A.
Appendix A
Figure A.1: Shadowgraph images of a fuel spray at 50bar injection pressure with a high
speed camera at a capture rate of 7500Hz and 20µs exposure time
0.13 ms
0.27 ms
0.40 ms
0.53 ms
0.67 ms
0.80 ms
0.93 ms
1.07 ms
1.20 ms
1.33 ms
1.47 ms
1.60 ms
1.73 ms
1.87 ms
2.00 ms
2.13 ms
2.27 ms
2.40 ms
2.53 ms
2.67 ms
2.80 ms
2.93 ms
3.07 ms
3.20 ms
3.33 ms
3.47 ms
3.60 ms
3.73 ms
3.87 ms
4.00 ms
4.13 ms
4.27 ms
192
Figure A.2: Shadowgraph images of a fuel spray at 110bar injection pressure with a high
speed camera at a capture rate of 7500Hz and 20µs exposure time
0.13 ms
0.27 ms
0.40 ms
0.53 ms
0.67 ms
0.80 ms
0.93 ms
1.07 ms
1.20 ms
1.33 ms
1.47 ms
1.60 ms
1.73 ms
1.87 ms
2.00 ms
2.13 ms
2.27 ms
2.40 ms
2.53 ms
2.67 ms
2.80 ms
2.93 ms
3.07 ms
3.20 ms
3.33 ms
3.47 ms
3.60 ms
3.73 ms
3.87 ms
4.00 ms
4.13 ms
4.27 ms
193
B.
Appendix B
Dimensional
Analysis
The variables considered relevant to drop size formation are as shown in Table
B.1. The dimensions M, L and T stand for mass, distance and time, respectively.
Variable
Symbol
Dimensions
Droplet diameter
d
L
Nozzle diameter
D
L
Fuel density
ρ
ML
-3
Fuel viscosity
µ
ML
-1
T
-1
Fuel surface tension
σ
MT
-2
Fuel spray velocity
u
LT
-1
Table B.1: Dimensions of assumed variables
Following the Buckingham’s Pi theorem [264], the 6 variables and 3
dimensions give 6 – 3 = 3 dimensionless groups. The three groups were identified as
follows:
𝛱
1
=
𝑑
𝐷
B.1
𝛱
2
=
𝜌𝑢𝐷
µ
B.2
𝛱
3
=
𝜌𝑢
2
𝐷
𝜎
B.3
Therefore, the relationship between the groups can be expressed as:
194
𝑑
𝐷
= 𝑓 [
𝜌𝑢𝐷
µ
,
𝜌𝑢
2
𝑑
𝜎
]
B.4
This relates well to the theory on atomisation as described in Section 2.3 as Π
2
is the Reynolds number and Π
3
the Weber number. A further variable that could be
considered is fluid bulk modulus of elasticity and as such a further dimensionless
group of Mach number could be used in the analysis. However, as explained by Iqbal
et al. [265], Mach number is only relevant if fluid velocity is greater than the speed of
sound. As could be seen in Figure 3.5 (page 78), the fuel spray velocity reached in the
current study remains below 52 m/s for all pressures up to 110 bar, meaning the Mach
number could be excluded from the current study. It is worthwhile noting that the
spray tip velocity was calculated from high speed images based on the fuel reaching
the measuring point at x = -10 mm and z = 60 mm. Initial spray velocity within the
injector nozzle and leaving the nozzle is likely to be higher, thus higher Weber and
Reynolds numbers could be computed. Nevertheless, it is a reasonable assumption that
the spray velocity would not exceed at any point the speed of sound of 343 m/s at
ambient temperature in air.
Figure B.1: d/D change with We and Re for sprays at 50-110 bar injection pressure
Figure B.1 displays the change in dimensionless droplet size with respect to
the Weber and Reynolds numbers. In both cases, it is evident that an increase in
injection pressure results in increased Weber and Reynolds number. This in turn
25.00
30.00
35.00
40.00
45.00
50.00
0.01
0.015
0.02
0.025
0.03
16,000
17,000
18,000
19,000
20,000
21,000
22,000
23,000
24,000
Weber Number
d/D
Reynolds Number
Reynolds Number
Weber Number(gas)
195
induces a reduction in droplet size. The calculated Weber number values in the range
26 < We < 47 suggest that secondary droplet break-up occurs. According to Table 2.2
(Page 47) the mechanisms responsible for the secondary break-up are bag-and-stamen
and chaotic.
196
Do'stlaringiz bilan baham: |