Proceedings of ASME Turbo Expo 2013
GT2013
June 3-7, 2013, San Antonio, Texas USA
GT2013-94183
EFFECT OF CHANGE IN FUEL COMPOSITIONS AND HEATING VALUE ON
IGNITION AND PERFORMANCE FOR SIEMENS SGT-400 DRY LOW EMISSION
COMBUSTION SYSTEM
Kexin Liu
Siemens Industrial Turbomachinery Limited
Lincoln, LN5 7FD, UK
kexin.liu@siemens.com
Pete Martin
Siemens Industrial Turbomachinery Limited
Lincoln, LN5 7FD, UK
pete.martin@siemens.com
Victoria Sanderson
Siemens Industrial Turbomachinery Limited
Lincoln, LN5 7FD, UK
victoria.sanderson@siemens.com
Phill Hubbard
Siemens Industrial Turbomachinery Limited
Lincoln, LN5 7FD, UK
phill.hubbard-gtec@siemens.com
ABSTRACT
The influence of changes in fuel composition and heating
value on the performance of an industrial gas turbine
combustor was investigated. The combustor tested was a single
cannular combustor for Siemens SGT-400 13.4 MW dry low
emission (DLE) engine. Ignition, engine starting, emissions,
combustion dynamics and flash back through burner metal
temperature monitoring were among the parameters
investigated to evaluate the impact of fuel flexibility on
combustor performance.
Lean ignition and extinction limits were measured for
three fuels with different heat values in term of Wobbe Index
(WI): 25, 28.9 and 45 MJ/Sm3 (natural gas). The test results
show that the air fuel ratio (AFR) at lean ignition/extinction
limits decreases and the margin between the two limits tends to
be smaller as fuel heat value decreases. Engine start tests were
also performed with a lower heating value fuel and results were
found to be comparable to those for engine starting with
natural gas.
The combustor was further tested in a high pressure air
facility at real engine operating conditions with different fuels
covering WIs from 17.5 to 70 MJ/Sm3. The variation in fuel
composition and heating value was achieved in a gas mixing
plant by blending natural gas with CO2, CO, N2 and H2 (for the
fuel with WI lower than natural gas) and C3H8 (for the fuel
with WI higher than natural gas).
Test results show that a benefit in NOx reduction can be
seen for the lower WI fuels without H2 presence in the fuel and
there are no adverse impacts on combustor performance except
for the requirement of higher fuel supply pressure, however,
this can be easily resolved by minor modification through the
fuel injection design.
Test results for the H2 enriched and higher WI fuels show
that NOx, combustion dynamics and flash back have been
adversely affected and major change in burner design is
required. For the H2 enriched fuel, the effect of CO and H2 on
combustor performance was also investigated for the fuels
having a fixed WI of 29 MJ/Sm3. It is found that H2 dominates
the adverse impact on combustor performance. The chemical
kinetic study shows that H2 has significant effect on flame
speed change and CO has significant effect on flame
temperature change.
Although the tests were performed on the Siemens SGT400 combustion system, the results provide general guidance
for the challenge of industrial gas turbine burner design for
fuel flexibility.
INTRODUCTION
To tackle the issues associated with fuel security and
environment, fuel flexibility has become increasingly
important for industrial gas turbines for power generation and
oil and gas industries. This presents a design challenge for
industrial gas turbines to ensure reliability with varying fuels
across an operational range. In this paper, the extension to the
standard natural gas fuel range is considered. The
considerations for such non-standard fuels for industrial gas
turbine combustion systems are the robustness of the
combustor to auto-ignition, flashback, acceptable combustion
1
Copyright © Siemens Industrial Turbomachinery Limited
dynamic pressure amplitude and emissions and neutral
flameholding [1-8].
Non-standard gas fuels normally have a different Wobbe
Index (WI) to natural gas, either higher or lower. WI is an
indicator of the interchangeability of fuel gases and is defined
as the ratio of lower heating value (LHV) to the square root of
the specific gravity (SG) of the fuel, WI LHV / SG . WI is
used for the comparison of energy release from gaseous fuels
with different compositions; for example, if two fuels have
identical WI then the energy release will also be identical given
the same volumetric flow rate. This is used as the primary
factor in evaluation of the gas fuel flexibility on a gas turbine.
In this paper a fuel with WI between 20 and 37 MJ/Sm3 is
named as MCV (Medium Calorific Value) fuel and a fuel with
WI higher than 49 MJ/Sm3 is classified as HCV (Higher
Calorific Value) fuel. MCV fuel can be further divided as
normal MCV fuel and hydrogen enriched MCV fuel. The
impact of non-standard fuel on combustor performance
depends on the compositions in the fuel and is reviewed here.
The WI values reported here are corrected to ISO fuel
temperature (15 C hence the unit is MJ/Sm3)
Normal MCV fuel
Normal MCV fuels include landfill gas, a by-product of
decomposing organic waste, and biogas, produced by chemical
process which includes anaerobic digestion or fermentation of
biodegradable materials. The quality of such fuel is highly
dependent on its feedstock and the main compositions in such
fuels are methane (CH4), diluents such as carbon dioxide (CO2)
and nitrogen (N2). The application for these types of gases as a
renewable energy source, instead of releasing or flaring to the
atmosphere offers environmental benefits. The reduction in
methane as a greenhouse gas (GHG) is important as this is
approximately 21 times more harmful to the atmosphere than
CO2.
The benefit of NOx reduction by using normal MCV fuels
was widely reported [1, 9-14], for example: 20% reduction
with 35% N2 in the fuel [1], 40% reduction with 50% N2 in the
fuel [2] and about 50% reduction with 35% CO2 in the fuel [6].
NOx is also sensitive to the selection of diluents, generally
more NOx is produced if N2 is selected as diluent [7, 9]. There
are a number of factors that can contribute to the reduction in
NOx using such kinds of fuel. One is the reduction in flame
temperature due to the dilution in the reactant [14-19], which
slows the chemical reaction rate and leads to the reduction in
laminar flame speed. Another factor may be due to a change in
chemical reaction by adding CO2 or N2 in the reactant. With
the presence of CO2 in the reactant, CO2 will compete with O2
for H atom via the reaction of CO OH CO2 H . As a
consequence, the reaction rate of most significant chain
branching reaction H O2 OH O is suppressed [20, 21].
This suppression effectively reduces the size of the O/H/OH
radical pool [22] and suppresses the formation of NOx. Also
with the presence of CO2 and/or N2 in the fuel, the reactant
becomes less active due to the reduction of oxygen
concentration in the reactant [11]. For industrial lean premixed system the unmixedness is a factor that can affect NOx.
The penetration of the fuel jet into the air stream can, in turn
affect this unmixedness. At a given operating temperature in a
given burner geometry, fuel penetration will increase with
reducing WI due to the increase in mass flow. This could
reduce the level of unmixedness in the burner and produce less
NOx due to the increased fuel jet penetration [23]. The final
factor is associated with the pressure loss across the
combustion system. An increase in mass flow rate of fuel is
required to provide the same heat release as the WI is reduced.
Increased combustor pressure drop will enhance the air and
fuel mixing by possible downstream displacement of the flame
[24]. This will reduce the burnt gas residence time and thermal
NOx from Zeldovich mechanism is reduced.
With respect to auto-ignition and flashback, inert gases in
the fuel tend to prevent the onset of a spontaneous combustion
chain reaction. Combustion dynamics, lean blow out and
flameholding are among the key issues. Flameholding is about
the flame attachment to burner and has direct impact on flame
instability. If flameholding is too weak, the flame may be
blown out, and if it’s too strong, the burner metal temperature
may be higher, which in turn may affect the burner mechanical
integrity. Lafay et al [15] identified that flame instabilities,
resulting from low frequency pressure oscillation, can appear
below a critical equivalence ratio value when diluting premixed flames with CO2. Bulat et al [12] showed the frequency
of pressure pulsation in engine operation was in the same
range. The frequency was found to reduce with decreasing WI
and the amplitude increased. The engine test and single
combustor test of Asti et al [1] showed that the combustion
dynamics behaved differently according to the initial
amplitudes and percentage of N2 dilution in the fuel. Several
critical points associated to the percentage of CO2 or N2 in the
fuel have been identified for different burner geometries and
when more diluent is added beyond this critical point,
combustion dynamic pressure amplitude is increased. This
critical point is highly dependent on burner design, for
example: 25% of N2 [1]; 55% of N2 [2] with frequency less
than 100 Hz (between 45-50% of N2 there is a reasonable
safety margin for instability); WI 27 MJ/Sm3 [6] corresponding
to 22.8% CO2 and 8.2% N2 in the fuel; 20% CO2 for a 10 MW
engine test [13] with a frequency shift towards to the lower
value for the frequency band 40 to 45 Hz. This lower range of
frequency is typical of the lean blow out frequency. Periodic
detachment of the flame front from its flame holder near the
lean blow out limit was observed in the study of Thiruchengode
et al [25] and the low frequency flame pulsation was found to
be dependent on pilot split.
The mechanisms which induce instabilities are linked to
the laminar flame speed, flame thickness and weak
flameholding. The increase of flame thickness can be seen in
2
Copyright © Siemens Industrial Turbomachinery Limited
CO2 diluted flames: for a given equivalence ratio, the higher
the dilution rate, the thicker the flame [15, 16]. A flame diluted
with CO2 showed a weaker anchorage than for the methane-air
flame [26], and flame resistance to extinction was weaker as
extinction strain was reduced as a result of reduced burning
velocity through dilution and radiation effect [17]. Weaker
flameholding was also found by Liu et al [6, 7] with burner
metal temperatures reducing as more diluent was added, and
which also indicates the change in flame position and size [6,
18, 24].
For normal MCV fuel the problems associated with lean
blowout, such as weaker flameholding and higher combustion
dynamic pressure amplitude when diluent concentration
approaches the critical point, can be controlled by pilot fuel
schedule [6, 7, 13] hence the operation of such fuel should not
be a problem, however, the increased mass flow of fuel with
increasing diluent leads to a requirement for higher supply
pressure [6, 7] and minor modifications of the fuel injection
design may be required [6].
Hydrogen enriched MCV fuel
Hydrogen enriched MCV fuels include Syngas, Blast
Furnace Gas (BFG), Coke Oven Gas (COG) and many others.
Syngas mainly contains H2, CO and diluents such as CO2 and
N2 and is produced through gasification of coal and biomass or
steam reforming of natural gas and liquid hydrocarbon. The
advantage of using Syngas is that it is potentially more
efficient than the direct use of the original fuel, and it has a
greater potential for Carbon Capture and Storage (CCS). COG
is a by-product of the coking process that is rich in H2, CO and
CH4 with smaller amounts of CO2 and non-CH4 higher order
hydrocarbons. BFG is a by-product in blast furnaces as a result
of the iron ore reacting with air. The heating value for such
fuel is quite low and it typically contains 25-35% CO, low
level of H2 (1-5%) [3] and diluents of CO2 and N2. The flaring
of COG and BFG is of increasing concern due to the damage to
the environment from the GHG emission from CH4. The
demand for COG and BFG to power an industrial gas turbine
is increasing. In many applications BFG can be blended with
COG to enhance the fuel heating value.
The presence of H2 in this type of fuel can significantly
broaden the flammability region and extend the flammability
limit of MCV fuel, as the chemical reactivity of hydrogen
enriched fuel is increased due to the more reactive species of
H2 in the fuel. The increased reactivity is a challenge for
combustor design to avoid flashback, auto-ignition, high
combustion dynamic pressure, increased flameholding and
acceptable NOx emissions.
The increase in NOx emissions and reduction in CO
emissions is generally reported by burning hydrogen enriched
fuel [1-3]. NOx was increased by almost 25% when about 10%
H2 was added [1]; NOx was increased 50% at 70% H2 and CO
was reduced 50% at 90% H2. The chemical mechanism for the
increase in NOx is the increase emission of NO through the
enhanced NNH reaction path, NNH O NH NO , due to
the higher concentration of super-equilibrium oxygen atom for
H2 rich flame [27, 28] and the increase emission of N2O is
through intermediate route [28].
Combustion dynamic pressure amplitude remained
comparable to that of natural gas when 0-14% H2 was added
[1]. Singh et al [5] found combustion dynamic pressure for H2
enriched mixtures had higher frequencies and amplitudes
compared to that of natural gas. This is because the flame for
such mixtures is compact and the heat release is distributed
over a smaller axial volume due to significant higher flame
speed. This concentrated source causes this type of mixture
flame to excite significantly higher frequency modes compared
to methane. The higher volumetric heat-release rate for H2
makes these modes more unstable than CH4.
For a given air mass flow rate, flashback is primarily
dominated by the H2 percentage in the fuel [29]. The higher
the H2 percentage in the fuel, the more susceptible the system
is to flashback. The diluent of CO2 and N2 can retard the
flashback limit and it is found that CO2 is more effective due to
its restricting the active free radicals and increasing the
recombination reaction rather than chain transfer reaction
[29]. Flashback and flameholding did occur when 90% H2 was
added [2] and it was possible to operate with such high H2
level with carefully control through pilot split and modification
of operation parameter, for example, flame temperature.
Generally flameholding increases as more H2 is added, due to
higher reactivity of the fuel, which causes burner metal
temperature to increase [4].
With careful control, the combustor may operate with high
levels of H2, however, flashback, auto-ignition, high
combustion dynamic pressure, increased flameholding and
acceptable emissions are still key issues. Major modification of
burner design and operation parameters is required.
HCV fuel
Liquefied Petroleum Gas (LPG) and Refinery Off-Gases
(ROG) are among the HCV fuels. LPG is synthetically
produced by refining petroleum and is either primarily C3H8,
or primarily C4H8, most commonly the mixture of both. The
exact proportion between the two is dependant on the season,
in winter there is more C3H8, in summer more C4H8. Propylene
and butylenes are usually also present in small portions. ROG
is the flaring gas from oil production rigs, refineries and
chemical plants and typically composed of higher order
paraffins (C2H6, C3H8, C5H12 and C6H14) and olefins (C2H4,
C3H6, C4H8).
These fuels are significantly more reactive and energetic
than natural gas. Similar to H2 enriched fuel, flashback, autoignition, high combustion dynamic pressure, increased
flameholding and acceptable emissions are major challenges
for using such fuels. Increasing levels of these higher order
hydrocarbons generally leads to increased flame speeds [30]
and higher probability of flashback and flameholding.
3
Copyright © Siemens Industrial Turbomachinery Limited
According to [30] CH4 has the lowest burning velocity, C2H6
has the highest and C3H8 and C4H10 lie in-between. Flames
with the same laminar flame speeds may have significantly
different turbulent flame [31].
Similar to hydrogen enriched fuel, increase in NOx and
reduction in CO is generally reported [1, 3, 8]. NOx was
increased 35% when about 38% C3H8 was added [1].
Generally the laminar flame speed for C2H6, C3H8 and
C4H10 is higher than CH4 [30] for lean premixed mixtures. The
enhanced flame speed for higher order hydrocarbons may lead
to higher turbulent flame speeds, and the risk of flashback is
increased.
Auto-ignition delay time is reduced as auto-ignition
temperature is decreased as more C3H8 and C4H10 added. With
10% C4H10 the auto-ignition temperature was reduced 45%
and with 20% C3H8 there was about 25% reduction in autoignition temperature. There was not much reduction in autoignition temperature with C2H6 added [1]. As the auto-ignition
delay time is shorter as more higher orders of hydrocarbons are
added to the fuel, the risk of flashback is also increased [3].
Generally, an increase in combustion dynamic pressure
amplitude is reported [1] with increasing content of higher
order hydrocarbons in the fuel. When C3H8 content was higher
than 25% the dynamic pressure amplitude started to increase
sharply, while below this value the amplitude was consistent
with natural gas. Interestingly, different frequencies associated
with dynamic pressure may respond differently when the C5H12
content in the fuel increases [8]. Low frequency combustion
dynamic pressure was not sensitive to the adding of C5H12, but
high frequency combustion dynamic pressure showed a small
increase in amplitude. There was not a particular increase in
frequency , but there was a general increase in across the
whole frequency range.
Increased flameholding is another issue [4] with more
C3H8 in the fuel due to higher reactivity and fuel density of the
fuel, which causes the burner metal temperature to increase.
Also, the higher density of fuel resulted in lower fuel jet
velocity, affecting fuel and air mixing.
Increased risks associated to flashback, shorter autoignition time, increased flameholding and higher combustion
dynamics are major challenges for the application of such fuel
to gas turbines.
Current work
Current work covers the fuel flexibility test for the
development and production combustion hardware for Siemens
Industrial Turbomachinery Limited in Lincoln (SITL) SGT400A which is a 13.4 MW engine with compressor pressure
ratio of 16.8:1 for both power generation and mechanical
drive. SGT-400B is also available for 15 MW with compressor
pressure ratio of 18.9:1. The combustion system of the SGT400 has three different configurations to cover different fuel
heating values from WI 25 to 49 MJ/Sm3. The standard
combustion system covers the fuel with WI from 37 to 49
MJ/Sm3, MCV1 configuration covers WI from 25 to 30
MJ/Sm3 and MCV2 configuration covers WI from 30 to 37
MJ/Sm3. The major differences of the two MCV combustion
systems compared to the standard system are that the MCV
fuel burners have larger fuel injection holes and some
modification of fuelling system. With such minor to medium
modifications, fuel supply pressure and injection
characteristics remain similar across the three different
combustion systems.
The Atmospheric Pressure Air Facility (APAF) was used to
investigate the impact of fuel heating value and burner
geometry on lean extinction and ignition limits of a single
combustion system. Engine start testing was also performed at
WI 29 MJ/Sm3 using the MCV1 production hardware and the
results were compared with those from the standard hardware
with natural gas.
The High Pressure Air Facility (HPAF) was used for single
combustion system performance test in terms of emissions,
combustion dynamics and flameholding at true engine
operating conditions, such as combustion air inlet mass flow
rate, pressure, temperature and turbine entry temperature
(TET). Normal MCV fuel was tested using production MCV1
and MCV2 configurations to extend their fuel flexibility
capacity. H2 enriched MCV fuel and HCV fuel were tested
using standard and modified configurations.
SITL DLE COMBUSTOR SYSTEM
SITL DLE combustion technology is applied across the
range of small-scale industrial gas turbines from 5MW to
15MW (SGT-100 to SGT-400) [12, 24, 32]. The operational
experience on these DLE engines has accumulated more than
20 million hours. Figure 1 shows the schematic view of the
SITL DLE combustor.
The SITL DLE combustion system is designed on the
reversal flow technology which consists of three main sections
(Figure 1 left): [i] the pilot body, which houses the pilot fuel
galleries and injectors for both gaseous and liquid fuel; [ii] the
main burner, which houses the main air swirler and main gas
and liquid fuel systems and; [iii] the combustor, which includes
a narrow inlet duct, called the pre-chamber. The combustor
constitutes a double skin wall and is cooled using impingement
jet cooling. This cooling air is then exhausted into the main
combustion zone through effusion cooling holes. The
compressed air is also flushed through the dilution holes
downstream of the main reaction zone to reduce the emissions
like CO. The transition duct located downstream of the
combustor, acts as flow conditioner designed from a circular
combustor exit to a sector of the turbine entry annulus.
4
Copyright © Siemens Industrial Turbomachinery Limited
Main Burner
Double Skin Impingement
Cooled Combustor
Pilot Burner
Testing was conducted in an atmospheric rig in SITL. The
compositions of the fuels and their WI are listed in Table 1.
The two MCV gases were supplied from bottles. During the
test inlet air temperature was kept at a constant 15 C.
Table 1 Fuel WI and its compositions (vol %) for ignition testing
WI
(MJ/Sm3)
Composition
CH4
C2H6
C3H8
C4H10
C5H12
CO2
N2
PreChamber
Radial Swirler
Figure 1 The Dry Low Emission combustor sectional view
IGNITION AND ENGINE START
The lean extinction limit was determined by establishing a
stable flame at each of the air flow rates studied and reducing
the amount of fuel until extinction occurred. At this point the
air and fuel flow rates were recorded. The lean ignition limit
was found by activating the igniter and then gradually
increasing the fuel flow rate from its extinction point until
ignition took place.
The determination of the extinction and ignition limit on
the fuel rich side is a completely opposite way. The rich
extinction limit was determined by establishing a stable flame
at each of the air flow rates studied and increasing the amount
of fuel until extinction occurred. The rich ignition limit was
found by activating the igniter and then gradually decreasing
the fuel flow rate from its extinction point until ignition took
place.
Ignition was deemed to have occurred when the flame
persisted without the presence of a spark. Each test was
repeated three times to ensure consistency and compensate for
noise factors such as fluctuations in the air or fuel flow rates.
28.9
Natural gas
45
41
6
7
2
0
37
7
69.09
0.93
0.17
0.02
0.04
4.26
25.49
94.2
3.2
0.6
0.2
0.1
0.5
1.2
Shown in Figure 2 are the lean extinction and ignition
limits for the three different WIs listed in Table 1 for different
air mass flow rates. All the fuel was supplied through the pilot
burner and with no fuel was supplied from the main burner.
The air fuel ratio (AFR) at the limits was normalised to the
value of the natural gas lean extinction limit at an air mass
flow rate of 0.2 kg/s. The results show that AFRs at lean
extinction and ignition limits decrease as WI decreases and air
mass flow rate increases. The gap of AFR between lean
extinction and ignition limits becomes smaller as WI decreases
and air mass flow rate increases. At higher air mass flow rates
the AFR at lean extinction is close to its ignition limit for all
the three WI fuels. For WI 25 MJ/Sm3, the AFR at lean
extinction is very close to its ignition limit for all air mass flow
rates investigated. This would suggest the AFR for engine
starting should avoid approaching its ignition limit at high air
mass flow, otherwise, engine will suffer from light up problems
due to fuel supply fluctuation. This is particularly important for
low WI fuels.
1.1
Lean extinction limit
0.9
natural gas
0.7
Lean ignition limit
AFR
The radial swirled DLE combustor design is sufficient to
generate high swirl number, greater than 0.6 for a typical gas
turbine combustor, to create a vortex breakdown reverse flow
feature along the axis of the combustor [12, 24, 32]. This flow
feature is called as internal reverse flow zone or central
recirculation zone (CRZ). In this DLE concept, the reverse
flow zone remains attached to the face of the pilot burner,
thereby establishing a firm aerodynamic base for flame
stabilisation. In the wake of the sudden expansion, an external
reverse flow zone is established. The flame is stabilised in the
aerodynamically generated shear layers around the internal
and external reverse flow regions (ERZ).
Fuel is introduced in two stages for both gas and liquid
fuels. Firstly, the main injection, which has a high degree of
premixing and hence low NOx emissions and secondly the
pilot, which is increased steadily as the load demand decreases,
in order to ensure flame stability [12, 24, 32].
25
0.5
WI 29 MJ/Sm3
0.3
WI 25 MJ/Sm
3
0.1
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Air mass flow rate ( kg/s )
Figure 2 Effect of WI on lean extinction and ignition limits for
standard burner
Figure 3 shows the influence of burner design on the lean
extinction and ignition limits at WI 29 MJ/Sm3. The MCV
5
Copyright © Siemens Industrial Turbomachinery Limited
burner was the SGT-400 MCV1 burner designed for WI from
25 to 30 MJ/Sm3. AFR was normalised and the reference point
was the same as shown in Figure 2. Compared with the
standard burner operated at low WI fuel, the lean extinction
and ignition limits for the MCV1 burner are broadened and the
margin between the extinction and ignition is also increased at
low air mass flow rate compared to the standard burner.
1.2
(a)
Fuel demand
1.0
•
reference point
0.8
0.6
0.4
WI 29
NG
0.2
0.0
1.5
0
1.3
MCV
burner
Lean extinction limit
Presssure
AFR
4000
6000
8000
10000
12000
14000
(b)
1.2
Lean ignition limit
1.1
standard
burner
0.9
2000
1.4
1.0
0.8
reference point
0.6
WI 29
NG
0.4
0.7
0.2
0
2000
0.5
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Air mass flow rate ( kg/s )
Figure 3 Impact of burner design on lean extinction and ignition
limits at WI 29 MJ/Sm3
The effect of pilot split on lean and rich ignition limits is
shown in Figure 4 for the MCV1 burner. Compared with 100%
fuel through pilot burner, any reduction in fuel through pilot
burner would result in reduction of AFR for the two limits.
1.1
100%
0.9
Lean ignition limit
80%
AFR
0.7
0.5
100%
0.3
80%
Rich ignition limit
0.1
0.1
0.2
0.3
0.4
0.5
0.6
0.7
4000
6000
8000 10000
Engine speed ( rpm )
12000
14000
Figure 5 Comparison of engine start between engines configured
with standard and MCV1 burners
COMBUSTOR PERFORMANCE
Investigation into the operability of production combustion
systems and its modifications for extended fuel flexibility was
undertaken in the HPAF using a single combustion system. The
HPAF facility has been a standard part of combustion
development process for over 20 years and a high level of
confidence in the transfer from a single combustion system
tested in the rig to the multiple combustion systems in the
engine has been demonstrated and achieved [6]. This rig
includes a mixing plant which allows the blending of various
fuel components to achieve a WI with a target fuel
composition. Schematic of the HPAF and the mixing plant is
shown in Figure 6. This testing considered natural gas blended
with CO2 and/or N2 for normal MCV fuel, CO, H2, CO2 and N2
for H2 enriched MCV fuel and C3H8 for HCV fuel. The fuels
tested had a range of WIs varying from 17.5 to 70 MJ/Sm3.
The WI was measured by a gas chromatograph and the fuel
temperature was kept constant.
Air mass flow rate ( kg/s )
Figure 4 Effect of pilot split on ignition limits, MCV burner 1 at
WI 29 MJ/Sm3
LPG
tank
Figure 5 is the comparison of fuel demand (in KW) and
compressor delivered pressure for a standard engine with
natural gas and the engine with MCV1 burners at WI 29
MJ/Sm3 supplied by bottles as aforementioned. The results
were normalised to the values of natural gas at 10000 rpm
(revolution per minute). The results are comparable and
demonstrate that engine start is not affected by the fuel heating
value and burner geometry change.
H2
trailer
Control & data
acquisition
Liquid LPG
LPG
vaporiser
air
CO
trailer
combustion
air
N2 bottle
banks
CO2 bottle
banks
air
heater
GT
compressor
Alternative fuel
such as Naptha,
unleaded petrol
& biodiesel
Natrual
gas
cooling water
Emissions
measurement
Tank 1 Tank 2 Tank 3
Diesel tank
6
Copyright © Siemens Industrial Turbomachinery Limited
Figure 6 Schematic of High Pressure Air Facility combustion rig
17.5
20
22.5
25
27.5
28.7
30
32.5
35
NG
51.5
49.0
57.3
62.7
60.0
67.9
67.0
66.0
72.7
75.0
77.3
79.0
85.7
83.0
CO2
48.5
40.0
42.7
37.3
22.0
32.1
27.0
20.0
27.3
25.0
22.7
10.0
14.3
12.0
N2
0
11.0
0
0
18.0
0
6.0
14.0
0
0
0
11.0
0
5.0
0
0
20
WI 45
WI 40
WI 35
WI 30
WI 28.7
WI 25
0
20
30
40
50
60
70
80
90
100
110
Load ( % )
Figure 7 NOx emissions against engine load for various WI fuels
for MCV2 burner
10
Table 2 MCV fuels (without H2& CO) tested and their
compositions (vol %)
WI
6.8
0
Shown in Figure 7 are the NOx emissions for the MCV2
combustion system, for different engine load conditions at
various WIs. The lowest WI for full load is 28.7 MJ/Sm3 and
25 MJ/Sm3 for part load. NOx emissions are lower than 20
ppmv (@15% O2) for all conditions tested. Figure 8 shows the
NOx emissions at full load for WI from 28.7 to 45 MJ/Sm3 and
NOx is less than 10 ppmv and not sensitive to changes in WI.
The insensitivity of NOx to WI change may be due to less pilot
split being used when the WI was higher than 40 MJ/Sm3.
Standard pilot split was used for the WI between 28.7 to 35
MJ/Sm3.
NOx ( ppmv @ 15% O2 )
Normal MCV Fuel
The extension of fuel flexibility for the two production
MCV combustion systems, MCV1 originally designed for WI
25 to 30 MJ/Sm3 and MCV2 for WI 30 to 37 MJ/Sm3, was
tested on the HPAF. WI for MCV2 was extended as low as 28.7
MJ/Sm3 (25% CO2 in the fuel) and as high as 45 MJ/Sm3
(100% natural gas) at full load. For the MCV2 system, a WI
lower than 28.7 MJ/Sm3 at full load could not be achieved at
that time due to the fuel mass flow capacity limitation of the
mixing plant. After the MCV2 system was tested, the mixing
plant was updated to allow more diluents to be blended with
natural gas, then the MCV1 combustion system was tested and
the WI tested was as low as 17.5 MJ/Sm3 corresponding to
48.5% CO2 blended with natural gas. Comprehensive
investigation of the change in fuel heating value on emissions,
combustion dynamics, flame holding and fuel pressure drop
across burner was carried out on the MCV1 combustion
system. Details about the fuel compositions and associated WI
for the two MCV combustion systems tested in HPAF are listed
in Table 2.
93.2
100
NOx ( ppmv @ 15 % O2 )
Emissions were recorded, along with other information
such as combustion dynamics, fuel pressure and burner and
combustor metal temperature, enabling the combustor design
to be evaluated. The technique for emissions measurement was
described in [3l] and equipment used was compliant with ISO11042.
The predicted performance of the engine at various load
and ambient conditions was used to define the test condition
parameters such as air pressure, temperature, mass flow rate
and turbine entry temperature (TET). An interesting
consideration in the engine performance when WI is decreased
for the same TET, is the increase in power achieved due to the
increased mass flow of fuel.
40
45
0
25
30
35
40
45
50
3
WI ( MJ/Sm )
Figure 8 Effect of WI on NOx at full load for MCV2 burner
Presented in Figure 9 are the NOx and CO emissions for
the MCV1 combustion system for WI from 17.5 to 35 MJ/Sm3
at different engine load conditions. When engine load is higher
than 30%, NOx is lower than 15 ppmv. CO is un-measurable
for the load from 50% to full power for all WIs investigated.
7
Copyright © Siemens Industrial Turbomachinery Limited
WI 17.5
WI 20
WI 22.5
WI 25
WI 27.5
WI 32.5
WI 35
10
20
0
15
30
40
50
60
70
80
90
Load ( % )
100 110
WI 17.5
WI 20
WI 22.5
WI 25
WI 27.5
WI 32.5
WI 35
10
(b)
0
0
10
20
20
25
30
35
2.0
Comb. dynamics pressure
0
10
30
40
50
60
70
80
90
Load ( % )
Figure 9: NOx and CO emissions vs. engine load for various WI
fuels for MCV1 burner
The impact of change in fuel heating value on combustor
performance was investigated and the results are shown in
Figure 10 in terms of NOx emissions, combustion dynamics
pressure, flameholding in term of burner metal temperature
and main fuel pressure drop across burner at full load
condition.
1.4
1.2
1.0
0.8
0.6
0.4
1.10
2.0
15
20
25
30
35
40
(c)
1.05
1.00
0.95
0.90
15
100 110
40
(b)
1.8
1.6
Burner metal temperature
0
CO ( ppmv @ 15% O2 )
NOx ( ppmv @ 15% O2 )
15
(a)
Main fuel pressure drop
NOx ( ppmv @ 15% O2 )
(a)
20
25
30
35
40
(d)
1.8
1.6
1.4
1.2
Reference point
1.0
0.8
0.6
0.4
15
20
25
30
35
40
WI ( MJ/Sm 3 )
Figure 10 Impact of WI on NOx, combustion dynamics, burner
metal temperature and main fuel pressure drop at full load for
MCV1 burner
During the testing, standard engine pilot schedule which is
the pilot schedule for production engines was used. A reduction
in NOx emission is obvious as WI decreases. Combustion
dynamics pressure, burner metal temperature and main fuel
pressure drop were normalised to those at WI 25 MJ/Sm3. It
seems that combustion dynamics pressure slightly increases
with WI increase for WI from 25 to 35 MJ/Sm3 and decreases
with WI decrease for WI from 17.6 to 22.5 MJ/Sm3. There is a
step change for combustion dynamic pressure amplitude when
WI change from 25 to 22.5 MJ/Sm3, however, the actual
amplitude still remains at low level. The amplitude changes
are 30% increase at WI 20 MJ/Sm3, 40% increase at WI 22.5
MJ/Sm3 and 20% increase for WI 35 MJ/Sm3. Burner metal
temperature is not sensitive to WI change and the scatters are
within ± 2% for WI change from 17.5 to 35 MJ/Sm3 which
indicates that the flameholding characteristics do not change
8
Copyright © Siemens Industrial Turbomachinery Limited
1.3
15
1.2
NOx
1.1
1
dynamics
0.9
Nominal pilot
10
dynamics
0.9
WI 25 MJ/Sm , full load
1.6
dynamics
1.3
1
4
8
12
16
N 2 in fuel ( vol % )
Figure 12 Influence of N2 concentration on NOx and combustion
dynamics at WI 25 MJ/Sm3 for MCV1 burner
Figure 13 is a summary figure of the impact of change in
fuel heating value on NOx emissions at full load, for MCV1
and MCV2 combustion systems. The results shown in this
figure are for fuels using only CO2 as the diluent in natural gas
to provide more fundamental information. For the MCV2
combustion system NOx is not so sensitive to CO2 addition
because less pilot was used when less CO2 was added. For the
MCV1 combustion system, NOx reduction with more CO2
addition is clearly seen.
Comb. dynamics pressure
1.9
0.8
0
WI 35 MJ/Sm
NOx
Comb. dynamic pressure
1
0
3
NOx ( ppmv @ 15% O2 )
1.1
3
0.8
2.2
0
45
41.3
37.8
34.6
31.5
28.7
26.1
23.6
21.2
19
40
45
3
WI ( MJ/Sm )
MCV burner 2
MCV burner 1
0
0
5
10
15
20
25
30
35
CO2 in fuel ( vol % )
Nominal pilot
0
1.2
NOx
10
15
1.3
NOx ( ppmv @ 15% O2 )
NOx ( ppmv @ 15% O2 )
Unit A Tamb = 31 to 37 °C
Unit B Tamb = 36 to 40 °C
Comb. Dynamics pressure
WI 27.5 MJ/Sm3
25 MJ/Sm3. Although there are some variations of NOx and
combustion dynamic pressure with N2 content change in the
fuel, the actual amplitude change is small. The impact of fuel
composition on emissions and combustion dynamics is not
significant for the N2 concentration variation from 0% to 15%.
NOx ( ppmv @ 15% O2 )
significantly with fuel heating value change. The most
noticeable change is the main fuel pressure drop across the
burner with an 80% higher when WI decreases from 25 to 17.5
MJ/Sm3 and 60% reduction in pressure drop when WI
increases from 25 to 35 MJ/Sm3.
The effect of pilot split on NOx emissions and combustion
dynamics was investigated at WI 27.5 MJ/Sm3 and 35 MJ/Sm3
and results are shown in Figure 11. The impact of pilot split on
NOx emissions and combustion dynamics is commonly seen
for industrial gas turbines at WI 27.5 MJ/Sm3: NOx increases
and combustion dynamic pressure decreases with increases in
pilot split. However, for WI 35 MJ/Sm3, NOx behaves as
normal, the increase in combustion dynamic pressure as pilot
split increases is not normally seen. The change in behaviour
of combustion dynamics pressure with pilot split at higher WI
may be due to the reduced fuel pressure drop, and as a
consequence, the fuel injection characteristic is changed.
Shown in Figure 11 are also the comparisons of NOx
emissions between HPAF and two field engines at the same WI
of 27.5 MJ/Sm3. The NOx emissions are comparable between
rig and engines and are below 15 ppmv.
0.7
Pilot split ( % )
Figure 11 Effect of pilot split on NOx and combustion dynamics
at WI 27.5 and 35 MJ/Sm3 for MCV1 burner
The influence of fuel composition, CO2 vs N2, at the same
WI on NOx emissions and combustion dynamics pressure were
also investigated and the results are shown in Figure 12 for WI
Figure 13 Influence of CO2 addition into natural gas on NOx at
full load for two MCV burners
H2 enriched MCV Fuel
For combustion testing with H2 enriched MCV fuel in the
HPAF, fuels at a fixed WI of 29 MJ/Sm3 were tested and the
variations in compositions are listed in Table 3. The variation
in H2 and CO was from 0 to 35.1% and 0 to 40%, respectively.
9
Copyright © Siemens Industrial Turbomachinery Limited
Actual WI’s tested varied from 28.8 to 29.6 MJ/Sm3 at ISO
fuel temperature. The impact of fuel compositions on
combustion performance was identified by such variations in
H2 and CO concentration and keeping the WI at the same
value.
Table 3 H2 enriched MCV fuels and their compositions, vol %,
tested at WI 29 MJ/Sm3
NG
73.5
66.1
59.8
54.0
49.0
65.9
60.3
57.3
53.6
67.5
71.9
63.9
62.3
58.8
H2
0
10.1
20.2
29.3
35.1
7.1
12.2
15.7
18.9
0
0
0
0
0
CO
0
0
0
0
0
6.6
12.4
15.2
19.4
10.0
10.8
21.6
29.9
40.0
N2
13.3
11.8
9.9
8.3
8.2
10.1
7.4
6.0
3.9
22.4
0
14.5
3.8
0.4
CO2
13.2
12.0
10.1
8.4
7.8
10.3
7.8
5.8
4.2
0
17.3
0.1
4.1
0.7
WI
29.2
28.8
29.1
29.4
28.8
29.0
29.0
29.3
29.3
29.3
29.6
29.3
29.3
29.6
Figure 14 is the effect of H2 concentration in the fuel on
NOx emissions for the standard and modified combustion
systems. The difference of the modified combustion system to
the standard one is the modification of the fuel injection holes
to allow the main fuel pressure drop across the burner to
remain at a similar level to the standard burner when using
natural gas fuel. The increase in NOx emissions with
increasing levels of H2 in the fuel is obvious, however, with
35% H2 in the fuel, NOx emissions are still below 20 ppmv.
NOx ( ppmv @ 15% O2 )
20
modified burner
standard burner
0
0
5
10
15
20
25
30
35
40
H2 ( Vol % )
Figure 14 Impact of H2 on NOx for standard and modified
burners
Further tests were focused on the modified combustion
system and the results are shown in Figure 15. Figure 15(a),
(b) and (c) (3 figures on the left hand side) show the impact of
H2 composition in the fuel on NOx emissions, combustion
dynamics and flameholding and Figure 15(d), (e) and (f) (3
figures on the right hand side) show the effect of CO
composition in the fuel. Apart from the impact from pure H2 or
CO on combustor performance, in each figure the combined
effect of H2 and CO is also shown. The origin in fuel
composition indicates the normal MCV fuel and acts as a
reference point to investigate the effect of addition of H2 and/or
CO. Combustion dynamic pressure and burner metal
temperatures were normalised at this reference point.
As shown in Figure 15(a) and (d), the effect of H2 on NOx
is more significant than that of CO as: (1) the gradient of NOx
changing with CO is quite small compared with that of H2, (2)
with the existence of CO in the fuel, any addition of H2
increases NOx dramatically. However, with the existence of H2,
an increase in NOx can be found with addition of CO in the
fuel, particularly with higher percentages of CO in the fuel,
this can be explained as the flame temperature increases when
more CO is added in the fuel as shown in the Figure 16(c) and
(d). Similar phenomenon can be found that the adverse impact
of H2 on combustion dynamics and burner metal temperature is
more significant than that of CO as shown in Figure 15(b) and
(e), Figure 15(c) and (f). Interestingly, when mixtures with up
to 20% of either H2 or CO were tested, the increase of burner
metal temperature was less than 10%, however, with the
existence of H2, the addition of CO can increase the burner
metal temperature significantly and vice versa.
To provide more fundamental understanding, equivalence
ratio, flame speed and adiabatic flame temperature were
calculated by the GRI 3.0 full chemical mechanism [33]. The
calculated results are shown in Figure 16. Equivalence ratios
were normalised at the reference point aforementioned.
Generally equivalence ratio decreases as more H2 and/or CO is
added although some scatter can be seen, but the overall
change in equivalence ratio is less than 5% when 35% H2 or
40% CO is added. Flame speed increases as more H2 and/or
CO is added, and the increase with H2 is more than that with
CO. With the existence either H2 or CO in the fuel and
addition the other in the fuel, flame speed increases sharply as
shown in Figure 16(b) and (e). The change in flame
temperature is within +/- 10 C as up to 35% H2 is added in the
fuel if the fuel does not consist of CO, however, with the
existence of H2, the addition of CO in the fuel increases flame
temperature significantly as shown in Figure 16(c). Without H2
in the fuel, flame temperature increases about 30 C as 40%
CO is added in. Hence CO has significant impact on flame
temperature change.
As aforementioned without CO, small to mediate
percentage addition of H2 into the CH4-air flame has little
effect on flame temperature, with 35% H2 added, flame
temperature variation is less than 10 C, this is also confirmed
by [34, 35] as only 15 C increase can be seen as 40% H2 is
added in. However, if 100% H2 replaced CH4, flame
temperature will be 120 C higher. Without H2, CO has a
slightly larger effect on flame temperature, as about 40 C
10
Copyright © Siemens Industrial Turbomachinery Limited
NOx ( ppmv @ 15% O2 )
NOx ( ppmv @ 15% O2 )
increase in flame temperature with 40% CO in, this is also
confirmed by [35] as 12 C increase with 20% CO in and 27 C
increase for 50% added in, and about 120 C increase in CH4
replaced by CO. When CO is low (8%) heat release is through
the reaction path of:
(a)
25
CO % = H2 %
0% CO
0
5
10
15
20
25
30
35
CH 2O
H
H2
OH
H
H 2O
H2
CO
(d)
25
H2 % = CO %
0% H2
40
8
(b)
CO % = H2 %
7
0
Comb. dynamics pressure
9
0% CO
6
5
4
3
2
1
0
5
10
15
20
25
30
35
1.6
(c)
1.5
CO % = H2 %
1.4
0% CO
1.3
1.2
1.1
1
0
5
10
15
20
25
30
35
5
10
15
20
25
30
35
40
45
9
(e)
8
H2 % = CO %
7
6
5
4
0% H2
3
2
1
0
40
Burner metal temperature
Comb. dynamics pressure
H
O CH 3
0
0
Burner metal temperature
O CH 3
5
10
15
20
25
30
35
40
45
1.6
(f)
1.5
H2 % = CO %
1.4
1.3
0% H2
1.2
1.1
1
40
0
H2 ( vol % )
5
10
15
20
25
30
35
40
45
CO ( vol % )
Figure 15 Impact of H2 and CO on NOx, combustion dynamics and burner metal temperature
11
Copyright © Siemens Industrial Turbomachinery Limited
1.03
1.02
1.01
1.00
0.99
0.98
0.97
0.96
0.95
0%CO
CO% = H2%
10
15
20
H
25
30
40
35
( vol % )
(b)
CO % = H 2 %
0% CO
0
5
10
15
20
25
30
40
(c)
CO % = H2 %
30
20
10
0
0% CO
-10
-20
0
5
10
15
20
25
30
35
H2% = CO%
0% H2
5
10
15
26
25
24
23
22
21
20
19
18
40
35
50
(d)
0
Flame speed ( cm/s )
5
26
25
24
23
22
21
20
19
18
20
25
30
35
40
45
(e)
H2 % = CO %
0 % H2
0
Flame Temp. change ( C )
Flame speed ( cm/s )
0
Flame Temp. change ( C )
1.03
1.02
1.01
1.00
0.99
0.98
0.97
0.96
0.95
(a)
5
10
15
20
25
30
35
40
45
CO ( vol % )
50
40
H2 % = CO %
(f)
0 % H2
30
20
10
0
-10
-20
0
40
5
10
15
20
25
30
35
40
45
CO ( vol % )
H2 ( vol % )
Figure 16 Influence of H2 and CO on equivalence ratio, flame speed and flame temperature
when CO is increased, the following reactions become active
and contribute significant amount of heat release [35] and play
a dominant role affecting the laminar burning velocity:
OH
CO
H
HO2
H
OH
CO2
OH
Flame position is also altered with CO in the fuel, with
more CO in the fuel the flame is closer to flame holder. Both
H2 and CO increase laminar flame speed, H2 has a larger
impact on this compared to CO. Higher laminar flame speed
would result in higher turbulent flame speed. Venkateswaran et
al [36] pointed out that even at same laminar flame speed the
turbulent flame speed increases with the increase level of H2 in
the fuel. The higher laminar and turbulent flame speed is one
of the major factors causes flashback in H2 enriched fuel.
HCV fuel
Shown in Figure 17 is the effect of adding C3H8 into
natural gas on combustor performance such as NOx emissions,
flameholding and combustion dynamics at full load conditions.
Metal temperature and combustion dynamic pressure
amplitude were normalised at standard burner operated with
natural gas. The standard burner and two modifications are
shown. Modification 1 is with minor modification to the fuel
injection holes and modification 2 is with some geometry
modification in the swirler vane passage.
An increase in NOx emissions is noticeable as more C3H8
is added in the fuel, with modification 1 having similar levels
of NOx to the standard burner, and modification 2 having less
NOx. Even with 100% C3H8, NOx is still less than 25 ppmv.
Similar to NOx behaviour, burner metal temperature increases
with more addition of C3H8 in the fuel. Modification 1 has a
similar metal temperature to the standard burner throughout
and modification 2 has lower metal temperature when C3H8 is
less than 65% and has comparable level of temperature with
the standard burner once C3H8 is higher than this level. For
combustion dynamics, modification 1 has much lower dynamic
pressure amplitude throughout. Modification 2 has slightly
higher amplitude when operated with natural gas and the
amplitude decreases with more C3H8 added when C3H8 is less
than 55%. Once this level is approached, the amplitude
increases with more C3H8 in the fuel. The results indicate that
for HCV fuel, major changes to the burner fuel injection and
air passage geometry may be needed.
12
Copyright © Siemens Industrial Turbomachinery Limited
C3H8 ( vol %)
0
NOx ( ppmv @ 15% O2 )
28
16.1
34.7
54.9
76.9
100
(a)
25
23
18
13
Standard burner
Modification 1
Modification 2
8
03
40
45
50
55
60
65
70
0
16.1
34.7
54.9
76.9
100
75
Burner metal temperature
1.6
1.5
(b)
1.4
1.3
1.2
1.1
Standard burner
Modification 1
Modification 2
1.0
0.9
40
45
50
0
16.1
55
8
60
65
70
34.7
54.9
76.9
100
3
75
Combustion dynamics pressure
2.5
(c)
2.0
Standard burner
1.5
Modification 2
1.0
0.5
Modification 1
0.0
40
45
50
55
60
65
70
75
3
WI ( MJ/Sm )
Figure 17 Impact of C3H8 addition in natural gas on NOx,
burner metal temperature and combustion dynamics for three
different burner designs
CONCLUSIONS
The impact of change in fuel compositions and heating
value on combustor performance has been reviewed. Ignition
and combustor performance were tested in an atmospheric and
a high pressure rig, separately.
The results of ignition testing show that the air fuel ratio
at lean ignition/extinction limits decreases and the margin
between the two limits tends to be smaller as fuel heat value
decreases. Engine start testing performed with a lower heating
value fuel shows that engine start is not significantly changed.
For the normal MCV fuel (without H2) testing, a benefit in
NOx reduction can be seen for the lower WI fuels and there are
no adverse impacts on combustor performance except for the
requirement of higher fuel supply pressure, however, this can
be easily resolved by minor modification through the fuel
injection design. The problems associated with lean blowout,
weaker flameholding and higher combustion dynamic pressure
amplitude can be controlled by pilot fuel schedule.
Testing results for the H2/CO enriched MCV fuels and
higher WI fuels show that NOx, combustion dynamics and
flash back have been adversely affected and major change in
burner design is required. For the H2/CO enriched fuel, it is
found that H2 dominates the adverse impact on combustor
performance. The chemical kinetic study shows that H2 has
significant effect on flame speed change and CO has
significant effect on flame temperature change.
For the purpose of engine build standards for the fuel with
lower H2 content (<5%) the use of three combustor variations
will be considered:
(1) Standard hardware: covering WI range 37 - 49 MJ/Sm3;
(2) MCV2 hardware: covering WI range 28 - 40 MJ/Sm3;
(3) MCV1 hardware: covering WI range 20 - 30 MJ/Sm3;
But with facility, if required, of starting on a richer gas,
including natural gas may be considered. This arrangement
offers the best combination of supply pressure requirements
and acceptable operational parameters such as NOx,
combustion dynamics, burner pressure drop and burner metal
temperature.
REFERENCES
[1] Asti, A., Stewart, J. F., Forte, A., Yilmaz, E. and D’Ercole,
M., Enlarging the fuel flexibility boundaries: theoretical and
experimental application to a new heavy-duty gas turbine
(MS5002E), GT2008-50773, Proc. ASME Turbo Expo 2008.
[2] Carrera, A. M., Andersson, M., and Nasvall, H.,
experimental investigation of the 4th generation DLE burner
concept: emissions and fuel flexibility performance at
atmospheric conditions, GT2011-46387, Proc. ASME Turbo
Expo 2011.
[3] Campbell, A., Goldmeer, J., Healy, T., Washman, R.,
Moliere, M., and Citeno, J., heavy duty gas turbines fuel
flexibility, GT2008-51368, Proc. ASME Turbo Expo 2008.
[4] Blouch, J., Li, H., Mueller, M., Hook, R., fuel flexibility in
LM2500 and LM6000 dry low emission engines, ASME,
GT2011-45387, Proc. ASME Turbo Expo 2011.
[5] Singh, K., Varatharajan, B., Yilmaz, E., Han, F., Kim, K.,
effect of hydrogen combustion on the combustion dynamics
of a natural gas combustor, GT2008-51343, Proc. ASME
Turbo Expo 2008,
[6] Liu, K., Alexander, V., Sanderson, V., Bulat, G., ASME,
extension of fuel flexibility in the Siemens dry low emissions
SGT-300-1S to cover a Wobbe Index range of 15 to 49
MJ/Sm3, GT2012-68838, Proc. ASME Turbo Expo 2012.
[7] Liu, K., Sanderson, V., the influence of changes in fuel
calorific value to combustion performance for Siemens SGT300 dry low emission combustion system, submitted to Fuel.
13
Copyright © Siemens Industrial Turbomachinery Limited
[8] Andersson, M., Larsson, A., Carrera, A., pentane rich fuels
for standard Siemens DLE gas turbines, GT2011-46099,
Proc. ASME Turbo Expo 2011.
[9] Rokke PE, Hustad JE. Exhaust gas recirculation in gas
turbines for reduction of CO2 emissions; combustion testing
with focus on stability and emissions. Int. J. of
Thermodynamics 2005;8:167-173.
[10] Valler K, Wopera A, Palotas AB, Whitty KJ. NOx
formation by synthesis gas – natural gas co-firing. Proc.
European Combustion Meeting 2007.
[11] Elkady AM, Brand AR, Vandervort CL, Evulet AT. Exhaust
gas recirculation performance in dry low emissions
combustors. GT2011-46482, Proc. ASME Turbo Expo 2011.
[12] Bulat G, Liu K, Brickwood G, Sanderson V, Igoe B.
Intelligent operation of Siemens (SGT-300) DLE gas turbine
combustion system over an extended fuel range with low
emissions. GT2011-46103, Proc. ASME Turbo Expo 2011.
[13] Cocchi S, Provenzale M, Ceccherini G. Fuel flexibility test
campaign on a 10 MW class gas turbine equiped with a drylow-NOx combustion system. GT2007-27154, Proc. ASME
Turbo Expo 2007.
[14] Ren JY, Egolfopoulos FN, Mak H, Tsotsis TT. NOx
emission control of lean methane-air combustion with
addition of methane reforming products. Combust. Sci. and
Tech. 2002;174:181-205.
[15] Lafay, Y., Renou, B., Cabot, G. and Boukhalfa, M.,
Experimental determination of laminar flame thickness for
CO2 and H2 diluted methane/air flames, Proc. European
Combustion Meeting 2007.
[16] Cohe, C., Chauveau, C. and Gokalp, I., addition effect in
high pressure CH4-air turbulent premixed flames, Proc.
Combust. Inst. 32 (2009) 1803-1810.
[17] Ren, J. Y., Qin, W., Egolfopoulos, F. N., Mak, H. and
Tsotsis, T. T., methane reforming and its potential effect on
the efficiency and pollutant emissions of lean methane-air
combustion, Chemical Engineering Science 2001, 56, 15411549.
[18] Qin, Q., Egolfopoulos, F. N. and Tsotsis, T. T., fundamental
and environmental aspects of landfill gas utilization for
power generation, Chemical Engineering Science 2001, 82,
157-172.
[19] Kishore, V. R., Duan, N., Ravi, M. R. and Ray, A.,
measurement of adiabatic burning velocity in natural gas
like mixtures, Experimental Thermal and Fluid Science, 33
(2008) 10-16.
[20] Liu, F., Guo, H. and Smallwood, G., the chemical effect of
CO2 replacement of N2 in air on the burning velocity of
CH4 and H2 premixed flames, Combustion and Flame 133
(2003) 495-497.
[21] Glarborg, P. and Bentzen, L. L. B., chemical effects of a
high CO2 concentration in oxy-fuel combustion of methane,
Energy and Fuel 22 (2008) 291-296.
[22] Fackler, K. B., Karalus, M. F., Novosselov, I. V., Kramlich,
J. C. and Malte, P. C., experimental and numerical study of
NOx formation from the lean premixed combustion of CH4
mixed with CO2 and N2, GT2011-45090, Proc. ASME
Turbo Expo 2011.
[23] Dobbeling, K., Meeuwissen, T., Zajadatz, M. and Flohr,
P., fuel flexibility of the Alstom GT132E medium sized gas
turbine, GT2008-50950, Proc. ASME Turbo Expo 2008,
[24] Liu, K., Wood, J. P., Buchanan, E., Martin, P., and
Sanderson, V. E., Biodiesel as an alternative fuel in
Siemens dry low emissions combustors: atmospheric and
high pressure rig testing, Journal of Engineering for Gas
Turbine and Power, 132 (2010) 011501-1.
[25] Thiruchengode, M., Nair, S., Prakash, S., Scarborough,
D., Neumeier, Y., Lieuwen, T., Jagoda, J., Seitzman, J. and
Zinn, B., an active control system for LBO margin
reduction in turbine engines, Proc AIAA (2003) 41st
Aerospace Science Meeting and Exhibit, AIAA 2003-1008.
[26] Lafay, Y., Cabot, G. and Boukhalfa, A., experimental study
of biogas combustion using a gas turbine configuration, 13 th
Int Symp on Application of Laser Techniques to fluid
Mechanics, 2006
[27] Lin, Y. C., Matuszewski, M., Daniele, S., Jansohn, P.,
Boulouchos, K., NOx emission for combustion systems
relevant to zero emissions power concepts, Proceedings of
the European Combustion, 2011.
[28] Guo, H., Smallwood, G. J., Liu, F., Ju., Y., Güulder, Ö.,
the effect of hydrogen addition on flammability limit and
NOx emission in ultra-lean conterflow CH4/air premixed
flame, Proceedings of the combustion institute (30), 303311, 2005.
[29] Dam, B., Corona, G., Hayder, M., Choudhuri, A., Effects
of syngas composition on combustion induced vortex
breakdown (CIVB) flashback in a swirl stabilized
combustor, Fuel 90 (2011) 3274-3284.
[30] Dirrenberger, P., Glaude, P. A., Gall, H. L., Bounaceur, R.,
Herbinet, O., Battin-Leclerc, F., Konnov, A. A., Laminar
flame velocity of components of natural gas, GT201146312, Proc. ASME Turbo Expo 2011,
[31] Kido, H., Nakahara, M., Nakashima, K., Hashimoto, J.,
influence of local flame displacement velocity on turbulent
burning velocity. Proceeding of the combustion institute
(29), pp. 1855-1861, 2002.
[32] Kowkabie, M., Noden, R. and De Pietro, S., The
development of a dry low NOx combustion system for the
EGT Typhoon, 97-GT-60, Proc. ASME Turbo Expo 1997.
[33] GRI-MECH 3.0, http://www.me.berkeley.edu/gri_mech/
[34] Hu, E., Huang, Z., Zheng, J., Li, Q., He, J., Numerical
study on laminar burning velocity and NO formation of
premixed methane-hydrogen-air flames, International
journal of hydrogen energy 34(2009) pp. 6545-7557.
[35]Cheng, T. S., Chang, Y. C., Chao, Y. C ., Chen, G. B., Li,
Y. H., Wu, C. Y., An experimental and numerical study on
characteristics of laminar premixed H2/CO/CH4/air flames,
International journal of hydrogen energy, 36 (2011), pp.
13207-13217.
14
Copyright © Siemens Industrial Turbomachinery Limited
[36] Venkateswaran, P., Marshall, A. D., Noble, D. R.,
Seitzman, J. M., Lieuwen, T., Turbulent consumption speed
scaling of H2/CO blends, GT2011-45401, Proc. ASME
Turbo Expo 2011.
15
Copyright © Siemens Industrial Turbomachinery Limited