Xin Q. Diesel Engine System Design

Подождите немного. Документ загружается.

468 Diesel engine system design

can be used to assist DPF regeneration. However, there are challenges on

the post injection timing. Too early post injection may cause difculty in

engine torque control, while too late post injection may cause oil dilution

or bore-washing problems.

Engine system design is closely related to the following fuel system

areas:

∑ the effect of the conventional diesel fuel properties on engine performance,

emissions, and durability

∑ alternative-fuel and dual-fuel diesel engine performance and durability

∑ life cycle cost and benet analysis for the alternative-fuel and dual-fuel

engines

∑ the effect of fuel injection on engine performance, combustion, emissions,

and noise; and the strategy of the required fuel injection rate prole

∑ fuel injection spray characteristics

∑ fuel injection system hydraulic dynamics for the predictive modeling of

fuel injection rate shape and engine system optimization

∑ parameterization of fuel injection rate proles for the engine steady-state

and transient cycle simulations

∑ fuel system parasitic power losses, heat rejection, and the impact on

engine BSFC and vehicle fuel economy

∑ injector tip temperature and injector coking

∑ real-time modeling of fuel system hydraulic dynamics for hardware-

in-the-loop controls development (Woermann et al., 1999; discussed in

Chapter 14)

∑ fuel delivery unevenness detection and misre detection, and their model-

based controls based on system performance and dynamics parameters

(Macián et al., 2006; discussed in Chapter 14)

∑ model-based fuel path and governor controls (e.g., engine speed control

for better stability and drivability; discussed in Chapter 14).

Engine cycle simulation is closely related to fuel system design and

matching. The fuel injection rate prole used in the engine cycle simulation

can be estimated empirically based on the fuel system bench test data or

the hydraulic/dynamic simulation data. Fuel spray can be simulated to

predict the air entrainment, the evaporation and the combustion by using

the phenomenological or KIVA models. The effects of fuel system design

and combustion bowl matching on fuel spray pattern, heat release and air

utilization in the cylinder can be analyzed. The injector tip temperature is

affected by the in-cylinder gas temperature, the heat ux, and the fuel ow

rate. Injector coking is directly affected by the metal temperature and the

additives in the fuel. The engine cycle simulation in system design may

provide the simulated thermal boundary conditions for the cylinder head and

the injector in the entire engine speed–load domain under different operating

Diesel-Xin-07.indd 468 5/5/11 11:51:26 AM



469Combustion, emissions, and calibration for system design

conditions. This can help develop engine control algorithms to alleviate the

injector coking problem.

The ability of diesel fuels to lubricate the fuel injection components

is referred to as its lubricity. Fuel lubricity (SAE J2265, 1995; Matzke et

al., 2009) and wear durability of the fuel system is mainly a component

design issue. However, the impact of lubricity additives in the diesel fuel

on engine emissions and aftertreatment performance should be considered

at the system design level. Modern fuel systems offer very high injection

pressures and hence the loaded tribological contact conditions in the fuel

injection equipment will become more severe.

The following literature on diesel fuels and fuel systems can help an engine

system design engineer to acquire the necessary knowledge of fuel system

selection and matching. The chemistry of diesel fuels is introduced by Oven

and Trevor (1995) and Song et al. (2000). Diesel fuel properties are reviewed

by Batts and Zuhdan-Fathoni (1991), Majewski and Khair (2006), Ribeiro et

al. (2007), and Matzke et al. (2009), and also explained in SAE J313 (2004)

and J1498 (2005).

The effects of diesel fuels on emissions are reported by Den Ouden et al.

(1994), Singal and Pundir (1996), Nylund et al. (1997), Boesel et al. (2003),

Matthews et al. (2005), Kono et al. (2005), Hara et al. (2006), Zannis et al.

(2008), Fanick (2008), Nanjundaswamy et al. (2009), and Hochhauser (2009).

The sulfur impact on diesel emissions control is reviewed by Corro (2002).

Diesel fuel system design and performance are summarized by Cuenca (1993),

Gill and Herzog (1996), Bauer (1999), Stan (1999) and Zhao (2010).

The fundamentals of fuel injection system dynamics are introduced by

Marcic (1993, 1995). Advanced simulation models of fuel injection system

dynamics are developed by Kouremenos et al. (1999), Desantes et al. (1999),

Yamanishi (2003), Gullaksen (2004), Mulemane et al. (2004, with the software

AMESim), and Kolade et al. (2004, with GT-FUEL). The models were able

to predict the fuel injection pressure, the needle lift, the injection ow rate

shape, the hydraulic pressure uctuation in the system, and the fuel spray

condition at the nozzle exit. A Design-of-Experiments (DoE) simulation

analysis applied to fuel injection system dynamics was presented by Amoia

et al. (1997). Fuel injection system instability and cavitation were studied by

Ficarella et al. (1999) with dynamic modeling. The effects of the geometry of

the rail-to-injector connection pipe on the injection pressure oscillation and

the injection rate in a common rail system were experimentally investigated

by Beierer et al. (2007).

Simplied uid dynamics models may produce substantial errors in fuel

injection pressure and rate predictions if fuel cavitation in the high pressure

system and the variations in bulk modulus with temperature and pressure are

not considered (Lee et al., 2002). Fuel spray behavior and injector nozzle

ow cavitation are reviewed by Schmidt and Corradini (2001).

Diesel-Xin-07.indd 469 5/5/11 11:51:26 AM



470 Diesel engine system design

7.2.4 Combustion chamber design

In direct injection diesel combustion system design, the combustion chamber

shape and the engine compression ratio are important for in-cylinder air motion

and emissions. The compression ratio has a direct impact on system design

because it affects unaided cold start, peak cylinder pressure at the full load,

the allowable air–fuel ratio and EGR rate, friction losses and mechanical

efciency, emissions, engine cycle indicated thermodynamic efciency, and

BSFC. For low emissions and low fuel consumption the heat release rate

during the premixed combustion phase needs to be reduced in order to lower

the combustion temperature and reduce the NO

emissions (for example, by

using pilot injection). The heat release rate during the diffusion combustion

phase needs to be raised in order to keep fast combustion rate and a proper

in-cylinder temperature to reduce the HC, the CO, and the PM. The late

stage of the diffusion combustion phase needs to be shortened in order to

reduce the fuel consumption and the exhaust temperature.

Key issues in combustion design include the following:

1. The engine compression ratio needs to be selected to ensure acceptable

peak cylinder pressure and reliable unaided cold start. During the cold

start, self-ignition is achieved by a combination of sufcient time and the

required temperature (Gardner and Henein, 1988). They are affected by a

number of factors, including compression ratio, ambient temperature, fuel

injection amount and timing, starting speed, the speed-related gas leakage

past the piston rings and heat transfer losses, and other starting aids (e.g.,

excess fuel injection, glow plug, heaters, and ether). A low compression

ratio causes a reduction in engine thermodynamic cycle efciency and

possibly undesirable increase in the ignition delay. Note that the indicated

fuel conversion efciency of the constant-volume cycle is

= 1 – W

1–k

7.1

where W is the compression ratio and k is the ratio of the specic heat

capacities, k = c

> 1. An excessively high compression ratio results

in increased mechanical friction and higher NO

emissions. When intake

valve closing timing is changed, the ‘effective’ engine compression

ratio may vary although the effect is not exactly the same as changing

the geometric compression ratio. In general, a high compression ratio

is better for low HC emission at the low-speed, low-load operation. At

high speeds or high loads, a low compression ratio may reduce smoke

and fuel consumption via calibration changes.

2. The combustion chamber design needs to match the high-pressure fuel

injection system and the intake swirl for good mixing. The combustion

chamber shape variation along the radial or the axial direction controls

the swirl, the vortex, and the turbulence of the in-cylinder airow and

Diesel-Xin-07.indd 470 5/5/11 11:51:27 AM



471Combustion, emissions, and calibration for system design

the wall impingement of the fuel spray to achieve better air utilization

and reduce the air–fuel ratio required.

3. Optimum swirl intensity has a tendency to decrease so that the design

of a more quiescent combustion bowl with a higher injection pressure

is more favored for low to medium speed engines. Wall impingement

of the fuel spray should be generally avoided in the design. Low swirl

leads to low turbulence in the combustion chamber, resulting in low

heat rejection through the wall of the combustion chamber.

4. A low surface-to-volume ratio is preferred in order to minimize the heat

rejection losses (Siewert, 1978; Filipi and Assanis, 2000).

5. The metal temperature at the edge of the combustion bowl is related to

the bowl shape. It needs to be optimized for the best trade-off between

the airow turbulence enhancement and the thermal fatigue life.

6. The combustion noise may be reduced by turbocharging and fuel injection

strategies through a reduced rate of increase of the cylinder pressure. The

combustion noise can also be reduced by the attenuation of structural

damping. This topic is discussed in more detail in Chapter 11.

More information about diesel engine combustion chamber design can be

found in Hikosaka (1997), Lu et al. (2000), Montajir et al. (2000), Mori et

al. (2000), Bianchi et al. (2000), Regueiro (2001), Wickman et al. (2001),

and Cursente et al. (2008). Diesel cold start performance was investigated

by Gonzalez et al. (1991), Mitchell (1993), Liu et al. (2003), Zhong et al.

(2007), Peng et al. (2008), MacMillan et al. (2008), and Pacaud et al. (2008).

Variable compression ratio (VCR) engines have been researched by Roberts

(2003), Rabhi et al. (2004), Hountalas et al. (2006), Tomita et al. (2007),

Tsuchida et al. (2007), and Gérard et al. (2008).

An engine cycle simulation is presented in Fig. 7.2 to illustrate the

effect of start-of-combustion (SOC) timing and EGR rate on engine system

performance. Figure 7.3 shows the simulation result of the effect of combustion

duration on engine performance at peak torque and rated power. Figure 7.4

illustrates the effects of the rise rate and the duration of the heat release

rate on exhaust manifold gas temperature and coolant heat rejection. It is

found that the initial reduction of the coolant heat rejection is caused by a

reduction in the in-cylinder heat transfer as the heat release rate duration

is stretched longer. As the combustion duration increases further, the nal

sharp increase in the coolant heat rejection is due to the much higher fueling

rate required to maintain the same rated power caused by the large increase

in the BSFC.

7.2.5 Fundamental combustion/emissions test and air

system requirement

Either single-cylinder or multi-cylinder engine emissions experiments can

be conducted in order to determine the combustion recipe and the air system

Diesel-Xin-07.indd 471 5/5/11 11:51:27 AM



1900 rpm, A/F = 26 for 330 hp and A/F = 24 for 370 hp

Solid curves: 330 hp

Dotted curves: 370 hp

Solid curves: 330 hp

Dotted curves: 370 hp

Solid curves: 330 hp

Dotted curves: 370 hp

Solid curves: 330 hp

Dotted curves: 370 hp

0 5 10 15 20

EGR rate (%)

0 5 10 15 20

EGR rate (%)

0 5 10 15 20

EGR rate (%)

0 5 10 15 20

EGR rate (%)

SOC = 10

deg BTDC

SOC = 7

deg BTDC

SOC = 4

deg BTDC

SOC = 1

deg BTDC

SOC = 10

deg BTDC

SOC = 7

deg BTDC

SOC = 4

deg BTDC

SOC = 1

deg BTDC

SOC = 10

deg BTDC

SOC = 7

deg BTDC

SOC = 4

deg BTDC

SOC = 1

deg BTDC

SOC = 10

deg BTDC

SOC = 7

deg BTDC

SOC = 4

deg BTDC

SOC = 1

deg BTDC

Peak cylinder gas temperature

(°F)

Exhaust manifold gas temperature

(°F)

BSFC (lb/(hp.hr))

Peak cylinder pressure (psia)

3000

2900

2800

2700

2600

2500

2400

1200

1150

1100

1050

1000

950

900

0.38

0.37

0.36

0.35

0.34

0.33

0.32

3500

3400

3300

3200

3100

3000

2900

2800

2700

2600

2500

2400

2300

2200

2100

2000

7.2 DoE simulation of the effect of start-of-combustion (SOC) timing and EGR rate.

Diesel-Xin-07.indd 472 5/5/11 11:51:27 AM



473Combustion, emissions, and calibration for system design

–20 –10 0 10 20 30 40 50 60 70 80 90 100 110

Crank angle (degree)

0.045

0.040

0.035

0.030

0.025

0.020

0.015

0.010

0.005

0.000

–0.005

Normalized apparent heat release

rate (1/degree)

At rated power 2000 rpm 470 hp

Heat release rate duration multiplier = 0.8

Heat release rate duration multiplier = 1

Heat release rate duration multiplier = 0.6

A/F ratio

Percentage of fuel energy lost to base

engine coolant heat rejection

Peak torque

Rated power

Peak torque

Rated power

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4

Heat release rate duration multiplier

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4

Heat release rate duration multiplier

7.3 Simulation of the effect of heat release rate duration on engine

performance (non-EGR high-power-density diesel engine).

Diesel-Xin-07.indd 473 5/5/11 11:51:27 AM



474 Diesel engine system design

Peak torque

Rated power

Peak torque

Rated power

Peak torque

Rated power

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4

Heat release rate duration multiplier

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4

Heat release rate duration multiplier

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4

BSFC (lb/(hp.hr))

Exhaust manifold gas temperature (°F)

Engine delta P (back minus boost, mbar)

BSFC (lb/hp.hr)

0.43

0.42

0.41

0.40

0.39

0.38

0.37

0.36

0.35

0.34

1600

1550

1500

1450

1400

1350

1300

1250

1200

800

600

400

200

–200

–400

–600

–800

Heat release rate duration multiplier

7.3 Continued

Diesel-Xin-07.indd 474 5/5/11 11:51:28 AM



475Combustion, emissions, and calibration for system design

requirement at the minimum engine delta P (pumping loss) and BSFC. At a

given speed and torque, the air–fuel ratio and EGR rate required to meet the

target emission requirements are contingent upon the BSFC produced in the

test. Therefore, the engine delta P and the BSFC used in the emissions recipe

test need to be as close to their respective optimums as possible. In the test,

the EGR circuit hardware needs to be selected with the lowest possible ow

restriction for the EGR cooler and the EGR valve. Hardware screening with

the DoE method can be conducted by both the KIVA simulation and engine

testing in order to select the optimum compression ratio, combustion bowl

design, intake swirl ratio, and fuel injector nozzle design. Then, the DoE

tuning test may be conducted at the critical speed–load modes used for the

heavy-duty emissions certication. In the single-cylinder engine test, the DoE

factors are intake manifold gas temperature (varied by EGR or charge air

cooling), intake manifold pressure, exhaust manifold pressure, EGR valve

opening, fuel injection timing and pressure. In the multi-cylinder engine

test, the DoE factors are the same except for using the VGT vane opening

or the turbine wastegate opening to replace the intake manifold pressure,

and using the exhaust restriction setting to replace the exhaust manifold

pressure. Intake throttle valve opening is an optional DoE factor that may

be used near the peak torque region to drive the EGR ow or used in the

high-speed, low-load region to largely reduce the air–fuel ratio in order to

reduce the NO

. The intake throttle may also be used to achieve the low

Engine coolant heat rejection (Including

EGR cooler heat rejection, Btu/min)

8700

8600

8500

8400

8300

8200

8100

8000

7900

7800

7700

7600

7500

Rated power, A/F = 23, EGR = 20%, ﬁxed fuel injection timing

Longer combustion duration

Slower heat

release rate

Steeper

(faster) heat

release rate

800 900 1000 1100 1200 1300

Exhaust manifold gas temperature (°F)

7.4 Effect of combustion heat release rate on exhaust temperature

and coolant heat rejection.

Diesel-Xin-07.indd 475 5/5/11 11:51:28 AM



476 Diesel engine system design

minimum ow rates for smooth engine shutdown with a minimum NVH, or

used in the fuel-rich regeneration of a NO

adsorber. More DoE factors may

be included for more exible air systems (e.g., variable valve actuation).

Those DoE factors are very similar to those used in the production engine

calibration stage.

Before running the multi-cylinder engine test, in order to check whether

the test can cover a desirable range of EGR rate and air–fuel ratio, the

capability of the available turbocharger hardware needs to be checked by

an engine cycle simulation through sweeping the VGT vane or the turbine

wastegate opening, the exhaust restriction, or the intake throttle opening.

The parametric sweeping simulation data in Fig. 7.5 illustrate how to build

a sufciently wide and controllable range of air–fuel ratio and EGR rate in

such a fundamental emissions test. Note that each data point in Fig. 7.5 can

220

200

180

160

140

120

0.95

0.9

0.85

0.8

0.75

0.7

0.386

0.384

0.382

0.38

0.378

A/F ratio

A/F ratioA/F ratio

A/F ratio

1 inch Hg = 0.033864 bar

1 lb/hp.hr = 608.3g/kW.hr

VNT vane opening (–)

BSFC (lb/(hp.hr)) Engine delta P (mbar)

Turbine wastegate fully closed, EGR valve fully open, intake throttle valve fully open

Exhaust restriction pressure drop (inch Hg)

28 30 32 34 36 38

EGR rate (%)

28 30 32 34 36 38

EGR rate (%)

28 30 32 34 36 38

EGR rate (%)

28 30 32 34 36 38

EGR rate (%)

7.5 Effects of turbine area and exhaust restriction at rated power.

Diesel-Xin-07.indd 476 5/5/11 11:51:28 AM



477Combustion, emissions, and calibration for system design

be solved by using the equations in Chapter 4. If the air–fuel ratio is too low,

the exhaust restriction pressure drop can be articially lowered. If the EGR

rate is too high with the VGT (VNT) vane fully open, a turbine wastegate

passage can be built in the test or cell to reduce the engine delta P in order

to reduce the EGR rate.

After the test data are obtained, DoE emulators need to be built and an

optimization needs to be conducted to compute the engine performance

contour maps as shown in Fig. 7.6. The parameters on the contour maps

usually include EGR rate, air–fuel ratio, intake manifold gas temperature,

fuel injection timing and pressure, engine delta P, BSFC, and combustion

noise. The required emissions recipe for a given emissions target of NO

, PM,

HC, and CO can be found in the ‘PM or soot vs. NO

’, ‘PM vs. NO

+HC’

or the ‘air–fuel ratio vs. EGR rate’ domains. In order to address the possible

differences in turbocharger efciency and EGR circuit restriction in a future

design stage, two sets of contour plotting are useful: (1) an ideal situation with

the EGR valve set at fully open, which produces the minimum engine delta

P (pumping loss) and BSFC; (2) a more practical or deteriorated situation

with the EGR valve set at partially open, which produces a higher pumping

loss. The degree of EGR valve throttling depends on speed, torque, turbine

area, and turbocharger efciency.

The above emissions test may also be conducted to mimic the high-altitude

and hot-ambient emissions performance for the NTE emissions requirements.

The DoE data of the extreme ambient conditions can be processed using

the same approach to obtain the air system requirements for NTE. During

the process of determining the combustion/emissions recipe, the correlation

between the steady-state and the transient emissions with proper engine control

strategies should be factored in. These air system functional requirements of

air–fuel ratio, EGR rate, and intake manifold gas temperature are then sent

to the system design engineer. Note that this combustion/emissions recipe is

associated with a certain heat release rate and start-of-combustion timing and

BSFC, as well as certain assumptions in the combustion–fuel system matching

used by the combustion engineer. Engine system design specications are

then computed and optimized by the system engineer for hardware sizing

of the turbocharger, the EGR circuit ow restriction, and the cooler size.

The lower the air–fuel ratio and the EGR rate required for combustion, the

easier the air system design. It should be noted that multiple turbochargers

or cooling congurations may reach the same air system functional target.

However, it is the system analysis engineer’s responsibility to optimize the

hardware selection. An example of combustion development can be found on

Navistar’s model year 2004 6.0 L V8 engine (Zhu et al., 2004). A procedure

of engine air system design based on combustion/emissions requirements is

illustrated in Fig. 7.7.

It should be noted that engine system design often needs to translate the

Diesel-Xin-07.indd 477 5/5/11 11:51:28 AM

