0% found this document useful (0 votes)
275 views35 pages

Compressible Aerodynamics - Chapter10 - v2 PDF

This document discusses compressible flow through nozzles and diffusers. It covers governing equations for quasi-1D compressible flows, convergent-divergent nozzle flows, and diffusers. Key points include: the area-velocity relation for nozzle flows, choked mass flow in nozzles, effects of under-expanded and over-expanded flows, and how diffusers are used to slow supersonic flows but introduce entropy increases. Examples and problems are provided to demonstrate these concepts.

Uploaded by

Pythonraptor
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
275 views35 pages

Compressible Aerodynamics - Chapter10 - v2 PDF

This document discusses compressible flow through nozzles and diffusers. It covers governing equations for quasi-1D compressible flows, convergent-divergent nozzle flows, and diffusers. Key points include: the area-velocity relation for nozzle flows, choked mass flow in nozzles, effects of under-expanded and over-expanded flows, and how diffusers are used to slow supersonic flows but introduce entropy increases. Examples and problems are provided to demonstrate these concepts.

Uploaded by

Pythonraptor
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 35

Chapter 10

Compressible flows though nozzles

• Governing equations for quasi 1D flows


• Convergent – divergent nozzle flows
• Diffusers

1
Compressible flow through nozzles
Relevance
• Rocket and jet engines
• High-speed wind tunnels

Governing equations for quasi-1D flow


L dA
A2 x << 1
A A dx
Vy Vz
A1
Vx ¹ 0 << 1 << 1
Vx Vx
p, r , T , u uniform along each section
A = A ( x ) , p = p ( x ) , etc
2
Recall the governing equations for steady, inviscid, adiabatic
and quasi 1-D flow
rVA = const A
p1 A1 + r1V12 A1 + ò pdA = p2 A2 + r2V22 A2
A1
H = const
dr dV dA
From rVA = const + + =0
r V A
From momentum equation
( )(
pA + ρV 2 A + pdA = p + dp A + dA +… )
…( ρ + d ρ ) (V + dV ) ( A + dA)
2

Adp + AV 2 d ρ + ρV 2 dA + 2 ρVAdV = 0
3
from continuity ρVA = const
d ( ρVA) = 0 V ⋅ d ( ρVA) = 0
rV 2 dA + rVAdV + AV 2 d r = 0

combined with the previous result (momentum)


Adp + AV 2 d ρ + ρV 2 dA + 2 ρVAdV = 0

dp + rVdV = 0 Euler’s equation

From the energy equation


dH = 0 or dh + VdV = 0
The objective is to obtain a relation between dV dA
4
Starting from
dr dV dA
+ + =0
r V A
dp dp dr 2 dr
= =a
r dr r r
From Euler equation dp = -VdV
r
dr dr dV
a 2
= -VdV = -M 2
r r V

dA dV
= ( M - 1)
2
Area-Velocity relation
A V

5
dA dV
= ( M - 1)
2

Consequences: A V
For M < 1, dA < 0 For M < 1, dA > 0

V V V V

For M > 1, dA < 0 For M > 1, dA > 0

V V V V

At M = 1 even if dV ¹ 0
either a max or a min of A. Physically will
correspond to Amin
6
Convergent-divergent nozzle
supersonic nozzle (accelerator)

M <1 V increases M >1

supersonic diffuser (decelerator)

M >1 V decreases M <1

7
Nozzle flows
Objective: determine a relation starting from the Velocity-
Area relation

Amin = A* A
M ** = 1 * M
u =a u
Sonic throat generic section

8
from continuity r *u* A* = ruA
A r *u * r * a* r * r0 a*
= = =
A*
ru ru r0 r u
1

but r = æ 2 ö
* g -1

r0 çè g + 1 ÷ø
1
ρ0 ⎛ γ − 1 2 ⎞ γ −1
= ⎜ 1+ M ⎟
ρ ⎝ 2 ⎠

a* 1
M*
2
=
( g + 1) M 2
= *
u M 2 + ( g - 1) M 2
g +1
2
æ A ö IF 1 é 2 æ g - 1 2 ö ù g -1 Mach-Area relation
ç *÷ = 2ê ç1 + M ÷ú
è A ø M ëg +1è 2 øû given in Appendix A
9
æ Aö
M =Mç *÷
èA ø
and A = f ( M )
A*
1) M < 1
A
however at given
A* 2) M > 1
The solution depends on the boundary condition:
pressure ratio across the nozzle

10
Recall the Mach-Area relation g +1
2
æ A ö IF 1 é 2 æ g - 1 2 ö ù g -1
ç *÷ = 2ê ç1 + M ÷ú
è A ø M ëg +1è 2 øû

æ Aö
M = f çg , * ÷
è A ø
where always A > A*
and for a given value of A A* we have 2 possible solutions
1) M < 1 ; 2) M > 1

e.g. look into the table for IF


M = 0.31
at A * = 2
A M = 2.2
11
Consider a Laval nozzle
A = A( x) Ae
M =1 A*
inlet
p0 , T0 A = A*

assuming supersonic exit


M e6
M M =M A ( A)
*

p
-g
p0 isentropic flow p æ g -1 2 ö g -1
0.528 = ç1 + M ÷
pe 6 p0 è 2 ø
p0
T -1
T0 T æ g -1 2 ö
0.833 = ç1 + M ÷
Te6 T0 è 2 ø
T0
12
A pressure difference is needed to obtain flow in the nozzle
pe < p0
for isentropic flow pe must be exactly pe (IF)
6

What happens if pe ¹ pe6 ?

13
At pe
p0

1
M e3
M e2
M e1

p
pe1 p0
p0 pe2 p0
pe3 p0

14
When pe = pe3 At = A*
m = ρ ∗u∗ A∗ = ρ ∗u∗ At

If pe < pe3 still the throat conditions are sonic


m is fixed “choked mass flow regime”

m

pe
pe3 p0
15
When pe < pe3 and pe > pe6
NSW is formed either inside the nozzle or at its exit

p0 At NSW pe

1 NSW M e4

p
p0
pe4

16
When pe further decreases
NSW travels downstream

We call pe5 the pressure where a NSW is placed at the nozzle


exit

If pe6 < pe < pe5 a NSW is too strong and OSW appear
“over-expanded flow regime”

If pe < pe6 the flow needs to expand further to adapt the pressure
Prandtl-Meyer expansion fans appear
“under-expanded flow regime”
17
If pe6 < pe < pe5 over-expanded p = pe

pe < pe6 p = pe
If under-expanded

18
19
Summarizing
p
p0
subsonic flow pe
3

pe
5

pe
6

exit
Examples: 10.1, 10.2, 10.3

Problems: 10.2, 10.5, 10.7,10.8, 10.9


20
Diffusers
Any duct designed to slow down the flow to lower velocity is
a diffuser

The objective of the application of diffusers is to recover as


high as possible total pressure in the stream

The ideal diffuser performs an isentropic compression


Ds = 0 p0,2 = p0,1

M1 > 1 M =1 M2 <1
p0,2 = p0,1
A* s2 = s1
21
an isentropic deceleration is not realistic due to SW and
viscous effects (boundary layer)

Actual supersonic diffuser

At
M2 <1
M1 > 1 NSW
p0,2 < p0,1
(weak)
s2 > s1

Due to entropy increase across SWs


At in the real diffuser is larger than A*
22
Supersonic wind tunnels
Objective: generate a uniform supersonic stream in a laboratory
e.g. Model testing required at M = 2.5
1st choice
A
Take a convergent-divergent nozzle with e* = 2.64
p0 A
We need to establish = 17.1 to exit in the
pe
ambient at pe6
p A* Ae
p0
M e = 2.5 model
pe = 1 atm
pe
6

x* xe
23
1st choice
Ae
Take a convergent-divergent nozzle with * = 2.64
p0 A
We need to establish = 17.1 to exit in the ambient at pe6
pe

p A* Ae
p0
M e = 2.5 model
p0 > 17 atm
pe = 1 atm
pe
6

x* xe

Consequences:
-need for air storage at p0 > 17 atm
-large mass flow m = ρ ∗u∗ A∗

24
2nd choice
exit test section pb = 1 atm
throat
M e = 2.5 M2 <1
p0 = 2.4 atm p2 = pb = 1 atm
pe = 0.14 atm
NSW

The nozzle exhaust into a constant area duct.


Flow terminates with a NSW

Consequences:
-only p0 = 2.4 atm is needed reduction in mass flow
-NSW acts as a diffuser low efficiency of deceleration
25
3rd choice Laval nozzle + diffuser

nozzle test section diffuser

Me
p0 pe
model

Consequences:
less entropy production decelerating with multiple OSWs
p0 can be further lowered

26
pe
The required pressure ratio p0 can be obtained:

1) Pressurized storage vessel ( p0 > patm )


2) Vacuum at the outlet ( p0 = patm )

In reality losses due to BL have to be accounted for (especially


if separation occurs)

27
Design criteria for a supersonic wind tunnel
1) operating Mach number M e
simply by Ae
A*
Given the size of the test section Ae the first throat A1* or At1
is determined

2) Calculation of the second throat At2 (diffuser)

Consider m = ρuA = ρ1 u1 A1 = ρ2 u2 A2
∗ ∗ ∗ ∗ ∗ ∗

assuming sonic conditions in both throats

28
At2 r1*u1* r1*a1* AF r1*
= * *= * * = *
At1
r2 u2 r2 a2 r2
AF: adiabatic flow

p1* RT2* p1* p0,1 ( )


g
2 g -1
At2 g +1
= * *= *=
0,2 ( g +1 )
g
At1 p2 RT1 p2 p 2 g -1

At2 p0,1
=
At1 p0,2
The second throat must always be larger than the first one
p0,2 At2
If is known
p0,1 At1

29
p0,2
In a preliminary design phase one may assume from
p0,1
a NSW at M e
At2 p0,1
If < tunnel choking
At1 p0,2

Example: determine the SW position for a nozzle operating


with pe5 < pe < pe3

Consider that the nozzle exit is followed by another duct with


throat At2 with sonic conditions

30
* NSW e
A1 A2*

t1 t2
r1* a1* A1* = r 2*a2* A2*
æ r* T* ö æ r *
T * ö
ç r0 g R T0 A ÷ = ç r0 g R
* *
T0 A ÷
ç r0 T0 ÷ ç r0 T0 ÷
è ø1 è ø2

p0 = r0 RT0 ; ç ÷ = ç ÷ and ç ÷ = ç ö÷
æ r ö æ r ö æ ö æ
* * * *
T T
è r 0 ø1 è r 0 ø 2 è T0 ø1 è T0 ø2
A2* p0,1 T0,2 and for AF A2* p0,1
*
= *
=
A1 p0,2 T0,1 A1 p0,2
31
2 2 2
æ Ae ö æ r ö æ r0 ö T * T0
*

ç * ÷ =ç ÷ ç ÷
è A1 ø è r0 ø è re ø T0 M eTe
with p = r RT
2 2 2
æ Ae ö æ r ö æ p0 ö TeT * 1
*

ç * ÷ =ç ÷ ç ÷
A r
è 1ø è 0ø è eø 0 p T 2
M 2
e
-1
and with e = æ1 +
T g - 1 2ö
ç M e ÷
T0 è 2 ø
2 2
æ g -1 ö æ r ö æ p A ö T** *
M e2 ç1 + M e2 ÷ = ç ÷ ç 0 ÷
è 2 r
ø è 0 ø è e e ø T0
p A
g +1 2
æ g -1 2 ö æ 2 ö g -1 æ p0 A ö *
M ç1 +
2
Me ÷ = ç ÷ ç ÷
e
è 2 ø è g + 1 ø p A
è e eø
32
æ p0 A* ö
manipulating the expression we derive M e = M e ç g , , ÷
è pe Ae ø
⎡⎛ γ +1
1
2⎞ 2

AF
1 ⎢⎜ ⎛ 2 ⎞ γ −1 ⎛ A∗ p0 ⎞ ⎟ ⎥
Me = ⎢ 1+ 2 (γ − 1) ⎜ − 1⎥
2

γ − 1 ⎢⎜ ⎟ ⎜ ⎟
⎝ γ + 1⎠ ⎝ Ae pe ⎠ ⎟
⎝ ⎠ ⎥
⎢⎣ ⎥⎦
valid for both isentropic flow and adiabatic flow (with shocks)

Adiabatic nozzle flow


Even in presence of SWs we can obtain a relation returning the
Mach number M e
Useful in the range pe5 < pe < pe3

33
p
p0
pe p0
3

0.528 pe p0
5

pe p0
6

t ( *) e
In the only assumption of adiabatic flow (constant total
enthalpy)
ρeue Ae = ρ ∗u∗ A∗ = ρ ∗a∗ A∗

ue ue ae a0 Te T0
*
= *
= Me
a ae a0 a T0 T *
34
To determine shock position As and strength M s proceed as
follows:
pe Ae AF
known and * Me
p0,1 A1
IF Ae
known M e
A2*
*
Ae Ae A
known and 2
A1* A2*
A1*
A2* AF p0,1 NSW
known * Ms
A1 p0,2
IF As
known M s
A1*
35

You might also like