Module 3
Module 3
Module - 3
ENGINE PERFORMANCE
Syllabus:
Design & off - design Performance. Surge margin requirements, surge margin stack up.
Transient performance. Qualitative characteristics quantities. Transient working lines.
Starting process & Wind milling of Engines. Thrust engine start envelope. Starting
torque and speed requirements Calculations for design and off-design performance from
given test data – (case study for a single shaft Jet Engine). Engine performance
monitoring.
For initial definition work, the operating condition where an engine will spend most
time has been traditionally chosen as the engine design point. For an industrial unit this
would normally be ISO base load, or for an aero-engine cruise at altitude on an ISA day.
Alternatively some important high power condition may be chosen. Either way, at the
design point the engine configuration, component design and cycle parameters are
optimized. The method used is the design point performance calculation. Each time input
parameters are changed and this calculation procedure is repeated, the resulting change to
the engine design requires a different engine geometry, at the fixed operating condition.
With the engine geometry fixed by the design point calculation, the performance at
other key operating conditions can be evaluated, such as ISA SLS takeoff for an aero-
engine. In this instance the calculation procedure is the off design performance calculation.
Here geometry is fixed and operating conditions are changing.
A number of key parameters that define overall engine performance are utilized to
assess the suitability of a given engine design to the application, or compare several
possible engine designs.
Output power or net thrust: It is evaluated via the overall cycle calculation. The
term effective or equivalent power is used for turboprops and turboshafts, where
any residual thrust in the exhaust is converted to power.
Exhaust gas power: For a turboshaft engine core this is the output power that
would be produced by a power turbine of 100% efficiency. It is of interest when
engine cores are tested or supplied without their free power turbine.
Specific power or thrust: This is the amount of output power or thrust per unit of
mass flow entering the engine. It provides a good first-order indication of the engine
weight, frontal area and volume.
Specific fuel consumption: This is the mass of fuel burnt per unit time per unit of
output power or thrust. It is important to minimize SFC for applications where the
weight and/or cost of the fuel is significant versus the penalties of doing so.
Thermal efficiency for shaft power engines: This is the engine power output
divided by the rate of fuel energy input, usually expressed as a percentage.
Heat rate for shaft power cycles: Heat rate is a parameter used only in the power
generation industry, and is the rate of fuel energy input divided by the useful power
output. Hence it is comparable to SFC but is independent of fuel calorific value.
Exhaust temperature: For engines used in combined cycle for industrial power
generation, high exhaust temperature is vital in maximizing overall efficiency. For
combined heat and power the optimum value depends on the relative demand of
heat versus power.
Exhaust mass flow: For engines used in combined cycle or combined heat and
power applications the exhaust mass flow is important in indicating the heat
available in the gas turbine exhaust, and hence the overall plant thermal efficiency.
Thermal efficiency: Thermal efficiency for aircraft thrust engines is defined as the
rate of addition of kinetic energy to the air divided by the rate of fuel energy
supplied, usually expressed as a percentage.
This immediately show that the changes in pressure and temperature that the
working fluid experiences strongly affect the engine performance parameters. The degree
of change of pressure and temperature are reflected via the following cycle design
parameters. Changes in component performance parameters have a secondary effect on the
optimum values of engine cycle design parameters.
Cyclic life (low cycle fatigue) as a function of material type and metal temperatures
Disc and blade tensile stress as a function of rim speed or AN2
Casing rupture as a function of compressor delivery pressure
Choke or stall flutter as a function of fan or compressor referred speed
Vibration (high cycle fatigue) of rotating components as a function of rotational
speed and excitation parameters such as upstream blade numbers and pressure
levels
Shaft critical speeds
Life parameters
The TBO is governed mainly by creep and oxidation life, while cyclic life is dictated
by thermal stress levels. Typical life requirements for the major gas turbine applications
are as follows.
Fuel type
Kerosene is the standard aviation fuel while marine engines burn diesel and most
industrial applications use natural gas. The highly distilled forms of diesel used make little
difference to performance compared with kerosene, but natural gas gives performance
improvements because of the higher resulting specific heat of the combustion products.
This is the internationally accepted SAE definition, though others have been used
in previous years.
The minimum steady state surge margin required will depend upon the engine
configuration and application requirements.
The power or thrust level at which the minimum surge margin occurs will also vary.
During the engine concept design phase the steady state model can be used to
predict surge margin ‘pinch points’.
For each engine application the worst operating conditions and transient
requirements vary and it is not possible to cover all combinations here. Generally
surge margin stack ups are conducted at these key operating conditions.
Once the required margins there have been achieved, the values resulting at some
other single operating condition may be compared for different engine types. This
is usually at ISA sea level static, maximum rating.
The required surge margin is evaluated from a surge margin stack up where a range
of issues, including transient working line excursions, must be addressed at the
worst operating condition.
These issues are listed below, together with typical values for a civil aero-engine
HP compressor at ISA SLS and rated power or thrust.
The required surge margin is calculated by adding the arithmetic sum of the
systematic deviances, to the root sum square of the random variances, as per.
For example:
Typical surge margin requirements at ISA SLS (Sea-Level Static condition), maximum
rating
The required surge margin at ISA SLS and maximum rating varies greatly, being
dependent upon accel and decel times required, engine configuration, whether
centrifugal or axial compressors are applied, whether bleed valves or VSVs
(variable stator vanes) are employed at part load, etc.
The levels listed below are a first-order guide:
For a fan, the biggest single contributor to the requirement is inlet distortion where
up to 5% surge margin must be allowed.
During transient operation there are significant net heat fluxes between the working
fluid and the engine metal, unlike for steady state operation where there is negligible
net heat transfer.
For example, due to an accel from idle to full power or thrust the engine carcass
must soak to a new higher steady state operating temperature, which absorbs
typically 30% of the excess fuel energy.
This net heat transfer from the working fluid to the metal is termed heat soakage
and has a significant effect on engine performance.
Where heat exchangers are employed the impact of heat soakage during transient
operation can be dramatic, due to the large thermal inertias. Of conventional engine
components the combustor has the largest effect, due to its large surface area,
thermal mass and temperature changes.
Volume packing:
During steady state operation the mass flow entering a given volume, such as a duct,
is equal to that leaving.
This is no longer true under transient operation as the pressure, temperature and
hence density of the fluid change with time.
This is known as volume packing and can have a notable impact upon an engine’s
transient performance, especially for the largest volumes such as ducts and heat
exchangers.
For most other engines the combustor has the largest volume and is the primary
concern, though other components must also be considered for fast transients.
During an accel the thermal growth of the compressor or turbine discs is slower
than the pressure and thermal growth of casings, causing blade tip clearances to be
temporarily increased.
The converse is true during a decel which can lead to rubs. This change in
compressor geometry affects its map, the main issue being lower surge lines. There
is also a second-order reduction in flow and efficiency at a speed.
Where a single map is used to model a multi-stage component such as an axial flow
compressor, net heat transfer will have a second-order effect upon the map during
a transient.
This is due to its effect upon gas temperature through the component and hence
stage matching, as it changes the referred speed and hence flow capabilities of the
rear stages.
Combustion delay:
There is a time delay between the fuel leaving the injector and actually burning to
release heat within the combustor.
For steady state performance this is irrelevant, however for transient performance
it should be considered.
Figure 8.1 shows the typical response versus time of gas generator performance
parameters to a slam (step) increase, or decrease in power lever angle (PLA).
In the simplest control system each level of PLA corresponds to a given speed
demand. Following the step increase in PLA to initiate an accel (acceleration) the
speed demand is far higher than the actual engine speed.
The control system responds by increasing fuel flow at a defined limiting rate until
the demanded speed is achieved. The over-fueling is typically between 20 and 100%
of the steady state value for the current speed.
Owing to the additional fuel flow the turbine produces more power than the
compressor requires.
For a decel (deceleration) the opposite occurs.
The compressor working lines differ from steady state operation, as shown in Figs
8.2 and 8.3.
The high temperatures associated with a slam accel are of such short duration that
they do not affect the creep or oxidation life. However as transient times are reduced
so is the cyclic life, due to the severe thermal stresses induced.
One cycle is usually defined as a start, a holding period at idle, an accel to full thrust
or power and eventually the corresponding engine shut down. Though the main
damage is that from starting, fast accel times also contribute.
Whenever longer engine response times than those for slam manoeuvres are
acceptable to the application, PLA and hence fuel flow are changed at a slow rate.
This greatly eases the operability concerns described later, as well as increasing
engine cyclic life.
The hot reslam is a particularly severe maneuver described in Fig. 8.4 and is only
used in service during an emergency.
It is also referred to as a Bodie, being named after a US air force pilot who first used
the maneuver during engine flight trials.
First the engine is held at a high power condition for at least 5 minutes to ensure the
carcass has soaked to its hot condition.
A slam decel to around idle is followed immediately by a reslam back to high power,
allowing no time for the carcass to thermally soak at the low speed.
In the combustor and turbines heat soakage is akin to additional fuel flow, and in
the HP compressor it lowers the surge line.
The adverse impact on the transient HP compressor working line is also shown in
Fig. 8.4.
This manoeuvre is used during engine development programmes to give the engine
harder operation than it will normally see in service to search for any potential surge
margin deficiencies.
Here a turbine drives both the engine compressor and the output load. When power
is extracted from a shaft two referred parameter groups, rather than just one, must be fixed
in order to fix all others. Referred power is used as a base parameter for the charts for this
configuration alone, because it is almost solely employed for power generation where the
shaft must rotate at synchronous speed irrespective of power level.
Figure 7.1 shows that for a given day temperature as fuel flow, hence SOT and
output power, are increased the compressor operating point moves up the constant referred
speed line. Equally if day temperature increases referred speed falls. If the engine is flat
rated to hold constant power then on hot days surge margin will reduce, as referred SOT
must increase.
Subsonic operation:
Fig. 7.6 the compressor working line. An additional parameter relative to the land
based engines is flight Mach number. The engine has a unique referred running line for
each level of flight Mach number. Only once the propelling nozzle chokes do these running
lines become coincident. The thrust level at which choking occurs depends on flight Mach
number, which produces different levels of nozzle pressure ratio. Referred fuel flow, air
flow and turbine temperatures show a strong variation versus Mach number at lower
referred speeds, via variation of turbine expansion ratio into the unchoked propelling
nozzle.
As Mach number increases at a referred speed level referred mass flow increases,
referred fuel flow reduces and the referred turbine temperatures are therefore lower. The
exact variation of referred P3 depends on the compressor map shape, the small amount
shown corresponds to relatively flat speed lines.
Unlike the land based engines, SFC improves significantly down to around 50%
thrust due to increasing propulsive efficiency outweighing falling thermal efficiency of the
core engine cycle. This is due to lower exhaust velocities and temperatures, and hence less
energy used for any level of exhaust momentum (i.e. thrust). At lower thrust levels SFC
worsens again, due to rapidly deteriorating thermal efficiency.
The largest effect of flight Mach number is via inlet momentum drag, which reduces
net thrust and therefore worsens SFC. The ‘ram’ compression partly offsets this, and the
available physical (rather than referred) thrust is high at cruise altitude. For early civil
airliners with turbojets rather than turbofans, achieving takeoff thrust rather than cruise
sized the engines.
Start phases
The key phases of a start are briefly defined below. Each is then comprehensively
described in the ensuing sections.
Dry cranking:
The engine HP shaft is rotated by the starter with no fuel being metered to the
combustor.
The purpose of dry cranking is to develop sufficient pressure and mass flow in the
combustor to permit light off. At start initiation the starter is energized and applies
torque to the HP spool.
To minimize shock torque loads torque may be applied gradually, for example via
slow opening of air valves for a turbine starter.
The HP spool then rotates and accelerates due to the excess starter assistance power.
The airflow induced by the HP compressor causes the LP spool and if applicable
eventually the free power turbine to break away from the oil at the bearings.
As is apparent from Fig. 9.2, starter torque peaks shortly after start initiation, while
starter power typically peaks at about 50% of the idle speed.
In the dry crank phase the engine provides a resistance on the HP spool which
increases with cranking speed, the turbine output power being less than that taken
by the compressor, auxiliaries, bearings and disc windage.
At all times the LP compressor power input is provided entirely by the LP turbine.
It is usual to have handling bleed valves open during starting to lower working lines
and, for inter stage bleeds, to raise the surge line.
The achievement of adequate driving pressure ratio for the bleed valves to pass flow
is crucial, levels should be assessed early in the design phase.
At low LP spool speeds LP compressor delivery bleed valves actually suck air in.
Invariably the starter cranks the HP spool rather than the LP, this being the most
efficient way to provide the combustor mass flow and pressure to enable ignition
and light around.
Starting is eased for engines which have a high fraction of their pressure ratio
developed by the HP compressor, as accelerating these stages via direct input shaft
power avoids energy loss to the low efficiency of the LP turbine at these conditions.
A related effect is that a relatively high LP spool inertia lengthens the time for that
spool to accelerate, and hence the whole start sequence.
The alternative of cranking an LP spool would incur worse pressure losses upstream
of the combustor in the HP compressor.
Purging:
This ensures that there is no fuel from previous operation or failed start attempts in
the engine gas path or exhaust that may ignite and cause damage.
The engine is dead cranked at the maximum speed the starter can sustain, which
purges any fuel into the atmosphere.
Purging is required for all starts and restarts with gas fuel, and may be used for
liquid fuels following a failed start or emergency shut down.
Where required purging typically lasts for 1–10 minutes depending on engine size
and the type of unburnt fuel to remove.
The dead crank phase where purging is performed does not appear on Fig. 9.2.
There the combustor has been lit before the HP spool has accelerated to a point
where the engine net resistance equals the starter assistance.
Fuel is metered to the combustor, and ignitors are energized. This causes ignition
locally within the combustor, and then light around of all the burners.
Here the ignitors are activated and a constant light off flat of fuel flow is metered
to the combustor by the control system.
The light off flat may be as low as 300kW for a small RPV turbojet and up to 5000
kW for a large turbofan.
Once fuel has ignited local to the ignitor, the flame must propagate and stabilize
circumferentially around the combustor.
The HP speed at which the combustor conditions are suitable for light off must be
found from a combination of modelling, combustor rig testing and finally engine
testing.
Light off is a key combustor condition with many practical issues to be overcome
such as the atomization of kerosene or highly viscous diesel on cold days.
Figure 9.2 shows that on light off there is a step reduction in engine resistance,
however starter assistance is still required to continue HP spool acceleration.
Turbine power output is still usually less than the sum of compressor input power,
bearing and windage losses, and auxiliary requirements.
Figure 9.3 shows that both compressor working lines also show a step upwards due
to light off, with the HP compressor being the closest to the rotating stall drop in
line.
It is essential that the starter motor size, hence HP speed at the top of crank, and the
light off flat are chosen such that the HP compressor does not go into rotating stall.
The control system usually detects ignition and light around by means of
thermocouples placed rearwards in the turbines.
If light off does not occur within some specified time (e.g. 10 seconds) the control
system aborts the start, shuts off fuel flow and commences a purging phase.
Acceleration to idle:
This is achieved via a steady increase in fuel flow, and continuing starter assistance.
Fuel flow is steadily increased, causing the engine to accelerate towards idle very
much as per the above idle accels.
The starter motor continues to provide crank assistance well after light around. As
speed increases the engine assistance eventually dwarfs that of the starter, which
cuts out before idle via de-energization and declutching.
As shown on Fig. 9.2, the engine resistance crosses the ‘X’ axis and becomes
assistance shortly after light off. This point is called self-sustain and theoretically if
the starter motor were cut the engine could operate there steady state.
However combustor exit temperature profiles make this impractical with respect to
turbine life, and hence this speed must be passed through transiently.
During acceleration fuel flow is scheduled such that the compressor working lines
run approximately parallel to the rotating stall drop in line.
Many combustion systems employ separate, lower flow injectors for starting. This
is because at low fuel flows the main injectors may not produce adequate
atomization; here the combustor stability and efficiency depend strongly on which
injectors are in use.
In many cases below a certain threshold the main burner system may in fact pass
no flow at all. The point of changeover between systems must be chosen with these
issues in mind, to avoid extinction, stall or hang.
On reaching idle, fuel flow is cut back and the engine assistance/resistance becomes
zero; no unbalanced power is required for steady state idle operation.
The idle point on the HP compressor map is below the transient start working line,
whereas for the LP compressor it is higher.
Heat soakage, can have a very significant impact on working lines for hot restarts
or cold soaked starts. For an immediate restart following shut down heat transfer
from the carcass is akin to additional fuel flow, and will push the compressor
towards rotating stall and increase turbine temperatures.
In addition, the compressor surge lines are lowered,. Conversely, after a prolonged
cold soak, heat transfer to the carcass is akin to reduced fuel flow and may drive the
engine towards hang.
Thermal soakage:
Engines are often held at idle to allow the carcass to thermally soak to the new
temperature to preserve cyclic life.
3.9 Windmilling
Windmilling occurs when air flowing through an unlit engine causes spool rotation.
This phenomenon applies mostly to aircraft engines, where it is caused by ram pressure.
Examples include when an engine has flamed out during flight, or an unmanned air
launched vehicle is being carried by a parent aircraft prior to launch. The direction of
rotation is the same as for normal operation. Under certain conditions windmilling also
occurs for land based and marine engines.
Free windmilling is where all the engine spools are free to rotate. Locked rotor
windmilling is where the HP spool is mechanically prevented from rotating.
The aircraft designers must know the engine drag (i.e. negative thrust) during
windmilling. Drag is caused by air slowing down as it passes through the engine.
Figure 10.1 shows schematically how the usual non-dimensional relationships can
be extended to the windmill regime. Referred fuel flow and mass flow are shown versus
referred speed for a high and low flight Mach number; a locus of windmill points is apparent
at the windmill condition of zero fuel flow. Mass flow increases with flight Mach number,
as the higher the Mach number the higher the referred speed. Operation on the lower parts
of the curves may not be practical since the combustor is likely to weak extinct in this
regime, causing the engine to decelerate to the windmill point. Other referred parameter
groups may be plotted in this manner. Finally, Fig. 10.1 shows the locus of operating points
for free and locked rotor windmilling on the low speed compressor map.
The pressure and temperature ratios at key stations through a turbojet while
windmilling. The compressor behavior depends on flight Mach number:
As flight Mach number is increased from zero, compressor pressure ratio initially
falls from the value of one. The temperature ratio is greater than one however, and
hence the compressor is operating as a stirrer or paddle. For a multi-stage axial flow
compressor some front stages may actually perform as a turbine but overall the
machine has a net work input.
As Mach number increases further, pressure ratio increases and eventually exceeds
one. This is particularly true for high design pressure ratios and multiple axial
stages. Here the compressor overall, and certainly the back stages, function in the
true compressor fashion. Typically the higher the design pressure ratio, the higher
the windmill pressure ratio at a given flight Mach number. For a given design
pressure ratio centrifugal compressors tend to have a lower pressure ratio at a flight
Mach number, due to the absence of back stages able to operate normally.
During steady state windmilling the compressor always absorbs power overall.
Otherwise the shaft would accelerate as there is no other significant power absorption
mechanism, bearing and windage losses being small. The turbine is able to supply the
compressor input power because there is an expansion ratio remaining after all pressure
drops due to ducts, combustor and possibly compressor(s) have been deducted from the
ram pressure ratio. The combustor, intake, exhaust and other ducts impose pressure losses
with no change to total temperature.
The expansion ratio across the propelling nozzle is small and is typically
independent of turbojet design pressure ratio. Total temperature at the nozzle is slightly
lower than the ram temperature. This small drop is due to any work extracted by the
bearings and engine auxiliaries, the effects of compressor and turbine work otherwise
cancelling each other.
The bypass duct presents the path of least resistance to the ram pressure at the fan
face, hence most flow takes this path. This means that the fan demands high work input
and therefore the core must match such that a high expansion ratio is available for the LP
turbine. The HP turbine therefore has a low power output, and the HP compressor has a
significantly lower pressure ratio than in an equivalent turbojet.
Windmill tests in an altitude facility on two single spool turboprops, with the
propeller pitch set for maximum windmill rotational speed. In this configuration the
propeller acts as a turbine, dropping pressure and temperature, and hence producing shaft
power. The result is that 100% referred rotational speed is achieved at a flight Mach number
of less than 0.4, and substantial customer power extraction is available. The compressor
pressure ratio is greater than one for all flight Mach numbers as it is driven by both the
propeller and turbine. As shown by Fig. 10.2, the compressor working line is lower than
the no load line during normal operation, due to zero fuel flow and hence zero combustor
temperature rise. For the 100% referred speed case pressure ratio is approximately 25% of
its ISO takeoff design point value. This pressure ratio, less combustor and duct pressure
losses, is available for expansion across the turbine. Referred air mass flow is
approximately twice that of a turbojet at the same flight Mach number.
The drag is predominantly created by the propeller. At high flight speeds the
magnitude of the drag would approach that of cruise thrust in normal operation. In actual
flight situations such a large drag makes it impractical to operate with the propeller pitch
as above. Hence engines are fitted with a ‘reverse torque switch’ in the gearbox, which
senses windmill operation by the change in direction of torque due to the propeller driving
the engine (as opposed to vice versa). The control system then coarsens the pitch to the
feathered position where the propeller blades are parallel to the direction of flight,
preventing any engine rotation and ensuring drag is minimal.
In this instance the propeller must be feathered immediately since it will otherwise
overspeed the power turbine. There is no connection to the compressor to absorb the output
power from the propeller and the power turbine itself. Only a small fraction of the available
power is absorbed by auxiliaries, bearings and windage.
3.11 Formulae
a. What are the design point performance parameters that are involved in gas turbine
engine? (10M)
b. Write the steps involved in starting of jet engine. (3M)
c. Draw and explain a typical restart envelope for a civil turbofan engine. (7M)
d. A turbo jet engine performance data is given below: (10M)
RPM = 9500; EGT = 4500 C; Wf (fuel consumption) = 1800 kg/hr; Wa (air
consumption) = 91 kg/sec; TSFC = 0.5. The test is carried out at a pressure of 102.6 KPa
and ambient temperature of 300 C. Correct the test data for ISA conditions (Pressure
101.3 KPa and temperature 150 C). Take Fn (Net thrust) = 4510 kg.