40
Undergraduate Team Engine Student Design Competition 2014/15 An Ultra-High Bypass Ratio Turbofan Engine for the Future - Request for Proposal - September 13, 2014

Undergraduate Team Engine Student Design Competition 2014 ... · Undergraduate Team – Engine Student Design Competition 2014/15 ... a new LP/IP system that fits around it, ... Number

Embed Size (px)

Citation preview

Undergraduate Team – Engine

Student Design Competition 2014/15

An Ultra-High Bypass Ratio Turbofan

Engine for the Future

- Request for Proposal -

September 13, 2014

2

Abstract

Major engine manufacturers are continually assessing and revising their technical & business

plans to ensure that their vision reaches into the next decade. In the commercial aviation market,

replacement engines for new generations of the Boeing 787 and Airbus A380 & A350 airplanes

are currently being considered. Very recently Rolls-Royce revealed its road map for the future(*)

,

whereby it will extend its Trent 1000 and Trent XWB engine programs to address significantly

higher bypass ratios, further improvements in propulsive efficiencies at cruise and reduced fuel

burn & emissions for long range travel in 2025 and beyond. This Request for Proposal asks that

you also look to 2025 and design a new geared, 3-spool, high bypass ratio turbofan for entry into

service around that time for use on twin-engine, wide-body passenger and freight aircraft. Your

primary objective is also reduced fuel burn, as a result of higher propulsive efficiency at cruise

conditions.

A generic model, representative of 3-spool current systems is supplied as a baseline engine. This

model has been generated solely on the basis of publically-available information. You should

model this engine with your design system to provide a viable reference from which to gauge

your improvements. You are then required to retain the core of the baseline design and generate

a new LP/IP system that fits around it, using aerodynamic similarity. A simple but typical,

multi-segment, extended mission should be constructed that covers both design-point and off-

design engine operations. Such a mission will also test propulsive efficiencies at cruise and

reduced fuel burn specifically and should be “flown” using both engines. The performance

characteristics and total fuel consumption of both engines should be estimated over the mission

and stated clearly in the proposal. The benefits of the new design should be clearly stated.

Special attention should be paid to engine mass, dimensions & integration with the aircraft.

Technical feasibility is critical and operating costs should also be considered.

Dr. Ian Halliwell

AIAA Air Breathing Propulsion Group and IGTI Aircraft Engines & Education Committees

Principal Engineer, PSM-Alstom

E-mail: [email protected]

(*)

Aviation Week and Space Technology, August 25, 2014.

3

CONTENTS

Page

1. Introduction 4

2. Design Objectives & Requirements 6

3. Baseline Engine Model 8

3.1 Overall Characteristics 9

3.2 Inlet 17

3.3 Fan 17

3.4 Intermediate-Pressure Compressor 19

3.5 Inter-Compressor Duct 20

3.6 High-Pressure Compressor

20

3.7 Combustor 22

3.8 High-Pressure Turbine 22

3.9 Inter-Turbine Duct 1 25

3.10 Intermediate-Pressure Turbine 25

3.11 Inter-Turbine Duct 2 28

3.12 Low-Pressure Turbine 28

3.13 Core Exhaust & Nozzle 31

3.14 Bypass Duct

32

4. Hints & Suggestions 32

5. Competition Expectations 33

References 34

Suggested Reading 35

Available Software & Reference Material 35

Appendix 1. Letter of Intent 37

Appendix 2. Rules and Guidelines 38

I. General Rules 38

II. Copyright 39

III. Schedule & Activity Sequences 39

IV. Proposal Requirements 39

V. Basis for Judging 40

4

1. Introduction

The A350-900 is the first Airbus A350 model and seats 314 passengers in a three-class cabin and

9-abreast layout. It has a standard design range target of 15,000 km (8,100 nmi). The −900 is

designed to compete with the Boeing 777-200ER and replace the Airbus A340-300. A −900R

variant, which has been proposed but not yet launched, would feature higher engine thrust and a

strengthened structure.

Figure 1: Rolls-Royce Trent XWB Engines on the Airbus A350

The range of the A350-900R is estimated to be 17,600 km (9,500 nmi), which would be boosted

to about 19,100 km (10,315 nmi) by these design improvements to compete with the Boeing 777-

200LR and be capable of non-stop flight from London-Heathrow to Auckland. Rolls-Royce

agreed with Airbus to supply a new variant of the Trent engine - correspondingly named the

Trent XWB - for the A350 XWB aircraft. After low-speed wind tunnel tests, Airbus froze the

static thrust at sea level for all three proposed variants in the 330–420 kN (74,000–94,000 lbf)

range in 2010. This Request for Proposal is aimed at future engines for this type of aircraft and

for this type of mission.

5

Figures 1 and 2 are from the Airbus and Rolls-Royce websites respectively and show the A350

XWB aircraft and the installed engine.

Figure 2: A Closer View of the Rolls-Royce Trent XWB Engine Installation

General characteristics

Capacity 314 passengers (3-class)

Length 66.89 m

Wing span 69.8 m

Height 17.05 m

Wing area 443 m²

Max. take-off weight 268 t

Power plant 2 × high bypass ratio turbofans; 374 kN each at take-off

Performance

Cruise speed Mach 0.85

Range 17,600 km

Service ceiling 12.19 km

Table 1: General Characteristics of the Airbus A350-XWR Aircraft

6

Some typical aircraft characteristics are given in Table 1. At take-off, the total thrust needed

from each of the two engines is 374 kN. Table 2 contains a summary of basic engine

characteristics, taken mostly from References 1 and 2. This data provides target values for the

baseline engine model. It is emphasized that the model is intended to be only a rough generic

representation of the Rolls-Royce Trent XWB purely for the purposes of this exercise!

Design Features of the Baseline Engine

Engine Type Axial, turbofan

Number of fan/booster/compressor stages 1, 8, 6

Number of HP/LP turbine stages 1, 2, 7

Combustor type Annular

Maximum net thrust at sea level 400.4 kN

Specific fuel consumption at cruise at Mach 0.85 & 12.19 altitude 18.00 g/kN.s

Overall pressure ratio at max. power 50.0

Bypass ratio 9.3

Max. envelope diameter 2.997 m

Max. envelope length 4.064 m

Dry weight less tail-pipe 5,445 kg

Table 2: Baseline Engine Design Targets: Basic Data, Overall Geometry & Performance

The actual baseline engine model is described in some detail in Section 3.

2. Design Objectives & Requirements

A new engine design is required for future versions of the Airbus A350 and Boeing 777 and

787, with an entry-into-service date of 2025. The new engine should include a geared

fan. An explanation should be given for why a gear is needed, along with the pros and

cons!

The current flight envelope ranges from take-off at static sea-level conditions to cruise at

12,190 m/Mach 0.85. This is to be retained for the new engine, so these two flight

conditions should be used as the principal design points for candidate engines. Maximum

potential take-off thrust should match that of the baseline engine described later and the

actual take-off thrust given in Table 1 should be assumed. The range target outlined in

7

the Introduction should be borne in mind and it is hoped that the endurance might be

extended with the new derivative engine by reducing the fuel consumption and

minimizing engine mass.

In the baseline engine model, a nominal power off-take of only 50 kW has been assumed.

This is too low and should be increased to 200 kW.

The generic baseline engine model should be used as a starting point. The core of the

engine (HP compressor, combustor & HP turbine) should be retained and a new LP/IP

system should be designed around it. The bypass ratio should be increased to 15 by using

a geared fan. The overall pressure ratio should be increased to 60.

A turbine inlet temperature of 1784K has been assumed in the baseline engine. In the

new design, based on the entry into service date, assume that advances in materials and

cooling technology permit a T4 limit of 1930K. The development and potential

application of carbon matrix composites is of particular interest. Based on research of

available literature, justify carefully your choices of any new materials, their location

within the engine and the appropriate advances in design limits that they provide.

Aerodynamic similarity should be used to ensure compatibility between conditions at the

IPC exit and HPC inlet. The new engine design should be optimized for minimum

engine mass & fuel burn, based on trade studies to determine the best combination of fan

& intermediate compressor pressure ratios, bypass ratio, overall pressure ratio and turbine

entry temperature. Values of these four major design parameters should be compatible

with those expected to be available in 2025 and the selected design limits should be

justified in the proposal.

For the gear, assume a mass of 0.0036 kg per hp transmitted.

Design proposals must include engine mass, engine dimensions, net thrust values,

specific fuel consumption, thermal and propulsive efficiencies at take-off (standard sea-

level conditions) and cruise. Details of the major flow path components must be given.

These include inlet, fan, IP Compressor, HP compressor, combustor, HP turbine, IP

turbine, LP turbine, exhaust nozzle, bypass duct, and the connecting ducts.

Since reduced specific fuel consumption does not necessarily lead to reduced fuel

consumption should the new engine be heavier, the fuel burn over an assumed mission

must be determined by dividing it into suitable segments in terms of time at altitude and

Mach number and summing the incremental fuel burn estimates. This should be done for

both the baseline engine and the new derivative to determine the improvement.

8

3. Baseline Engine Model

As stated previously, the baseline engine is a 3-spool, high bypass ratio turbofan. A generic

model has been generated from publically-available information (References 1 & 2) using

GasTurb12. Some details of the baseline model are given below to assist with construction of

your baseline case and to provide some indication of typical values of design parameters.

Figure 3: An Unmixed, High Bypass Ratio Turbofan Engine Schematic

with Calculation Stations & Nominal Cooling Flows

Figure 3 contains a general schematic with relevant station numbers.

9

3.1 Overall Characteristics

Major Design Parameters

In a turbofan engine, the four primary design variables are turbine entry temperature (T4), overall

pressure ratio (OPR or P3/P2), fan pressure ratio (FPR or P21/P2) and bypass ratio (BPR). We

usually differentiate between the fan pressure ratios in the core & bypass streams. In a 3-spool

engine an additional variable is introduced in the form of pressure ratio generated by the

intermediate pressure compressor, so an optimum IPC/HPC pressure ratio split must therefore be

determined.

Table 3: Basic Input

Table 3 is the “Basic Input” for the GasTurb12 model of the baseline engine and the five primary

design variables are specified. To generate an acceptable replica of the engine, a unique

combination of the remainder must be estimated iteratively using performance figures which are

provided – namely the net thrust (FN) and specific fuel consumption (sfc) at cruises conditions -

as targets. Since the sfc target is not at the engine design point, this can only be checked

10

periodically once what is thought to be a satisfactory design point solution has first been

obtained.

Table 3 also contains some of the secondary inputs, while the remainder are addressed below.

The first row of Table 3 assumes negligible total pressure loss between the inlet leading edge and

the fan face. The inner and outer fan pressure ratios are then selected separately; there is more

blade speed at the fan tip than at its hub, so the inner & outer fan pressure ratios have been set at

1.4 & 1.43 respectively – fairly aggressive but not unreasonable for a modern single-stage

machine. A zero total pressure loss is then accounted for in the duct between the fan and the IP

compressor or booster. This is probably optimistic but not too much so, as the prevailing Mach

number is quite low. Knowing that the required overall pressure ratio is 50.0, results in a

pressure ratio across the remainder of the compression system of 35.714, allowing for losses.

This is distributed between the booster and the HP compressor with 6.3 across the former (over 8

stages) and 5.76 across the latter (over 6 stages). A 2.5% total pressure loss is assumed in the

bypass duct. Inter-turbine duct losses of 0.8% and zero have been used – again somewhat

optimistic!

Continuing with the input description, the design bypass ratio was set at 9.3. A value of 1783.33

K for the turbine exit temperature was taken as being reasonable for this engine with limited

cooling capacity and an expected long life for the HP turbine (say 5,000 hours). The temperature

is a guessed value as, understandably, engine manufacturers do not reveal such critical

information. In fact, this value of T4 was the result of an iterative process that involved

turbomachinery efficiencies and the target thrust. The next four parameters relate to the primary

combustor; they are all fairly conventional values by modern standards. The burner “part load

constant” is an element in the calculation of burner efficiency that is discussed in the GasTurb12

User Guide in Reference 3. Without expert knowledge, this is best left alone! The remaining

parameters in Table 3 may be considered as secondary influences and are discussed briefly

below.

Secondary Design Parameters

Cooling Air: Mention has already been made of bleed and cooling air flows – the

secondary flows. Only the overboard bleed is listed in Table 3 (although this is in fact

zero), however the secondary flows indicated in Figure 2 have been set via another “air

system” tab on the input screen as fractions of W25, the HP compressor entry flow.

Pressure Losses: A number of total pressure losses, mentioned earlier, are also specified

in Table 3 by inserting the appropriate pressure ratios across the inter-compressor duct,

the inter-turbine duct, the mixer and the primary combustor.

Turbomachinery Efficiencies: Efficiencies of the fan, HP compressor, HP turbine and

LP turbine are entered via their respective tabs on the input screen. The values are not

listed specifically in Table 3, but may be reviewed in the output summary presented later

in Table 4. The designer has the choice of either isentropic or polytropic values, so he or

she should be certain of their applicability and their definitions! Both values appear in

11

the output summary in Table 4. However, another option is available that has been used

here for both compressors & fan and turbines. It allows GasTurb12 to estimate turbine

efficiencies from data supplied – via values of stage loading and flow coefficients. For

turbines these values are used in a Smith Chart (Reference 4), assuming an equal work

spilt between stages. It is recommended that either this be used or initial values be taken

from Table 4.

Power Off-take: All engines have power extracted – in the Trent XWB it is taken from

the IP spool via a bevel gear and a tower shaft that passes through an enlarged vane or

strut in the frame between the IP and HP compressors. This is often preferred to the use

of a separate auxiliary power unit, depending on how much power is required for

airframe use. In the application currently under consideration, considerable auxiliary

power may be needed for avionics and passenger equipment and this usage is growing

rapidly in modern aircraft. We have selected a nominal power off-take of 50 kW from

our baseline engine but 200kW has been requested in Design Objectives in Section 2.

A limited study has been made of the influence of a number of secondary parameters and it was

determined that the default values present in the GasTurb12 generic model should be retained,

based on the known expertise of the author of the code.

Dimensions: Diameters & Lengths

The engine cycle may be defined purely on the basis of thermodynamics. Geometrically, we

define a “rubber engine” initially - where performance is delivered in terms of a net thrust of

400.50 kN given in Table 4 once the engine scale has been determined. We also have a target

dimensional envelope to fit into, namely a maximum casing diameter of 2.997 m and length of

4.064 m, although the latter is very much open to interpretation. The diameter can be determined

via the mass flow rate; the length is a separate issue that is dealt with by manipulation of vane &

blade aspect ratios and axial gaps in the turbomachinery and by suitable selection of duct lengths,

usually defined as fractions of the corresponding entry radii. Once the correct thrust has been

reached, the maximum radius is determined by setting an inlet radius ratio and then varying the

Mach number at entry to the fan. These values are input on the primary input screen under the

LP compressor tab, where a Mach number of 0.574 was combined with a fan inlet radius ratio of

0.283 and a fan tip speed of 411.18 m/s were found to be appropriate. This sets the general

radial dimension for the complete engine, although in fact downstream of the fan, the entry radii

of the IP and HP compressors are determined independently. The HP & LP turbine radii follow

from the exit values of the respective upstream components. For the ducts, radial dimensions are

keyed off the inner wall with the blade spans being superimposed. For the overall engine length,

early adjustments are made by eye (My personal philosophy is that if it looks right, it probably is

right!), with final manipulations being added as the target dimension is approached. The fan

diameter turned out to be 3.02 m (compared to the target value of 2.997 m in Table 2.) while the

overall diameter of the engine model is 3.558 m, which allows for the thickness of the nacelle.

The engine model length of 7.944 m includes the exhaust system and cone so this is deemed to

be satisfactory. The “target” length of 4.064 m in Table 2 may be interpreted as a “flange-to-

12

flange” length that extends from the fan face to just downstream of the rear frame, just aft of the

LP turbine in the plot or our engine model in Figure 5. The target value was in fact taken from

the Trent 1000, as the Trent XWB data was not available in Reference 2. (We are not cutting

metal here folks, so we are probably OK!)

Table 4: A Summary of the Baseline Engine Model

Materials & Weights

As far as possible, use was made of the materials database in the GasTurb12 design code. For

proprietary reasons many advanced materials are not included. Examples of these are: polymeric

composites used in cold parts of the engine, such as the inlet and fan; metal matrix composites,

which might be expected in the exhaust system; carbon-matrix-composites, again intended for

use in hot sections. All of these materials are considerably lighter than conventional alternatives,

although it should be noted they may not yet have found their way into the baseline engine,

where long life and reliability are critical. However, within the component models, material

13

densities can be modified independently of the database and I have taken advantage of this

feature in some cases where I believe that “advanced” materials of lower density are appropriate.

Use has also been made of the materials data in Reference 5, interpolating and extrapolating

where necessary.

In GasTurb12 component weights are calculated by multiplying the effective volumes by the

corresponding material densities. Of course, only the major elements which are designed

directly are weighed and there are many more constituents. Nuts, bolts, washers, seals and other

much larger elements such as fuel lines, oil lines, pumps and control systems still must be

accounted for. In the engine industry, this is done usually, at the preliminary design stage, by the

application of a multiplier or adder whose value is based on decades of experience. In general, a

multiplication factor of 1.3 is recommended in the GasTurb12 manual, but for an engine as large

as the Trent XWB I reduced this to a “net mass factor” of 1.15 in Table 5 mainly because it got

me closer to the gross engine weight I was looking for! The total mass of the engine shown in

Table 5 (5,576.61 kg) is 2.4% over the 5,445 kg target in Table 2, but it should be remembered

that the “tail pipe” is not accounted for in the latter and in our model the core nozzle weighs

110.27 kg when the mass factor has been applied. Conveniently, this accounts for most of the

discrepancy!

A summary of the baseline engine model is presented in Table 4 and Table 5 is a more detailed

“Overall Output Table”.

14

Table 5: Baseline Engine Detailed Output

A cutaway of the baseline engine is shown in Figure 4.

15

Figure 4: A Cutaway View of the Rolls-Royce Trent XWB Engine

A plot of the GasTurb12 baseline engine model appears in Figure 5.

16

Figure 5: GasTurb12 Model of the Rolls-Royce Trent XWB - the Baseline Engine

Some details of the component models now follow.

17

3.2 Inlet

The inlet is designed with a conical center body (Figure 5). In practice, a single-stage fan can be

cantilevered from a bearing located in the main frame of the engine. The outer diameter of the

inlet has been determined from that of the fan.

Table 6: Inlet Design

Pertinent characteristics of the inlet are shown in Table 6. At 109.725 kg, the inlet is fairly light

and this is because, based on the density, we have taken a typical Ti-Al alloy as our choice of

materials. It is noteworthy that the GasTurb “inlet” is merely the portion of the casing (plus

center body) immediately upstream of the fan. The GasTurb12 model begins at the “upstream

flange”, which is located further forward of the central cone than shown in the real engine in

Figure 4.

3.3 Fan

Table 7: Fan: Detailed Overview

18

The fan characteristics are given in Tables 7 & 8. The radius ratio and inlet Mach number are of

particular interest because, when taken with mass flow rate, they define the fan tip radius. Based

on tip radius, the blade tip speed sets the rotational speed of the LP spool. The value of corrected

flow per unit area (197.18 kg/m2 or 40.37 lbm/ft

2) is fairly conventional and corresponds to the

input value of Mach number (0.574).

Table 8: Fan General Output

On September 12, 2014 three new parameters were added to the LPC input of GasTurb12 to

control the inlet duct to the IPC. The new inputs are indicated in red in Table 8. If necessary, an

update to the code should be acquired by users.

19

3.4 Intermediate-Pressure Compressor

Inputs for the intermediate pressure compressor are provided in Tables 7 & 8. To maintain

access to the engine geometry and plot, it may be necessary to s switch to the “efficiency known”

option and insert the estimated isentropic value.

Table 7: Intermediate Compressor - Detailed Overview

Table 9: Intermediate Compressor - General Output

20

3.5 Inter-Compressor Duct

Input and output for the inter-compressor duct are given in Table 10.

Table 10: Inter-Compressor Duct

Notice that in addition to using an overall net mass factor to adjust the engine weight, individual

net mass factors may be applied to the components or net mass adders may be used, although this

remains at a value of unity for the inter-compressor duct since very little of the structure is left

unaccounted for in the simple model.

3.6 High Pressure Compressor

Table 11: High Pressure Compressor - Detailed Overview

Again, we set the speed of the HP spool via the tip speed and the corresponding radius. The

general characteristics of the HP compressor are given in Table 11. Input and output parameters

are shown in Table 12.

21

Table 12: High Pressure Compressor - General Output

22

3.7 Combustor

A fairly conventional annular combustor is used and details are given in Table 13. The high

density of its material corresponds to the necessary thermal properties. The combustor is a major

structural component, linked closely to the HP turbine first vane assembly.

Table 13: Combustor

3.8 High-Pressure Turbine

Table 14: High Pressure Turbine – Basis for Efficiency Estimate

As stated in Section 3.1, the efficiency of the high pressure turbine was estimated by GasTurb12

on the basis of the data shown in Table 14, which is made available once that efficiency option is

selected. As a result of that selection, the details of the HP turbine in Table 14 appear.

23

Table 15: HPT Summary

A general summary of the HP turbine is given in Table 16, followed by the velocity diagrams

and Smith Chart in Figure 6.

24

Table 16: High Pressure Turbine – General Output

Figure 6: High Pressure Turbine Velocity Diagrams & Smith Chart

25

3.9 Inter-Turbine Duct 1

Table 17 contains details of the inter-turbine duct between the HP and IP turbines. Its relatively

short length allows the two turbines to be close-coupled and the exit-to-inlet radius ratio of 1.1

emphasizes this. The intermediate shaft rotates counter to those of the LP and HP systems

although this is not indicated in the velocity diagrams shown here.

Table 17: Inter-Turbine Duct 1

3.10 Intermediate-Pressure Turbine

Table 18: Intermediate Pressure Turbine – Basis for Efficiency Estimate

Table 18 contains the input data used when the option to calculate turbine efficiency is selected.

The warning on the high exit radius ratio appears because the value is beyond conventional

limits but it is due to the high value of bypass ratio and the relatively small size of the inner

engine flowpath. To maintain access to the engine geometry and plot, it may be necessary later

to switch to the “efficiency known” option and insert the calculated isentropic value.

26

Table 19: Intermediate Pressure Turbine Summary

As a result of the efficiency calculation option, Table 19 appears in the IP turbine output. The

stage loading coefficient is fairly conventional but the stage flow coefficient is quite high1.

These observations are reflected in the velocity diagram in Figure 7. The rotational speed of the

IP spool was set primarily by turbine disk stress considerations, but an increase in axial velocity

could have improved IP turbine performance. Additional input and output characteristics of the

IP turbine are given in Table 20.

1 Loading coefficient (Ψ) = ΔH/U

2. Flow coefficient (Φ) = Vax/U.

27

Table 20: Intermediate Pressure Turbine – General Output

Figure 7: Intermediate Pressure Turbine Velocity Diagrams & Smith Chart

28

3.11 Inter-Turbine Duct 2

Table 21: Inter-Turbine Duct 2

Table 21 contains input and output information for the second inter-turbine duct between the IP

and LP turbines. The exit/inlet radius ratio increases the radial location of the LP turbine and

results in higher blade speeds, lower loading coefficients and hence improved efficiencies.

3.12 Low-Pressure Turbine

Characteristics of the low pressure turbine are presented in Tables 22 - 24 and Figure 8. Figure

8 contains velocity diagrams for the first and last stages. The flared nature of the LP turbine

flowpath ensures that meanline radii are maximized, stage loading coefficients are minimized

and stage efficiencies are fairly. However, it may be seen from Figure 8 that the common design

point for all seven stages is too far to the left on the Smith Chart due mainly to the high mean

blade speed and improvements in the form of higher efficiency and smaller disks could be

obtained by reducing rpm. It should be noted that the efficiency contours in Figure 8 (and

Figure 7 & 9) are expressed as fractions of the maximum value on the chart! The true value of

the average stage efficiency is 91.89%, which corresponds to the value in the engine

performance summary in Table 4.

Table 22: Basis for LP Turbine Calculated Efficiency

29

Table 23: LPT Summary

30

Table 24: Low Pressure Turbine: General Output

Figure 8: Low Pressure Turbine Velocity Diagrams & Smith Chart

31

3.13 Core Exhaust & Core Nozzle

The core exhaust is directly downstream of the low pressure turbine. It is comprised of an outer

casing, an inner casing, and an inner cone that closes off the casing, and a strut or frame. In

Figure 5 on page 16, the core exhaust extends to about 5.6 m. The “core exhaust” in GasTurb12

does not include the convergent portion or the core nozzle. Table 25 contains the input and

output details of the core exhaust while Table 26 covers the remainder, termed the “core nozzle”.

Table 25: Core Exhaust

The core nozzle is the part of the engine that converges to its exit area at about 6.33 m in Figure

4. The casing material density in the core nozzle is the same as that for the core exhaust,

although a lighter material most likely could have been used owing to the local temperatures.

Table 26: Core Nozzle

The cone ends in the exhaust duct

32

3.14 Bypass Duct

Table 27 defines the input and output parameters for the bypass duct. The shape and geometric

continuity of the bypass duct with adjacent structures depends critically on the values of the

parameters indicated by the blue box.

Table 27: Bypass Duct

4. Hints & Suggestions

You should first model the baseline engine with the same software that you will use for

your new engine design. Your results may not match the generic baseline model exactly

but will provide an essential starting point for a valid comparison of weights and

performance for your new engine.

In general, subsonic commercial engines tend to be sized at take-off rather than at “top-

of-climb” (the beginning of cruise). However, since the major objective in this exercise

is to minimize fuel burn at cruise – where most of the fuel will be burned – it is essential

that off-design performance (particularly for the turbines) be given special attention.

The efficiencies of the turbomachinery components may be assumed to be the same as

those of the baseline engine, and be input directly or the “calculate efficiency” mode of

GasTurb12 may be invoked.

This is not an aircraft design competition, so credit will not be given for detailed

derivation of aircraft flight characteristics, but some reasonable assumption should be

made - and clearly stated - concerning the thrust needed by the airplane compared to the

engine capabilities at a particular Mach number and altitude.

33

The use of design codes from industrial or government contacts, that are not accessible to

all participating teams, is not allowed.

Even though the date for submission of Letters of Intent is stated as November 1, 2014 on

pages 37 and 39, it is recommended that teams who know that they will enter the

competition inform AIAA, ASME-IGTI or Dr. Ian Halliwell ([email protected])

as soon as possible, so that assistance may be given and access to design codes may be

arranged, where appropriate (See page 35).

Questions will be taken by volunteers from the AIAA Air Breathing Propulsion Technical

Group or the IGTI Aircraft Engines Technical Committee, whose contact information will be

provided to teams who submit a letter of intent.

5. Competition Expectations

The existing rules and guidelines for the Student Design Competition shall be observed and these

are provided in Appendix 2. In addition, the following specific suggestions are offered for the

event.

This is a preliminary engine design. It is not expected that student teams produce design

solutions of industrial quality, however it is hoped that attention will be paid to the practical

difficulties encountered in a real-world design situation and that these will be recognized and

acknowledged. If such difficulties can be resolved quantitatively, appropriate credit will be

given. If suitable design tools and/or knowledge are not available, then a qualitative description

of an approach to address the issues is quite acceptable.

In a preliminary engine design the following features must be provided:

Definition and justification of the mission and the critical mission point(s) that drive the

candidate propulsion system design.

Clear and concise demonstration that the overall engine performance satisfies the mission

requirements.

Documentation of the trade studies conducted to determine the preferred engine cycle

parameters such as fan pressure ratio, bypass ratio, overall pressure ratio, turbine inlet

temperature, etc.

An engine configuration with a plot of the flow path that shows how the major

components fit together, with comments on operability at different mission points.

34

A clear demonstration of design feasibility, with attention having been paid to

technology limits. Examples of some, but not all, velocity diagrams are important to

demonstrate viability of turbomachinery components.

Stage count estimates, again, with attention having been paid to technology limits.

Estimates of component performance and overall engine performance to show that the

assumptions made in the cycle have been achieved.

While only the preliminary design of major components in the engine flow path is expected to be

addressed quantitatively in the proposals, it is intended that the role of secondary systems such as

fuel & lubrication be given serious consideration in terms of modifications and how they would

be integrated in to the new engine design. Credit will be given for clear descriptions of how any

appropriate upgrades would be incorporated and how they would affect the engine cycle.

Each proposal should contain a brief discussion of any computer codes or Microsoft Excel

spreadsheets used to perform engine design & analysis, with emphasis on any additional special

features generated by the team.

Proposals should be limited to fifty pages, which will not include the

administrative/contents or the “signature” pages.

References

1. “Road Map: Rolls-Royce’s future turbofan strategy will leverage European, national and

company research.”

Aviation Week & Space Technology. August 25, 2014.

2. “Aerospace Source Book.”

Aviation Week & Space Technology. January 26, 2009.

3. “GasTurb 12: A Design & Off-Design Performance Program for Gas Turbines”

<http://www.gasturb.de>

Joachim Kurzke, 2012.

4. “A Simple Correlation of Turbine Efficiency”

S. F. Smith

Journal of the Royal Aeronautical Society. Volume 69. 1965.

5. “Aeronautical Vest Pocket Handbook”. Pratt & Whitney Aircraft. Circa 1980

35

Suggested Reading

1. “Gas Turbine Theory”

H.I.H Saravanamuttoo, G.F.C Rogers &.H. Cohen,

Prentice Hall. 5th

Edition 2001.

2. “Aircraft Engine Design”

J.D.Mattingly, W.H. Heiser, & D.H. Daley

AIAA Education Series. 1987.

3. “Elements of Propulsion – Gas Turbines and Rockets”

J.D. Mattingly.

AIAA Education Series. 2006.

4. “Jet Propulsion”

N. Cumpsty.

Cambridge University Press. 2000.

5. “Gas Turbine Performance”

P. Walsh & P. Fletcher.

Blackwell/ASME Press. 2nd

Edition, 2004.

6. “Fundamentals of Jet Propulsion with Applications”

Ronald D. Flack

Cambridge University Press. 2005.

7. “The Jet Engine”

Rolls-Royce plc. 2005.

8. “Aircraft Propulsion” 2nd

Edition.

Saeed Farokhi.

John Wiley & Sons Ltd. 2014.

Available Software & Additional Reference Material

GasTurb 12 is a comprehensive code for the preliminary design of propulsion and industrial gas

turbine engines (Reference 3). It encompasses design point and off-design performance, based

on extensive libraries of engine architectures and component performance maps, all coupled to

impressive graphics. A materials database and plotting capabilities enable a detailed engine

model to be generated, with stressed disks and component weights. A student license for this

code is available at a very low price directly from [email protected] strictly for academic work

only.

36

AxSTREAM is the first design & analysis code that permits the topic of propulsion and power

generation by gas & steam turbine to progress beyond velocity diagrams in the course of

university class. A suite of compressor and turbine modules cover the design steps from

meanline and streamline solutions to detailed design of airfoils. Use of this code is also

supported fully by excellent graphics. SoftInWay Inc. recently announced the availability of

AxSTREAM Lite to students that covers the design of turbines. However, an expanded license

will be provided to participants in the Undergraduate Team Engine Design Competition that also

includes fans and compressors for an appropriate time period prior to submission of proposals.

Once a Letter of Intent has been received, the names of team members will be

recognized as being eligible to be granted access to the AxSTREAM software.

Students must then apply to SoftInWay Inc. SoftInWay will not contact team

members.

GSP is NLR's (www.nlr.nl) primary gas turbine performance simulation tool

(www.GSPteam.com). It is a component based modeling environment based on a flexible object-

oriented architecture that allows modelers to simulate steady-state and transient performance of

virtually any gas turbine configuration using a user-friendly drag-and-drop interface. GSP has

been used for a variety of applications such as various types of off-design performance analysis,

emission calculations, control system design and diagnostics of both aircraft and industrial gas

turbines. All team managers or supervisors of the competing design teams are welcome to

request a free team license.

The offers above are subject to ITAR restrictions.

37

Appendix 1. Letter of Intent

Undergraduate Team Engine Design Competition 2014/15

Request for Proposal: An Ultra-High Bypass Ratio Turbofan Engine for the Future

Title of Design Proposal: _________________________________________________________

Name of School: _______________________________________________________________

Designer’s Name AIAA or ASME Graduation Date Degree

______________________ ______________ ______________ _________________

Team Leader

______________________ ______________ ______________ _________________ Team Leader E-mail

________________________ ________________ ________________ ___________________

________________________ ________________ ________________ ___________________

________________________ ________________ ________________ ___________________

________________________ ________________ ________________ ___________________

AIAA Foundation will act as the administrator for this competition.

In order to be eligible for the 2014/2015 Undergraduate Team Engine Design Competition, you

must complete this form and return it electronically to the AIAA Student Programs Coordinator,

Rachel Andino ([email protected]) before November 1, 2014, at AIAA Headquarters, as noted in

Appendix 2, Section III, “Schedule and Activity Sequences.”

Signature of Faculty Advisor Signature of Project Advisor Date

Faculty Advisor – Printed Project Advisor – Printed Date

38

Appendix 2. Rules and Guidelines

I. General Rules

1. All undergraduate AIAA or ASME branch or at-large Student Members are eligible and

encouraged to participate.

2. Teams will be groups of not more than four students.

3. An electronic copy of the report in MS Word or Adobe PDF format must be submitted on a

CD or DVD to AIAA Student Programs. Total size of the file(s) cannot exceed 60 MB, which

must also fit on 50 double spaced, 12 point font pages when printed. The file title should

include the team name and/or university. A “Signature” page must be included in the

report and indicate all participants, including faculty and project advisors, along with their

AIAA or ASME member numbers. Designs that are submitted must be the work of the

students, but guidance may come from the Faculty/Project Advisor and should be accurately

acknowledged. Graduate student participation in any form is prohibited.

4. Design projects that are used as part of an organized classroom requirement are eligible and

encouraged for competition.

5. More than one design may be submitted from students at any one school.

6. If a design group withdraws their project from the competition, the team chairman must notify

AIAA Headquarters immediately.

7. Judging will be in two parts.

First, the written proposals will be assessed by a judging panel comprised of members of

AIAA and IGTI organizing committees from the industrial and government communities.

Second, the best three teams will be invited to present their work to a second judging

panel at a special technical session. The in person presentation will either be at the ASME

TurboExpo in Montreal, Canada in June 2015 or the AIAA Propulsion and Energy Forum

in Orlando, FL in July 2015. The results of the presentations will be combined with the

earlier scores from the proposals to determine first, second and third places.

8. Certificates will be presented to the winning design teams for display at their university and a

certificate will also be presented to each team member and the faculty/project advisor. The

finishing order will be announced immediately following the three presentations.

Certificates and recognition in a press release will be the only prizes for this

competition. There will be neither prize money nor travel assistance to attend

the final presentation.

39

II. Copyright

All submissions to the competition shall be the original work of the team members.

Any submission that does not contain a copyright notice shall become the property of AIAA. A

team desiring to maintain copyright ownership may so indicate on the signature page but

nevertheless, by submitting a proposal, grants an irrevocable license to AIAA to copy, display,

publish, and distribute the work and to use it for all of AIAA’s current and future print and

electronic uses (e.g. “Copyright © 20__ by _____. Published by the American Institute of

Aeronautics and Astronautics, Inc., with permission.).

Any submission purporting to limit or deny AIAA licensure (or copyright) will not be eligible

for prizes.

III. Schedule & Activity Sequences

Significant activities, dates, and addresses for submission of proposal and related materials are as

follows:

A. Letter of Intent – November 1, 2014

B. Receipt of Proposal – April 1, 2015

C. Proposal evaluations completed - April 30, 2015

D. Round 2 Proposal Presentations & Announcement of Winners – June or July 2015. See

the website for updates as to location of the final presentation.

The finished proposal must be received at AIAA Headquarters on or before the date specified

above for the Receipt of Proposal (Item B).

IV. Proposal Requirements

A technical proposal is the most important criterion in the award of a contract. It should be

specific and complete. While it is realized that all of the technical factors cannot be included in

advance, the following should be included and keyed accordingly:

1. Demonstrate a thorough understanding of the Request for Proposal (RFP) requirements.

2. Describe the proposed technical approaches to comply with each of the requirements specified

in the RFP, including phasing of tasks. Legibility, clarity, and completeness of the technical

approach are primary factors in evaluation of the proposals.

3. Particular emphasis should be directed at identification of critical, technical problem areas.

Descriptions, sketches, drawings, systems analysis, method of attack, and discussions of new

techniques should be presented in sufficient detail to permit engineering evaluation of the

proposal. Exceptions to proposed technical requirements should be identified and explained.

40

4. Include tradeoff studies performed to arrive at the final design.

5. Provide a description of automated design tools used to develop the design.

V. Basis for Judging

Round 1: Proposal

1. Technical Content (35 points)

This concerns the correctness of theory, validity of reasoning used, apparent understanding and

grasp of the subject, etc. Are all major factors considered and a reasonably accurate evaluation of

these factors presented?

2. Organization and Presentation (20 points)

The description of the design as an instrument of communication is a strong factor on judging.

Organization of written design, clarity, and inclusion of pertinent information are major factors.

3. Originality (20 points)

The design proposal should avoid standard textbook information, and should show independence

of thinking or a fresh approach to the project. Does the method and treatment of the problem

show imagination? Does the approach show an adaptation or creation of automated design

tools?

4. Practical Application and Feasibility (25 points)

The proposal should present conclusions or recommendations that are feasible and practical, and

not merely lead the evaluators into further difficult or insolvable problems.

Round 2: Presentation

Each team will have 30 minutes to present a summary of its proposal to the judging panel and

answer questions. In addition to the categories above, the presentations will be assessed for

clarity, effectiveness and the ability to sell the teams’ ideas. Scores from the presentation will be

added to those from the proposal. The presentation score will be adjusted so that it is worth 30%

of the overall value.