Lefebvre A.H., Ballal D.R. Gas Turbine Combustion: Alternative Fuels and Emissions
Подождите немного. Документ загружается.
170 Gas Turbine Combustion: Alternative Fuels and Emissions, Third Edition
that the combustion engineer be fully aware of all the factors that govern
lean blowout limits in practical combustion systems.
5.7.3 Water injection Technique
A much cheaper alternative to the determination of stability performance,
as described above, is to use the water injection technique. This technique
allows complete stability loops to be drawn for full-scale combustors
while operating within their normal range of velocities and fuel/air ratios
[2,18–24]. Air is supplied from a fan at atmospheric pressure, and low pres-
sures are simulated by introducing water into the combustion zone. The
essence of the method is the theoretical equivalence, on a global reaction
rate basis, between a reduction in reaction pressure and a reduction in reac-
tion temperature, which, in this instance, is accomplished by the addition
of water [2,18].
One of the most useful applications of the technique is in obtaining
blowout data for various designs of ameholder. The ameholder under test
is mounted in a pipe that is connected to the outlet of a fan via a preheat
combustion chamber. Upstream of the ameholder are manifolds designed
to inject fuel and water uniformly across the gas stream. Usually, the fuel
and water are premixed, as shown in Figure 5.13, but this is not essential to
the method. The preheat temperature should be preset to a value that is high
enough to ensure that the fuel and water are fully vaporized upstream of the
ameholder.
Burning
range
Mach number
Altitude
Figure 5.12
Stability performance of an aircraft combustor.
Combustion Performance 171
The test procedure is quite simple. The velocity and temperature of the
gas owing over the stabilizer are adjusted to the desired values; the fuel is
turned on, and a ame is established in the recirculation zone downstream
of the stabilizer. Water is then gradually mixed with the fuel in increasing
amounts until extinction occurs. This process is repeated at a sufcient num-
ber of fuel ow rates for a complete stability loop to be drawn. A typical
stability loop is shown in Figure 5.14, in which the ordinate represents the
Air supplied
from fan at
atmospheric
pressure
Preheat
chamber
Fuel
Duplex atomizer
Water
Stabilizer
Uniform injection
of fuel water mixture
Figure 5.13
Basic rig requirements for stability tests using water injection technique.
1.3
1.2
1.1
25.4 mm
Combustion possible
within this loop
1.0
0.9
0.8
0.7
0.6
0.5
0.4 0.5 0.6 0.7
U
g
= 168 m/s
T
o
= 773 K
0.8
Water/fuel mass ratio
Equivalence ratio
0.9 1.0 1.1 1.2
Figure 5.14
Typical stability loop obtained using water injection technique.
172 Gas Turbine Combustion: Alternative Fuels and Emissions, Third Edition
equivalence ratio of the kerosine/air mixture, and the abscissa denotes the
mass ratio of water ow to kerosine ow. Curves of this type provide useful
data whereby the basic stability of various designs of ameholder may be
compared. The only assumption involved is a reasonable one, namely, that
the gutter requiring the largest amount of water to cause ame extinction
has the best stability. The value of the technique is further enhanced by a
relationship (derived from global reaction rate theory) between the water/
fuel mass ratio and the equivalent reduction in combustion pressure. This
relationship is illustrated in Figure 5.15. The curves shown are for octane
fuel and represent calculations by Taylor [19]. Almost identical curves were
obtained for kerosine [18]. In general, it is found that injecting equal amounts
of water and liquid fuel into the combustion zone is equivalent to halving the
combustion pressure.
The simulation of low pressures by nitrogen gas dilution has also been
used successfully by several workers, including Norster [25] and Sturgess
et al. [26]. Nitrogen is clearly more costly than water, but it does not require
a preheat combustor, which makes it attractive for studies on small-scale
ameholders.
1.0
φ = 0.9
1.0
1.1
1.2
0.1
T
o
= 773 K
01
Water/fuel mass ratio
Effective pressure ratio
23
Figure 5.15
Relationship between water-fuel ratio and equivalent pressure ratio for isooctane fuel. (From
Taylor, J.S., “Large-Scale Bluff Body Flame Stabilization,” MSc thesis, School of Mechanical
Engineering, Purdue University, 1980.)
Combustion Performance 173
5.8 Bluff-Body Flameholders
Bluff-body ameholders are widely used to stabilize ames in owing
combustible mixtures, and their many practical applications include ram-
jet and turbojet afterburner systems. The practical importance of bluff-body
stabilizers has given rise to a large number of theoretical and experimental
studies. Much of our present understanding of the ame stabilization pro-
cess is due to the pioneering studies carried out in the 1950s by Longwell
et al. [27], Zukowski and Marble [28], Barrère and Mestre [29], De Zubay
[30], and Spalding [31]. More recent studies include those of Lefebvre et al.
[20–24,32–34], whose work culminated in equations for predicting stability
limits in terms of bluff-body dimensions, blockage ratio, and the pressure,
temperature, velocity, turbulence properties, and equivalence ratio of the
incoming fresh mixture. Plee and Mellor [35] have successfully correlated
lean blowout data for bluff-body stabilized ames, using a characteristic-
time model.
5.8.1 experimental Findings on Bluff-Body Flame Stabilization
The ameholding properties of bluff-body stabilizers have been studied
extensively for both homogeneous gaseous fuel–air mixtures and for com-
bustion zones supplied with heterogeneous mixtures of fuel drops and air.
5.8.1.1 Homogeneous Mixtures
Ballal and Lefebvre [32] investigated the effects of inlet air temperature,
pressure, velocity, and turbulence on the lean blowout performance of ame-
holders supplied with homogeneous mixtures of gaseous propane and air.
The apparatus employed comprised a ameholder in the form of a hollow
cone that was located at the center of a circular pipe with its apex pointing
upstream. Fourteen geometrically similar conical bafes were manufactured
to various sizes and used in conjunction with three different pipe diameters
to allow the effects of bafe size and blockage ratio to be studied indepen-
dently over wide ranges of operating conditions.
The inuence of ameholder size on blowout limits is illustrated in
Figure 5.16. In this gure, measured values of weak extinction equivalence
ratio are shown plotted against ameholder diameter for two different levels
of approach stream velocity and three different values of blockage, B
g
. (Note
that B
g
is dened as the ratio of the ameholder cross-sectional area to that
of the pipe.) The improvement in stability with an increase in ameholder
diameter is attributed to the longer residence time of the reactants in the
recirculation zone. Figures 5.17 and 5.18 contain similar data to illustrate the
effects of inlet air temperature and approach velocity on lean blowout limits.
All the experimental data obtained by Ballal and Lefebvre [32] are consistent
174 Gas Turbine Combustion: Alternative Fuels and Emissions, Third Edition
0.6
P = 9×10
4
Pa
T
0
= 300 K
B
g
0.25
0.11
0.04
U = 16 m/s
U = 10 m/s
0.5
0.4
φ
WE
φ
WE
0.3
0.6
0.5
0.4
0.3
04 8
Flameholder diameter, cm
12
Figure 5.16
Inuence of ameholder size on weak extinction limits for propane–air mixtures. (From
Ballal, D.R. and Lefebvre, A.H., Journal of Engineering for Power, 101(3), 343–348, 1979. With
permission.)
0.8
0.25
0.11
0.04
φ
WE
U = 50m/s
D
p
= 0.15 m
P = 10
5
Pa
B
g
0.7
0.6
0.5
0.4
0.3
300 400
Inlet air temperature, K
500 600
Figure 5.17
Inuence of temperature on weak extinction limits for propane–air mixtures. (From Ballal, D.R.
and Lefebvre, A.H., Journal of Engineering for Power, 101(3), 343–348, 1979. With permission.)
Combustion Performance 175
with the notion that lean blowout limits are improved (i.e., extended to lower
values of equivalence ratio) by increases in pressure and temperature (via
increase in reaction rates) and by an increase in ameholder size and/or
reduction in approach velocity (via increase in residence time).
Baxter and Lefebvre [24] employed fully vaporized kerosine–air mixtures
to examine the effects of various ow parameters on the lean blowout limits
of Vee-gutter ameholders of various widths and included angles. A novel
feature of their apparatus is that it includes both a preheat combustor and a
heat exchanger in order to allow the effects of inlet air vitiation on stability
limits to be examined independently from those of inlet gas temperature.
The adverse effect on ame stability of increasing the degree of vitiation
while maintaining a constant inlet temperature is illustrated in Figure 5.19.
The data contained in this gure were obtained using two Vee-gutter ame-
holders, both of 60 degree included angle. The gutter widths and ow condi-
tions were selected to represent those encountered in modern afterburner
systems. An interesting feature of this gure is that the slopes of the lines
drawn through the data points become increasingly steep with an increase
in upstream vitiation, as indicated by the preheater equivalence ratio, ϕ
p
.
This is due to the rapid depletion of available oxygen with the increase in ϕ
p
.
The strong adverse effect of inlet air vitiation on weak extinction limits, as
illustrated in Figure 5.19, is clearly of practical importance to the design of
afterburner systems, where the hot gas stream approaching the ameholder
array has experienced appreciable oxygen depletion because of prior com-
bustion in the main combustor.
The shape of a bluff-body ameholder affects its stability characteristics
through its inuence on the size and shape of the wake region. The effect of
0.34
B
g
0.25
0.11
0.04
D
p
= 10 cms
P = 9×10
4
Pa
T
o
= 300 K
0.7
0.8
0.6
0.5
0.4
0204060
Mainstream velocity, m/s
80 100
φ
WE
Figure 5.18
Inuence of velocity on weak extinction limits for propane–air mixtures. (From Ballal, D.R. and
Lefebvre, A.H., Journal of Engineering for Power, 101(3), 343–348, 1979. With permission.)
176 Gas Turbine Combustion: Alternative Fuels and Emissions, Third Edition
shape is illustrated in Figure 5.20, which shows lean blowout data for three
gutters having different included angles, but the same projected width.
These data conrm the results of Barrère and Mestre [29] in showing that the
characteristic dimension of a bluff-body ameholder should not be its geo-
metric width, w, but rather the maximum aerodynamic width of the wake
created behind it, w
a
. The ratio w
a
/w increases with an increase in gutter
included angle, thereby improving stability performance by enlarging the
recirculation-zone volume.
0.44
T
o
= 823 K
Mn = 0.23
θ = 60°
w, mm
30.2
39.7
0.42
0.40
0.38
0.36
0.34
0.32
0.30
0.1 0.2 0.3
Preheater equivalence ratio
Weak extinction equivalence ratio
0.4 0.5 0.6 0.7 0.8
Figure 5.19
Inuence of inlet-air vitiation on weak extinction limits. (From Baxter, M.R. and Lefebvre, A.H.,
Journal of Engineering for Gas Turbines and Power, 114(4), 776–782, 1992. With permission.)
θ
0.38
0.36
0.34
0.32
0.30
0.28
0.18
0.26
0.20
Approach stream mach number
Weak extinction equivalence ratio
0.22
T
o
= 733 K
φ
p
= 0.19
w = 39.7 mm
θ
0.24 0.26
45°
60°
90°
Figure 5.20
Inuence of approach stream velocity and ameholder shape on weak extinction limits. (From
Baxter, M.R. and Lefebvre, A.H., Journal of Engineering for Gas Turbines and Power, 114(4), 776–782,
1992. With permission.)
Combustion Performance 177
Surprisingly perhaps, the inuence of combustion pressure on lean blowout
limits is quite small. Ballal and Lefebvre [32] observed only a slight increase
in ϕ
WE
as the inlet air pressure was reduced from 1.0 to 0.2 bars.
5.8.1.2 Heterogeneous Mixtures
Ballal and Lefebvre [33] examined the factors governing the lean blowout
limits of bluff-body stabilized ames supplied with owing mixtures of
liquid fuel drops and air. Their test program included wide variations in
inlet air pressure and velocity, and also covered wide ranges of fuel volatil-
ity and mean fuel drop size. The experimental apparatus was essentially the
same as that used previously for homogeneous fuel/air mixtures, except that
means were provided for spraying fuel drops directly into the combustion
zone through a simplex pressure nozzle that was located inside the coni-
cal ameholder. Ten fuel nozzles of different ow numbers were used in
order to cover wide ranges of air velocity, air pressure, and mean drop size.
Some of the results obtained in this investigation are shown in Figures 5.21
through 5.23. They show that lean blowout limits are improved, i.e., extended
to lower equivalence ratios, by increases in fuel volatility and air pressure,
and by reductions in mainstream velocity and mean drop size. Figure 5.21
indicates a strong effect of pressure on ϕ
WE
, which is in marked contrast
to the small effect observed with homogeneous mixtures. This result is of
special interest because it shows that the inuence of pressure on the lean
0.8
Heavy fuel oil
Theory
T
o
= 300 K
SMD = 100 µm
U = 30 m/s
Diesel
i-octane
0.6
0.4
0 0.5
P, Pa ×10
5
1.0
φ
WE
Figure 5.21
Inuence of pressure on weak extinction limits. (From Ballal, D.R. and Lefebvre, A.H., Journal
of Engineering for Power, 102(2), 416–421, 1980. With permission.)
178 Gas Turbine Combustion: Alternative Fuels and Emissions, Third Edition
Theory
1.0
0.6
0.2
0 40
U, m/s
80
P = 10
5
Pa
T
o
= 300 K
SMD = 100 µm
Heavy fuel oil
Diesel
oil
i-octane
φ
WE
Figure 5.22
Inuence of velocity on weak extinction limits. (From Ballal, D.R. and Lefebvre, A.H., Journal of
Engineering for Power, 102(2), 416–421, 1980. With permission.)
Heavy fuel oil
Diesel oil
isooctane
Theory
0.8
0.7
0.6
0.5
0.4
050 100
SMD, microns
150 200
T
o
= 300 K
P = 100 kPa
U = 15 m/s
φ
WE
Figure 5.23
Inuence of mean drop size on weak extinction limits. (From Ballal, D.R. and Lefebvre, A.H.,
Journal of Engineering for Power, 102(2), 416–421, 1980. With permission.)
Combustion Performance 179
blowout limits of heterogeneous mixtures is manifested primarily through
fuel evaporation rates rather than chemical reaction rates. Also worthy of
note is that in Figure 5.23 the curves of ϕ
WE
vs. SMD become horizontal at low
values of SMD. This denotes that drop sizes are small enough for evapora-
tion to be complete within the reaction zone. Thus, over the range of SMDs,
where ϕ
WE
is independent of SMD, the combustion zone operates effectively
as a homogeneous stirred reactor.
5.8.2 Summary of experimental Findings
A considerable body of evidence has been accumulated on the factors
governing ame stability, and several broad conclusions can be drawn. In
general, stability limits are extended by:
1. A reduction in approach stream velocity
2. An increase in approach stream temperature
3. An increase in gas pressure
4. A reduction in turbulence intensity
5. Any change in equivalence ratio toward unity
6. An increase in ameholder size
7. An increase in ameholder base-drag coefcient
8. A reduction in ameholder blockage (for a constant ameholder
size)
For liquid fuels, stability is further improved by:
1. An increase in fuel volatility
2. Finer atomization, i.e., reduction of mean drop size
5.9 Mechanisms of Flame Stabilization
Many theoretical studies have been carried out on bluff-body ame stabiliza-
tion and several models have been proposed to account for the experimental
observations on ame extinction. Most of these models tend to fall into two
main categories. One of these, following Longwell et al. [27], views the wake
region of a bluff-body essentially as a homogeneous chemical reactor. Flame
extinction occurs when the time available for chemical reaction becomes
less than the time required to generate sufcient heat to raise the fresh mix-
ture up to its ignition temperature. The other category includes models in
which attention is focused mainly on the shear layer surrounding the wake