analysis of oblique shocks p m v subbarao associate professor mechanical engineering department i i...
TRANSCRIPT
Analysis of Oblique Shocks
P M V SubbaraoAssociate Professor
Mechanical Engineering DepartmentI I T Delhi
A Mild, Efficient and Compact Compressor ….
Mach Waves, Revisited
• A ‘’point-mass’’ object moving with Supersonic velocity Generates an infinitesimally weak “mach wave”.• The direction of flow remains unchanged across Mach wave.
V t
c t
sin c t
V t
1
M sin 1 1
M
Oblique Shock Wave
• When generating object is larger than a “point”, shockwave is stronger than mach wave …. Oblique shock wave
• -- shock angle
• -- turning or “wedge angle”
Oblique Shock Wave Geometry
• Shock is A CV & Must satisfyi) continuityii) momentumiii) energy
Tangential NormalAhead wx, Mtx ux, Mnx
Of ShockBehind wy, Mty uy, Mny
Shock
w x &
M tx
ux &
Mnx
Vx & Mx
Vx & Mx
V y & M y
w y &
M tyu
y & M
ny
Vy & My
Continuity Equation
• For Steady Flow
w2
u2
w1
u1
ds
ds
V ds
C.S.
0 xuxA
yuyA xux
yuy
w x
ux
Vx
w y
uy
Vy
x y
Momentum Equation
• For Steady Flow w/o Body Forces V
ds
C .S . V
p
C .S . dS
• Tangential Component
xuxwxA
yuywy A 0
xux
yuy wx wy
Tangential velocity isConstant across obliqueShock wave• But from continuity
SdpdAVuSCSC
....
• Normal Component
Tangential velocity isConstant across obliqueShock wave
xux2 A yuy
2 A py px A
px xux2 py yuy
2
SdpdAVuSCSC
....
Energy Equation
•
• thus …
Write Velocity in terms of components
Vx2 ux
2 wx2 Vx
2 uy2 wy
2 wx wy
hx ux2
2hx ux
2
2
hx Vx
2
2hy
Vy2
2
Collected Oblique Shock Equations
• Continuity
wx wy
px x ux
2 py
y uy2
x ux y uy
cpTx ux2
2cpTy uy
2
2
ux
uy
wx
wy
• Momentum
• Energy
w x &
M tx
ux &
Mnx
Vx & Mx
Vx & Mx
V y & M y
w y &
M ty
uy &
Mny
Vy & My
• Defining:
Mnx=Mxsin(
Mtx=Mxcos(
• Then by similarity we can write the solution
12
2)1(2
22
nx
nxny
M
MM
• Similarity Solution Letting
1
12 2
nx
x
y M
p
p
Mnx= Mxsin(
22
22
1
1221
nx
nxnx
x
y
M
MM
T
T
21
12
2
nx
nx
x
y
M
M
• Properties across an Oblique Shock wave ~ f(Mx, )
1
1sin2 2
x
x
y M
p
p
22
22
sin1
1sin22sin1
x
xx
x
y
M
MM
T
T
2sin1
sin12
2
x
x
x
y
M
M
Total Mach Number Downstream of Oblique Shock
w1 w2
Tangential velocity isConstant across obliqueShock wave
w1 w2 Mt1c1 Mt2c2 M1 cos()c1
Mt2 M1 cos()c1
c2
M1 cos()T1
T2
M 2 Mt22 Mn2
2
Tangential velocity isConstant across obliqueShock wave
M 2 Mt22 Mn2
2 Mn2 1
1 2
M1 sin( ) 2
M1 sin( ) 2 1
2
M 2 M1 cos( ) 2 T1
T2
1
1 2
M1 sin( ) 2
M1 sin( ) 2 1
2
Tangential velocity is Constant across oblique Shock wave
• Or … More simply .. If we consider geometric arguments
M 2 Mn2
sin
M3M2
Mn2Mt2
Oblique Shock Wave Angle
• Properties across Oblique Shock wave ~ f(M1, )
• is the geometric angle that “forces” the flow
• How do we relate to
Oblique Shock Wave Angle (cont’d)
• Solving for the ratio u2/u1
Implicit relationship for shock angle in terms ofFree stream mach number and “wedge angle”
u2
u1
tan
tan 1
2
2
1
1 Mn1
2
2 1 Mn12
tan
tan 2 1 M1 sin
2 1 M1 sin
2
• Solve explicitly for tan()
tan tan tan tan2 tan
2 1 M1 sin 2
1 M1 sin 2
tan
tan 1 M1 sin
2 2 1 M1 sin 2
1 M1 sin 2 tan2 2 1 M1 sin
2
Oblique Shock Wave Angle (cont’d)
• Simplify Numerator
tan 1 M1 sin 2 2 1 M1 sin
2
tan M1 sin 2 M1 sin
2 2 M1 sin
2 M1 sin
2
tan 2 M1 sin 2 1
• Simplify Denominator
1 M1 sin 2 tan2 2 1 M1 sin
2
tan2 1 M1
sin tan
2
2 1 M1 sin 2
tan2 1 M1 cos 2 2 1 M1 sin
2
tan2 1 M12 1 sin2 2 1 M1
2 sin2
tan2 2 1 M12 1 M1
2 sin2 1 M12 sin2
tan2 2 1 M12 2M1
2 sin2 tan2 2 1 M12 2M1
2 sin2
tan2 2 M12 M1
2 1 2sin2 tan2 2 M12 M1
2 cos2 sin2
tan2 2 M12 cos 2
Oblique Shock Wave Angle (cont’d)
• Collect terms
tan 2 tan M1 sin
2 1
tan2 2 M12 cos 2
2 M1
2 sin2 1 tan 2 M1
2 cos 2 • “Wedge Angle” Given explicitly as function of shock angle and freestream Mach number
• Two Solutions “weak” and “strong” shock wave. In reality weak shock typically occurs;
strong only occurs under very Specialized circumstances .e.g near stagnation point for a detached Shock.
Oblique Shock Wave Angle
M1=1.5 M1=2.0 M1=2.5
M1=3.0
“strong shock”
“weak shock”
M1=4.0
M1=5.0
1.4
max
curve
Solving for Oblique Shock Wave Angle in Terms of Wedge Angle
• As derived
• “Wedge Angle” Given explicitly as function of shock angle and freestream Mach number
• For most practical applications, the geometric deflection angle (wedge angle) and Mach number are prescribed .. Need in terms of and M1
• Obvious Approach …. Numerical Solution using Newton’s method
tan 2 M1
2 sin2 1 tan 2 M1
2 cos 2
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (cont’d)• Newton method
2 M12 sin2 1
tan 2 M12 cos 2
tan f () 0
f () f (( j ) ) f
( j )
( j ) O( 2 ) ....
( j1) ( j )
2 M12 sin2 1
tan 2 M12 cos 2
tan
f
( j )
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (cont’d)• Newton method (continued)
• Iterate until convergence
f
2 M14 sin2 1 cos 2 M1
2 2 cos 2 1 2 sin2 2 M1
2 cos 2 2
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (cont’d)
f
• “Flat spot”Causes potentialConvergence Problems withNewton Method
IncreasingMach
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (cont’d)• Newton method … Convergence can often be slow (because of low derivative slope)
• Converged solution
true 60.26o
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (concluded)• Newton method … or can “toggle” to strong shock solution
• Strong shock solution
strong 71.87o
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)• Because of the slow convergence of Newton’s method for this implicit function… explicit solution … (if possible) .. Or better behaved .. Method very desirable
tan 2 M1
2 sin2 1 tan 2 M1
2 cos 2
2 M12 sin2 1
tan 2 M12 M1
2 cos2 sin2
cos 2 cos2 sin2 Substitute
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• But, since 1 cos2 sin2
2 M12 sin2 1
tan 2 M12 M1
2 cos2 sin2
2 M12 sin2 sin2 cos2
tan 2 M12 M1
2 cos2 sin2
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Simplify and collect terms2 M1
2 sin2 sin2 cos2 tan 2 M1
2 M12 cos2 sin2
M12 1 sin2 cos2
tan 12
M12
1
2M1
2 cos2 sin2
M12 1 sin2 cos2
tan 12
M12
1
2M1
2 cos2 sin2
M12 1 sin2 cos2
tan 1 cos2 sin2
2M1
2
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Again, Since
M12 1 sin2 cos2
tan 1 cos2 sin2
2M1
2
M12 1 sin2 cos2
tan cos2 sin2 cos2 sin2 cos2 sin2
2M1
2
1 cos2 sin2
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Regroup and collect terms
M12 1 sin2 cos2
tan cos2 sin2 cos2 sin2 cos2 sin2
2M1
2
M12 1 tan2 1
tan 1 tan2 1 tan2 1 tan2
2M1
2
M12 1 tan2 1
tan 1 1
2M1
2 tan2 tan2 tan2
2M1
2
M12 1 tan2 1
tan 1 1
2M1
2
tan2 1
1
2M1
2
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Finally
• Regrouping in terms of powers of tan()
tan M1
2 1 tan2 1 1
1
2M1
2
tan 1 1
2M1
2
tan3
1 1
2M
1
2
tan
tan3 1 1
2M
1
2
tan
tan M1
2 1 tan2 1 0
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Letting
• Result is a cubic equation of the form
a 1 1
2M1
2
tan
b M12 1
c 1 1
2M1
2
tan
x tan
ax3 bx2 cx 1 0
• Polynomial has 3 real rootsi) weak shockii) strong shockiii) meaningless solution
( < 0)
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Numerical Solution of Cubic (Newton’s method)
ax3 bx2 cx 1 f (x) 0
0 f (x j ) f (x)
x j
x j1 x j o(x2 )
x j1 x j f (x j )
f (x)
x j
x j ax j
3 bx j2 cx j 1
3ax j2 2bx j c
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Collecting terms
x j ax j
3 bx j2 cx j 1
3ax j2 2bx j c
3ax j3 2bx j
2 cx j ax j3 bx j
2 cx j 1 3ax j
2 2bx j c
2ax j3 bx j
2 1
3ax j2 2bx j c
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Solution Algorithm (iterate to convergence)
x j1 2ax j
3 bx j2 1
3ax j2 2bx j c
• Where again a 1 1
2M1
2
tan
b M12 1
c 1 1
2M1
2
tan
x tan
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Properties of Solver algorithm are much improved
Improved AlgorithmOriginal Algorithm
true 60.26o
• Original algorithm• Improved algorithm
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Three Solutions always returned depending on start condition
Original Algorithm Improved Algorithmtrue 60.26o
• Weak Shock Solution
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Three Solutions always returned depending on start condition
Improved Algorithm• Strong Shock Solutionstrong 71.87o
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (improved solution)
(cont’d)• Three Solutions always returned depending on start condition
Improved Algorithm• Meaningless Solution meaningless 0o
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (explicit solution)
Improved Algorithm• Meaningless Solution
• Explicit Solution … Using guidance from numerical algorithm, can we findExplicit (non -iterative) solution for shock angle?
• Cubic equation has three explicit solutionsi) weak shockii) Strong shockiii) non-physical solution
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (explicit solution)
Improved Algorithm
• Explicit Solution … Using guidance from numerical algorithm, can we find Explicit (non -iterative) solution for shock angle?
• Root 1: tan[]=
• Root 2: tan[]=
• Root 3: tan[]=
• Break solutions down into manageable form
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (explicit solution)
Improved Algorithm• Meaningless Solution
• Explicit Solution (From Anderson, pp. 142,143) …
1 1
2M1
2
tan
tan3 M12 1 tan2 1
1
2M1
2
tan
tan 1 0
tan M1
2 1 2 cos4 cos 1
3
3 1 1
2M1
2
tan
M12 1 2
3 1 1
2M1
2
1 1
2M1
2
tan2
M1
2 1 3 9 1
1
2M1
2
1 1
2M1
2 1
4M1
4
tan2
3
= 0 ---> Strong Shock
= 1 ---> Weak Shock
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (explicit solution)
Improved Algorithm• Meaningless Solution
• Explicit Solution Check … let {M=5, , =40
M12 1 2
3 1 1
2M1
2
1 1
2M1
2
tan2 =
= 13.5321
52 1 2
3 11.4 1
252+
11.4 1+
252+
180
40 tan
2
0.5
1.4
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (explicit solution)
Improved Algorithm• Meaningless Solution
• Explicit Solution Check … let {M=5, , =40
=
M12 1 3
9 1 1
2M1
2
1 1
2M1
2 1
4M1
4
tan2
3
= -0.267118
52 1 3
9 11.4 1
252+
11.4 1
252 1.4 1+
454+ +
180
40 tan
2
13.53213
1.4
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (explicit solution)
Improved Algorithm• Meaningless Solutiontan
M12 1 2 cos
4 cos 1 3
3 1 1
2M1
2
tan
= 1 ---> Weak Shock
=
180
52 1 2 13.5321 4 1 0.26712 acos+
3 cos+
3 11.4 1
252+
180
40 tan
atan
= 60.259 Check!
• Explicit Solution Check … let {M=5, , =40 1.4
Solving for Oblique Shock Wave Angle in Terms of Wedge
Angle (explicit solution)
• Meaningless Solutiontan
M12 1 2 cos
4 cos 1 3
3 1 1
2M1
2
tan
= 0 ---> Strong Shock
=
= 71.869 Check!
180
52 1 2 13.5321 4 0 0.26712 acos+
3 cos+
3 11.4 1
252+
180
40 tan
atan
… OK .. This works and isClearly the Best method
(concluded)
• Explicit Solution Check … let {M=5, , =40 1.4
Oblique Shock Waves:Collected Algorithm
• Properties across Oblique Shock wave ~ f(M1, )
• is the geometric angle that “forces” the flow
tan 2 M1
2 sin2 1 tan 2 M1
2 cos 2
Oblique Shock Waves:Collected Algorithm (cont’d)
• Can be re-written as third order polynomial in tan()
• “Very Easy” numerical solution
1 1
2M
1
2
tan
tan3 1 1
2M
1
2
tan
tan M1
2 1 tan2 1 0
x j1 2ax j
3 bx j2 1
3ax j2 2bx j c
a 1 1
2M1
2
tan
b M12 1
c 1 1
2M1
2
tan
x tan
• Cubic equation has three solutionsi) weak shockii) Strong shockiii) non-physical solution
Oblique Shock Waves:Collected Algorithm (cont’d)
• “Less Obvious” explicit solution
tan M1
2 1 2 cos4 cos 1
3
3 1 1
2M1
2
tan
M12 1 2
3 1 1
2M1
2
1 1
2M1
2
tan2
M1
2 1 3 9 1
1
2M1
2
1 1
2M1
2 1
4M1
4
tan2
3
= 0 ---> Strong Shock
= 1 ---> Weak Shock
Oblique Shock Waves:Collected Algorithm (cont’d)
• ... and the rest of the story …
2
1
1 M1 sin 2
2 1 M1 sin 2 p2
p1
12
1 M1 sin 2 1
T2
T1
12
1 M1 sin 2 1
2 1 M1 sin 2 1 M1 sin 2
Oblique Shock Waves:Collected Algorithm (concluded)
• ... and the rest of the story …
Mn2 1
1 2
M1 sin 2
M1 sin 2 1
2
M 2 Mn2
sin
P02
P01
2
1 M1 sin 2
1 2
1
1 2
M1 sin
2
1 1
2M1 sin 2
1
Example:•M1 = 3.0, p1=1atm, T1=288K, =20=1.4,
M1
M2
• Compute shock wave angle (weak)
• Compute P02, T02, p2, T2, M2 … Behind Shockwave
Example: (cont’d)
•M1 = 3.0, p1=1atm, =1.4, T1=288K, =20
• Explicit Solver for
M12 1 2
3 1 1
2M1
2
1 1
2M1
2
tan2 =7.13226
M1
2 1 3 9 1
1
2M1
2
1 1
2M1
2 1
4M1
4
tan2
3=0.93825
Example: (cont’d)
•M1 = 3.0, p1=1atm, =1.4, T1=288K, =20
• = 1 (weak shock)
180
32 1 2 7.132264 1 0.93825 acos+
3 cos+
3 11.4 1
232+
180
20 tan
atan
tan M1
2 1 2 cos4 cos 1
3
3 1 1
2M1
2
tan
=
37.764
Example: (cont’d)
•M1 = 3.0, p1=1atm, =1.4, T1=288K, =20
• Normal Component of Free stream mach Number
p2
p1
12
1 Mn12 1
Mn1 M1 sin 3
18037.7636
sin =1.837
Normal Shock Solver
p2 = 3.771(1 atm) = 3.771 atm
• Compute Pressure ratio across shock
• Flow is compressed
Example: (cont’d)
•M1 = 3.0, p1=1atm, =1.4, T1=288K, =20
• Temperature ratio Across Shock
Normal Shock Solver
T2 = 1.5596(288 K) = 449.2 K
T2
T1
12
1 Mn12 1
2 1 Mn12
1 Mn12
Example: (cont’d)
•M1 = 3.0, p1=1atm, =1.4, T1=288K, =20
• Stagnation Pressure ratio across shock
Normal Shock Solver
0.7961
Mn1 M1 sin 3
18037.7636
sin =1.837
P02
P01
Example: (cont’d)
•M1 = 3.0, p1=1atm, =1.4, T1=288K, =20
• Stagnation Pressure ratio (alternate method)
P02
P01
2
1 Mn12
1 2
1
1 2
Mn12
2
1 1
2Mn1
2
1
2
1.4 1+ 1.4 3
18037.7636
sin
2 1.4 1 2
1
1.4 1
1.4 1+ 2
2
3
18037.7636
sin
2
11.4 1
2 3
180
37.7636 sin
2
+
1.4
1.4 1
=0.7961
Example: (cont’d)
•M1 = 3.0, p1=1atm, =1.4, T1=288K, =20
• Stagnation Pressure
=29.24 atm
P02 P02
P01
P01
p1
p1 P02
P01
1 1
2M1
2
1
p1
0.7961 11.4 1
232
1.4
1.4 11atm
Example: (cont’d)
•M1 = 3.0, p1=1atm, =1.4, T1=288K, =20
• Stagnation Temperature behind shock
=806.4 oK
T02 T01 T01
T1
T1 1 1
2M1
2
T1
11.4 1
232
288o K
What Happens when•M1 = 3.0, p1=1atm, =1.4, T1=288K, =0.00001
• Explicit Solver for
M12 1 2
3 1 1
2M1
2
1 1
2M1
2
tan2 =8.0
M1
2 1 3 9 1
1
2M1
2
1 1
2M1
2 1
4M1
4
tan2
3=1.0
What Happens when (cont’d) •M1 = 3.0, p1=1atm, =1.4, T1=288K, =0.00001
tan M1
2 1 2 cos4 cos 1
3
3 1 1
2M1
2
tan
=19.47 180
sin 1 1
M1
19.47o
• “mach line”
What Happens when (cont’d) •M1 = 3.0, p1=1atm, =1.4, T1=288K, =0.00001
• Normal Component of Free stream mach Number
Mn1 M1 sin 1.0000
• p2
p1
12
1 Mn12 1 = 1.0 (NO COMPRESSION!)