0% found this document useful (0 votes)
63 views49 pages

Normal Shock Relations in Compressible Flow

Uploaded by

Umer Cheema
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)
63 views49 pages

Normal Shock Relations in Compressible Flow

Uploaded by

Umer Cheema
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

Normal Shock Relations

Solving for M*, gives

Substituting

Solving the equation for M22

52
Normal Shock Relations

• The equation demonstrates that, for a calorically perfect gas with a constant value
of γ, the Mach number behind the shock is a function of only the Mach number
ahead of the shock.
• It also shows that when M1 = 1, then M2 = 1. This is the case of an infinitely weak
normal shock, which is defined as a Mach wave.
• In contrast, as M1 increases above 1, the normal shock becomes stronger and M2
becomes progressively less than 1.
• However, in the limit, as , M2 approaches a finite minimum value,
• , which for air is 0.378.

53
Normal Shock Relations
• The upstream Mach number M1 is a powerful parameter which dictates shock wave properties.
• Ratios of other properties across the shock can also be found in terms of M1.
• For example, from two equations,

Substituting

54
Normal Shock Relations
To obtain the pressure ratio, return to the momentum equation,

combined with yields

Dividing it by p1, and recalling that

55
Normal Shock Relations
Substitute u1∕u2 into the previous equation

56
Normal Shock Relations
To obtain the temperature ratio, recall the equation of state, p = ρRT. Hence,

57
Normal Shock Relations

• For a calorically perfect gas with a given γ, the equations give M2, ρ2 ∕ρ1, p2 ∕p1, and T2 ∕T1 as functions of M1
only. This is our first major demonstration of the importance of Mach number in the quantitative
governance of compressible flowfields.
• In contrast, for an equilibrium thermally perfect gas, the changes across a normal shock depend on both M1
and T1, whereas for an equilibrium chemically reacting gas, they depend on M1, T1, and p1.
• Moreover, for such high-temperature reacting gas cases, closed-form expressions such as equations are
generally not possible, and the normal shock properties must be calculated numerically.
• Fortunately, the results of this section hold reasonably accurately up to approximately M1 = 5 in air at
standard conditions. Beyond Mach 5, the temperature behind the normal shock becomes high enough that
γ is no longer constant.
58
Normal Shock Relations
Earlier in this section, it was stated that the flow ahead of the normal shock wave must be
supersonic. In order to prove that these equations have physical meaning only when M1 > 1,
we must appeal to the second law of thermodynamics.

59
Normal Shock Relations

• Equation demonstrates that the entropy change across the normal shock is also a function of M1 only
• It shows that, if M1 = 1 then s2 − s1 = 0, if M1 < 1 then s2 − s1 < 0, and if M1 > 1 then s2 − s1 > 0
• Therefore, since it is necessary that s2 − s1 ≥ 0 from the second law, the upstream Mach number M1 must be
greater than or equal to 1
• Here is another example of how the second law tells us the direction in which a physical process will proceed. If
M1 is subsonic, then it says that the entropy decreases across the normal shock—an impossible situation
• The only physically possible case is M1 ≥ 1, which in turn dictates that M2 ≤ 1, ρ2 ∕ρ1 ≥ 1, p2 ∕p1 ≥ 1, and T2 ∕T1 ≥ 1
• Thus, across a normal shock wave, the pressure, density, and temperature increase, whereas the velocity
decreases and the Mach number decreases to a subsonic value

60
Normal Shock Relations
How do the total (stagnation) conditions vary across a normal shock wave?

Illustration of total (stagnation) conditions ahead of and behind a normal shock wave

61
Normal Shock Relations

The total temperature is given by

Thus,

Total temperature is constant across a stationary normal shock wave


• For a stationary normal shock, the total enthalpy is always constant across the shock wave, which for
calorically or thermally perfect gases translates into a constant total temperature across the shock
• However, for a chemically reacting gas, the total temperature is not constant across the shock
• If the shock wave is not stationary—if it is moving through space—neither the total enthalpy nor total
temperature is constant across the wave 62
Normal Shock Relations

Considering

or

• The ratio of total pressures across the normal shock depends on M1 only
• Because s2 > s1, the equation shows that the total pressure decreases across a shock wave
63
Normal Shock Relations

Properties behind a normal shock wave


as a function of upstream Mach number

• As M1 becomes very large, T2 ∕T1 and p2 ∕p1 also become very large
• whereas ρ2 ∕ρ1 and M2 approach finite limits 64
Chapter 3.7
Hugoniot Equation
Hugoniot Equation
• The static pressure always increases across a shock wave.
• The wave itself can also be visualized as a thermodynamic device which compresses the gas.
• The changes across a normal shock wave can be expressed in terms of purely thermodynamic variables
without explicit reference to a velocity or Mach number, as follows.

From the continuity equation,

Substitute above equation into the momentum equation

Solve for u12

66
Hugoniot Equation
From the continuity equation, and substituting to the momentum equation,

From the energy equation,

recalling that by definition , h = e +pv, we have

67
Hugoniot Equation
Substituting u1 and u2

This simplifies to

The equation is called the


Hugoniot equation

• It has certain advantages because it relates only thermodynamic quantities across the shock.
• We have made no assumption about the type of gas. This is a general relation that holds for a perfect
gas, chemically reacting gas, real gas, etc.
• Note that the equation has the form of Δe = −paveΔv, i.e., the change in internal energy equals the mean
pressure across the shock times the change in specific volume.
• This strongly reminds us of the first law of thermodynamics in the form of the equation with δq = 0 for
the adiabatic process across the shock. 68
Hugoniot Equation

• A plot of this relation on a pv graph is called the Hugoniot


curve
• This curve is the locus of all possible pressure-volume
conditions behind normal shocks of various strengths for
one specific set of upstream values for p1 and v1
• Each point on the Hugoniot curve represents a different
shock with a different upstream velocity u1
• Shock wave compression is a very effective (not necessarily
efficient, but effective) process.

69
Hugoniot Equation

• Compare the isentropic and Hugoniot curves drawn through


the same initial point (p1, v1). At this point, both curves have
the same slope.
• However, as v decreases, the Hugoniot curve climbs above
the isentropic curve.
• Therefore, for a given decrease in specific volume, a shock
wave creates a higher pressure increase than an isentropic
compression.
• However, the shock wave costs more because of the entropy
increase and consequent total pressure loss
• The shock compression is less efficient than the isentropic
compression.

70
Chapter 3.8
One-Dimensional Flow
with Heat Addition
One-Dimensional Flow with Heat Addition
• In the previous sections, inside this control volume some action is
occurring, which causes the flow properties in region 2 to be different
than in region 1.
• This action has been due to a normal shock wave, where the large
gradients inside the shock structure ultimately result in an increase in
entropy via the effects of viscosity and thermal conduction.
• The action occurring inside the control volume can be caused by effects
other than a shock wave. For example, if the flow is through a duct,
friction between the moving fluid and the stationary walls of the duct
causes changes between regions 1 and 2. This can be particularly
important in long pipelines transferring gases over miles of land.
• Another source of change in a one-dimensional flow is heat addition. If
Rectangular control volume for one heat is added to or taken away from the gas inside the control volume,
dimensional flow the properties in region 2 will be different from those in region 1.
• Properties changes by both friction and heat addition without the
presence of a shock wave.
• This is a governing phenomenon in turbojet and ramjet engine burners,
where heat is added in the form of fuel–air combustion.
72
One-Dimensional Flow with Heat Addition
Consider the one-dimensional flow, with heat addition (or extraction)
taking place between regions 1 and 2

Solving energy equation for q, with h = cpT,

From the definition of total temperature,

Equation clearly indicates that the effect of heat addition is to directly change the total
temperature of the flow. (If heat is added, To increases; if heat is extracted, To decreases) 73
One-Dimensional Flow with Heat Addition
Find the ratios of properties between regions 1 and 2 in terms of the Mach numbers M1 and M2.

From the momentum equation

and noting that

Hence

74
One-Dimensional Flow with Heat Addition
From the perfect gas equation of state and mass conservation

From sonic speed equation and the definition of Mach number,

75
One-Dimensional Flow with Heat Addition

Since

76
One-Dimensional Flow with Heat Addition

• Once M2 is known, then p2 ∕p1, T2 ∕T1, and ρ2 ∕ρ1 are directly obtained.
• This is a straightforward procedure; however, the solution of equation above(To2/To1) for M2 must be
found by trial and error.
• A more direct method of solving the problem of one-dimensional flow with heat addition is given in
the next slide. 77
One-Dimensional Flow with Heat Addition
• For convenience of calculation, we use sonic flow as a reference condition.
• Let M1 = 1; the corresponding flow properties are denoted by

• The flow properties at any other value of M are then obtained by inserting M1 = 1 and M2 = M into
Equations below, yielding

78
One-Dimensional Flow with Heat Addition

79
One-Dimensional Flow with Heat Addition
• Equations are tabulated as a function of M for γ = 1.4 in
Table A.3.
• Note that, for a given flow, no matter what the local flow
properties are, the reference sonic conditions are
constant values.
• These starred values, although defined as conditions that
exist at Mach 1, are fundamentally different from T *, p*,
and ρ* defined in previous section.
• There, T* was defined as the temperature that would
exist at a point in the flow if the flow at that point were
imagined to be locally slowed down (for a supersonic
case) or speeded up (for a subsonic case) to Mach 1
adiabatically.
• In the present section, we are dealing with a one-
dimensional flow with heat addition, a nonadiabatic
process. Here, T*, p*, and ρ* are those conditions in a
Illustration of the meaning of the starred quantities at one-dimensional flow that would exist if enough heat is
Mach 1 for one-dimensional flow with heat addition added to achieve Mach 1.
80
One-Dimensional Flow with Heat Addition
• To see this more clearly, consider two different locations in a
one-dimensional flow with heat addition, denoted by stations
1 and 2 (Figure a). The flow at station 1 is given by M1, p1, and
T1. For the sake of discussion, let M1 = 3
• Now, let an amount of heat q1 be added to this flow between
stations 1 and 2. As a result, the flow properties at location 2
are M2, p2, and T2. Assume that q1 was a sufficient amount to
result in M2 = 1.5 (Adding heat to a supersonic flow reduces
the Mach number of the flow)
• Return to station 1, where the local Mach number is M1 = 3.
Imagine that we add enough heat q1* (>q1) downstream of
this station to cause the flow to slow down to Mach 1 as
shown in Fig b
• The conditions in the duct where M = 1 after q1* is added are
denoted by T*, p*, etc. Return to station 2, where M2 = 1.5

Illustration of the meaning of the starred quantities at • Imagine that we add enough heat downstream of this station
Mach 1 for one-dimensional flow with heat addition to cause the flow to slow down to Mach 1 as sketched in Fig. c;
denote this amount of heat by q2*
81
One-Dimensional Flow with Heat Addition
• The conditions in the duct where M = 1 after q2* is added are
denoted by T*, p*, etc.
• These are precisely the same values that were obtained by
adding downstream of station 1.
• In other words, for a given one-dimensional flow, the values of
T*, p*, etc, achieved when enough heat is added to bring the
flow to Mach 1, are the same values, no matter whether the
heat is added as downstream of station 1 or as downstream of
station 2.
• This is why, in the last five equations, the starred quantities
are simply reference quantities that are fixed values for a given
flow entering a one-dimensional duct with heat addition.
• With this concept, the equations, or rather the tabulated
values in Table A.3 obtained from these equations, simplify the
calculation of problems involving one-dimensional flow with
Illustration of the meaning of the starred quantities at heat addition.
Mach 1 for one-dimensional flow with heat addition
82
One-Dimensional Flow with Heat Addition

For heat extraction (cooling of the flow), all of the above trends are opposite.

83
One-Dimensional Flow with Heat Addition

• It is important to note that heat addition always drives the Mach


numbers toward 1, decelerating a supersonic flow and accelerating a
subsonic flow.
• This is emphasized in the figure, which is a Mollier diagram (enthalpy
versus entropy) of the one-dimensional heat-addition process.
• The curve is called the Rayleigh curve, and is drawn for a set of given
initial conditions.
• If the conditions in region 1 are given by point 1, then the particular
Rayleigh curve through point 1 is the locus of all possible states in
region 2.
• Each point on the curve corresponds to a different value of q added or
taken away.

The Rayleigh curve • Point a corresponds to maximum entropy; also at point a the flow is
sonic.

84
One-Dimensional Flow with Heat Addition
• The lower branch of the Rayleigh curve below point a corresponds to
supersonic flow.
• The upper branch above point a corresponds to subsonic flow.
• If the flow in region 1 of the previous figure is supersonic and
corresponds to point 1 in the left figure, then heat addition will cause
conditions in region 2 to move closer to point a, with a consequent
decrease of Mach number toward unity.
• As q is made larger, conditions in region 2 get closer and closer to point
a.
• Finally, for a certain value of q, the flow will become sonic in region 2.
For this condition, the flow is said to be choked, because any further
increase in q is not possible without a drastic revision of the upstream
conditions in region 1.
• For example, if the initial supersonic conditions in region 1 were
The Rayleigh curve obtained by expansion through a supersonic nozzle, and if a value of q
is added to the flow that allowed for attaining Mach 1 in region 2, then
a normal shock will form inside the nozzle, and conditions in region 1
will suddenly become subsonic.
85
One-Dimensional Flow with Heat Addition

• Now consider an alternative case where the initial flow in region 1 in


the figure is subsonic, say given by point 1′ in the earlier figure.
• If heat is added to the flow, conditions in the downstream region 2 will
move closer to point a.
• If q is increased to a sufficiently high value, then point a will be reached
and the flow in region 2 will be sonic.
• The flow is again choked, and any further increase in q is impossible
without an adjustment of the initial conditions in region 1.
The Rayleigh curve
• If q is increased above this value, then a series of pressure waves will
propagate upstream, and nature will adjust the conditions in region 1 to
a lower subsonic Mach number, to the left of point 1′.

86
One-Dimensional Flow with Heat Addition

• Note from the Rayleigh curve that it is theoretically possible to


decelerate a supersonic flow to a subsonic value, by first heating it
until sonic flow (point a) is reached, and then cooling it thereafter.
• Similarly, an initially subsonic flow can be made supersonic, by first
heating it until sonic flow (point a) is reached, and then cooling it
thereafter.
• Just as in the case of a normal shock wave, heat addition to a flow
The Rayleigh curve (subsonic or supersonic) always decreases the total pressure.
• This effect is of prime importance in the design of jet engines and in
the pressure recovery attainable in gasdynamic and chemical lasers.

87
Chapter 3.9
One-Dimensional Flow
with Friction
One-Dimensional Flow with Friction
• Consider the one-dimensional flow of a compressible
inviscid fluid in a constant-area duct.
• If the flow is steady, adiabatic, and shockless, the three
governing equations yield the trivial solution of constant
property flow everywhere along the duct.
• However, in reality, all fluids are viscous, and the friction
between the moving fluid and the stationary walls of the
duct causes the flow properties to change along the duct.
• Although viscous flows are not the subject of this course, if
the frictional effect is modeled as a shear stress at the wall
acting on a fluid with uniform properties over any cross
section, then the equations developed with one
Model of one-dimensional flow with friction modification, describe the mean properties of frictional flow
in constant-area ducts.
• The analysis and results are analogous to one-dimensional
flow with heat addition.

89
One-Dimensional Flow with Friction
• The aforementioned modification applies to the momentum equation.
• The frictional shear stress τw acts on the surface of the cylindrical control volume, thus contributing an
additional surface force in the integral formulation of the momentum equation.
• The x direction momentum equation for an inviscid gas; with the shear stress included, becomes

Applied to the cylindrical control volume of diameter D and length L

Since A = π D2 / 4

90
One-Dimensional Flow with Friction

As L shrinks to dx

From Hence

The shear stress can be expressed in terms of a friction coefficient f, defined as

Recalling that

91
One-Dimensional Flow with Friction

The equation relates the Mach numbers at two different sections to the integrated
effect of friction between the sections.
The ratios of static temperature, pressure, density, and total pressure between the two sections
are readily obtained. The flow is adiabatic; hence, To = const.

Since

92
One-Dimensional Flow with Friction

Since

Then the total pressure ratio is

93
One-Dimensional Flow with Friction
Analogous to our previous discussion of one-dimensional flow with heat addition, calculations of
flow with friction are expedited by using sonic flow reference conditions, where the flow
properties are denoted by p*, ρ*, T*, and p0*.

94
One-Dimensional Flow with Friction

If we define x = L* as the station where M = 1, then

or

where f¯ is an average friction coefficient defined as

• Equations are tabulated versus Mach number in Table A.4 for γ = 1.4.
• The local friction coefficient f depends on whether the flow is laminar or turbulent, and is a function
of Mach number, Reynolds number, and surface roughness, among other variables.
• In almost all practical cases, the flow is turbulent, and the variation of f must be obtained
empirically.
• For our purposes, it is reasonable to assume an approximate constant value of f = 0.005, which
holds for Re > 105 and a surface roughness of 0.001 D. 95
One-Dimensional Flow with Friction

96
One-Dimensional Flow with Friction
Certain physical trends reflected by the numbers obtained from such solutions are summarized here:

97
One-Dimensional Flow with Friction

• Note that friction always drives the Mach number toward 1,


decelerating a supersonic flow and accelerating a subsonic
flow.
• The figure on the left is a Mollier diagram of one-
dimensional flow with friction.
• The curve is called the Fanno curve, and is drawn for a set of
given initial conditions.
• Point a corresponds to maximum entropy, where the flow is
sonic.
• This point splits the Fanno curve into subsonic (upper) and
supersonic (lower) portions.

The Fanno curve

98
One-Dimensional Flow with Friction

• If the inlet flow is supersonic and corresponds to point 1 in the


figure, then friction causes the downstream flow to move
closer to point a, with a consequent decrease of Mach number
toward unity.
• Each point on the curve between points 1 and a corresponds to
a certain duct length L. As L is made larger, the conditions at
the exit move closer to point a.
• For a certain value of L, the flow becomes sonic. For this
condition, the flow is choked, because any further increase in L
is not possible without a drastic revision of the inlet conditions.
• If the inlet conditions at point 1 were obtained by expansion
through a supersonic nozzle, and if L were larger than that
allowed for attaining Mach 1 at the exit, then a normal shock
The Fanno curve would form inside the nozzle, and the duct inlet conditions
would suddenly become subsonic.

99
One-Dimensional Flow with Friction
• Consider a case where the inlet flow is subsonic(point 1′). As L
increases, the exit conditions move closer to point a.
• If L is increased to a sufficiently large value, then point a is
reached and the flow at the exit becomes sonic.
• The flow is again choked, and any further increase in L is
impossible without an adjustment of the inlet conditions to a
lower inlet Mach number (without moving the inlet conditions
to the left of point 1′ in the figure).
• Note that friction always causes the total pressure to decrease
whether the inlet flow is subsonic or supersonic.
• Unlike the Rayleigh curve for flow with heating and cooling,
the upper and lower portions of the Fanno curve cannot be
traversed by the same one-dimensional flow.
• Within the framework of one-dimensional theory, it is not
possible to first slow a supersonic flow to sonic conditions by
The Fanno curve
friction, and then further slow it to subsonic speeds also by
friction. Such a subsonic deceleration would violate the second
law of thermodynamics.
100

You might also like