You are on page 1of 42

THERMODYNAMICS, S.A. Klein and G.F.

Nellis, Cambridge University Press, 2011

Chapter 16 COMPRESSIBLE FLOW


Nozzles were first discussed in Section 4.4.4 and the concept of a nozzle efficiency was
introduced in Section 6.6.4. The purpose of a nozzle is to increase the velocity of a fluid as
its pressure is reduced. Nozzles are used in many applications. They are used in gas and
steam turbine engines to allow the conversion of a high pressure gas into a high kinetic
energy flow stream that can then be converted into mechanical power by the blades of a
turbine. Nozzles are used directly to provide thrust in jet engines and rockets. In addition,
the unique behavior of nozzles makes them very useful for flow rate measurement. This
chapter provides a more detailed analysis of nozzle behavior than the discussion provided in
Section 6.6.4. Very high velocities can be achieved in a properly designed nozzle; it is even
possible to achieve velocities that greatly exceed the speed of sound.

16.1 Speed of Sound


The phenomenon that we perceive as sound is the result of small pressure variations or waves
propagating through a medium. We can determine the speed that sound can travel through a
medium by applying mass and momentum balances to a control volume.
The sound wave is caused by a disturbance, for example the motion of a piston as shown in
Figure 16-1. The disturbance propagates to the right in the figure at the speed of sound, c.
The fluid through which the disturbance has already passed (on the left side of the wave front
in Figure 16-1) has experienced a change in velocity ( V ), as well as a change in pressure
(P), temperature (T), and density ().
wave front moving at velocity c
disturbed fluid
T T , P P

c V

undisturbed fluid
T, P,
c
A

Figure 16-1: Disturbance in a fluid propagating to the right at speed c.

A system can be chosen that encompasses the wave front so that it moves with the wave at
velocity c. From the perspective of an observer that is moving with the wave, undisturbed
fluid with density steadily enters this system from the right at velocity c. Disturbed fluid
with density ( + ) exits this system on the left at velocity (c - V ). A steady-state mass
balance for the system is:

E20-1

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

A c V Ac m

(16-1)

where A is the cross-sectional area of the disturbance. Rearranging Eq. (16-1) shows that the
change in velocity is related to the change in density according to:


V c

(16-2)

The pressure wave that produces sound is ordinarily very small and it results in a small
relative change in density (/ << 1). Therefore, Eq. (16-2) shows that the change in the
velocity of the fluid through which the disturbance has passed is much lower than the speed at
which the disturbance propagates through the fluid ( V c ).
Newtons Second Law balances the forces on a system with the momentum flows into and out
of the system and the rate of momentum change in the system. The momentum of the fluid
passing through a control surface is the product of the fluid mass flow rate and velocity. A
momentum balance on the system defined around the wave front, shown in Figure 16-2,
balances the change in the pressure force with the momentum flows:

P P A m c V P A m c

(16-3)

wave front moving at velocity c


(P - P) A

m c V

PA

mc

Figure 16-2: A momentum balance on a system that encompasses and moves with the wave front.

Equation (16-3) can be simplified:

m
V
A

(16-4)

P c V

(16-5)

P
Substituting Eq. (16-1) into Eq. (16-4) provides:

The velocity change in the disturbed fluid, V , can be eliminated from Eq. (16-5) using Eq.
(16-2), which results in:

E20-2

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011


P c 2

(16-6)

Equation (16-6) can be rearranged:

c2

P
1

(16-7)

Taking the limit of Eq. (16-7) for a small disturbance, for which the relative change in density
approaches zero, results in a general relation for the speed at which the disturbance
propagates, i.e., the sound speed:

P P

1
0

c 2 lim

(16-8)

The right side of Eq. (16-8) is shown as a partial derivative because additional information is
needed to evaluate the derivative. The disturbance passes through the fluid quickly, which
does not allow time for heat to transfer. Since the disturbance itself is small, the process can
be approximated as adiabatic and reversible, and therefore isentropic. The partial derivative
in Eq. (16-8) should be taken at constant entropy.

P
c2

(16-9)

Solving Eq. (16-9) for the speed of sound provides:


P
c

(16-10)

The speed of sound can also be expressed in terms of specific volume using the chain rule:

P P dv


s v s d

(16-11)

Specific volume is the inverse of density; the derivative of specific volume with respect to
density is therefore:

dv
1
2 v 2

E20-3

(16-12)

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011
Substituting Eqs. (16-12) and (16-11) into Eq. (16-10) provides:

P
2
c
v
v s

(16-13)

The speed of sound defined by Eqs. (16-10) and (16-13) is a thermodynamic property that is
available in EES using the function SoundSpeed. The function SoundSpeed is accessed in the
same way as other thermodynamic functions in EES, such as Enthalpy or Entropy. For
example, the speed of sound in superheated steam at 300C and 250 kPa can be determined
using the following EES code:
$UnitSystem SI Mass Pa J K
T=converttemp(C,K,300 [C])
P=250 [kPa]*convert(kPa,Pa)
c=SoundSpeed(Water,T=T,P=P)

"temperature"
"pressure"
"sound speed"

which returns c = 584.5 m/s.


Equation (16-10) can be simplified for a fluid that obeys the ideal gas law. The ideal gas law
is:

RT
v

(16-14)

The chain rule in Eq. (10-104) is needed to transform the partial derivative in Eq. (16-13) into
a form with independent variables v and T, as provided in the ideal gas law. The partial
derivative of pressure with respect to specific volume at constant entropy is

P P T P v





v s
v s T v v s v T

(16-15)

Substituting Eq. (16-14) into Eq. (16-15) provides:

R T RT
P

2
v

v
v
v s

T v

(16-16)

P

v T

Equation (6-21) provides a relation for the specific entropy of an ideal gas in terms of
temperature and specific volume.

ds

cv
R
dT dv
T
v

E20-4

(16-17)

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011
If the process is isentropic then ds = 0 and Eq. (16-17) provides:
RT
T


cv v
v s

(16-18)

Substituting Eq. (16-18) into Eq. (16-16) results in:


R RT RT
P
2


v s v cv v v

(16-19)

Equation (16-19) can be rearranged:


RT R
P

2 1
v cv
v s

(16-20)

Recall from Eq. (6-26) that cP cv R . Inserting this relation into Eq. (16-20) leads to:

RT cP cv
RT
P
1 2

2
v cv
v
v s

cP

cv

(16-21)

or
RT
P

2 k
v
v s

(16-22)

where k is the ratio of cP to cv. Substituting Eq. (16-22) into Eq. (16-13) produces the final
result:
c k RT for an ideal gas

(16-23)

The speed of sound through water represented as an ideal gas (using the substance H2O in
EES) at 300C and 250 kPa can be evaluated using Eq. (16-23).
c_P=cP(H2O,T=T)
c_v=cv(H2O,T=T)
k=c_P/c_v
R=R#/MolarMass(H2O)
c_IG=sqrt(k*R*T)

"specific heat capacity at constant pressure"


"specific heat capacity at constant volume"
"ratio of specific heat capacities"
"ideal gas constant"
"sound speed, assuming ideal gas"

The sound speed determined with these equations is c = 586.6 m/s, which compares well with
the value of 584.5 m/s that is returned by the built-in function, SoundSpeed, for the substance

E20-5

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011
Water at the same conditions. The difference is due to the slight non-ideal gas behavior of

steam at these conditions.

16.2 Flow Behavior in Nozzles


The fluid velocity at the outlet of a nozzle may be lower than, equal to, or greater than the
local speed of sound, depending on the nozzle design. Fluid velocities that are lower than the
speed of sound are termed subsonic. Supersonic velocities are greater than the speed of
sound. A nozzle must have a converging-diverging geometry in order to achieve supersonic
velocities, as shown by the analysis in Section 16.3.2. The fluid velocity at the exit of a
converging nozzle is limited to the speed of sound. The flow behavior occurring in a nozzle
is not intuitive to most engineers; therefore, it is useful to qualitatively discuss this behavior
before we analyze it more formally.
inlet

exit
throat

Pressure
P in
a

critical pressure

c
d
e
f design pressure
g

Position
Figure 16-3: Sketch of pressure as a function of position for flow in converging-diverging nozzle with
various values of the nozzle exit pressure.

Figure 16-3 illustrates the pressure as a function of position for a compressible flow through a
converging-diverging nozzle. The geometry of the nozzle, which has a converging section
followed by a diverging section, is illustrated at the top of the figure. The different plots in
Figure 16-3 (labeled a through g) are each associated with a different operating condition.
Each operating condition has the same inlet pressure (Pin) but a different discharge (or back)
pressure. The discharge pressure is highest for case a and lowest for case g.
When the discharge pressure is slightly lower than the inlet pressure (corresponding to case
a), the flow is subsonic throughout the nozzle. In this case, the lowest pressure and highest

E20-6

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

velocity occur at the nozzle throat where the cross-sectional area reaches a minimum. As the
discharge pressure is decreased, the velocity at the throat continues to increase until
eventually it reaches the local speed of sound at the throat, corresponding to case b. The ratio
of the local velocity to the local speed of sound is called the Mach number, M:
M

V
c

(16-24)

The Mach number at the throat for case b is unity; this corresponds to the highest velocity that
a fluid can achieve in a converging nozzle. The highest back pressure for which the Mach
number is unity is referred to as the critical back pressure. When the Mach number is unity,
the nozzle is choked and the mass flow rate reaches its maximum value for the given set of
inlet conditions. Under these conditions, the mass flow rate is insensitive to the discharge
pressure. As the fluid progresses through the diverging section of the nozzle for case b, its
pressure slowly increases as the velocity decreases until the fluid exits the nozzle at the
critical back pressure. No discontinuities in pressure are observed for case b.
As the discharge pressure is reduced still further (case c in Figure 16-3), the mass flow rate
through the nozzle remains constant but the flow accelerates from the speed of sound at the
throat to supersonic velocities in the diverging section of the nozzle. At some point
downstream of the throat, the flow suddenly changes from being supersonic to being
subsonic. This change in velocity is accompanied by an abrupt increase in pressure. The
behavior occurring at this pressure discontinuity is referred to as a shock wave. The shock
wave occurring inside the nozzle is called a normal shock because it is perpendicular to the
flow direction. Note that shock waves can only occur at locations where the upstream flow is
supersonic. Downstream of the shock wave, the flow must be subsonic. The pressure
smoothly increases and the velocity decreases as the flow travels from the shock wave to the
nozzle exit. The flow emerges at subsonic velocity at the imposed back pressure.
The location of the normal shock moves towards the nozzle exit as the discharge pressure is
further reduced. Case d in Figure 16-3 illustrates the behavior for the back pressure value that
causes a shock wave exactly at the nozzle exit. If the back pressure is decreased still further,
as shown in case e, an oblique shock wave occurs outside of the nozzle as the pressure adjusts
from the supersonic condition at the nozzle exit to the imposed back pressure. The oblique
shock wave will, in general, not be perpendicular to the flow. This condition of a discharge
pressure that is lower than the imposed back pressure is referred to as overexpansion.
When the back pressure is reduced to the design pressure, corresponding to case f, supersonic
flow leaves the nozzle with an exit pressure that is exactly equal to the applied back pressure
and no shock occurs. The flow is isentropic throughout the nozzle. If the back pressure is
reduced to a value that is below the design pressure, corresponding to case g, an oblique

E20-7

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

expansion wave results. This behavior is not a pressure discontinuity but rather a continuous
wave that is a result of the flow exiting the nozzle coming to equilibrium with the low
external pressure. This situation is called under-expansion.

16.3 Thermodynamic Analysis of Isentropic Flow in Nozzles


The non-intuitive nozzle behavior that is illustrated in Figure 16-3 and discussed in Section
16.2 can be explained by applying mass, energy, and entropy balances to the flow in the
nozzle. It is also necessary to account for the compressibility (i.e., the density change) of the
fluid that results from changes in its pressure and temperature. Liquids flowing through
nozzles are nearly incompressible. However, the density of a gas may change significantly as
it flows through a nozzle.

16.3.1 Stagnation Properties


Nozzles typically operate at steady-state with one inlet and one outlet, as shown in Figure 164. We first consider a nozzle in which no shock wave occurs. Therefore, the outlet pressure
must either be equal to the critical pressure (if the outlet velocity is subsonic) or equal to the
design pressure (if the outlet velocity is supersonic).
inlet

exit
throat
Tout , Pout ,Vout

Tin , Pin ,Vin

Figure 16-4: Nozzle operating at steady state.

A mass balance on the nozzle requires that the mass flow rate at the inlet must be equal to the
mass flow rate at the exit:
m

Ain Vin Aout Vout

vin
vout

(16-25)

where Ain and Aout are the cross-sectional areas at the nozzle inlet and outlet, respectively.
The specific volume of the fluid can vary significantly between the inlet and outlet, depending
on the differences in temperature and pressure that occur across the nozzle. The specific
volume is a function of temperature and pressure. The fluid may or may not be described by

E20-8

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

the ideal gas law, but regardless, the specific volume can be determined if the temperature and
pressure are known.
vin f Tin , Pin

(16-26)

vout f Tout , Pout

(16-27)

The time required for the fluid to pass through the nozzle is typically very short and therefore
there is little opportunity for heat transfer and so we can assume that the nozzle operation is
adiabatic. A steady-state energy balance for these conditions is:
2

Vin2
Vout

m hin
m hout

2
2

(16-28)

Dividing Eq. (16-28) by the mass flow rate provides:


2
Vin2
Vout
hin
hout
2
2

(16-29)

It is customary to define a stagnation state in compressible flow applications. The stagnation


state is defined as the state that the fluid would achieve if its velocity were reversibly and
adiabatically reduced to zero. The stagnation state is defined by the entropy of the fluid,
which must remain constant during this reversible adiabatic process, and the stagnation
specific enthalpy, ho , which, according to an energy balance on the deceleration process, must
be:
V 2
ho h
2

(16-30)

Equation (16-28) shows that the stagnation specific enthalpy of the fluid passing through the
nozzle must be constant:
2

Vin2
Vout
ho hin
hout

2
2

(16-31)

For an ideal gas, specific enthalpy is only a function of temperature. Therefore, the stagnation
temperature (To) will also be constant within the nozzle for a fluid that obeys the ideal gas
law. The state of the fluid at the nozzle inlet is usually known. The velocity of the fluid at the
nozzle inlet is small and the corresponding kinetic energy at the inlet is often negligible. In

E20-9

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

this case, the stagnation inlet state is approximately the same as the actual inlet state. In any
case, the inlet state of the nozzle is known and the purpose of our analysis is to determine the
temperature, pressure and velocity of the fluid at the nozzle exit. Equation (16-28) provides
one equation with two unknowns (hout and Vout ).
An entropy balance on the nozzle under steady flow and adiabatic conditions is:
m sin Sgen m sout

(16-32)

The only irreversible processes occurring in adiabatic flow are fluid friction and shock waves.
Fluid friction can be minimized in a well-designed nozzle. If no shock waves occur then the
rate of entropy generation in the nozzle is nearly zero and the flow is therefore isentropic, as
indicated in Eq. (16-33). A consideration of the effects of shock waves, which cause the flow
to be non-isentrocpic, is provided in Section 16.7.

sin sout

(16-33)

The stagnation state is defined by the specific entropy and stagnation specific enthalpy. For
an isentropic nozzle, both of these quantities remain constant at any location within the
nozzle. Therefore, all of the stagnation properties, including the stagnation pressure of the
fluid (Po), are constant within the nozzle. Nozzle measurements indicate that the isentropic
flow model is reasonably accurate for a well-designed nozzle operating under conditions at
which no shock waves occur. If significant fluid friction or shock waves occur within the
nozzle then the specific entropy of the fluid will necessarily increase and therefore the
stagnation state will change in the nozzle. However, the stagnation specific enthalpy remains
constant even with these effects provided that the nozzle is adiabatic.

16.3.2 Explanation for Flow Behavior in the Diverging Section


The velocity of a fluid flowing in the converging section of a nozzle increases as the crosssectional area decreases. If the downstream pressure is sufficiently low, this velocity will
reach the local speed of sound at the location where the cross-sectional area is smallest (i.e.,
at the nozzle throat). The velocity can continue to increase into the supersonic regime, but
only if the nozzle geometry provides a diverging section downstream of the throat. The
explanation for this non-intuitive behavior can be understood from an examination of the
governing equations.
The energy balance on the nozzle, Eq. (16-31), assumes steady adiabatic flow and can be
written as:

E20-10

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

V 2 Vin2
h hin

0
2
2

(16-34)

where h is the specific enthalpy and V is the velocity of the fluid anywhere in the nozzle.
The specific enthalpy and velocity at the nozzle inlet are fixed. Therefore, the total derivative
of Eq. (16-34) is:

dh V dV 0

(16-35)

The fundamental property relation introduced in Eq. (6-16) is:

dh T ds v dP

dP

(16-36)

Since the flow is assumed to be isentropic, ds = 0 in Eq. (16-36). Equations (16-35) and
(16-36) are combined to provide:

dP

V dV 0

(16-37)

The mass flow rate at any point in the nozzle is:

m A V

(16-38)

where V and are the local velocity and density in the nozzle at a location where the crosssectional area is A. The mass flow rate is constant for steady flow, although the area,
velocity, and density all vary with position. The total derivative of Eq. (16-38) is then:

dm A V d A dV V dA 0

(16-39)

Sound speed is defined by Eq. (16-10):


P
c2
s

(16-40)

Since the flow is assumed to be isentropic in this analysis, the constraint in Eq. (16-40) is
satisfied and therefore:

E20-11

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

dP
for isentropic flow
c2

(16-41)

Substituting Eq. (16-41) into the first term on the right side of Eq. (16-39) and dividing the
entire equation by AV results in:

dP dV dA

0
A
c 2 V

(16-42)

Eliminating dP in Eq. (16-42) using Eq. (16-37) and introducing the definition of Mach
number, M V / c , provides the final result.

dA dV

M 2 1
A
V

(16-43)

No assumptions regarding ideal gas behavior or constant specific heat capacity were made in
deriving Eq. (16-43). Equation (16-43) shows that if the fluid velocity increases ( dV >0)
when under subsonic conditions (i.e., M < 1) then dA < 0 and therefore the nozzle geometry
must be converging (i.e., the cross-sectional area is decreasing). Alternatively, if the fluid
velocity is increasing under supersonic conditions (i.e., M > 1) then dA > 0 and the nozzle
geometry must be diverging. When the fluid velocity is equal to the speed of sound (M = 1)
then Eq. (16-43) indicates that dA = 0; therefore, the fluid velocity can only equal the local
sound speed at a location for which the cross-sectional area is at a minimum (i.e., the throat of
the nozzle).

16.4 Isentropic Flow Relations for Ideal Gases


The mass flow rate, stagnation specific enthalpy, and specific entropy of the fluid flowing
through an isentropic nozzle are constant:

AV
constant
v

(16-44)

V 2
ho h
constant
2

(16-45)

s constant

(16-46)

Equations (16-44), (16-45), and (16-46) can be solved analytically if it is assumed that the
fluid obeys the ideal gas law everywhere within the nozzle and also that the specific heat

E20-12

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

capacities, cP and cv, are constant. The solution provides the fluid state at any point in the
nozzle as a function of the local Mach number, M, defined in Eq. (16-24).
The relation between pressure and temperature for an ideal gas with constant specific heat
capacities undergoing an isentropic process is provided by Eq. (6-35):

T2 P2

T1 P1

k 1
k

(16-47)

where k is the ratio of specific heat capacities, cP/cv. A fluid that is reversibly and
adiabatically decelerated to its stagnation state (To, Po) undergoes an isentropic process;
therefore the temperature and pressure are related to the stagnation temperature and pressure
according to:

T P

To Po

k 1
k

(16-48)

The stagnation specific enthalpy is given by Eq. (16-30):

ho h

V 2
2

(16-49)

For an ideal gas with constant cP, Eq. (16-49) can be expressed in terms of temperatures:
V 2 2 cP To T

(16-50)

Equation (16-50) is expressed in terms of the Mach number:

V 2 2 cP To T
M 2
c
c2
2

(16-51)

For an ideal gas, the speed of sound is given by Eq. (16-23), which is substituted in Eq.
(16-51):

M2

2 cP To T
k RT

Equation (16-52) can be rearranged:

E20-13

(16-52)

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

kR 2
T 1
M To
2 cP

(16-53)

and solved for the ratio of the local temperature (T, sometimes referred to as the static
temperature) to the stagnation temperature (To):
T
kR 2
M
1
To 2 cP

(16-54)

Substituting R = cP - cv from Eq. (6-26) into Eq. (16-54) provides:


T k 1 2
1
M
To
2

(16-55)

The temperature ratio on the left side of Eq. (16-55) can be replaced by the ratio of the local
(static) pressure to the stagnation pressure using Eq. (16-48):
P k 1 2
1
M
2
Po

k / k 1

(16-56)

The critical area, A*, is defined as the cross-sectional area at which the local Mach number
(M) is unity (i.e., the flow velocity reaches the speed of sound). The mass flow rate
everywhere in the nozzle must be the same according Eq. (16-44):
m

A* V * AV

v*
v

(16-57)

where the superscript * indicates that the quantity is evaluated at the location associated with
the critical area. Solving Eq. (16-57) for the ratio of the area to the critical area provides:
A v V *

A* v* V

(16-58)

Substituting the ideal gas law into Eq. (16-58) provides:


A T P* V *

A* T * P V

(16-59)

By definition, the velocity at the critical area is equal to the local speed of sound at the critical
area:

E20-14

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

V * k RT *

(16-60)

The velocity at any location in the nozzle can be expressed as a product of the local Mach
number and the local speed of sound:
V M k RT

(16-61)

Substituting Eqs. (16-60) and (16-61) into (16-59) provides:

A
P*
T

*
A
M P T*

(16-62)

The pressure and temperature ratios in Eq. (16-62) can be expressed in terms of the Mach
number. The temperature ratio can be equivalently written as:

T / T
T
* o
*
T
T / To

(16-63)

Substituting Eq. (16-55) into Eq. (16-63) and recognizing the M is unity at the critical area
provides:

k 1 2
M
1
2
T

1
T*
k 1
1

(16-64)

The pressure ratio can be written as:

P / P
P
* o
*
P
P / Po

(16-65)

Substituting Eq. (16-56) into Eq. (16-65) provides:


k / k 1

k 1 2
M
1
2
P

*
k / k 1
P
k 1
1

E20-15

(16-66)

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Substituting Eqs. (16-64) and (16-66) into Eq. (16-62) provides the area ratio as a function of
the Mach number.
k 1

A
1

*
A
M

2 k 1 2 2 k 1
k 1 1 2 M

(16-67)

Figure 16-5 illustrates the area ratio, A/A*, as a function of the Mach number for k = 1.4.
Notice that the area ratio reaches a minimum value when M = 1. Therefore, if the Mach
number is greater than unity anywhere in the nozzle, the critical area will correspond to the
cross-sectional area at the throat. For subsonic nozzles or for flow after a shock wave where
the Mach number is less than unity, the critical area A* does not physically exist. Note that
there may be two real values of M, corresponding to subsonic and supersonic flow, that both
satisfy Eq. (16-67) for a specified area ratio.
3

Area ratio, A/A

2.5
2
1.5
1
0.5
0
0

0.5

1.5

2.5

Mach number

Figure 16-5: Ratio of the area to the critical area as a function of the Mach number for k = 1.4.

EXAMPLE 16.4-1: Ideal Gas Analysis of a Converging-Diverging Nozzle


A converging-diverging nozzle is designed to provide supersonic flow at its outlet. The
nozzle has a circular cross-sectional area. The diameters at various locations along the flow
direction are provided in Table 1. The diameter at the throat of the nozzle is Dthroat = 6.132
mm. Air is provided to the nozzle from a large supply tank at To = 25C and Po = 500 kPa.
Assume that air behaves as an ideal gas with constant specific heat capacities and a specific
heat capacity ratio of k = 1.4.

E20-16

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Table 1: Nozzle diameter as a function of position in the flow direction.

a.) Prepare plots of the Mach number, local (static) pressure, and local (static) temperature as
a function of position in the flow direction.
The inputs are entered in EES. Air enters the nozzle from a supply tank where the velocity is
low so the temperature and pressure in the supply tank are essentially equal to the stagnation
temperature and pressure.
$UnitSystem SI Mass J K Pa
D_throat=6.132 [mm]*convert(mm,m)
T_o=converttemp(C,K,25[C])
P_o=500 [kPa]*convert(kPa,Pa)
k=1.4 [-]

"throat diameter"
"stagnation temperature"
"stagnation pressure"
"specific heat capacity ratio"

The geometry data describing the nozzle in Table 1 are entered into a Lookup Table. Select
New Lookup Table from the Tables menu and specify 2 columns and 18 rows in the New
Lookup Table dialog (Figure 1). Name the table Nozzle.

Figure 1: New Lookup Table dialog.

E20-17

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Right click in the header of the first column of the table and change the name to x_mm and set
the units to mm. Change the name and units of the second column to D_mm and mm,
respectively. Enter all of the data into the table so that it appears as shown in Figure 2.

Figure 2: Lookup Table.

The cross-sectional area of the throat determines the critical area of the nozzle.

A*

2
Dthroat

A|star=pi*D_throat^2/4

4
"critical area"

The data in the Lookup table can be retrieved using the Lookup command, which reads data at
the specified row and column. The row can be indicated using the TableRun# function.
TableRun# is a built-in function (with no arguments) that returns the row in the Parametric
table for which the calculations are being done. (TableRun# returns 0 if the calculations are
not being done using a Parametric table.) The column can be indicated by providing its name
in quotes. The value of the variables x_mm and D_mm are obtained according to:
x_mm=Lookup('Nozzle',TableRun#,'x_mm')
D_mm=Lookup('Nozzle',TableRun#,'D_mm')

"position, in mm"
"diameter, in mm"

Note that if the equations are solved at this point without using the Parametric table, the value
of TableRun# will be zero. Row zero does not exist so the Lookup function will return the
values found in the first row of the table (x_mm = 0 mm and D_mm = 20.42 mm) and issue a
warning message. The diameter is converted to base SI units and used to compute the area:

E20-18

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

D2

D=D_mm*convert(mm,m)
A=pi*D^2/4

4
"diameter"
"area"

Equation (16-67) can be used to obtain the Mach number based on the ratio of the local area
to the critical area:
A
1

*
A
M

2 k 1 2
k 1 1 2 M

k 1
2 k 1

(1)

Equations (16-55) and (16-56) can be used to determine the local temperature and pressure
based on the Mach number and the stagnation temperature and pressure:

T k 1 2
1
M
To
2

P k 1 2
1
M
Po
2

k / k 1

A/A|star=((2/(k+1))*(1+(k-1)*M^2/2))^((k+1)/(2*(k-1)))/M "Mach number"


T=T_o*(1+(k-1)*M^2/2)^(-1)
"static temperature"
P=P_o*(1+(k-1)*M^2/2)^(-k/(k-1))
"static pressure"

Solving the EES code at this point may result in either M = 0.05227 or M = 4.038, depending
on the guess value used for the Mach number. The first entry in the Lookup table is clearly in
the converging section of the nozzle and therefore the first (subsonic) solution is correct.
However, there are two roots to Eq. (1), as evident in Figure 16-5. We can control which root
is returned by setting an appropriate guess value for M. Define a new variable, M_guess, and
set its value to a number that is less than 1.
M_guess=0.1 "guess value for the Mach number"

Select Variable Info from the Options window. Scroll down to the information for the
variable M and set the guess value for M to be the variable M_guess, as shown in Figure 3.
Using a variable to set the guess value is convenient because it facilitates the control over
which root is returned by EES.

E20-19

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Figure 3: Setting the guess value using a variable in the Variable Information dialog.

The local speed of sound is computed using Eq. (16-23):


c k RT

and the local velocity is obtained according to:

V M c
The specific volume is obtained from the ideal gas law:

RT
P

AV
v

The mass flow rate is:

R=R#/MolarMass(Air)
c=sqrt(k*R*T)
Vel=c*M
v=R*T/P
m_dot=Vel*A/v

"gas constant"
"local speed of sound"
"velocity"
"specific volume"
"mass flow rate"

Generate a Parametric table with 18 rows (one for each of the rows in the Lookup table) and
include columns for the variables x_mm, T, P, M, m_dot, and M_guess. Comment out the value
of M_guess that is set in the Equations window:
{M_guess=0.1

"guess value for the Mach number"}

E20-20

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Set the value of M_guess in the Parametric table so that EES returns the correct Mach number.
For the converging section (rows 1-5) set a low value of M_guess and in the diverging section
(rows 7-18) set a high value. The Mach number at row 6 is one. Run the table in order to
obtain the results shown in Figure 4.

Figure 4: Parametric table.

Note that Mach number is less than unity in the converging section and greater than unity in
the diverging section, as it should be. The mass flow rate is constant at all positions. Figure 5
illustrates the Mach number as a function of position and Figure 6 illustrates the pressure and
temperature as a function of position.
1.8
1.6

Mach number

1.4
1.2
1
0.8
0.6
0.4
0.2
0
0

10

15

20

Position (mm)

Figure 5: Mach number as a function of position.

E20-21

25

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

5.0x105

300

4.5x105

280
260

3.5x105
3.0x105

240

2.5x105
2.0x10

1.5x10

temperature

220

Temperature (K)

Pressure (Pa)

4.0x105

pressure

200

1.0x105
0

10

15

20

180
25

Position (mm)

Figure 6: Pressure and temperature as a function of position.

16.5 Flow Measurement using Nozzles


Nozzles that are used for flow measurement ordinarily consist only of a converging section.
The reason nozzles are used for flow measurement is that an analytical equation can be
determined for the flow rate that requires only knowledge of the throat diameter, the nozzle
inlet and outlet pressures, and the inlet air density. If sonic velocity occurs at the nozzle
throat (i.e., the nozzle is choked), then mass flow rate is independent of the outlet pressure
and therefore the inlet pressure, temperature, and throat diameter completely determine the
mass flow rate.
The analytical equations for the mass flow rate can be derived using the isentropic flow
equations, assuming ideal gas behavior with constant specific heat capacities. The mass flow
rate at any point in the nozzle is:
P
m AV A Mc
constant
RT

(16-68)

We will first determine an expression for the mass flow rate at the critical area, A*, which is
the area at which the Mach number is 1. If the nozzle is choked, the critical area is the area at
the throat of the nozzle; note that the throat of a converging nozzle is the exit area. If the
back pressure is sufficiently high so that sonic velocity is not achieved at the exit, then the
critical area does not physically exist. However, the critical area will be useful for deriving
an expression for the mass flow rate in any case.

E20-22

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

At the critical area, M =1 and the pressure and temperature at this point are designated as P*
and T*. The local speed of sound at the location of the critical area, from Eq. (16-23), is

k RT * . Substituting this information into Eq. (16-68) allows the mass flow rate to be
expressed as:
m A*

k *
P
RT *

(16-69)

Assuming isentropic flow, the temperature and pressure at any point in the nozzle are given
by Eqs. (16-55) and (16-56). Applying these equations at the critical area (for which M = 1)
determines T * and P* :

T * k 1
1

To
2
P* k 1
1

2
Po

(16-70)

k / k 1

(16-71)

Substituting Eqs. (16-70) and (16-71) into Eq. (16-69) and rearranging results in:
m A

2 k RTo 1 k 1 k
Po
1 k 2 RTo 2 2

k /( k 1)

(16-72)

If the nozzle is choked (i.e., M exit =1) at the exit of the converging section, then A* =Aexit and
the mass flow rate is completely determined for a known fluid in terms of the Aexit and the
stagnation temperature (To) and pressure (Po), which are the conditions upstream of the
nozzle. Equation (16-71) shows that the pressure at the exit of the nozzle has no effect on the
mass flow rate, which is the basis of the term, choked flow. For a fixed inlet condition, the
mass flow rate is at its maximum possible value under these conditions.
If the flow is subsonic at the nozzle exit then Aexit > A* . We can determine the Mach number
at the nozzle exit, M exit , knowing the exit pressure of the nozzle. Assuming isentropic flow,
the Mach number at the nozzle exit is provided by algebraic rearrangement of Eq. (16-56):

M exit

MIN 1,

( k 1)

2
Pexit

1
Po
k 1

E20-23

(16-73)

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Note, that if the calculated value of M exit is greater than one, it must be set to one as indicated
in Eq. (16-73) since the maximum possible value of Mach number for a converging nozzle is
1. The critical area, A* , can be calculated as a function of M exit with Eq. (16-67):
k 1

2
Aexit
1 2 k 1 M exit 2 k 1

A*
M exit
k 1

(16-74)

If M exit =1 then Aexit = A* . If M exit <1 then Aexit > A* and a cross-sectional area in the nozzle
equal to A* does not physically exist. The mass flow rate through the nozzle is still given by
Eq. (16-72), but in this case it is convenient to eliminate A* . This elimination is
accomplished by solving Eq. (16-74) for A* and substituting it into Eq. (16-72), which results
in:
m

Aexit
2 k RTo
1 k
M exit
k 1 Po
2 R To
k 1
2 2

k
k 1

2M k M

k 1

2
exit

2
exit

k 1
2( k 1)

(16-75)

Equation (16-75) will provide the mass flow rate for both choked and unchoked flow. For
choked flow, M exit =1. For unchoked flow, M exit depends on the exit pressure of the nozzle.
The mass flow rate depends only on the pressure and temperature upstream of the nozzle, the
pressure at the nozzle exit which determines the exit Mach number with Eq. (16-73), and the
cross-sectional area of the nozzle exit. All of these quantities can easily be measured, which
explains why nozzles are useful for mass flow measurement.

EXAMPLE 16.5-1: Mass Flow in a Converging Nozzle


Air (k = 1.4) enters a converging nozzle from a plenum tank at To = 25C and Po = 500 kPa.
The diameter at the exit of the nozzle is Dexit = 6.131 mm.
a.) Prepare a plot of the mass flow rate of air through the nozzle as a function of the nozzle
exit pressure for exit pressures ranging from 490 kPa to 100 kPa.
Enter the known information into EES.
$UnitSystem SI K Pa J
T_o=convertTemp(C,K,25 [C])
P_o_kPa = 500 [kPa]*convert(kPa,Pa)
P_o = P_o_kPa*convert(kPa,Pa)
k = 1.4
R = R#/MolarMass(Air)
D_exit=6.131 [mm]*convert(mm,m)

"stagnation temperature"
"stagnation pressure in kPa"
"stagnation pressure"
"isentropic index for air"
"gas constant for air"
"diameter of nozzle exit"

E20-24

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

The exit area of the nozzle is computed:

Aexit

2
Dexit

A_exit=pi*D_exit^2/4

(16-76)

"area at nozzle exit"

The pressure at the exit of the nozzle will be varied in a table. For now, set it to any
reasonable value.
P_exit_kPa=400 [kPa]
P_exit=P_kPa*convert(kPa,Pa)

"exit pressure in kPa"


"exit pressure"

The Mach number at the outlet of the nozzle is calculated with Eq. (16-73). Note that the
nozzle geometry is converging and so the maximum possible Mach number is 1. The Min
function in EES limits the value of the calculated Mach number to 1.
M_exit=min(1, sqrt(((P_exit/P_o)^(-(k-1)/k)-1)*2/(k-1)))

"Mach number at exit"

With M exit known, the mass flow rate can be directly determined with Eq. (16-75).
m_dot=A_exit/(2*R*T_o)*M_exit*sqrt((2*k*R*T_o)/(k+1))*(k+1)*P_o*(1/2+k/2)^(-k/(k-1))&
*((2+M_exit^2*k-M_exit^2)/(k+1))^(-(k+1)/(2*(k-1)))
"mass flow rate"

Comment out the specification for the exit pressure.


{P_exit_kPa=400 [kPa]

"exit pressure in kPa"}

Create a Parametric table with columns for P_exit_kPa, m_dot and M_exit. Fill the P_exit_kPa
column with values between 490 kPa and 110 kPa. A plot of the mass flow rate is shown in
Figure 1. Note that the mass flow rate reaches a maximum at an exit pressure of about 270
kPa, at which point the exit Mach number becomes 1.

E20-25

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Mass flow rate (kg/s)

0.035
0.03
0.025
0.02
0.015
0.01
0.005
100

150

200

250

300

350

400

450

500

Exit pressure (kPa)


Figure 1: Mass flow rate as a function of exit pressure.

b.) Prepare a plot of the mass flow rate as a function of the pressure at the inlet of the nozzle
for inlet pressures ranging from 500 kPa to 110 kPa with the exit pressure of the nozzle
equal to 100 kPa. Explain the behavior of this plot.
Set the exit pressure to 100 kPa and comment out the equation that set the inlet stagnation
pressure (or use a $IF directive as shown).
$ifnot ParametricTable='Table 2'
P_o_kPa=500 [kPa]

"stagnation pressure at inlet in kPa"

$endif
P_exit_kPa=100 [kPa]

"exit pressure in kPa"

Set P_o_kPa to range from 500 kPa to 110 kPa in a new Parametric table. Include m_dot and
M_exit in the Parametric table. Solve and plot the mass flow rate as a function of the inlet
pressure. Also plot the Mach number at the exit as a function of the stagnation pressure.

E20-26

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011
0.035

1
0.9
Mach number

0.025

0.8

0.02

0.7
mass flow rate

0.015

0.6

0.01

0.5

0.005

0.4

0
100

150

200

250

300

350

400

450

Mach number at exit

Mass flow rate (kg/s)

0.03

0.3
500

Inlet stagnation pressure (kPa)


Figure 2: Mass flow rate and exit Mach number as a function of inlet stagnation pressure

Note that the mass flow rate changes with changing inlet stagnation pressure, even when
choked flow (Mach number = 1) occurs at the nozzle exit. This behavior occurs because the
density of the air entering the nozzle varies with changes in the inlet stagnation pressure.

16.6 Analysis of Isentropic Nozzles with Real Gases


The thermodynamic analysis provided in Sections 16.4 and 16.5 assumes that the fluid
flowing through a nozzle is an ideal gas with constant specific heat capacity. As the pressure
increases, the behavior of all gases departs from the ideal gas model. In addition, large
temperature changes can occur within a nozzle, as shown in Figure 6 of Example 16.4-1.
Therefore, the assumption of constant specific heat capacities may not be applicable even if
the ideal gas model remains valid. Although it greatly simplifies the solution, it is not
necessary to assume ideal gas behavior with constant specific heat capacities. The governing
relations for steady isentropic flow in a nozzle, Eqs. (16-25) through (16-33), can be solved
for any fluid. In this section, we will prepare a general Subprogram in EES that solves these
equations for real fluids flowing through a reversible, adiabatic nozzle.
A Subprogram in EES is a separate EES program that communicates with the main EES
program through its parameter list. In this section we will develop the Subprogram
IsentropicFlow , which will provide the pressure, temperature, and Mach number for a real
fluid in an isentropic and adiabatic converging-diverging nozzle. The Subprogram
IsentropicFlow takes as inputs a string variable containing the name of the gas (Gas$), the inlet
temperature (T[1]), inlet pressure (P[1]), inlet velocity (Vel[1]), the nozzle cross-sectional area

E20-27

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

at the throat (A|star), the cross-sectional area at any location in the nozzle (A[2]), and a guess
value for the Mach number at this location (M2_guess). The purpose of this guess value will
be explained shortly. Note that state 2 does not necessarily designate the outlet of the nozzle,
but rather any specified location within the nozzle. The desired outputs of the Subprogram
are the temperature (T[2]), pressure (P[2]) and Mach number (M[2]) at the location where the
area is A[2].
Subprogram IsentropicFlow(Gas$, T[1], P[1], Vel[1], A|star, A[2], M2_guess: T[2], P[2], M[2])

We are assuming that Gas$ is a real fluid (not an ideal gas) so that the specific enthalpy and
the sound speed depend on both temperature and pressure. The temperature and pressure at
the inlet are known and they can be used directly to calculate the specific enthalpy and
specific entropy at the nozzle inlet (state 1).
h[1]=enthalpy(Gas$, T[1],P[1])
s[1]=entropy(Gas$,T=T[1], P=P[1])

"specific enthalpy at nozzle inlet"


"specific entropy at nozzle inlet"

The mass flow rate is constant throughout the nozzle. However, the mass flow rate is not
provided as an input. The mass flow rate can be determined at the throat, where the velocity
must be equal to the sound speed (M = 1) if supersonic velocities occur in the diverging
section. Properties at the nozzle throat are designated with the star superscript. The entire
nozzle is assumed to be isentropic. An energy balance and entropy balance determine the
specific enthalpy and specific entropy at the throat, which allow the temperature (T|star),
pressure (P|star), specific volume (v|star), and sound speed (c|star) at the throat to be
determined. The mass flow rate at the throat is calculated knowing the cross-sectional area of
the nozzle throat.
h[1]+Vel[1]^2/2 =h|star+c|star^2/2
throat"
h|star=enthalpy(Gas$,T=T|star,P=P|star)
s|star=s[1]
throat"
s|star=entropy(Gas$,T=T|star,P=P|star)
c|star=soundSpeed(Gas$,T=T|star,P=P|star)
v|star=volume(Gas$,T=T|star,P=P|star)
m_dot=A|star*c|star/v|star

"energy balance between the inlet and


"specific enthalpy at throat"
"entropy balance between the inlet and
"specific entropy at throat"
"velocity at throat - assumes M=1"
"specific volume at throat"
"mass flow rate of fluid"

We need to calculate the temperature, pressure and Mach number at state 2. Three
independent relations are needed to determine these three properties. The mass flow rate at
the specified location (state 2) is the same value as the mass flow rate at the throat, providing
one relation. The remaining two relations are provided by an energy balance and an entropy
balance between states 1 and 2.
m_dot=A[2]*M[2]*c[2]/v[2]
h[1]+Vel[1]^2/2 =h[2]+Vel[2]^2/2
h[2]=enthalpy(Gas$,T=T[2],P=P[2])

"same mass flow rate at state 2"


"energy balance on adiabatic nozzle"
"specific enthalpy at state 2"

E20-28

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011
s[2]=s[1]
s[2]=entropy(Gas$,T=T[2],P=P[2])
v[2]=volume(Gas$,T=T[2],P=P[2])
c[2]=soundSpeed(Gas$,T=T[2],P=P[2])

"entropy balance on isentropic nozzle"


"specific volume at state 2"
"specific volume at state 2"
"local sound speed"

The equations in this subprogram are much more difficult to solve than the corresponding
isentropic ideal gas flow equations in Eqs. (16-55), (16-56), and (16-67) because they require
solution of the rather complicated equation of state that relates the thermodynamic properties
for real gases. In addition, there are two solutions for the outlet Mach number given the area
of the nozzle at the outlet (corresponding to subsonic and supersonic flows).
The only way to ensure that the calculations return the desired solution is to provide an
appropriate guess value and appropriate limits for the Mach number. We will use the Mach
number provided to the Subprogram (M2_guess) as a guess value for the real-gas solution. A
good way to obtain a guess value for M2 is to use the ideal gas solution provided in Section
16.4. The value of the variable M2_guess needs to be assigned as the guess value for variable
M[2] in the IsentropicFlow Subprogram. However, setting an appropriate guess value for Mach
number does not necessarily guarantee that the correct solution will be obtained. It is also
necessary to set appropriate limits for M2. The variable M_low is the lower limit for the Mach
number and is defined with the IF function. The IF function sets the value of M_low to 0.01 if
M2_guess is less than 1 and to 1.0001 if M2_guess is greater than 1. Variable M_low must be
specified to be the lower limit for the variable M[2]. The assignment of the lower limit and
guess value can be accomplished using the Variable Information dialog. Select the variables
for the IsentropicFlow Subprogram using the drop down control at the top center of the dialog.
Set the guess value for M[2] to be M2_guess and set the lower limit for M[2] to be M_low. Also,
to help with the solution, set the lower limit and upper limits for P[2] and P|star to 1x105 Pa
and P[1], respectively and set the guess values of these variables to 1x105 Pa. Lower and
upper limits of 100 K and T[1] are appropriate for the variables T[2] and T|star with a guess
value of T[1]. Set lower limits of 1 m/s for the variables c[2] and Vel[2], as shown in Figure 166. The IsentropicFlow Subprogram is used in Example 16.6-1 to compare the results of ideal
and real gas analysis of an isentropic nozzle.

E20-29

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Figure 16-6: Variable Information dialog window for the IsentropicFlow Subprogram.

EXAMPLE 16.6-1: Real Gas Analysis of an Isentropic Nozzle


a.) Determine the pressure, temperature and Mach number at the outlet of the nozzle
described in Example 16.4-1 using real gas thermodynamic properties and compare the
results with those obtained using the ideal gas, constant specific heat capacity
assumptions in Example 16.4-1.
It is convenient to start with the EES program developed for Example 16.4-1 for this example,
since it provides the Lookup table with the nozzle geometry as well as the ideal gas analysis.
Open the EES code for Example 16.4-1 and save it with a different name. The IsentropicFlow
Subprogram is placed at the top of the Equations window. Enter the guess values, limits and
units for the variables in the subprogram as shown in Figure 16-6. A listing of the
Subprogram and the ideal gas analysis is shown below.
$UnitSystem SI Mass J K Pa
Subprogram IsentropicFlow(Gas$, T[1], P[1], Vel[1], A|star, A[2], M2_guess: T[2], P[2], M[2])
h[1]=enthalpy(Gas$, T=T[1],P=P[1])
"specific enthalpy at nozzle inlet"
s[1]=entropy(Gas$,T=T[1], P=P[1])
"specific entropy at nozzle inlet"
h[1]+Vel[1]^2/2 =h|star+c|star^2/2
"energy balance between the inlet and
throat"
h|star=enthalpy(Gas$,T=T|star,P=P|star)
"specific enthalpy at throat"
s|star=s[1]
"entropy balance between the inlet and
throat"

E20-30

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011
s|star=entropy(Gas$,T=T|star,P=P|star)
c|star=soundSpeed(Gas$,T=T|star,P=P|star)
v|star=volume(Gas$,T=T|star,P=P|star)
m_dot=A|star*c|star/v|star
m_dot=A[2]*M[2]*c[2]/v[2]
h[1]+Vel[1]^2/2 =h[2]+Vel[2]^2/2
h[2]=enthalpy(Gas$,T=T[2],P=P[2])
s[2]=s[1]
s[2]=entropy(Gas$,T=T[2],P=P[2])
v[2]=volume(Gas$,T=T[2],P=P[2])
c[2]=soundSpeed(Gas$,T=T[2],P=P[2])
M_low=if(M2_guess,1,0.01,1,1.0001)
M[2]=Vel[2]/c[2]
end

"specific entropy at throat"


"velocity at throat - assumes M=1"
"specific volume at throat"
"mass flow rate of fluid"
"same mass flow rate at state 2"
"energy balance on adiabatic nozzle"
"specific enthalpy at state 2"
"entropy balance on isentropic nozzle"
"specific volume at state 2"
"specific volume at state 2"
"local sound speed"
"set a lower limit on Mach number"
"Mach number at state 2"

"Ideal gas analysis from Example 16.4-1"


D_throat=6.132 [mm]*convert(mm,m)
"throat diameter"
T_o=converttemp(C,K,25[C])
"stagnation temperature"
P_o=500 [kPa]*convert(kPa,Pa)
"stagnation pressure"
k=1.4 [-]
"specific heat capacity ratio"
A_star=pi*D_throat^2/4
"critical area"
x_mm=Lookup('Nozzle',TableRun#,'x_mm')
"position, in mm"
D_mm=Lookup('Nozzle',TableRun#,'D_mm')
"diameter, in mm"
D=D_mm*convert(mm,m)
"diameter"
A=pi*D^2/4
"area"
A/A_star=((2/(k+1))*(1+(k-1)*M^2/2))^((k+1)/(2*(k-1)))/M "Mach number"
T=T_o*(1+(k-1)*M^2/2)^(-1)
"static temperature"
P=P_o*(1+(k-1)*M^2/2)^(-k/(k-1))
"static pressure"
{M_guess=0.1
"guess value for the Mach number"}
R=R#/MolarMass(Air)
"gas constant"
c=sqrt(k*R*T)
"local speed of sound"
Vel=c*M
"velocity"
v=R*T/P
"specific volume"
m_dot=Vel*A/v
"mass flow rate"

Next, we will add the real gas analysis to this code for the fluid Air_ha. The fluid Air_ha
provides high accuracy real gas properties for air. The real gas analysis relies on a call to the
IsentropicFlow Subprogram. The inlet temperature (T[1]) and pressure P[1]) of the air are set to
the stagnation conditions, as was assumed in Example 16.4-1. The inlet velocity at the
stagnation state is approximately 0 m/s. The cross-sectional area at the throat (A|star) and at
each axial position in the Lookup table (A[2]) are provided as inputs to the IsentropicFlow
subprogram along with M_guess, which is the guess for the Mach number at the specified
position that was used in the ideal gas analyses in Example 16.4-1. M_guess will be set to
M2_guess in the Subprogram and used as a guess value for M[2]. The Call statement relays
these parameters to the IsentropicFlow Subprogram. The additional EES code needed for the
real gas analysis is entered below the code for the ideal gas analysis.
"Real gas analysis of the nozzle"
A[2]=A
"area at position 2"
call IsentropicFlow('Air_ha', T_o, P_o, 0 [m/s], A_star, A[2], M: T[2], P[2], M[2])

E20-31

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

A Parametric table was used in Example 16.4-1 to do the calculations for each of the 18 axial
positions indicated in the Lookup table. Add three columns to this existing Parametric table
for the variables T[2], P[2], and M[2]. Solve the Parametric table. The calculated temperature,
pressure and Mach number are overlaid on the plots prepared in Example 16.4-1. The real
gas results are very similar to the results obtained assuming ideal gas with constant specific
heat capacity.
1.8
1.6

Mach number

1.4
1.2
1
0.8
0.6
Ideal gas
Real gas

0.4
0.2
0
0

10

15

20

25

Axial position (mm)


Figure 1: Mach number as a function of position for ideal and real gas solutions with air.
5.0x105

300
Ideal gas
Real gas

4.5x105

260

3.5x105
3.0x105

240

2.5x105

temperature

220

2.0x105
pressure

200

1.5x105
1.0x105
0

10

Temperature (K)

Pressure (Pa)

4.0x10

280

15

20

180
25

Position (mm)
Figure 2: Static temperature and static pressure as a function of position for ideal and real gas solutions
with air and a 500 kPa inlet pressure.

The stagnation conditions in this example were set to Po = 500 kPa and To = 25C; these are
conditions for which air closely follows the ideal gas law. Therefore, it is not surprising that
the results of the ideal and real gas analyses shown in Figures 1 and 2 are similar. Differences

E20-32

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

between the ideal gas and real gas analysis become more apparent when the inlet pressure is
increased to 10 MPa. Change the value of the variable P_o from 500 kPa to 10 MPa.
{P_o = 500 [kPa]*convert(kPa,Pa)
P_o=10 [MPa]*convert(MPa,Pa)

"stagnation pressure"}
"stagnation pressure"

Solve the Parametric table again. Figure 3 illustrates the resulting static temperature and
static pressure as a function of position and shows small differences between the ideal gas and
real gas analyses.
7

300

10

Ideal gas
Real gas

9x10

280
260

7x10

240

6x10

5x10

temperature

220

4x10

pressure

200

3x10

2x10

Temperature (K)

Pressure (Pa)

6
8x10

10

15

20

180
25

Position (mm)
Figure 3: Static pressure and static temperature as a function of position for ideal and real gas solutions
for air with an inlet pressure of 10 MPa.

16.7 Analysis of Nozzle Flow with Shock Waves


The discharge pressures for cases c and d shown in Figure 16-3 are lower than the critical
back pressure and therefore supersonic flow occurs in the diverging section of the nozzle.
However, at some point downstream of the throat, the flow suddenly switches from
supersonic to subsonic. At this point, the temperature and pressure of the fluid exhibit an
abrupt and discontinuous change. The change occurs in a plane normal to the flow direction
and it is described as a normal shock. After the shock, the flow proceeds at subsonic speeds
with properties varying isentropically such that the fluid exits the nozzle at the discharge
pressure. This section reviews the principles that relate the velocities and other properties
upstream and downstream of the normal shock.
The states of the fluid immediately upstream and downstream of the normal shock are
designated with the subscripts u and d, respectively. Mass, energy, and momentum must be
conserved as flow proceeds across the shock. The width of the shock is very small and
therefore the nozzle cross-sectional area is the same on either side of the shock. The mass
balance, from Eq. (16-25), is therefore:

E20-33

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

m Vu Vd

A vu vd

(16-77)

The energy balance is the same as Eq. (16-28):

Vu2
Vd2
h

u
d

2
2

(16-78)

A momentum balance for a control volume around the shock front requires that

u Pd A mV
d
Pu A mV

(16-79)

where Pu and Pd are the pressures just upstream and just downstream of the shock,
respectively. Substituting the mass flow rate from Eq. (16-77) allows the momentum balance
to be written as

Pu Pd

Vd2 Vu2

vd vu

(16-80)

The pressure, temperature, and velocity upstream of the shock are known. The mass, energy,
and momentum equations fix the pressure, temperature, and velocity downstream of the
shock. The specific entropy of the fluid increases across the shock front.
An analytical solution for the pressure, temperature and velocity downstream of the shock can
be obtained for an ideal gas with constant specific heat capacities. The specific volumes in
Eq. (16-80) can be eliminated using the ideal gas law:

vu

RTu
Pu

(16-81)

vd

RTd
Pd

(16-82)

which results in

Pu Pd

Pd Vd2 Pu Vu2

RTd
RTu

(16-83)

where Tu and Td are the static temperatures just upstream and just downstream of the shock,
respectively. The local speed of sound upstream and downstream of the shock is determined
with Eq. (16-23).

E20-34

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

cu2 k RTu

(16-84)

cd2 k RTd

(16-85)

Substituting Eq. (16-85) into Eq. (16-83) and applying the definition of Mach number:

Mu

Vu
cu

(16-86)

Md

Vd
cd

(16-87)

allows Eq. (16-83) to be represented as:

Pu Pd k Pd M d2 k Pu M u2

(16-88)

Pd 1 k M u2

Pu 1 k M d2

(16-89)

or

Algebraic manipulation of Eqs. (16-77) through (16-87) leads to Eqs. (16-90) and (16-91),
which provide the downstream temperature and Mach number as a function fluid state just
upstream of the shock:
2
Td 2 k 1 M u

Tu 2 k 1 M d2

M d2

2 k 1 M u2
2 k M u2 k 1

(16-90)

(16-91)

EXAMPLE 16.7-1: Analysis of a Normal Shock


Air (k = 1.4) enters the converging-diverging nozzle described in Example 16.4-1 from a
large supply tank at To = 25C and Po = 500 kPa. The pressure at the nozzle outlet is adjusted
from 500 kPa to the design pressure.
a.) Determine the critical and design pressures.

E20-35

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

The conditions for this problem are consistent with the assumptions of ideal gas behavior with
constant specific heat capacities, as demonstrated in Example 16.6-1. Consequently, we will
use the ideal gas isentropic and shock equations for this analysis.
Enter the known information into EES. Also enter the nozzle geometry into a Lookup table
named Nozzle, as was done in Example 16.4-1.
$UnitSystem SI K Pa J mass
k=1.4
R=R#/molarMass(Air)
D_t=6.131 [mm]*convert(mm,m)
A|star=pi*D_t^2/4
D_exit=7.015 [mm]*convert(mm,m)
A_exit=pi*D_exit^2/4
T_o=convertTemp(C,K,25 [C])
P_o=500 [kPa]*convert(kPa,Pa)

"ratio of specific heat capacities"


"gas constant for air"
"diameter of nozzle throat"
"critical area = area at the thoat"
"diameter at exit of nozzle"
"cross-sectional area at nozzle exit"
"stagnation temperature at inlet"
"stagation pressure at inlet"

Equation (16-67) provides the relation between the Mach number and the cross-sectional area
of the nozzle, assuming isentropic flow. We will apply this equation at the exit of the nozzle.

Aexit
1 2 k 1 M

*
A
M exit
k 1

2
exit

k 1

2 k 1

Note that there are two solutions to Eq. (2) for the exit Mach number, corresponding to
subsonic and supersonic conditions; either condition can occur at the nozzle exit depending
on the back pressure. We will apply the equation twice and obtain both answers by providing
a guess value for the exit subsonic Mach number (Mexit,sub) that is less than 1 and a guess
value for the exit supersonic Mach number (Mexit,sup) that is greater than 1.
A_exit/A|star=1/M_exit_sub*((2+(k-1)*M_exit_sub^2)/(k+1))^((k+1)/(2*(k-1))) "M for critical pressure"
A_exit/A|star=1/M_exit_sup*((2+(k-1)*M_exit_sup^2)/(k+1))^((k+1)/(2*(k-1))) "M for design pressure"

Set the guess values for the variables M_exit_sub and M_exit_sup in the Variable Information
dialog, as shown in Figure 1.

E20-36

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

Figure 1: Variable Information dialog showing the guess values for the variables M_sub and M_sup.

Solve the problem in order to obtain the solution Mexit,sub = 0.5167 and Mexit,sup = 1.669 for the
subsonic and supersonic cases, respectively.
The critical pressure is the back pressure that results in the subsonic Mach number solution.
The design pressure is the back pressure that results in the supersonic Mach number solution.
In both cases, the flow is entirely isentropic throughout the nozzle. Equation (16-56) relates
the pressure and the Mach number and it is applied twice to determine both pressures.

Pcrit k 1 2
1
M exit , sub
Po
2

k / k 1

Pcrit k 1 2
1
M exit , sup
Po
2

k / k 1

P_crit/P_o=(1+(k-1)/2*M_exit_sub^2)^(-k/(k-1))
P_design/P_o=(1+(k-1)/2*M_exit_sup^2)^(-k/(k-1))

"critical back pressure"


"design back pressure"

The critical and design pressures are Pcrit = 416.7 kPa and Pdesign = 106.1 kPa, respectively.
b.) Determine the static temperature and static pressure at the nozzle throat and the
corresponding mass flow rate.
The Mach number at the throat will be equal to unity assuming that the pressure at the outlet
of the nozzle is lower than the critical pressure. Equations (16-55) and (16-56) determine the
static temperature and static pressure at the nozzle throat when the Mach number, M, is set to
1.

E20-37

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

T * k 1
1

To
2
P* k 1
1

2
Po

k / k 1

The specific volume at the throat is obtained from the ideal gas law:

v*

RT *
P*

The velocity at the throat is the local speed of sound, as determined by Eq. (16-23):

c* k RT *
The mass flow rate is the product of the area of the throat and the local sound speed divided
by the specific volume at the throat:

A* c*
v*

T|star/T_o=1/(1+(k-1)/2)
P|star/P_o=(1+(k-1)/2)^(-k/(k-1))
v|star=R*T|star/P|star
c|star=sqrt(k*R*T|star)
m_dot=A|star*c|star/v|star

"temperature at the throat"


"pressure at the throat"
"specific volume at throat"
"sound speed at throat"
"mass flow rate through nozzle"

c.) Prepare a plot showing the location along the flow direction where a normal shock
appears as function of the pressure at the nozzle outlet for an ideal gas.
A shock will only occur when the upstream conditions are supersonic. Consequently a shock
can only occur in the diverging section of the nozzle between the throat and the exit. In this
case, this limitation requires the shock to occur at axial locations between x = 13.5 mm and x
= 33.5 mm. Isentropic flow occurs between the nozzle inlet and the point just upstream of the
shock. Select a location, e.g., 20 mm, as the shock location. We will later vary this location
between 13.5 and 33.5 mm. The diameter (Ds) and area (As) at the selected location are
obtained from the table Nozzle using the Interpolate function.
Location=20 [mm]
D_s_mm=interpolate('Nozzle',D,x,x=Location)

E20-38

"location of shock"
"diameter of nozzle at shock location"

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011
D_s=D_s_mm*convert(mm,m)
A_s=pi*D_s^2/4

"diameter in m"
"cross-sectional area at shock location"

The Mach number just upstream of the shock is determined by Eq. (16-67):
k 1

As
1 2 k 1 2 2 k 1
M u

1
A* M u k 1
2

The pressure and temperature at this location are determined using the upstream Mach
number with Eqs. (16-55) and (16-56):

Tu k 1 2
1
Mu
To
2

Pu k 1 2
1
Mu
Po
2

k / k 1

Note that the application of these equations assumes that the flow is isentropic from the inlet
to the shock location. The upstream Mach number must be greater than one.
A_s/A|star=1/M_u*((2+(k-1)*M_u^2)/(k+1))^((k+1)/(2*(k-1))) "area at shock location"
T_u/T_o=1/(1+(k-1)/2*M_u^2)
"temperature upstream of shock"
P_u/P_o=(1+(k-1)/2*M_u^2)^(-k/(k-1))
"pressure upstream of shock"

Equations (16-89) through (16-91) are used to determine the pressure, temperature and Mach
number downstream of the shock.
M
2
d

2 k 1 M u2

2 k M u2 k 1

Pd 1 k M u2

Pu 1 k M d2
2
Td 2 k 1 M u

Tu 2 k 1 M d2

"Normal shock"
M_d^2=(2+(k-1)*M_u^2)/(2*k*M_u^2-(k-1))
P_d/P_u=(1+k*M_u^2)/(1+k*M_d^2)
T_d/T_u=(2+(k-1)*M_u^2)/(2+(k-1)*M_d^2)

"Mach number downstream of shock"


"pressure downstream of shock"
"temperature downstream of shock"

E20-39

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

The stagnation temperature is not affected by the shock, provided that the nozzle operation is
adiabatic. We can calculate the stagnation temperature by applying an energy balance for the
process of adiabatically slowing the fluid just downstream of the shock to zero velocity. An
energy balance for this process is:

Vd2
hd ho,d 2 0

(1)

where the subscript d indicates a location just downstream of the shock and ho ,d is the
stagnation specific enthalpy at this location. Equation (1) is divided by the square of the local
speed of sound just downstream of the shock (cd) computed using Eq. (16-23):

cd2 kRTd
so that

ho , d
cd2

Vd2
2 0
2 cd

(2)

The specific enthalpy difference in Eq. (2) can be expressed as cP Td To , d where cP is the
specific heat capacity at constant pressure, which is assumed constant, Td is the static
temperature just downstream of the shock, and To,d is the stagnation temperature at this
location. The sound speed is eliminated in the first term of Eq. (2) to yield:
cP Td To , d 1 Vd2

0
2
2
k
R
T
c
d
d

1/( k 1)

M d2

Note that:

cp
kR

1
k 1

and

Vd2
M d2
cd2

E20-40

(3)

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

which allows Eq. (3) to be simplified:

To , d

k 1 Td

1/(k-1)*(T_d-T_od)/T_d+M_d^2/2=0

M d2
0
2

"stagnation temperature downstream of shock"

Solving this equation shows that T_od = 298.1 K, which is identical to the stagnation
temperature at the nozzle inlet. The stagnation pressure, however, does change downstream
of the shock. The stagnation pressure can be found by recognizing that the specific entropy at
the location just downstream of the shock is equal to the stagnation specific entropy on this
side of the shock:

so ,d sd 0

(4)

Assuming ideal gas behavior with constant specific heat capacity, Eq. (4) can be written as
T
cP ln o , d
Td

Po ,d
R ln

Pd

or

To ,d

Td
T_od/T_d=(P_od/P_d)^(1-1/k)

11/ k

Po ,d

Pd

"stagnation pressure downstream of shock"

The flow is isentropic from a point downstream of the shock to the exit. The exit temperature
and pressure can be found using Eqs. (16-55) and (16-56).

Texit k 1 2
1
M exit
2
To ,d

Pexit k 1 2
1
M exit
Po ,d
2

k / k 1

Note that the stagnation pressure downstream of the shock (P_od) must be used, rather than
the stagnation pressure at the nozzle inlet. The specific volume at the nozzle exit is:

E20-41

THERMODYNAMICS, S.A. Klein and G.F. Nellis, Cambridge University Press, 2011

vexit

RTexit
Pexit

The local speed of sound at the exit is:

cexit k RTexit
The velocity at the exit is given by:

Vexit cexit M exit


The mass flow rate is the product of the area of the exit and the local velocity divided by the
specific volume:

Aexit cexit
vexit

P_exit/P_od=(1+(k-1)/2*M_exit^2)^(-k/(k-1))
T_exit/T_od=1/(1+(k-1)/2*M_exit^2)
v_exit=R*T_exit/P_exit
c_exit=sqrt(k*R*T_exit)
m_dot=A_exit*Vel_exit/v_exit
Vel_exit=M_exit*c_exit
P_exit_kPa=P_exit*convert(Pa,kPa)

"pressure at the exit"


"temperature at the exit"
"specific volume at exit"
"sound speed at exit"
"mass flow rate at exit"
"velocity at nozzle exit"
"convert to kPa for plotting"

Solving shows that the exit pressure is Pexit = 396.1 kPa. The shock location will be varied
between the throat and the nozzle exit in a Parametric table. Create a Parametric table with
columns for the variables Location, P_exit_kPa, M_exit. Comment out the equation that sets the
location of the shock.
{Location=20 [mm]

"location of shock"}

Solve the Parametric table to obtain the exit pressure as a function of the location of the
shock. Figure 2 illustrates the shock location and exit Mach number as a function of the back
pressure.

E20-42

You might also like