Optical Techniques in Gasoline Engine Performance and Emissions Development

Transcription

1 Optical Techniques in Gasoline Engine Performance and Emissions Development TC GDI engines: analysis and development techniques to solve pre-ignition and soot formation issues Ernst Winklhofer AVL List GmbH, Graz Austria June 2014 Gaydon, UK contact: 1

2 TC GDI engines some topics As liquid fuel enters the combustion chamber of an SI engine otherwise: 1. it must evaporate and mix with air to support formation of a premixed flame 2. it must keep away from cold surfaces to avoid fuel film formation oil dilution deposit formation see movies on soot formation m1, m2: premixed + diffusion m3, m4: premixed and examples for pre-ignition 2

3 TC GDI engines PN particle number emissions 3 what is our target? Euro 6c in 2017: PN < 66*10 11 in the 20 minute NEDC test 66*10 11 PN 6*10 11 /km allows PN 66*10 11 /test 3

4 TC GDI engines PN particle number emissions 4 The PN Euro 6c limit eliminate the root causes: find measures to avoid soot formation Achieving the Euro 6c target: Spray and flame movies show that avoiding diffusion flames is key to soot free combustion. How can we use this in practical engine development?

5 Spray and flame movies show that avoiding diffusion flames is key to soot free combustion. How can we use this in practical engine development? Step 1 select correct hardware for airflow fuel injection piston surface correct hardware what is it? 5

6 Injector piston combinations Step 1 select correct hardware for airflow.port development for volumetric efficiency and tumble fuel injection: injector piston surface: piston bowl Injector piston combinations workflow on the optical engine: we take an injector piston combination and use spray / flame movies to guide our selection of injection parameters for soot free combustion. Result: tradeoff trends on stability and emissions for given hardware combination 6

7 Injector piston combinations injection variations: pressure, timing, multiple, etc. Injector piston combinations catalyst heating soot stability tradeoff Step 1 select correct hardware for airflow.port development for volumetric efficiency and tumble fuel injection: injector piston surface: piston bowl our criterion for correct : - lowest engine out soot at - lowest IMEP fluctuations here: data show tradeoff in cat heating point 7

8 Selecting correct hardware under guidance of the optical engine tests Injector piston combinations TC GDI: the correct package of development techniques enables soot free combustion. What is in this package? injection variations: pressure, timing, multiple, etc. We use optical engine tests for visual evaluation of flame injection relations catalyst heating soot stability tradeoff Step 1 select correct hardware for airflow fuel injection piston surface in reference operating points at the optical engine 8

9 Step 1 select correct hardware at optical engine TC GDI: the correct package of development techniques enables soot free combustion. What is in this package? Step 2 multicylinder engine development to meet all combustion requirements including PN and PI pre-ignition quality Results step 1: injector + piston + a set of parameters to start combustion developmnet in multicylinder engine correct package of development techniques [1] Fuchs H. et al.: Methods and criteria for fuel injector integration in boosted gasoline direct injection engines. IMECHE, May 2009, London Step 3 Vehicle calibration for the NEDC test [2] Fuchs H. et al: MIXTURE QUALITY EVALUATION FOR TRANSIENT MODE GASOLINE ENGINE CALIBRATION, ICEF , ASME 2010 [3] Fraidl, G., Winklhofer E.: Entwicklungsmethodik für Euro 6 GDI Motoren, 10. Internationales Symposium für Verbrennungsdiagnostik, Baden Baden Mai 2012 [4] Dobes Th. et al.: Optical Combustion Analysis with Spark Plug Sensors for Particulate-optimized EU6 Calibration, Haus der Technik, Berlin Dez. 12,

10 Step 1 select correct hardware at optical engine TC GDI: the correct package of development techniques enables soot free combustion. What is in this package? Step 2 multicylinder engine development to meet all combustion requirements including PN and PI pre-ignition quality diagnostics in step 2 and 3: fiber optic spark plugs for flame radiation evaluation show PN sources in stationary and transient tests Step 3 Vehicle calibration for the NEDC test 10

11 Spark plug sensor type single channel Application / purpose Cylinder comparison, synchronous measurement result Best / worst cylinder 7+1 Find soot sources in transient test Injection calibration for transient test l 80# Refined local analysis, aging, long term stability Quality and risk assessment 11

12 accumulated flame radiation - rel. unit.access and exploit local information in normal engine operation cycle resolved data for Here: a 20 minute NEDC test script 1000 on intake side at high load 800 on piston at start Time - seconds 1. do not spray onto the cold piston 2. do not spray onto the intake valves 12

13 accumulated flame radiation - rel. unit on intake side at high load on piston at start Time - seconds With such guidance we apply modifications to injector and injection parameters Less high load sooting Smaller transient peaks tailpipe PN measurement Smaller start hills Before / After modifications limit target range Confidential 13

14 PN Summary Select correct hardware use flame analysis to guide combustion development use in-vehicle flame sensors to calibrate emissions tests The result: achieve development targets well below EU6c limit Euro 6 PN limits 14

15 Pre Ignition 15

16 PCYL [bar] Irregular combustion: development and calibration of highly boosted SI engines Current Limits of Turbocharged Gasoline Engines Limitations: Knock Pre-Ignition Kurbelwinkel [Grad] Internal 16

17 BMEP Avoid PI events: PI classes LSPI low speed pre-ignition: high temperature acts over time to initiate combustion. It sets a natural limit to high load low speed operation parameter Load Speed deposits Engine Speed [rpm] Internal 17

18 BMEP Avoid PI events alltogether: PI classes Thermal pre-ignition: heat transfer from surfaces to mixture initiates ignition. Engine load is most dominant. parameter Load Speed deposits Engine Speed [rpm] Internal 18

19 BMEP Avoid PI events alltogether: PI classes Sporadic pre-ignition: residual deposits initiate ignition. Deposits formation history is most dominant. parameter Load Speed deposits Engine Speed [rpm] Internal 19

20 On engine test bed first task is to avoid PI events at all The combustion analysis task: understand dominant parameter to define suitable improvement Internal 20

21 PI analysis on engine test bed PI event analysis: Flame data identify ignition centers Event statistics together with test variants analysis provide guidance for combustion system improvement the spark plug flame sensor to find PI ignition spots Internal 21

22 group 1 group 2 group 4 group 3 group 3 group 2 group 1 PI events are detected with a fiberoptic multichannel spark plug sensor Spark Cyl pressure -20 deg CA 20 group 4 Ex Ex deg CA 20 3 In In PI location: 1. in topmost sensor group: near cylinder head 2. in direction of center intake 3. ~ 8 mm from spark plug border 8 mm 8 mm Flame moves from ignition point to spark plug in 4 deg CA ( 8 mm) Flame moves across spark plug in 4 deg CA ( = 8 mm) PI location is derived from flame data and sensor channel configuration Internal 22

23 PI analysis on engine test bed from single event data Test of variants shows: Ex In Ex In to PI distribution statistics Combustion chamber temperature is foremost PI parameter The first choice in a GDI engine: use late injection for charge cooling Internal 23

24 PI analysis on engine test bed from single event data Ex Ex to PI distribution statistics The first choice in a GDI engine: use late injection for charge cooling Ex Ex In In retard injection In In Internal 24

25 PI analysis on engine test bed Ex from single event data Spray on piston impingement puts a limit to what late injection can achieve Ex to PI distribution statistics BUT: The first choice in a GDI engine: use late injection for charge cooling Ex Ex In In retard injection In In Internal 25

26 PI tuning: find the balance between temperature and deposits effects the degrees of freedom: A: cooling B: wallfilm C: injector Internal 26

27 PI tuning: find the balance between temperature and deposits effects the degrees of freedom: A: cooling B: wallfilm C: injector Internal 27

28 FSN, D-Flame on piston PI Events / hr High load soot is pointer towards PI issues the wallfilm issue: engine out soot and diffusion flame signatures are pointers for PI risk Engine operation variant 0 Smoke Number Diffusion flame on piston PI Events The actions: find the balance between charge cooling benefits and fuel deposit formation we use guidance available from flame measurement Internal 28

29 Mixture formation quality is best evident in fiber optic flame measurement data B A A early injection late injection Diffusion flame signatures show spray impact on valves (A) and on piston (B). Target: late injection for effective charge cooling just short of spray piston deposit formation Internal 29

30 test bed result injection parameters (fuel pressure, SOI1, DOI1, SOI2, DOI2) in balance between late injection charge cooling at still acceptable mixture quality is there any benefit in the vehicle tests? Internal 30

31 see vehicle test improvement in reference Martin Ogris, Ernst Winklhofer: Irregular combustion: development and calibration of highly boosted SI engines, 4th International Conference on Knocking in Gasoline Engines, Berlin, Internal 31

32 thank you 32