Combustion Systems What we might have learned IMechE ADSC, 6 December 2012 Chris Whelan
Contents Engines Big & Small Carnot, Otto & Diesel Thermodynamic Cycles Combustion Process & Systems Diesel & Otto Comparison Invention & Re-discovery Missed Opportunities Conclusions Dec 2012 Chris Whelan 2
Engines Big & Small They all obey the same rules Wartsila 50: 2050 litres 22,300 hp 355,000 kg Saito 56: 0.0092 litre 0.9 hp 0.041 kg Dec 2012 Chris Whelan 3
Carnot, Otto & Diesel Carnot (1824) defined the basic operation of all piston IC engines 1824 Thesis, no experiments (*) Theoretical efficiency: 62%, achieved 2% (*) see next slide Otto (1867) defined the spark ignition (SI) engine: Constant volume compression Experimental work Early success with compression engine Combustion controlled by stratified charge Theoretical efficiency: 82%, achieved 14% Diesel (1893) compression ignition (CI) engine: Constant pressure combustion Theory first, then (many) experiments Coal dust fuel Isothermal compression Insulated cylinder Theoretical efficiency: 73%, achieved; 26% Dec 2012 Chris Whelan 4
Carnot s Engine Carnot never built an engine If one have ben constructed with the technology & materials of the day, the efficiency would have been about 2% Theoretical efficiency 62% Actual efficiency 2% Dec 2012 Chris Whelan 5
Thermodynamic Cycles Ideal cycles & thermodynamic processes: Process Compression Heat Addition Expansion Heat Rejection Carnot isentropic isothermal isentropic isothermal Ericsson (First, 1833) adiabatic isobaric adiabatic isobaric Ericsson (Second, 1853) isothermal isobaric isothermal isobaric Brayton adiabatic isobaric adiabatic isobaric Cycle efficiencies: Novikov is Carnot with irreversible heat transfer into the working fluid. It is representative of the real efficiency of large scale power plant Brayton is usually expressed as a function of the pressure ratio, not temperature 80 Cycle Efficiencies (Tcold = 27 C) 40 Cycle Efficiencies 70 35 60 30 Efficiency [%] 50 40 30 Carnot 20 Stirling 10 Novikov 0 200 250 300 350 400 450 500 550 600 Thot [ C] Efficiency [%] 25 20 15 10 Brayton (1) 5 0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Pressure Ratio [-] Dec 2012 Chris Whelan 6
The Combustion Process Diesel (CI): Rate determined by evaporation of fuel droplets & rate of air-fuel mixing Controlled by: Droplet size Compression temperature Oxygen present Limited by: Droplet size & oxygen present (smoke, particulates, HC NO) Peak temperature (NOx) Otto (SI): Rate determined by flame speed through prepared air-fuel mixture Controlled by: Air motion Temperature Limited by: Detonation ( CR, pressure, temperatures) Misfire (mixture, rich or lean) Dec 2012 Chris Whelan 7
SI Combustion Chamber Peugeot (1914), Cosworth (1967), Ferrari (2006): 4 valve Narrowing valve angle Central spark plug Dec 2012 Chris Whelan 8
CI Combustion Chamber IDI: Vehicles: 1930 ~ 2000 Pre-chamber in cyl. head High swirl (~25:1 @ TDC) Low fuel pressure (~400 bar) High CR (21~23:1) DI: Vehicles: 1988 ~ today Combustion chamber in piston Low swirl (4~6:1 @TDC) High fuel pressure (~1800 bar) Medium CR (15~19:1) Dec 2012 Chris Whelan 9
Chris Diesel Law Efficiency depends on the speed of mixing fuel-air mixing The air can mix with the fuel (high swirl), or the fuel with the air (high pressure) The mixing time available depends on the engine speed Dec 2012 Chris Whelan 10
Fun with Wiebe In the 1950s Wiebe developed a simple equation to describe combustion heat release It can be adjusted to represent most combustion types Wiebe used this to show that DIFFERENT heat releases, if timed correctly give the SAME efficiency Dec 2012 Chris Whelan 11
Wiebe was right... Two versions of the same 4-stroke stationary power generator engine: Diesel CI: low swirl open chamber, short combustion Natural gas SI: pre-chamber lean-burn, long combustion Similar BMEP, CR & AFR With optimum timing, the efficiency was almost IDENTICAL despite very different combustion rates BUT the NOx is much lower for the gas engine Dec 2012 Chris Whelan 12
CI & SI Differences Key features of current CI & SI engines which influence efficiency Parameter Timeframe CI Current Eff. Impact Current SI Eff. Impact Pmax High Low BMEP High Low CR High Low AFR High Low Fuel Pressure High Low Friction High Low Efficiency (total) High Low Dec 2012 Chris Whelan 13
...& Similarities Key parameters of SI & CI engines are converging Efficiency trend is up for SI & neutral for CI Parameter Timeframe CI Current Trend Eff. Impact Current SI Trend Eff. Impact Pmax High >> Low High BMEP High >> Low High CR High Low Low High AFR High Low Low Low Fuel Pressure High >> Low High Friction High >> Low >> Efficiency (total) High >> Low High Dec 2012 Chris Whelan 14
SI Invention & Re-discovery Dates show first engine application, then vehicle mass production 1867 ~ today: Otto engine 1867, 1984: Lean-burn combustion 1908, 1982: 4 valves per cylinder 1915, 20??: WOT EGR 1919, 1984: Controlled air motion 1927, 1984: Turbocharging 1940, 1986: Direct injection 1942, 20??: Turbo-compound 1945, 1987: Miller cycle 1983 ~ today: VVT Dec 2012 Chris Whelan 15
CI Invention & Re-discovery Dates show first engine application, then vehicle mass production 1893 ~ today: Diesel engine 1893, 1983: Solid fuel 1893, 1988: Direct injection 1919 ~ today: Pre-chamber combustion 1919, 2000: Common rail fuel system 1938 ~ today: Turbocharging 1988, 2012: VVT Dec 2012 Chris Whelan 16
Missed Opportunities: EGR at WOT 1915: Ricardo: Increase in BMEP Improved detonation limit (*) Some experiments on Supercharging in a High-Speed Engine, Inst. of Automotive Engineers 1921 (1.3 litre/cyl. 1500 rpm) 2009: Mahle Powertrain: Increase in detonation limit Improved combustion Reduced fuel enrichment (*) Mahle.com 2009 (0.4 litre/cyl. 6000 rpm) Std. dev. of IMEP (σimep in %) & % MFB as functions of the combustion mixture dilution with EGR or fuel, 3500 RPM (0 % fuel dilution equal to λ = 1, 10 % fuel dilution λ = 0,9) Dec 2012 Chris Whelan 17
Others... Isothermal compression: Compression at constant temperature Diesel s concept in 1893, using water injection Demonstrated in 1998 (npower Iso-engine) Adiabatic expansion: Expansion with no heat loss Diesel s concept in 1893, using insulated cylinder Demonstrated in 1980s ( insulated diesels) BUT cold compression & hot expansion requires a Split Cycle engine: Scuderi concept in 2004 (but not intent) Industry R&D in 2014?? Dec 2012 Chris Whelan 18
Conclusions The theoretical basis of what we do now is more than 100 years old The rules don t change Much current research has previously been either proposed or tested There s a lot still to be re-discovered & made to work, using modern design, analysis, materials, controls etc. Diesel & Otto engine combustion systems will merge It s always worth doing the literature search Dec 2012 Chris Whelan 19
Recommended Reading Anything by Harry Ricardo Internal Fire : Lyle Cummins The original Wiebe publication The Heat Engine: Ivo Koln Dec 2012 Chris Whelan 20