a high specific power or to increase the efficiency at part load. A variation of the intake runner length is also shown in Fig. 4 (right). The valve timings (intake and exhaust) have been optimized for each intake runner length separately to optimize the gas exchange. An intake runner length of 450 mm shows the highest achievable power. In contrast, the highest engine efficiency can be reached with a shorter intake runner length of 40 mm. The higher efficiency results from the higher cylinder filling due to an optimized gas exchange and therefore higher EGR capability.

Power

Rated power

D = 5 kW/l

Efficiency

Sweet spot

D = 0.5 %-points 11

13 15 17 Compression ratio / 1

19 0

200 400 600 800 Intake runner length / mm

Fig. 4. Trade-off between thermal efficiency and maximum power at Lambda = 1 depending on the compression ratio and the intake runner length

The global optimum for highest engine efficiency can be found using a Design of Experiment (DoE) approach to find the best combination for valve timings, intake and exhaust runner lengths as well as compression ratio. The test plan and evaluation of the simulation results has been done with FEV xCal [4]. An optimum compression ratio of 16 has been selected to reach high thermal efficiencies in the relevant map region and the 80 kW target power at Lambda = 1 with a target displacement not exceeding 2.0 l. One further key to high efficiency at the sweet spot is lowering the EGR temperature to reduce the knocking tendency even further and to allow higher EGR rates in operation points with full de-throttling. This leads to an increased engine efficiency as shown

6

in Fig. 5. A reduction of the EGR temperature below coolant temperature requires on the one hand side a separate cooling circuit for the EGR cooler, which however is available in the hybrid system. On the other hand, the EGR system needs to be designed in a way that effectively ensures that all condensate is drained during engine operation, e.g. via distribution of the EGR and EGR condensate via an EGR rail.

Efficiency

D = 0.5 %-points

20

30

40

50

60 70 80 90 100 110 120 130 140 EGR temperature / °C

Fig. 5. Benefit of EGR cooling with low temperature cooling circuit at sweet spot operating point

Furthermore, a large cylinder displacement is beneficial in terms of wall heat losses and losses due to unburned fuel. Thus the naturally aspirated engine is realized as a 3-cylinder engine. The corresponding efficiency map is shown in Fig. 6. The above mentioned technologies, as e.g. the high compression ratio combined with external EGR and high stroke-bore-ratio lead to a peak break thermal efficiency of 42.5 %. The peak efficiency operating point is located at medium speeds and full load since the concept avoids reducing the external EGR rate for even better performance but worse real life fuel consumption. The valve timings and the external EGR rate have been optimized for each operating point to achieve maximum efficiency. The rated power target of 80 kW can be reached between 5000-5500 1/min while operating with cooled external EGR and thus high efficiency.

7 250 Efficiency / % 80 kW

Engine torque / Nm

200 42.5

150

100

40

35

50

0

30

1000

2000

3000 4000 Engine speed / (1/min)

5000

6000

Fig. 6. Efficiency map of the dedicated 2.0 l NA GDI hybrid engine

3.2

Turbocharged engine

Also a turbocharged engine for dedicated hybrid application holds considerable potential to achieve high thermal efficiency. In order to understand this potential, some of the most important requirements will be discussed next. Without addressing the topic of gear ratio layout in this section, achieving high vehicle aceleration requires a high load capability of the internal combustion engine at low engine speeds. This is also referred to as good low-end-torque capability. The low-end-torque capability mainly depends on the choice of the turbocharger size and the knock resistance of the base engine. A turbocharger with a turbine, which is small in size, delivers high boost pressure and acts with a lower inertia during the acceleration process. A small turbine, however, is in conflict with the requirement for high turbine flow capacity, which is required to achieve high engine power output. Thus, the optimal turbine size for a conventional internal combustion engine is always a compromise between low-end-torque and maximum power output requirements. Another important limitation for the torque output is the knock resistance of an internal combustion engine. From a fuel quality point of view, a high octane rating improves the knock resistance of the engine at high load operation. From a combustion system point of view, the knock resistance can be improved with an efficient combustion system, early or late intake valve closure, also referred to as Miller cycle, and a low compression ratio. The Miller cycle, though it may improve the knock tendency of an engine, further increases the boost pressure demand. This stresses the above described conflicts regarding the turbocharger layout even further. A low compression ratio, also comes along with a limitation: higher fuel consumption at low to medium engine loads.

8

In conventional engines, the fuel consumption increases at high power operation due to the enrichment of the air/fuel ratio which is required for thermal component protection. A turbocharged internal combustion engine, which is intended for hybrid powertrain application, opens up the possibility for a different layout of the engine torque characteristic. This is shown schematically in Fig. 7.

Fig. 7. Potentials to tailor the torque characteristics of dedicated internal combustions engines for hybrid powertrain applications

On the right-hand side of Fig. 7, a typical efficiency map of a turbocharged engine with 1.5 l displacement is shown. The speed, at which the top left and top right edge of this engine’s map is reached, displays the conflicting requirements for the turbocharger layout. Break thermal efficiency of such an engine nowadays reaches peak values of more than 38 %. In a hybrid architecture, the maximum torque output can be limited to a comparably smaller speed range. Two exemplary efficiency maps of such engines and their corresponding full load curves are shown on the right-hand side of Fig. 7. In the context of a parallel hybrid system (here a P2 position of the electrical machine is selected exemplary), the torque output of the conventional engine at low engine speed is decreased. At low engine speed, the missing torque output for vehicle propulsion is delivered by the P2-machine. The additional power output of the electrical machine reduces the power of the internal combustion engine. Combined with an appropriate

9

gear ratio selection, engine operation at lower speeds is the result. These modifications of the conventional turbocharged engine lead to an optimal turbocharger turbine, which is larger in size. One effect of this increased size is higher thermal efficiency at higher engine loads, due to decreased exhaust back pressure. In addition, the compressor map can be tailored to higher efficiency, while accepting a narrower map width. Also the addition of cooled external exhaust gas recirculation becomes less challenging with the dedicated hybrid engine. An increased stroke-bore ratio improves the combustion quality and the thermal efficiency of this engine even further. Another possible hybrid powertrain architecture is a serial concept layout. For such a hybrid powertrain, the operation range of the internal combustion engine becomes even more reduced in terms of the speed and load ranges. Tailoring the engine to such a limited operating range reduces friction losses and increases the combustion process efficiency compared to the dedicated P2 hybrid engine. For the TwinRex powertrain, a dedicated parallel hybrid turbocharged engine is considered, which incorporates several combustion system optimization steps. These optimizations are stressed to an extent which is considered appropriate for an outlook into near future series engine applications. The efficiency map of this turbocharged engine is shown in Fig. 8. In order to meet the key vehicle attributes, the 1.5 l turbocharged engine in the TwinRex system achieves a maximum power output of 80 kW. This relatively low specific power output enables a high geometric compression ratio of more than 13.5 combined with Miller valve timings and cooled external EGR even at rated power. A highefficiency waste-gate turbocharger, the increased stroke-bore ratio of above 1.3 and a high energy ignition system help to run the engine with highest thermal efficiency and stable combustion, even at high EGR levels. The thermal efficiency is further improved by friction optimization steps such as the application of an electrical water pump combined with an electric split cooling valve or electrically activated piston cooling jets. This allows operation in the sweet spot region with active piston cooling at a low oil pressure level reducing the oil pump drive power. Adding up the efficiency impact of all of these technologies and specifications leads to a peak break thermal efficiency of 43 % and Diesel-engine like fuel consumption in the rated power region.

10 250 Efficiency / % 80 kW

Engine torque / Nm

200

43

150

40

100

35

50

30

0

1000

2000

3000 4000 Engine speed / (1/min)

5000

6000

Fig. 8. Efficiency map of a dedicated 1.5 l TC GDI hybrid engine for the TwinRex hybrid system

4

How the engine choice influences the TwinRex concept

4.1

Sizing of electric machines

The initial layout of the TwinRex concept in [2] was intended to enable attractive fuel consumption savings at moderate costs. The vehicle driving performance was chosen to fit into the competitive environment, but this is explicitly not the approach to create a top performance powertrain within a vehicle lineup with this hybrid variant. Table 1summarizes the main requirements defined upfront for the vehicle of interest which belongs to the C-segment. In order to make a fair comparison between the two hybrid engine concepts, the size of the electric machines must be adjusted in such a way that the driving performance remains comparable. The TwinRex concept from [2] with the 1.5 l TC GDI engine is taken as a basis. This results in a time of 11.3 s for the acceleration from 0 to 100 km/h, see Fig. 9.

11 Table 1. Simplified requirements list and affected components

Maximum velocity  180 km/h shall be fulfilled at all times at a slope of 0 %

 Engine  Transmission system  Electric drive system

Gradeability while driving  120 km/h shall be fulfilled at all times at a slope of 4 %  100 km/h shall be fulfilled at all times at a slope of 6 %  50 km/h shall be fulfilled at all times at a slope of 15 %

 Engine  Transmission system  Electric drive system

Launch torque and acceleration  The vehicle shall be able to operate at a slope of 30 %  0 to 100 km/h shall be reached within a maximum of 12 seconds on a leveled surface

 Engine  Transmission  Electric drive system

CO2 savings according to EU regulations  The hybrid system shall save at least 15 % of CO2 compared to a state-of-the-art competitor

 Engine  Transmission system  Electric drive system

100

400

75

300

50

200

25

100

75

150

50

100

25

50

0

0 8000

75

6000

50

4000

25

2000 0

0

0

0

10

20 30 Time / s

40

0 50

Engine speed / (1/min)

100

200

Battery energy / Wh

100

Vehicle speed / (km/h)

 Engine  Transmission system  Electric drive system

Combined power / kW

Simplification  Use the electric motor as often as possible  Reduce mechanical complexity

Engine power / kW

Affected components

Battery power / kW

Requirements

Fig. 9. Key parameters during full load acceleration with the 1.5 l TC GDI engine

12

400

75

300

50

200

25

100

150

50

100

25

50

0

0 8000

75

6000

50

4000

25

2000

0

0

0

0

10

20 30 Time / s

40

0 50

Vehicle speed / (km/h)

100

75

Engine speed / (1/min)

Engine power / kW

100

200

Battery energy / Wh

Combined power / kW

100

Battery power / kW

The naturally aspirated engine variant has less torque than the turbocharged variant despite the same rated power at all other speeds below the rated speed. As a consequence, the electric motors and battery must be designed to be much more powerful in order to allow the same acceleration time from 0 to 100 km/h and to be able to meet the other driving performance requirements. The peak power capability of the battery must increase from 45 to 60 kW to enable sufficient power boosting during the 0 to 100 km/h acceleration, see Fig. 10.

Fig. 10. Key parameters during full load acceleration with the 2.0 l NA GDI engine

With the lower torque of the NA combustion engine, several other performance requirements cannot be fulfilled in parallel mode anymore. For example, the requirement to enable a gradeability of 15 % at 50 km/h can only be fulfilled in serial hybrid mode whereas it could be fulfilled in parallel mode with the TC GDI engine. Thus, the power of the MG1 must increase from 55 kW to 80 kW and that of the MG2 from 45 kW to 75 kW. These performance figures correspond to those of the TwinRex concept considered in [2] with only one mechanical gear where several performance attributes could only be met in serial hybrid mode as well.

13

4.2

Vehicle fuel consumption

If the hybrid operation strategy is strongly orientated towards achieving optimal fuel consumption, the result is a control strategy with a large serial hybrid operation share close to the minimum fuel consumption range. Since both engines show a similar peak efficiency, the resulting CO2 emissions are comparable. The NEDC as well as the WLTP CO2 emissions of the powertrain with the TC GDI engine are about 0.5 g/km lower than those of the powertrain with the NA engine. The WLTP fuel consumption is higher than that of the NEDC mainly because of the different coast down regulations which result in higher driving resistances for the WLTC.

Fuel consumption (normalized) / %

150

140

130

120

110

100

90

1.5 l TC GDI NEDC

2.0 l NA GDI NEDC

1.5 l TC GDI WLTP

2.0 l NA GDI WLTP

Fig. 11. NEDC and WLTP fuel consumption of a TwinRex system with TC and NA engines

4.3

Cost

The before mentioned 1.5 l TC GDI engine is very cost effective in areas where there is no downside or even an advantage regarding thermal efficiency. This refers e.g. to a low number of cylinders, a grey-cast iron crankcase, a cooled integrated exhaust manifold and the application of a standard wastegate turbocharger. For very cost-sensitive markets, the exhaust camphaser could be omitted. This would result in an additional cost benefit with only a slightly higher fuel consumption in the lowest part load operation region and in the best point of the fuel map. The 2.0 l NA GDI engine itself is even less expensive because it features the same low number of cylinders but has no boosting system. But as it has been pointed out above, the electric machines together with their inverters and the battery need to be more powerful to achieve the same vehicle performance with the 2.0 l NA GDI engine.

14

Overall, the cost balance therefore looks better for the TwinRex powertrain with the 1.5 l TC GDI engine, see Fig. 12.

Cost

+6%

Transmission MG2 MG1 Double Inverter Battery ICE TwinRex with 1.5 l TC GDI

TwinRex with 2.0 l NA GDI

Fig. 12. Cost comparison between a TwinRex system with NA and TC engine

In some markets the car tax which the customers have to pay depend on the engine displacement. For example in China for the engine displacement class up to 1.5 l a tax of 3 % is charged on the vehicle sales price. For the displacement class up to 2.0 l the car tax is already at 5 %. Fig. 13 shows that in such a case that the TwinRex concept with the TC GDI engine would allow for even a higher OEM profit, which increases with a rising vehicle sales price.

20000

D = 100 €

D = 0.2 % 25000

30000

35000 40000 Sales price / €

45000

50000

Potential additional profit margin of TwinRex hybrid with TC instead of NA engine

Potential additional absolute profit of TwinRex hybrid with TC instead of NA engine

15

55000

Fig. 13. Assessment of additional OEM profit potential comparing the TwinRex concept with the TC GDI and the NA GDI engines for the Chinese market

5

NVH development

With the target to develop low cost dedicated engines which achieve a high thermal efficiency, engines with only 3 cylinders are obviously one key parameter to reduce cost and increase efficiency at the same time. On the other hand, such engines are more challenging regarding NVH, and a dedicated NVH development is essential. Compared to engines with 4 or more cylinders, 3-cylinder engines are typically associated with a rougher noise character, low speed booming, and free first order mass moments as long as no balancing shaft is intended to be used. On the other hand, there are advantages with regard to high speed booming and potentially a lower overall airborne sound pressure level, see also Fig. 14.

16

Fig. 14. Interior noise characteristics caused by a 3-cylinder and a 4-cylinder engine in the same vehicle

The different characteristics result mainly from the higher speed fluctuations that occur in three-cylinder engines compared to four-cylinder engines due to the lower number of combustion processes at the same engine speed. The appropriate countermeasures are of course somewhat different. Low speed booming is a driveline effect that can be addressed by devices to manage the dynamics as well as by insulation. Typical measures are two mass flywheels and centrifugal pendulum absorbers. In a hybrid powertrain there are even opportunities by fast in-cycle torque smoothening via the electric machine. The operation strategy of the TwinRex system also enables NVH improvements by avoiding high load operation in the critical low engine speed region unless absolutely mandatory (e.g. during a hard 0-100 km/h acceleration). Gear rattling noises in transmissions occur when during long firing gaps gear tooth flank contacts change. Gear drives within the engine in particular are not decoupled by a dual-mass flywheel and are therefore particularly sensitive. For the drive of the oil pump, for example, a timing belt in oil is therefore a good choice, which, in addition to the NVH advantages, is also low in friction. The drive of a mass balance shaft can be decoupled e.g. via elastomer elements in the gear wheels. Roughness is especially caused by the crankshaft dynamics and the amplification due to resonances. An internal combustion engine optimized for a hybrid powertrain makes it easier to find a good compromise between NVH and efficiency. Engine operation at low loads can be avoided as far as possible so that friction is somewhat less of a problem. Hence, larger main and conrod bearing diameters can be realized, which tend to shift the prioritization towards better NVH. A friction optimization potential also exists through the use of lubricating oils with lower viscosity.

17

Fig. 15. Root causes for different NVH phenomena of 3-cylinder engines

Combined with a cast iron crankcase and suspended GDI injectors, a very good NVH behavior is achieved, also compared to 4-cylinder hybrid engines, which serve as benchmark in Fig. 16.

Fig. 16. NVH rating of the 3-cylinder TwinRex 1.5 l TC GDI engine in comparison to benchmark 4-cylinder engines

18

6

Emission development towards Post-Euro 6

6.1

Expectations for Post-Euro 6 emission standards

FEV’s hypothesis on the key challenges of a Post-Euro 6 emission legislation consists of the following main issues:  General reduction of the emission limits to a level as low as e.g. the China 6b emission limits CO: 500 mg/km HC: 50 mg/km NOx: 35 mg/km  Requirement for Lambda = 1 stoichiometric combustion in the complete engine map (ban on high load air/fuel enrichment for thermal component protection)  Measurement of particle emissions even down to smaller sizes of 10 nm instead of 23 nm  Extension of the real driving emission (RDE) legislation framework to incorporate further emission components as well as stricter boundary conditions to meet the emission limits even for very short urban drive trips. 6.2

Impact of Post-Euro 6 emission standards

Short urban trips will pose the biggest challenge for powertrains with gasoline engines, since most gaseous emissions occur in the first seconds of operation after a cold start until the catalyst has reached sufficient conversion efficiency. Moreover, also the particulate number (PN) emissions are highest directly after engine start while the combustion chamber walls are still cold. PN emission measurement even below 23 nm particle size would further increase the measured emissions by 10–20 %. Further increased fuel injection pressures of 500 bar or even higher can contribute to a PN emission reduction of at least a similar order of magnitude. Also second generation gasoline particle filters (GPF) offering higher filtration efficiency, ash deposition on the GPF before installation and pre-sooting as well as soot load control are effective measures to lower the PN emissions. The TwinRex hybrid system can also help to reduce PN emissions. With conventional vehicles, a big contribution to the total PN emissions is resulting from the first vehicle accelerations directly after engine start. The TwinRex concept can mitigate such sharp load gradients of the internal combustion engine especially after cold start by operation in serial hybrid mode. Meeting the expected gaseous emission limits under worst-case RDE conditions will be the bigger challenge. This refers especially to the hydrocarbon (HC) emission limit at low ambient temperatures during short trips. Fig. 17 depicts the scatterband for engine raw emissions during the first seconds of the US Cold CO test procedure at -7 °C. This test is chosen because the vehicle only starts driving from second 20 onwards. Thus, it allows for an assessment which emissions could be achieved in serial hybrid operation mode where initial catalyst heating could be fully de-coupled from the vehicle speed. Thus, emissions as they occur until ~ second 15 could be expected for the

19

entire period of catalyst heating. The horizontal line shows the HC emission level that would at least needed to be achieved under consideration of a typical cat heating duration to meet the emission limit of 50 mg/km in an 8 km short distance drive scenario. 0.20 0.18

HC emissions / (g/s)

0.16 0.14 0.12 0.10 Estimated minimum requirement to meet Post-Euro 6 with conventional EATS

0.08 0.06 0.04 0.02 0.00

0

2

4

6

8 10 Time / s

12

14

16

18

Fig. 17. HC emissions vs. time after a cold start at -7 °C (US Cold CO test)

For some engines with not too big displacements it seems feasible to achieve such improvements. For other engines, measures to pre-heat the catalyst will become mandatory. This can be achieved using electrically heated catalysts or a burner in the exhaust system. With optimized raw emissions due to catalyst heating in serial hybrid operation mode, electrically heated catalysts on a 12 V voltage level might be sufficient, reducing costs compared to a 48 V electrically heated catalyst which would need an additional DC/DC converter in a high voltage hybrid. It has to be noted though that the cost driven conceptual layout of the TwinRex hybrid powertrain with rather small electric machines will limit the driving performance that can be achieved in serial hybrid operation mode. This has to be accepted during the first seconds of engine operation after a cold start.

7

Summary

The TwinRex hybrid system with two electric machines and a two speed transmission, which uses simple gear shifting elements and electrical actuators had been presented before with the following key conclusions:

20

 The electric machines of the two gear hybrid system can be significantly smaller and thus more cost-effective compared to the classical serial-parallel hybrid concepts with just one mechanical gear.  Drivetrain efficiency is enhanced.  The lack of power-shift capability can be addressed in almost all driving situations with predictive strategies based on simple camera systems. Within this paper, an assessment of the ideal internal combustion engine for the TwinRex hybrid system was carried out. The key conclusions are as follows:  A dedicated turbocharged 3-cylinder gasoline engine with 500 cm³ cylinder displacement achieves thermal efficiencies of more than 43 % with stoichiometric combustion, making it a very attractive alternative to Diesel engines.  A dedicated naturally aspirated engine has a lower efficiency potential and needs a higher compression ratio for this, since its sweet spot is at lower mechanical efficiency.  A naturally aspirated engine benefits from the potential of cooling the external EGR with the same low temperature cooling circuit that also cools other hybrid components. This allows similar intake manifold temperatures as they can be achieved with a dedicated turbocharged engine.  The TwinRex concept with a dedicted turbocharged engine allows for smaller electric machines to achieve the same vehicle performance.  Fuel consumption differs only slightly between the TwinRex concepts with a naturally aspirated engine and a turbocharged engine.  The TwinRex concept can be realized at more attractive costs with the turbocharged 3-cylinder engine and can still achieve very good NVH performance. Thus making it the recommended choice.

References 1. Bundesministerium der Justiz und für Verbraucherschutz: Deutsches Einkommensteuergesetz, https://www.gesetze-im-internet.de/estg/__6.html, last accessed 2019/05/01 2. Savelsberg, R., Maiterth, J. M., Keizer, R., Odeim, F., Birmes, G., Balazs, A., Thewes, M., Scharf, J., Sehr, A.: TwinRex – The twin concept for a serial parallel hybrid system with an excellent cost value index. In: Der Antrieb von morgen 2020, 14. Internationale MTZFachtagung Zukunftsantriebe (2020). 3. Franzke, B., Adomeit, P., Morcinkowski, B., Hoppe, P., Esposito, S.: Potenziale der 1D/3D-Kopplung in der Brennverfahrensentwicklung. In: Ladungswechsel im Verbrennungsmotor, 10. MTZ-Fachtagung (2017) 4. FEV STS Homepage, https://www.fev-sts.com/en/solutions/products/calibration/xcaldoe.html, last accessed 2019/12/31.

Future diesel powertrain in LCV and SUV – electrified, modular platform with focus on emission, efficiency and cost (Zukünftige Dieselantriebe in LCV und SUV – Elektrifizierte, modulare Plattform mit Fokus auf Emission, Effizienz und Kosten) Dr. Wolfgang Schöffmann, Michael Howlett, DI (FH) Bernhard Enzi, DI Stefan Krapf, DI Christoph Sams, DI Hannes Wancura, DI Michael Weißbäck, Dr. Helfried Sorger AVL List GmbH, Hans-List-Platz 1, A-8020 Graz, Österreich

Abstract. Considering worldwide future emission and CO2-legislation for the Light Commercial Vehicle segment a wide range of powertrain variants has to be expected. Dependent on the application use case all powertrain combinations from Diesel-propulsion only via various hybrids to pure battery electric variants will be introduced. Under this aspect as well as facing global legislation updates and market reguirements a modular approach is shown for the LCV and SUV Segment in the displacement range of 2.0L to 2.3L, which allows flexible adjustment for different requirements. In regard of commercial boundaries tailored technology packages, engine related technical features towards emission- and fuel consumption improvement, as well as electrification measures, in particular 48V-MHEV variants, are defined and compared, with cost-effectiveness and efficiency as essential criteria. Kurzfassung. Betrachtet man das Fahrzeugsegment der LCV global für zukünftige weltweite Emissions- und CO2-Gesetzgebung, so wird es hier einen großen Variantenmix betreffend Antriebsstrang geben. Von einem reinen dieselmotorischen Antrieb über verschiedene Hybridisierungsstufen bis hin zu reinen BEV Varianten wird hier je nach Einsatzzweck alle Spielarten geben. Unter diesem Aspekt sowie global unterschiedlichen Gesetzgebungen und Marktanforderungen wird für das Segment LCV und SUV mit 2.0L bis 2.3L Hubvolumen ein modularer Ansatz dargestellt, der flexibel den unterschiedlichsten Anforderungen angepasst werden kann. Hinsichtlich wirtschaftlicher Randbedingungen werden maßgeschneiderte Technologiepakete, direkte MotorenTechnikpakete zur Emissions- und Verbrauchsreduktion, sowie Elektrifizierungsmaßnahmen, insbesondere auch 48V-MHEV, auf Basis einer einheitlichen, globalen Motorenplattform gegenübergestellt, wobei Kosten und Wirtschaftlichkeit wesentliche Kriterien darstellen.

© Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2020 J. Liebl et al. (Hrsg.), Internationaler Motorenkongress 2020, Proceedings, https://doi.org/10.1007/978-3-658-30500-0_3

2

1

Introduction

Future worldwide CO2 scenarios with stringent fleet average fuel consumption targets, increasing customer demand for fuel efficient vehicles as well as significantly enhanced emission limits with more severe testing cycles and boundaries are the most relevant drivers in the development of future powertrain concepts in passenger cars as well as light commercial vehicles (LCV). Looking at the CO2 fleet targets for 2025 (-15%) and 2030 (-31%) compared to baseline which will be defined in 2021 based on WLTP and the subsequent CO2 target reduction, it is obvious that Diesel will be an essential fleet component for cost-effective target achievement for LCV and VANs. The actual fleet values for EU28 of 159 g CO2/km @ approx. 1850 kg based on EEA for 2018 [1] are above the 2020 target of 147 g CO2/km @ 1766 kg within NEDC – see also Fig. 1.

Fig. 1. LCV/Vans CO2 vs. weight with segment share – EU28 Fleet 2018

The actual available data for 2018 indicate an increase of CO2 fleet values within WLTP by approx. 20 to 25% compared to NEDC test cycle [2] – Fig. 2 (left side). Therefore the 2021 WLTP target is expected to be in the range of 175 to 185 g CO2/km (NEDC base 147 g CO2/km). The current diesel shares are shown here in a segment-related manner. Overall in 2018 in the EU28 95% of all utility vehicles (N1 Class) were driven by Diesel powertrains for all segments. Especially in the larger D- & E-Segment the Diesel Share was > 99%. Approx. 60% of all N1-class LCV/VANs belong to the D & E segment in Europe. In 2018 0.75% of all LCV/VANs were sold as BEV. Besides CO2 reduction the introduction of EU7 (exact timing and limit values not defined yet) will be a major factor for the future technology requirements for the LCV powertrains. Additionally, those requirements are influenced by incentivation for electrification combined with possible further access limitation for combustion engines into cities.

3

For EU7 emission achievement we see a 48V system in combination with P0 configuration of the e-motor as baseline. Additionally, a corresponding portion of 48V hybrid powertrains with P2 or P4 configuration and, of course, pure electric vehicles will also be required for future CO2 target achievement.

Fig. 2. LCV/Vans Fleet CO2 Simulation for EU28

For the future line-up of LCV/Vans we see a reduced number of variants of engine and vehicle combinations. The Diesel engine line-up will be streamlined to 2.0 to 2.3L engine for the larger Segment. Depending on sales volumes and worldwide synergies some OEMs will still use 1.5/1.6L Diesel engines in smaller segments. Generally, the modularity of tech packages to meet worldwide requirements in combination with cost efficient hybridization will get more and more important. Fig. 2 (right side) shows an example of a possible technology scenario to achieve 2025 (-15%) and 2030 (-31%). Within this scenario 24% BEV are required in 2030 combined with 76% ICE based powertrains, still mainly Diesel in the larger segments.

2

Emission Technologies for LCV

LCVs up to 3.5t GVW (Class N1) are, like cars with up to 8 seats (M1), certified on the chassis dynamometer. Technology to meet the current EU6d legislation with M1 and N1 includes EGR to reduce engine-out NOx, combined with close-coupled diesel particle filter and NOx reduction by SCR with Urea injection. With reduced RDE conformity factors and in-service conformity requirements, an extension of the SCR capability with a second underfloor substrate and dosing unit becomes necessary to guarantee conversion at higher loads and temperatures. Temperature of the EAS will be managed at cost of a CO2 increase, by deliberately reducing combustion efficiency to generate heat. Based on todays’ known Eu6d standards the discussions for possible future Eu7 limits are ongoing. Different scenarios are under investigation and do differ mainly in the consideration of the nitrogen oxide limits and the length of the RDE city trip.

4

Nitrogen oxide (NOx) limits between 35–50mg/km are seen realistic. For the RDE city trip length distances in the range from 8–10km seem to be very likely. In order to be well prepared for the future, a NOx limit of 35mg/km also for LCV applications combined with 8km distance for the RDE city part, are going to be used in the following considerations. For the engine and exhaust gas aftertreatment system layout the commonization of technology packages between LCV and PC is also seen as a trend. Technologies currently in use on PC applications will also become mandatory for LCV variants. To allow sufficient engine out emission control a combined LP-/HP-EGR system is seen as main route. The exhaust gas aftertreatment system is featuring the SDPF/SCR technology in combination with a double dosing system to sufficiently cover the entire operation range of the vehicle. The essential requirement is a very robust temperature management for the complete EAS, to ensure high NOx-conversion rates immediately after engine start and compliance in a shortened city driving cycle. In this regard, an electrical heating element upstream of the EAS (e-Cat) offers higher flexibility in operating strategy, as well as an improved NOx-CO2 balance, compared to heating via engine combustion [3]. Due to the heating power required (~4kW), a 48V electrical system will be mandatory to realize full potential (Fig. 3).

Fig. 3. Technology package to fulfil future emissions legislation [18]

5

An appropriate layout of the 48V e-machine allows additional functions. The high e-machine torque at low speed supports down-speeding of the ICE, shifting to more efficient operating points in the engine map. Recuperation of kinetic energy during braking contributes further to the CO2 balance and maintaining the battery’s state of charge (SOC). Electrical boost of engine torque reduces engine transients in dynamic driving, eliminating NOx emission peaks, avoiding the need for a storage catalyst. The 48V system is therefore seen as a key technology for future compliance with emissions legislation in the LCV segment, and a widespread introduction is expected – on the one hand allowing the operation of a 4kW electrical heated catalyst – and on the other hand to keep the CO2 benefit of the Diesel propulsion on a high level. Simulations performed for a typical N1ClassIII vehicle in the 2200kg weight class in RDE operation highlights the emission reduction potential in combination with improved CO2 figures derived from the operation strategy optimization provided by a 48V P0 hybrid (Fig. 4).

Fig. 4. RDE results – Full & city trip – Conventional powertrain / + e-cat / + P0 + e-cat

Having here identified measures to cover the next step of emission compliance and allow the achievement of the short term CO2 targets, the further envisaged CO2 fleet reduction targets will also ask for further improvements of the base engine and the possible introduction of P2 and P4 hybridization. Due to convergence in worldwide legislation, the described technology package will be broadly applicable also in other major diesel markets such as China and India, as well as the EU28.

3

Powertrain Layout

The base variants of current ICE powertrains already consistently implement start-stop systems. In many applications the auxiliary drive is equipped with a 12 V start-stop generator and a/c compressor, whereas the power steering servo is generally electrically supported. The same layout is used for the 48 V belt starter-generator (BSG) system that is a prerequisite for electric heated exhaust gas aftertreatment as well as electric supported charging systems, as e-supercharger, respectively e-Turbocharger. A P2-Configuration is the next level in powertrain electrification, still at the 48V level. The parallel configuration as shown in Fig. 5 allows the integration in transverse installations, usually very sensitive regarding overall powertrain length.

6

The configurations with increased electrical performance allow replacement of mechanical functions by electrical solutions, as well as extended recuperation capability, torque assist, load point shifting of the ICE and full electric drive. The trend towards extended demand control lead to increasing electrification of all auxiliaries, thus allowing a simplification of the ICE auxiliary drive. A beltless ICE without auxiliary drives is a consequent route in hybrid PT [12,13, 14]. Integrating the e-Motor into the transmission architecture allows function integration of electrical and mechanical functions in P2 or P2,5 architectures (Fig. 5 Right).

Fig. 5. Electrification Variants 48V – P0 & P2 as well as Hybrid Transmission [17]

AVL is developing a modular Hybrid transmission family, allowing the integration of 48V, as well as HV-E-motors. The concept was already presented at previous conferences [15]. For LCV application two options of a 48V-System, P0 or P2, have most potential [4]. A P0 Hybrid with a 12kW BSG offers start-stop functionality, limited electrical boosting and recuperation capacity. A benefit of this system is that it can be implemented as an “add-on” to a conventional engine architecture, the BSG replacing the 12V alternator. On the downside, increased belt loads lead to higher parasitic losses in the drive. The functionality is also limited because the generator is driving the powertrain via the ICE Crankshaft. To overcome these disadvantages, a P2 Hybrid can be implemented as a compact unit between ICE and gearbox [5]. An electric power of 30kW coupled with increased battery capacity opens up limited electric driving, for example for delivery traffic in cities (Fig. 6). Electrification of auxiliaries like A/C, Vacuum and Steering pumps is possible and when the vehicle can be operated while the ICE is stopped, necessary.

7

Fig. 6. System layout for future LCV powertrains [18]

The add-on cost of the P2 system can be partially compensated by simplification of the ICE [5], for example electrical water and oil pumps, eliminating mechanical drives. These measures also contribute to an increase in overall system efficiency mainly due to enhanced controllability. The CO2 potential of the 48V system will strongly depend on the overall operating strategy as well as battery sizing, since SOC has to be maintained across RDE cycles of unpredictable duration [3,6,7,8].

4

Efficiency and CO2 Reduction Measures

Mandated reductions in fleet CO2 emissions must be met, to avoid penalty payments. This will make technologies previously too expensive for the LCV market more attractive. For example add-on components like variable flow and pressure oil pump, and variable or switchable water pumps have shown their potential and robustness in PCengines and will increasingly be carried over into LCV applications. 4.1

Base Engine Architecture

Wide-ranging conceptual proposals to reduce mechanical losses need to be considered in the product definition phase to cover future fuel economy requirements, since they have a significant influence on the engine concept and the production line. Base layout criteria are:

8

1. Long stroke 2. Crankshaft offset 3. Conrod length (L/r > 3,3) 4. Minimisation of bore distortion, shape honing 5. Steel pistons 6. Valvetrain with low friction (RFF+HLA) 7. Minimised diameter of main bearings 8. Switchable, map controlled piston cooling jets 9. Split cooling 10. Chain driven oil pump, pressure and volume controlled The list of necessary additional measures that can be incorporated with minimal changes to the production and assembly lines, even for existing production facilities, includes: 1. Electronically controlled thermostat/ Thermomanagement module 2. Mass balancing with roller bearings 3. Friction reducing coatings (piston rings, piston pins) 4. Camshaft roller bearings (1st bearing) 5. Switchable high performance water pump (seal, impeller) This efficiency package [16] is the basis for further fuel economy measures for all variants. 4.2

Base Engine Friction Optimization

An equally mechanical and thermally effective concept is required for diesel engines with high peak firing pressure requirements. An aluminium closed deck variant is possible that represents a lightweight and reliable concept. The lowest possible cylinder bore deformation is a prerequisite for the reduction of piston ring pre-tension and hence the reduction of friction. A plate honing process can almost completely compensate for assembly distortion of the cylinder. This process has recently established itself in large volume production, particularly for engines with high firing pressures and hence increased assembly forces. With regard to friction minimisation, the development of honing processes, additionally compensating the thermal distortion came into focus and was introduced in volume production already (Fig. 7). Bore coatings are becoming main stream for all aluminium variants that offer significant advantages in terms of friction in combination with weight reduction and improved cooling conditions, particularly for the high thermal loads present on the bore bridges. A clear target to reduce crankshaft friction is the reduction of the main bearing diameter and crank pin. A natural consequence of reducing the diameter is the increase in specific loading of the bearing with corresponding demands placed on the bearing material. Further optimization of bearing friction with polishing as well as DLCcoatings is in market introduction or in production already.

9

Fig. 7. Friction Optimized Cranktrain [17]

Overall minimization of engine friction by design optimization of cranktrain and cylinder-liner interface is a must. The biggest single contributor to engine FMEP remains the piston-liner interface. The use of a steel piston and a long conrod (r/L 1,2) (Fig. 11).

Fig. 11. K-Factor dependent on compression ratio and stroke/bore ratio

A smaller bore also reduces the maximum gas force on the crankshaft, sofar limiting the mechanical loads under firing and is favourable for the structural stiffness of the head and block. Fig. 12 shows the gas force in the target peak firing pressure range dependent on stroke to bore ratio. On the other hand an extreme long stroke approach is more critical for crankshaft bending- and torsional stiffness and mass balancing.

13

Fig. 12. Gas Force dependent on stroke/bore ratio and peak firing pressure

A steel piston is a proven efficiency measure at current passenger car and light commercial diesel engines with 200 bar PFP. At significantly increased peak pressure levels beyond 220 bar, required for next efficiency improvements however the switch from aluminium to steel is seen even mandatory from mechanical point of view. Of sure with additional benefits of lower compression height, higher combustion chamber wall temperatures, and lower friction. The increase of compression ratio also requires a reduction of the dead volume achievable with lower top land height as well as lower tolerances due to the reduced thermal expansion (Fig. 13).

Fig. 13. Comparison of aluminium versus steel piston

14

Typically, the Crankshaft layout of the diesel engine is limited by the bearing oil film thickness at the Low End Torque point around 1500-1750rpm. In hybrid configurations this point is shifted to higher speeds, low end torque being supported efficiently by the electric motor. Similarly, a reduction in rated speed mitigates the second critical point for the crankshaft layout at high speed and high load. Supporting and optimizing these operation areas with the electric power in hybrid applications allows to achieve a robust design for the increased cylinder pressure, without significantly increasing crankshaft dimensions and hence increase in cranktrain friction (Fig. 14).

Fig. 14. Mechanically Critical Operation areas in the engine map – supported by hybrid system

Best BSFC in the sweet spot is obtained using a Miller approach with asymmetric compression and expansion, using late intake valve closing. For higher performance variants, a switchable inlet valve lift would be beneficial to reduce pumping losses. A flexible valvetrain module allows implementation of roller bearings at the camshafts and can be adapted to add variable intake valve lift if required (Fig. 15). With the combination of above measures a further significant reduction of the Diesel engine efficiency by 10% in best point BSFC from 200 to 180g/kWh is achievable (Fig. 15).

15

Fig. 15. BSFC-Map of an optimized 2,2L-I4-Miller-Diesel Engine for hybrid application

5

Modular engine and powertrain

Implementation of the optimization measures, as summarized above in Fig. 8, in a modular powertrain platform gives flexibility for production and packaging in multiple variants. In the reference example, as shown in Fig. 16, a platform based on a current production engine, with step-by-step development to reach the future emissions and CO2 targets is described. Basis is a 2.2L engine, fulfilling current EU6d legislation in delivery van application. The engine is equipped with a close-coupled sDPF and a high-pressure cooled EGR System. The next development step is to fulfil reduced RDE conformity factors, by addition of an underfloor SCR catalyst with its own Urea-dosing system and an updated EGR Calibration. Adding low-pressure EGR gives a CO2 benefit for the same NOx level; the HP EGR is still required for cold engine conditions. For EU7, a 48V System will be introduced. A step in CO2-efficiency is possible by using the e-Cat to manage EAS temperature instead of purely by internal engine measures like variable exhaust valve timing.

16

Fig. 16. Modular powertrain family: Overview [18]

Within the engine family, different power outputs can be realized through the charging system (Fig. 16 left side). The base rating covers a large market share. A higher power variant can be offered, moving from cost effective fixed geometry WG-TC to a VGT Turbocharger. With 48V available, the variable 2-stage charging system of current engines with high power ratings are replaced by electric-assisted charging systems, either e-Supercharger in combination with a VGT or directly integrated e-VGT. The main engine components and dimensions remain common across the range of applications and performance variants. While the focus of the platform is the N1 segment, variants with higher power ratings can also be applied to larger passenger (M1) vehicles, for example minivans, pick-ups or SUVs. In combination with P2 hybridisation, the ICE can be further optimized as a beltless engine with electric auxiliaries, providing the ideal powertrain for low CO2 at long distance highway driving. The same base engine with high cylinder pressure capability can also be applied in heavier N2 class vehicles above 3,5t. Although the base engine is protected for these applications, the requirements and new or modified components are not directly discussed in this paper.

17

6

Cost Evaluation of the Powertrain Variants

The additional technology packages required to meet future emissions, respectively robustness regarding RDE-requirements, as well as CO2 targets naturally have an impact on the production cost of the overall powertrain. Some of this additional cost can be set off against the penalties in case the CO2 targets are not met. Due to the continuous lowering of the CO2 targets and the reduction of piece cost for advanced technologies, as the production volumes increase, the cost-benefit analysis is dynamic and requires frequent review. Some of the proposed technology elements such as hybridisation represent a major on-cost, however also bring some customer benefits such as electric driving range or city access, which may justify a higher sales price. Furthermore, the introduction of these elements can lead to a reduction in features or requirements on the ICE side, partially offsetting the on-cost. In the following table (Fig. 17) the powertrain architectures and technology packages are summarized.

Fig. 17. Powertrain architectures and technology packages considered for the cost comparison [18].

Based on the technical features of the powertrain variants described above, the relative costs were assessed. The results are summarised in Fig. 18: Main boundaries for the cost comparison are:  The presented cost data are calculated on BOM basis, components as delivered to the assembly line.  Cost changes are calculated based on a volume of >300.000 units per year produced in 2025 in western Europe with an established suppliers base. The baseline (100% Cost line) for the comparison is a current (2020) Diesel engine based powertrain with 6-Speed manual transmission and 12V electrical system, compliant with Euro 6d (temp) legislation.

18

As can be seen in the chart, the main cost driver for reaching Euro7 emissions is the aftertreatment system. The on-costs for the 48V system, include BSG, power network and battery. The add on of the 4kW-e-Cat is included in the EAS-cost in these variants, the cost at least partially compensated by the removal of the exhaust side VVL system, which alternatively controls exhaust gas temperature for emission compliance. The eCat provides significantly higher potential in heating power at considerably lower CO2 penalty. The fully optimized dedicated hybrid Diesel engine as described as D and E is focused on a narrower operating range, which allows a simplification of the Turbocharging system. In combination with a P2 hybrid the auxiliary drive belt is deleted.

Fig. 18. Powertrain Cost and Technology Packages for Performance variants as well as Hybridization variants

Although the contribution of individual measures and features to both CO2 reduction and emission compliance is not exactly measureable, all options considered, fall significantly below the proposed penalty line for all vehicle sements investigated and so become potential options to meet future legislation targets. However, the final selection of the most appropriate technology combinations for each particular OEM would perhaps also need to consider the main powertrain options for the respective vehicle lines.

7

Summary – Outlook

An increasing number of different PT architectures as well as stringent worldwide future emission and CO2-legislation are boundaries for future ICE development for the Light Commercial Vehicle segment.

19

An approach for an affordable lineup of powertrain versions is a modular common engine family architecture with common machining and assembly concepts, as well as the feasibility to integrate different technology packages The powertrain family, as a reference in the paper for LCV as well as SUV applications, requires optimized targets for each individual variant, rather than a traditional common parts concept. So far a modular technology component box is the consequence to cope with future fuel consumption and emission limits, in particular when considering an uncertain distribution of variants in future vehicle platforms. With the evolutionary measures presented above, including 48V electrification, diesel powertrains will be developed to meet future emissions scenarios and contribute significant CO2 reductions to support fleet average targets, at a competitive cost. While BEV-based vehicles will be of increasing interest in inner-city use, the diesel will remain an important player in the worldwide LCV market.

8

Rererences

1. Monitoring of CO2 emissions from vans – Regulation 510/2011, Data 2018 – Provisional “CO2_vans_v13“, https://www.eea.europa.eu/data-and-maps/data/vans-12; Last modified 09 Jul 2019 2. Monitoring of CO2 emissions from Passenger Cars – Regulation 443/2009, Data 2018 – Provisional “CO2_passenger_cars_v17“, https://www.eea.europa.eu/data-andmaps/data/co2-cars-emission-16; Last modified 05 Dec. 2019 3. Mitterecker, H.; Wieser, M.; Weissbäck, M.; Wancura, H.: Dieselmotor als wichtiger Baustein zur CO2-Flottenzielerreichung. MTZ 07/2018 4. Fuckar, G.; Böhm, H.; Schöffmann, W.; Hoffmann, S.: Das 48V P2 Modul für den Quermotor – Von der Auslegung zur Umsetzung. MTZ 09/2019 5. Schöffmann, W.; Sorger, H.; Weissbäck, M.: Effiziente und kostenoptimierte Antriebseinheit für 48V – Systeme. MTZ 05/2017 6. Weissbaeck, M.; Kaup, C.; Mitterecker, H.: Cost efficient CO 2 Reduction – Vehicle Results based on 12V & 48V Architectures. Rouen, SIA Conference 2016 7. Küpper, K.; Pels, T.; Deiml, M.; Angermaier, A.; Bürger, T.; “Efficient Powertrain Solutions for 12V up to 800V” Graz, 27th International AVL Conference “Engine & Environment”, 2015 8. Winkler, M.; Hoffmann, S.; Unterberger, B.; Kaup, C.; Weissbäck, M: Can 48V bridge the gap between 12V and 800V? Graz, 27. International AVL Conference "Engine and Environment", 2015 9. List, H.O.: Propulsion Systems in Transition. Vienna, 39. International Vienna Motor Symposium, 2018 10. Fraidl, G.; Kapus, P.; Mitterecker, H.; Prevedel, K.; Teuschl, G.; Weissbäck, M.: Internal Combustion Engine 4.0. Vienna, 39. International Vienna Motor Symposium, 2018 11. Schöffmann, W.; Sorger, H.; Zieher, F.; Kapus, P.; Weissbäck, M.; von Falck, G.; Rehrl, C.; Hammer, M.; Kukuca, M.; Seiringer, C.; Howlett, M.F.; Prevedel, K..: Realization of Gasoline and Diesel High Performance Engines. Vienna, 36. International Vienna Motor Symposium, 2015

20 12. Sorger, H.; Schöffmann, W.; Schöggl, P.; Hütter, M.; Krenek, T.; Fuckar, G.; Hood, J.; Graf, B.: Vehicle Integration of a new engine concept for 48 Volts – Opportunities for Efficiency improvement and optimization of the overall system complexity. Baden-Baden, 3. International Engine Congress 2016 13. Schöffmann, W.; Sorger, H.; Weissbäck, M.; Pels, T.; Kaup, C.; Brunner, M.: The tailored powertrain for 48V – Options for the Gasoline Engine – Chance for future Diesel Engines. Baden-Baden, 4. International Engine Congress 2017 14. Pels, T.; Davydov, V.; Ellinger, R.; Kaup, C.; Schöffmann, W.; 48V – where to place the e-machine ? Frankfurt, 11th International MTZ Conference on Future Powertrains, 2017 15. Andrašec, I.; Jeitler, B.: AVL’s Future Hybrid X Mode – a modular hybrid transmission family concept for 12V, 48V, HEV and PHEV. 12h International CTI Symposium Automotive Transmissions, HEV and EV Drives, USA, 2018 16. Schöffmann, W.; Sorger, H.; Ennemoser, A.; Priestner, C.; Hütter, M.; Klarin, B.: The impact of 48V to friction and efficiency optimization of the base engine – Approach for quantification in future driving cycles. Esslingen, 5. ATZ Fachtagung Reibungsminimierung im Antriebsstrang 2016 17. Schöffmann, W.; Sorger, H.; Fürhapter, A.; Kapus, P.; Teuschl, G.; Sams, C.: The ICE in the electrified powertrain – modular approach within a common platform between Cost and CO2 optimization. Baden-Baden, 6. International Engine Congress 2019 18. Howlett, M.; Krapf, S.; Enzi, B.; Schöffmann, W.: Modular Platform for Electrified Diesel-Powertrains in Light Commercial Vehicles and SUVs. MTZ 01/2020

MPI valves for use in large engine applications – challenges in the development and derived benefits for operation Peter Christiner1, Claudia Hengstberger 1, Markus Schmitzberger 1, Michael Köhler 1 1

Robert Bosch AG, Linz, Austria

Abstract. Different factors have to be considered during the development of ported fuel injection valves. Customer specifications as well as legal requirements and economical goals pose a challenge during the development process. In this paper the authors presents aspects of an integrated approach for the optimization of ported fuel injection valves for large bore engines in order to meet those requirements with special focus given on current challenges from the market and the achieved benefits for engine operation. Engine concepts like dual-fuel engines help to combine the advantages of diesel and pure natural gas engines, but also increase the demands in regards of robustness and durability of the used fuel injection equipment especially in terms of MPI valves. High EGR rates in combination with a dynamic engine operation lead to an increased need of robust valve design. In order to cope with this market demands and to derive the best possible benefit for the engine operator an integrated approach of component optimization and later validation with component as well as engine tests was chosen to verify the expected benefits of the chosen development approach. Benefits like long term operation of the developed MPI valves and thereby reduced TCO for the engine operator, minimized valve leakage in combination with high valve dynamics and robust behavior under all kind of operation conditions were achieved. Those benefits are described in the paper and linked to the integrated development process. Keywords: MPI Valve, Large Engine, Alternative Fuels

1

Introduction

Due to increasing fuel prices, the share of gas engines on the overall engine market is growing. As the demand for efficient and robust engine solutions is increasing, gas engines have become increasingly important to provide reliable solutions for decentral power supply. Based on continuously growing demand for high power-densities and thereof growing challenges in the area of combustion control concepts in conjunction with requirements of dynamic engine operation, engine concepts with MPI-valves become more important. MPI valves (Multi Point Injection) can help engine manufacturers to cope © Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2020 J. Liebl et al. (Hrsg.), Internationaler Motorenkongress 2020, Proceedings, https://doi.org/10.1007/978-3-658-30500-0_4

2

with steadily shrinking operational windows by offering additional means of air-fuel ratio control. This paper describes the benefits of an integrated development methodology taking customer requirements into account and optimizing the valve concepts to meet this requirements. By applying the described methodology it is possible the derive robust valve solutions, that combine high reliability with optimal adaption to the engine concept providing highest benefits for the customer.

2

Integrated development approach

2.1

Used development methodology

During the development of a new product, the usage of a well-structured development process improves the overall product quality by simultaneously lowering development costs. At Bosch the Bosch Engineering System (BES) a comprehensive development method is widely used to cover all technical aspects over the lifetime of a product, covering the product development phase, the market entry phase and the useful lifetime in order to develop market driven products. Two main phases can be distinguished during the development of a new product. A first phase with high focus on the product itself and the achievement of the defined requirements. A second phase with higher focus on application aspects, mainly how the achieved functional features can be translated to customer benefits. In addition, a phase with a more market driven focus can be added combining both aspects to an integrated approach for development. In Fig. 1 the connection between the pure product development phase and the later application phase is displayed. During the application engineering phase factors like the adaption of the valve size to the customer’s needs and the evaluation of interference of the interface geometry with the valve have to be considered.

Fig. 1. Integrated development methodology.

3

2.2

Meeting customer needs

Product flexibility, as well as a strong customer focus are key factors for success in the gas engine business. The Bosch large engine gas admission valve concept (LEGV) combined with years of experience in development and serial manufacturing allows to react on changing customer requirements rapidly and to provide customized products in short time. Fig. 2 lists the necessary working steps for first sample delivery. Depending on the specific requirements, it is possible to receive first sample parts of the gas admission valve just 3-4 months after first customer inquiry.

Fig. 2. Workflow from request to sample delivery.

2.3

Features of MPI Valves by Bosch

Bosch is the leading provider of a new generation of MPI valves for the usage in large bore gas engines in order to meet the market demand for robust gas admission valves for multipoint port injection (MPI valves). The by Bosch provided valves are capable of dosing fuel gas to the intake port for engine applications with a cylinder specific power output power between 50 and 600+ kW/Cyl. The housing of the valves is costumer specific and can be adapted in a broad range to meet the demand for different interface geometries. The geometric flow area and thus the injected gas mass is defined costumer specific in a range between 50 to 500+ mm2. The used solenoid assembly is the same for all valves The main features of the developed valve family are:       

Cylinder-individual gas admission (MPI) for SI & DF engines Modular design with identical functional group in different applications incl. Marine Customer individual flow rates (engine power 50…600+ kW/cyl.) compatibility with existing solutions Gas tight solenoid (safety) Easy serviceability Optimized installation space with co-axial or side-feed gas flow

4

The valves are capable to be used with both CNG and LNG. The flow rate can be adjusted to any needed mass flow in a range between 80 to 1200 kg/h depending on the geometric flow area of the valve and the operation conditions. In addition the operation conditions regarding pressure and temperature are summarized. Several customer benefits could be obtained throughout the extended development process. An increased lifetime of 720 Mio. operation cycles could be achieved by optimizing the stress level on the main components of the valve. Therefor extended service intervals and increased engine on time with reduced TCO is possible. The taken measures will be described in detail later in this paper. Additionally the internal leakage of the valves was minimized. The leakage is only 0.1% of the maximum static mass flow providing optimized efficiency and reduced engine cycle to cycle variations (COV IMEP). As result engine systems costs can be reduced (venting devices of engine intake port,..). By optimizing the layout of the magnetic circuit of the solenoid within the valve, high valve dynamics were attained. Therefore accurate gas admission strategies and improved engine load change behavior is possible.

3

Development of filters for improved robustness

All variants of the LEGV are delivered with a filter at the gas inlet. This filter acts as a last chance filter. It ensures a correct function of the valve, even in case of particles in the fuel, which have not been caught by the filters on the engine. This filter is one of the most important components of the LEGV and Bosch suggests to never using a gas valve without a filter. This applies, regardless which gaseous fuel is used. In Fig. 3 all available filter variants for all valve types produced by Bosch are depicted.

5

Fig. 3. Picture all available filter variants.

3.1

The importance of filters for gas valves

For understanding why this last chance filter is so important, one has to look at the inside of the LEGV and its function. To inject fuel into the engine manifold, the energized magnet pulls the moveable valve plate away from the stationary valve seat. At the end of the injection process, the energizing of the magnet stops and spring forces and pressure forces push the valve plate back onto the valve seat. However, if there are particles in the fuel, a particle could get stuck between the valve plate and the valve seat. This prevents the valve from fully closing, which leads to over fueling and could damage the engine. This needs to be avoided in any case. The mesh size of the LEGV last chance filter is designed to keep all particles from entering the valve, which are big enough to keep the valve open unintentionally. Particles smaller than the mesh size cannot get stuck somewhere inside the valve. 3.2

Designing filters

When designing and dimensioning filters there are three main criteria to consider. These are the chemical resistance against the fuel including possible contaminates, the long-term durability and the throttling of the filter. The choice of materials and the dimensions of the filter mesh are the variables that are defined in the development process in order to meet the requirements.

6

Filter material. First the material, for both the filter mesh as well as the filter frame, has to be selected The filter is subjected to a wide range of temperatures and is also exposed to all different kinds of fuels, since the LEGV can be operated with various types of gas. No corrosion at all is allowed on the material, because this would decrease the mechanical strength of the filter. In addition, the filter is also exposed to high forces, due to pressure pulsations. Further restrictions for the material also come from requirements for certification of the valves apply, especially if the valve is used in marine applications. As part of the development process Bosch verifies the correct choice of materials in different endurance runs. These endurance runs include end-of-life tests at the LEGV test benches at the Bosch site in Linz, Austria as well as vibration tests and pressure tests. Furthermore, there are material tests, similar to the ones described in chapter 4.2. Filter dimensions. After choosing the right material, the dimension and type of filter mesh is defined. Different types of filter meshes result from different ways of weaving the mesh and also depend on how many layers of mesh is used. Both the size and the type of the filter mesh influence its mechanical strength and the mass flow through the filter and subsequently the mass flow through the gas valve. Looking at the mass flow through the valve, the mesh of the filter should be as big as possible. In that case, the throttling effect of the filter on the mass flow would be as low as possible. However, to achieve a high mechanical strength, a smaller mesh size is preferable. These two requirements of minimum throttling and maximum mechanical strength are somewhat conflicting. Fig. 8 shows the decrease of mass flow through the valve over a decrease in mesh size. It also shows the increase of the mechanical strength of the filter over a decrease in mesh size. To fulfill both requirements best, one has to look at the crossing of the two lines.

Mass Flow; Mechanical Strength

7

Optimum

Smaller mesh size

Mesh Size Mass Flow

Mechanical Strength

Fig. 4. Mass flow through the valve for different mesh sizes and mechanical strength of the filter for different mesh sizes.

To confirm and verify the correct dimensioning, mainly functional tests, but also endurance runs are conducted. In the functional tests, the mass flow through the valves is analyzed. Even different meshes can be compared to evaluate their influence on the mass flow. Moreover, in the functional tests also the opening and closing times of the valves are determined and investigated. For comparison, also valves without filters can be measured, to get an idea of the influence of the filter on the valve. All these investigations are performed on the LEGV tests benches.

4

Considering Dual-Fuel Applications and EGR in the development of the LEGV

The Bosch gas valve is suitable for both monovalent gas engines, as well as dual-fuel applications. In the development of the LEGV this influences the choice of the materials of the valve parts. The most challenging requirements concerning corrosion and chemical resistance derive from high EGR rates, which occur especially in dual-fuel engines. Furthermore, considering dual-fuel applications also influences the design of the compressions springs between the valve plate and the valve seat. These springs are necessary to withstand a negative differential pressure and keep the valve from opening unintentionally in diesel mode. 4.1

Dimensioning compression springs

During conventional use of the LEGV, i.e. on a monovalent engine or during gas mode on dual-fuel engines, there is a positive differential pressure between the gas pipe and

8

the engine manifold. However, in case of using diesel mode on a dual-fuel engine, there is no pressure in the gas pipe, but still pressure in the engine manifold. This leads to a negative differential pressure at the gas valve. Thus, there are no pressure forces inside the valve pushing the valve plate onto the valve seat. The pressure in the engine manifold could even possibly push the valve plate away from the valve seat. Air from the manifold could enter the valve and the gas pipe. To prevent this, there are several compression springs, which push the valve plate onto the valve seat and keep the valve fully closed. Choosing the right spring force. Designing the springs and dimensioning their force is a complex optimization task. On the one hand, the force should be as high as possible to withstand high negative differential pressures. This would also be an advantage in order to achieve short closing times of the valve. On the other hand, high compression forces lead to higher wear over lifetime and longer opening times of the valve. Wear between the valve plate and the valve seat leads to an increase in mass flow and to an increase of internal leakage over lifetime. This, of course, has to be avoided.

Fig. 5. Simulation model to calculate the contact pressure due to spring forces.

For determining the necessary forces of the springs, both an analytical approach as well as a simulation is used. The simulation model is shown in Fig. 5. It determines the contact pressure between the valve plate and valve seat at the sealing grooves due to the compression springs. With the help of the simulation model, different designs can be investigated easily. It is possible, for example, to investigate the influence of changing the number of grooves in the valve seat. Verifying the correct dimensions. To validate the theoretical results of the spring design, functional tests on the LEGV test bench are performed. These tests determine the maximum negative differential pressure, up to which the valve strays closed.

9

In all test points, the magnet is not energized, to simulate the diesel mode on dualfuel engines. In the first test point the differential pressure is ∆p = 0 bar. In all following test points, the differential pressure is further decreased in steps of 0.1 bar. In each test point the mass flow through the valve is measured. As long as there is no measureable mass flow through the valve, the valve is still closed. The test point before the first one with a measureable mass flow through the valve provides the information of the maximum negative differential pressure. Another part of the verification process are endurance runs. At the end of the endurance runs, wear on all parts of the valve is investigated. If the wear on the valve plate and the valve seat is high, this could indicate that the force of the compression springs is too strong. 4.2

Choosing robust materials

In order to make the LEGV suitable for dual-fuel application and EGR usage, the Bosch development team put a lot of effort into the choice of the materials at the very beginning of the gas valve development. Especially the valve parts close to the engine manifold need to have a high corrosion resistance. However, also the other parts of the valve need to have robust materials. Bosch allows the usage of a wide range of different gaseous fuels. This leads to challenging tasks during the development, due to varying gas composition and gas quality. To test materials for its chemical resistance to corrosion, standardized material samples are stored in different acid solutions. Both the material samples, as well as the storage in the sample container can be seen in Fig. 6. The solutions represent different, but precisely specified types of exhaust gas condensate. The solutions used cover all possible actual occurring exhaust gases. To simulate worst-case conditions the containers are put inside an oven and are subjected to high temperatures during the test. At the end of each test cycle the material loss of all samples is measured and the surface of the samples are thoroughly analyzed with the help of different methods, including a scanning electron microscope (SEM).

Fig. 6. Material samples (on the left) and sample containers inside an oven (on the right).

10

5

Impact factors on optimum valve performance

From Bosch point of view, it is not sufficient to develop and deliver a gas admission valve of fixed valve size and leave the adaption of this standard product to the customer. The main challenge is to support the customer to find a perfect alignment of gas admission valve and gas engine to achieve maximum overall valve performance and maximum advantages for the customer. To ensure this strong information share between Bosch and engine manufacturer is necessary to take all different impact factors on valve performance into account. 5.1

Impact of engine map

The engine map is the main information source, to derive the boundary conditions for valve operation. Especially for the use in Genset applications, the customer is interested in good performance within the engine full load point and smooth operation during Idle mode. For the gas admission valve, each operation point is characterized by a defined required gas amount (gas/shot) and suction pipe pressure (p2). The left part of Fig. 7 shows a LEGV performance map for specified full load conditions. Each graph represents a different valve size (LEGVMax1 > LEGV3 > LEGVMin) and the necessary combinations of energizing time and differential pressure to supply the requested gas amount/shot for the specified suction pipe pressure. In general a maximum allowed energizing time tE_max_engine is specified, as well as the maximum available gas supply pressure within the engine system (p1 => ∆pmax_engine = p1max-p2). The figure shows, by increasing valve size and differential pressure, the necessary energizing time for the full load point can be minimized (tE_LEGVMax 51). The AFIDA features a constant volume combustion chamber method and shows good agreement with the CN measurement from the CFR engine [11]. Using a high-pressure injection system a fuel sample was compressed up to 1000 bar and injected into a heated and pressurized combustion chamber using a piezo injector. The calibration of the device was carried out with seven primary reference fuel blends, covering an ICN range from 35 to 85. The measurements were conducted under standard measurement conditions with air (20.9 ± 0.5 % O2) compressed up to 17.5 bar at an end-temperature of 580 °C. These conditions ensured single-stage combustion of the primary reference fuel blends. Details of the AFIDA device and testing method have been summarized in previous literature [11]. The ignition delay was measured for twelve shots which resulted in an acceptable standard deviation of less than 1 %. 2.2

Characterization of the Fuel Properties

As aforementioned, the density was used to assess the volumetric energy content of the different blends [12]. Additionally, the viscosity was measured in order to provide inputs for the 3D-CFD models. The density was measured in accordance with DIN EN ISO 12185:1996 with a Stabinger Viscometer (SVMTM 3001 manufactured by Anton-Paar GmbH, Austria) [13]. The viscosity was measured coincidently with the same device. Both properties were determined over a temperature range of 288 K to 398 K. All the relevant properties normed in EN590 standards for diesel [14] were analyzed in a preliminary work and are not discussed here [10].

4

2.3

Heavy Duty Single Cylinder Engine

The HD SCE was derived from a six-cylinder heavy duty commercial vehicle engine of N3 class complaint to Euro VI stage C. The engine specifications are listed in Table 1. The engine featured a common rail fuel injector with a built-in pressure intensifier and an in-house developed prototype electronic control unit with a model-based fuel path control. Such a prototype control system provided a wide range of flexibility especially in adjusting relative separation and energizing duration of main injection, pilot injection and pressure intensifier. The exhaust gas recirculation (EGR) rate was derived from the CO2 concentration in the intake runner. The exhaust back pressure (pexh) was regulated with two flow control butterfly valves: one valve for a faster control and the other for a finer control of pexh. The regulated emissions were measured at the engine exhaust. The measurement line for the unburned hydrocarbon (HC), carbon monoxide (CO) and nitrogen oxides (NOx) was pre-heated to a temperature of 200 °C to avoid condensation. The measurement devices used are listed in Table 2. Table 1. HD SCE specifications.

Parameter Displacement Stroke Bore Compression ratio Max. cylinder pressure Max. injection pressure Max. rail pressure EGR Injection system Injection nozzle Number of holes Injection nozzle cone angle

Unit liters mm mm bar bar bar cm³/30s °

Value 2.13 156 132 18.3 250 2700 1200 Cooled high pressure EGR CRIN 4.2 Bosch 850 8 142

Table 2. Measuring devices for the HD SCE.

Parameter CO, CO2, NOx, HC

Device FEVER NDIR, CLD, FID

Filter Smoke Number (FSN) Fuel flow

AVL 415S Emerson CMF010 Coriolis Kistler 6044 A

Combustion pressure sensor

Range CO ~ 0 – 5000 ppm CO2 ~ 0 – 20 %-vol NO/ NOx ~ 0 – 3000 ppm THC ~ 0 – 3000 ppm C3 0 – 10 FSN 0 – 120 kg/h 0 – 300 bar

5

Heavy Duty Single Cylinder Engine Testing Approach. The combustion and emission behavior of the fuels under consideration were assessed by testing these fuels at a given load for different EGR rates. Among the four selected load points presented in Fig. 1, ranging from low part loads up to rated power operation, the fuels were screened in this first stage of the REDIFUEL project at cruise point operation. At this engine part load, the in-cylinder thermodynamic conditions like temperature, pressure, and turbulence are expected to be moderate enough to permit the assessment of the ignition behavior of the fuels. This part load was preferred because it is relevant for both the world harmonized stationary cycle (WHSC) and world harmonized transient cycle (WHTC) for emission regulation. Hence, significant changes in emissions like particulate matter (PM), CO and HC can be detected. The given fuel were screened at a Euro VI-c base indicated specific nitrogen oxides (ISNOx) level of 5.8 g/kWh. While performing the EGR sweeps, a constant center of combustion (Q50) was kept in order to avoid impact on emissions of retarded combustion. The fuels were compared at the same engine indicated mean effective pressure (IMEP) and the relevant engine calibration parameters like, injection pressure (pinj), boost pressure (pboost) and temperature (Tboost) defined in Fig. 1.

Fig. 1. Explicative engine load map in break mean effective pressure (BMEP) vs engine speed (a) and engine calibration parameters at cruise point operation (b).

2.4

Numerical Modeling

Reynolds-averaged Navier–Stokes numerical simulations were performed using the CFD code provided by CONVERGE. CONVERGE is a general-purpose CFD tool that automates the mesh generation process and the adaptive mesh refinement (AMR) algorithm [15]. In particular, the AMR delivers small grid size where high temperature and velocity gradients are calculated without significantly increasing the total number of computational cells. A full mesh was adopted during the 3D-CFD gas exchange simulations, while a sector mesh was used for the mixture formation simulations, as shown in Fig. 2 (left). The mixture formation cases simulated here featured a base grid of 1.4 mm and an additional mesh refinement yielding a local grid of minimum 0.35 mm.

6

The Renormalization Group k-ε equations were chosen to model the turbulence. The Kelvin-Helmholtz and Rayleigh-Taylor model was used to capture the fuel spray breakup [16]. The mass, momentum, and energy were calculated at each node of the grid using the unsteady Navier- Stokes equations, which were implicitly discretized based on a finite volume method over the Cartesian grid [17]. For the analysis of the droplet evaporation, the FROSSLING model was selected [18]. The droplet collision was analyzed by using the no time counter collision model [18] and the blob injection model was used to inject liquid parcels inside the computational domain [18]. For solving the pressure-velocity coupling the pressure implicit for the splitting of operator algorithm was used [17]. The O’Rourke and Amsden model was set for the wall heat transfer calculations [18]. For the analysis of the spray wall interactions, the Rebound/slide model was adopted [18]. To assess the quality of the mixture formation process the characteristic numbers Air utilisation (AU) and Oxidation Potential Number (OPN) were used. The AU indicates the volumetric fraction of air inside the combustion chamber, sorted per air/fuel equivalence ratios (λ) ranges [19]. The AU is evaluated at a given engine operating condition, from starting of injection (SOI) to exhaust valve opening (EVO). The OPN is a number based on the AU. It represents the share of lean mixture (i.e. Air utilization between 1 < 𝜆൏2) divided by the rich one (𝜆 < 1) and by the unused air (𝜆 > 2) [20] as shown in Fig. 2 (right). The higher the OPN, the better the expected soot oxidation pontential. Mesh embedding during valve opening (cell size 0.5 mm) Sector Mesh

CFD calculation w/o combustion

Unused air ( > 2.0) Lean mixture (1.0 <  < 2.0) Rich mixture ( < 1.0)

Air Utilisation / 1

Liner

Permanent fixed embedding at boundary wall (cell size 0.5 mm)

SOI

EVO

1.0 0.8 0.6 Soot oxidation 0.4 0.2 Soot formation 0.0 160 180 200 220 240 260 280 300  / °CA ABDC

Fig. 2. Computational mesh for gas exchange and mixture formation simulations illustrating the fixed embedding (left) and air utilization (right).

2.5

Estimation of Fuel Properties

Since the real fuel blends are a complex mixture of hundreds of molecules, surrogate liquid molecules, which represent the group properties of the fuel blends were used for the CFD simulations. The thermo-physical (i.e., specific heat capacity, thermal conductivity, and heat of vaporization) and physicochemical properties (i.e., viscosity, density, and surface tension) are required to model the droplet break-up and evaporation in CFD. These properties are necessary over a wide range of temperatures, i.e. from the melting

7

temperature (Tm) to the critical temperature (Tc). In addition, molecules with similar ignition delay and combustion properties are needed to model the gas phase properties. Here, the AFIDA was adopted to find surrogates for the gas phase properties of the blends under consideration. The reaction kinetics was not modeled in this study, due to the unavailability of the reaction mechanism for the multi-component mixtures. In this paper, simulations were performed for a mixture of diesel, GtL, and SAM (C6-C11 alcohols). For diesel fuel, diesel-2 was taken as a liquid phase surrogate, which is widely used [21]. As 1-octanol is roughly in the middle of the composition of the SAM, it was chosen as its liquid surrogate. Since GtL is a mixture of different hydrocarbons and finding its exact composition is out of the scope of this paper, a composition determined in a previous literature work was adopted. Herein, GtL was modeled as 45 vol% cyclo-octane, 51 vol% iso-cetane and 4 vol% n-decane a mixture of cyclooctane, iso-cetane, and n-decane was used as a GTL liquid surrogate [22]. The thermo-physical and physicochemical properties for GtL and SAM were estimated and extrapolated using different models and mixing rules available in the literature [23, 24]. The measurements of the dynamic viscosity and density, explained earlier in part 2.2, were used to validate the prediction and extrapolation models adopted in this study. For the estimation of dynamic viscosity of pure components, the Andrade equation [25] was used. 𝜇 = 𝑒 (𝐴+𝐵/𝑇)

(1)

where: ─ 𝜇 is the dynamic viscosity; ─ T is the temperature; ─ A and B are constants to be determined using Eqs. 1.1-1.2 show which requires at least two sets of data for viscosity/temperature. A = ln ( 𝜇1 ) − ( 𝜇

1 ) 𝑇1

1

1

2

1

B = ln ( 𝜇2) / ( 𝑇 − 𝑇 ) 1

(1.1) (1.2)

The density for pure components was estimated using Yaws regression equation [24]. 𝑇 𝑛

𝜌 =𝐴𝐵 (1−C)

(2)

where: ─ ρ is density; ─ T is temperature; ─ A, B, C, and n are regression constants. The other properties (i.e. vapor pressure, surface tension, thermal conductivity specific heat and heat of vaporization, etc.) were modeled using equations listed in Table 3. However, these properties could not be validated against measurements, due to the unavailability of experimental data. Hence, only the density and viscosity matching was

8

used as a criterion to select the proper liquid fuel surrogate. The properties of each multi-component fuel mixture were estimated with the following Arrhenius mixing rule presented in Eq. 3 [31]. 𝑥

𝑥

(3)

𝜇 = 𝜇𝐴𝐴 𝜇𝐵𝐵 Table 3. Equations used for estimation of properties for pure components.

Property

Equation used

Viscosity (𝜇) Density (ρ)

[25]

𝜇= 𝑒

[24]

(𝐴+

𝐵 ) 𝑇

𝜌 =𝐴𝐵 (1−

Vapor pressure (pv) Surface tension (σ)

𝑇 𝑛 ) 𝐶

𝐵 ( 𝐴 – (𝐶+𝑇) )

[26] [27]

𝑝𝑉 = 10

[24]

𝜎 = 𝐴 (1−

𝑇 𝑛 ) 𝐵 2

Thermal conductivity (λ)

[28] [29]

𝑇 (7) =𝐴+𝐵 (1− ) 𝐶 A ( 1 − 𝑇𝑟 )0.38

λ 𝑙𝑜𝑔10

λ=

1 ( )

𝑇𝑟 6 Specific heat (cp)

[30]

Heat of vaporization (∆Hv)

3

𝑐𝑝 = 𝐴 + 𝐵 𝑇 + 𝐶 𝑇 2 + 𝐷 𝑇 3 + 𝐸 𝑇 4 [24]

𝛥𝐻𝑣 = A ( 1 −

𝑇 𝑛 ) 𝑇𝑐

Results

In this section, the experimental and numerical results are discussed. First, the results of ignition delay and density measurements are discussed. Based on these results the most promising blends were tested at the HD SCE and also simulated in 3D-CFD, to have an in-depth understanding of the mixture formation process. 3.1

Ignition Delay Characterization

As aforementioned, SAM was used to represent the real end-product and mainly CN and density were analyzed to select drop-in mixtures compliant with the EN590 norm. Table 4 lists the first results for the density and viscosity screening and provides evidences that SAM is a good representative for the real alcohol mixture (RAM) produced on a lab-scale level. The left-hand side of Fig. 3 presents the characterization of the

9

ignition behavior of different blends of SAM with GtL. Here, each fuel and blend is represented with a scatter and distinguished with symbol type. Additionally, 1-octanol and a mixture of n-heptane and n-dodecane are analyzed too. Table 4. Comparison of the alcohol mixtures.

Characteristics Density (kg/m3) Viscosity (mm2/s) Cetane number

Surrogate Alcohol Mixture (SAM) 828.1 5.66 34.1

Real Alcohol Mixture (RAM) 829.5 5.62 33.8

This first screening served to assess the impact of different blending proportions between the two main REDIFUEL components (i.e. SAM and GtL) on the ignition delay behavior. An increase in ignition delay proportional to an increase of SAM volumetric content in GtL can be observed. As a CN of 51 is the minimum requirement in the EN590 norm, a reference horizontal line is represented. Among the different blends tested, only the ones featuring a maximum of 40 vol% SAM can meet the CN limit. However, since the fuel production process can ensure a maximum of 30 vol% of RAM in the end-product, this blend cannot be further investigated and represents only a theoretical maximum. Moreover, it can be also seen that SAM and RAM feature very similar ignition behavior, with respectively 34.1 and 33.8 ICN. 1-Octanol scoring an ICN of 34 confirms to be a good surrogate for SAM for the gas phase properties. The gasphase properties of the alcoholic components are modeled with 1-octanol, due to the very close ICN as shown in Fig. 3. n-Heptane is chosen as a surrogate for diesel [23]. A mixture of 20 vol% n-heptane and n-dodecane 80 vol% resulted in the same ICN as GtL (see Fig. 3 left plane) and it is chosen as a gas phase surrogate for this fuel. The right-hand side of Fig. 3 shows a trade-off between ICN and density at 15 °C. Here, vertical reference lines are introduced to indicate the EN590 minimum density value of 820 kg/m3 in summer time [16] and the artic grading value of 800 g/m³ for the class 1A. Due to the high shares of paraffins in the blends, the product’s density for the RFA30P70 (791.7 kg/m³) is lower than the minimum limit for EN590 as well as below the artic grading value of 800 kg/m³. Fig. 4 presents the results of the RFA30P70 mixed in different blending proportions with diesel, termed as B0 in the plots to indicate that it did not feature biofuel content. Since 100% B0 featured an ICN of 52.6, in the left-hand side of Fig. 4 it can be seen that all blends satisfy the minimum requirement for CN of the EN590 norms but that the ICN decreases with B0 content, as neat RFA30P70 features a ICN of 56.9. On the right-hand side of Fig. 4 it can be seen that the density of the blends increases with an increasing share of B0. The blend with 60 vol% diesel and 40 vol% RFA30P70 satisfies both the minimum requirement for CN and density stated by the EN590 norm. Thus, this blend was chosen for future engine testing and 3D-CFD simulations work.

10

100 % SAM 70 % SAM / 30 % GtL 50 % SAM / 50 % GtL 40 % SAM / 60 % GtL 15 % SAM / 85 % GtL 100 % GtL 20 % n-heptane / 80 % n-dodecane

80 70 60

EN 590 min CN = 51

50 40 30 1.50

2.30 3.10 3.90 Ignition delay / ms

4.70

Indicated Cetane Number / -

Indicated Cetane Number / -

100 % RAM 60 % SAM / 40 % GtL 30 % SAM / 70 % GtL 100 % 1-Octanol

80 70

EN 590 min density 15°C

60 50 40 30 770

790 810 830 Density / (kg/m3)

850

Fig. 3. Indicated Cetane Number determination of Surrogate alcoholic mixture (SAM), real product alcohol mixture (RAM) and blends of SAM and GtL.

75

Calibration curve

70 65 60 55

EN 590 min CN = 51

50 1.50

1.75 2.00 2.25 Ignition delay / ms

2.50

Indicated Cetane Number / -

Indicated Cetane Number / -

Blends of B0 (EN590 diesel) and RFA30P70 (Blend of 30 % SAM and 70 % GtL) 20 % B0 / 80 % RFA30P70 50 % B0 / 50 % RFA30P70 100 % RFA30P70 100 % B0 60 % B0 / 40 % RFA30P70 80 % B0 / 20 % RFA30P70 100 % GtL 75 70

EN 590 min density 15°C

65 60 55 50 770

790 810 830 Density / (kg/m3)

850

Fig. 4. Indicated Cetane Number determination of blends of RFA30P70 with diesel.

3.2

Test Bench Results

In the following section, the results from the limited screening of the fuel blends on the HD SCE are discussed. In Table 5, the relevant fuel properties of the blends are mentioned, as the CN and oxygen content increase, the carbon content and the calorific value decrease with an increase in the RFA30P70 share. In Fig. 5, the results for the cruise point operation at the base ISNOx of 5.8 g/kWh are presented. The different fuel blends are displayed in bar charts and distinguished by hatching style. The indicated specific particulate matter emissions (ISPM) are presented in the top-left plot in Fig. 5. Generally, adding the RFA30P70 to diesel reduces the ISPM emission.

11 Table 5. Fuel blend properties.

Carbon mass fraction 0.865

Hydrogen mass fraction 0.138

Oxygen mass fraction % 0

0.855

0.140

60 % B0 + 40 % RFA30P70

0.844

50 % B0 + 50 % RFA30P70 20 % B0 + 80 % RFA30P70

Fuel

100 % B0 Diesel 80 % B0 + 20 % RFA30P70

Calorific value

Cetane number

MJ/kg 42.9

52.1

0.734

42.7

52.5

0.142

1.467

42.5

53.6

0.839

0.143

1.83

42.3

54.1

0.823

0.147

2.94

42.0

56.7

This can be attributed to the presence of oxygen and paraffinic molecules in the RFA30P70. Less soot precursors are generally formed when diluting diesel with a straightchained oxygenated mixture, as shown in the literature [32, 33]. Moreover, this renewable drop-in fuel might positively affect also soot oxidation, due to its fast ignition chemistry and enhanced mixture formation properties [33]. As the 40 vol% RFA30P70 in diesel is compliant with EN590 in accordance with previous discussions, in this result section the relative changes with reference to diesel in performance and emissions are presented for this blend only. A relative reduction of ISPM by up to 12 % is achieved with a blending proportion of 40 vol% RFA30P70 in diesel. The indicated specific carbon monoxide (ISCO) emissions are shown in the middleleft corner of Fig. 5. With an increase in RFA30P70 substitution in diesel, lower ISCO emissions are observed. This could be attributed to the faster ignition, which would prevents over-leaning in air-fuel mixture and to the inherent oxygen moieties included in the renewable fuel. Moreover, an enhanced mixture formation (due to better atomization owing to the paraffinic content) is also beneficial to ensure a proper oxygen entrainment [35]. The indicated specific hydrocarbon emissions (ISHC), shown in the lower-left bar plot in Fig. 5, reduce with increasing RFA30P70 share. Similarly to ISCO emissions, an improved mixture formation might be the reason for the observed trend [35]. Due to reduced ISCO and ISHC emissions, a marginal rise in the indicated thermal efficiency (ITE) by 0.6 % was noticed for the blend with 40 vol% RFA30P70 in diesel, as shown in top-right diagram of Fig. 5. A relative reduction in ISCO and ISHC by up to 6 % and 18 % was respectively noticed for the blend under consideration. Generally, an increase in RFA30P70 share leads to a direct reduction in indicated specific carbon dioxide (ISCO2) emissions. A relative reduction in ISCO2 by up to 2.6 % is seen for the blend with 40 vol% RFA30P70 in diesel, refer middle-right diagram of Fig. 5. To explain this ISCO2 reduction, a parameter named as theoretical fuel carbon flow rate (TFCFR) is introduced. The TFCFR is defined as the fuel carbon mass fraction times the fuel mass flow rate. It represents the theoretical carbon mass flow rate that is available

12

for a complete combustion of the fuel at a given engine load, accounting for variations in injected fuel mass due to changes in calorific fuel content and ITE. The lower-right corner of Fig. 5 shows a TFCFR reduction by 2.3 % for the blend with 40 vol% RFA30P70 in diesel, which nearly agrees with the aforementioned relative ISCO2 reduction. The heat release analysis for the cruise point at the base ISNOx level is shown in Fig. 6. For the sake of clarity, the Fig. 6 presents the heat release analysis for diesel and blends of RFA30P70 with 40 vol% and 80 vol% in diesel. Due to a slightly lower ignition delay time (i.e. slightly higher CN) of the RFA30P70 blends with diesel, the injection timing was slightly retarded to maintain a constant Q50. This can be seen in the cumulative heat release plot at ~2000 J, refer lower-right corner of Fig. 6. The heat release rate (HRR) of the RFA30P70 blends, depicted on the right side in Fig. 6, is similar to that of diesel. The crank angle position corresponding to the maximum heat released during combustion is slightly retarded for the RFA30P70 blends. This can be attributed to an increased injector energizing time – necessary to achieve the load matching (i.e., increased injected mass to compensate IMEP = 10 bar, n = 1200 pinjretarded = 1280 bar, Q50 =timing. 6.2 °CAIncreased afTDC, injected the reduced fuel calorific value) – andrpm, to the injection Tboosta relatively = 41°C, pboost 1789 mbar, pexh =and 1875a subsequent mbar fuel mass causes later =end of combustion shift in heat release.

0.00

50 49 48 47 46 45

0.40

600

ITE / %

0.03 0.02

ISCO2 / (g/kWh)

0.01

0.30 0.20 0.10 0.00 0.20

TFCFR / (kg/h)

ISHC / (g/kWh)

ISCO / (g/kWh)

ISPM / (g/kWh)

ISNOX = 5.8 g/kWh

0.04

0.15 0.10 0.05 0.00

575 550 525 500

3.45 3.40 3.35 3.30 3.25

100% B0 Diesel 80% B0 Diesel / 20% RFA30P70 60% B0 Diesel / 40% RFA30P70 50% B0 Diesel / 50% RFA30P70 20% B0 Diesel / 80% RFA30P70 Fig. 5. Test results for diesel and REDIFUEL blends at cruise point for base NOx level, IMEP = 10 bar, n = 1200 1/min, pinj = 1280 bar, Q5 0= 6.2°CA afTDC, Tboost = 41°C, pboost = 1789 mbar, pexh = 1875 mbar, ISNOx = 5.8 g/kWh.

Cummulative heat release / J

120 90 60 30

30

0 0 -20 -10 0 10 20 30 40 Crank angle / ° CA

Heat release rate / (J/ °CA)

150

Energizing current / A

Cylinder pressure / bar

IMEP = 10 bar, n = 1200 rpm, pinj = 1280 bar, Q50 = 6.2 °CA afTDC, Tboost = 41 °C, pboost = 1789 mbar, pexh = 1875 mbar, ISNOX = 5.8 g/kWh

13

250 200 150 100 50 0 -50 -20 -10 0 10 20 30 40 Crank angle / ° CA

5000 4000 3000 2000 1000

100% B0 Diesel 60% B0 Diesel / 40% RFA30P70 20% B0 Diesel / 80% RFA30P70

0 -20 -10 0 10 20 30 40 Crank angle / ° CA

Fig. 6. Heat release analysis for diesel and REDIFUEL blends at cruise point IMEP = 10 bar, n = 1200 1/min, pinj = 1280 bar, Q50 = 6.2 °CA afTDC, Tboost = 41°C, pboost = 1789 mbar, pexh = 1875 mbar, ISNOx = 5.8 g/kWh.

3.3

3D-CFD Simulation Results

In this section, the numerical results are discussed. Firstly, an overview of the surrogate fuel properties estimations for the numerical simulation is provided. Secondly, the 3DCFD simulations results for the most promising blends of RFA30P70 with diesel are presented. Estimation of Fuel Properties with Surrogates. Fig. 7 shows measured and modeled data for viscosity and density of 1-octanol. The modeled data show very good agreement with the measured data, the same model parameters were used to extrapolate the data in the higher temperatures range. To corroborate the selection of 1-octanol as a surrogate (see Par. 2.5) for the liquid properties of the SAM, Fig. 8 presents a comparison of the measured density of SAM v/s modeled data of 1-octanol. It can be observed that 1-octanol modeled data show a very good agreement with measured SAM data. To prove that also the surrogate selected to model GtL agrees with the experimental data, Fig. 9 shows the comparison for viscosity and density. Here, it can be noticed that both properties matches well with the measured ones. For modeling, the properties of GtL Arrhenius mixing rule was used, as explained in Par. 2.5.

14

Viscosity of 1-Octanol

Density of 1-Octanol

Measured data Modeled data

0.010 0.008 0.006 0.004 0.002 0.000 260

360 460 560 Temperature / K

860

Density / (kg/m3)

Viscosity / (Ns/m2)

0.012

760 660 560 460 360 260 260

660

Measured data Modeled data 360 460 560 Temperature / K

660

Fig. 7. Comparison of measured viscosity and density of the 1-octanol with modeled data. Viscosity of SAM

Density of SAM

Measured data SAM Modeled data 1-Octanol

0.010 0.008 0.006 0.004 0.002 0.000 260

360 460 560 Temperature / K

860

Density / (kg/m3)

Viscosity / (Ns/m2)

0.012

760 660 560 460 360 260 260

660

Measured data SAM Modeled data 1-Octanol 360 460 560 Temperature / K

660

Fig. 8. Comparison of measured viscosity of the surrogate alcohol mixture (SAM) with modeled viscosity of 1-octanol. Viscosity of GTL - Arrhenius mixing rule Measured data Modeled data

0.0035 0.0028 0.0021 0.0014 0.0007 0.0000 280

300

320 340 360 Temperature / K

380

Density of GTL - Arrhenius mixing rule 900

Density / (kg/m3)

Viscosity / (Ns/m2)

0.0042

Measured data Modeled data

830 760 690 620 550 480 280

300

320 340 360 Temperature / K

380

Fig. 9. Comparison of measured GtL viscosity and density with its surrogate modeled fuel mixture of 45 vol% cyclo-octane, 51 vol% iso-cetane and 4 vol% n-decane.

Preliminary Screening of Drop-in Biofuel Using 3D-CFD. To evaluate the mixture formation and emission reduction potential of the fuels under consideration, numerical

15

3D-CFD simulations were performed. In this study, the mixture formation during the high-pressure cycle under inert mixing condition was analyzed. To model the mixture formation of RFA30P70 blends with B0 with a good confidence level, firstly the CFD model was calibrated using diesel experimental data from the HD SCE at cruise point operation. Here, the same ISNOx level as shown in the test bench results section (i.e. 5.8 g/kWh) was chosen. Successively, the physico-chemical properties of diesel were exchanged with those of the blends of RFA30P70 and B0 diesel. Two different blends of RFA30P70 with B0 were studied. A first blend of 60 vol% B0 and 40 vol% RFA30P70 was selected, owing to is CN and density meeting the EN590 norm. Furthermore, a second blend, having a higher share of RFA30P70 (i.e. 80 vol%) with B0, was simulated to derive trends at increasing renewable substitution share in the fuel mixture. Fig. 10 presents the comparison of the air utilization for different blends of RFA30P70 and diesel at cruise point operation. It can be seen that with an increasing blending proportion of RFA30P70 in diesel, the share of lean mixture (i.e. AU between 1 < 𝜆 < 2) is larger. Hence, a higher degree of air/fuel mixing and a high potential to oxidize soot during combustion is expected. The bottom of Fig. 10 shows the cut-sections of the piston bowl modeled in CONVERGE that corroborate the trends of the air utilization curves. At the selected crank angle of 22 °CA a larger share of lean equivalence ratios can be seen and a smaller share of rich zones (𝜆 < 1). This indicates that the soot is expected to be better oxidized in case of the blends with RFA30P70. Further, it can be seen from Table 6 that at the cruise point operation, the RFA30P70 blends resulted in higher OPN as compared to diesel. Impact of the most promising blends were also studied at the rated power point, as shown in Fig. 11. Similarly to cruise point operation, increasing RFA30P70 blending proportion with diesel yields larger lean mixture shares (i.e AU between 1 < 𝜆 < 2), indicating a higher degree of air/fuel mixing and a high potential to oxidize soot during combustion. Table 6. Comparison of Oxidation Potential Number (OPN) for different blends of B0 and RFA30P30 at cruise point and rated power operation. Fuel 100 % B0 60 % B0 / 40 % RFA30P70 20 % B0 / 80 % RFA30P70

OPN Cruise point 22.18 25.16 28.60

OPN Rated power point 9.32 12.84 17.69

16

Air utilization / -

1.0

Cruise point, IMEP = 10.33 bar -1 l < 2 n = 1200 min Pilot =13° CA bTDC Main = 9.8° CA bTDC, Injection pressure = 1500 bar

0.8 0.6 0.4 0.2 0.0 -40

l 110/100 459 kJ/kg 4.9 kg/kg

Due to LHV of DMC+, a larger amount of fuel mass must be supplied to the system to obtain a similar performance. As a result, a large quantity of fuel is available for evaporation. This has a direct effect on the intake air, which cools down considerably. However, the volumetric consumption is relatively lower due to the density which is 37%

3

higher. Taking the improved efficiency into account (as demonstrated later), this means on balance that the fuel tank has to be roughly 1.5 times bigger than in a gasolinepowered vehicle to achieve the same cruising range. As can be shown the enthalpy of vaporization is ca. 9 % higher for DMC+ compared to gasoline E10, which further enhances the effect described above. In addition, the stoichiometric air consumption is lower with the same fuel energy, which means that less intake air has to be cooled. The significantly greater knock resistance is expressed by the RON, which is increased by more than 15. 1.3

Ignition Delay DMC+

Accurate prediction of engine knocking is essential for reliable simulation results. Therefore, the ignition delay in the quasidimensional models must be calculated correctly. In order to facilitate the numerical simulation of ignition delay times and laminar flame speeds of DMC+, a chemical mechanism was developed as part of this study. The well-validated chemical mechanism [6] for a set of C0-C4 hydrocarbon species including ethanol served as the starting point of the model development. The specificmechanisms of DMC and MeFo were extracted from the mechanisms of Dooley et al. [7] and Sun et al. [8], respectively, and integrated into the mechanism of Blanquart et al. [6]. The mechanism was validated successfully against the literature data of ignition delay times and laminar flame speeds of dimethyl carbonate, methyl formate, and ethanol following called Cai Mechanism (see also [9]). Fig. 2 shows the ignition delay times for Gasoline RON 95 E10 and the investigated fuel DMC+ (preliminary results from reaction-kinetics calculations) at a stoichiometric air ratio and without EGR. DMC+, unlike gasoline, does not show a two-stage ignition (TSI)/ negative temperature coefficient (NTC) behavior. This has an extreme effect on the area of temperatures lower than 1.2 1000/K. The ignition delay times increase by over an order of magnitude below this value. Since the relevant part of the engine operation is between approx. 0.8 and 1.2 1000/K at pressures higher than 20 bar, the significantly lower knocking tendency can be seen directly.

4

Fig. 2. Ignition delay times simulated with Cai Mechanism of DMC+ and Gasoline RON95 E10.

As the DMC+ results were only preliminary and a quick estimation of the knock behavior of DMC+ was to be achieved, an adaption of existing knock models was used at first (the implementation of a proper DMC+ knock model is intended for the upcoming months). It was decided to use the older Schmid model [10] rather than the newer and generally much better Fandakov model [11], [12] due to the lack of TSI/NTC behavior, allowing to use an Arrhenius equation for the ignition delay times. Fig. 3 shows the ignition delay times already shown above with the corresponding curve of an Arrhenius equations. While with E10 the representation of the ignition delay times via an Arrhenius approach is insufficiently accurate, this is much better suited for DMC+ due to the missing two-stage ignition behavior. In order not to underestimate the knocking tendency, the coefficients for the Arrhenius equation are assumed conservatively. The 50 and 100 bar curves are always below the simulated values. This makes the modelled fuel more prone to knocking.

5

Fig. 3. Arrhenius equation curves for the fuel DMC+.

1.4

Laminar Flame Speed DMC+

The laminar flame speed is determined using the Heywood equation [13] with parameters adapted for DMC+ in the 0D/1D-Simulation model. Fig. 4 shows the agreement of the laminar flame speed for the Heywood equation with results from a reaction kinetic simulation for DMC+. Parameters for the Heywood equation depend on pressure and temperature.

Fig. 4. Laminar flame speed represented with Heywood equation vs. reaction kinetic simulation.

6

The simulation was performed for DMC with the model according to Sun et al. [8], while MeFo was performed with the model according to Dooley et al. [7]. In the upcoming months, the Heywood correlation will be substituted with a reaction-kinetics based approach similar to [14], [15] and [16], which will ensure even more accurate results. In the range of lean mixtures until rich mixtures to values of Φ ≈ 1.2 deviation of laminar flame speed is less than 15% for boundary conditions relevant for engine operation (50 bar 800 K, 100 bar 1000 K). In this paper only Φ = 1 is considered, with a deviation of less than 4 %. However, it is interesting to note that laminar flame speed for lean mixtures does not drop as steep as it happens for gasoline, making DMC+ also suitable for lean combustion concepts (s. Fig. 5), which will be investigated in ongoing work on the matter.

Fig. 5. Laminar flame speed of DMC+ compared to Gasoline RON95 E10.

2

Engine models

In the first step, the potential of DMC+ is determined with the help of an efficient state of the art gasoline engine, which is adapted for operating with the synthetic fuel. Then a concept is discussed that is optimized for the special properties of DMC+ to further highlight the potential of the fuel. The engine model presented below serves as the basic model. 2.1

Basic Model

VW EA211 TSI evo is a direct-injection gasoline engine characterized by a high efficiency over a wide range of the characteristic map (s. Fig. 6). Decisive factors are the VNT, Miller process, a high compression rate and cylinder deactivation in part

7

load [17]. This engine can be arguably seen as state-of-the-art for current gasoline engines in terms of efficiency. The specifications are shown in Table 2. Table 2. Specification of VW EA211 TSI evo [17]. Max. Power Max. Torque Displacement volume Compression Ratio Stoichiometric Ratio EGR

96 kW@5500 RPM 200 Nm 1500 cc 12.5:1 1 internal

Fig. 6 shows the brake efficiency as a function of torque and engine speed for the engine operated with gasoline RON95 E10. The deviation of the simulation from the measurements in the relevant map area is less than 2 %. The engine achieves its peak efficiency of 38.4% at 3500 RPM and 120 Nm.

Fig. 6. Simulation of the characteristic map of VW EA211 evo RON95 E10.

2.2

Basic Model with DMC+

In order to be able to generate results with the synthetic fuel, modifications described in section fluid properties must be done. Due to the low heat value of DMC+ the parameters of the injector model must be adjusted. Injection delivery rate is increased by factor of three. This is the rate of fuel

8

injection when the injector is held open. In practice, adaptions of the fuel injection system will be unavoidable, for instance by combining direct injection with port fuel injection or by adapting the injector design. Fig. 7 shows the change in brake efficiency between the Gasoline and DMC+ simulation.

Fig. 7. Growth in brake efficiency of VW Evo EA211 with DMC+.

The increased efficiency of DMC+ is clearly visible in the area of high load. At high engine speed and low loads efficiency decrease and the simulation with gasoline becomes more efficient. Three load points marked with X in Fig. 7 will be used to analyze the differences between an engine powered by gasoline or DMC+. Analysis of specific load points. Fig. 8 shows aforementioned load points with their fractionation of fuel energy in the cylinder at a stationary engine operating point. All values are related to the supplied fuel energy for one cycle (100% energy).

9

Fig. 8. Analysis of three engine load points gasoline vs. DMC+.

A comparison of the full load point shows that the energy content of the exhaust gas is significantly lower for DMC+. This is achieved through a better process control due to the higher knock resistance. The heat loss is slightly increased by ca 1 % of the fuel energy because of the earlier combustion center of DMC+. The other energy losses are similar. This means that a large part of fuel energy can be converted into mechanical work for DMC+ which is equivalent to a higher brake efficiency. The combustion centers of the second load point are similar for both fuels. Nevertheless, a slight efficiency advantage can be seen for DMC+. Due to the significantly higher fuel mass that has to be supplied to the system, a cooling effect of the combustion chamber through fuel evaporation can be seen. As a result, the heat loss is reduced which leads to a slightly higher efficiency. At the third load point, DMC+ suffers a combustion loss. This can be attributed to the lower flame speed compared to gasoline. Especially at high engine speed and low loads, this effect is visible. The flame does not reach all combustion chamber walls until the end of cycle. This results in a 2% lower efficiency compared to gasoline for the load point under consideration. Full load curve. Due to the fact that the Schmid Model does not predict knocking for DMC+ in the above presented results, the theoretically possible full load curve shall now be investigated.

10

Fig. 9. Full load curve Gasoline vs. DMC+.

Ignoring a possible peak pressure limit (simulated values can be up to 260 bar), rated power can be increased by a factor of 1.8 and the max. torque can be increased by a factor of 2.6 even. Basically, DMC+ allows the design of a knock-free SI engine with a liquid fuel, a very attractive combination. At the same time, these results obviously point to the conclusion that a DMC+ engine can be downsized heavily if rated power is to be held constant, which is investigated in the following section. 2.3

DMC+ adapted Engine

In order to take advantage of the knock resistance of DMC+, the engine displacement and the compression ratio are adjusted. Since the available exhaust gas enthalpy changes with the process management and fluid properties, the turbocharger must be adapted, too. Characteristic map. Fig. 10 shows the characteristic map of the downsized engine with DMC+. It is noticeable that the only area where the efficiency is worse is in the area of low end torque, which can be explained by a number of design choices that are not compulsory. An optimization of the cam profile is needed. As for the remaining operating range, a considerable improvement in brake efficiency up to 20% can be observed.

11

Fig. 10. Growth in effective efficiency of downsized VW Evo EA211 with DMC+.

The potential of DMC+ is to be clarified again in Fig. 11 by means of three load points. A moderate partial load point in blue, the point with highest brake efficiency of gasoline in green and the full load point in yellow. Even for the best point of gasoline there has been a slight improvement with the switch to DMC+. With the downsizing engine, the efficiency increases by more than 10 %. The highest potential for increasing brake efficiency is offered by the full load point due to the fuels knock resistance.

12

Fig. 11. Comparison of DMC+ engine concepts based on three load points.

3

Summary/Conclusions

The engine of the future must be highly efficient and emit as few pollutants as possible. Furthermore, the fuel should be CO2 neutral. DMC+ is predestined as a synthetic fuel to meet these demands. The most important properties of DMC+ are the high knock resistance and the large cooling potential. The brake efficiency already improves when simply exchanging the used fuel in the simulation model. In order to exploit the full potential of DMC+, the engine design has to be adapted. In this way, as has been shown, downsizing and higher compression ratios can be used to improve brake efficiency even further, leading to more than 20% higher efficiencies compared to a state-of-the-art gasoline engine. In order to generate even more reliable results, the models must be adapted to accurately predict the behavior of DMC+ in comparison with measurements on the test bench. This work will be done in the upcoming months and will presumably provide even more impressive results, as so far all model adjustments were done conservatively. As it is an engine concept of the future, in the next step, a similar procedure will be carried out with a high performance engine, featuring a high turbulence concept and an efficiency-improved turbocharger, among others. Additionally, the potential of lean combustion will be investigated. All in all, peak gross indicated efficiency values higher than 50% should be easily feasible in such a concept.

13

References 1. European Union, “A Clean Planet for all; A European strategic long-term vision for a prosperous, modern, competitive and climate neutral economy,” COM(2018) 773 final, 2018. 2. Härtl, M., Stadler, A., Blochum, S., Pélerin, D, “DMC+ als partikelfreier und potenziell nachhaltiger Kraftstoff für DI Ottomotoren,” Zukünftige Kraftstoffe, 2019, doi:10.1007/978-3-662-58006-6. 3. Grill, M. and Bargende, M., “The cylinder module,” MTZ Worldwide 70(10):60–66, 2009, doi:10.1007/BF03227984. 4. Grill, M., Schmid, A., Chiodi, M., Berner, H.-J. et al., “Calculating the Properties of UserDefined Working Fluids for Real Working-Process Simulations,” SAE Technical Paper Series, SAE Technical Paper Series, SAE World Congress & Exhibition, APR. 16, 2007, SAE International400 Commonwealth Drive, Warrendale, PA, United States, 2007. 5. Grill, M. and Bargende, M., “The Development of an Highly Modular Designed ZeroDimensional Engine Process Calculation Code,” SAE Int. J. Engines 3(1):1–11, 2010, doi:10.4271/2010-01-0149. 6. Blanquart, G., Pepiot-Desjardins, P., and Pitsch, H., “Chemical mechanism for high temperature combustion of engine relevant fuels with emphasis on soot precursors,” Combustion and Flame 156(3):588–607, 2009, doi:10.1016/j.combustflame.2008.12.007. 7. Dooley, S., Burke, M.P., Chaos, M., Stein, Y. et al., “Methyl formate oxidation: Speciation data, laminar burning velocities, ignition delay times, and a validated chemical kinetic model,” Int. J. Chem. Kinet. 42(9):527–549, 2010, doi:10.1002/kin.20512. 8. Sun, H., Yang, S.I., Jomaas, G., and Law, C.K., “High-pressure laminar flame speeds and kinetic modeling of carbon monoxide/hydrogen combustion,” Proceedings of the Combustion Institute 31(1):439–446, 2007, doi:10.1016/j.proci.2006.07.193. 9. Cai, L., Ramalingam, A., Minwegen, H., Alexander Heufer, K. et al., “Impact of exhaust gas recirculation on ignition delay times of gasoline fuel: An experimental and modeling study,” Proceedings of the Combustion Institute 37(1):639–647, 2019, doi:10.1016/j.proci.2018.05.032. 10. Schmid, A., Grill, M., Berner, H., Bargende, M, “Ein neuer Ansatz zur Vorhersage des ottomotorischen Klopfens,” 3. IAV Tagung: Ottomotorisches Klopfen, Tagungsband, Berlin, 2010. 11. Fandakov, A., Grill, M., Bargende, M., and Kulzer, A.C., “Two-Stage Ignition Occurrence in the End Gas and Modeling Its Influence on Engine Knock,” SAE Int. J. Engines 10(4):2109–2128, 2017, doi:10.4271/2017-24-0001. 12. Fandakov, A., Grill, M., Bargende, M., and Kulzer, A.C., “A Two-Stage Knock Model for the Development of Future SI Engine Concepts,” SAE Technical Paper Series, SAE Technical Paper Series, WCX World Congress Experience, APR. 10, 2018, SAE International400 Commonwealth Drive, Warrendale, PA, United States, 2018. 13. Heywood, J.B., “Internal combustion engine fundamentals,” Mechanical engineering, ISBN 978-1260116106, 2018. 14. Hann, S., Grill, M., and Bargende, M., “Reaction Kinetics Calculations and Modeling of the Laminar Flame Speeds of Gasoline Fuels,” SAE Technical Paper Series, SAE Technical Paper Series, WCX World Congress Experience, APR. 10, 2018, SAE International400 Commonwealth Drive, Warrendale, PA, United States, 2018.

14 15. Hann, S., Urban, L., Grill, M., and Bargende, M., “Prediction of burn rate, knocking and cycle-to-cycle variations of binary compressed natural gas substitutes in consideration of reaction kinetics influences,” International Journal of Engine Research 19(1):21–32, 2018, doi:10.1177/1468087417732883. 16. Hann, S., Grill, M., Bargende, M., “A Quasi-Dimensional SI Combustion Model Predicting the Effects of Changing Fuel, Air-Fuel-Ratio, EGR and Water Injection,” in: SAE Technical Paper Series 2020. 17. Eichler, F., Demmelbauer-Ebner, W., Theobald, J., Stiebels, B., et al., “Der neue EA211 TSI®evo von Volkswagen,”(37. International Vienna Motor Symposium), 2016.

Acknowledgments The research project is financed by BMBF (Federal Ministry of Education and Research), due to a decision of the German Bundestag. The authors would like to thank the BMBF for providing financing. Supported by:

On the basis of a decision by the German Bundestag Also the authors would like to thank Martin Härtl and Sebastian Blochum from LVK at TUM for exchange of project progress.

Definitions/Abbreviations BEV FCEV LHV RON MON A/F ratio TSI NTC VNT EGR DMC MeFo FL PL FTDC EVO DOE MFB50

Battery Electric Vehicle Fuel Cell Electric Vehicle Lower Heat Value Research Octane Number Motor Octane Number Air to fuel ratio Two-Stage Ignition Negative Temperature Coefficient Variable Nozzle Turbocharger Exhaust Gas Recirculation Dimethyl carbonate Methyl formate Full Load Part Load Firing Top Dead Center Exhaust Valve Opens Design of Experiment Mass Fraction Burned 50%