You are on page 1of 13

Last Rev.

: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

Page 1

Grading Sheet
~~~~~~~~~~~~~~
MIME 3470Thermal Science Laboratory
~~~~~~~~~~~~~~
Experiment 18
~~~~~~~~~~~~~~

BRAYTON CYCLE JET ENGINE


~~~~~~~~~~~~~~

Students Names / Section


POINTS
PRESENTATIONApplicable to Both MS Word and Mathcad Sections

5
10
5

GENERAL APPEARANCE
ORGANIZATION
ENGLISH / GRAMMAR
ORDERED DATA, CALCULATIONS & RESULTS
PLOT 5 PRESSURES, 5 TEMPERATURES, & RPMS
(vs. TIME) ON ONE PLOT.
FOR THE 3 STEADY-STATE OPERATING POINTS SELECTED
CALCULATE COMPRESSOR ISENTROPIC EFFICIENCY
CALCULATE TURBINE ISENTROPIC EFFICIENCY
CALCULATE THERMODYNAMIC ACTUAL EFFICIENCY
CALCULATE THERMODYNAMIC ISENTROPIC EFFICIENCY
PLOT THE 4 EFFICIENCIES vs. RPMs

5
10
10
10
10
5

TECHNICAL WRITTEN CONTENT


DISCUSSION
HOW DOES THE ACTUAL CYCLE EFFICIENCY COMPARE
WITH THE IDEAL BRAYTON CYCLE?
HOW DO THE COMPRESSOR AND TUBINE EFFICIENCIES
AFFECT THE CYCLE EFFICIENCY?
CONCLUSIONS
NO DATA REQUIRED/WE HAVE YOUR DATA FILES ALREADY
TOTAL
COMMENTS

GRADER

10
10
10
0
100

SCORE

TOTAL

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

MIME 3470Thermal Science Laboratory


~~~~~~~~~~~~~~
Experiment 18

BRAYTON CYCLE JET ENGINE


~~~~~~~~~~~~~~

LAB PARTNERS: NAME


NAME
NAME
SECTION

EXPERIMENT TIME/DATE:

NAME
NAME
NAME
TIME, DATE

~~~~~~~~~~~~~~

OBJECTIVES of this experiment are to:


1. Understand the basic operation of a Brayton cycle 1.
2. Determine the performance (efficiencies) of an actual turbine
components and the cycle.
Q
in

Combustor
(Heat Exchanger)

Compressor

purpose and provides acceptable quantitative results for gas turbines.


The foremost assumption of this model is that air is the working fluid
treated as an ideal gas throughout the cycle. Thus, neither the mass
of injected fuel nor the different chemical make up and properties of
the exhaust gases are considered.
The Brayton cycle (Figure 2b) used to model a steady-flow gas
turbine further assumes the following idealized processes:
States 1 to 2s Isentropic compression of air.
States 2s to 3 Reversible, constant-pressure heat-addition to the air
no actual combustion is considered and the products
of combustions (exhaust) is considered to be air.
States 3 to 4s Isentropic expansion of air.
States 4s to 1 As the model is a closed cycle, a process between
States 4s and 1 must be considered. This is modeled
as a reversible, constant-pressure heat rejection at
ambient pressure.
T p p2 const

W cycle

4a

p p1 const

Heat Exchanger

Q out
Figure 1Basic Brayton Cycle Model of a Generic Gas Turbine

THEORY
A simple gas turbine is comprised of three man-made components
and one implied component when considering it as a closed cycle
(see Figure 1). The implied componentthe lower heat exchanger
of the figure operating between States 4 and 1will be discussed
later but is added when considering the gas turbine as a closed,
ideal cycle. The three man-made components are
1. Compressorlow-pressure (ambient) air of State 1 is
compressed to State 2.
2. Combustorfuel is added to the compressed air and ignited.
3. Turbinethe hot combustion gases expand through and produce

work by the turbine, W


turb . Part of this work is used to drive the

compressor, W
comp . The net outputted work of the entire cycle,

W
cycle , is shaft work and can be used to power machineryi.e.,

generators or helicopters. Any kinetic energy of the exhaust gases


at State 4 is considered lost energy in this case. However, if a gas
turbine is used as a jet engine, then thrust is the desired output and
just enough work is produced by the turbine to drive the compressor and produce any needed auxiliary power. Then the exhaust
gases are expanded through a nozzle to create a high-velocity
flow, i.e., thrust.
To analyze the cycle, we need to evaluate all the states as completely
as possible. The Brayton air-standard model2 is very useful for this
1
2

In European literature, this is called the Joule cycle.


air-standard models: provide useful quantitative results for realworld processes such as gas turbines and spark and compression ignition
engines. In these models the real processes are simulated by a closed cycle
of air (treated as an ideal gas) where combustion is modeled as heat
addition without actually combusting an additional mass of fuel. Thus,
there are no exhaust gasses. Air-standard processes include the Otto,
Diesel, dual, and Brayton cycles. The Otto cycle is a constant-volume heat
addition (near top dead center) and is used to simulate spark ignition
engines while the Diesel cycle is a constant-pressure heat addition to
simulate compression ignition engines. Since the p-V diagrams of actual
internal combustion engines are not described well by either the Otto or

T p p3 const

2a

Turbine

Page 2

(a )

2s 2a

4a
4s
p const
(b )

Figure 2Irreversibility Effects in Simple, Closed-Cycle Gas Turbines

Irreversibilities
The principal states points of a simple closed-cycle gas turbine might
be shown more realistically in Figure 2a. Because of irreversibilities
within the compressor and turbine, the working fluid would experience increases in specific entropy across these components. Owing to
irreversibilities (friction), there also would be pressure drops as the
working fluid passes through the heat exchangers (or the combustion
chamber of the open-cycle gas turbine). However, because frictional
pressure drops are less significant sources of irreversibility, we ignore
them in subsequent discussions and for simplicity show the flow
through the heat exchangers as occurring at constant pressuresee
Figure 2b. Stray heat transfers from the power plant components to
the surroundings represent losses, but these effects are usually of
secon-dary importance and are also ignored in subsequent
discussions.
As the effect of irreversibilities in the turbine and compressor
becomes more pronounced, the work developed by the turbine
decreases and the work input to the compressor increases,
resulting in a marked decrease in the net work of the gas turbine.
Accordingly, if an appreciable amount of net work is to be
developed by the gas turbine, relatively high turbine and
compressor efficiencies are required. After decades of

Diesel cycles, the dual cycle was developed and has a constant-volume heat
addition immediately followed by a constant-pressure heat addition. The
Brayton cycle is used to simulate gas turbines and has a constant-pressure
(atmospheric) heat addition. All of these air-standard cycles include a heat
rejection process to close the cycle. Further, in all of these cycles, the
compression and expansion of the working fluid is considered as
isentropic. The real engines these cycles simulate do not operate in a cycle;
but, instead the intake air and fuel at specific locations and eject exhaust to
the atmosphere. To further put air-standard cycles in perspective, before a
student is introduced to the concepts of entropy and isentropic processes,
he/she is usually introduced to the Carnot cycle which is the basis of
studying refrigeration, heat pumps, and power plants. In the Carnot cycle, a
gas is compressed and expanded adiabatically with isothermal heat
addition and rejection.

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

developmental effort, efficiencies of 80 to 90% can now be


achieved for the turbine and compressor components.

h T

Tref

Cold Air-Standard Analysis


In cold air-standard analyses of air-standard models, the specific
heats of air are assumed constant (perfect gas model) throughout the
entire cycle. The values of the specific heats are determined at room
temperature. Such an assumption leads to closed form solutions
having only a small loss of accuracy. It also enables one to study quailtatively the influence of major parameters on the performance of the
actual cycle. In this lab, we will not use the cold air-standard model.
A Truer Analysis of the Ideal Cycle
Instead, the effect of temperature on the specific heat can be
included in the analysis at a modest increase in effort. To perform
the thermodynamic analysis on the cycle, we consider control
volumes containing each component of the cycle shown in Figure
1. These components are addressed below.
Compressor
Consider the compressor of Figure 3 and the energy flows across a
control volume (c.v.) around it.

Compressor

(5)

Use this relation to determine all enthalpies. Mathcad will solve


this integralsee further on.
Now, to determine h2, remember that we are assuming an ideal
isentropic compression process. Thus, we can use the relation

0 s 2 T2 s , p 2 s1 T1 , p1

T2 s

c p T

T1

p
dT
R ln 2
T
p1

(6)

where, R is the gas constant for air. We know the pressure at States 1
and 2 and the temperature at State 1. To solve for T2s, we need to
determine where the function

F T2 s

T2 s

p
dT
R ln 2
T
p1

c p T

T1

(7)

Finding Roots______ _________________________


.
Root(f(var1, var2, ), var1, [a,b]) Returns the value of var1
lying between a and b at which the function f is equal to zero.

W comp

c.v.

Figure 3Compressor and Associated Control Volume

Note that ideally there is no heat transfer from the control volume
to the surroundings. Under steady-state conditions, and neglecting
the kinetic and potential energy effects, the first law of
thermodynamics for this control volume is then written as

W
comp H 2 H1

(1)

represents enthalpy flow and the power to compress the air


where H

is Wcomp . Note that, thermodynamically, work performed on a

process is negative. However, W


comp as expressed in Equation 1 is

. As we have the same mass


positive because H 2 is larger than H
1
flow rate both into and out of the control volume, we may write the
specific form of the first law as
wcomp m
h2 m
h1
m
wcomp h2 h1 (2)
or

Constant pressure specific heat is a function of temperature only


c p(T) = (h/T)p; thus
T

c p T dT

(3)

Tref

where Tref is some reference temperature. To solve the integral, we


need some relation for cp since specific heat is a function of
temperature. For air, we will define specific heat as [2]
a bT cT 2 dT 3 kJ
c p T
28.97
kg K

(273K < T < 1800K)

(4)

where

a bT cT 2 dT 3
kJ
dT
28.97
kg

is zeroi.e., we want to calculate the value of T2s for which a plot of


the function crosses the abscissa. This is called finding the root of the
function. Mathcads help option explains the root function as:

h T

Page 3

T Temperature in Kelvin degrees


28.97 Molecular weight of air
a 28.11
b 0.1967 102
c 0.4802 105
d 1.966 109.
Thus, enthalpy is

The Mathcad example below demonstrates using the integral


function (as promised above) and the Mathcad library function root.
Dummy Data (Temperatures in Degrees C & Pressures In psig;
Want Degrees K & psia--But DO NOT Apply Units):
Define New Units:
P atm 14.7
kJ 1000 J
T 1 20.1 273.15
P 1 0.1 P atm
P 2 8.9 P atm
Molec ular Weight of Air:
M 28.97
Gas Constant for Air:

8.31434
M

kJ
kg K

Func tion for Spec ific Heat of Air:


2
5
9
a 28.11 b 0.1967 10
c 0.4802 10
d 1.966 10
cp ( T )

a b T cT

d T

M
Compressor Isentropic Outlet Enthalpy-Refer to the tool bar for the integral func tion:

T 2s root

cp ( T )

T1

P2

P1

h ( T )

T 300 600

T 2s 335.012
To find the enthalpy, integrate the func tion use
As a referenc e temperature, use

kJ
kg K

d T R ln

dh = cp dT

T ref 273.15

T
ref

cp ( T ) K d T

The units of Degrees Kelvin have been added to the function.


This is bec ause no units have previously been c ommitted to
temperature but we have given units to the func tion c p(T) and
we want the units of enthalpy to be kJ/kg. So we supply the
units for the dT term.
Thus, the isentropic enthalpy of State 2s is
kJ
h 2s h T 2s
h 2s 62.135
kg

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

The irreversibilities present in the real process can be represented


by introducing the compressor isentropic efficiency,

w h
comp isen wcomp s h2
comp a

h1

2a

h1

(8)

where the subscripts s and isen both refer to the isentropic process
and the subscript a refers to the actual process.
Combustor
Now consider the combustor and its associated control volume of
Figure 4. Under steady-state conditions, neglecting kinetic and
potential energy effects, and following the procedure used for the
compressor, the first law for this control volume is written as

qin h3 h2 a

(9)

Q in

where,

cp specific heat
ho stagnation enthalpy
h static enthalpy
To stagnation temperature
T static temperature.
For constant cp (cold air-standard model assuming an ideal gas)
this becomes

ho h c p To T

c.v.

Combustor
(Heat Exchanger)

Figure 4 Combustor and Associated Control Volume

The isentropic turbine efficiency is developed as was done for the


compressor:

turb isen h4a h3

(11)

h4 s h3

Compressor

c.v.

W cycle
See Figure 1

V 2

(12)

where,

po stagnation pressure
p static pressure
fluid density
V fluid velocity.
The stagnation temperature is determined from the stagnation
enthalpy which is

ho h

1V2
2

or ho h

To

c p T dT 12 V

1V2
2

V 47.74

m
V
s

This converts to about 108mph or 175km/hr. So, for low velocity


flows such as Stations 1 and 2, we can ignore the effects of
velocity and say that T To. We can also ignore the velocity of the
gas at Station 3just after the constant pressure heat addition
where gases have not yet been allowed to expand to a lower
pressure and reach a higher velocity. Further, since the purpose of
the turbine is to extract shaft work and not to accelerate the flow,
we also ignore the velocity of the gas at Station 4. So, the only
location where the gas velocity is significant enough that we must
differentiate between static (thermodynamic) temperature and total
temperature is at Station 5 (see Figure 6).
Now, what about differentiating between static and stagnation
pressure? A gas moving at less than Mach 0.3 is usually
considered as incompressibleconstant density. If the gas is air at
room temperature (say, 300K), its density is 1.1614kg/m3. Further
if the gas is at rest, its static and stagnation pressures are the same
po = p = 1.01325 105 N/m2. Using Equation 12 and giving the
gas the velocity we computed above in Equation 14 of 47.74m/s
the static pressure of the gas becomes

1atm

Comments on Pressure and Temperature Measurements


Thermodynamic properties such as enthalpy are usually tabulated as
functions of static (thermodynamic) temperature, T, and pressure, p.
Yet in this experiment, the only static pressures are p1 and p3 and the
only static temperature is T1. All other measured temperatures and
pressures are stagnation (or total) values, usually denoted with the
subscript of o. Stagnation pressure is defined by
1
2

kJ
m2
1140

kg
s2

N
p o 1.01325 10 5
p 12 V 2
2
m

Figure 5Turbine and Associated Control Volume

po p

(14)

(15)

Turbine
Next, consider the turbine and its control volume as shown in Figure 5.
Following the example of the previous components of the jet engine:
for steady-state conditions and neglecting the kinetic and potential
energy effects, the first law for this control volume is
wturb wcomp wcycle h3 h4 a
(10)

W comp

1V 2
2

NOW, we ask: How fast must the flow be moving before the
velocity term causes noticeable differences between To and T?
To answer this, we note that in the thermodynamic tables for air in
the vicinity of 1000K that a 1K difference in temperature leads to
about a 1.14kJ/kg change in enthalpy. Converting this to velocity
using Equation 13

ho h 1.14

Page 4

(13)

kg
m
p 12 1.1614 47.74
3
s
m
1.32348103

N
m2

(16)

0.01306 atm

We see that the velocity must be large before static and total
pressures are appreciably different.
We conclude that for our engineering calculations, the total pressure and temperature data values may be substituted with static
values for Stations 1 thru 4. The high velocity of the expanded gas
will only be significant at Station 5the nozzle outlet.
The student may be a bit confused by the term expand. In an engineering application whenever a gas is expanded the desire is to do such
reversiblyi.e., get as much desired effect from the working fluid. If a
gas were to simply explodea very irreversible process little useful
work could be captured from the process. Both the turbine and the
nozzle of a jet engine are designed to recover as much useful work as
possible from an expanding gas. In the case of the turbine, if the
expan-sion process caused higher gas velocity, the turbine would be

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

conver-ting thermal energy into kinetic energy instead of its desired


job of generating shaft power. So we do not want to increase the
working fluids velocity through the turbine. On the other hand, the
job of a nozzle is to convert thermal energy to thrust energy i.e.,
velocity.
Cycle Efficiency for a Jet Engine
Thermodynamic efficiency, th, is defined as

th

Desired Energy Output


(17)
Energy That Costs Input Energy

The reader will remember that gas turbines are used for two different functions. The first is where shaft power is needed to propel
an aircraft or a fast boat or to drive an electric generator. In such
cases, the turbine is designed to absorb as much of the energy
from the exhaust gases as possible. Some of this energy is used to
drive the compressor and the rest is net shaft work for the cycle,

W
cycle . In this usage, the thermo-dynamic efficiency for the
cycle is

th

W cycle w turb w comp h3 h4 a h2 a h1

q in
h3 h2a
Q in
(18)

The second use of a gas turbine is as a jet engine where the desired
output is thrust3i.e., a high gas velocity at Station 5, which is
nozzle exit in Figure 6. The thermodynamic efficiency in this case
is expressed as
h h5a
W
V52a / 2
th thrust
o5 a
(19)

h3 h2a h3 h2a
Qin
where

s 0

W comp

W thrust
c.v.

Figure 6Turbine and Nozzle and Associated Control Volume

Remember that in this experiment, we measure the total or stagnation


temperature and pressure at Stations 2, 4 and 5 and static pressure and
total temperature at Station 3. Further, because of the small flow
velocities at Stations 2, 3, and 4, the stagnation values are essentially
the same as the static values. Thus, we only have to determine h5a
from the stagnation value ho5a = f(To5a).
h
po5a
ho5a

h5a

p5a p atm
s

Figure 7h-s (Mollier) Diagram for State 5

We will assume that the increase in pressure between States 5a


and o5a is due to an isentropic compression as shown in Figure 7.

A jet engine on a real aircraft would have as desired outputted


work both thrust and shaft work to run, say, an electric generator to
power avionics.

To5 a

c p T

T5a

p
dT
R ln o5a
T
p 5a

(20)

Now, we know all the temperatures associated with the enthalpies


of Equation 19 and can compute the thermodynamic efficiency.
Finally, to determine the ideal Brayton cycle thermodynamic efficiency, we consider only isentropic expansions thus we need to
compute T5s from T5a just as we computed T2s and T4s. Thus,
T5 s

0 s 5 T5 s , p 5a s 4 T4 s , p 4a

c p T

T4 s

p
dT
R ln atm
T
p 4a

(21)

To compute the ideal Brayton cycle efficiency, we do not use Equation


19 and apply isentropic values instead of actual values of enthalpy.
Ideally, the shaft work absorbed by the turbine is entirely used to drive
the compressor. However, only 70 to 90 % of the actual energy
absorbed by the turbine is actually passed on to the compres-sor. This
lost energy could be used in an ideal cycle to produce thrust. Thus, the
ideal thermodynamic efficiency is defined as

th isen h3 h5s h2s h1


h3 h2s
T
2s 2a

Knowing po5a, p5a, and To5a, we can solve for T5a using Equation
20 and using a root solver as we did with Equation 7

ho5a actual total enthalpy at Station 5, and


h5a actual enthalpy at Station 5.

Page 5

(22)

p3

4 s 4a

p4

p patm
5
5a
5s
s

Figure 8Irreversibility Effects When Cycle Is Used As a Jet Engine

For either static or stagnation conditions, Equation 5 can be used


to determine enthalpy.
EXPERIMENTAL SETUP
The laboratory setup (Figure 9) is a self-contained, turnkey, portable
propulsion laboratory manufactured by Turbine Technologies Ltd.
called TTL Mini-Lab. The Mini-Lab consists of a real jet enginethe
same safety concerns of running any jet engine are present. Care
must be taken to follow all the safety procedures precisely as
outlined in the laboratory and stated by your lab instructors. The
following description of the setup is provided by the manufacturer.
The Model SR-30 turbojet engine is the primary system component.
Operational sound and smell are hard to distinguish from any idling,
small business jet. The engines axial turbine wheel and vane guide
ring are vacuum investment castings produced from modern, high
cobalt and nickel content super alloys (MAR -M-247 and Inconnel
718). The combustion chamber consists of an annular, counterflow system, including internal film cooling strips. Fuel and oil
tanks, filters, oil cooler, all necessary wiring and plumbing are
located in the lower part of the Mini-Lab structure. A throttle
lever is located on the right side of the operator on the slanted
instrument panel. The throttle enables the operator to perform
smooth power changes between idle and maximum. Digital
engine RPM and exhaust gas temperature (EGT) gauges, Oil,
Fuel, Air start pressure gauges are also part of the standard panel.
Annunciator lights indicate low oil pressure, igniter on, and airstart bus. Other panel-mounted switches control igniter, air start,

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

and activate fuel flow. The SR-30 engines fuel system is very
similar to large-scale enginesfuel atomization via 6 return flow
high-pressure nozzles that allow operation with a wide variety of
kerosene based liquid fuels (e.g., diesel, Jet A, JP-4 through 8).

Figure 9MiniLab Jet Engine and Experimental Setup

Engine Components and Measurement Locations


The engine consists of a single stage radial compressor, a counterflow annular combustor and a single stage axial turbine which

directs the combustion products into a converging nozzle for


further expansion. Details of the engine may be viewed from the
cutaway provided in Figure 10.
Instrumentation, Data Acquisition, and Data Import to Mathcad
The sensors are routed to a central access panel and interfaced
with data acquisition hardware and software from National
Instruments. The manufacturer provides the following description
of the sensors and their location. Along with fuel flow and digital
thrust readouts the data acquisition system (LabVIEW) creates the
following eleven data files as functions of time:
Shaft rpm (SPEED)
Compressor Inlet:
Static Pressure (P1)
Static Temperature (T1)
Compressor Outlet:
Stagnation Pressure (P2)
Stagnation Temperature (T2)
Combustion chamber/Turbine Inlet:
Static Pressure (P3)
Stagnation Temperature (T3)
Turbine Outlet:
Stagnation Pressure (P4)
Stagnation Temperature (T4)
Thrust Nozzle Exit:
Stagnation Pressure (P5)
Stagnation Temperature (T5)

TURBINE

RADIAL
COMPRESSOR

Page 6

COMBUSTOR

INLET
NOZZLE

OUTLET
NOZZLE

OIL PORT
Compressor

Inlet

Compressor
Outlet

COMPRESSOR/TURBINE
SHAFT

Turbine Inlet

Turbine Outlet Figure 10Cut-Away View of Turbine Technologies SR-30 Engine


To
read LabVIEWs
data Exit
files (*.lvm) into Mathcad:
Nozzle
1. With Mathcad open, select from the menu
INSERT / DATA /FILE INPUT.
2. The dialog box below will appear

FUEL
INJECTOR

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

Page 7

data file P1_Data above into pressure and time vectors,


proceed as shown below. Remember: Temperature data is in
degrees Celsius and must be converted to Kelvin degrees by
adding 273.15K. Further, pressure data is in psig and must be
converted to psia by adding 14.696psi.
Compressor Inlet

0
tP1 P1_Data
1
Pressure Data:
P1 P1_Data
14.696
The length of these vectors is:
length tP1 364
Time Data:

0
0

The File Format should be Text. The second field contains the
file path name relative to the C Directory, i.e., c:\**.
3. Click NEXT and get the following dialog box. Change nothing

4.Click FINISH and the item below appears

Note: when the Mathcad object is closed, the file icon remains but
the path name to the file is not displayed. In the place holder type
P1_data. Now if one lists P1_data (as shown below), the times will
appear in Column 0 and the pressures in Column 1.

0
0

14.696

0.36

14.696

0.621

14.696

0.891

14.696

1.162

14.696

1.432

14.696

1.722

14.696

tP1 7
P1 7
2.003
14.696
In the above
example
we
have
used
the
Mathcad
8
2.303
8
14.696 function
length. It merely tells us how many elements are in the vector
9
14.696
starting9with the2.594
0th element.
10
2.894
10
14.696
EXPERIMENTAL
PROCEDURE
11
3.194
11
14.696
SAFETY NOTES
12you are3.495
12 If 14.696
1. Make sure
wearing ear protection.
you are not sure
how the earplugs
are
properly used, ask 13
you lab14.696
instructor for a
13
3.805
demonstration. NEVER stay in the laboratory without ear
14
4.116
14
14.696
protection while the engine is in operation.
15engine4.426
15 speeds.
14.696Although
2. The SR-30
operates at high rotational
there is a protective pane that separates the engine from the
operator, make certain that you do not lean too close to this pane.
3. Make sure the low-oil-pressure light goes off immediately
after an engine start. If it stays on or comes on at any time
during the engine operation cut off the fuel flow immediately.
4. There is a vibration sensor whose indicator is to the far right of
the operators panel. If this indicator shows any activity
(increase in voltage) shut-off the engine immediately.
5. If at any time you suspect something is wrong shut off the fuel
immediately and notify the lab instructor.
6. If the engine is hung (starts but does not speed up to idle speed
of about 50,000 rpm) turn the air-start back on for a short
while until the engine speeds up to about 30,000 rpm. Then
turn off the air-start switch.

MAKE SURE NEITHER YOU NOR ANY OF YOUR BELONGINGS


4. This experiment measures many temperature and pressures as
well as rpms. LabVIEW is instructed to sample all of these
items every so many seconds. However, when these items are
measured, they are measured one at a time and the exact time
of each measurement recorded. Thus, p1 will be measured at a
slightly different time than T1. To separate the data in the
individual columns into vectors, choose the Matrix icon,

from the Math Tool Bar. From the Matrix Pallet,


,
we will use the function noted as M<>. M<i >chooses the ith
column of the matrix. REMEMBER: Matrix and vector
indicing begins from 0 and not from 1. Thus, to partition the

ARE PLACED IN FRONT OF THE INTAKE TO OR THE EXHAUST


FROM THE ENGINE WHEN THE ENGINE IS RUNNING.

PROCEDURE
1. Ask your TA to load the data acquisition program and run the
preprogrammed LabVIEW VI for this lab. The screen should
display readings from all sensors. Review the readouts to make
sure they are working properly.
2. Make sure that the air pressure in the compressed-air-start line
is at least 100 psia (not exceeding 120 psia). Ask your lab
instructor to check the oil level.
3. Have your lab instructor turn on the system and start the engine.
After starting the engine, you must first allow it to achieve the idle
speed before making any measurements. Make sure the throttle is
at its lowest point. The idle position is nearly vertical, and is close
to the engine (away from the operator).

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

4. Slowly open the throttle. Make sure that you allow the engine
time to reach steady state by monitoring the digital engine rpm
indicator on the panel. The reading fluctuates somewhat so use
your judgment.
5. Take data at 3 different engine speeds. You will use the data to
study how cycle and component efficiencies change with speed.
6. After you are done taking data, turn off the fuel flow switch first.
7. The data will be stored in LabVIEW files (*.lvm) which can
be read by Mathcad.
DATA ANALYSIS & REPORT
1. First, read your data files and plot all 11 data values (5
pressures, 5 temperatures, and rpms vs. time) on one plot
similar to that shown below.

Page 8

2. From this plot, select three points in time where you consider
the gas turbine to be operating at steady state. Then determine
the indices of the data vectors corresponding to these times.
For the example plot above, the first steady-state time chosen
was 95sec. In the previous column, a time vector of tP1 was
established. The student needs to find (by trial and error) what
index of the vector corresponds to about 95sec (or what ever
time the student chooses from his or her data).
3. For the 3 steady-state operating conditions, calculate com-

pressor isentropic efficiency, turbine isentropic efficiency, actual thermodynamic efficiency (based on thrust),
and the ideal Brayton cycle thermodynamic efficiency
based on thrust for the jet engine. For the reference
temperature in computing enthalpy, use Tref = 273.15K.

4. On one graph, plot comp isen vs. rpm, turb isen vs. rpm,

th vs. rpm and th isen vs. rpm for the 3 steady-state

conditions.
FOR THE DISCUSSION
1. How do the compressor and turbine efficiencies affect the
cycle efficiency?
2. How does the actual cycle efficiency compare with the ideal
Brayton cycle efficiency?
REFERENCES
1. Turbine Technologies, Ltd., Brayton CycleJet Engine Experiment, http://www.turbinetechnologies.com/minilab/Technical Papers/ Univ
of Toledo.pdf

2. engel, Yunus A., and Boles, Michael A., Thermodynamics


An Engineering Approach, 4th ed., McGraw-Hill, 2002
3. Moran, Michael J., and Shapiro, Howard N., Fundamentals of
Engineering Thermodynamics, 2nd ed., John Wiley & Sons, 1992

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

Page 9

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

ORDERED DATA, CALCULATIONS, and RESULTS


Mathcad object, DOUBLE CLICK to open
Data:

Temperatures are in degrees Celcius and Pressures are in psig--change both to absolute values

Ambient Pressure (1 atm):

Patm 14.696

Celsius to Kelvin Conversion:

Kelvin 273.15

Page 10

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

DISCUSSION OF RESULTS
1. How do the compressor and turbine efficiencies affect the
cycle efficiency?
Answer:
2. How does the actual cycle efficiency compare with the ideal
Brayton cycle efficiency?
Answer:

CONCLUSIONS

Page 11

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

Page 12

APPENDICES
APPENDIX AGEORGE BAILEY BRAYTON
George Bailey Brayton
b. 3 October 1830, Compton, Rhode Island
d. 17 December 1892, London, England.
Brayton was the son of a cotton mill superintendent who
was himself an inventor. He was fascinated with engines
and began seriously experimenting with combustion in a
cylinder at age thirteen. After minimal public schooling, he
apprenticed in a machine shop in Providence and became a master machinist.
At age 18 he invented a new type of steam boiler, and later worked for the
Corliss Machine Works that produced the great Corliss steam engines. By age
24, he was already experimenting with a concept for an internal combustion
engine that could run on liquid fuels; he would work on the idea for 18 years
before receiving a patent on it in 1872 for the Ready Motor gas engine. The
Patent Office identifies this, 2-cycle engine as a hot-air engine that ran quietly
with kerosene. Brayton's engine was an interesting one. It used two cylinders,
connected, with the pistons running in opposite phase. One was the compression cylinder, which compressed the fuel-air mixture to a somewhat higher
pressure than the pressure in the power cylinder. Introducing the new principle
of fuel injection, it pumped the combustible mixture into the power cylinder,
where it was continuously ignited and burned during the power stroke, keeping
the pressure up in the cylinder as the piston was displaced, thus accomplishing
work per unit of fuel. However, much of the efficiency gained by this method was
lost due to the lack of an adequate method of compressing the fuel mixture prior
to ignition. The power cylinder, operating at a slightly lower pressure than the
compression cylinder, was quite a bit larger. Although this engine was not very
successful, it was considered the first safe and practical oil engine.
These engines were commercially available gas or oil burning hot air designs
from which the Brayton, or isobaric combustion, cycle originated.
A gas turbine, if you think about it, operates much the same way. This constantpressure combustion cycle is known by engineers as the Brayton cycle, though
few could draw a picture of a Brayton engine.
Another source reports: George Bailey Brayton was an inventor of engines. He
constructed a number of different patterns of these engines, some of which he
put into small boats or launches; they were the primitive naphtha launches now
in general use. Probably the most highly finished engine he ever built was sent
to the Centennial at Philadelphia, afterwards it ran the shop on Potter Street,
Providence, RI, and later it was sent to Sayles' Bleachery; he invented an eyelet
and rivet machine, The patents of these machines were probably the most
remunerative of any he ever obtained, netting him nearly fifty thousand dollars.
He went to England on business in Oct., 1892, and while there died; his body
was brought home in 1893. His home was Boston; his family still resides there."
Birth:
1839, East Greenwich, Kent, Rhode Island
Death:
Bet Oct 1892 and 1893, Leeds, West Yorkshire,
England
Father:
William H. Brayton
Mother
Minerva Bailey
Married:
Rhonda V. Dean, 23 Oct 1862 In Providence, RI
Daughter:
Mavelle Clifton Brayton
Evolution of the Internal-Combustion Engine
Braytons engine was displayed at the 1876 Philadelphia Centennial Exhibition.
Although more impressive steam engines were displayed, Braytons engine
pointed to the future. The Otto & Company engine, patented in 1876 was not
ready in time to be displayed at Philadelphia. Ottos engine for first time placed
internal combustion on a soundly competitive footing with steam power. It was
on display in Paris in 1878.
Inspired by Braytons mammoth internal combustion engine at the Centennial
Exposition, George B. Seldon (inventor and lawyer, 1846-1922) began working
on a smaller lighter version, succeeding by 1878 in producing a one cylinder, 2
HP, 370 pound version which featured an enclosed crankshaftthe Road
Engine. He filed for a patent in 1879not just for the engine but the entire
concept of an automobile. Through legal maneuvers, this was granted in 1895
poised to reap royalties from the fledgling American auto industry. George
Selden, despite never actually producing a working model of an automobile,
had a credible claim to have patented the automobile.
The first person to experiment with an internal-combustion engine was the
Dutch physicist Christian Huygens, about 1680. But no effective gasolinepowered engine was developed until 1859, when the French engineer J. J.
tienne Lenoir built a double-acting, spark-ignition engine that could be
operated continuously. In 1862 Alphonse Beau de Rochas, a French scientist,
patented but did not build a four-stroke engine; sixteen years later, when
Nikolaus A. Otto built a successful four-stroke engine, it became known as the
Otto cycle. The first successful two-stroke engine was completed in the same
year by Sir Dougald Clerk, in a form which (simplified somewhat by Joseph Day
in 1891) remains in use today. George Brayton, an American engineer, had
developed a two-stroke kerosene engine in 1873, but it was too large and too
slow to be commercially successful. In 1885 Gottlieb Daimler constructed what
is generally recognized as the prototype of the modern gas engine: small and
fast, with a vertical cylinder, it used gasoline injected through a carburetor. In
1889 Daimler introduced a four-stroke engine with mushroom-shaped valves

and two cylinders arranged in a V, having a much higher power-to-weight ratio;


with the exception of electric starting, which would not be introduced until 1924,
most modern gasoline engines are descended from Daimler's engines.
Evolution of the Turbine Engine
John Barber received the first patent for a turbine engine in England in 1791.
His design was for propelling a 'horseless carriage.' The turbine was designed
with a chain-driven, reciprocating type of compressor. It had a compressor, a
combustion chamber, and a turbine. The gas-turbine engine was first
successfully tested by F. Whittle in 1937, and first applied by the Heinkel Aircraft
Company in 1939. Today, gas-turbines are used by practically all aircraft except
smaller ones, by many fast boats, and increasingly been used for stationary
power generation, particularly when both power and heat are of interest.

Braytons Ready Motor


Exhibit Title:

Brayton, Geo. B., Philadelphia, Pa., Exhibit #590b,


Machinery Hall,
Bldg. #2.George B. Brayton's hydro-carbon Ready
Motor engine.
Centennial Photographic Co. / Centennial Exhibition Digital Collection
http://genweb.whipple.org/d0041/I55267.html
http://libwww.library.phila.gov/CenCol/cedcimgview.taf?CEDCNo=c012140
Culp, John S., M.D., http://www.atis.net/stationary-engine/digest/v03.n502
http://imartinez.etsin.upm.es/bk3/c17/Power.htm
http://www.asme.org/history/brochures/h135.pdf
http://www.nationmaster.com/encyclopedia/George-B.-Selden
http://personalwebs.oakland.edu/~leidel/SAE_PAPER_970068.pdf
http://www.infoplease.com/ce6/sci/A0858862.html
http://www.asme.org/history/biography.html#Brayton
http://www.cre8tivenergy.com/brayton.html
http://www.hw.ac.uk/mecWWW/research/whm/term2_2000/part2.PDF
http://www.maritime.org/fleetsub/diesel/chap1.htm

Last Rev.: 17 JUL 08

BRAYTON CYCLEJET ENGINE : MIME 3470

Page 13