AFAPL-TR-75-71

; '' 'CAN-TYPE COMBUSTOR DESIGN FOR

;A LOW COST SUBSONIC RAMJET ENGINEA
'D

26A

CN EER OMUS ORDEINF

'i
AIR FOLO ROST ASUBOTIC LARAM ENGN

Approved for public release: dis tribution un!imited

'D D C
Uf,h .. 26 qr 7::

AIR FORCE AERO PROPULSION LABORATORY A

" AIR FORCE WRIGHT AERONAUTICAL LABORATORIES
Air Force Systems Command
Wrlght.Patterson Air Force Bose, Ohio 45433
 T,

NOTICE

When Government drawings, specifications, or other data are used for

any purpose other than in connection with a definitely related Government
procurement operation, the United States Government thereby incurs no
responsibility nor any obligation whatsoever; and the fact that the
Government may have formulated, furnished, or in any way supplied the
said drawings, specifications, or other data, is not to be regarded by
implication or otherwise as in any manner licensing the holder or any
other person or corporation, or conveying any rights or permission to
manufacture, use, or sell any patented invention that may in any way be
related thereto.

This report contains the results of an effort to design, fabricate,

and test a can-type combustor for the low cost subsonic ramjet. The work
was performed in the Ramet Engine Division of the Air Force Aero Propulsion
Laboratory, Air Force Systems Command, Wright-Patterson AFB, Ohio under
Project 3012, Task 301208, and Work Unit 30120829. The effort was con-
ducted by lLt Charles E. Hornbeck during the period April 1974 to April
1975.

This report has been reviewed by the Information Office, (ASD/OIP)

and is releasable to the National Technical Information Service (NTIS).
At NTIS, it will be available to the general public, including foreign
nations.

This Technical report has been reviewed and is approved for publication.

CHARLES E. HORNBECK, Lt, USAF

Project Engineer

FOR THE COMMANDER

WILLIAM E. suPP, Tec a

Strategic Missile PropulsiinA

Copies of this report should not be returned unless return is required by

security considerations, contractual obligations, or notice on a specific document.
 UNCLASSIFIED
SECURITY C. ASSIFICATION OF THIS PAGE (*%en Data FnIered)
" READ INSTRUCTIONS
REPORT DOCUMENTATION PAGE BEFORE COMPLETING FORM
... i" RER
' T-NUM-ER ..-.... 12.GOVT ACCS, RECIPENT'S CATALOG NUMBER

CAN-TYPE COMBUSTOR DESIGN FOR A LOW COST I echnlcalRet

l~~Apr T974 - 1 Apr I} a !1
SUBSONIC RAMJET ENGINE.

7. AUTHOR(.) . CONTRACT OR GRANT NUMBER(3)

- Charles E./Hornbeck ILt, USAF

M9
PERFORMING A GANI$ATION AM4E AND ADORESS -10 PROGRAM ELEMENT. PROJECT. TASK
Air orce Aero Propu sion Laboratory (RJA) /ER
Wright-Patterson AFB, OH 45433 AF

11. CONTROLLING OFFICE NAME ANI) ADDRESS . 12. REPORT DATE

Air Force Aero Propulsion Laboratory (RJA) '1 ..........

Wright-Patterson AFB, Or! 45433
14 MONITORING AGENCY NAME & ADDRESS(if different Iron Controtllng Office) IS. SECURITY CLASS. (of this report)

. ,Unclassified
.,a l". DECLASSIFICATION DOWNGRADING
SCHEDULE

16. DISTRI BdTTIOhiJAA&.4f Ns)

Approved for public release; distribution unlimited.

i7. DISTRIBU rION STATEMENT (of the abstfaec entered In Block 20, if dlfferent from Report)

I0. SUPP'LEMENTARY NOTES

i9. KEY WORDS (Continue on reverse tide If necesary and identify by block number)

Subsonic Ramjets
Can-Type Combustor
Ramjet Combustion Performance

20. ;X TRACT (Cnntinae nn rovers* side It necessary and Identify by block number)

This study investigates the feasibility of using a low cost can combustor
in a subsonic ramjet engine in the flight envelope of Mach 0.7 to 0.9,
altitude 20,000 to 30,000 feet. A simple can combustor without close fabri-
cation tolerances was designed and fabricated. The can was installed in a
subsonic ramjet which was ground tested in a direct-connect pipe set-up.
The combustion efficiency and thrust were determined at conditions ,-
representative of the flight envelooe. These results are compared with
DD ,JANm, 1473 EDITION OF I NOV 6 IS OBSOLETE
UNCLASS I AFEIdED
SE[CURITY CLASSFICATION OF THIS PAGE (*%*on Dole E£ntered)
 UNCLASSIFIED
SECJRIT CLASSIFICATION OF THIS PAGE(*?W... Det. Enter.d)

results of a model subsonic ramjet engine cycle analysis. The end result
of the program is a developed combustor system for a subsonic ramjet
which is an inexpensive and attractive propulsion device for one-mission
remotely piloted vehicles and target drones.

\I

- i

UNCLASSI FI ED I

SECURITY CLASSIFICA'ION OF THIS PAGEI'I., fire FflIoed)

AFAPL-TR-75-71

TABLE OF CONTENTS

SECTION PAGE
INTRODUCTION 1
1. GENERAL DESCRIPTION 1
2. STATEMENT OF THE PROBLEM 2
3. APPROACH 5
II CAN-TYPE COMBUSTOR DESIGN 7

1. GENERAL LAYOUT 7
2. PILOT STABILIZATION 10
3. FUEL-AIR MIXTURE DISTRIBUTION 14
4. FUEL-INJECTION 15
5. IGNITION 21
6. BASELINE DESIGN 21
III EXPERIMENTAL TESTS AND PROCEDURES 25

1. TEST OUTLINE AND TEST CONDITIONS 25

2. TEST INSTALLATION 27
3. INSTRUMENTATION 29
4. DATA REDUCTION AND CALCULATION METHODS 32

IV TEST RESULTS 43

1. SUMMARY 43
2. TESTS FOR COMBUSTION EFFICIENCY OPTIMIZATION 44

a. Effect of Main Fuel Injector Position 45

b. Effect of Pilot Fuel Nozzle Type 48
c. Effect of Pilot Fuel-Air Ratio 50

3. DESIGN CHANGES TO ACHIEVE COMBUSTOR OPERATION

AT INCREASED ALTITUDE 52
a. Baseline Pilot Can Blow-Off Data
Correlation 52
b. Combustor Design Changes 54
4. CAN COMBUSTOR PERFORMANCE 57
5. MODEL AND ACTUAL RAMJET THRUST PERFORMANCE 64

V CONCLUSIONS 70

REFERENCES 72

1
AFAPL-TR-75-71

LIST OF ILLUSTRATIONS

FIGURE PAGE
I Schematic Diagram of a Subsonic Ramjet Engine 3
2 Subsonic Ramjet Flight Envelope 3

3 Flameholder Types 4

4 The Low Cost Subsonic Ramjet 8

5 Cross-Section of the Low Cost Subsonic Ramjet with

a Can Combustor 9

6 Assembly Drawing of the Low Cost Subsonic Ramjet

with a Can Combustor 9
7 Pilot Recirculation Zone 12

8 Pilot Can Blow-Off Data Correlation 12

9a Equivalent Disk Diameter 16

9b Blow-Off Data Correlation 16

10 Variable Main Fuel Injectors 19
11 Pilot Fuel Injector Types 20

12 Can Flameholder and Shroud 22

13 Can Flameholder and Shroud 23

14 Subsonic Ramjet Test Set-up Schematic 28

15 Instrumentation Locations 31

16 Ideal Temperature Rise for Given Fuel-Air Ratio

and Inlet Temperature 36
17 Molecular Weight of Combustion Products for
Given Fuel-Air Ratio and Inlet Temperature 37

18 Ratio of Specific Heats of Combustion Products

for Given Fuel-Air Ratio and Inlet Temperature 38

19 Exit Nozzle Discharge Coefficient 39

20 Effect of Main Fuel Injector Position 46

ii
AFAPL-TR-75- 7 1

LIST OF ILLUSTRATIONS (Concluded)

FIGURE PAGE

21 Effect of Main Fuel Injector Position 47

22 Effect of Pilot Fuel Injector Type 49

23 Effect of Pilot Fuel-Air Ratio 51

24 Stability Limits of Baseline Design 53

25 Pilot Can Blow-Off D&ta Correlation 55

26 Shroud Extension 56

27 Stability Limits of Baseline Design with Shroud

Extension 56

28 First Stage Pilot Can Air Scoops 58

29 Stability Limits of Baseline Design with Shroud

Extension and Air Scoops 59

30 Combustion Efficiency vs. Engine Fuel-Air Ratio 60

31 Net Jet Thrust vs. Engine Fuel Flow 62

32 Total Pressure Recovery vs. Engine Fuel-Air Ratio 63

33 Model and Actual Thrust Performance 65

34 Model and Actual Thrust Performance 66

35 Model and Actual Thrust Performance 67

36 Model and Actual Thrust Performance 68

111
AFAPL-TR-75-71

LIST OF TABLES

PAGE

TABLE
13
I Pilot Blow-Off Correlation
17
II Pilot Can Flameholder Design
17
Il ri,,,:nCan Flameholder Design

Test Conditions Computed for Standard Day 26

(59°F at Sea Level)
26
V, Test Outline
30
VI Instrumentation List

ii
IJ

AFAPL-TR-75-71

LIST OF SYMBOLS

A Cross-Sectional area, ft2 I.

CDB Combustor drag coefficient b/O.5Y 2P2A2M2

CF Thrust coefficient - FNj/O.5yoPoAoM2 f

CN Exit nozzle mass flow discharge coefficient

d Hole diameter, ft

f/a Ratio of mass rate of fuel flow to mass rate of air flow
FNJ Net jet thrust, lbf
2
g Gravitational constant, 32.174 ft-lbm/lbf-sec

KT Inlet air temperature correction factor

L Length, ft
2
m Mass flow function g/ - M [1+It'M 2

M Mach number

MW Molecular weight

P Static pressure, psia

PT Stagnation (or total) pressure, psia
R Gas constant, ft-lbf/Ibm-0 R
SFC Specific fuel consumption, Ib/hr-lbf
T Static temperature, OR

TT Stagnation (or total) temperature, OR i

ATT I Ideal temperature rise, aR

Vol Volume, ft3

V Average velocity, ft/sec

WMass flow, Ibm/sec
WA Air flow, lbm/sec
AFAPL-TR-75-71

LIST OF SYMBOLS (Concluded)

WF Fuel flow, lb /sec

F m +1I

XMach function y
M~T 1+ M2 2y-

Y Mach function H X/Z

Z Mach function'[ 1 + M2] /[I+ 2- M2

3
Density, lbm/ft

y Ratio of specific heats

nc Thermal combustion efficiency

Subscripts

0-5 Engine stations (see Figure 1)

ALT Altitude condition

I Ideal condition

D Disk plane

j Jet through combustor holes

S Stage of holes
T Stagnation (or total) state
P Pilot flameholder

M Main flameholder

rz Recirculation zone

vi
AFAPL-TR-75-71

TABLE OF CONTENTS

SECTION PAGE

I INTRODUCTION 1

1. GENERAL DESCRIPTION 1
2. STATEMENT OF THE PROBLEM 2
3. APPROACH 5

II CAN-TYPE COMBUSTOR DESIGN 7

1. GENERAL LAYOUT 7
2. PILOT STABILIZATION 10
3. FUEL-AIR MIXTURE DISTRIBUTION 14
4. FUEL-INJECTION 15
5. IGNITION 21
6. BASELINE DESIGN 21

III EXPERIMENTAL TESTS AND PROCEDURES 25

1. TEST OUTLINE AND TEST CONDI'iT _NS 25

2. TEST INSTALLATION 27
3. INSTRUMENTATION 29
4. DATA REDUCTION AND CALCULATION METHODS 32

IV TEST RESULTS 43

1. SUMMARY 43
2. TESTS FOR COMBUSTION EFFICIENCY OPTIMIZATION 44

a. Effect of Main Fuel Injector Position 45

b. Effect of Pilot Fuel Nozzle Type 48
c. Effect of Pilot Fuel-Air Ratio 50

3. DESIGN CHANGES TO ACHIEVE COMBUSTOR OPERATION

AT INCREASED ALTITUDE 52

a. Baseline Pilot Can Blow-Off Data

Correlation 52
b. Combustor Design Changes 54

4. CAN COMBUSTOR PERFORMANCE 57

5. MODEL AND ACTUAL RAMJET THRUST PERFORMANCE 64

V CONCLUSIONS 70

REFERENCES 72
AFAPL-TR-75-71

processes is the increase of momentum of the air and fuel, resulting in

a thrust force.

Because of the simplicity of the ramjet, it can be produced for a

very low cost, making it an attractive propulsion device for one-mission

designed tactical missiles. Recently there has been a great deal of

interest in remotely piloted vehicles and target drones in the subsonic

flight speed regime as shown in Figure 2. Because the accomplishment of

the air compression is entirely done by fluid dynamic means, the ramjet
is best suited for high speeds (M > 2). However, due to the necessity
of low cost for throw-away missiles, the subsonic ramjet, with only
modest performance, is still an attractive propulsion device for these

one-mission vehicles.

2. STATEMENT OF THE PROBLEM

The purpose of this investigation was to design, fabricate, and ex--

perimentally determine the performance of a can-type combustor for use

in a subsonic ramjet engine. The subsonic ramjet uses a simple diverg-

ing inlet diffuser and a converging exit nozzle, both of which can be

designed to be manufactured easily and cheaply. The combustor flame-

holder therefore becomes the important factor in the performance and

cost of the subsonic engine.

In the past, flameholders in subsonic ramjets have been of the so-

called bluff-body or gutter type. They consist of hollow cones and V-

shaped gutters, as shown in Figure 3. The simplicity and compactness of

the design has provided for very low cost. However, adequate thermal

2
AFAPL-TR-75-71

NOZZLE
DIFFUSER COMBJSTIQN CHAMBER
<
- - o~~ FLAMEHOLDER

0 134 5

FUEL INJECTION

Figure 1. Schematic Diagram of a Subsonic Ramjet Engine

30
FLIGHT
o ENVELOPE

20

--
_M

0.7 0.9
MACH NUMBER

Figure 2. Subsonic Ramjet Flight Envelope

3 _
AFAPL-rFR- 75-71

BLUFF-BODY TYPE

I CAN TYE

II .-
Ell

IFigure 3. Fidmeholder Types

AFAPL-TR-75-71

combustion efficiencies have not been attainable using bluff-body flame-

holders.

A can in its simplest form, as shown in Figure 3, is a perforated

cone which is actually a multitude of bluff-bodies to hold flame. The

separation of the flow of gas through the perforations promotes intense

air-fuel mixing and rapid burning. The advantage of the can over the
bluff-body flameholder is that the designer has closer control over the

distribution of the combustible mixture into the combustion zone. As a

result, wider operating limits of fuel-air ratio and higher combustion

efficiency are possible over wider ranges of inlet air temperature and

pressure. The can flameholder has had its greatest utility as the com-
bustion stabilizer for turbojet engines, where the flameholder cost is

not as significant when compared to the total engine cost.

The can's wide operating range makes it a likely solution to the

severe operating conditions present in the subsonic ramjet engine. The

study described in this report was utndertaken to demonstrate that the

advantages of the can could be utilized in the subsonic ramjet at a low
cost by incorporating a simple design, free of close fabrication toler-

ance requirements.

3. APPROACH

This study was designed to show whether the concept of low cost can
combustors is feasible. The first phase of the study was to design a

simple can which would be easy and inexpensive to place in production.

AFAPL-TR-75-71

The can was then to be fabricated and tested in a subsonic ramjet engine

to determine the combustion performance.

The aspects of combustor performance considered in the design were

stability limits, combustion efficiency, and pressure loss. The design

of the can is given in detail in Section II. The design is based to an
appreciable degree upon completely empirical and semi-theoretical me-

thods. An extensive literature search was conducted to determine the

important design considerations. A number of these were incorporated
into the can during design. Also, a great deal of redesign was accom-
plished in the testing portion of the program.

The testing was done in the AF Aero Propulsion Laboratory's Ramjet

Test Facility located at Wright-Patterson Air Force Base, Ohio. A de-
scription of the test facility, test procedures, measurement instrumen-

tation, and data reduction techniques are given in Section III.

The test results are given in Section IV. The initial testing was

to optimize a fuel injection system to maximize the combustion efficien-

cy. Using the optimized fuel injector system, tests were conducted at

the Mach number and altitude conditions of the flight envelope of Figure
2. Design changes were incorporated into the can during this phase of

testing to provide stable operation at the Mach number and altitude

boundaries of the flight envelope. The performance of the can obtained
in testing is then compared to the model ramjet performance. Section V
presents the final conclusions of the can design study and test program.

6
AFAPL-TR-75-71
SECTION II
CAN-TYPE COMBUSTOR DESIGN

1. GENERAL LAYOUT
The general layout of the can was mae by examining the 15-inch
diameter low cost ramjet engine of Figure 4. The combustor design
chosen contained a two-section conical can flameholder, a pilot can, and
a main can (Figures 5 and 6). The pilot can had 6 stages with 8 holes

per stage and the main can had 5 stages with 8 holes per stage. As
shown, a stage is a row of holes, all in the same plane, which provides
an entry path for the fuel-air mixture into the turbulent mixing region

of the can. How the number of stages and number of holes per stage was
arrived at will be described later in Section 11.3. A shroud enclosed

the pilot can completely. The shroud divided the air flow for the pilot
can from the main can air flow. The pilot can and main can also had

separate fuel injection systems upstream to provide a pre-mixed fuel-air

mixture. The overall length of the pilot and main can combustor was

47.5 inches with a tailpipe turbulent mixing region of 22.5 inches.

This region, with an L/D of 1.5, was needed to allow the combustion
process to complete behind the last two stages. A 12-inch length was
left between the engine inlet and the shroud inlet plane in order to "
keep the shroud inlet out of the engine inlet flow field. The diameter
of the exit of the main can (Station 3) was 14.0 inches, which allowed a

0.275 gap between the flameholder and the tailpipe. This gap allowed,
after thermal expansion of the can flameholder, enough film cooling to
prevent a hot spot on the tailpipe. The diameter of the junction of the

main and pilot can (Station 2a) was chosen from pressure loss considera-

tion. Grobman (Reference 13) showed for a conical can that the pressure

7
______
 -T

AFAPL-TR-75-71

Figure 4. The Low Cost Subsonic Ramjet

8
 AFAPL-TR-75-71l

MAIN FUEL
SHROUD SPRAY BAR IAI

3.00

Ramjet~ ~ obso
C CihaCn

PILO

figure . AssbleDraiong of the Low Cost Subsonic

Ramjet with a Can Combustor

_~~MI FUEL_ _ _
 AFAPL-TR-75-71

drop is a minimum when the cross-sectional area just upstream of the

first staqe of holes is equal to 0.3 of the cross-sectional area of the

duct around the flameholder. The diameter at Station 2A, therefore, was
7.0 inches. The same pressure loss consideration was used to obtain a
diameter at the pilot can first stage. However, this deals with the

subject of pilot stabilization which will be discussed as a separate

topic.
0

2. PILOT STABILIZATION
It was anticipated that it would be difficult to stabilize and
maintain a flame in the combustor under the operating conditions of low

pressure and temperature. Therefore, it was decided that a flame pilot

would be necessary to maintain combustion. This was done by designing
the can so that a recirculation zone would appear ahead of the first

stage of air entry holes, as shown in Figure 7. This zone was designed
to maintain combustion throughout the engine flight envelope. The pilot
zone then would maintain the combustion in the remainder of the pilot

can which would in turn maintain the combustion in the main can.

The pilot can air flow rate was chosen to be 1/4 of the total

engine air flow by designing the shroud inlet (Figure 6) to be 5.5

inches in diameter. This dimension and Grobman's low pressure drop

results (Reference 13) of a 0.3 ratio of can cross-sectional area to

duct cross-sectional area established the pilot can diameter at the

upstream edge of the first stage of holes at 4.0 inches.

10
 AFAPL-TR-75-71
I
To establish the size of the holes of the first stage and the
volume of the pilot recirculation zone, the pilot can blow-off data
correlation of Figure 8 was used. This is an experimental correlation
for 3.0- and 5.0-inch diameter cylindrical pilot can flameholders having

one row of holes (Reference 9). In the study described in Reference 9

it was also determined that 16 percent of the first stage fuel-air

mixture enters the pilot recirculation zone. The rest of the mixture is

burned in the recirculation zone downstream of the holes.

The design of the low cost combustor pilot recirculation zone was
performed for the 30,000 feet altitude conditions. These conditions are
the most severe due to the lower pressure associated with the higher al-

titude. The pilot recirculation volume was chosen to be 46.838 cubic

inches with 0.5-inch diameter holes in the first stage. The pilot can

blow-off data correlation parameter for this design is given in Table I.

The effective inlet temperature correction factor, KT, was used to ac-

count for the inlet temperature effect on the flameholding stability.

The parameter W
rz is the fuel-air mixture flow rate entering the pilot
recirculation zone volume. The parameter Prz is the pressure of the

mixture entering the recirculation zone volume. The value of these

parameters becomes known when the fuel-air mixture distribution is de-

cided upon as discussed in the next section. The values given in Table I

are for the final design. From Figure 8, it can be determined that this

design should provide a stable operating pilot zone in the equivalence

ratio operating range of 0.51 to 2.0, which for JP-4 corresponds to the
fuel-air ratio in the range of 0.034 to 0.135.

11.
 AFAPL-TR-75-71

..- PILOT RECIRCULATION ZONE

Figure 7. Pilot Recirculation Zone

1.6
RICH
BLOW-OFF
1.4

1.2
STABLF '
OPERATION
.~1.0

.8

LEAN
.6 BLOW-OFF

.4
1 2 5 10 20 50 100

W /PVol (lb )I(s)(atm)2 (ft)3

rz rz rz m

Figure 8. Pilot Can Blow-off Data Correlation

(From Ref. 9)
AF APL -TR - 75-71

TABLE I

PILOT BLOW-OFF CORRELATION

rz
Condition z rT

M .7 at 30 KFT 3.228im/e
atm2 ft3

M .9 at 30 KFT 2.877 2
atm ft

J~

13
F -- -i - --

AFAPL-TR-75-71

At the same time the pilot recirculation zone was designed, a

design was established also for the remaining stages of the pilot can

and the main can. The blow-off data correlation of Zelinski et al

(Reference 8) was used and the design is discussed in the next section.

3. FUEL-AIR MIXTURE DISTRIBUTION

The fuel-air mixture should be distributed to the pilot and main

can in such a way that each stage of the cans operates at a pressure and

velocity which will provide stable operation. The distribution can be

controlled by varying the number of hole stages, the number of holes per

stage, and the diameter of the holes. There are a number of criteria

which have been established from experience which help govern the combi-

nations of the above design variables. These are:

1. Total hole area of entire can should be roughly 120% of com-

bustor cross-sectional area. This leads to a small pressure

drop through the holes and to the lowest possible mixture ve-

locity through the holes for a given flow rate (Reference 13).

2. Low pressure drop through the holes and a low velocity are

also obtained if each stage has from 40 to 60 percent of its

perimeter open (Reference 3).

3. Since too large an addition of reactants after the first stage

for a given axial distance may result in quenching of the

flame, the minimum distance between stages should be about 2.0

inches (Reference 2).

74
AFAPL-TR-75-71

The above criteria were used to establish a number of possible designs

with different hole sizes, different number of stages, and different

number of holes per stage. These designs were then evaluated using the

blow-off data correlation of Zelinski et al, which is shown in Figure 9.

The design that was finally chosen for the pilot and main can

flameholders is shown in Tables II and Ill. This design was chosen

because of its simplicity and because it gave the largest equivalent

disk blockage areas (see Figure 9). The large disk blockage would

provide large recirculation zones. Low velocities are important so that

fuel does not enter the combustion region faster than the combustion

reaction can occur. Finally, this design provided the highest mixture

pressure through the holes and, therefore, the highest reaction rate.

4. FUEL INJECTION

It is important that the fuel and air mixing process provide a near

homogeneous stoichiometric mixture to the can. The combustor design, as

shown in Figures 5 and 6, provides for a separate fuel injection system

for the pilot and the main cans.

The primary objective of the main fuel injector system is to pro-

vide the flameholder element with the proper amount of fuel in a pattern

that will result in efficient combustion. The main fuel injectors also

provide the capability of varying the overall engine fuel-air ratio in

order that the engine thrust level can be varied from acceleration to

cruise conditions. With a can flameholder it is important to have a

homogeneous fuel-air mixture and not allow the local fuel-air ratio in

15
---
- -- -- -- --- -- --
k AFAPL-TR-75-71

BLOCKEDAE WETTED PERIMETER

D AREA

Figure 9a. Equivalent Disk Diameter

r (Zelinski et al, Ref. 8)

N. HOLES/STAGE P
4 6 8 PSIA

01. 04

-. 9

-J

0Y REF. 7CURVE
10l 20 40 60 80 150

21T (WAd (b520 DE-625)

Figure 9b. Blow-off Data Correlation

(Data of Zelinski et al, Ref. 8)

16
 AFAPL-TR-75-71

TABLE II

PILOT CAN FLAMEHOLDER DESIGN

Distance
No. of Hole to Next
stage Holes/Stage Diameter (in.) Stage (in.)
1 8 0.500 3.0 t
2 8 0.750 2.0
3 8 l. 2.0
4 8 1.125 2.0

5 8 1.250 2.0
6 8 1.375 4.0

TABLE III
MAIN CAN FLAMEHOLDER DESIGN

Distance
No. of Hole to Next
Stage Holes/Stage Diameter (in. Stae (in.)
1 8 1.500 4.0

2 8 1.750 4.0
3 8 2.000 4.0
4 8 2.250 4.0
5 8 2.500 ---

17
• - -" .
- "" - - ' ""' - .. - -- -- -___ -. - = - , - -. _ . ,. : -,- , ,
AFAPL-TR-75-71

any of the recirculation zones, created behind the can, to become greater

than stoichiometric. If the local fuel-air ratio exceeds its stoichio-

metric value in any zone, then there is more fuel present than can be

completely burned and the efficiency is lowered. The main fuel injec-

tors shown in Figure 10 consisted of 12 tubes located circumferentially It

with each tube having one 0.022-inch diameter spray hole. The 12 injec-

tors -,e:e designed so they could be varied in the radial direction.

This feature was necessary for the tests described in Section I1. For

stable combustion, it is important to have a homogeneous fuel-air mix-

ture, in vapor form, without large droplets of fuel. At the conditions

at which the engine operates, the inlet temperatures are from -7.8°F to

59.8 0 F and good fuel vaporization cannot be obtained easily. Therefore,

the main fuel injectors were placed such that the fuel spray was directed

upstream. This gives the fuel droplets a large relative velocity with

respect to the air stream and the droplets break up into smaller drop-

lets. This increases the surface area where evaporation of the fuel can

take place.

The pilot fuel injection is not only important for good combustion

efficienc,, but also for combustion stability of the engine over the en-

tire flight envelope. The pilot fuel injection system also sprays up-

stream to break up the fuel droplets. To obtain good atomization, two

types of spray nozzles were tested (Figure 11). The first type was a

full-cone spray nozzle which uses pressure across an orifice to atomize

the fuel. The nozzle also contains a swirl plate to give the fuel ra-

dial velocity to help facilitate atomization. The second type, a hol-

low-cone spray nozzle also uses pressure drop across an orifice for

1P
AFAPL-TR-75-71

41

19
AFAPL-rR- 75-71

SWIRL PLATE

ccF

FULL-CONE NOZZLE

SPRING-LOADED IMPACT PLUG

0I

HOLLOW-CONE NOZZLE

Figure 11. Pilot Fuel Injector Types

20
AFAPL-TR-75-71

atomizing. In addition, this nozzle has an impact plug which mechani-

cally facilitates atomization.

Experimental tests were run to derine the fuel injection system for

optimum combustion efficiency. These tests are described in detail in

Section IV.2.

The fuel chosen for this study was JP-4. This fuel was chosen pri-

marily because of its low cost and accessibility. Also, JP-4 has a high

heat of combustion, 18,701.3 BTU/lbm , for good combustion performance

and is a volatile fuel even at low temperatures.

5. IGNITION

The engine was ignited by a triethylborane (TEB) ignitor system.

Triethylborane is a pyrophoric liquid with a high flame speed. Ignition

of the engine was accomplished by supplying a mixture of fuel and air at

a ratio of 0.04 to the combustor. The triethylborane was then forced

with nitrogen under pressure into the recirculation zone area of the can

until the fuel-air mixture ignited and sustained combustion.

6. BASELINE DESIGN

Photographs of the final design and fabricated low cost can flame-

holder and shroud are shown in Figures 12 and 13. All the material used

in the design was either 321 stainless steel sheet or 321 stainless

steel stock. The 321 stainless steel was used because of its excellent

yield strength characteristics at high temperature. The can was fabri-

cated from 0.050-inch thick stainless steel sheet. The holes were

21
I AFAPL-TR-75-71

A -Can B -Shroud C -Spacers

Figure 12. Can Flameholder and Shroud

22
AFAPL-TR-75-71

WO

.4

-
, LO7
1,

a.

I,
I 234 C"

0,!

Ci
AFAPL-TR-75-71

punched in the sheets and the sheets were then rolled and seam welded.

The shroud was also rolled from a 0.050-inch thick stainless steel

sheet. The struts on the shroud, fabricated from two sections of sheet

material, provided support from the engine wall. Eight struts of solid

stock stainless steel supplied support for the flameholder to the shroud.

These struts were upstream of the combustion and remained at ambient

temperature. The can flameholder was not secured at the downstream end

(base of the cone). Instead, the diameter of the base of the cone was

less than the engine tailpipe diameter. The difference was made up with

12 spacers. This allowed the flameholder to freely move in the axial

and radial direction with thermal growth during combustion. The axial

growth due to thermal expansion was calculated to be 0.374 inch. The

most severe thermal growth problem is at the downstream end of the

shroud which is near the hot pilot flameholder wall. The shroud is near

ambient temperature due to the cool, unburned fuel-air mixture flowing

on the internal and external surface. The pilot can wall is at high

temperature due to the combustion process. The radial thermal expansion

of the pilot can at this station was calculated to be 0.042 inch; there-

fore, the combustor designed allowed a 0.050-inch gap between the shroud

wall and the pilot can wall. This allowed the pilot can to expand

radially without being restrained by the shroud, thus avoiding the

chance of buckling of the oilot can.

The design was successful in achieving low cost. The total cost of

the can combustor was $2468; this included $500 for material and fabri-

cation labor costs of $1968.

24
AFAPL-TR-75-71

SECTION III

EXPERIMENTAL TESTS AND PROCEDURES

1. TEST OUTLINE AND TEST CONDITIONS

The major objective of the tests was to optimize the combustion ef-

ficiency of the designed can combustor by varying the position and type

of fuel injectors. Once this was accomplished, the combustion efficien-

cy, stability characteristics, and the engine thrust were measured at

the intersection of the Mach number and altitude boundaries of the

flight envelope of Figure 2. The test conditions to simulate these four

flight envelope points are given in Table IV.

The main fuel injector positions could be varied in the radial

direction by sliding the fuel injectors in or out as previously shown in

Figure 10. The radial position of the injectors was measured as the

distance from the engine wall to the injector spray hole. The pilot in-

jector was not varied in position but was varied in type. The two types

tested were the hollow-cone nozzle and the full-cone nozzle, as pre-

viously shown in Figure 11.

Another variable was the pilot injector fuel flow setting. It was

expected that the pilot flameholder would have an optimum fuel flow set-

ting which would provide the maximum combustion efficiency. The com-

bustion efficiency optimization tests were to be done at the 30,000 feet

altitude conditions since these would be the most severe due to lower

pressure. A test outline is shown in Table V.

25
AFAPL-TR-75-71

TABLE IV

TEST CONDITIONS COMPUTED FOR STANDARD DAY

(590 F at Sea Level)

( TTI(R) TT I(F)
M0 LT(f_) PALT(psia) PT1 (psia TALT R)

.7 20,000 6.732 9.338 447.4 491.2 31.2

.9 20,000 6.732 11.385 447.4 519.8 59.8

.7 30,000 4.356 6.042 411.8 452.2 -7.8

.9 20,000 4.356 7.367 411.8 478.5 18.5

TABLE V
TEST OUTLINE

Test Objective Variables Conditions

1 Check test None Mo = .7 Alt 20K

facility and
instrumentation.

2 Optimize combustion Main fuel injector M0 = .7 Alt 30K

efficiency. Check positions: 1.5, 2.0, M .9 Alt 30K
combustion stability, and 2.5 inches.

3 Optimize combustion Pilot fuel injector M .7 Alt 30K

efficiency. Check type M0 .9 Alt 30K
combustion stability. 1. Hollow cone
2. Full cone
4 Optimize combustion Pilot can Mo = .7 Alt 30K
efficiency. Check fuel-air ratio Mo .9 Alt 30K
combustion stability.

5 Measure combustor Altitude M = .7

efficiency at the Mach number °Alt 20K & 30K
intersection points M = .g
of the flight envelope °Alt 20K & 30K
with best fuel injector
configuration.

26
AFAPL-TR-75-71

2. TEST INSTALLATION

The experimental tests were conducted in the Air Force Aero Propul-

sion Laboratory Ramjet Test Facility at Wright-Patterson AFB, Ohio. The

test equipment is shown in Figure 14. The engine tailpipe section was

installed in a cabin which was connected to exhausters which provided

back pressures representative of altitude conditions. The inlet of the

engine was connected to outside ambient air through a 12-inch diameter

pipe. A total pressure rake, designed to ASME standards, was placed in

the inlet of the engine and at the exit of the tailpipe. The total

pressure associated with the flight Mach number being simulated was set

at the inlet total pressure rake. This was done with a pneumatically A

cortrolled butterfly valve placed 28 feet upstream of the engine inlet.

A subsonic flange tap orifice was used to measure engine air flow. The

orifice was placed in relation with the valve and a set of flow straight-

eners in accordance with ASME standard such that the velocity gradient

created by the valve would not affect air flow measurement accuracy.

The bellows were installed to allow expansion of the pipe installation.

The tests were conducted by setting the cabin pressure and the en-

gine inlet total pressure to the appropriate PALT and P for the re-

quired Mach numbers and altitibde simulation. While there was no means

of conditioning the air for correct temperature, testing during the

winter months provided air near the desired value. The normal test se-

quence was to start taking data at low fuel flow rates to the engine and

then increase the main fuel flow until rich flame blow-off occurred.

The P1 and PALT values were corrected after each fuel flow setting to

maintain correct Mach number and altitude. Pressure, temperature, ano

27
AFAPL-TR-75-71

ccU

u 4
J

Ln)
wU

cr--

LU CL

U, U
LiLS

rCD cr.

kn I--

U5-

C)r

28
AFAPL-TR-75-71
fuel flow data were recorded at each fuel flow setting. The results
were then translated into combustion efficiency and engine thrust as a

function of fuel-air ratio for each fuel injector configuration at a

given Mach number 3nd altitude.

3. INSTRUMENTATION
The engine was instrumented for temperature and pressure measure-

ments to calculate combustion efficiency and engine thrust. Tempera-

tures were measured with thermocouples and pressures with transducers.

Fuel flow was measured with a rotometer and a frequency converter. The

analog signals from the thermocouples, transducers, and frequency con-

verters were all recorded on a magnetic tape with the use of a data ac-

quisition system. The magnetic tape was run in conjunction with a com-

puter program which contained the required calibration data to change

the recorded voltage signals to engineering units. Table VI gives a

list of the parameters measured and Figure 15 shows the instrumentation

locations.

Four total pressure pitot tubes were placed at the engine inlet,

the diffuserexit, and the engine exit. At each station the four total

pressures were averaged. The inlet total temperature was measured with

the average of three total temperature probes. The tailpipe exit nozzle
was instrumented with four thermocouples which were averaged.

Real-time color movies were taken from behind the engine looking

into the combustion chamber. A telescopic lens was used which enabled

zones of flameholding to be clearly distinguished. The lens was used

either with'the movie camera or for direct visual observation.

29
AFAPL-TR-75-71
TABLE VI

INSTRUMENTATION LIST

Parameter Averages Description Location

P Flange Tap Pressure, Top Orifice, Upstream

P20 Orifice, Downstream
P30 Flange Tap Pressure. Bottom Orifice, Upstream
P40 Orifice, Downstream
P10 1 Inlet Static Pressure Engine Inlet
P P ,,
P102 1
P10 3 ,I
104
P Inlet Total Pressure Engine Inlet
P 112 ,T
If It
P113 of I
114
P121 Diffuser Static Pressure Diffuser Exit
p 12 2
p 123 2 . ,
"124
Pl31 Diffuser Total Pressure
P132 ,
p 133 2, ,,
,,
"134
P Combustor Static Pressure Combustor Wall
P141 If
142
PP162 P5" Exit Static Pressure Exit Nozzle

P 163 II
P164
P1 Exit Total Pressure
171
P172 p ,, ,,
p173,,,
174
P P Altitude Pressure Cabin Wall
P191 PALT to
193
T Ambient Air Temperature
TT Total Temperature Inlet
TT2 12 TT
T 22 1
T823 Nozzle Wall Temperature Exit Nozzle Ring
T8 2 TEXIT ', II
T83,,,
*84

F Main Fuel Flow

4M Pilot Fuel Flow
rp

30
AFAPL-TR-75-71

I-

+r

LUn

I0-

0 0 0[

+ -

0.

000 L

100 /-- .6-

000

+31
AFAPL-TR-75-71

4. DATA REDUCTION AND CALCULATION METHODS

A computer program, LOWCOST 2, was written to perform the engine

performance calculations using the measured engine pressure and tempera-

ture data of Table VI. The performance calculations are discussed
below.

The engine air flow, AT, is calculated using the pressures Po

10

P2 0 ' P30 ' and P4 0 of the orifice plate. Next, the Mach number at engine
Station 1 (see Figure 15 for engine station nomenclature) is calculated

from
1/2
Y- 1

M 1 =( Y)LP) -]

In this expression PT and P1 have values obtained through measurements.

1
The ratio of specific heats y, = 1.4.

The division of the air flow to the main and pilot flameholders is
calculated next. The air flow to the main flameholder is calculated

using the measured conditions at engine Station 2. For an ideal gas,

the rate of mass flow

WA pAV (2)

32
AFAPL-TR-75-71

can be rearranged to yield

WA = PAMs RR (3)

where M = V// lyFT

is the Mach number.

Using

TT +L - 1M2 (4)
T 2

the value of T can be found and substituting it into Equation (3) gives

1 1 2 (5)
OA =PAMg V T [+ Y -2

Now defining

0 1+ 2 1/2 (6)

the mass flow relation becomes

A;mF (7)

It is this form in which mass flow rate will be most useful in the data

reduction calculations.

33
 AFAPL-TR-75-71

From Equation (7), the main flameholder air flow is given by

W AM
AT2 = (8)

where P2 is obtained by measurement, m2 is calculated from

M2 = f(PT2/P 2 ) and Equation (6), TT is assumed equal to TT which is

measured, and A2 has a value of 68.666 in2 .

21

The pilot flameholder air flow is given by the conservation of mass

relation

WA = WA - WAM (9)

The fuel-air ratios are now calculated using the measured fuel

flows

(f\ F F (10)
a /T WA

()M =
WA M

<f)WFpp (12)

aP
WA3

!| 34
 AFAPL-TR-75-71

Next the conditions at engine Station 5 are calculated. In these

calculations the real gas thermodynamic properties after combustion were

used. Figures 16, 17 and 18 show the ideal temperature rise (ATT ),
I
the molecular weight (MW5 ), and the ratio of specific heats (y5 ) for the

combustion products. The nozzle exit Mach number is given by

/

Y5 112

Y5

M~(7 ?T (P (13)

where PT5 and P5 have been measured and Y5 is obtained from Figure 18.
The nozzle exit total temperature can be calculated using the mass flow

relation at Station 5, which is

W5 PACN m T5 (14)
5 5 fT

Or rearranging

0 2

T = AW5 (15) .
5 Li

where P5 is obtained by measurement, A5 has the value of 108.99 in2 ,

m is calculated from M5 using Equation (6), W5 is the sum of IA and

WFT, and CN is the nozzle discharge coefficient which was found ex-

perimentally and is presented in Figure 19.

35
 AFAPL-TR-75-71

JP.4
1.( H .0)PRESSURE =1.00 ATM
4000 '* *~ ~

3600

TTO
3200
400
700
2800 001000
1300
1600
1900
Z 2400 20

2500
S 2000

1600

1200

80

0.0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
FUEL-AIR RATIO
Figure 16. Ideal Temperature Rise for Given
Fuel-Air Ratio and Inlet Temperature
A

36
 II,.

AFAPL-TR-75-71

JP-4 (C ".0 H 2.0) PRESSURE = 1.00 ATM

29.5 * . !.

29.0

28.5

° i
28.0

27.5

27.0
T=

TTO i
I.
26.5 400

2500

26.0
0.0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10

FUEL-AIR RATIO

Figure 17. Molecular Weight of Combustion Products for

Given Fuel-Air Ratio and Inlet Temperature

37
AFAPL-TR-75-71

JP-4 (C 1.0 H 2.0)

PRESSURE 1.00 ATM

1.40

1.35

1.30
Tr

0 400
1.25

1.20
2500

1.15

I - 0
0.0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10

FUEL-AIR RATIO

Figure 18. Ratio of Specific Heats of Combustion Products

for Given Fuel-Air Ratio and Inlet Temperature

.38
AFAPL-TR-75-71

.0

2.0 3.0
.0

Figure 19. Exit Nozzle Discharge Coefficient

39
 i ,

AFAPL-TR-75-71 p

The thermal combustion efficienry is now defined

•i as 4

TT5 - TT

n 1 (16)
c ATT

where ATT is from Figure 16, TT is measured and TT is Equation (15).

5

Thrust performance for ramjet engines is normally presented as net

jet thrust, which is calculated on the control volume from Station 0 to

Station 5. In these tests, the freestream Mach number, Mo , is being

simulated by setting the correct total pressure at Station 1. In order

that the test data can be compared to other results, the simulated free-

stream M and A0 will be calculated. The simulated freestream Mach num-

ber is given by

- 1 1/2

M = Y, " (17)-

which assumes PT = T1 "

O

The simulated area of the air flow stream at Station 0 can be found

using mass conservation

4AWA
=

40
 AFAPL-TR-75-71

Therefore, using Equation (7) and assuming PT 0 PT1 and To =Tthe

capture area ratio is given by

Ao0 I P1
A1 mo PALT (18)

00

where P1 and PALT are obtained by measurement, ml is calculated from M1

using Equation (6), mo from M using Equation (6), and A1 has a value of

62.072 in2 .

The net jet thrust is given by

F pPsNJP5+-55
A- PALTAO(I
A + P!O2
A
ALTk5 - A)(19)

where P5 and PALT are obtained by measurement, A5 has the value of

108.99 in2 , y0 has the value of 1.4, 5 is obtained from Figure 18, M 5
is Equation (13), Mo is Equation (17), A is Equation (18), and n. is
0 0

the nozzle discharge efficiency with a value of 0.98.

In order to present thrust performance as dimensionless, the thrust

coefficient, CF, is defined as

F
C
CF~
F NJ 2
2 Yo PALT' o

where FNJ is Equation (19), Y0 has the value of 1.4, PALT is obtained by
I measurement, A3 has the value of 167.7 in2 , and M is Equation (17).

41
AFAPL-TR-75-71

Another important ramjet performance parameter is thesecfcul

per lbf Of
consumption which is the measure of lbm of fuel used per hour
thrust and is defined as

-SFC T
FNJ

424

ii ii
 AFAPL-TR-75-71

SECTION IV

TEST RESULTS

I. SUMMARY

During the test program, 36 hours of combustor operation time were

totaled. The same hardware was used throughout the test program with no

hardware failures. The initial test of Table V. Test Outline, was a

cold flow run without combustion at the Macn 0.7, altitude 20,000 feet

condition. This test was made to check the facility and instrumentation.

The main result of the initial test was the establishment of the fact

that the air flow pressure and velocity supplied to the engine inlet did

not vary with radial position in the pipe. The next three tests of

Table V were to be fuel injector variations to optimize combustion

efficiency at the most severe conditions, which were the Mach 0.7 and

0.9, altitude 30,000 feet conditions. However, attempts to run the can

at the 30,000 feet test conditions failed because combustion could not

be sustained. To achieve stable combustion at these conditions, some

design changes would be necessary. Before making these changes, however,

the combustion efficiency optimization tests were conducted at the

20,000 feet altitude condition, since the combustor would operate smooth-

ly at this condition. With this experience, dn optimum fuel injector

*system could be used when attempts would be made at the 30,000 feet test

conditions. The results of the combustion efficiency optimization Tests

_2, 3, and 4 are given in Section IV.2.

Between Test 4 and Test 5 of Table V, an experimental design change

program was undertaken. The design program consisted of a blow-off data

43
K

AFAPL-TR-75-71

correlation study oF the pilot can, two pilot can design changes, and

experimental verification of the design changes. The above design

changes resulted in successful operation of the can at the 30,000 feet

test conditions. These design changes are discussed in Section IV.3.

Once the design changes were complete, Test 5 was conducted which

was the measuring of the combustor performance at the test conditions of

Mach 0.7 and 0.9, altitude 20,000 feet and 30,000 feet. The results of

Test 5 are discussed in detail in Section IV.4.

In order to check the validity of the test results, the actual

ramjet performance was compared to model ramjet performance which was

obtained with a methematical model. This comparison is discussed and

shown graphically in Section IV.5.

2. TESTS FOR COMBUSTION EFFICIENCY OPTIMIZATION

After the can is designed and constructed with a fixed number of

stages, a fixed number of holes per stage, and fixed hole diameters, the

only design variable left is the fuel injection system. The distribution

of the fuel to the combustor greatly influences the combustion efficiency

in that the fuel must be distributed so there is not more fuel present

in any stage of the combustor than can be burned. Also, the method of

fuel injection of the fuel affects the vaporization of the fuel in the

air stream and therefore affects the reaction time necessary to complete
the combustion process. The best method for obtaining a good fuel in-
jection system is by experiment. The variables of the fuel system are

the main fuel injector position (radial distance from the engine wall to

44
AFAPL-TR-75-71
the spray hole), the pilot fuel nozzle type, and the pilot can fuel-air
ratio. The objective of the combustion efficiency optimization tests
was to exercise the above variables independently to obtain the maximum

combustion efficiency. The optimization tests were conducted at the

Mach 0.7 and 0.9, altitude 20,000 feet engine operating conditions.

a. Effect of Main Fuel Injector Position

The main fuel injectors must provide a near homogeneous mix-

ture to the main can. For the highest combustion efficiency, the main

fuel injectors must not allow the local fuel-air ratio in any of the
recirculation zones to become greater than stoichiometric. The variable

of the main fuel injector design was the radial distance of the spray
hole from the engine wall, as shown previously in Figure 10. Three
distances were tested: 2.5 inches, 2.0 inches, and 1.5 inches. The
2.5-inch distance was as far as the injectors could be positioned without
interfering with the pilot can shroud. As shown by Figure 20, the 2.5-
inch position provided the best combustion efficiency throughout the

total engine fuel-air traverse for the Mach 0.7, altitude 20,000 feet

test condition. Similar results were obtained at the Mach 0.9, altitude
20,000 feet test condition, as shown in Figure 21. The 1.5-inch posi-

tion was not tested at the Mach 0.9, altitude 20,000 feet condition
since the trend of poorer combustion efficiency with less injector

distance was being repeated.

To substantiate the results of Figures 20 and 21, the color

movies and visual observation showed more traces of red fuel-rich burn-

ing taking place outside of the engine tailpipe as the fuel injectors
were brought closer to the wall. Fuel burning outside of the engine in
45

- - - - - - - - - - -
AFAPL-TR-75-71

MACH NUMBER -0.7

ALTITUDE - 20K

FULL-CONE PILOT NOZZLE - 400 lbm/hr FLOW RATE

FUEL - JP-4

1.0

0.8

I.-
S0.6
U-

= 0.4 MAIN FUEL INJECTOR DISTANCE

FROM ENGINE WALL

0 2.5 inch
0.2
0 2.0 inch

0 1.5 inch

0.0
0.02 0.03 0.04 0.05 0.06 0.07

ENGINE FUEL-AIR RATIO

Figure 20. Effect of Main Fuel Injector Position

46
-AFAPL-TR-75-71

MACH NUMBER -0.9

ALT ITUDE - 20K

FULL-CONE PILOT NOZZLE -500 lbn/hr FLOW RATE
FUEL -JP-4

1.0

0.8

S0.6
U
MAIN FUEL INJECTOR DISTANCE
~ 0.4FROM ENGINE WALL

0 2.5 inch I
0.2 0 2.0 inch

0.0
0.02 0.03 0.04 0.05 0.06 0.07
ENGINE FUEL-AIR RATIO

Figure 21. Effect of Main Fuel Injector Position

47
AFAPL-TR-75-71

the exhaust plume naturally results in engine combustion efficiency

loss. The probable cause of the 1.5- and 2.0-inch positions providing

less combustion efficiency is that these two configurations provide more

fuel to other radial main can hole stages. Because these stages are

also nearer the exit nozzle, the recirculation zones behind these stages
are shorter. Therefore, there is less time for complete combustion of

all the fuel to take place.

Based on the above results, the 2.5-inch position for the main

fuel injectors was used in all subsequent tests.

b. Effect of Pilot Fuel Nozzle Type

The pilot fuel nozzle must provide a near homogeneous mixture

to the pilot can. Since the pilot fuel nozzle must spray in the small

restricted area of the 5.5-inch diameter shroud, it is important that

fuel not spray too far in the radial direction, causing the fuel flow to

impinge on the shroud wall. This would result in a flowing liquid

stream'of fuel down the wall rather than a well-vaporized fuel-air mix-
ture which in turn would result in lowered combustion efficiency.

As has been mentioned, two types of nozzles were tested and

compared at the Mach 0.7, altitude 20,000 feet test condition. These

were the hollow-cone and full-cone types. Figure 22 shows that better

combustion efficiency was obtained with the full-cone nozzle throughout

the total engine fuel-air ratio range. Also, during the tests, combus-

tion instability was noted with the hollow-cone nozzle at the high fuel

flow rates. The instability resulted in the flame blowing out on two

48
AFAPL-TR-75-71

MACH NUMBER - 0.7

ALTITUDE - 20K

MAIN INJECTORS AT 2.5 INCH

PILOT FLOW RATE - 400 Ibm/hr

FUEL - JP-4
1.0

LO v
0.8

S0.6
°ii
LW~

S0.4

~~PILOT NOZZLE TYPE )

0.2 -0 Full Cone
0.4
0Hollow
_ _ Cone__
_ _

0.0 I I I i

0.02 0.03 0.04 0.05 0.06 0.07

ENGINE FUEL-AIR RATIO

Figure 22. Effect of Pilot Fuel Injector Type I

49
 1,1

AFAPL-TR-75-71

occasions. Since for the Mach 0.9, altitude 20,000 feet condition

higher fuel flow rates would be required, the hollow-cone nozzle which

previously resulted in an instability problen was not compared to the

full-cone nozzle at Mach 0.9 test condition. The full-cone nozzle was

used successfully at the Mach 0.9, 20,000 feet condition during the

tests to determine the influence of main fuel injector position on corn-

bustion efficiency. Therefore, the full-cone nozzle was selected for

subsequent tests also.

C. Effect of Pilot Fuel-Air Ratio

The effect on combustion efficiency by the variation of the

pilot fuel-air ratio is shown in Figure 23 for the Mach 0.7, altitude

20,000 feet test condition. It would be expected that the maximum en-

gine combustion efficiency would be achieved if the pilot can was operat-

ing near stoichiometric. This is shown to be true in Figure 23. This

test was run by holding the pilot fuel flow constant at one of the three

fuel settings (400, 350, and 500 lb/hr) while the main fuel flow was

varied to change the overall engine fuel-air ratio. As the main fuel

flow is increased, and thus the total engine fuel-air ratio increased,

there is less air flow through the engine. Therefore, with the fixed

pilot fuel flow and decreasing pilot air flow, the pilot fuel-air ratio

also increases. The pilot fuel-air ratio range for the 400 lb/hr set-

ting was 0.061 to 0.065, for the 350 lb/hr setting the range was 0.055

to 0.057, and for the 500 lb/hr setting the range was 0.073 to 0.079.

As can be seen, the near stoichiometric pilot fuel-air ratio range of

the 400 lb/hr fuel setting provided the best combustion efficiency over

50
II .

I
AFAPL-TR-75-71

MACH NUMBER - 0.7

ALTITUDE - 20K

MAIN INJECTORS AT 2.5 INCH

HOLLOW-CONE PILOT NOZZLE V

FUEL - JP-4

1.0

0.8

La

S0.6
Uj
Ln
z
I.-
= 0.4
PILOT FUEL-AIR RATIO

o 0.061-0.065
0 0.055-0.057

V 0.073-0.079

0.0 L
I I
0.02 0.03 0.04 0.05 0.06 0.07

ENGINE FUEL-AIR RATIO

Figure 23. Effect of Pilot Fuel-Air Ratio

51
AFAPL-TR-75-71

the entire range of total engine fuel-air ratios tested. The pilot
fuel-air ratio was therefore set as near stoichiornetric as possible for
subsequent tests.

3. DESIGN CHANGES TO ACHIEVE COMBUSTOR OPERATION AT INCREASED ALTITUDE

Attempts to run the combustor at the 30,000 feet test conditions

with the optimized fuel injection system resulted in combustion insta-

bility and flame blow-off. A stability limit curve of the altitude and
fuel-air ratio region which would provice stable combustion is shown in

Figure 24. This curve is for a constant Mach number of 0.7. Visual
observation of the engine flame-outs showed that initial flame blow-off

was occurring in the center of the engine which is the pilot can; there-
fore, a redesign of the pilot can was made.

a. Baseline Pilot Can Blow-off Data Correlation A-5

A blow-off data correlation was made by operating the engine

with the pilot can and pilot fuel injector only. The purpose of the
test was to determine the effect of pilot combustor pressure, pilot air
flow rate, and pilot fuel-air ratio on the blow-off limits of the pilot

can. The range of variables was: pressure, 5.988 to 11.269 psia; air A

flow, 1.382 to 2.849 lb /sec; and fuel-air ratio, 0.048 to 0.119. The 4

inlet total temperature for this test varied only slightly from 528.2 to
531.9°R. The blow-off limits were found by setting the pilot can pres-
sure and pilot can air flow. The pilot fuel flow setting was then

reduced until lean blow-off occurred. The combustor was then reignited '
tf A
and the pilot fuel flow setting was increased until rich blow-off oc-

curred. The Longwell and Weiss correlation parameter without the recir-
culation zone volume term was used to correlate the data. The volume
52
AFAPL-TR-75-71

MACH NUMBER - 0.7

FUEL - JP-4
301

, 25 FLAME FLAME
BLOW-OFF BLOWF-OFF

-°
Lii
"-
S20 STABLE OPERAT10N ""

15
0.02 0.03 0.04 0.05 0.06 0.07
ENGINE FUEL-AIR RATIO
Figure 24. Stability Limits of Baseline Design

53i
AFAPL-TR-75-71

term was not included since it was fixed and therefore would not change

and effect the blow-off limits. The result of the blow-off data cor- -

relation is shown in Figure 25. The parameter is the pilot can air
P -

flow and P2 is the pressure at the pilot can en:rance. The significance

of the pilot can blow-off correlation is that for : known pressure the
air flow which will provide stable combustion in the pilot can flame-

holder can be found. The combustor entrance pressure representative of

the Mach 0.7 and 0.9, altitude 30,000 feet test conditions was used with

the correlation to find an air flow rate which would provide a correla-

tion parameter value of 3.0, thus providing wide stable operating limits.

This air flow rate was found to be lower than the pilot can air flow
rates being provided by the baseline design. Therefore, a design change

was necessary to reduce the pilot can air flow rate.

b. Combustor Design Changes

To reduce the pilot air flow an extension was added to the

shroud which reduced the shroud inlet from a 5.5-inch diameter to a 3.0-

inch diameter (Figure 26). The engine was tested with the shroud exten-

sion. The result was increased stable operation to an altitude con-

dition of 27,000 feet, as shown in Figure 27. This still was not satis-

factory. It was thought that with the lower pilot can air flow rates,

the fuel-air mixture entering the first stage of holes was not pene-

trating into the pilot recirculation zone due to the lower velocities.

A well-defined recirculation zone must be formed or stable flameholding

cannot be achieved because there will not be a turbulent mixing region

for the combustion to occur. Therefore, a set of air scoops was designed,

54
AFAPL-TR- 75-71

.14

.12 RICH BLOW-OFF

06

LEA

0 BLOW-OFF
10 2o 30 40 5.0 6.01

/P1 .8 (lb /sec)/(atm) 1.8

Figure 25. Pilot Can Blow-off Data Correlation

SHROUD EXTENSION

Figure 26. Shroud Extension

MACH NUMBER -0.7

FUEL -JP-4

30 *

FLAMEI
25 BLOW-OFF * BLOW-OFF

STABLE OPERATION
-j
~-20

15
0.02 0.03 0.04 0.05 0.06 0.07

ENGINE FUEL-AIR RATIO

Figure 27. Stability Limits of Baseline

Design with Shroud Extension
56
AFAPL-TR-75-71
fabricated, and placed on the downstream side of the first stage of

pilot can holes. The scoops shown in Figure 28 were 0.5 inch tall and

0.5 inch wide. The purpose of the air scoops was to help change the

direction of the oncoming fuel-air mixture and force the mixture to

I
penetrate and form a hot recirculation zone. Tne engine was then tested

with the shroud extension and air scoops. The combination was success-

ful in extending the stable operating region of the can combustor into

the Mach 0.07, altitude 30,000 feet test condition, as shown in Figure

29. The condition of Mach 0.9, altitude 30,000 feet was not attainable,

partially due to facility problems at this high Mach number and alti-

tude. This condition was near the operating limit of the facility and

the test conditions were difficult to set and stabilize. The chanying

test condition setting added to the instability of the combustor. i

Therefore, performance was taken at Mach 0.8, altitude 27,000 feet. The

next section discusses the documentation of the can combustor perform-

ance at the four corners of the flight envelope.
I

4. CAN COMBUSTOR PERFORMANCE

After the design changes were complete, the can combustor perform-

ance at Mach 0.7 and 0.9, altitude 20,000 feet and 30,000 feet was de-

fined. These tests were conducted with the baseline can design plus the

shroud extension and the first stage air scoops described in Section

IV.3.b. The fuel injection system used was the result of the combustion

efficiency optimization tests of Section IV.2.

The combustion efficiency results are shown in Figure 30. Since

the combustor could not be operated at the Mach 0.9, altitude 30,000

feet test condition, data was taken instead at a Mach 0.8, altitude

57
 i-i

AFAPL-TR-75-71 -~

-~

iJ.~
~.

a.
4 f

~ ij~

ii

Figure 28. First Stage Pilot Can Air Scoops

ii

58
T._ _ _ _ _ _

-AFAPL-TR-75-71-

cz 0
00

CLJ

Ln --

0 L

C"I
CDC

o59
AFAPL-TR-75-71

c)w

cc0

U- e

C~i

CD cc

UJ,

>.E
01
uI

~~4--
o a 0E

i.j

U-

*JI
I I I C~J

- 0 0l Ci C0 0 C0 C) CD

0 NI11UJJ3 N'JI1SflOWOJ AON]11J3 N0IJ.sfl8wo

60
AFAPL-TR-75-71
27,000 feet test condition. In each of the four combustion efficiency
curves, the lower fuel-air ratio is the lowest fuel-air ratio which

would sustain combustion. The highest fuel-air ratio indicated on each

curve is not the upper limit on sustained combustion. The combustor was

not tested above a fuel-air ratio of 0.06 at the 20,000 feet test con-
ditions due to the high engine tailpipe temperatures experienced at near

stoichiometric combustion. In the test set-up, adequate tailpipe cool-

ing was not available. However, there is no reason to believe, due to

the near constant slope of the curves, that the combustion efficiency at

the near stoichiometric fuel-air ratio would be much different in value.

At the higher altitude test conditions, the fuel-air ratio was not ex-

tended much beyond the stoichiometric fuel-air ratio since the perform-
ance begins to drop off at this point. One can see that the higher com-

bustor air flow velocities associated with low fuel-air ratios are much

more critical at the higher altitude conditions. At the higher altitude

conditions, combustion could not be sustained at the low fuel-air ratios.

Figure 31 s9rows the net jet thrust performance of the engine. The

figure shows the increase in thrust due to higher dynamic pressures and
higher mass flow rates associated with the higher Mach number test

conditions. Also shown is the decrease in thrust with the lower pres-
sures and lower mass flow rates associated with the higher altitude

conditions.

The engine total pressure recovery is shown in Figure 32. The high
pressure recoveries indicate that a low drag and low pressure loss

design was achieved. Low pressure loss is very important in order that
the thrust level can be as high as possible.
61
 V4
j AFAPL-TR-75-71

o 00

S-l

oE 0o

C Li. L..

L L J. L . I I L

C0 o C 00 C g
C > C) 00

(Jt PNA (Iqi) rN~ 0.

LUI
0. 0

(~I 0000) ionii iiv.-

oC0

I...

LL. o LL.

C) 0
p-J C)-J
LL..

C14cJ - Li.

L Iq ii 0~ I I I 0

o 02
AFAPL-TR-75-71

0 > r

Cl -
0i CII
C3 I-
~. CD
JL
LAo CD N

1dl d dl

uj o
=0i

4 4.4

0(1 01 0aill

L- L-

C's.) C ..

L d/ Gd 1 d/
d

63
V

AFAPL-TR-75-71

5. MODEL AND ACTUAL RAMJET THRUST PERFORMANCE COMPARISON

in order to show that the assumptions of the performance calcula-
tions were valid and the can combustor thrust performance was near the
maximum performance which can be expected from the ramjet cycle for the
combustion efficiencies obtained, the test data was plotted on model
performance curves. The model predicted performance was obtained with
use of the subsonic engine performance computer program of Reference 18.
The engine burner drag coefficient was calculated from the test data to
be 3.0. This burner drag coefficient and the actual engine dimensions
were used to obtain a parametric plot of net jet thrust coefficient vs.
specific fuel consumption for all fuel-air ratios and combustion effi-
ciencies. This was done for each of the four test conditions, as shown
in Figures 33, 34, 35, and 36. On each of the parametric plots the
combustion efficiency obtained at a given fuel-air ratio in testing was
plotted as a solid line. The solid lines thus represent the thrust
performance of the model ramjet for the combustion efficiency ranges and
fuel-air ratios that were obtained during testing. The actual thrust
coefficient and specific fuel consumption test results are shown as data
points on the figures. The only other factor which affects internal
ramjet performance other than combustion efficiency is internal aerody-
namics. The distance along the constant fuel-air ratio lines between
the data points and the solid line represents the aerodynamic effects.
At the Mach 0.7, altitude 20,000 and 30,000 feet test conditions and
Mach 0.8, altitude 27,000 feet test condition (Figures 33, 34, and 35),
the calculated thrust performance corresponds very well to the predicted
values of the solid line. At the Mach 0.9, altitude 20,000 feet test
condition (Figure 36), the calculated thrust and specific fuel consumption

64
AFAPL-TR-75-71

MACH NUMBER - 0.7

ALTITUDE - 20 K

FUEL - JP-4
0.6 1 : 1.0
- .- .,0.9

C( " . 0.06
o, / .:.~o
%o / 0.05
,or..0.04

04. 7 0.03
0.4
,

O"'"n.O
- 0.3

L-

i~ 0.2
I-
i

0.1-

0.0 I I I
4.0 5.0 6.0 7.0 8.0 9.0

SPECIFIC FUEL CONSUMPTION, SFC

Figure 33. Model and Actual Thrust Performance

65
AFAPL-TR-75-71

MACH NUMBER - 0.7

ALTITUDE - 30K
FUEL - JP-4
0.6
nc
--
,<o .7"

2 - -. ="_
- - 0
0.5 . e _/ _"' _ b .07
C.,,/

/ /5 V f/a +i
/ / , +
0. 4 /

I 7 . 03
'- '
" S.0

0.3 S

.02
I,
I-.,
LI-
~ 0.2

0.1

0.0I I I I
4.0 5.0 6.0 7.0 8.0 9.0 10.0
SPECIFIC FUEL CONSUMPTION, SFC

Figure 34. Model and Actual Thrust Performance

66
AFAPL-TR-75- 71

MACH NUMBER -0.8

ALTITUDE - 27K
1
1
FUEL - JP-4

0.60

,. , " " .07

- .06
06

0.5 05f/a

L" 0.4 1.03

g ~ .02
- 0.3 -

0.1
0.2

0.0 I I "

4.0 5.0 6.0 7.0 8.0 9.0

SPECIFIC FUEL CONSUMPTION, SFC

Figure 35. Mudel and Actual Thrust Performance

67
I: - - ~ ~ .

AFAPL-TR-75-71

MACH NUMBER - 0.9

ALTITUDE - 20K

FUEL - JP-4

0.6 .9
-b- .8=n

0.5 / %~ 06
, ~ '%.
%% / 05 .If/a
C-) / .04
'.0 .07

o 0.4 -
S 02.02
-- 0.3 '

S 0.2
U-

0.1

0.0 I I
3.0 4.0 5.0 6.0 7.0 8.0

SPECIFIC FUEL CONSUMPTION, SFC

Figure 36. Model and Actual Thrust Performance

68
k AFAPL-TR-75-71

for the higher fuel-air ratios agreed with the predicted values. At the

lower fuel-air ratios, aerodynamic losses lowered the thrust values even A

though good combustion efficiency was obtained. ..

69
AFAPL-TR-75-71

SECTIUN V

CONCLUSIONS

The significant accomplishments of this study are summarized below:

1. A ramjet can combustor can be designed to operate in the

flight envelope bounded by the flight conditions of Mach 0.7

and 0.9, altitude 20,000 feet, Mach 0.7, altitude 30,000 feet,

and Mach 0.8, altitude 27,000 feet.

2. The design of the combustor can be free of close fabrication

tolerance requirements, and combustion efficiencies of 80 to

90 percent are still attainable. Because of the uncomplicated

design, the combustor can be easily and inexpensively placed

in production.

3. In a can combustor designed to operate at pressures of one-

half atmosphere and very low incoming air temperatures, the

most important aspect is a well-designed piloting system.

4. Minimum burner drag, and thus low pressure drop, can be ob-

tained by allowing a large total hole open area without sacri-

ficing recirculation zone volume. This can be accomplished

with two conical cans in series, which provides a large recir-

culation volume.

70
AFAPL-TR-75-71

5. The fuel injection system is of great importance at the low

temperature and pressure operating conditions. The fuel

should be added near the center of the engine and allowed to

spread radially to the outer stage air-entry holes of the

combustor. The full-cone fuel nozzle provides better atomi-

zation and pilot combustor performance than the hollow-cone

nozzle. The best overall combustion efficiency is obtained

when the pilot can is operating at near stoichiometric.

6. The addition of air scoops on the first stage of air-entry

holes increases the low pressure stable operation range of the

can combustor.

7. The direct-connect pipe test set-up provides a valid perform-

ance evaluation of the subsonic ramjet engine. Dependable

thrust calculations from pressure measurements can be made.

The final result of this program has been the direct-connect pipe test-

ing of a subsonic ramjet engine with a can combustor. The program has

demonstrated that the ramjet with a can combustor will provide an inex-

pensive propulsion system with modest performance characteristics.

71
AFAPL-TR-75-71 _-4

REFERENCES
-I

1. Watson, K. A., "X, Y, Z Mach Number Functions for One-Dimensional

Compressible Flow", APRA-TM-70-16, May 1970.

2. Farley, J. M., Smith, R. E., and Dovolny, J. H., "Preliminary

Experiments with Pilot Burners for Ramjet Combustors", NACA
RM-E52J23, Lewis Research Center, Cleveland, Ohio, January, 1953.
:
3. Feleznik, F. J. and Gordon, S., "A General IBM 704 or 7090 Computer
Program for Computation of Chemical Equilibrium Compositions,
Rocket Performance, and Chapman-Jouguet Detonations", NASA TND-1454,
Lewis Research Center, Cleveland, Ohio, October, 1962. j
4. Gordon, S. and Feleznik, F. J., "A General Computation Program for| I
Computation of Chemical Equilibrium Coipositions, Rocket Perform-
ance, and Chapman-Jouguet Detonations; Supplement I", NASA TND-1737,
Lewis Research Center, Cleveland, Ohio, October 1963.

5. Ross, J. L., "A Fuel Data Standardization Study for JP-4, JP-5, JP-7,
and RJ-5 Combusted in Air", AFAPL-TR-74-22, Wright-Patterson Air
Force Base, Ohio, March, 1974.

6. DeZubay, E. A., "Characteristics of Disk Controlled Flame", Aero

Digest, July 7, 1950, p. 54.

7. Longwell, J. P. and Weiss, M. A., "High Temperature Reaction Rates

in Hydrocarbon Combustion", Ind. Eng. Chem., Vol. 47, August 4, 1955,
p. 1634.

8. Zelinski, J. J., Mathews, L. J. III, Falk, F., and Bagnall, E. C.,

"The Design of Combustors for Ramjet Engines,", Combustion and
Flame, Vol. 4, 1960, p. 343.
g. Ozawa, R. I., "Survey of Basic Data on Flame Stabilization and
Propagation for High Speed Combustion Systems", AFAPL-TR-70-81,
Wright-Patterson Air Force Base, Ohio, 1971.

10. Levenspiel, 0., Chemical Reaction Engineering, John Wiley & Sons,
Inc., 1962.
11. Vincenti, W. G., and Kruger, C. H., Physical Gas Dynamics, John
Wiley & Sons, Inc., 1967, p. 210.

12. Edelman, R. and Fort.ine, 0., "A Quasi-Global Chemical Kinetic Model
for the Finite Ratc Combustion of Hydrocarbon Fuels", AIAA 69-86.

13. Grobman, J. S., "Comparison of Calculated ird Experimental Total

Pressure Loss and Airflow Distribution .r.Tubular Turbojet Combus-
tors with Tapered Liners", NASA TM-1l-26-58E, 1959.

72
AFAPL-TR-75-71

REFERENCES (Concluded)

14. Dittrich, R. T. and Graves, C. C., "Discharge Coefficients for Com-

bustor-Liner Air-Entry Holes", NACA Technical Note 3663, April, 1956.

15. Shapiro, A. II., The Dynamics and Thermodynamics of Compressible

Fluid Flow, Vol. 1, The Ronald Press Company, 1953.

16. Zucrow, M. J., Aircraft and Missile Propulsion, Vol. II, John Wiley
and Sons, Inc., 1964.

17. Benedict, R. P., Fundamentals of Temperature, Pressure, and Flow

Measurements, John Wiley and Sons, Inc., 1969.

18. Hornbeck, C. E., Can-Type Combustor Design and Evaluation for a Low
Cost Ramjet, Thesis: The Ohio State University, 1975.

73
*(~wg~q41
', i IPI'~ r~,r u