**2. Experimental characterisation of the GDI spray dynamics from multi-hole injectors**

Either the desired charge stratification around the spark plug in lean-mixture operation, or the greatest homogeneity under stoichiometric conditions, is achievable in GDI engines through different modes of gasoline-air mixture formation. In the so-called *wall-guided* mode the gasoline spray is directed towards the piston, which exhibits a properly shaped "nose" deflecting the mixture cloud in the vicinity of the spark plug. In the *air-guided* mode the mixture richer region is brought towards the ignition location by the tumble motion of the air entering from the intake ducts. Finally, in the *jet-guided* or *spray-guided* mode typically the spacing between the injector and the spark is smaller, with the fuel spray injected close to the ignition location [Stan, 2000].

Several kinds of injectors for GDI applications are today available. The earliest solution to reduce the rapidly changing fuel concentration gradients as the fuel passes the spark location during the injection period, hence to increase the combustion robustness, relies on the adoption of air-assisted injection systems, such as the one developed by Orbital Engine [Cathcart and Railton, 2001]. This technology is today still applied, because it offers an additional degree of freedom constituted by the direct injection of air, that allows a more effective control of local oxygen concentration, temperature and charge motion through the cycle [Shim *et al.*, 2008]. Alternative solutions, better meeting the requirements for the development of more efficient GDI engines, are the high pressure injectors: the swirl type injector generates an hollow-cone fuel spray by providing a swirl rotational motion to the fuel, that widely disperses and well-atomizes the spray at moderate injection pressures [Brewster *et al.*, 2008]; the multi-hole configuration, on the other hand, exhibits flexible spray patterns that reduce the fuel impingement on the cylinder walls and improve the spray stability (cone shape) with respect to the existing backpressure.

Three commercial multi-hole injectors suitable to be mounted on high-performances SI engines are tested within the present work. As mentioned in the Introduction, the major aim is the assessment of a complete database for the development of a 3D numerical model for the spray dynamics. Table 1 reports the holes number and diameter, as well as the exact flow rates of the considered injectors. The axes of the single jets coming from the nozzles are configured to depict different spray footprint structures. Two injectors are manufactured by Bosch, type HDEV 5.1, differing for the holes number, six for Injector #1, seven for Injector # 2, distributed regularly on a circumference to form an ellipsoidal-like hollow-cone geometry. The third injector is a six-hole Continental device, with five holes distributed over

Numerical Modelling and Optimization of the

pressures.

achieved when the spray is completely developed (t ~ 500 s).

Mixture Formation Process by Multi-Hole Injectors in a GDI Engine 181

quiescent vessel containing air at atmospheric backpressure and ambient temperature. The jets are enlightened by powerful flashes at different instants from SOI. Images are captured through a high resolution CCD camera, 0.5 s shutter time, 12 bit, at different times from SOI. The optical axis of the CCD is oriented either in a parallel or in an orthogonal way with respect to the spray propagation. Alignments of the jet directions with respect to the camera axis are actuated by a wet seal spherical holder enabling to tilt the injector in the angular range +/- 15°. The tip penetration of the considered jet, as well as the cone angle, is collected as a function of time. The images processing is based on background subtraction, filtering and edges determination. All the measurements are made on five-image averaged pictures for a statistical analysis of the cycle-to-cycle dispersion. A plateau value of the cone angle is

The injection strategies in the experimental campaign cover the entire injection pressure range for the three injectors. The pulse durations are calibrated to deliver 10, 20 and 50 mg of gasoline at different injection pressures. Some single injection tests are reported in Table 2. Fig.2 reports a typical energizing current signal to the solenoid for injecting 20 mg of fuel at the pressure of 10 MPa, and the correspondent fuel injection rate signals, as collected for the three injectors. The signals are averaged over one hundred shots. A shift of 0.35 ms is registered between the start of energizing current and the exiting of fuel from the nozzle, indicating a postponed answer of the mechanical parts. This delay remains practically unchanged for all the three devices. Differently, the fuel injection rate signals show different rise times: for both the Bosch injectors it is of about 70 s, while for the Continental one it is of about 170 s. Furthermore, the closure time for Injector #3 is longer than for the others, namely it is of about 70 s. The precise overlapping between signals relevant to Injectors #1 and #2 is indicative of an analogue behaviour of the moving equipment, while the Injector #3 has a slight larger inertia, hence greater opening and closing delays. This implies a

Pinj(MPa) 3 6 10 10 15 20 23 tinj (s) 1000 1900 1450 3600 2900 2600 2500 mf (mg) 10 20 20 50 50 50 50 Table 2. Time durations of the pulses for the desired fuel amounts at the indicated injection

0,0 0,5 1,0 1,5 2,0 2,5 3,0

time [ms]

Fig. 2. Energizing solenoid current (top) and fuel injection rates (bottom) for the three

Pinj 10 MPa <sup>t</sup>

Qinj = 20.00 mg/str

inj = 1,45 ms

Inj. 1 - 6 holes Inj. 2 - 7 holes Inj. 3 - 6 holes

different promptness availability of the fuel with the same command signals.

0.35 ms

solenoid current [A]

fuel inj. rate [mg/ms]

considered injectors at Pinj=10 MPa, mf = 20 mg.

a circumference and the sixth one in central position. Fig. 1 represents three sketches, each drawing the position of the holes on the relevant injector and the footprint of the spray axes on a plane placed at a distance of 30 mm from the holes themselves.


Table 1. Geometrical and flow rate characteristics of the three tested injectors.

Fig. 1. Holes distribution and spray footprint on a plane placed at 30 mm from the injector tip. Injector #1 (top left), Injector #2 (top right) and Injector #3 (bottom).

The maximum operating pressure for all the three injectors is forced up to 25 MPa for Injector #1 and #2, up to 20 MPa for Injector #3. Commercial gasoline is used (=740 kg/m3) delivered by a hydro-pneumatic injection system without rotating organs. The system is managed by a programmable electronic control unit (PECU) enabling to define the strategy typology in terms of number of injection events, durations and dwell times.

Two types of analysis are conducted: instantaneous mass flow rates of gasoline are measured by means of an AVL meter operating on the Bosch principle [Bosch, 1966; Wallace, 2002] under both single and double injection strategies; image processing techniques are applied to derive the single jet penetration length and cone angle over time in the single injection case. The measured instantaneous mass flow rate profile is integrated over the injection interval of time to gain the total injected mass, and to verify that the value of this last quantity is in accordance with that measured by means of a precision balance. The study of the fuel dispersion, instead, is realized in an optically-accessible high-pressure

a circumference and the sixth one in central position. Fig. 1 represents three sketches, each drawing the position of the holes on the relevant injector and the footprint of the spray axes

> Hole diameter (mm)

Static Flow at 10 MPa (g/s)

on a plane placed at a distance of 30 mm from the holes themselves.

number

Table 1. Geometrical and flow rate characteristics of the three tested injectors.

BOSCH HDEV 5.1 6 0.193 13.7

BOSCH HDEV 5.1 7 0.179 13.7

CONTINENTAL 6 0.190 13.1

Fig. 1. Holes distribution and spray footprint on a plane placed at 30 mm from the injector

The maximum operating pressure for all the three injectors is forced up to 25 MPa for Injector #1 and #2, up to 20 MPa for Injector #3. Commercial gasoline is used (=740 kg/m3) delivered by a hydro-pneumatic injection system without rotating organs. The system is managed by a programmable electronic control unit (PECU) enabling to define the strategy

Two types of analysis are conducted: instantaneous mass flow rates of gasoline are measured by means of an AVL meter operating on the Bosch principle [Bosch, 1966; Wallace, 2002] under both single and double injection strategies; image processing techniques are applied to derive the single jet penetration length and cone angle over time in the single injection case. The measured instantaneous mass flow rate profile is integrated over the injection interval of time to gain the total injected mass, and to verify that the value of this last quantity is in accordance with that measured by means of a precision balance. The study of the fuel dispersion, instead, is realized in an optically-accessible high-pressure

tip. Injector #1 (top left), Injector #2 (top right) and Injector #3 (bottom).

typology in terms of number of injection events, durations and dwell times.

Injector Type holes

**Injector #1** 

**Injector # 2** 

**Injector # 3** 

quiescent vessel containing air at atmospheric backpressure and ambient temperature. The jets are enlightened by powerful flashes at different instants from SOI. Images are captured through a high resolution CCD camera, 0.5 s shutter time, 12 bit, at different times from SOI. The optical axis of the CCD is oriented either in a parallel or in an orthogonal way with respect to the spray propagation. Alignments of the jet directions with respect to the camera axis are actuated by a wet seal spherical holder enabling to tilt the injector in the angular range +/- 15°. The tip penetration of the considered jet, as well as the cone angle, is collected as a function of time. The images processing is based on background subtraction, filtering and edges determination. All the measurements are made on five-image averaged pictures for a statistical analysis of the cycle-to-cycle dispersion. A plateau value of the cone angle is achieved when the spray is completely developed (t ~ 500 s).

The injection strategies in the experimental campaign cover the entire injection pressure range for the three injectors. The pulse durations are calibrated to deliver 10, 20 and 50 mg of gasoline at different injection pressures. Some single injection tests are reported in Table 2. Fig.2 reports a typical energizing current signal to the solenoid for injecting 20 mg of fuel at the pressure of 10 MPa, and the correspondent fuel injection rate signals, as collected for the three injectors. The signals are averaged over one hundred shots. A shift of 0.35 ms is registered between the start of energizing current and the exiting of fuel from the nozzle, indicating a postponed answer of the mechanical parts. This delay remains practically unchanged for all the three devices. Differently, the fuel injection rate signals show different rise times: for both the Bosch injectors it is of about 70 s, while for the Continental one it is of about 170 s. Furthermore, the closure time for Injector #3 is longer than for the others, namely it is of about 70 s. The precise overlapping between signals relevant to Injectors #1 and #2 is indicative of an analogue behaviour of the moving equipment, while the Injector #3 has a slight larger inertia, hence greater opening and closing delays. This implies a different promptness availability of the fuel with the same command signals.


Table 2. Time durations of the pulses for the desired fuel amounts at the indicated injection pressures.

Fig. 2. Energizing solenoid current (top) and fuel injection rates (bottom) for the three considered injectors at Pinj=10 MPa, mf = 20 mg.

Numerical Modelling and Optimization of the

Mixture Formation Process by Multi-Hole Injectors in a GDI Engine 183

Fig.3 is drawn to give an idea of the operation of one of the considered injectors, namely Injector #1, under different injection pressures and with current signals set to deliver two injected quantities, mf=10 mg and mf=50 mg. The injection pressure is equal to 3 and 6 MPa for mf=10 mg, and 10 and 20 MPa for mf=50 mg. The figure highlights the great flexibility of the injector in its capability to range from what are low to high load engine conditions. Images of the jets evolving in the optically accessible vessel at different instants from SOI are reported in Figs. 4-6. Fig. 4 is a sequence of the propagating jets produced by the Injector #1 at 50, 100, 200, 500 and 700 s from the SOI at the injection pressure of 10 MPa and 50 mg delivered gasoline. The lateral view of the spray allows distinguishing the origin of the single jets close to the nozzle exit. Four single well-confined arrows appear in the CCD view plane, while the last two are in the back side. The chosen orientation of the injector enables a complete view of the jet placed in the bottom part of the figure (horizontal), and permits a precise determination of its length, under the main hypothesis that all the jets behave in a similar way. In Figs. 5 and 6 the propagating sprays for the Injector #2 and Injector #3 are reported, respectively, under the same injection conditions of Fig. 4. Differences in the structure of the global spray appear due to the different number of holes and directions, although the overall behaviour appears almost unchanged. Injectors #1 and #2 assume analogues total angle while Injector #3 has a larger one. Detectable differences should

appear from the punctual measurements of the penetrations and cone-angles.

The spray images highlight a complex structure of the evolving jets with inner bunches or fuel pockets picked out by the highest intensities of the scattered light. This aspect is indicative of a non homogeneous distribution of the fuel and is peculiar of the injection process. Fig. 7 reports an example of the front-view images of the sprays from the three injectors taken at 700 s from the SOI, 10 MPa injection pressure and 50 mg injected fuel. The number of jets, their directions and the difference in the footprint figures for the diverse devices are evident. The six jets of Injector #1 are gathered together with respect to the other two. Injector #2 has a wider rose of the jets, with the seven sprays well distinguishable.

Injector #1 – 6 holes Injector #2 – 7 holes Injector #3 – 6 holes

Fig. 7. Frontal-view images for the sprays issuing from the three considered injectors at 700

An idea of the behaviour of Injectors #1 and #2 under double strategies is given in Figs. 8 and 9. Fig. 8 reports the fuel injection rate signal collected for a double-pulse strategy at the injection pressure of 6 MPa for Injector #1, together with the timing of the solenoid driving current. Each pulse is equal to 0.9 ms in duration, hence the gasoline injected mass is split in percentages equal to 50% plus 50% of the total amount. Stability and repetitiveness of the injection events is studied by varying the value of the dwell time, dw, from the minimum value up to 1.5 ms. The minimum value of this variable, below which the opening of the

s from the SOI, Pinj = 10MPa, mf = 50 mg.

Fig. 3. Measured mass flow rates for four injection pressures and two values of the total injected mass for Injector #1.

Fig. 4. Sequence of sprays from Injector #1 taken at different time from the SOI at Pinj = 10 MPa, mf = 50mg.

Fig. 5. Sequence of sprays from Injector #2 taken at different time from the SOI at Pinj = 10 MPa, mf = 50mg.

Fig. 6. Sequence of sprays from Injector #3 taken at different time from the SOI at Pinj = 10 MPa, mf = 50mg.

182 Computational Simulations and Applications

Pinj=3 MPa mf

Pinj=6 MPa mf

Pinj=10 MPa mf

Pinj=20 MPa mf

=10 mg

=10 mg

=50 mg

=50 mg

0 0.001 0.002 0.003 0.004 Time (s)

Fig. 3. Measured mass flow rates for four injection pressures and two values of the total

Fig. 4. Sequence of sprays from Injector #1 taken at different time from the SOI at

Fig. 5. Sequence of sprays from Injector #2 taken at different time from the SOI at

Fig. 6. Sequence of sprays from Injector #3 taken at different time from the SOI at

0

injected mass for Injector #1.

Pinj = 10 MPa, mf = 50mg.

Pinj = 10 MPa, mf = 50mg.

Pinj = 10 MPa, mf = 50mg.

10

20

Injected mass flow rate (kg/s)

30

40

Fig.3 is drawn to give an idea of the operation of one of the considered injectors, namely Injector #1, under different injection pressures and with current signals set to deliver two injected quantities, mf=10 mg and mf=50 mg. The injection pressure is equal to 3 and 6 MPa for mf=10 mg, and 10 and 20 MPa for mf=50 mg. The figure highlights the great flexibility of the injector in its capability to range from what are low to high load engine conditions.

Images of the jets evolving in the optically accessible vessel at different instants from SOI are reported in Figs. 4-6. Fig. 4 is a sequence of the propagating jets produced by the Injector #1 at 50, 100, 200, 500 and 700 s from the SOI at the injection pressure of 10 MPa and 50 mg delivered gasoline. The lateral view of the spray allows distinguishing the origin of the single jets close to the nozzle exit. Four single well-confined arrows appear in the CCD view plane, while the last two are in the back side. The chosen orientation of the injector enables a complete view of the jet placed in the bottom part of the figure (horizontal), and permits a precise determination of its length, under the main hypothesis that all the jets behave in a similar way. In Figs. 5 and 6 the propagating sprays for the Injector #2 and Injector #3 are reported, respectively, under the same injection conditions of Fig. 4. Differences in the structure of the global spray appear due to the different number of holes and directions, although the overall behaviour appears almost unchanged. Injectors #1 and #2 assume analogues total angle while Injector #3 has a larger one. Detectable differences should appear from the punctual measurements of the penetrations and cone-angles.

The spray images highlight a complex structure of the evolving jets with inner bunches or fuel pockets picked out by the highest intensities of the scattered light. This aspect is indicative of a non homogeneous distribution of the fuel and is peculiar of the injection process. Fig. 7 reports an example of the front-view images of the sprays from the three injectors taken at 700 s from the SOI, 10 MPa injection pressure and 50 mg injected fuel. The number of jets, their directions and the difference in the footprint figures for the diverse devices are evident. The six jets of Injector #1 are gathered together with respect to the other two. Injector #2 has a wider rose of the jets, with the seven sprays well distinguishable.

Fig. 7. Frontal-view images for the sprays issuing from the three considered injectors at 700 s from the SOI, Pinj = 10MPa, mf = 50 mg.

An idea of the behaviour of Injectors #1 and #2 under double strategies is given in Figs. 8 and 9. Fig. 8 reports the fuel injection rate signal collected for a double-pulse strategy at the injection pressure of 6 MPa for Injector #1, together with the timing of the solenoid driving current. Each pulse is equal to 0.9 ms in duration, hence the gasoline injected mass is split in percentages equal to 50% plus 50% of the total amount. Stability and repetitiveness of the injection events is studied by varying the value of the dwell time, dw, from the minimum value up to 1.5 ms. The minimum value of this variable, below which the opening of the

Numerical Modelling and Optimization of the


convergence criterion is satisfied.

50+50%, 30+70% and 70+30% for Injector #2.

**3. Numerical simulation of the GDI spray dynamics** 

Mixture Formation Process by Multi-Hole Injectors in a GDI Engine 185

minj m = 9.38 mg/str inj = 9.17 mg/str

minj = 7.00 mg/str minj = 13.27 mg/str

minj = 14.29 mg/str minj = 5.61 mg/str

**Pinj = 10 MPa** minj = 20.06 mg/str

0 0.5 1 1.5 2 2.5 3 3.5 Time (ms)

Fig. 9. Fuel injection rates for a single injection and for double injection strategies split at

Reducing development time, improving performances and reliability of numerical models is of crucial importance for the design of new engine components. The use of optimization methods coupled with modern CFD tools is today very effective to accomplish these tasks, especially where uncertainty exists about a number of involved constants. Numerical procedures, in fact, may be used to generate a series of progressively improved solutions to the optimization problem, starting from an initial one. The process is terminated when some

In the present section the assessment of a simulation tool reproducing the spatio-temporal dynamics of sprays issuing from new generation high pressure injectors under various operating conditions is presented. The model, developed within the AVL FireTM code environment, is conceived to exploit the previously described experimental data in part as

In order to numerically simulate the effected tests, the spray is hypotesised to enter the top surface of a properly dimensioned computational domain of cylindrical shape, where the injector is supposed to be placed in central position. According to the discrete droplet method (DDM), the spray is considered as a train of droplets of given size, suffering various

input parameters, in part as terms of comparison for the numerical results.

second injection event interferes with the closing of the previous one, due to the electrohydraulic inertia of the internal mobile equipment, is equal to 320 s. Fig. 9 represents four different strategies of Injector #2, all delivering a total mass equal to 20 mg at the injection pressure of 10 MPa. It is to be remarked that also for this injector the minimum dwell time allowing distinguishable events is equal to 320 s. The top of Fig. 9 shows the single pulse strategy. The setting for a 50% fuel injected during the first pulse and the remaining 50% delivered during the second one is then plotted. The successive strategy exhibits a 30+70% splitting, while at the bottom of the figure the 70+30% case is represented. This last could serve to realize, in the engine working cycle, a homogeneous combustion followed by a post-injection aimed at improving the exhaust conditions for a more effective catalytic conversion. Note that these percentages are just indicative and can be varied at will by modulating the energizing currents duration. Since no memory of the first pulse is induced in the second one above the minimum dwell time, the regulation of the percentages is a mere question of current settings.

Fig. 8. Fuel injection rate for a double injection strategy at the minimum dw (top) with the corresponding exciting solenoid currents (bottom) for Injector #1. Pinj= 6 MPa.

second injection event interferes with the closing of the previous one, due to the electrohydraulic inertia of the internal mobile equipment, is equal to 320 s. Fig. 9 represents four different strategies of Injector #2, all delivering a total mass equal to 20 mg at the injection pressure of 10 MPa. It is to be remarked that also for this injector the minimum dwell time allowing distinguishable events is equal to 320 s. The top of Fig. 9 shows the single pulse strategy. The setting for a 50% fuel injected during the first pulse and the remaining 50% delivered during the second one is then plotted. The successive strategy exhibits a 30+70% splitting, while at the bottom of the figure the 70+30% case is represented. This last could serve to realize, in the engine working cycle, a homogeneous combustion followed by a post-injection aimed at improving the exhaust conditions for a more effective catalytic conversion. Note that these percentages are just indicative and can be varied at will by modulating the energizing currents duration. Since no memory of the first pulse is induced in the second one above the minimum dwell time, the regulation of the percentages is a

> -0.5 0 0.5 1 1.5 2 2.5 3 3.5 Time (ms)

Fig. 8. Fuel injection rate for a double injection strategy at the minimum dw (top) with the

corresponding exciting solenoid currents (bottom) for Injector #1. Pinj= 6 MPa.

**0.9 ms 0.9 ms**

**0.32 ms**

**0.4 ms**

mere question of current settings.


0

5

10

Solenoid current (A)

15

20

25


0

10

Fuel injection rate (mg/ms)

20

30

Fig. 9. Fuel injection rates for a single injection and for double injection strategies split at 50+50%, 30+70% and 70+30% for Injector #2.
