Introduction to the Method of

Characteristics and the Minimum

Length Nozzle
Stephen A. Whitmore
Mechanical and Aerospace Engineering Department

Material Taken from Anderson, Chapter 11, pp. 377-403

1
MAE 5540 - Propulsion Systems
 An Introduction to the
Two-Dimensional Method of Characteristics

• Anderson,
Chapter 11 pp. 377-403

2
MAE 5540 - Propulsion Systems
 “Method of Characteristics”
• Basic principle of Methods of Characteristics

-- If supersonic flow properties

are known at two points in a flow field,

-- There is one and only one set of properties

compatible* with these at a third point,

-- Determined by the intersection

of characteristics, or mach waves, from the two
original points.
*Root of term “compatibility equations”

3
MAE 5540 - Propulsion Systems
 “Method of Characteristics” (cont’d)
• Compatibility Equations relate the velocity magnitude
and direction along the characteristic line.

• In 2-D and quasi 1-D flow, compatibility equations are

Independent of spatial position, in 3-D methods, space
Becomes a player and complexity goes up considerably

• Computational Machinery for applying the method of

Characteristics are the so-called “unit processes”

• By repeated application of unit processes, flow field

Can be solved in entirety

4
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Field
θ + ν (M ) = Const ≡ K −
1
3 C+ θ − ν (M ) = Const ≡ K +

C-
2

• Conditions Known at Points {1, 2}

• Point {3} is at intersection of {C+, C-} characteristics

5
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Field (cont’d)

1 C+
3

C-
2

Point{1} → {M 1 ,θ1 } known →

γ + 1 −1 &( γ − 1 *(
ν1 =
γ −1
tan ' ( M 1
2
− 1) + − tan −1
M 2
1 −1
() γ + 1 (,

Along {C− } → θ1 + ν1 = const = ( K − )1

6
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Field (cont’d)

1 C+
3

C-
2

Point{2} → {M 2 ,θ 2 } known →

γ + 1 −1 &( γ − 1 *(
ν2 =
γ −1
tan ' ( M 2
2
− 1) + − tan −1
M 2
2 −1
() γ + 1 (,

Along {C+ } → θ 2 − ν 2 = const = ( K + )2

7
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Field (cont’d)

Mach and
1
3 C+
Flow Direction solved for at
2
C- Point 3
θ + ν (M ) = Const ≡ K −
θ − ν (M ) = Const ≡ K +

% (θ1 + ν1 ) + (θ 2 − ν 2 ) ( K − )1 + ( K + )2 (
' θ3 = = *
θ
% 1 + ν 1 = θ 3 + ν 3 ( ' 2 2 *
Point{3} → ' * →
&θ 2 − ν 2 = θ 3 − ν 3 ) ' ( θ1 + ν1 ) − (θ 2 − ν 2 ) ( K − )1 − ( K + )2 *
'ν 3 = = *
& 2 2 )
% γ + 1 −1 ,. γ − 1 0. (
M 3 = Solve 'ν 3 =
γ −1
tan - M 32 − 1( ) 1 − tan
−1 2
M 3 − 1*
'& ./ γ + 1 .2 *)
8
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Field (cont’d)

But where is
Point {3} ?

• {M,q} known at points {1,2,3}

---> {µ1,µ2,µ3} known

9
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Field (concluded)

But where is
Point {3} ?

• Slope of characteristics lines approximated by:

slope {C− } =
(θ1 − µ1 ) + (θ 3 − µ3 )
2 Intersection locates point 3

slope {C+ } =
(θ 2 + µ2 ) + (θ 3 + µ3 )
2
10
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Solve for {x3,y3}

" y3 − y1 $
& x − x = tan "# slope {C− } $% ' " y3 = ( x3 − x1 ) tan " slope {C− } $ + y1 $
& 3 1 '→ & # %
'
& y3 − y2 ' & y3 = ( x3 − x2 ) tan "# slope {C+ } $% + y2 '
& x − x = tan "# slope {C+ } $% ' # %
# 3 2 %
11
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Solve for {x3,y3}

- x1 ⋅ tan #$slope {C− }%& − x2 ⋅ tan #$slope {C+ }%& + ( y2 − y1 )

x3 =

tan #$slope {C− }%& ⋅ tan #$slope {C+ }%& ⋅ ( x1 − x2 ) + tan #$slope {C− }%& ⋅ y2 − tan #$slope {C+ }%& ⋅ y1
y3 =
tan #$slope {C− }%& − tan #$slope {C+ }%&

12
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example

M 1 = 2.0, θ1 = 10 , {x1 ,y1 } = {1.0, 2.0}

o

M 2 = 1.75, θ 2 = 5 , {x 2 ,y 2 } = {1.5,1.0}
o

13
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Point 1, compute

{ν , µ , ( K ) }
1 1 − 1

γ + 1 −1 $& γ − 1 (&
ν1 =
γ −1
tan % ( 2.0 2
− 1) ) − tan −1
2.0 2
− 1 = 26.37976 o

&' γ + 1 &*
180 −1 , 1 / o
µ1 = sin . = 30
π - 2.0 10
( K − )1 = θ1 + ν1 = 10 o + 26.37976 o = 36.37976 o
14
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Point 2, compute
{ν , µ , ( K ) }
2 2 + 2

γ + 1 −1 $& γ − 1 (&
ν2 =
γ −1
tan % (1.75 − 1) ) − tan −1 1.75 2 − 1 = 19.27319 o
2

&' γ + 1 &*
180 −1 , 1 / o
µ2 = sin . = 34.84990
π - 1.75 10
( K + )2 = θ 2 − ν 2 = 5 o − 19.27319 o = −14.27319 o
15
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Point 3 Solve for

{θ ,ν }
3 3

θ3 =
( K − )1 + ( K + )2
=
(
56.37976 o + −14.27319 o ) = 21.0533
deg. o

2 2

ν3 =
( K − )1 − ( K + )2
=
(
56.37976 o − −14.27319 o ) = 35.3265deg. o

2 2

16
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Point 3 Solve for

{ M 3 , µ3 }

+ π γ + 1 $& γ − 1 (& .
25.3265
M 3 = Solve - 35.3265
180
=
γ −1
−1
tan % ( )
M 32 − 1 ) − tan
−1 2
M 3 − 10
-, &' γ + 1 &* 0/

M3 = 1.96198
1
sin ( µ ) = ---> µ3= 30.6431o
M
17
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Locate Point 3

• Line Slope Angles

θ1 − µ1 ) + (θ 3 − µ3 ) (10 − 30 ) + ( 21.0533 − 25.2776 )

o o o o

slope {C− } =
( = = −12.1122 o deg
2 2
(θ 2 + µ2 ) + (θ 3 + µ3 ) ( ) (
5 o + 34.8499 o + 21.0533o + 25.2776 o ) = 43.0904
slope {C+ } = = o

2 2

deg

18
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Solve for {x3,y3}

x3= -

= 2.17091

19
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (cont’d)

• Solve for {x3,y3}

y3=

=1.57856

20
MAE 5540 - Propulsion Systems
 Unit Process 1: Internal Flow Example (concluded)

" M 1 % " 2.0 % " M 2 % "1.75 % " M 3 % "!2.3419

1.96198%$
$θ ' $10 o ' $θ ' $ 5 o ' $θ ' $#21.0533o '&
$ 1 '= $ '→ $ 2 '= $ ' → $ 3 ' = $#11.0533'&
$ x1 ' $ 1.0 ' $ x2 ' $1.5 ' $ x3 ' $#2.170912.2745 '&
$ ' $ ' $ ' $ ' $ ' $ '
# y1 & # 2.0 & # y2 & # 1 & # y3 & ##1.57856 1.726 &&
" %
21
MAE 5540 - Propulsion Systems
 Using MOC for Supersonic Nozzle Design

22
MAE 5540 - Propulsion Systems
 What happens when a nozzle expands
too quickly?
• i.e. …. A mess … for a given
Operating condition there is
only so fast we can expand a
Conventional Nozzle

23
MAE 5540 - Propulsion Systems
 Supersonic Nozzle Design

• Strategic contouring will “absorb” mach waves to give

isentropic flow in divergent section

24
MAE 5540 - Propulsion Systems
 Using Method of Characteristics to Design a Bell Nozzle
• This approach “prescribes” the expansion section of the nozzle,
and then uses M.O.C to design turning section to achieve wave
cancellation at wall …. And ensure isentropic flow

25
MAE 5540 - Propulsion Systems
 Supersonic Nozzle Design (cont’d)

• Rocket Nozzle
(Minimum Length)

• Bell Nozzle
(gradual expansion)

• Use compatibility eqs. to

design boundary with shock
free flow
26
MAE 5540 - Propulsion Systems
 Method of Characteristics
• Supersonic “compatibility” equations

• Apply along “characteristic lines” in flow field, and

insure isentropic flow …

27
MAE 5540 - Propulsion Systems
 Minimum Length Nozzle Design (cont’d)

• Along C+ characteristic {d,c, exit}

• Find minimum length nozzle
with shock-free flow àqexit = 0

• Along C- characteristic {a,c, exit}

C+

• Add
C-

28
MAE 5540 - Propulsion Systems
 Minimum Length Nozzle Design (cont’d)

• Find minimum length nozzle • Along C characteristic {a,c}

-
with shock-free flow at point a θc = 0

C+ • But from Prandtl-Meyer

expansion at point a
0

C-

29
MAE 5540 - Propulsion Systems
 Minimum Length Nozzle Design (cont’d)

But as already shown

C+
θw −ν a = 0
max

C-

30
MAE 5540 - Propulsion Systems
 Minimum Length Nozzle Design (concluded)

• Criterion for Minimum

Length Nozzle

31
MAE 5540 - Propulsion Systems
 Typical Conical Nozzle Contour
Y
N θnozzle
(X N,Y N)
Dt X De Dt

R1
LN R1 R1
e --> expansion ratio (Aexit/A*) θnozzle θnozzle
YN
Dt = Throat diameter XN
R1 = Radius of curvature of nozzle contraction
N = Transition point from circular contraction to conical nozzle
LN = Nozzle Length
De = Exit diameter

• R1 ~ 0.75Dt is typical
Credit: 32
MAE 5540 - Propulsion Systems Georgia Tech
 Typical Conical Nozzle Contour
(Cont’d)
Y
N θnozzle • Solve for Nozzle length
(X N,Y N)
Dt X De Dt in terms of other parameters

R1
LN

33
MAE 5540 - Propulsion Systems 33
 Typical Conical Nozzle Contour
• Using trig identities (Cont’d)

34
• R5540
MAE 1 ~ -0.75D t is Systems
Propulsion typical
 Minimum Length Conical Nozzle (1)
Y
N θnozzle
(X N,Y N)
Dt X De Dt

R1
LN
• Modify characteristic along C+ line from Cl to exit plane for non-zero Exit angle

• From earlier Minimum Length Nozzle derivation ,,,

35
MAE 5540 - Propulsion Systems
 Minimum Length Conical Nozzle (2)
Y
N θnozzle
(X N,Y N)
Dt X De Dt

R1
LN
• Simplify

“Two-thirds rule-of-thumb”
Applies strictly for conical nozzles
Generally applied as “safety factor”
for most nozzles
36
MAE 5540 - Propulsion Systems
 Minimum Length Conical Nozzle
• Example… given
Dthroat = 1 cm
Ae/A* = 8
g = 1.2
γ +1
A 1 )" 2 % " (γ − 1) 2 % , 2 (γ −1)
= 8.0 =
*
= +$ ' $ 1+ M '.
A M *# γ + 1& # 2 &-

1.2 + 1
## 2 $ # 1.2 − 1 2 $ $ 2 ( 1.2 − 1 )
!! " ! 1 + ( 3.122 )""
1.2 + 1 2 Mexit = 3.122
3.122

37
MAE 5540 - Propulsion Systems
 Minimum Length Conical Nozzle
(cont’d)

Mexit = 3.122

180 % % 1.2 + 1 & 0.5 % % 1.2 − 1 &

0.5&
0.5 &
= '# atan ' #
2 2
( 3.122 − 1 ) $ ( − atan ( ( 3.122 − 1 ) ) (
π # 1.2 − 1 $ # 1.2 + 1 $ $

ν exit
= 67.06° θ w Max = = 33.53°
2
Apply 2/3’rds rule

38
MAE 5540 - Propulsion Systems
 Minimum Length Conical Nozzle
(cont’d)

• R1 ~ 0.75Dt is typical …. R1=0.75 cm • Any shorter

and you have
“problems”

2.372 cm
1.6055 cm

1.0000 cm 2.8284 cm
=

= 2.372 cm
39
MAE 5540 - Propulsion Systems
 Comparison of Cone and Bell Nozzles

Credit: Georgia Tech 40

MAE 5540 - Propulsion Systems
 Bell Nozzle Contour Design

LN
e --> expansion ratio (Aexit/A*)

•
•
Credit: Georgia Tech 41
MAE 5540 - Propulsion Systems
 Bell Nozzle Contour Design (cont’d)
•
•
e --> expansion ratio (Aexit/A*)

qN

LN
• Boundary Conditions
• qe
Given
• qN
Credit: Georgia Tech 42
MAE 5540 - Propulsion Systems
 Bell Nozzle Contour Design (cont’d)
• Evaluate position
boundary condition at N

• Evaluate slope boundary condition at N

43
MAE 5540 - Propulsion Systems
 Bell Nozzle Contour Design (cont’d)

• Rearranging slope boundary condition at N

• Evaluate Slope Boundary condition at e

44
MAE 5540 - Propulsion Systems
 Bell Nozzle Contour Design (cont’d)

• Evaluate Position Boundary Condition at e

• And the Collection expressions are

45
MAE 5540 - Propulsion Systems
 Bell Nozzle Contour Design (cont’d)
1)
• 4 equations in 4 unknowns
2)
• Analytical Solution is a
3) Mess getting there .. But result
is OK
4)

46
MAE 5540 - Propulsion Systems
 SSME Nozzle example

47
MAE 5540 - Propulsion Systems
 SSME Nozzle example (cont’d)
• Fit with Parabolic bell profile

48
MAE 5540 - Propulsion Systems
 SSME Nozzle example (cont’d)
• Fit with Parabolic bell profile

BOUNDARY
CONDITIONS
qe =10°
qN =35 °
Dthroat =24.5 cm
Ae/A* =77.5
R1 =4.681cm

• Pretty
good model
49
MAE 5540 - Propulsion Systems
 SSME Nozzle example (Cont’d)

• Mexit = 4.677

180 % % 1.196 + 1 & 0.5 % % 1.196 − 1 2 &

0.5&
2 0.5 &
= '# atan ' # ( 4.677 − 1 ) $ ( − atan ( ( 4.677 − 1 ) ) (
π # 1.196 − 1 $ # 1.196 + 1 $ $

ν exit
= 102.34° θ w Max = = 51.17°
2

50
MAE 5540 - Propulsion Systems
 SSME Nozzle example (cont’d)
ν exit
θ w Max = = 51.17°
2
0
51.17

• SSME is
definitely not 35
0

a minimum length
nozzle

35/51.7 = 0.677

“two thirds rule”

51
MAE 5540 - Propulsion Systems
 SSME Nozzle example (cont’d)
ν exit • ~ “minimum length SSME Nozzle
θ w Max = = 51.17°
2

Rule of Thumb
Use qN < 2/3 qmax

“two thirds rule”

52
MAE 5540 - Propulsion Systems