PDF processed with CutePDF evaluation edition www.CutePDF.

com
 Fatigue II

Lecture 11

Engineering 473
Machine Design
 Finite Life Estimates

How can the life

of a part be
estimated if the
Alternating S yt
Finite Life mean stress-
Stress, σ a
(Cycles to failure?) alternating stress
Se pair lie above the
Goodman line?
Infinite Life
Stress State

S yt Sut
Mean Stress, σ m

Goodman Diagram
 S-N Curve

The S-N curve gives

the cycles to failure
for a completely
reversed (R=-1)
uniaxial stress state.

What do you do if
the stress state is not
completely
reversed?
Completely reversed cyclic
stress, UNS G41200 steel

Shigley, Fig. 7-6

 Definitions
Stress Range
σ r = σ max − σ min

Alternating Stress
σ max − σ min
σa =
2
Mean Stress
σ max + σ min
σm =
2
Stress Ratio Amplitude Ratio
Note that R=-1 for a
σ min σa completely reversed
R= A=
σ max σm stress state with zero
mean stress.
Fluctuating-Stress Failure
Interaction Curves
The interaction curves
provide relationships between
alternating stress and mean
stress.
When the mean stress is
zero, the alternating
component is equal to the
endurance limit.

The interaction curves are

for infinite life or a large
number of cycles.
Shigley, Fig. 7-16
 Goodman Interaction Line
Any combination of mean and
k f Sa S m alternating stress that lies on or
+ =1 below Goodman line will have
Se Sut infinite life.

Factor of Safety Format

k f Sa Sm 1
+ =
Se Sut N f
Note that the fatigue stress
concentration factor is applied
only to the alternating
component.
Master Fatigue Plot

Constant
cycles till
failure
interaction
curves.

Shigley, Fig. 7-15

 Equivalent Alternating Stress

σa Alternating stress at zero

σ m =0 mean stress that fails the part
Alternating S yt in the same number of cycles
Stress, σ a as the original stress state.
Se
105 cycles

106 cycles

S yt Sut
Mean Stress, σ m

The red and blue lines are estimated fatigue interaction curves
associated with a specific number of cycles to failure.
Number of Cycles to
Failure

Once the equivalent

alternating stress is
found, the S-N curve
may be used to find
the number of cycles
to failure.
 Equivalent Alternating Stress
Formula
k f σa σm 1
+ = Goodman Line
Se Sut N f

σa ≡ Equivalent completely reversed

k f σa σm 1 σ m =0
+ = (R = -1) stress that causes fatigue
σ a σ =0 Sut N f
m
failure in the same number of cycles
as the original σ a and σ m pair.
k f σa
σa =
σ m =0 1 σm
−
N f Sut
 Example
5 in 5 in Pmax = 3000 lb
Pmin = 2000 lb
1 2

Material UNS
1.5 in. dia. 0.875 in. dia. G41200 Steel
0.125 in. rad. Notch sensitivity
q=0.3

π 4 π I1 0.249 in 4
D 1 = (1.5) = 0.249 in 4 S1 = = = 0.332 in 3
4
I1 =
64 64 c1 0.75 in
π 4 π I 2 0.088 in 4
D 2 = (0.875) = 0.088 in 4
4
I2 = S2 = = = 0.201 in 3
64 64 c 2 0.438 in
 Example
(Continued)
5 in 5 in Pmax = 3000 lb
Pmin = 2000 lb
1 2

Material UNS
1.5 in. dia. 0.875 in. dia. G41200 Steel
0.125 in. rad. Notch sensitivity
q=0.3

kf −1 D 1.5 in k t = 1.61
q= = = 1.71
k t −1 d 0.875 in k f = 1 + q(k t − 1)
r 0.125
k f = 1 + q(k t − 1) = = 0.143 = 1 + 0.3(1.61 − 1)
d 0.875
= 1.18
Ref. Peterson
 Example
(Continued)
Pmax = 3000 lb
5 in 5 in
Pmin = 2000 lb
1 2

Material UNS
1.5 in. dia. 0.875 in. dia. G41200 Steel
0.125 in. rad. Notch sensitivity
q=0.3
Section 1 (Base)

M1 (3000 lb )(10 in ) σ max − σ min

σ max = = 3
= 90.4 ksi σa = = 15.1 ksi
S1 0.332 in 2
M1 (2000 lb )(10 in ) σ max + σ min
σ min = = = 60.2 ksi
S1 0.332 in 3 σm = = 75.3 ksi
2
 Example
(Continued)
5 in 5 in Pmax = 3000 lb
Pmin = 2000 lb
1 2

Material UNS
1.5 in. dia. 0.875 in. dia.
G41200 Steel
0.125 in. rad. Notch sensitivity
Section 2 (Fillet) q=0.3

M1 (3000 lb )(5 in ) σ max − σ min

σ max = = 3
= 74.6 ksi σa = = 12.4 ksi
S1 0.201 in 2
M1 (2000 lb )(5 in ) σ max + σ min
σ min = = 3
= 49.8 ksi σm = = 62.2 ksi
S1 0.201 in 2
 Example
(Continued)

Sut = 116 ksi

S′e = 30 ksi = Se

k f σa σm 1
+ =
Se Sult N f

Completely reversed cyclic

stress, UNS G41200 steel
Shigley, Fig. 7-6
 Example
(Continued)
Section 1 (Base)

M (3000 lb)(10 in ) = 90.4 ksi σ max − σ min

σ max = 1= σa = = 15.1 ksi
S1 0.332 in 3 2
M1 (2000 lb )(10 in ) σ max + σ min
σ min = = 3
= 60.2 ksi σm = = 75.3 ksi
S1 0.332 in
2

Sut = 116 ksi Nf = 1

S′e = 30 ksi = Se
1.0(15.1 ksi ) 75.3 ksi
k f σa σm 1 + = 1.15
+ = 30 ksi 116 ksi
Se Sult N f Part has finite life at base.
 Example
(Continued)

Section 2 (Fillet)

M1 (3000 lb )(5 in ) σ max − σ min

σ max = = 3
= 74.6 ksi σa = = 12.4 ksi
S1 0.201 in 2
M1 (2000 lb )(5 in ) σ max + σ min
σ min = = 3
= 49.8 ksi σm = = 62.2 ksi
S1 0.201 in 2

Sut = 116 ksi Nf = 1

S′e = 30 ksi = Se
1.18(12.4 ksi ) 62.2 ksi
+ = 1.02
k f σa σm 1 30 ksi 116 ksi
+ =
Se Sult N f Part has finite life.
 Calculation of Equivalent
Alternating Stress

k f σa
σa =
σ m =0 1 σm
−
N f Sut
Base Fillet
σ a = 15.1 ksi σ a = 12.4 ksi
σ m = 75.3 ksi σ m = 62.2 ksi

(1.0)15.1 (1.18)12.4
σ a σ =0 = σ a σ =0 =
1 75.3 m 1 62.2
−
m
−
1.0 116 1.0 116
= 43.0 ksi = 31.5 ksi
 Cycles to Failure Estimate

90
70

50 Base

30 Fillet
20

10
 Multi-axis Fluctuating
Stress States
Everything presented on fatigue has been based on
experiments involving a single stress component.

What do you do for problems in which there are

more than one stress component?
 Marin Load Factor, kc

Se = k a ⋅ k b ⋅ k c ⋅ k d ⋅ S′e
The endurance limit is a function of the
load/stress component used in the test.

ì0.923 Axial loading Sut ≤ 220 ksi (1520 MPa)

ï 1 Axial loading Sut > 220 ksi (1520 MPa)
ï
kc = í
ï 1 Bending
ïî0.577 Torsion and shear
 Alternating and Mean Von
Mises Stresses
1. Increase the stress caused by an axial force by
1/kc.
2. Multiply each stress component by the
appropriate fatigue stress concentration factor.
3. Compute the maximum and minimum von Mises
stresses.
4. Compute the alternating and mean stresses based
on the maximum and minimum values of the von
Mises stress.
5. Use the Goodman alternating and mean stress
interaction curve and S-N curve to estimate the
number of cycles to failure. Use the reversed
bending endurance limit.
 Complex Loads

A part is subjected
σ, F to completely reversed
σ2 stresses as follows
σ1 σ3 σ1 for n1 cycles,
σ 2 for n 2 cycles,
t, time σ 3 for n 3 cycles,
M
σ m for n m cycles,

What is the cumulative effect of these different load cycles?

 Minor’s Rule
Cumulative Damage Law

n1 n 2 n 3 nm
+ + +K + =C
N1 N 2 N 3 Nm
n i ≡ number of cycles for stress level i
N i ≡ cycles to failure at stress level i
C ≡ Constant ranging from 0.7 to 2.2.
C is usually taken as 1.0
Minor’s Rule is the simplest and most widely used
Cumulative Damage Law
 Example

Stress Cycles Life n

State (n) (N) N
1 1,000 2,000 0.5
2 5,000 10,000 0.5
3 10,000 100,000 0.1
1.1 Part will fail
 Assignment
(Problem No. 1)

A rotating shaft is made of 42 x 4 mm AISI 1020 cold-drawn

steel tubing and has a 6-mm diameter hole drilled transversely
through it. Estimate the factor of safety guarding against
fatigue failure when the shaft is subjected to a completely
reversed torque of 120 N-m in phase with a completely
reversed bending moment of 150 N-m. Use the stress
concentration factor tables found in the appendices, and
estimate the Marin factors using information in the body of the
text.
 Assignment
(Problem No. 2)

A solid circular bar with a 5/8 inch diameter is subjected to a

reversed bending moment of 1200 in-lb for 2000 cycles,
1000 in-lb for 100,000 cycles and 900 in-lb for 10,000
cycles. Use the S-N curve used in this lecture. Determine
whether the bar will fail due to fatigue. Assume all Marin
factors are equal to 1.0.
 Assignment
(Problem No. 3)

Same as Problem No. 2 except there is a constant

axial force of 5,000 lb acting on the bar in
addition to the completely reversed bending
moment.
Crack Propagation

Lecture 12

Engineering 473
Machine Design
 Evidence of Crack Growth

Crack Growth Crack Growth

Striations in shaft with Striations in
keyway crankshaft
Norton, Fig. 6-4
Crack Propagation Striations

Concentric pattern is characteristic

of fatigue failures in which a
crack propagates (grows) under
cyclic loading.

Norton, Fig. 6-3

Varying Alternating Stress
Beachmarks
The ripples are called
“beachmarks” and can often
be distinguished by the
unaided eye. In some cases
they can only be seen with the
aided of and electron
microscope. These fine
markings represent stepwise
advances of the crack front.
These beachmarks show the
progression of a crack front
due to an alternating stress
having different magnitudes.
Bela I. Sandor, Fundamentals of Cyclic Stress and
Strain, The University of Wisconsin Press, 1972, p. 143.
 Crack Opening
Displacement (COD) Tests
COD Tests are used to determine the change in crack size in
compact tension specimens subjected to cyclic loads.

www.testresources.com
Stress Intensity Factor
Range
Stress Intensity Equation

K = Yσ nom π ⋅ a
(Refer to Class 7 Notes)

Stress Intensity Factor Range

K max = Yσ max π ⋅ a
K min = Yσ min π ⋅ a
∆K = Y π ⋅ a ⋅ (σ max − σ min )

If K min < 0, ∆K = K max

Norton, Fig. 5-15
 Crack Growth Rate Curve
Phase III
æ da ö
logç ÷
è dN ø
K c is reached
Phase I
Phase II

∆Kth log(∆K )

Phase I: Crack initiation

Phase II: Stable crack-growth
Phase III: Unstable crack-growth (fracture)
Example da/dN Curve

Phase II crack
growth data for 316
and 317 stainless
steel at –452 oF.

Stainless steels are used

extensively at cryogenic
temperatures.
 Griffith-Irwin Formula
Empirical equation used to fit da/dN data

da C(∆K − ∆K th )
m
=
dN (1 − R )K c − ∆K

C, m ≡ empirical constants
σ min
R ≡ stress amplitude ratio =
σ max
K c ≡ Fracture toughness
∆K th ≡ threshold value, crack growth is not
expected for values of ∆K less than ∆K th.
Griffith-Irwin Formula Plots

Norton, Fig. 6-20

 Stable Crack Growth Regime

da C(∆K − ∆K th )
m
= Griffith-Irwin formula
dN (1 − R )K c − ∆K

da
= C ⋅ ∆K m Phase II simplified formula
dN

These equations can be integrated (usually numerically)

to obtain the number of cycles to failure. There are
several commercially available programs containing
material libraries that make this a fairly straight forward
process. NASA uses a program called FLAWGRO.
Thick Walled Cylinders

Lecture 13

Engineering 473
Machine Design
 Axisymmetric Equation of
Equilibrium
(Geometry)

p i ≡ internal pressure
θ ≡ angular position coordinate
r ≡ radial position coordinate
u ≡ displacement in r - direction
v ≡ displacement in θ - direction

Axisymmetric ≡ Nothing varies in the θ - direction.

∂
=0
∂θ
Ugural, Fig. 8.1(a)
 Axisymmetric Equation of
Equilibrium
(Differential Element)

∂τ rθ
τ rθ + dθ
∂θ

τ rθ

∂τ rθ
= 0, due to axisymmetric constraint
∂θ
τ rθ = 0, due to stress compatibility
Ugural, Fig. 8.1(b)
 Axisymmetric Equation of
Equilibrium
æ ∂σ r ö æ dθ ö
ç σr + dr ÷ ⋅ (r + dr )dθ ⋅ dz - 2σ θ sin ç ÷drdz - σ r ⋅ rdθ ⋅ dz + Fr rdθ ⋅ dr ⋅ dz = 0
è ∂r ø è 2ø

dσ r
r + σ r − σ θ + rFr = 0
dr

dσ r σ r − σ θ
+ + Fr = 0
dr r

Fr ≡ radial body force per unit volume

Strain Displacement Equations

∂u
dr + dr − dr
∂r du
εr = =
dr dr

εθ =
(r + u )dθ - rdθ u
=
rdθ r

du u
εr = , εθ =
dr r
 Constitutive Equations
Hooke’s Law
Stress-Strain equations are often
1 referred to as constitutive
εr = (σ r − νσ θ ) equations, because they depend
E
on what the part is made of.
1 The equilibrium and strain-
ε θ = (σ θ − νσ r ) displacement equations are
E
independent of the material.

E Webster, “constitutive - making

σr = (ε r + νε θ )
1− ν 2
a thing what it is, essential”

E
σθ = (ε θ + νε r )
1− ν 2
 Summary of
Axisymmetric Equations

Equilibrium Equation Constitutive Equations

1
dσ r σ r − σ θ
+ + Fr = 0 ε r = (σ r − νσ θ )
dr r E

1
ε θ = (σ θ − νσ r )
Strain-Displacement Equations E

du u
εr = , εθ =
dr r
 Thick Walled Cylinders
(Displacement Differential Equation)
E
σr = (ε r + νε θ )
1− ν 2

E
σθ = (ε θ + νε r )
1− ν 2

E æ du uö
σr = 2 ç
+ν ÷
1 − ν è dr rø
E æu du ö
σθ = 2 ç
+ν ÷
1− ν è r dr ø
a ≡ inside radius
b ≡ outside radius
d 2 u 1 du u
p i ≡ internal pressure 2
+ − 2 =0
dr r dr r
p o ≡ external pressure
Ugural, Fig. 8.2
 Thick Walled Cylinders
(General Solution & Boundary Conditions)

E æ du uö
σr = 2 ç
+ ν ÷
1 − ν è dr rø
E æu du ö
σθ = 2 ç
+ν ÷
1− ν è r dr ø

E é æ 1 − ν öù
êC1 (1 + ν ) − C 2 ç r 2 ÷ú
d 2 u 1 du u σr =
+ − 2 =0
dr 2
r dr r 1− ν2 ë è øû
E é æ 1 − ν öù
General Solution
σθ =
1− ν2 êC1 (1 + ν ) + C 2 ç r 2 ÷ú
ë è øû
C2
u = C1r +
r Ugural, Fig. 8.2
 Thick Walled Cylinders
(Boundary Conditions)

E é æ 1 − ν öù
σr = C
ê 1 (1 + ν ) − C 2ç 2 ÷ú
1− ν2 ë è r øû

E é æ 1 − ν öù
− pi = C
ê 1 (1 + ν ) − C 2ç 2 ÷ú
1− ν2
ë è a øû
E é æ 1 − ν öù
− po = C (1 + ν ) − C 2 ç 2 ÷ú
2 ê 1
1− ν ë è b øû
Boundary Conditions
1 − ν é a 2 pi − b 2 po ù
σr r =a
= −p i C1 = ê ú
E ë b2 − a 2 û
1 + ν é a 2 b 2 (p i − p o )ù
σr = −p o C2 = ê ú
r =b E ë b2 − a 2 û
Ugural, Fig. 8.2
Thick Walled Cylinders
(Lame’ Equations)

a 2 p i − b 2 p o (p i − p o )a 2 b 2
σr = −
b −a
2 2
(
b2 − a 2 r 2)
a 2 p i − b 2 p o (p i − p o )a 2 b 2
σθ = +
b −a
2 2
(
b2 − a 2 r 2)

u=
( )
1 − ν a 2 p i − b 2 p o r 1 + ν (p i − p o )a 2 b 2
+
E b −a
2 2
E (
b2 − a 2 r )

Ugural, Fig. 8.2

 Longitudinal Strain
(Unconstrained and Open Ends)
1
ε z = (σ z − νσ r − νσ θ )
E
2E
σ r + σθ = [C1 (1 + ν )]
1− ν 2
Ends are unconstrained
and open, σ z = 0
− 2ν ⋅ C1
εz =
1− ν
ν
ε z = − (σ r + σ θ )
E
2ν æ a 2 p i − b 2 p o ö
E é 1 − ν ù ε z = − çç ÷÷
æ ö E è b −a ø
( )
2 2
σr = C 1 + ν − C ç ÷
1 − ν 2 êë ú
1 2 2
è r øû
E é æ 1 − ν öù
σθ = 2 ê 1
C (1 + ν ) + C 2 ç 2 ÷ú Note that σ r + σ θ = Constant
1− ν ë è r øû
 Longitudinal Stress
(Constrained Ends)

2E
1 σ r + σθ = [C1 (1 + ν )]
ε z = 0 = (σ z − νσ r − νσ θ ) 1− ν 2

æ 2EC1 ö
σ z = ν(σ r + σ θ ) σ z = νç ÷
è 1− ν ø

æ a 2pi − b2po ö
σ z = 2νçç ÷÷
E é æ 1 − ν öù
êC1 (1 + ν ) − C 2 ç r 2 ÷ú è b −a ø
2 2
σr =
1− ν2 ë è øû
E é æ 1 − ν öù
σθ = C (1 + ν ) + C 2 ç 2 ÷ú
2 ê 1 Note that σ z = Constant
1− ν ë è r øû
 Longitudinal Stress
(Closed and Unconstrained Ends)

σz
po pi
σz

( )
σ z π b 2 − a 2 + po π ⋅ b 2 − pi π ⋅ a 2 = 0

pi a 2 − p o b 2
σz =
b2 − a 2
 Special Cases
Internal Pressure Only External Pressure Only
a 2 pi æ b 2 ö b2po æ a 2 ö
σr = 2 ç1 − 2 ÷÷
2 ç σr = − 2 ç1 − 2 ÷÷
b −a è r ø 2 ç
b −a è r ø

a 2 pi æ b 2 ö b2po æ a 2 ö
σθ = 2 ç1 + 2 ÷÷
2 ç
σθ = − 2 ç1 + 2 ÷÷
2 ç
b −a è r ø b −a è r ø

σ z = 0, unconstrained σ z = 0, unconstrained
2ν ⋅ a p i
2 2ν ⋅ b 2
po
σz = 2 , constrained σ z = − , constrained
b −a 2 b −a
2 2

2
2
a pi b po
σz = 2 , closed and unconstrained σz = − 2 , closed & unconstrained
b − a2 b −a 2
 Stress Variation
b/a=4

Internal Pressure Only External Pressure Only

Ugural, Fig. 8.3

 Assignment

1. Show that the Lame’ equations for the case of

internal pressure reduce to the equations for a thin
walled cylinder when the ratio b/a approaches 1.
2. A thick walled cylinder with 12 and 16 inch internal
and external diameters is fabricated of a material
whose tensile yield strength is 36 ksi and Poisson’s
ratio is 0.3. Calculate the von Mises stress when the
internal pressure is 10 ksi. The cylinder has closed
but unconstrained ends. Will the material yield?
Compound Cylinders
&
Discontinuity Stresses

Lecture 14

Engineering 473
Machine Design
Lame’ Equations for Thick
Walled Cylinders

a 2 p i − b 2 p o (p i − p o )a 2 b 2
σr = −
b −a
2 2
(
b2 − a 2 r 2)
a 2 p i − b 2 p o (p i − p o )a 2 b 2
σθ = +
b −a
2 2
(
b2 − a 2 r 2)

u=
( )
1 − ν a 2 p i − b 2 p o r 1 + ν (p i − p o )a 2 b 2
+
E b −a
2 2
E (
b2 − a 2 r )
 Compound Cylinders
Civil War Parrott Guns

Compound cylinders are used

to increase the pressure that
can be contained in cylinders.

www.wwd.net/steen
 Compound Cylinders
(Assembly)
Outer Cylinder (2)
Inner Cylinder (1)

c
a b−δ

The inside diameter of cylinder 2 is undersized by a small

amount (interference) and must be heated to fit over
cylinder 1. This is often referred to as a shrink fit.
 Compound Cylinders
(Interference Equations)
Lame’s Equation

u=
( )
1 − ν a 2 p i − b 2 p o r 1 + ν (p i − p o )a 2 b 2
+
E b −a
2 2
E (
b2 − a 2 r )
pb æ b 2 + a 2 ö Inner
u1 = ç
ç − ν i ÷÷
Ei è b − a
2 2
ø Cylinder

pb æ b 2 + c 2 ö Outer
u2 = çç 2 + ν o ÷÷
Eo è c − b 2
ø Cylinder

The interface pressure is

directly proportional to the pb æ b 2 + c 2 ö pb æ b 2 + a 2 ö
δ= çç 2 + ν o ÷÷ + çç 2 − ν i ÷÷
interference. Eo è c − b 2
ø Ei è b − a
2
ø
Compound Cylinders
(Interference Pressure)

pb æ b 2 + c 2 ö pb æ b 2 + a 2 ö
δ= ç
ç + νo ÷ + ç 2
÷ ç − ν i ÷÷
Eo è c − b
2 2
ø Ei è b − a
2
ø

p=
( )(
Eδ b 2 − a 2 c 2 − b 2 ) For same
b (
2b 2 c 2 − a 2 ) materials

The interference pressure is that

pressure needed to compress the
inner cylinder and expand the outer
cylinder so that the two cylinders
can be assembled.
Ugural, Fig. 8.5
Compound Cylinder
(Shrink Fit Stresses)
Hoop stress due
to interference
pressure

Hoop stress due

to an internal
pressure

Total Hoop Stress

Harvey, Fig. 2.23

 Example Problem
Determine the tangential (hoop) stresses at the inner, outer,
and mating surfaces of a compound cylinder subjected to an
internal pressure of 20,000 psi. The radii are: a=6 in, b=8 in,
and c=10 in. The material is steel with a modulus of
elasticity of E=30x106 psi, and the interference is 0.004 in.
 Example Problem
(Interference Pressure)

p=
( )(
Eδ b 2 − a 2 c 2 − b 2 )
b (
2b 2 c 2 − a 2)
p = 1,850 psi

a=6 in E = 30x10 6 psi

b=8 in δ = 0.004 in
c=10 in
 Example Problem
(Inner Cylinder)

Lame’ Equation
a 2 p i − b 2 p o (p i − p o )a 2 b 2
σθ = +
b −a
2 2
(
b2 − a 2 r 2)
External Pressure
− b2p − pa 2 b 2
σθ = 2 + 2
b −a 2
(
b − a2 r2 )
Interference Stresses
− 2pb 2
σθ = 2 = −8,440 psi
E = 30x10 psi
6 r =a
b −a 2
a=6 in
b=8 in δ = 0.004 in − 2p(b 2 + a 2 )
σθ = = −6,590psi
c=10 in
r =b
b −a
2 2
p = 1,850 psi
 Example Problem
(Outer Cylinder)

Lame’ Equation
b 2 p i − c 2 p o (p i − p o )b 2 c 2
σθ = + 2
c −b
2 2
(
c − b2 r 2 )
External Pressure
b2p pb 2 c 2
σθ = 2 + 2
c −b 2
(
c − b2 r 2 )
Interference Stresses
æ b2 + c2 ö
σ θ r = b = pçç 2 ÷ = 8,410 psi
2 ÷
a=6 in E = 30x10 6 psi èc −b ø
b=8 in δ = 0.004 in 2pb 2
c=10 in σ θ r =c = 2 = 6,560 psi
p = 1,850 psi c −b 2
 Example Problem
(Shrink-fit Stress Distribution)

Outer Cylinder

b c

Inner Cylinder
a=6 in E = 30x10 6 psi
b=8 in δ = 0.004 in
c=10 in p = 1,850 psi
Harvey, Fig. 2.23(a)
 Example Problem
(Internal Pressure)
Lame’ Equation
a 2 p i − c 2 p o (p i − p o )a 2 c 2
σθ = + 2
c −a
2 2
(
c − a2 r2 )
Internal Pressure
a 2 pi pia 2c 2
σθ = 2 + 2
c −a 2
(
c − a2 r2 )
Radial Locations
σθ r =a
= 42,500 psi
a=6 in E = 30x10 6 psi σθ = 28,830 psi
r =b
b=8 in δ = 0.004 in σθ = 22,500 psi
r =c
c=10 in p i = 20,000 psi
 Example Problem
(Internal Pressure Stress Distribution)

Outer Cylinder

c
b

a=6 in E = 30x10 6 psi Inner Cylinder

b=8 in δ = 0.004 in
c=10 in p i = 20,000 psi
Harvey, Fig. 2.23(b)
Example Problem
(Total Stresses)

Note that the total

stress on the inside of
the cylinder is
approximately 20%
less than what it
would have been
without the shrink-fit
stresses.

Harvey, Fig. 2.23

 Discontinuity Stresses

• The stresses in thick and thin walled cylinders (pressure

vessels) considered so far have considered only the
cylinder.
• There are often high stresses at geometric
discontinuities in cylinders.
• These high stresses are similar to stress concentrations.
 Discontinuity Stresses
(Example)

Discontinuity at Hemispherical Head and

Cylindrical Shell Juncture

The force and moment required to cause the head and shell to
displace and rotate the same amount will cause local bending
stresses in both the head and shell.
Harvey, Fig. 4.8
 Discontinuity Stresses
(Example)

Discontinuity at Flat Head and

Cylindrical Shell Juncture

Hand stress analysis or finite element methods may

be used to accurately compute the stresses around
geometric discontinuities.
 Assignment

1. What is the required thickness of a 6 ft inside

diameter cylinder, considering it as a thin wall
vessel, to withstand an internal pressure of 1,000 psi
if the allowable tangential stress is 20,000 psi.
2. A cylinder with a 48 in inside diameter and a 60 in
outside diameter is subjected to an internal pressure
of 5,000 psi. Determine value and place of
occurrence the maximum tangential stress, the
maximum radial stress, and the maximum shear
stress.
 Assignment
(Continued)

3. Determine the tangential (hoop) stresses at the inside

radius of a compound cylinder subjected to an internal
pressure of 32,000 psi. The radii are: a=10 in, b=12 in, and
c=13 in. The material is steel with a modulus of elasticity of
E=30x106 psi, and the interference is 0.005 in.
Interference Fits

Lecture 15

Engineering 473
Machine Design
 Standards

US Customary
Preferred Limits and Fits for Cylindrical
Parts, ANSI B4.1-1967.

Metric
Preferred Metric Limits and Fits, ANSI
B4.2-1978.
Limits and Fits
(Metric Nomenclature)

D ≡ basic size of hole

d ≡ basic size of shaft
δ u ≡ upper deviation
δ l ≡ lower deviation
δ F ≡ fundamental deviation
∆D ≡ tolerance grade for hole
∆d ≡ tolerance grade for shaft

Shigley, 4-9
Tolerance Grade Numbers

Tolerance – difference
between the maximum and
minimum size limits of a part.

International Tolerance
Grade Numbers are used to
specify the size of a tolerance
zone.

In the ANSI standard, the

tolerance is the same for both
the internal (hole) and external
(shaft) parts having the same
Tolerance Grade Numbers.

Shigley, 4-9
Tolerance Grade Numbers

IT0 through IT16 are

contained in the
standard.

Grade Number

IT11

International Tolerance

Shigley, 4-9
Tolerance Grades Table

Shigley, Table A-11

Fundamental Deviations
Example of Fit Specification
32H7 Hole
32g6 Shaft
Upper Case => Hole
Lower Case=> shaft

δ F , Fundamental Deviation

32g6

Tolerance Grade, IT6

D = d = 32 mm (Basic Size)
Shigley, 4-9
Fundamental Deviations
for Shafts

Shigley, Table A-12

Lower and Upper Deviations
Shaft letter codes c,d,f,g, and h
Upper deviation = fundamental
deviation
Lower deviation = upper deviation –
tolerance grade

Shaft letter codes k,n,p,s, and u

Lower deviation = fundamental
deviation
Upper deviation = lower deviation +
tolerance grade

Hole letter code H

Lower deviation = 0
Upper deviation = tolerance grade
Shigley, 4-9
Preferred Fits Using the
Basic-Hole System

Shigley Table 4-5

 Loose Running Fit
(Example)
Determine the “loose running fit” tolerances for a
shaft and hole that have a basic diameter of 32 mm.

From Table 4-5, Specification is 32H11/32c11

Hole Shaft
Tolerance Grade 0.160 mm 0.160 mm (0.0063 in)
Upper deviation 0.160 mm -0.120 mm
Lower deviation 0.000 mm -0.280 mm
Max Diameter 32.160 mm (1.266 in) 31.880 mm (1.255 in)
Min Diameter 32.000 mm (1.260 in) 31.720 mm (1.225 in)
Ave Diameter 32.080 mm (1.263 in) 31.800 mm (1.252 in)

Max Clearance C max = D max − d min = 0.44 mm (0.017 in )

Min Clearance C min = D min − d max = 0.12 mm (0.005 in)
 Loose Running Fit
(Example Continued)

Dimension Tolerances Shown on Drawing

Hole Shaft

32.080 +−00..080
080 31.800 +−00..080
080
 Force Fit
(Example)
Determine the “force fit” tolerances for a shaft and
hole that have a basic diameter of 32 mm.
From Table 4-5, Specification is 32H7/32u6
Hole Shaft
Tolerance Grade 0.025 mm (0.001 in) 0.016 mm (0.0006 in)
Upper deviation 0.025 mm 0.076 mm
Lower deviation 0.000 mm 0.060 mm
Max Diameter 32.025 mm (1.261 in) 32.076 mm (1.262 in)
Min Diameter 32.000 mm (1.260 in) 32.060 mm (1.262 in)
Ave Diameter 32.013 mm (1.260 in) 32.068 mm (1.263 in)

Max Clearance C max = D max − d min = −0.035 mm (- 0.001 in )

Min Clearance C min = D min − d max = −0.076 mm (-0.003 in)
 Force Fit
(Example Continued)

Dimension Tolerances Shown on Drawing

Hole Shaft

32.013+−00..012
013 32.068+−00..008
008
Interference Pressures
& Torques

How much pressure exists in a force

fit, and how much torque can it
transmit?
Deutschman, Fig. 18-2
Lame’ Equations for Thick
Walled Cylinders

a 2 p i − b 2 p o (p i − p o )a 2 b 2
σr = −
b −a
2 2
(
b2 − a 2 r 2)
a 2 p i − b 2 p o (p i − p o )a 2 b 2
σθ = +
b −a
2 2
(
b2 − a 2 r 2)

u=
( )
1 − ν a 2 p i − b 2 p o r 1 + ν (p i − p o )a 2 b 2
+
E b −a
2 2
E (
b2 − a 2 r )
Shaft Displacement and
Stresses

For external pressure and

zero inside radius, Lame’s
equations reduce to -
a
σ r = −p

p σ θ = −p

1 − νs
us = − pa
Es
Hub Displacement and
Stresses

a 2 - b2
σr = p 2
b − a2

a 2 + b2
σθ = p 2
b − a2

pa æ a 2 + b 2 ö
uh = çç 2 + ν h ÷÷
po = 0 Eh è b − a 2
ø
Interference Pressure Equation

æ b2 + a 2 ö
ç 2 + νh ÷
ç b − a 2 1 − νs ÷
C = u h − u s = ap +
ç Eh Es ÷
ç ÷
è ø

C
p=
æ b2 + a 2 ö
ç 2 + νh ÷
1 − νs ÷
aç b − a
2
+
ç Eh Es ÷
ç ÷
è ø
 Maximum Torque without
Slipping

Fn = p ⋅ 2 ⋅ π ⋅ a ⋅ L
a
Ff = µFn

p Torque = Ff ⋅ a

L= Hub Thickness
 Assignment
1. A 4-in diameter, 2-in face width, 20-tooth cast iron
pinion gear is to transmit a maximum torque of 1200 in-
lb at low speed. Find the required radial interference on
1 in diameter steel shaft and the stress in the gear due to
the press fit. Use the dedendum radius as the outside
radius of the pinion gear.
2. Determine the dimension and tolerance to be specified
on a drawing for a shaft and hole having a basic size of
50 mm. The fit must allow a snug fit but be freely
assembled and disassembled.
Stresses in Rotating Disks

Lecture 16

Engineering 473
Machine Design
 Summary of
Axisymmetric Equations
Equilibrium Equation Constitutive Equations
1
dσ r σ r − σ θ
+ + Fr = 0
εr = (σ r − υσθ )
E
dr r
1
ε θ = (σ θ − υσ r )
E
Strain-Displacement Equations
or
du u
εr = , εθ = E
dr r σr = (ε r + υε θ )
1− ν 2

E
σθ = (ε θ + υε r )
1− ν 2
 Rotating Disk
Fr Equilibrium Diff Equation
dθ
dσ r σ r − σ θ
θ + + Fr = 0
dr dr r
r ω
Fr = ρrω2

dσ r σ r − σθ
+ + ρrω2 = 0
dr r

Fr ≡ radial body force per unit volume

 Displacement Base Equilibrium
Equation
Equilibrium Equation Combining the equilibrium and
constitutive equations yields
dσ r σ r − σθ
+ + ρrω2 = 0
dr r
d 2 u 1 du u
( )
2
2 ρrω
2
+ − r = − 1− υ
Constitutive Equations dr r dr r E
E
σr = (ε r + υε θ ) This equation is the
1− ν 2

differential equation of
equilibrium written in terms
E
σθ = (εθ + υε r ) of the radial displacement
1− ν 2
component.
 General Solution
Differential Equation of Equilibrium

( )
2 2
d u 1 du u 2 ρrω
2
+ − r = − 1− υ
dr r dr r E
Homogeneous Solution Particular Solution

( )
3 2
C2 ρr ω
u h = C1r + u p = − 1 − υ2
r 8E
The homogeneous solution
is the same as the general General Solution
solution for the thick walled
( )
3 2
C2 ρr ω
cylinder. u = C1r + − 1− υ 2

r 8E
 Stress Distributions
Constitutive Equations General Solution - Displacement
E
σr = (ε r + υε θ )
( )
3 2
1− υ 2
C2 ρr ω
u = C1r + − 1− υ 2

r 8E
E
σθ = (εθ + υε r )
1− υ 2

Displacement Based General Solution - Stress

E æ du uö
σr = 2 ç
1 − υ è dr
+υ ÷
rø σr =
( )
E é − (3 + υ ) 1 − υ2 ρr 2ω2
+ (1 + υ )C − (1 − υ ) C2 ù
1 − υ2 êë r 2 úû
1
8E

σθ =
( )
E é − (1 + 3υ ) 1 − υ2 ρr 2ω2 C2 ù
+ (1 + υ )C1 + (1 − υ ) 2 ú
E æu du ö 2 ê
σθ = 2 ç
+ υ ÷ 1− υ ë 8E r û
1− υ è r dr ø
 Annular Rotating Disk

Boundary Conditions

σ r (a ) = 0
b
ω σ r (b ) = 0
a

This disk has a hole in the center.

Constant Determination for Annular
Rotating Disk
E é − (3 + υ )(1 − υ )ρa ω 2 2 2
C ù
σ (a ) = + (1 + υ )C − (1 − υ ) 2
1 − υ2 êë a úû
r 1 2
8E
=0

(
σr b ) =
(
E é − (3 + υ ) 1 − υ2 ρb 2ω2
+ ()+ ) − ( − )
1 υ C1 1 υ 2 ú
C2 ù
2 ê
1− υ ë 8E b û
=0
Multiplying the top equation by a2 and the bottom by b2
and then subtracting the two equations yields

C1 = ρω 2 (a 2
)
+ b 2 (1 − υ )(3 + υ )
E 8
 Constant Determination
(Continued)

( )
σr a =
E ( )
é − (3 + υ ) 1 − υ 2 ρa 2ω2 C2 ù
+ (1 + υ )C1 − (1 − υ ) 2 ú = 0
1 − υ2 ê 8E a û
ë

σ r (b ) =
(
E é − (3 + υ ) 1 − υ 2 ρb 2ω2) C2 ù
+ (1 + υ )C1 − (1 − υ ) 2 ú = 0
2 ê
1− υ ë 8E b û

C1 = ρω 2 (a 2
)
+ b 2 (1 − υ )(3 + υ )
E 8

2 a b ö (1 + υ )(3 + υ )
æ 2 2
C2 = ρω çç ÷÷
è E ø 8
Annular Rotating Disk Equations

3+ υæ 2 a b ö 2
2 2
σr = çç a + b − r − 2 ÷÷ρω
2 2

8 è r ø
3+ υæ 2 1 + 3υ a b ö 2
2 2
σθ = çç a + b −
2
r + 2 ÷÷ρω
2

8 è 3+ υ r ø

u=
(3 + υ )(1 − υ ) æ 2
ça + b −
2 1 + υ
r +
2 1 + υ a b ö
2 2
÷ρrω2
8E ç 3+ υ 1 − υ r 2 ÷ø
è

Note that r=a and r=b, that the radial

stress component is zero.
Stress and Displacement Variation
Through the Thickness

Ugural, Fig. 8.6

 Solid Rotating Disk

Boundary Conditions

b σ(b ) = 0
ω
u (0 ) = 0
 Solid Rotating Disk
(Continued)

( )
3 2
C2 ρr ω
u = C1r + − 1− ν 2

r 8E

σr =
E ( )
é − (3 + υ ) 1 − υ2 ρr 2ω2 C2 ù
+ (1 + υ )C1 − (1 − υ ) 2 ú
1 − υ2 ê 8E r û
ë

Since the displacement must be finite at r = 0, C2 = 0

C1 = ρω 2 b 2
(1 − υ )(3 + υ )
E 8
 Solid Rotating Disk Stress and
Displacement Equations
3+ υ 2 2
σr =
8
( )
b − r ρω2

3 + υ æ 2 1 + 3υ 2 ö 2
σθ = çb − r ÷ρω
8 è 3+ υ ø

u=
1− υ
8E
[ ]
(3 + υ)b 2 − (1 + υ)r 2 ρrω2

Note that these equations satisfy the boundary conditions.

Stress and Displacement Variation
 Other Solutions

Solutions to the governing differential equations exist

for variable thickness geometries and for constant
stress conditions.

Variable Thickness Constant Stress

Ugural, Fig. 8.8 & 8.9

 Complex Geometries

Complex geometries must be solved using numerical methods.

 Assignment
A flat 20 inch outer diameter, 4 inch inner diameter,
and 3 inch thick steel disk is shrunk onto a steel
shaft. If the assembly is to run safely at 6900 rpm,
determine: (a) the required interference (inches), (b)
the maximum stress when not rotating, and (c) the
maximum stress when rotating. The material
properties are ρ=0.00072 lb-sec2/in4, E=30x106 psi,
and ν=0.3.
Shaft Loading

Lecture 17

Engineering 473
Machine Design
 Shaft Design Issues
Shaft Rotating machine element
Shaft:
that transmits power. Loads
Material
Stationary
Se Sut Rotating
K IC Syt
Interfaces
RC q Press Fits
Environment Keyways
Temperature Splines
Corrosion Bearings
Magnetic
Assembly
Tolerances Stiffness
Shafts are one of the most commonly
encountered machine components.
Mott, Fig. 5-1
Parallel Shaft Gear Box

Shaft design spans most topics taught

in a Machine Design Course.
Mott, Fig. 15-7
Design Detail Needed to
Specify a Shaft
Significant detail is
required to completely
specify the geometry
needed to fabricate a
shaft.

Mott, Fig. 15-5

 Common Shaft
Loading Mechanisms
Spur Gears Chain Drives Unbalanced Mass

Helical Gears
Spiral Bevel Gears Belt Drives
Spur Gear Loads
T = 63,000 ⋅ P
n
T
Wt =
D
2
Wr = Wt ⋅ tanφ

P ≡ transmitted power [hp]

n ≡ rotational speed [rpm]
T ≡ shaft torque [in ⋅ lb]
D ≡ pitch diameter [in ]
φ ≡ pressure angle
Mott, Fig. 12-3
Helical Gear Loads

Mott Fig’s 10-3 & 10-4(a)

Helical Gear Loads
(Continued)
φ n ≡ Normal pressure angle
φ t ≡ Transverse pressure angle
ψ ≡ Helix angle
Wn ≡ Resultant normal force
Wt ≡ Transmitted force
Wr ≡ Radial force
Wx ≡ Axial force

tanφ n = tanφ t ⋅ cosψ

Mott Fig’s 10-4(a)

Helical Gear Loads
(Continued)

Mott Fig 10-4

Helical Gear Loads
(Continued)

T = 63,000 ⋅ P
n
T
Wt =
D
2

Wr = Wt ⋅ tan φ n /cos ψ

Wx = Wt ⋅ tan ψ

Mott Fig 10-4

Chain Drive Loads

Mott Fig 12-4

Belt Drive Loads

Net Driving Force

Fn = F1 − F2
T
Fn =
D
2
Total Bending Force
Fb = F1 + F2

Mott, Fig. 12-5

 Belt Drive Loads
(Bending Force)
Net Driving Force
FB F1 + F2
Fn = F1 − F2 C= =
Fn F1 − F2
T
Fn =
D
2 5F2 + F2
C= = 1.5 (V-belts)
Total Bending Force 5F2 − F2
Fb = F1 + F2 3F2 + F2
C= = 2.0 (Flat-belts)
3F2 − F2
Tension Ratio
F1
= 5.0 (V-belts)
FB = 1.5 Fn (V-belts)
F2
F1 FB = 2.0 Fn (Flat-belts)
= 3.0 (Flat-belts)
F2
 Stationary Loads
F2 F2

F1 F1

F2 F2

F1 F1
 Bending Stresses Due to
Stationary Loads
3
M 2 c 3 M 3c 2
σb = −
M3 I 22 I33

c2
c3 M 2 r sin θ M 3 r cos θ Eq. 1
θ 2 σb = −
I I
M2

∂σ b M 2 r/ cos θ M 3 r/ sin θ
= + =0
∂θ I/ I/
I 22 = I 33 = I
c 2 = r cos θ M2 Eq. 2
tan θ = -
c 3 = r sin θ M3
 Bending Stresses Due to
Stationary Loads
3 Eq. 3
M = M 22 + M 32
M3
Combining with Eq. 2,
c2
c3 2
θ − M2
M2 M sin θ
tan θ = =
M3 cos θ
M
M 2 r sin θ M 3 r cos θ Eq. 1
σb = −
I I − M2
sin θ =
M Eq. 4
M M3
tan θ = - 2 Eq. 2 cos θ =
M3 M
 3
Bending Stresses Due to
Stationary Loads
M3
Combining Eq’3 1,3, and 4
c2
c3 2
θ M 2 r sin θ M 3 r cos θ
σb = −
M2 I I

M r sin θ M 3 r cos θ
Eq. 1 M 22 r M 32 r
σb = 2 − σb = − −
I I I I
M = M 22 + M 32 Eq. 3
− M2 M 22 + M 32 ⋅ r
sin θ = σb = −
M I
Eq’s 4
M3
cos θ =
M
 Bending Stresses Due to
Stationary Loads

M 22 + M 32 ⋅ r M 22 + M 32 ⋅ r
σ b,max = σ b,min =−
I I

Mott, Fig. 5-3(e)

 Torsional Stresses Due to
Stationary Loads
3
M 1r
M3
τ=
J

r τ
2
M1 M2

time

The torsional stress at a point will be

constant under steady state conditions.
Axial Stresses Due to
Stationary Loads

Helical, worm, and spiral

gears will generate axial
loads in the shaft. Under
steady state conditions,
the axial stress from
these loads will be
constant.

Wx
σx =
A

Mott Fig 10-4

 Unbalanced Mass Loads

Bending stresses in a shaft due to in-balance loads are

complicated by whether the rotational speed is lower or
higher than the critical speeds of the shaft. In practice, the
in-balance loads are minimized by balancing the shaft and
attached components as a system. Rotordynamics theory is
required if the magnitudes of the stresses at a particular
operating speed is required.
 Synchronous Whirl
(Due to Unbalanced Mass)
m=unbalanced mass

φ
e

meω cos(ωt - φ )
2 meω 2sin (ωt - φ )
xs = ys =
(k − mω ) + (cω)
2 2 2 (k − mω ) + (cω)
2 2 2

cω
OS = x + y =
2 2 meω 2
tan φ =
k − mω 2
(k − mω ) + (cω)
s s
2 2 2

Thomson, Fig. 3.4-2

 Assignment
(Problem 1)

The shaft rotating at 550 rpm

carries a spur gear B having
96 teeth and a diametral pitch
of 6. The teeth are of the
20o, full-depth, involute
form. The gear receives 30
hp from a pinion directly
above it.

Compute the torque delivered

to the shaft and the tangential
and radial forces exerted on
the shaft by the gear.
Mott, Fig. 12-20
 Assignment
(Problem 2)

The shaft rotating at 200 rpm

carries a 20-in-diameter flat-
belt pulley at A that receives
10 hp from below.

Compute the torque delivered

by the pulley to the shaft and
the force exerted on the shaft
by the pulley.

Mott, Fig. 12-21

 Assignment
(Problem 3)
The shaft is rotating at 650 rpm and
receives 7.5 hp through a flexible
coupling. The power is delivered to
an adjacent shaft through a single
helical gear B having a normal
pressure angle of 20o and a helix
angle of 15o.

(a) draw free-body diagrams for the DB=4.141 in

shaft in both the vertical and
horizontal planes, (b) find the
magnitude of the forces shown, (c)
draw the shearing force and bending
moment diagrams for the shaft in
both planes. Mott, Fig. 12-29
 Assignment
(Problem 4)
The shaft rotating at 480 rpm
carries a 10-in-diameter chain
sprocket at C that receives 11 hp
from a mating sprocket below
and to the left as shown.

Compute the torque delivered to

the shaft by the sprocket and the
total force exerted on the shaft
by the sprocket. Resolve the
force into its horizontal and
vertical components, and show
the net forces acting on the shaft
at C in the vertical and
horizontal directions.

Mott, Fig. 12-22

Shaft Analysis

Lecture 18

Engineering 473
Machine Design
 3
Bending Stresses Due to
Stationary Loads
M3
Combining Eq’3 1,3, and 4
c2
c3 2
θ M 2 r sin θ M 3 r cos θ
σb = −
M2 I I

M r sin θ M 3 r cos θ
Eq. 1 M 22 r M 32 r
σb = 2 − σb = − −
I I I I
M = M 22 + M 32 Eq. 3
− M2 M 22 + M 32 ⋅ r
sin θ = σb = −
M I
Eq’s 4
M3
cos θ =
M
 Shaft Stress Equations
(Bending and Torsion Only)
Reference Bending Stress Torsional Stress
Mc Tc
σx = τ xy =
M = M 2y + M 2z I J

64Md 32 M Td
c=d 2 σx = = 2 = 16T
2π ⋅ d 4
πd 3 τ xy =
πd 4 32 πd 3
πd 4 Principal Stresses
I=
64 σx æ σx ö
2

σ1 , σ 2 = ± ç ÷ + τ 2xy
2 è 2 ø
πd 4 16M
2
æ 16M ö æ 16T ö
2
J = 2I = σ1 , σ 2 = ± ç 3 ÷ +ç 3 ÷
32 πd 3
è πd ø è πd ø
16
[
σ1 , σ 2 = 3 M ± M 2 + T 2
πd
]
Yielding in Ductile Material
(No Stress Concentration)

Distortion Energy Theory

Syt
(σ2
1 + σ − σ1σ 2
2
2 )
1
2
≤
ns
σ3 = 0

Circular Shaft
16
[
σ1 , σ 2 = 3 M ± M 2 + T 2
πd
]
Distortion Energy Theory for Shaft

Syt
16
πd 3
(
4M 2
+ 3T 2
) 1
3
≤
ns
 Static Loading Minimum
Diameter

Syt
16
πd 3
(4M 2
+ 3T 2
) 1
3
≤
ns

1 This equation is useful in

æ 32n s 3 2 ö÷
3
coming up with an initial
d=ç M + T
2
ç πS 4 ÷ø shaft size early in the
è yt
design process.
 Shaft Stress Equations
(Bending, Axial, and Torsional Loading)

Axial Stress
32M 4P
σx = 3
+ 2
πd πd

Principal Stresses
2
σ1 , σ 2 = 3 é8M + Pd ±
êë
(8M + Pd )2 + (8T )2 ùú
πd û

Distortion Energy Theory

4 Syt
3
(8M + Pd ) + 48T ≤
2 2

πd ns
Fatigue Analysis of Shafts
Bending Stresses

Torsional Stresses
τ
Shear Stress

Time
 Shaft von Mises Stress
Equation
2 2 2
Von Mises Equation æ σx ö æ σx ö æ σx ö æ σx ö
σ1 = ç ÷ + 2ç ÷ ç ÷ + τ xy + ç ÷ + τ 2xy
2 2

Syt
( 2
1
σ1 + σ 2 − σ1σ 2 ≤
2 2
)
ns
è 2 ø è 2 ø è 2 ø è 2 ø
2 2 2
æ σx ö æ σx ö æ σx ö æ σx ö
σ 2 = ç ÷ − 2ç ÷ ç ÷ + τ xy + ç ÷ + τ 2xy
2 2

è 2 ø è 2 ø è 2 ø è 2 ø
Principal Stress
æ σ x ö ìïæ σ x ö ü
2 2
Equation 2 ï
σ1σ 2 = ç ÷ − íç ÷ + τ xy ý = − τ 2xy
2 è 2 ø ïîè 2 ø ïþ
σx æσ ö
σ1 , σ 2 = ± ç x ÷ + τ 2xy
2 è 2 ø
Shaft Equivalent Stress

Syt
σ eff = σ + 3τ
2
x
2
xy ≤
ns
 Effective Alternating and
Mean Stresses

σ eff = σ 2x + 3τ 2xy Since the bending stress is

completely reversed, the only mean
stress component is due to the shear
σ eff,a = σ x stress. Since the shear stress is
constant, the only alternating is
equal to the maximum bending
σ eff,m = 3τ xy stress.
 ANSI Standard Fatigue
Curve
A

ANSI Standard B106.1M-1985, “Design of Transmission

Shafting”, American National Standards Institute, 1985, is
based on the ASME Elliptic Fatigue Interaction Curve.
 ASME Elliptic
Fatigue Equation

2 2
æ n sk f σa ö æ n sσm ö
çç ÷÷ + ç ÷ =1
S ç S ÷
è e ø è yt ø

This equation is used to determine whether the shaft will

have infinite life. Note that the fatigue stress concentration
factor has only been applied to the alternating stress. Also,
the Marin factors need to be used to estimate the endurance
limit.
 Minimum Diameter Equation
(Bending and Torsion Only)

πd 3 2 2
S= æ n s k f M ö æ n sT 3 ö
32 çç ÷÷ + ç ÷ =1
ç ÷
è S ⋅ Se ø è 2S ⋅ S yt ø

σ=M
S é 2 2
ù
n æ
2
k M
æ s ö êç f ÷ ö 3 æ T ö
ç ÷ ç + ç ÷ ú =1
è S ø êè Se ÷ø 4 çè S yt ÷ø ú
τ=T ë û
2S

2 2
æ kf M ö 3 æ T ö
2 2
æ n sk f σa ö æ n sσm ö 32n s
çç ÷÷ + ç ÷ =1 çç ÷÷ + ç ÷ = 1
ç ÷ πd 3 ç ÷
è Se ø è S yt ø è Se ø 4 è Syt ø
 Minimum Diameter Equation
(Bending and Torsion Only)

2 2
32n s æ kf M ö 3 æ T ö This equation gives the
çç ÷÷ + ç ÷ = 1
πd 3 ç ÷ minimum diameter
è Se ø 4 è S yt ø
shaft that will result in
infinite fatigue life,
1
é 2 2 ù 3 and appears in the
ê 32n s æ kf M ö 3 æ T ö ú ANSI Standard.
d= çç ÷÷ + ç ÷
ê π ç ÷ ú
êë è Se ø 4 è S yt ø úû
 Summary of Shaft Stress
Analysis Methodology
1
é 2 2 ù 3
32n s æ kf M ö 3 æ T ö
d=ê çç ÷÷ + ç ÷ ú
ê π ç ÷ ú
êë è Se ø 4 è S yt ø úû
1. Establish factor of safety that will be used in the design.
2. Draw axial,shear, bending, and torsion diagrams. May
require orthogonal shear and bending diagrams.
3. Determine loads acting on the shaft (M,P, and T) at
critical locations.
4. Compute stress concentration factors.
5. Estimate endurance limit using test data or 0.5Sut and
the Marin factors.
6. Use above equation or equivalent to determine if shaft is
acceptable.
 Assignment
The shaft shown in the figure receives 110 hp from a
water turbine through a chain sprocket at point C. The
gear pair at E delivers 80 hp to an electrical generator.
The V-belt sheave at A delivers 30 hp to a bucket
elevator that carries grain to an elevated hopper. The
shaft rotates at 1,700 rpm. The sprocket, sheave, and
gear are located axially by retaining rings. The sheave
and gear are keyed with sled runner keyseats, and there
is a profile keyseat at the sprocket. The shaft is made
from AISI 1040 cold-drawn steel, and has a yield
strength of 71 ksi and and ultimate strength of 80 ksi.
Using a design factor of safety of 3, determine the
minimum diameters at each section on of the shaft.
Assignment
(Continued)

Mott, Figure 12-13

Keys & Keyways

Lecture 19

Engineering 473
Machine Design
 Fundamental Problem in
Shaft Design

How do I connect stuff to the shaft?

Interference Fits

Keys & Keyseats Pins Hubs/Collars

Integral Shaft Splines/Polygons

 Interference Fits

Interference Fits – Hole is undersized and part is heated to

allow it to slide over shaft. Compressive interface pressure
develops when part cools. Reference Lecture 15 Notes.
Notes
 Keys and Keyseats

Keyseat

Keys are used to transmit torque

from a component to the shaft.

Mott, Figure 11-1

 Types of Keyseats

Keyseats are classified according to

the process by which they are made.
Olivo, Fig. 40-3
Keyway Fabrication Methods

End Mill on Vertical

Milling Machine

Key Cutter on Horizontal

Milling Machine

Chang, Fig. 5.8, Mott, Fig. 12-6

 Fillet Radii and Key Chamfers

45o chamfer
Not to Scale

Better Practice
Standard contains recommended fillet
radii and key chamfer combinations to
provide lower stress concentration
General Practice: Zero factors.
root fillet and chamfer “Keys and Keyseats,” ANSI
Standard B17.1-1967.
1967
 Fillet Radii Fabrication
R0.3125

A ‘Bull’ end mill can be used to

machine fillet radii in keyways.

Bull End Mill

MSC Catalog, Fanfara, Figure 3-6
 Square and Rectangular
Parallel Keys

The hub is slightly larger than the shaft and key to allow it to
slide over the shaft during assembly. The set screw is used to
take up the slack. The resulting friction is used to provide
resistance to axial motion. Thread adhesive may be required to
ensure that vibration doesn’t cause the set screw to loosen.
Mott, Figure 11-1
Square and Rectangular
Key Geometry
“Keys and Keyseats,” ANSI
Standard B17.1-1967.
1967

Width is approximately ¼
the diameter of the shaft.

Standard contains tables of

recommended key sizes versus
shaft diameter.
 Set Screws

Holding Power –
Resistance to axial or
rotary motion of the hub or
Flat Point Cup Point collar relative to the shaft.

Holding power is a
function of friction
Oval Point
between contacting
Cone Point
portions of hub or collar
and shaft and any
penetration of the setscrew
into the keyway or shaft.
Half-dog Point
Shigley, Fig. 8-26
Representative Holding
Power Values
Based on alloy steel screw
against steel shaft, class
3A coarse or fine threads
in class 2B holes, and cup-
point socket setscrews.

Shigley, Table 8-13

 Tapered Keys

Designed to be inserted from the end of the shaft after the hub
is in position. The taper will impart a compressive contact
pressure between the hub and the shaft. Friction will help
transmit torque and provide resistance to axial motion of the
hub relative to the shaft. Tapered keys do not require set
screws. Access to both ends of tapered keys are required so
that the key can be inserted and driven out when the key is
being removed.
Mott, Figure 11-3
 Gib Head Keys

Installation is similar to standard tapered keys.

The extended head provides a holding method for
removing the key by pulling instead of driving it
out.

Mott, Figure 11-3, Shigley, Fig. 8-28

 Woodruff Keys

Circular groove in shaft holds the key in place while the hub
is slid over the shaft. The Woodruff key will have less shear
strength than a rectangular or square key.

ANSI Standard B17.2-1967 lists recommended dimensions

for Woodruff Keys.
Mott, Figure 11-3
 Circular (Pin) Keys

Significantly lower stress concentration factors result from

this type of key as compared to parallel or tapered keys. A
ball end mill can be used to make the circular key seat.

Fanfara, Figure 3- 6, Mott, Figure 11-3

 Tapered Bushings
Tapered hub causes split bushing to be drawn down on
shaft. Higher strength alternative to set screws.

Key is used to transmit torque,

friction keeps system from
sliding axially along shaft.

Mott, Fig. 11-10, www.emerson-ept.com

 Stress Analysis of
Parallel Keys
A key has two failure mechanisms: 1) it can be
sheared off, and 2) it can be crushed due to the
compressive bearing forces.
Bearing Surface
Shear Plane
F
F

T
Shear Stress Analysis of Square
and Rectangular Parallel Keys
T
F=
D
2

A s = WL

F T
τ ave =
As
=
( )
D (WL )
2

2T
τ ave =
DWL

Mott, Fig. 11-4(b)

 Required Key Length
(Shear)

From Maximum Shear The minimum length of the

Stress Failure Theory, key can be found by setting
the shear yield strength the average shear stress equal
is given by: to the allowable shear stress.

Sys = 0.5Syt
2T 0.5S yt
τ ave = =
DWL N fs
0.5Syt
τ all =
N sf
4TN fs
Ls =
Syt DW
 Bearing Stress: Square and
Rectangular Parallel Keys

F 4TN fs
σb = KSyc Lb =
Ab σ b,all = KSyc DH
N fs

T
F= Triaxial Stress Factor
D
2
1.0 ≤ K ≤ 1.5

A b = HL
2 Mott, Fig. 11-4(a)
 Comparison of Shear and
Bearing Length Equations

Minimum Required Minimum Required

Length to Prevent Length to Prevent
Shear Failure Bearing Failure

4TN fs 4TN fs
Ls = Lb =
S yt DW KSyc DH

If K=1, these equations give the same result for a square key. In
general K will be greater than 1.0 and more shear failures will be
observed in the field. Keys are generally designed to fail before
overloads can cause damage to the shaft or attached component.
In this respect they act like a mechanical fuse.
 Stress Analysis of
Woodruff Keys

B
2
B
− ds
2

Shear Area ds

2 2
æ Bö æ B ö
ç ÷ = ç − ds ÷ + A
2

è2ø è2 ø
 Shear Analysis of
Woodruff Keys
2 2
æ Bö æ B ö
ç ÷ = ç − ds ÷ + A
2

B è2ø è2 ø
2
B
− ds
2 A = d s (B − d s )

A A s ≡ Shear Area = 2W d s (B − d s )

ds 0.5S yt
2T
τ ave = τ s,all =
DA s N fs
 Shear Pins

The strength analysis of shear pins is similar to that

used to find the strength of a fastener. We’ll defer the
strength analysis until we cover fasteners.

Mott, Fig. 11-9

 Stress Concentration Factors
Key seats create stress concentrations in the shaft. There are
different stress concentration factors for bending and torsional
loads. Peterson contains a compilation of stress concentration
factors that includes key seat geometries. For flat end mills,
Peterson gives Kt=2.14 for bending and Kt=2.62 for torsion.
These may be reduced by using key seats made with bull end
mills. The stress concentration factor for a sled runner key seat
is significantly lower than for a profile key seat. A circular key
and keyseat will have lower stress concentration factors any
other key geometry.

R.E. Peterson, Stress Concentration

Factors, Wiley, New York, 1974.
 Assignment

1. Determine the length of a parallel key for a gear to be

mounted on a shaft with a 2.00 in-diameter shaft. The
key is made from AISI 1020 cold-drawn steel. The gear
transmits 21,000 lb-in of torque and has a hub length of
4.00 inch.
2. A V-belt sheave transmits 1,112 lb-in of torque to a 1.75
in-diameter shaft. The sheave is made form ASTM class
20 cast iron and has a hub length of 1.75 in. Design a
parallel key and key seat. The key material is AISI 1020
cold-drawn steel. Create an AutoCAD drawing that
would enable a machinist to make the key seat.
 Splines

Lecture 20

Engineering 473
Machine Design
 Fundamental Problem in
Shaft Design

How do I connect stuff to the shaft?

Interference Fits

Keys & Keyseats Pins Hubs/Collars

Integral Shaft Splines/Polygons

Splines

G Splines can be thought of as

a series of axial keyways
with mating keys machined
onto a shaft.
G There are two major types of
splines used in industry: 1)
straight-sided splines, and 2)
involute splines.

G Splines provide a more

uniform circumferential
transfer of torque to the
shaft than a key.
Mott, Fig. 11-6
 Splined Shaft and Hub

External Spline

Internal Spline
www.advanceadapters.com
 Spline Standards

G ANSI B92.1-1970 (R1982), Involute Splines,

American National Standards Institute.

G ANSI B92.2-1980, Metric Module Involute

Splines, American National Standards Institute.

G SAE Straight Tooth Splines

Straight-Tooth Spline
Geometry
G SAE straight-tooth
splines usually contain
4,6,10, or 16 splines.

G Parameter dimensions
are controlled by the fit
needed for a particular
application.

Mott, Fig. 11-4

 Straight-Tooth Spline Strength

G The torque capacity per unit

length of an SAE spline is
based on a 1,000 psi bearing
stress on the sides.

G Depending on the class of fit, a

spline is able to accommodate
axial movement along the
shaft and still transmit torque.

Splines have the same

failure mechanisms as keys:
1) shear or 2) bearing.
 Straight-Tooth Spline Strength
(Continued)

T = 1,000 ⋅ N ⋅ R ⋅ h

1æD dö D+d
R= ç + ÷=
2è 2 2ø 4

h = (D − d )
1
2

D+d 1
T = 1,000 ⋅ N ⋅ ⋅ (D − d )
4 2
T = Torque per unit length
N = Number of teeth æ D2 − d 2 ö
T = 1,000 ⋅ N ⋅ çç ÷÷
D = Major spline diameter è 8 ø
d = Minor spine diameter
d = f (D)
 Torque Capacity Curves
(SAE Straight-Tooth Splines)

Note that an involute spline has a higher torque

capacity than does a straight-tooth spline of the same
major diameter.
Mott, Fig. 11-7
Involute Splines

Involute splines generally

have a 30o pressure angle.
Mott, Fig. 11-8
 Standard Diametral Pitches
and Lengths
Diametral Pitches

There are seventeen diametral pitches in common use:

2.5 3 4 5 6 8 10
12 16 20 24 32 40 48
64 80 128

Standard Lengths
Common designs use spline lengths of 0.75 D to 1.25 D,
where D is the pitch diameter of the spline. When these
standard lengths are used, the shear strength of the splines
will exceed that of the shaft from which they are made.
 Spline Manufacturing Methods
Splines are either “cut” (machined) or rolled. Rolled
splines are stronger than cut splines due to the cold
working of the metal. Nitriding is common to achieve
very hard surfaces which reduce wear.

Rolled Spline Process

Forged blank is rolled under tons

of pressure prior to heat treating.

The finished spline is more accurate

and stronger (35%) than cut spines.

www.drivetraindirect.com
 Spline Failure Example

Note the yielding of the shaft outside of the engagement area

due to a torsional load. The mating internal spline forced the
external slines to remain parallel. In this case the spline is
stronger than the shaft.
www.4wdonline.com
 Splined Linear Bearing

Circular shaped splines

have been combined with
ball bearings to create linear
bearings that can resist a
torsional load.

www.tsubaki.com
 Polygons
An alternative to splines that has significantly
lower stress concentration is the polygon. Four
and three lobed polygons are shown.

Design information on polygons is

available from General Polygon.

www.generalpolygon.com
 Retaining Rings
G Retaining rings are used on shafts to
maintain the axial position of
components.

G There are many types of retaining rings.

In general, they may be classified as: 1)
internal and 2) external. External

Internal

www.rotorclip.com
Different Types of Retaining
Rings

www.mdmetric.com
 Spring Loaded Retaining Rings
G“Bowed” retaining rings provide restoring
forces to the components being held.

GFlat retaining rings allow small amounts of

axial motion of the held component.

Bowed Internal Bowed External

Retaining Ring Retaining Ring
www.rotorclip.com
 Smalley Compression Spring
Retaining System

Higher restoring forces can be obtained

using compression rings manufactured by
Smalley.
www.smalley.com
Retaining Ring Stress
Concentrations
G External retaining rings used
on shafts require that grooves
be cut into the shaft.

G The grooves generally have

sharp corners or very small
fillet radii which result in
significant stress concentration
factors.

Mott, Fig. 11-5

Retaining Ring Stress
Concentration Factors
G The high stresses at the root of
the retaining ring groove will be
highly localized and will not
significantly effect the static
strength of a shaft made from a
ductile material.

G The stress concentration factors

will be important in determining
the life of the shaft and must be
included in life calculations.

Shigley, Fig. A15-14 & 15

 Retaining Ring Design

Dimensions and design guidelines for retaining rings

are contained in catalogs and literature published by
retaining ring manufacturers.

Rotoclip, Inc. Smalley Waldes Truarc, Inc.

Designs that use retaining rings must take into account

how the rings will be installed and make sure that
sufficient assembly clearance is provided.
Integral Shafts

G An alternative to attaching
components to shafts is to
machine the components
directly onto the shaft.

G This higher priced approach is

often the only approach
available when tight space
constraints exist.

G Complex combinations of
components can be obtained
using modern CNC turning
centers. www.astas.co.za/shafts.html
 Assignment

1) Make a drawing of an SAE straight-tooth- 4-spline

connection having a major diameter of 1.5000 in
and a class A fit. Show all critical dimensions.
What is the torque capacity of the spline?

2) Identify two applications of retaining rings used

in mechanical equipment. Describe the
applications and discuss why you think retaining
rings of the type used were chosen by the
designer.
Involute Gear Tooth Bending
Stress Analysis
Lecture 21

Engineering 473
Machine Design
 Gear Interaction
Base Circle Line of Centers

Line Tangent to
Base Circles

Line Normal to
Line of Centers

Base Circle
 1st Close Up of Meshed Teeth

Line of Contact

H
W

H
W

Line Tangent to Base Circles

2nd Close Up of Meshed Teeth
Line of Contact
Base Circle
H
W

H
W

Base Circle
3rd Close Up of Meshed Teeth
Base Circle Line of Contact

Base Circle
 Line of Action/Pressure Angle
Line of Action – Line tangent to both base circles
Pressure Angle – Angle between the line normal to the line of
centers and the line of action.
Pitch Point – Intersection of the line of centers with the line of
action
Line of Contact
Base Circle
Pitch Point
H
Line of Action
W Pressure
φ Angle
H
W

Base Circle Line of Centers

 Pitch Circle
Pitch Circle – Circle with origin at the gear center and
passing through the pitch point.

Base Circle
Pitch Circle

Base Circle
Pitch Circle
 Relationship Between Pitch
and Base Circles
rb = rcos(φ )
Base Circle
Pitch Circle

rb r
Base Circle
Pitch Circle
 Torque Relationship

Power P
T≡ =
Angular Velocity ω

P(hp) 550 lb ⋅ ft sec 1.0 rev 60 sec 12 in

T= ⋅ ⋅ ⋅ ⋅
n (rev/min) 1.0 hp 2π rad min ft

P
T = 63,000 (lb ⋅ in )
n
 Tooth Load Equations
T H
Wt = Wr = Wt ⋅ tanφ W = Wt cosφ
d/2

Line of Contact
Base Circle
H
Wr W
Line of Action φ Pressure
Angle
H Wt
W

Base Circle Line of Centers

 Gear Tooth Failure
Mechanisms
The primary failure mechanisms for involute gear teeth are:
1) excessive bending stresses at the base of the tooth and, 2)
excessive bearing or contact stress at the point of contact.

Deutschman, Fig. 10-20 Mott, Fig. 9-14

The American Gear Manufacturers Association

(AGMA) has developed standard methods for
addressing both failure mechanisms.
 AGMA Publications
Standand 1010-95, Nomenclature of Gear Tooth Failure
Modes, AGMA, Alexandria, VA, 1995.

Standard 6010-E88, Standard for Spur, Helical, Herringbone,

and Bevel Enclosed Drives, AGMA, Alexandria, VA, 1989.

Standard 2001-C95, Fundamental Rating Factors and

Calculation Methods for Involute Spur and Helical Gear
Teeth, AGMA, Alexandria, VA, 1994.

Standard 908-B89, Geometry Factors for Determining the

Pitting Resistance and Bending Strength of Spur, Helical
and Herringbone Gear Teeth, AGMA, Alexandria, VA 1989.
Lewis Equation

M
σ=
Ic

M = Wt ⋅ L

2
1 t = bt
I c = bt 3
12 2 6

6Wt L
σ=
bt 2
Deutschman, Fig. 10-18
Lewis Equation
(Continued)
6Wt L
σ=
bt 2

Wt 1 æ Wt öæ 1 öæ 1 ö
σ= 2
=ç ÷çç 2 ÷÷çç ÷÷
b t 6L è b øè t 4L øè 4 6 ø

t 2 L
=
x t 2

t2
x=
4L
Lewis Equation
(Continued)
æ Wt öæ 1 öæ 1 ö
σ=ç ÷çç 2 ÷÷çç ÷÷
è b øè t 4L øè 4 6 ø

t2
x= p ≡ circular pitch
4L

æ W öæ 1 öæ 1 öæ p ö
σ = ç t ÷ç ÷çç ÷÷çç ÷÷
è b øè x øè 2 3 øè p ø

2x Lewis Form
y=
3p Factor

Wt
σ=
bpy
 Lewis Equation
(Continued)

2x
y=
3p

Wt
σ=
bpy

P ≡ Diametral Pitch = π p
Y ≡ πy
Most
commonly
Wt P
σ= used form of
bY
Lewis
Y can be determined graphically Equation
or by a computer.
Lewis Form Factor
(Example Values)

Values are for a normal

pressure angle of 20
degrees, full-depth teeth,
and a diametral pitch of
one.

Wt P
σ=
bY

Shigley, Table 14-2

 Limitations of the Lewis
Equation

1. Assumes that maximum bending load occurs at the tip.

Maximum load occurs near the pitch circle when one
tooth carries all of the torque induced load.
2. Considers only bending component of the force acting
on the tooth. The radial force will cause a compressive
stress over the base cross section.
3. Doesn’t consider contact stresses.
4. Assumes that the loads are static.

The AGMA has developed a number of factors to be

used with the Lewis Equation that will lead to an
acceptable design.
 The AGMA Equations
Sat K L
Wt Pd σ all =
σ= ⋅ Ka ⋅ Ks ⋅ Km ⋅ K v KTKR
FJ

Sat ≡ AGMA Allowable

F ≡ face width (b)
Stress Number
K a ≡ Application factor
K L ≡ Life factor
K s ≡ Size factor
K T ≡ Temperature Factor
K m ≡ Load distribution factor
K R ≡ Reliability Factor
K v ≡ Dynamic factor
J ≡ Geometry factor Factors are used to adjust the
stress computed by the Lewis
Pd ≡ Diametral Pitch = N d equation. Factors are also used to
d ≡ Pitch Diameter adjust the strength due to various
N ≡ Number of Teeth environmental conditions.
Wt ≡ Tangential Load Shigley contains tables and charts
for many of these factors.
AGMA Form Factor

Note that the

AGMA Form
Factor will result
in a lower stress
than the Lewis
Equation.

Mott, Fig. 19-5

AGMA Allowable Bending
Stress Numbers

Grade 1 is the
basic or standard
material
classification.

Grade 2 requires
better than normal
microstructure
control.

Mott, Fig. 9-8

 AGMA Dynamic Factor
The AGMA Dynamic Factor is used to correct the bending
stress number for dynamic effects associated with:

1. Inaccuracies in tooth profile, tooth spacing, profile lead,

and run-out,
2. Vibration of the tooth during meshing due to tooth
stiffness;
3. Magnitude of the pitch-line velocity,
4. Dynamic unbalance of the rotating members,
5. Wear and permanent deformation of contacting
surfaces,
6. Shaft misalignment and deflection, and
7. Tooth friction.
Dynamic Factor Chart

Q v ≡ AGMA Quality Number

The AGMA standards contain

tolerances for each quality number.

Pitch Line Velocity = rω

The dynamic factor in

Shigley is equal to the
reciprocal of the dynamic
factor given in this chart.

Mott, Fig. 9-19

 Assignment

1. A spur pinion has a pitch of 6 teeth/in, 22 full-depth

teeth, and a 20 degree pressure angle. This pinion
runs at a speed of 1200 rev/min and transmits 15 hp
to a 60-tooth gear. If the face width is 2 in, estimate
the bending stress.
2. A steel spur pinion has a module of 1.25 mm, 18 full
depth teeth, a pressure angle of 20 degrees, and a
face width of 12 mm. At a speed of 1800 rev/min,
this pinion is expected to carry a steady load of 0.5
kW. Determine the resulting bending stress.
Involute Gear Tooth Contact
Stress Analysis
Class 20

Engineering 473
Machine Design
 Pitting In Gear Teeth
G Pitting – phenomenon in which
small particles are removed from the
surface of the tooth because of the
high contact forces that are present
between mating teeth.

G Pitting is actually the fatigue failure

of the tooth surface.

G Hardness is the primary property of

the gear tooth that provides
resistance to pitting.

www.xteknic.com/products/gearproducts
 Contact Stress Between
F Two Cylinders
Contact Zone
x
2b

d1 p ≡ Contact
L Pressure
y y

z
d2
Contact Force
b
F = 2L ò p(y)dy
0
F=total contact force
z
 Hertz Contact Stress
Equations
Contact Zone Contact Width
2b
p max
b=
( ) (
2F 1 − ν12 E1 + 1 − ν 22 E 2 )
πL 1 d1 + 1 d 2
y
Maximum Contact Pressure
2F
z p max =
πbL

Reference Shigley, Page 73-74

 Hertz Contact Stress
Equations
(Continued)
æ z 2
z ö
σ x = −2ν ⋅ p max ç 1 + 2 − ÷
ç b b ÷ø
è

éæ ö ù Z-axis Stress
êç 1
÷
z 2
z ú Components
σ y = − p max êç 2 − ÷ 1+ −2 ú
ê ç z 2
÷ b 2
bú
1+ 2
êëçè b
÷
ø úû

− p max
σz =
z2
1+ 2
b
Hertz Contact Stress Equations
(Continued)

This graph shows the

variation of the stress
components along
the z-axis.

Note that the

maximum shear
stress is much less
than the maximum
contact pressure.

Shigley, Fig. 2-35

 Hertz Contact Stress Equations
(Continued)

Von Mises stress

variation along the z-axis.

Note that the von Mises

stress is much less than
the maximum contact
pressure.

σ eff = 0.26 ⋅ p max

if N fs = 1.0
Syt then allowable p max ≈ 3.85 ⋅ S yt
= 0.26 ⋅ p max
N fs
Equivalent Contacting Cylinders

Line of Contact
Base Circle

r1
H
W r2

H
W

Base Circle
Equivalent
Equivalent Cylinder # 2
Cylinder # 1
 Radii of Equivalent Cylinders

d psinφ d p ≡ pinion pitch diameter

r1 =
2 d g ≡ gear pitch diameter
φ ≡ pressure angle
d g sinφ r1 ≡ radius of equivalent cylinder #1
r2 = r2 ≡ radius of equivalent cylinder #1
2
 Contact Stress in Gear Teeth

Contact Zone
Contact Width
2b
p max
b=
( ) ( )
2F 1 − ν12 E1 + 1 − ν 22 E 2
πL 1 d1 + 1 d 2

y
Maximum Contact Pressure
2F
z p max =
πbL
 Contact Stress

2F Elastic Coefficient
p max =− 1

πbL é ù 2

ê ú
ê 1 ú
Cp = ê
æ 1 − ν 2p 1 − ν g2 ö ú
b=
( ) ( )
2F 1 − ν12 E1 + 1 − ν 22 E 2 ê πç + ÷ú
ê çè E p E ÷ú
πL 1 d1 + 1 d 2 ë g øû

b=
(
F 1 é 1 − ν12
+
) (
1 − ν 22 ù) 4
ê ú 1 1 é F æ 1 1 öù
1
2
L π ë E1 E2 û + p max = −Cp ê çç + ÷÷ú
r1 r2 ë L è r1 r2 øû
 Contact Stress
(Continued)
1 dg
é F æ 1 1 öù 2 m g ≡ speed ratio = (external gears)
p max = −Cp ê çç + ÷÷ú dp
ë L è r1 r2 øû

1 1 2 æç m g + 1 ö÷
F = Wn = Wt cosφ + =
r1 r2 d p sinφ çè m g ÷ø

1
é Wt æ 1 1 öù 2
1
σ c = −C p ê çç + ÷÷ú é Wt ù 2

ë L cos(φ ) è r1 r2 øû σ c = −C p ê ú
êë d p LI úû
1 1 2 æç 1 1 ö÷
+ = +
r1 r2 sinφ è d p d g ÷ø
ç cosφ ⋅ sinφ m g + 1
I=
2 mg
 Contact Stress Summary
1
é Wt ù 2

σ c = −C p ê ú
êë d p LI úû Elastic Coefficient
1
é ù 2
Form Factor ê ú
cosφ ⋅ sinφ m g + 1 ê 1 ú
I= Cp = ê
2 mg æ 1 − ν 2p 1 − ν g2 ö ú
ê πç + ÷ú
ê çè E p E ÷ú
øû
ë g

dg
mg = (external gears) L ≡ face width
dp
d g ≡ gear pitch diameter Wt ≡ tangential force
d p ≡ pinion pitch diameter
φ ≡ pressure angle
 AGMA Contact Stress Formulas
1
æ Wt C a Cs C m C f ö 2 Sc C L C H
σ c = C p çç ÷÷ σ c,all =
è C v Fd I ø CT C R

σ c ≡ absolute value of contact stress σ c,all ≡ allowable contact stress

C p ≡ elastic coefficient C L ≡ life factor
C a ≡ application factor C H ≡ hardness ratio factor
C v ≡ dynamic factor C T ≡ temperature factor
Cs ≡ size factor C R ≡ reliability factor
d ≡ pitch diameter of pinion
C m ≡ load − distribution factor
C f ≡ surface - condition factor
I ≡ geometry factor
Allowable Contact Stress Values
Hardness Ratio Factor
 Assignment
Web reading assignments
www.xtekinc.com/products/gearproducts
(look under product information)

http://hghouston.com/case012.html
1. A speed reducer has 20 degree full-depth teeth and consists of a
22-tooth steel spur pinion driving a 60-tooth cast-iron gear. The
horsepower transmitted is 15 at a pinion speed of 1200 rev/min.
For a diametral pitch of 6 teeth/in and a face width of 2 in, find
the contact stress.
2. A gearset has a diametral pitch of 5 teeth/in, a 20 degree pressure
angle, and a 24-tooth cast iron spur pinion driving a 48-tooth cast
iron gear. The pinion is to rotate at 50 rev/min. What horsepower
input can be used with this gearset if the contact stress is limited
to 100 kpsi? and F=2.5 in?
Rolling Contact Bearings

Lecture 23

Engineering 473
Machine Design
 Why Rolling Contact
Bearings?

Rolling contact bearings are used to

minimize the friction associated with
relative motion performed under load.

Typical applications
include supporting
shafts.
 Bearing Nomenclature
Retainer
or Cage

Rolling Element
Outside
(Ball)
Diameter Bore
Inner Race

Outer Race

Inner and outer races are typically pressed onto the shaft or
hub with a slight interference fit to make them move with the
shaft (inner race) or remain stationary (outer race).
 Ball Bearings

Thrust
Force

Radial
Force
Radial Ball Angular Ball

Angular ball bearings have higher thrust load capacity

in one direction than due radial ball bearings.

www.torrington.com/products
 Roller Bearings

Radial Cylindrical Radial Tapered

Roller bearings have

higher load capacity
than ball bearings.
Thrust

www.torrington.com/products
 Needle Bearings

Heavy Duty
Drawn Cup

Thrust
Gage
Needle bearings have very high load
ratings and require less space.
www.torrington.com/products
 Other Types of Bearings

Sperical

Polymer
Bearings

www.igus.com/iglide/iglide.htm
Bronze Bushings
 Rolling Contact Bearing
Materials

The space shuttle uses silicon nitride balls

in the oxygen and hydrogen turbopumps.
Mott, Table 14-2
 Static Load Capacity

G The static load rating is the load at which permanent

deformation of a race or ball will occur.

G The bearing is not rotating when this measurement is

made.

G The static load rating is usually designated at C0.

 Bearing Life
Bearings of the same type, size, and material
will exhibit wide variations in life.

Life – number of revolutions (or hours of operation at

design speed) of the inner race that a certain
percentage of the bearings will survive at a known
load.

L10 Life - 10% of the bearings tested at or fail before a rated

number of revolutions of the inner race at the
rated load.
 Statistical Nature of Life
Estimates
10% of the bearings
tested had failed by
18 million cycles.

50% of the bearings

tested had failed by
100+ million cycles.

Typical Weibull Plot of Bearing Fatigue Failures

B.J. Hamrock and W.J. Anderson, Rolling-Element

Bearings, NASA Reference Publication 1105, 1983.
 Load/Life Relationship
k
L 2 æ F1 ö k = 3 for ball bearings
= çç ÷÷
L1 è F2 ø = 3.33 for roller bearings

Basic Dynamic Load Rating

L1 = 1,000,000 revolutions
F1 = C = Basic Dynamic Load Rating

The Basic Dynamic Load Rating is that load which will

cause 10% of a sample of bearings to fail at or before 1
million revolutions. (i.e. 90% of bearings would achieve at
least 1 million revolutions at this load).
 Rated Load/Life
Relationship

k
L 2 æ F1 ö Bearing manufacturers
= çç ÷÷ provide one set of data
L1 è F2 ø relating load and life.

k
æCö This equation is used to
L 2 = çç ÷÷ ×10 Revolutions find the life at different
6

è F2 ø loads.
Example Bearing Data

SKF Catalog
 Outer Race Rotation
Manufacturer’s data is normally based on a rotating
inner race and a stationary outer race.
A rotating outer race and a stationary inner race
will have a lower life.
The Anti-Friction Bearing Manufacturer’s
Association (AFBMA) has developed a standard
equation for computing an equivalent radial load
that takes this into account.

Fe = XVFr

V = 1.0 Rotating Inner Race

V = 1.2 Rotating Outer Race
 Combined Radial and
Thrust Loads
The AFBMA has also developed a standard equation
for computing an equivalent radial load.

Fe = XVFr + YFa
Fr
F α
V = 1.0 Rotating Inner Race
V = 1.2 Rotating Outer Race

Fa X and Y depend on the bearing

geometry and are given in
manufacturers data books
 Variable Loads
Fe,3 , n 3 Fe,4 , n 4
Fe,1 , n1
Fe,2 , n 2

T1 T2 T3 T4
Fe,i ≡ Equivalent radial load for ith event
n i ≡ Speed of the ith event
Ti ≡ Time period of the ith event

1k
é j
k ù
T n (
ê å i i e,i ú
F )
Fe ≡ ê i =1 j ú
ê ú
ê å T n
i i ú
ë i =1 û
 Mounting Bearings
Most manufacturer’s catalogs specify the limiting
dimensions for the shaft and housing bore. These are
generally controlled to within a few thousands of an
inch.

Mott, Fig. 14-13

 Assignment
1. A certain application requires a bearing to last for 1800 h
with a reliability of 90 percent. What should be the rated
life of the bearing?

2. A ball bearing is to be selected to withstand a radial load

of 4 kN and have an L10 life of 1200 h at a speed of 600
rev/min. The bearing maker’s catalog rating sheets are
based on an L10 life of 3800 h at 500 rev/min. What load
should be used to enter the catalog?

3. Read Shigley, Chapter 11

4. Read www.timken.com/bearings/fundamen/
Plane Surface Bearings

Lecture 24

Engineering 473
Machine Design
 Examples of Plain Surface
Bearings
In plain surface bearings the shaft moves relative to the
stationary bearing surface – there is sliding contact versus
rolling contact.

Bronze Bearings

Bronze and Polymer Bearings are Polymer Bearings

two examples of plain surface
bearings. www.igus.com/iglide/iglide.htm
 Journal Bearings
Plain Surface Bearings are often called journal bearings.

Journal bearings
usually employ a
lubricating fluid
between the bearing
and the journal.

Polymer bearings are

often self-lubricating
and do not employ a
lubricant.

Mott, Fig. 16-1

 Lubrication Zones
Boundary Mixed-film Hydrodynamic
Boundary Lubrication Lubrication Lubrication Lubrication

Coefficient of Friction
Contact between
journal and bearing

Mixed-film Lubrication
Intermittent contact

Hydrodynamic Lubrication Bearing Parameter

µn
Journal rides on a fluid Bearing Parameter ≡
film. Film is created by the p
motion of the journal.
µ ≡ dynamic viscosity, lb - sec/in 2
n ≡ rotational speed, rev/sec
p ≡ pressure (force/projected area), psi
 Hydrostatic Bearings
Pressure is used to lift the journal off of the bearing surface.

Mott, Fig. 16-9

Boundary Lubrication - Bearing
Materials

Bronze – copper with tin, lead, zinc, or

aluminum alloying elements

Babbit – lead or tin with copper and

antimony alloying elements

Aluminum

Powdered Metals (Sintered metals)

Polymers (plastics)
 Manufacture of Powdered
Metals

Metal granules are

formed by cooling liquid
metals in jets of water.

www.mpif.org
P/M Compaction Cycle

P/M Design Guide, Metal

Powder Industries Foundation
 P/M Sintering
During the sintering
process,
metallurgical bounds
are made between
the particles at a
temperature less than
the material’s
melting point.

Volatile liquids are Sintering is normally done in an inert

evaporated out during the environment to prevent oxidation of
delubrication stage. the material at the high temperatures.
www.mpif.org
 PV Factor
(Used for Boundary Lubrication Design)

PV ≤ (PV )all.

P ≡ Force/Projected Area [psi]

V ≡ Journal Surface Speed [ft/min]

PV defines the maximum

combination of pressure
and speed that a bearing
material is capable of
withstanding.
 Velocity Calculation

dω
V = rω =
2 V

r
æ rads ö æ rev öæ 2 ⋅ π rads ö
çω ÷ = çn ÷ç ÷
è min ø è min øè rev ø ω

1 æ ft öæ rev öæ 2 ⋅ π rads ö
V = (d in )ç ÷ç n ÷ç ÷
2 è 12 in øè min øè rev ø

V = π⋅d⋅n
12
 Temperature Adjustments

G PV factors are determined at a specific ambient

temperature.

G If an application is at a temperature significantly

different than that at which the PV factor was
determined, a temperature adjustment factor will be
required.

G Consult with the bearing manufacturer to obtain

appropriate values for a specific material.
Sample Manufacturer’s Data
Sample Dimensional Data
 Assignment
Design a plain surface bearing for a radial load of 300
lb on a 1.5 inch diameter shaft rotating at 625 rpm.
Use an L/D ratio of approximately 1.0-1.5. Consider
both bronze and a polymer material for your design
decision. Explain the advantages of the design using
one material over the other. You will need to find
allowable PV data from manufacturer’s data
(recommend looking at www.igus.com).

Review information provided by Metal Powder

Industries Foundation at www.mpif.org.
Hydrodynamic Bearings -
Theory

Lecture 25

Engineering 473
Machine Design
 Lubrication Zones
Boundary Mixed-film Hydrodynamic
Boundary Lubrication Lubrication Lubrication Lubrication

Coefficient of Friction
Contact between
journal and bearing

Mixed-film Lubrication
Intermittent contact

Hydrodynamic Lubrication Bearing Parameter

µn
Journal rides on a fluid Bearing Parameter ≡
film. Film is created by the p
motion of the journal.
µ ≡ dynamic viscosity, lb - sec/in 2
n ≡ rotational speed, rev/sec
p ≡ pressure (force/projected area), psi
 Stable/Unstable Lubrication
Hydrodynamic Lubrication
Boundary
Lubrication
Mixed-film
Lubrication
Hydrodynamic
Lubrication
is often referred to as stable
lubrication.
Coefficient of Friction

If the lubrication temperature

increases, the viscosity
drops. This results in a
lower coefficient of friction,
µn that causes the lubrication
Bearing Parameter
p temperature to drop. => Self
Correcting.

Mixed-film lubrication is unstable – an increase in lubrication

temperature causes further increases in lubrication temperature.
 Newtonian Fluid
A Newtonian fluid is any fluid whose shear stress and
transverse rate of deformation are related through the
equation.
du
τ=µ
dy
 Dynamic Viscosity

du
µ=τ
dy

Units
ips SI
lbf N
lbf − sec m2 N − sec
in 2 = = reyn =
in 2 m m2
in
sec sec
in m

Other common units are discussed in the text.

Pumping Action
When dry, friction will cause the
journal to try to climb bearing
inner wall.

When lubricant is introduced, the

“climbing action” and the viscosity of
the fluid will cause lubricant to be
drawn around the journal creating a
film between the journal and bearing.
The lubricant pressure will push the
journal to the side.
Journal Bearing Nomenclature

β is equal to 2π for a
full bearing

If β is less than 2π, it is

known as a partial
bearing.

We will only be
considering the full
bearing case.
Analysis Assumptions

1. Lubricant is a Newtonian fluid

2. Inertia forces of the lubricant are
negligible
3. Incompressible
4. Constant viscosity
5. Zero pressure gradient along the
length of the bearing
6. The radius of the journal is large
compared to the film thickness
 Analysis Geometry

Actual Geometry Unrolled Geometry

From boundary layer theory, the pressure

gradient in the y direction is constant.
 X-Momentum Equation

æ dp ö æ ∂τ ö
å Fx = 0 = çè p + dx dx ÷ødydz + τdxdz − çç τ + ∂y dy ÷÷dxdz − pdydz
è ø

dp ∂τ ∂u dp ∂ 2u
= τ=µ =µ 2
dx ∂y ∂y dx ∂y
 X-Momentum Equation
(Continued)
X-Momentum Eq.
dp ∂ 2u
=µ 2
dx ∂y

General Solution

∂ 2 u 1 dp
=
∂y 2
µ dx
∂u 1 dp Boundary Conditions
= y + C1 (x )
∂y µ dx y = 0, u = 0
1 dp 2 y = h(x), u = -U
u= y + C1 (x )y + C2 (x )
2µ dx
 X-Momentum Equation
(Continued)

1 dp 2
u= y + C1 (x )y + C2 (x )
2µ dx

y = 0, u = 0 C2 (x ) = 0

U h (x ) dp
y = h(x), u = -U C1 (x ) = − −
h (x ) 2µ dx

u=
1 dp 2
2µ dx
(
y − h (x )y −
U
)
h (x )
y
Note that h(x) and
dp/dx are not known
at this point.
 Mass Flow Rate

h (x )
 = ρ ò udy
m
0
h (x )
 =ρ
m ò
æ 1 dp 2
çç ( )
y − h (x )y −
U ö
y ÷÷dy
h (x ) ø
0 è 2µ dx
é h (x )3 dp Uh (x )ù
 = ρ ê−
m − ú
ë 12µ dx 2 û
 Conservation of Mass
é h (x )3 dp Uh (x ) ù
 = ρ ê−
m − ú
ë 12µ dx 2 û

Conservation of Mass Requires


dm d h (x ) dp ö U dh
æ 3
=0 − çç ÷− =0
dx ÷
dx è 12µ dx ø 2 dx

d h (x ) dp ö
æ 3
dh
ç ÷ = −6 U
dx è µ dx ÷ø
ç dx Reynold’s Equation
 h(x) Relationship

cr = radial clearance
θ
e
ε=
cr
h (θ ) = c r (1 + ε ⋅ cos θ )

e h min = c r (1 − ε )
h max = c r (1 + ε )

æ æ 2x ö ö
h (x ) = c r çç1 + ε ⋅ cos ç ÷ ÷÷
è è D øø
 Sommerfeld Solution
A. Sommerfeld solved these
d h (x ) dp ö
æ equations in 1904 to find the
3
dh
ç ÷ = −6 U pressure distribution around the
dx è µ dx ÷ø
ç dx bearing.

æ æ 2x ö ö It is known as a “long bearing”

h (x ) = c r çç1 + ε ⋅ cos ç ÷ ÷÷ solution because there is no flow
è è D øø in the axial direction.

µUr é 6ε ⋅ sin θ ⋅ (2 + ε cos θ )ù

p= 2 ê + po 0 ≤ θ ≤ π
( )
c r ë 2 + ε (1 + ε cos θ ) û
2 2 ú

r is the journal radius, ε is a chosen design parameter.

 Ocvirk Short-Bearing Solution
A “short bearing” allows lubricant flow in the longitudinal
direction, z, as well as in the circumferential direction, x.

∂ æ h (x ) dp ö ∂ æ h (x ) dp ö
3 3
∂h Governing
ç ÷− ç ÷ = −6U
∂x çè µ dx ÷ø ∂z çè µ dz ÷ø ∂x Equation

The Ocvirk solution (1955) neglects the first term

as being small compared to the axial flow.

µU æ l2 2ö 3ε ⋅ sinθ
p = 2 çç − z ÷÷ 0≤θ≤π
rcr è 4 ø (1 + ε ⋅ cosθ )3
Short-Bearing Pressure Distributions

Norton Fig. 10-8 & 10-9

 Short & Long Bearing
Comparisons

%
 Assignment

Use Matlab to plot the pressure distribution predicted by the

Sommerfeld equation for a journal bearing having a
clearance ratio of 0.0017, journal radius of 0.75 in, ε of 0.6,
µ=2.2µreyn, shaft rotational speed=20 rev/sec, and po=o.

First, generate the plot only for the range θ equals 0 to π.

Second, generate the plot for the range θ equals 0 to 2π.

What happens to the pressure distribution from π to 2π. Is
this physically possible? Discuss what would happen to the
lubricant if this pressure distribution occurred.
Hydrodynamic Bearings -
Design

Lecture 26

Engineering 473
Machine Design
 Governing Equation
In the previous lecture, the momentum and continuity
equations were used to develop the following equation

∂ æ h (x ) dp ö dh (x )
3
ç ÷ = −6U
∂x çè µ dx ÷ø dx

This equation was generalized to include lubricant flow

in both the circumferential and longitudinal directions
æ ( )3
∂ h x dp
ç
ö ( )
∂ h x dp ö
÷− ç
æ 3
÷ = −6U
dh (x )
∂x è µ dx ø ∂z è µ dz ÷ø
ç ÷ ç dx

This equation is generally solved using

specialty computer programs.
 Design Variables

Independent Variables Dependent Variables

Viscosity, µ Friction, f
Load, P (W/projected Area) Temperature rise, ∆T
Speed, N Volumetric flow rate, Q
Dimensions r, c, β, and L Minimum film thickness, ho

The objective of the design engineer is to select the

independent variables necessary to achieve desired
performance criteria. The dependent variables will
be dictated by the selections made for the
independent variables.
 Graphical Data
Raimondi and Boyd (1958) did extensive numerical studies
on the relationships between the various parameters that
govern the design of fluid film bearings and published
graphical data to facilitate the design of such bearings.

The charts presented in this lecture are for long bearings

with β=360o (full bearings).

Albert Raimondi and John Boyd, “A Solution for the Finite

Journal Bearing and Its Application to Analysis and Design,
Parts I, II, and III,” Transactions of American Society of
Lubrication Engineers, Vol. 1, No. 1, in Lubrication Science
and Technology, Pergamon, New York, 1958, 159-209.
Viscosity-Temperature Charts
The viscosity of lubricants
used in fluid-film bearings
are very temperature
dependent.

As work is done on the fluid

as it moves through the
bearing it heats up.

The viscosity used in the

design/analysis of a fluid-
film bearing should be based
on the average temperature.
Shigley, Fig. 12-11
Average Temperature
Tout = Tin + ∆T

Tin + Tout
Tave =
2

∆T
Tave = Tin +
2
An initial ∆T is estimated
at the start of an analysis.

Iteration will be required

based on the actual ∆T.
Shigley, Fig. 12-11
 Sommerfeld Number
The Sommerfeld Number is used extensively
in journal bearing design.

r journal radius
2 c clearance
æ r ö µN
S=ç ÷ µ dynamic viscosity
ècø P N rotational speed (rev/sec)
P bearing load/projected area

Note that consistent units must be used. The

unit for the Sommerfeld number is Rev.
A. Sommerfeld, “Zur Hdrodynamischen Theorie der Schmiermittel-
Reibung”, Z. Math. Physik, vol. 50, 1904, pp 97-155.
 Example Problem
Given design parameters:

SAE 30 Oil
Tin = 150 oF (oil inlet temperature)
N = 30 rev/sec (journal rotational speed)
W = 500 lb (total load acting on bearing)
r = 0.75 in (journal radius)
c = 0.0015 in (clearance between journal and bearing)
L = 1.50 in (length of bearing)

Use the Raimondi-Boyd charts to determine the

bearing performance parameters.
 Step 1 – Estimate Average
Lubricant Temperature
Assume a temperature rise in the oil of 34oF.

∆T
Tave = Tin +
2

34°F
Tave = 150°F + = 167°F
2
Step 2 – Find Average Viscosity

µ = 2.2 µreyn

Shigley, Fig. 12-11

 Step 3 – Compute Force/Projected
Area (P)

W 500 lb
P= =
2 r L 2 ⋅ 0.75 in ⋅ 1.5 in

P = 222 lb/in 2
Step 4 – Compute L/D and Bearing
Characteristic Number

1.5 in
L/D = = 1.0
2(0.75 in)

2
ærö æ µN ö
S=ç ÷ ç ÷
ècø è P ø
æ 0.75 in ö
=ç ÷
2
(
çç
)
æ 2.2x10−6 lb − sec/in 2 (30 rev/sec) ö
÷÷
2
è 0.0015 in ø è 222 lb/in ø
S = 0.0743
 Step 5 – Find Minimum Film
Thickness and Eccentricity
h0
= 0.28
c

e
ε= = 0.7
c

Contact Light Load

Shigley, Fig. 12-14
Step 5 – Continued

h0
= 0.28
c

=> h 0 = 0.28(0.0015 in ) = 0.00042 in

e
ε= = 0.72
c

=> e = 0.72(0.0015 in ) = 0.00108 in

Step 6 – Find Position of Minimum
Film Thickness

φ = 44 degrees
 Step 7 – Find the Maximum Film
Pressure

P
= 0.36
Pmax

Pmax = 222/0.36 = 617 psi

Step 8 – Find Location of Maximum
Pressure θ = 18°
p max

θ po = 65°

Shigley Fig 12-21

Step 9 – Find Coefficient of Friction

r
f = 2.5
c

æ 0.0015 in ö
f = 2.5ç ÷
è 0.75 in ø
f = 0.005
 Step 10 – Find Horsepower Required
to Overcome Friction
W

T = f ⋅W⋅r ff = f ⋅ W T = 0.005 ⋅ 500 lb ⋅ 0.75 in

T = 1.88 in - lb
Pwr(hp)
T (in - lb ) = 63,000
N (rev/min )
Pwr =
(1.88 )(30 ⋅ 60 )
T (in − lb ) ⋅ N (rev/min )
63,000
Pwr(hp) = Pwr = 0.054 hp
63,000
Step 11 – Find the Lubricant Flow
Rate
Q
= 3.5
r ⋅c⋅ N⋅L

Q = 3.5 ⋅ r ⋅ c ⋅ N ⋅ L
Q = 3.5 (0.75 in )(0.0015 in )
(30 rev/sec)(1.5 in )
3
Q = 0.177 in
sec
Step 12 – Find Side Flow Leakage
Qs
= 0.8
Q

3
Qs = 0.8 ⋅ 0.177 in
sec

3
Qs = 0.142 in
sec
Step 13 – Find Temperature Rise of
Lubricant
It is assumed that all of the frictional energy is
converted to heat and carried away by the lubricant.
 =Q
W 
f
Tf = f ⋅ W ⋅ r
 = T ⋅ω
W f f
ω = 2π ⋅ N
 =m
Q  c ∆Tp  = ρ⋅Q
m
2π ⋅ f ⋅ W ⋅ r ⋅ N
Tf ⋅ ω ∆T =
∆T = ρ ⋅ Q ⋅c p
m cp
 Temperature Rise
(Continued)
2π ⋅ f ⋅ W ⋅ r ⋅ N in 3
∆T = Q = 0.177
ρ ⋅ Q ⋅cp sec
9,338lbf ⋅ in = BTU
f = 0.005
W = 500 lbf
∆T = 16.4 °F
r = 0.75 in
N = 30 rev/sec Note that a temperature rise
of 34 oF was assumed when
c = 0.42 BTU the average temperature was
p lbf ⋅ °R
estimated.
γ = 0.86
æ lb öæ ft 3
ö The analysis needs to be
ρ H 2O = ç 62.4 f
3 ֍
ç ÷
3÷ repeated with an improved
è ft øè 1728 in ø
estimate for Tave.
 Assignment

A journal bearing has a diameter of 3 in and is 1.5 in

long; it supports a load of 800 lbf. The journal speed is
600 rev/min and the radial clearance is 0.0025 in. Find
the minimum oil-film thickness and the maximum film
pressure for both SAE 10 and SAE 40 lubricants if the
operating temperature is 150 oF.

Discuss why one has a larger film thickness than the

other.
 Mechanical Fasteners –
Tensile and Shear Stress Areas

Lecture 28

Engineering 473
Machine Design
 Threaded Fasteners

Bolt – Threaded fastener designed to

pass through holes in mating members
and to be secured by tightening a nut
from the end opposite the head of the
bolt.

Screw – Threaded fastener designed to

be inserted through a hole in one
member and into a threaded hole in a
mating member.
Mott, Fig. 18-1
Bolts

Mott, Fig. 18-2

Machine Screws

Mott, Fig. 18-3

 Sheet Metal and
Lag Screws

Sheet metal screws are often self-tapping.

Mott, Fig. 18-4

 Set Screws

Set screws are used to develop a normal force

between two objects (e.g. collar and shaft).

Mott, Fig. 18-5

 Thread Standards
(Inch Series)
American Standard B1.1-1949

First American standard to cover the Unified Thread

Series agreed upon by the United Kingdom, Canada,
and the United States. Represents the basic American
standard for fastener threads. Threads made to this
standard are called “unified threads”.

ANSI B1.1-1989/ASME B1.1-1989

Revised standard that still incorporates much of the
original standard.
 Thread Standards
(Metric Series)

ANSI B1.13M-1983 (R1989)

Contains system of metric threads for general
fastening purposes in mechanisms and structures.
Fasteners made to this standard are often referred
to as M-series.
 Thread Profiles

The pitch line or diameter is located at ½ the height of the

theoretical sharp v-thread profile.
 Thread Series
Thread Series – groups of diameter-pitch combinations
distinguished from each other by the number of threads per inch
applied to a specific diameter.
Unified Coarse-Thread Series (UNC or UNRC)
Most commonly used in the bulk production of bolts,
screws, nuts for general engineering applications.

Unified Fine-Thread Series (UNF or UNRF)

Use when more threads per inch are required (i.e.
where are short length of engagement is available).
M-Series
Metric system of diameters, pitches, and
tolerance/allowances.
 Thread Classes

Thread Classes – Define the amount of tolerance

and allowance associated with a particular thread.

Classes 1A, 2A, 3A – apply to external threads.

Class 2A is the most commonly used.

Classes 1B, 2B, 3B – apply to internal

threads. Class 2B is the most commonly used.
 Thread Designations
(Inch Series)
The following is an example of the standard method used to
designate bolt and screw thread requirements on a drawing or
in a specification.

Threads per inch

Thread Class

1 4 − 20UNC − 2A
External Thread

Nominal Size Thread Series

 Thread Designations
(Metric Series)
The following is an example of the standard method used to
designate bolt and screw thread requirements on a drawing or
in a specification.
Pitch (mm)
Nominal Diameter Lowercase=> external
thread
M6x1− 4g6g(22)
Tolerance Classification
Metric Series
Material and Strength
Designations

Mott, Table 18-1

Material and Strength
Designations
(Continued)

Mott, Table 18-2

Material and Strength
Designations

Mott, Table 18-3

 Tensile Stress Area
The average axial stress in a fastener is
computed using a “tensile stress area”.

F F ≡ Axial Force
σ ave =
At D r ≡ Root Diameter
2 D p ≡ Pitch Diameter
π é Dr + Dp ù
At = ê ú A t ≡ Tensile Stress Area
4ë 2 û
σ ave ≡ Average axial stress

Tests of threaded rods have shown that an unthreaded rod having a

diameter equal to the mean of the pitch diameter and the minor diameter
will have the same tensile strength as the threaded rod.
 Tensile Stress Area
(Continued)

D t ≡ diameter at critical plane

æ3 3 ö
D t = d b − 2ç H + H ÷ d b ≡ diameter of bolt
è8 16 ø
1 H ≡ theoretical height of thread
H= ⋅ tan (60°) n ≡ 1 p = threads/in
2n
Matt Scolforo, Sverdrup Technology
 Tensile Stress Area
(Continued)
π 2
æ3 3 ö At = ⋅ Dt
D t = d b − 2ç H + H ÷ 4
è8 16 ø

2
1 πæ 0.9743 ö
H= ⋅ tan (60°) At = ç db − ÷
2n 4è n ø

tan (60°) æ 3 3 ö
Dt = db − ç + ÷
n è 8 16 ø
This is the formula used by
manufacturers of inch series
9 3 fasteners to publish the tensile
Dt = db −
16n area in their catalogs.
 Tensile Stress Area
(Continued)

The following formula may be obtained in

a similar manner for metric series threads.

2
πæ 0.9328 ö
At = ç db − ÷
4è n ø
 Shear Area of External
Thread
Consideration of the interaction between mating threads must
be considered to establish the shear area of an external thread.

Matt Scolforo, Sverdrup Technology

 Shear Area of External Threads
(Continued)

A s,e = π ⋅ K n,max ⋅ t e ⋅ n A s,e ≡ shear area of external thread

K n,max ≡ maximum minor diameter of internal thread
0.5t e t e ≡ thickness of external thread at critical shear plane
tan (30°) =
0.75H − gap n ≡ threads per inch
 Shear Area of External
Threads
(Continued)
A s,e = π ⋅ K n,max ⋅ t e ⋅ n 1æ 1 3 ö
gap = çç K n,max + − E s,min ÷÷
2è 2 2n ø
0.5t e
tan (30°) =
Es,min=minimum pitch diameter of the external thread

0.75H − gap 1 é3 3 1 æ 1 3 öù
te = 2 ⋅ ê − çç K n,max + − E s,min ÷÷ú
3 êë 4 2n 2 è 2 2n øúû
1 3
H= tan (60°) =
2n 2n te =
1
+
1
(E s,min − K n,max )
2n 3

æ 3 ö
t e = 2 ⋅ tan (30°)çç 0.75 − gap ÷÷ The gap equation is based on
è 2n ø tolerance data.
 Shear Area of External
Threads
(Continued)

é1 1 ù
A s,e = π ⋅ n ⋅ K n,max ⋅ ê + (E s,min − K n,max )ú
ë 2n 3 û
This equation appears in the ANSI standards and gives
the shear area per unit length of engagement. It must be
multiplied by the length of engagement, Le, to obtain the
actual shear area. This area is often reported in
manufacturers data sheets for bolts and screws.

é1 1 ù
A s,e = π ⋅ n ⋅ K n,max ⋅ ê + (E s,min − K n,max )ú ⋅ Le
ë 2n 3 û
 Shear Area of Internal
Threads

Ds,min ≡ Minimum major diameter (external thread)

A s,i = π ⋅ D s,min ⋅ t i ⋅ n
t i ≡ thickness of internal thread (critical plane)

Matt Scolforo, Sverdrup Technology

 Shear Area of Internal
Threads
(Continued)

Similar to the previous derivation, an equation that takes

into account the tolerances of the thread system can be
derived to compute the shear area of the internal thread.

é1 ù
A s,i = π ⋅ D s,min ⋅ n ⋅ ê +
1
(Ds,min − E n,max )ú
ë 2n 3 û

E n,max ≡ maximum pitch diameter of the internal threads

 Length of Engagement
(Equal Strength Materials)

If the internal thread and external thread material have

the same strength, then

Tensile Strength Shear Strength

(External Thread) (Internal Thread)
Fmax Fmax
St = 0.5St =
At A s,i ⋅ L e

Fmax = St A t = 0.5St A s,i L e

2A t
Le =
A s,i
 Length of Engagement
(Unequal Strength Materials)

If the internal thread and external thread do not have

the same material, then

Tensile Strength Shear Strength

(External Thread) (Internal Thread)
Fmax Fmax
St,e = 0.5St,i =
At A s,i ⋅ L e

Fmax = St,e A t = 0.5St,i A s,i L e

2A t ⋅ St,e
Le =
A s,i ⋅ St,i
 Bolt/Nut Design Philosophy

ANSI standard bolts and nuts of equal grades are

designed to have the bolt fail before the threads
in the nut are stripped.

The engineer designing a machine element is

responsible for determining how something should
fail taking into account the safety of the operators
and public. Length of engagement is an important
consideration in designing machine elements with
machine screws.
 Assignment
A 5/16-18UNC-2A fastener is made from a material having a
yield strength of 120 ksi. The fastener will be engaged with a
nut made from the same material. Compute the tensile stress
area, shear stress area per length of engagement, and
minimum length of engagement. Dimensional information on
the threads is given below.

The minimum pitch diameter of the external thread is 0.2712

in., and the maximum minor diameter of the internal thread is
0.265 inch, minimum major diameter of the external thread is
0.3026 in, and the maximum pitch diameter of the internal
threads is 0.2817 (reference Table 4, page1544, Machinery’s
Handbook).
Mechanical Fasteners –
Tension Connections
Lecture 29

Engineering 473
Machine Design
Tension Connection
Ø A threaded fastener
connection has clearance gaps
that are used to assemble the
connection.

Ø A connection can be loaded in

either tension/compression or
shear.

Ø Because of the clearance gaps,

dowel pins are often used for
accurately positioning of
mating parts.
Shigley, Fig. 8-12
 Proof Load & Proof Stress

Proof Load : Maximum force that a fastener can withstand

without acquiring a permanent set.

Stress Proof Load divided by the tensile-stress-area

Proof Stress:
of the fastener.

The proof stress is slightly less than the yield stress

of the material due to stress concentrations in the
threads.
Bolt Preload
Ø Purpose of bolt is to clamp
two parts together.

Ø Turning the nut on the

threads will stretch the bolt

grip
to create a clamping force.

Ø The clamping force will

impart a compressive force
on the mating parts.

Ø The clamping force is called

preload or initial tension.
Shigley, Fig. 8-12
 Bolt Stiffness
δ = δd + δ t

Fi
δd =
kd
Ld kd
grip

Fi
δt =
kt
Lt kt
kdk t
Fi = δ
kd + kt

Fi = k b δ
Clamped Part Stiffness

Fi = k m δ

L1
1 1 1 1
= + +
L2 k m k1 k 2 k 3
L3
k1 = A1E L1
k 2 = A2E L2
k 3 = A 3E L3
Pressure-Cone Equations

Reference Shigley, page

339-340 for derivation
0.577π ⋅ Ed
k=
æ (1.15t + D − d )(D + d ) ö
lnçç ÷÷
è (1.15t + D + d )(D − d ) ø

Shigley, Fig. 8-14

External Load Interaction –
Preloaded Connection
Pb Pm
δ= = Fb = Pb + Fi
kb km

k bP
æ kb ö Fb = + Fi
Pb = Pm çç ÷÷ kb + km
è km ø
k mP
P = Pb + Pm Fm = − Fi
kb + km

k bP Fm < 0
Pb =
kb + km
Shigley, Fig. 8-12
 External Load Interaction –
Preloaded Connection
kb
Let C=
kb + km

Fb = CP + Fi

Fm = (1 − C )P − Fi

C is the ratio of the bolt stiffness to the total stiffness

(bolt plus clamped material).
 Sample Data
Bolt: ½-13 UNC-2A kb
C=
Material: Steel kb + km

Mlb/in
Bolt Grip
(in) kb kp C 1-C
2 2.57 12.69 0.168 0.832
3 1.79 11.33 0.136 0.864
4 1.37 10.63 0.114 0.886

ØMost of the external load, P, is carried by the

mating parts.

ØMaking the grip longer decreases the bolt’s

share of the the load.
 Bolt Force – Displacement
Diagram

When the mating

Fb
kb parts and bolt are
made from the same
Fm=0 material, an external
Fi kb load will not cause
kb + km an appreciable
kb increase in the bolt’s
total force until
separation of the
mating parts occurs.
δ
 Recommended Preload

Fp = Sp ⋅ A t Fp – Bolt proof load

Sp – Bolt proof strength
Disassembly Required At – Tensile stress area
Fi = 0.75 ⋅ Fp Fi – Bolt preload

Permanent Connection
These recommendations assume
Fi = 0.90 ⋅ Fp that the nut or threaded part has
sufficient shear strength in the
threads (i.e. steel bolts and steel
nuts).
 Bolt Failure Criteria
Fb = CP + Fi n ≡ Load Factor
Sp ≡ Proof Stress
kb
C= Sp A t − Fi
kb + km n=
CP

CP Fi
σb = +
At At

CnP Fi
Sp = +
At At
 Connection Separation
Criteria
Fm = (1 − C )P − Fi Let Po = nP, where
n is a load factor, then
At separation, Fm = 0
Fi
n=
0 = (1 − C )Po − Fi P(1 − C )

Fi It is necessary to check both bolt

Po =
1− C stress and connection separation
when designing a tension
Po ≡ External load that connection. Separation often
will cause separation occurs first.
 Assignment
In the figure, the bolts have a diameter of ½ inch and the cover
plate is steel, with D=1/2 inch. The cylinder is cast iron, with
E=5/8 inch and a modulus of elasticity of 18 Mpsi. The ½ inch
SAE washer to be used under the nut has OD=1.062 inch and is
0.095 inch thick. Find the stiffness of the bolt and the mating
members and the joint constant.
 Assignment
(Continued)

In addition to the information given in problem 1, the

dimensions of the cylinder are A=3.5 inch and an effective
seal diameter of 4.25 inch. The internal static pressure is
1,500 psi. The outside diameter of the head is C = 8 inch.
The diameter of the bolt circle is B = 6 inch, and a bolt
spacing in the range of 3 to 5 bolt diameters would require
8 to 13 bolts. Select ten SAE grade 5 bolts and find the
resulting load factor n.
 Mechanical Fasteners –
Torque Vs Preload Relationship
Lecture 30

Engineering 473
Machine Design
Tension Connection
Ø A threaded fastener connection
has clearance gaps that are used
to facilitate assembly of the
connection.

Ø A connection can be loaded in

either tension/compression or
shear.

Ø Because of the clearance gaps,

dowel pins are often used for
accurately positioning of mating
parts and prevent sliding motion.
Shigley, Fig. 8-12
 How Much Torque to Achieve
Preload Requirement?

Fb
kb

Fm=0
Fi kb
kb + km
kb

δ
In the previous lecture, it was shown that a high
preload is a very desirable in a tension connection.
 Torque-Preload Relationship

Square Tooth Modified for

Power Screw Thread Angle
Equation Alpha

FD p æ l + µπD p ö FDp æ l + µπDpsecα ö

Tu = ç ÷ T= ç ÷
2 çè πD p − µl ÷ø 2 çè πD p − µl ⋅ secα ÷ø

These equations give the torque α

required to impart an axial force
and overcome thread friction.
 Torque-Preload Relationship
(Continued)

Fi D p æ l + µπD psecα ö Fi µ n d n
Nut Surface T= ç ÷+
2 çè πD p − µl ⋅ secα ÷ø 2
Friction Torque

Fµ d d n ≈ 1.5D p (Standard Washer)

Tn = n n
2
T = kFi d
µ n = Coefficient of Friction
between nut and part éæ D p öæ tanλ + µ ⋅ secα ö ù
k = êçç ÷÷çç ÷÷ + 0.625µ c ú
d n ≡ mean annulus diameter ëè 2d øè 1 − µtanλ ⋅ secα ø û
of nut
l
tanλ =
πD p
 Experimental Data
Given Unlubricated
T=90 N-m Ave. Fi=34.3 kN
Std. Dev. = 4.91 kN
Measure 2σ
Preload, Fi = ±29%
Fm
Bolt Lubricated
M12x1.25 Ave. Fi=34.2 kN
Std. Dev. = 2.9 kN
There is considerable 2σ
= ±17%
scatter in torque-versus Fm
preload data.

J.C. Blake and H.J. Kurtz,”Uncertainties in Measuring Fastener Preload,”

Machine Design, Vol. 37, Sept. 30, 1965, pp. 128-131.
 Typical Values of K

Bolt Condition k
Black oxide finish 0.3
Zinc-plated 0.2
Lubricated 0.18

Bolt manufacturers often list recommended

k values with their product data.
 Maximum Torque Values

It is very easy to twist off a small diameter fastener ( < 5/16

inch) when preloading a connection.

Design organizations often establish maximum torque

values than can be applied to a fastener during installation.
 Assignment

An initial preload of 50 ksi is needed in a ½-13UNC-

2A steel fastener. The coefficient of friction for the
threads is estimated to be 0.4, and the coefficient of
friction between the bolt head and part is estimated to
be 0.3. What torque should you specify on the
drawing to assure that the fastener is installed with the
correct preload?
Shear Connections

Lecture 31

Engineering 473
Machine Design
 Fasteners Loaded
in Shear
Connections may be designed to resist tension,
shear, or combined tension and shear loads.

This figure shows a lap connection with a single rivet loaded

in shear. The analysis of shear connections containing rivets
and threaded fasteners is similar.
 Bending Loads in Lap
Connections
The offset in a lap connection creates a moment equal to
approximately M=Ft/2. This bending moment can cause
complex deformations and stresses in the connection that
affect the overall connection strength.

In most cases this offset moment is neglected and a

suitable factor of safety is used.

Deutschman, Fig. 16-3

 Fastener Failure in Shear
Shear Plane

Contact
Area F = ò σ c dA τ ave

Stress Formula τ ave ≡ average shear stress in fastener

F A s ≡ shear area of fastener
τ ave =
As F ≡ shear force
Static Failure N fs ≡ safety factor
Criteria Sall ≡ allowable shear stress
Sall If no threads are in the shear
τave ≤
plane, then A s = πd 4
2
N fs
Deutschman, Fig. 16-3
 Shear Planes
Single Shear Plane
F
τ ave =
As
Shear Plane
Double Shear Plane
Shear Plane
F
τ ave =
2 ⋅ As
Shear Plane
Number of shear planes
Deutschman, Fig. 16-4, 16-5
 Multiple Fasteners

F
τ ave = n ≡ number of fasteners
n ⋅ As
It is standard practice to assume that each fastener
carries an equal amount of the total load.
Deutschman, Fig. 16-10
 Friction

Ø Significant friction forces can exist between mating

parts in a connection if a high preload is used in the
fasteners.

Ø This friction force is normally not accounted for in

the shear connection design calculations unless
measures are taken to ensure that the preload is
maintained over the life of the connection.

Ø Vibration, local yielding, and relaxation in the

fasteners can cause a decrease in preload over time.
 Tensile Failure of
Connected Plates
Stress Formula
F
σ ave =
(w − d )t

t Static Failure
ØThe stress formula does not include stress Criteria
concentration factors, and local yielding S yt
may occur around the holes. Stress σ ave ≤
concentration factors are required if fatigue N fs
is a concern.
ØThe total net area is used if multiple
fasteners are used in the connection. Deutschman, Fig. 16-6
 Bearing Failure of Plate

Stress Formula
F
σb =
dt

Static Failure
Criteria
ØThe bearing stress is based on the projected 0.9S yt
bolt area normal to the load direction. σb ≤
N fs
ØThe triaxial nature of the actual stress in the
material (Hertz Contact Stress) results in high
allowable values for the contact stress. Deutschman, Fig. 16-7
 Shear Tear Out at Edge
of Plate

Stress Formula
F
τt =
2at

Prevention
This type of failure is normally controlled by specifying
minimum edge distances. An edge distance greater than
1.5d is commonly used.
Deutschman, Fig. 16-8
 Eccentric Loads

Load Lying in Load Offset from

Shear Plane Shear Plane

Mott, Fig. 20.4 Deutschman, Fig. 16-17

 Load Lying in Shear
Plane
M=Pa P

Equivalent Loads in the Shear Plane

System With
Eccentrically Loaded
Connection
 Load Lying In Shear Plane
(Continued)

F4
The force resisted by each
F1 fastener is assumed to be
d1
c.g. of bolt d 4 M=Pa inversely proportional to
d2 d F3 the distance from the bolt
group 3
group c.g.
F2

Pa = F1d1 + F2 d 2 + F3d 3 + F4 d 4 d2 d3
F2 = F1 F3 = F1
d1 d1
F1 F2 F3 F4 d4
= = = F4 = F1
d1 d 2 d 3 d 4 d1
 Load Lying in Shear Plane
(Continued)

F4
Pad i
F1 Fi = n
d1
c.g. of bolt d 4 M=Pa å j
d
j=1
2

d2 d F3
group 3

Force resisted
F2 by the ith bolt.

æ d1 ö æ d2 ö æ d3 ö æ d4 ö
Pa = F1d1 çç ÷÷ + F1d 2 çç ÷÷ + F1d 3 çç ÷÷ + F1d 4 çç ÷÷
è d1 ø è d1 ø è d1 ø è d1 ø
F
( )
Pa = 1 d12 + d 22 + d 32 + d 24
d1
 Load Lying in Shear Plane
(Continued)
P Maximum Resultant
M=Pa

Since all of the fasteners generally have the same

diameter it is only necessary to find the maximum
resultant for use in the strength calculations.
 Load Lying in Shear
Plane
πd 2
C.G. of Bolt Group Ai =
4
y n
A1 A4 åx A i i
x= i =1

(x 4 , y 4 )
n
(x 1 , y 1 ) åA i
i =1
A2 A3
1 n
x = å xi
(x 2 , y 2 ) (x 3 , y 3 ) n i =1

1 n
x y = å yi
n i =1
Assignment

Choose a threaded
fastener that will support
the eccentric load lying
in the shear plane of the
connection. Use a safety
factor of 2.0. Determine
the minimum edge
distances and plate
thickness based on the
fastener design criteria.

Mott, Fig. 20-5

Connections Loaded in Shear
and Tension
Lecture 32

Engineering 473
Machine Design
 Eccentric Loads

Load Lying in Load Offset from

Shear Plane Shear Plane

Mott, Fig. 20.4 Deutschman, Fig. 16-17

 Prying Forces
æbö
f b = ç ÷Pj
èaø

f a = f b − Pj

ØPrying in a connection
fa fb increases the bolt force.

a ØPrying is a joint
b separation phenomena.
Exaggerated Deformation Free-Body Diagram
in Joint
 Finite Element Analysis
Showing Joint Deformation

http://river.tay.ac.uk
 Separation Example
y
(x1 , y1 ) P
P A1 A4
a
A2 A5

A3 A6

x
Preload is not sufficient to prevent separation.
 Connections With
Separation
n
M = Pa = å f i d i
P i =1

a f1 f 2 f i
= =
d1 d 2 d i Md1
f1 = n

(d i ) f å i
d 2

i =1
fi = 1
d1
di
f1 n 2
M = å di
d1 i =1

d i ≡ distance from pry point to ith fastener

 Connections With
Separation
(Continued)

P Md i
fi =
a I

n
I = å d 2j
j=1

di
Note that assumptions about the stiffness
of the connection plate are required to
estimate the location of the pry point.
 Connections With
Significant Preload

c.g. of
fastener
group

Compressive Mating Force Proportionality Point

Force Due to Preload (AISC Standard Practice)
 Connections With
Significant Preload
(Continued)

Bolt Force f i,ext ≡ axial force in ith fastener due

to external moment.
Mc i c i ≡ distance from c.g. to ith fastener
f i,ext =
I I ≡ moment of inertia of fastener group
M ≡ eccentric moment
Location of C.G. Moment of Inertia
n n 2
1
y = å yi I xx = å (y i − y )
n i =1 i =1

These formulas must be applied for both the x and y-directions if

the connection is eccentrically loaded in both directions.
 Connections With
Significant Preload
The total design force in the fastener is the
sum of that from the external load and the
fastener’s preload.

f i = f i,ext + f i,pre

This design approach is conservative if no

separation occurs.
 Combined Shear and Tension
Interaction Curve
τ
2 2
τ all æ τ ö æ σ ö 1
1.0 çç ÷÷ + çç ÷÷ ≤
è τ all ø è σ all ø N fs

Note that just like

σ
fatigue interaction
σ all curves, there are other
1.0 interaction curves in
use.
 Authoritative Source of
Information
The most extensive research on
bolted connections has been
done by the Civil Engineering
community. The best source for
information contained in
standards may be found in the
Guide to Design Criteria for
Bolted and Riveted Joints.

www.boltcouncil.org/guide1.htm
 Assignment

The bracket is to be bolted to a vertical face by means of four

3/8-16UNC-2A SAE Grade 5 bolts. The bolts are preloaded to
90 percent of the proof load. The joint constant, C, is equal to
0.173. Compute the maximum tension stress in the bolts and
identify which bolt(s) in which it will occur. Compute the factor
of safety for the most severely load fastener taking into account
combined tension and shear.
 Welds

Lecture 33

Engineering 473
Machine Design
 What is welding?
Welding is a fusion process, in which metal parts are
heated to the melting point and fused together, usually
with a filler of the same material melted along with the
parent material.
Temperature
control is the
most important
parameter
associated with
obtaining a
good weld.

Haynes Welding Manual, 10445(2110)

 Ancient Welding Processes

Welding was using

in ancient (3,000
years ago) times by
pouring molten
metal into a cast
located on another
metal part. Sand
Sand Cast casting is still used
today.

Arc Welding Instructions for the Beginner, Lincoln Arc Welding Foundation, Fig. 6
 Modern Welding Processes

Oxy-acetylene Tungsten Inert

Gas Welding Gas (TIG)
or
Gas Tungsten
Arc Welding Arc Welding
(GTAW)

There are many welding

processes. Metal Inert Gas
(MIG)
The ability to produce pure or
oxygen @1900 was Gas Metal Arc
important to the Welding
development of welding. (GMAW)
 Arc Welding

Electric current flowing through a

high resistance air gap generates an
intense arc with temperatures ranging
from 6,000 to 10,000 oF.

Arc welding is the

most common
method used with
structural steel
welding.
Arc Welding Instructions for the Beginner, Lincoln Arc Welding Foundation, Fig. 9&13
 MIG or GMAW
Argon and CO2 are
common inert gases used
in the MIG process. A
mixture of 75% Argon
and 25% CO2 (C-25) is
also very common.

Small commercial
MIG welder.

www.hobartbrothers.com

Continuously fed filler wire makes MIG welding suited to high

volume, machine assisted industrial welding applications.
Haynes Welding Manual, 10445(2110)
 TIG or GTAW
The shielding gas in the TIG
process is Helium. This gives
rise to the term heli-arc
welding.

The TIG process is

considered the most
precise, most controllable,
and cleanest method of
welding.

The TIG process can be used to make welds of great strength

and quality in thin, light, and dissimilar materials.
Haynes Welding Manual, 10445(2110)
 Soldering and Brazing

Soldering and brazing are common methods of

joining two metals.

Soldering and brazing are not welding processes

because they do not melt the parent material.

In soldering and brazing, the parent metal is heated

and the the filler metal is melted into the joint where
it cools.

Lead based filler metals are used in soldering and

brass based filler metals are used in brazing.
 Filler Metals
Filler metal rods and wires are designated
by a system established by the American
Welding Society (AWS).

AWS Designation System

Electrode Position of the weld

E70XX

Minimum yield strength Type of process

of the completed weld
(ksi).
The static strength of a good weld is
often higher than the base material.
 Residual Stresses
Ø The uneven heating and cooling in welding operations gives
rise to significant residual stresses in the vicinity of the weld.

Ø The residual stresses may distort the part.

Ø The residual stresses and discontinuities in a weld have a

significant influence on the fatigue life of a part.

Ø The fatigue life of a weld is often less than the base material.

Ø Residual stresses are influenced by the % carbon found in

the material being joined.

Ø Residual stresses may be removed by annealing.

 Butt Welds

Design Criteria
t
P Syp
≤
Lt N fs
 Welds Loaded in Shear

Fillet Weld Geometry

Real Surface
Weld Throat
Potential
Shear Failure Theoretical
Planes Surface

A s = cos 45° ⋅ h ⋅ L
h
= 0.7071⋅ h ⋅ L Deutschman, Fig. 17-5
 Welds Loaded in Shear
(Continued)

Design Criteria
P Sys
=
two shear 2A s N fs
Weld Shear Area
planes
A s = cos 45° ⋅ h ⋅ L
Sys ≈ 0.6Syt
= 0.7071⋅ h ⋅ L
Deutschman, Fig. 17-5
 Assignment
Read sections 9-1 through 9-3 in Shigley.

The figure shows a

horizontal steel bar, 3/8 in
thick, loaded in tension and
welded to a vertical support.
Find the load F that will
cause a shear stress in the
welds of 20 kpsi.
Assignment
(Continued)

A 5/16 in steel bar is welded to

a vertical support as shown.
What is the shear stress in the
welds if the force is F=32 kip?
Eccentrically Loaded Welds

Lecture 34

Engineering 473
Machine Design
 Eccentrically Loaded
Welds
a
P

M = Pa

Actual Loading Equivalent Loading

 Eccentrically Loaded Welds
(Continued)
Shear stress per
y
dl τ = cd unit length and
throat width.
d τ
c ≡ proportionality constant
c.g. M = Pa x
L
M = ò τ ⋅ d ⋅ (1) ⋅ dl
0

Shear stress is proportional M = c ò d 2 ⋅ (1) ⋅ dl

to the distance from the c.g. 0
 Eccentrically Loaded Welds
(Continued)
τ = cd
y
dl
L

d τ M = c ò d 2 ⋅ (1) ⋅ dl
0
c.g. M = Pa x
M
c= L

ò ⋅ (1)⋅ dl
Polar Moment of 2
Inertia of Weld d
0
Line
L
( )
J = ò x 2 + y 2 (1) ⋅ dl
τ=
M ⋅d
0
J
 Eccentrically Loaded Welds
(Continued)

τ τy Moment Contribution
d y My
x τx = − τ = −
d J
y
τx y x Mx
τy = τ =
d J
Py d Component Totals
y
My Px
τx = − +
J L
M Px Mx Py
c.g. x x
τy = +
J L
Note sign convention
 Eccentrically Loaded Welds
(Continued)
Component Totals
Design Criteria
My Px τ Sys
τx = − + ≤
J L 0.7071 ⋅ h N fs
Mx Py
τy = + or
J L τ ⋅N fs
h≥
0.7071 ⋅ Sys
Total Shear Stress
where
Sys ≈ 0.6 ⋅ Syt
τ = τ 2x + τ 2y
h is the minimum leg
τ is the shear stress per unit length
length of the fillet weld
and unit throat width.
 Example
y Weld C.G. Location ~
x=
å x ⋅L
åL

=
(2 in )(4 in ) + (0 in )(6 in )
(4 in ) + (6 in )

= 0.8 in
~y ⋅ L
y=
å
åL

=
(6 in )(4 in ) + (3 in )(6 in )
x (4 in ) + (6 in )

= 4.2 in
 Example
Weld Polar Moment of Inertia
y J = I xx + I yy

1
I xx = 4(1)(1.8) + (1)(6) + 6(1)(1.2 )
2 3 2

12
x = 39.6 in 4

1
(1)(4) + 4(1)(1.18) + 6(1)(0.82 )
3 2 2
I yy =
12
= 14.9 in 4

J = 39.6 + 14.9 = 54.5 in 4

 Example
Weld Stress
y Find Shear Stress at C
M = −5,000(6 )
= −30,000 lb - in
Py = −5,000 lb
x
5,000 lb Px = 0
x = −0.82
y = -4.2 in
L = 10 in
My Px J = 54.6 in 4
τx = − +
J L τ x = −2,310. psi
Mx Py τ y = −49.5 psi
τy = +
J L
 Example
Minimum Weld Size

τ= (- 2,310) + (− 49.5)
2 2
h≥
τ ⋅N fs
= 2.31 ksi 0.7071 ⋅ Sys

N fs = 2.0 h=
(2.31 ksi )(2 )
(0.7071)(36.0 ksi )
Syt = 60.0 ksi
h = 0.181 in

Sys = 0.6(60.0 ) = 36.0 ksi A 3/16 in fillet weld size

would be acceptable.
 Assignment
A 3/8 inch steel bar, to be used as a beam, is welded to a
vertical support using two fillet welds as illustrated. Find
the safe bending force F if the permissible shear stress in the
welds is 20 ksi.
Helical Springs

Lecture 35

Engineering 473
Machine Design
Types of Helical Springs

Extension Springs –
oppose extension

Compression Springs –
oppose compression

Torsional Springs –
oppose rotational motion
Compression Spring Terminology
Spring Internal Forces

The cross section of a spring wire

is exposed to a shear force and a
torsion moment.

Shigley, Fig. 10-1

 Stresses in Helical Springs
π 4
Tr F J = 2I = d
τ max = + 32
J A
π 4
A= d
Tr 4
τ=
J Wire
d Diameter
r=
2
Pitch
FD Diameter
τ=
F T=
A 2
 Stresses In Helical Springs
(Continued)
Tr
τ=
J Tr F
τ max = +
J A

8FD 4F
F τ max = 3
+ 2
τ= πd πd
A

π 4 π 2 d FD
J = 2I = d A= d r= T=
32 4 2 2
 Stresses in Helical Springs
(Continued)

8FD 4F 8FD 8DF æ 1 ö

τ max = + 2 τ max = 3
+ 3ç ÷
πd 3
πd πd πd è 2C ø

8FD 4F æ 2D öæ d ö 8FD æ 2C + 1 ö
τ max = 3
+ 2ç ÷ç ÷ τ max = 3 ç ÷
πd πd è 2D øè d ø πd è 2C ø

8FD 8DF æ d ö 8FD

τ max = 3
+ 3ç ÷ τ max = Ks
πd πd è 2D ø πd 3

D
Let C = , then 2C + 1
Ks =
d 2C
 Curvature Stress Concentration
Factor
A stress concentration occurs on the inside of a helical
spring due to the curvature of the wire.

The formula for the stress concentration factor, Kc, is

given by

2C(4C + 2) 8FD
Kc = τ max = K cKs
(4C − 3)(2C + 1) πd3
 Curvature Stress Concentration
Factor
(Continued)
D
C=
d

Kc

C
Note that for large C, Kc approaches 1.
Kc is important for fatigue calculations.
 Deflection/Stiffness
Strain Energy/Unit Volume Strain energy from a spring
comes from two sources: 1)
torsion, and 2) shear loading.
σ
Torsion Strain Energy

1
U T = ò τ ⋅ γ ⋅ dvol
2 vol
Tr τ
ε τ= ,γ=
J G
1 1 æ Tr ö æ Tr ö
U = ò σ ⋅ ε ⋅ dvol UT = ò ç ÷ ⋅ ç ÷ ⋅ dvol
2 vol 2 volè J ø è JG ø
 Deflection/Stiffness
(Continued)

Torsion Strain Energy Shear Load Strain Energy

(Continued)
1
U T = ò τ ⋅ γ ⋅ dvol
1 æ Tr ö æ Tr ö 2 vol
UT = ò ç ÷ ⋅ ç ÷ ⋅ dvol
2 volè J ø è JG ø F τ
τ= ,γ=
1 T2 A G
UT = ò ò r dA ⋅ dL
2
2 1 æFö æ F ö
2 J G LA UF = ò ç ÷ ⋅ ç ÷ ⋅ dvol
2 volè A ø è AG ø
1 T 2L
UT = 1 F2 L
2 JG UF =
2 AG
 Deflection/Stiffness
(Continued)
Total Strain Energy Castigliano’s Theorem
U = UT + U F
∂U
y=
∂F
1 T 2 L 1 F2 L
U= +
2 JG 2 AG
FD2 L FL
y= +
T = F⋅ D 2 4JG AG

2 2 2
y => spring deflection due to
1FDL 1FL load F
U= +
2 4JG 2 AG
 Deflection/Stiffness
(Continued)

2
FD L FL 3
8FD N æ 1 ö
y= + y= 4
4JG AG ç1 + 2 ÷
d G è 2C ø
π 4
J = 2I = d
32 8FD3 N
π 2 y≈ 4
A= d dG
4
D d 4G Equation for
C= k= stiffness of a
d 8D N3
spring.
 Extension Spring Ends

The end loop of an extension

spring is weaker than the coils.

A stress concentration factor of

K=rm/ri is present in the loop.

Shigley, Fig’s 10.2 & 10.3

Compression Spring Ends

Shigley, Fig. 10.5

Compression Spring Ends
(Continued)

These formulas show how the

type of ends effect the
number of coils and length of
the spring.
Spring Materials

www.acxesspring.com/properties.htm
Mechanical Properties of
Ductile Metallic Materials
Lecture 1

Engineering 473
Machine Design
 Mechanical Properties
(Static Strength – Monotonic Elongation)
P

Sut
SFt
S yt
Set
Stress
lo σ = P/A 0

εe ε y εu εF
0.002 li − l0
P ε=
l0
 Mechanical Properties
(Static Strength Nomenclature)
Subscripts
y ≡ 0.2% offset yield
Sut
u ≡ ultimate SFt
S yt
e ≡ elastic Set
F ≡ fracture Stress
t ≡ tension σ = P/A 0
c ≡ compression

Syt & Sut are εe ε y εu εF

generally given li − l0
0.002 ε=
in handbooks.
l0
 Mechanical Properties
(True Stress & True Strain)

Logarithmic Strain σF
σu
dl
dε =
l

True Stress
li
dl li
ε = ò = ln
lo
l lo
True Stress
P
σ=
Ai εu εF
Logarithmic Strain
Mechanical Properties
(Example Data)

True Stress-Logarithmic
Strain Curves for Several
Metallic Materials

H. Schwartzbart, W.F. Brown,

Jr., “Notch-Bar Tensile
Properties of Various Materials
and their Relation to the Unnotch
Flow Curve and Notch
Sharpness,” Trans. ASM, 46, 998,
1954.
Mechanical Properties
(High Strain Rates)

Stress-Strain Curves for

Mild Steel at Room
Temperatures at Various
Rates of Strain

Manjoine, M.J., “Influence of Rate of

Strain and Temperature on Yield Stresses
of Mild Steel,” Journal of Applied
Mechanics, 11(A):211-218, December
1944.
 Mechanical Properties
(High Strain Rates & High Temperatures)

Experimental Data for

6061-T6 Aluminum

Hoge, K.G., “Influence of Strain

Rate on Mechanical Properties of
6061-T6 Aluminum under
Uniaxial and Biaxial States of
Stress,” Experimental Mechanics,
6:204-211, April 1966.
 Mechanical Properties
(Monotonic Compression)

P li − l0
ε=
l0 εu ε y εe 0.002

Stress
lo σ = P/A 0

Sec
S yc
Suc
P
 Mechanical Properties
(Work Hardening or Cold Working)

Syt
Syt
σ

ε
 Mechanical Properties
(Reverse Loading)

Bauschinger’s Effect

Yield stress in compression

may decrease after an initial
load application past the
tension yield point.

This phenomena is Monotonic

Compression Curve
an important topic in
plasticity theory.
Mechanical Properties
(Stress Controlled Cyclic Loading)

Materials can demonstrate three

characteristics: 1) cyclic hardening, 2)
cyclic softening, and 3) cyclic strain
accumulation (ratcheting).

Skrzypek, J.J., Plasticity and Creep:

Theory, Examples, and Problems, CRC
Press, 1993, 130.
Mechanical Properties
(Strain Controlled Cyclic Loading)

Materials can demonstrate

two characteristics: 1) cyclic
hardening and 2) cyclic
softening.

Skrzypek, J.J., Plasticity and Creep:

Theory, Examples, and Problems, CRC
Press, 1993, 130.
 Mechanical Properties
(Creep)

Creep is most
pronounced at high ε Failure strain σ, T
temperatures. It may
also occur at room
temperatures when
the stress level is
close to the yield
strength. time
Primary Secondary Tertiary
Creep Creep Creep

Typical curves obtained from constant stress/temperature tests.

 Summary
The strength of ductile metallic materials is
dependent on several parameters.

1. Load Direction (Tensile or Compressive)

2. Strain Rate (Slow or Fast)
3. Temperature (Hot or Cold)
4. Load History (Monotonic or Cyclic)
5. Fabrication Process (Next Class)

• Metals are complex materials when used throughout

their total response envelope.
• Fortunately their elastic properties are most
commonly used.
 Assignment
Read pages 25-34 in Mott.
 Fatigue
Lecture 10

Engineering 473
Machine Design
 Load Histories and
Design Objectives

Failure
σ, F σ, F
Failure

t, time t, time

Monotonic, Static, or Steady Dynamic, Cyclic, or Unsteady

Design for Strength Design for Life

 Rotating Beam Fatigue Testing

Fatigue Dynamics, Inc. rotating

beam test equipment.

Mott, Fig. 5-2 & 5-3 www.fdinc.com

 S-N Curve

Completely reversed cyclic stress, UNS G41200 steel

Shigley, Fig. 7-6
 Fatigue Strength
The Fatigue Strength, Sf(N), is the stress level that a
material can endure for N cycles.

The stress level at which the material can withstand an

infinite number of cycles is call the Endurance Limit.

The Endurance Limit is

observed as a horizontal
line on the S-N curve.

Shigley, Fig. 7-6

 Representative S-N Curves

Note that non-ferrous materials often exhibit no

endurance limit.
Mott, Fig. 5-7
 Endurance Limit
Vs Tensile Strength

Conservative
Lower Bound
for Ferrous
Materials
S′e = 0.3Sut

S′e ≡ Endurance Limit of Test Specimen

Sut ≡ Tensile Strength of Test Specimen
Shigley, Fig. 7-7
 Endurance Limit
Multiplying Factors
(Marin Factors)

Se = k a ⋅ k b ⋅ k c ⋅ k d ⋅ k e ⋅ S′e
There are several
factors that are known
Se ≡ Endurance limit of part to result in differences
between the endurance
S′e ≡ Endurance limit of test specimen
limits in test specimens
k a ≡ Surface factor and those found in
k b ≡ Size factor machine elements.
k c ≡ Load factor
k d ≡ Temperature factor See sections 7-8 & 7-9 in
k e ≡ Miscellaneous - effects factor Shigley for a discussion on
each factor.
Mean Stress Effects
• The S-N curve obtained from a
rotating beam test has completely
reversed stress states.

• Many stress histories will not have

completely reversed stress states.

Shigley, Fig. 7-12

 Definitions
Stress Range
σ r = σ max − σ min

Alternating Stress
σ max − σ min
σa =
2
Mean Stress
σ max + σ min
σm =
2
Stress Ratio Amplitude Ratio
Note that R=-1 for a
σ min σa completely reversed
R= A=
σ max σm stress state with zero
mean stress.
 Mean Stress Fatigue Testing

Fatigue Dynamics, Inc.,

fluctuating fatigue stress
testing equipment.

www.fdinc.com
 Fluctuating Stress Failure Data

Note that a tensile

mean stress results
in a significantly
lower fatigue
strength for a given
number of cycles to
failure.

Note that a curved

line passes through
This plot shows the fatigue strength of
the mean of the
several steels as a function of mean
data.
stress for a constant number of cycles to
failure.
Shigley, Fig. 7-14
Master Fatigue Plot

Shigley, Fig. 7-15

Fluctuating Stress Failure
Interaction Curves

Shigley, Fig. 7-16

 Soderberg Interaction Line
Any combination of mean and
k f Sa S m alternating stress that lies on or
+ =1 below the Solderberg line will
Se S yt
have infinite life.

Factor of Safety Format

k f Sa S m 1
+ =
Se S yt N f

Note that the fatigue stress

concentration factor is applied
only to the alternating
component.
 Goodman Interaction Line

k f Sa S m Any combination of mean and

+ =1 alternating stress that lies on or
Se Sut below the Goodman line will
have infinite life.

Factor of Safety Format

k f Sa Sm 1
+ =
Se Sut N f
Note that the fatigue stress
concentration factor is applied
only to the alternating
component.
 Gerber Interaction Line
Any combination of mean and
2 alternating stress that lies on or
k f Sa æ S m ö below the Gerber line will have
+ çç ÷÷ = 1 infinite life.
Se è Sut ø
Factor of Safety Format
2
k f N f Sa æ N f S m ö
+ çç ÷÷ = 1
Se è Sut ø

Note that the fatigue stress

concentration factor is applied
only to the alternating
component.
Modified-Goodman
Interaction Line

The Modified-Goodman
Interaction Line never exceeds
the yield line.
 Example No. 1

A 1.5-inch round bar has been machined from AISI 1050 cold-
drawn round bar. This part is to withstand a fluctuating tensile
load varying from 0 to 16 kip. Because of the design of the
ends and the fillet radius, a fatigue stress-concentration factor
of 1.85 exists. The remaining Marin factors have been worked
out, and are ka=0.797, kb=kd=1, and kc=0.923. Find the factor
of safety using the Goodman interaction line.

Shigley, Example 7-5

 Example No. 1
(Continued)
Sut = 100. ksi σ max + σ min
S′e ≈ 0.50 ⋅ Sut = 50. ksi σm = = 4.52 ksi
2

π ⋅ d2 Se = k a k b k c k dS′e
A= = 1.77 in 2
4
= (0.797 )(1)(0.923)(1)(50 ksi )
16 kip Se = 36.8 ksi
σ max = 2
= 9.04 ksi
1.77 in
σ min = 0 ksi

σ max − σ min
σa = = 4.52 ksi
2
 Example No. 1
(Continued)

k f σa σm 1
+ =
Se Sut N f

1.85 ⋅ 4.52 ksi 4.52 ksi 1

+ = 0.272 =
36.8 ksi 100. ksi Nf

N f = 3.67
 Example
5 in 5 in Pmax = 1000 lb
Pmin = 350 lb
1 2

Material UNS
1.5 in. dia. 0.875 in. dia. G41200 Steel
0.125 in. rad. Notch sensitivity
Will the beam have infinite life? q=0.3

π 4 π I1 0.249 in 4
D 1 = (1.5) = 0.249 in 4 S1 = = = 0.332 in 3
4
I1 =
64 64 c1 0.75 in
π 4 π I 2 0.088 in 4
D 2 = (0.875) = 0.088 in 4
4
I2 = S2 = = = 0.201 in 3
64 64 c 2 0.438 in
 Example
(Continued)
5 in 5 in Pmax = 1000 lb
Pmin = 350 lb
1 2

Material UNS
1.5 in. dia. 0.875 in. dia. G41200 Steel
0.125 in. rad. Notch sensitivity
q=0.3

kf −1 D 1.5 in k t = 1.61
q= = = 1.71
k t −1 d 0.875 in k f = 1 + q(k t − 1)
r 0.125
k f = 1 + q(k t − 1) = = 0.143 = 1 + 0.3(1.61 − 1)
d 0.875
= 1.18
Ref. Peterson
 Example
(Continued)
5 in 5 in Pmax = 1000 lb
Pmin = 350 lb
1 2

Material UNS
1.5 in. dia. 0.875 in. dia. G41200 Steel
0.125 in. rad. Notch sensitivity
q=0.3
Section 1 (Base)

M1 (1000 lb )(10 in ) σ max − σ min

σ max = = = 30.1 ksi σa = = 9.8 ksi
S1 0.332 in 3
2
M1 (350 lb )(10 in ) σ max + σ min
σ min = = 3
= 10.5 ksi σm = = 20.3 ksi
S1 0.332 in
2
 Example
(Continued)
5 in 5 in Pmax = 1000 lb
Pmin = 350 lb
1 2

Material UNS
1.5 in. dia. 0.875 in. dia.
G41200 Steel
0.125 in. rad. Notch sensitivity
Section 2 (Fillet) q=0.3

M1 (1000 lb )(5 in ) σ max − σ min

σ max = = = 24.9 ksi σa = = 8.10 ksi
S1 0.201 in 3
2
M1 (350 lb )(5 in ) σ max + σ min
σ min = = 3
= 8.71 ksi σm = = 16.8 ksi
S1 0.201 in 2
 Example
(Continued)
Section 1 (Base)

M (1000 lb)(10 in ) = 30.1 ksi σ max − σ min

σ max = 1= σa = = 9.8 ksi
S1 0.332 in 3 2
M1 (350 lb )(10 in ) σ max + σ min
σ min = = 3
= 10.5 ksi σm = = 20.3 ksi
S1 0.332 in 2

Sut = 116 ksi 1.0(9.8 ksi ) 20.3 ksi

+ = 0.502
S′e = 30 ksi = Se 30 ksi 116 ksi

k f σa σm 1 1
+ = Nf = = 1.99
Se Sult N f 0.502
Part has infinite life.
 Example
(Continued)

Section 2 (Fillet)

σ max
M
= 1=
(1000 lb)(5 in ) = 24.9 ksi σa =
σ max − σ min
= 8.10 ksi
S1 0.201 in 3 2
M1 (350 lb )(5 in ) σ max + σ min
σ min = = 3
= 8.71 ksi σm = = 16.8 ksi
S1 0.201 in 2

Sut = 116 ksi 1.18(8.10 ksi ) 16.8 ksi

+ = 0.463
S′e = 30 ksi = Se 30 ksi 116 ksi

k f σa σm 1 1
+ = Nf = = 2.16
Se Sult N f 0.463

Part has infinite life.

 Assignment
Problem 1
 Assignment
(Continued)
Problem 2
 Influence of Fabrication
Processes on the Strength of
Metals
Lecture 2

Engineering 473
Machine Design
Things that Affect Metal Strength

The strength of ductile metallic materials is

dependent on several parameters.

1. Load Direction (Tensile or Compressive)

2. Strain Rate (Slow or Fast)
3. Temperature (Hot or Cold)
4. Load History (Monotonic or Cyclic)
5. Fabrication Process
 Common Fabrication Processes

Cold-working
Casting
Heading
Sand Casting
Roll threading
Investment Casting
Spinning
Shell Molding
Stamping
Powder-Metallurgy
Hot-working Heat Treatment
Annealing
Hot rolling
Extrusion Quenching
Forging Tempering
Case Hardening
 Hot Working

Hot working of metals is done for two reasons

1. Plastically mold the metal into the

desired shape
2. Improve the properties of the metal as
compared to the as-cast condition
 Microstructure Changes due to Hot
Rolling

Large coarse
Smaller grains
grain structure

The granular structure of the material

is changed during hot rolling.
Allen, Fig. 16-14
Hot Working Temperatures
Material Temperature Range (oF)
Aluminum 650-900
Aluminum Alloys 750-900
Beryllium 700-1300
Brass 1200-1475
Cooper 1200-1650
High Speed Steels 1900-2200
Inconel 1850-2350
Magnesium Alloys 400-750
Monel 1850-2150
Nickel 1600-2300
Refractory Metals & Alloys 1800-3000
Steel: Carbon 1900-2400
Low Alloy 1800-2300
Stainless 1900-2200
Titanium 1400-1800
Zinc Alloys 425-550
Allen, Table 16-1
 Example of Microstructure Changes

Low carbon cast steel

(A) As cast (dendritic structure)

(B) After hot rolling (reduced grain size)
(C) After temper rolling (elongated
grains) Directional Properties

Allen, Fig, 16-18.

 Beneficial Effects of Hot Rolling

Typical defects in cast metals which

are minimized in hot worked metals

1. Large grain size (due to slow cooling)

2. Porosity (voids due to shrinkage)
3. Blow holes (due to gas evolution during
solidification)
4. Segregation (due to limited solubility in the solid
state)
5. Dirt and slag inclusions
6. Poor surface condition (due to oxides and scale)

The strength of hot rolled metals is higher than cast metals.

Allen, pg 508.
 Forging

• A hot working process

• Metal flows under high
compressive stresses
• May be used with or
without die cavity to
obtain a specific shape

A blacksmith uses a hammer and

an anvil to forge metallic parts.
Forged Workpiece

The curvature on the

sides of a forged
product is due to
friction between the
ram and the workpiece.

Allen, Fig. 16-19

Directional Nature of Forged Material
Properties

Flow lines in upset forging of 1.5” dia. AISI 1045

steel specimen at 1800 oF.
Flow lines are caused by the elongation of
slag particles or non-metallic inclusions.
Allen, Fig. 16-23
 Strength of Forged Materials

• Forged products generally have substantially

higher strength properties than cast products.

• Cast products have material properties that are

approximately the same in all directions (isotropic).

• Forged products have material properties that are

different in each direction. Transverse properties are
significantly less than the longitudinal direction
(orthotropic or anisotropic)
Extrusion

Allen, Fig. 16-25

Example of Extruded Aluminum
Cross Sections

Allen, Fig. 16-24

Directional Nature of Extrusion
Material Properties

Flow Lines in Extruded Section

Allen, Fig’s 16-26 and 16-27

Extrusion Conditions for Typical
Metals

Allen, Table 16-2

Strength of Extruded Materials

• High degree of grain flow in the direction

parallel to the axis of extrusion.

• High strength properties in the direction

parallel to the axis of extrusion.

• Lower strength properties in the direction

transverse to the axis of extrusion.
 Spinning

Conventional Spin Forming Shear Spin Forming

(No change in material thickness) (Significant material thickness changes)

Used to produce rocket motor casings and missile nose cones.

Allen, Fig. 16-43
Directional Nature of Spin Formed
Material Properties

Grid Flow Lines in Shear Spun Copper Cone

Allen, Fig. 16-44

 Effect of Cold Working on
Microstructure

Grain boundaries in
3003 aluminum alloy.
Strength of Spin Formed Materials

• Spin formed products have increased strength in the

longitudinal direction

• Strength properties in the transverse direction (through

thickness) may be significantly different.
Heat Treatment

Heat Treating
Processes

• Annealing
• Quenching
• Tempering
• Case Hardening
 Annealing

Heat treating operation used to:

1) Refine the grain structure,

2) Relieve residual stresses,
3) Increase ductility.
Annealing Effects
Recrystallization
The growth of new stress-free
equiaxed crystals in cold worked
materials. Occurs after a
critical (recrystallization)
temperature is reached.

Equiaxed Crystals
Have equivalent dimensions
in all directions (i.e. not
longer in one direction)

Flinn, Fig. 3-19

Fabrication Processes Summary
• Hot and cold working fabrication processes
have significant influence on the materials
strength.
• Cast materials generally have uniform or
isotropic material strength.

• Cold and hot worked materials generally have

higher strengths. Strength properties are dependent
on direction (orthotropic or anisotropic)

• Standard practice is to obtain/verify material

properties from sample product in the direction
of highest stress/strain.
 Fabrication Processes Summary
(Continued)

• Annealing may be used on hot and cold worked materials

to obtain uniform properties and to relieve fabrication
induced stresses.

• Heat treating may be performed to obtain strength

properties and characteristics higher than the annealed
state.
 Summary
The strength of ductile metallic materials is
dependent on several parameters.

1. Load Direction (Tensile or Compressive)

2. Strain Rate (Slow or Fast)
3. Temperature (Hot or Cold)
4. Load History (Monotonic or Cyclic)
5. Fabrication Process (Hot or cold working
and/or heat treatment)
Assignment

Read pages 35-51

Stress at a Point
Lecture 3

Engineering 473
Machine Design
 Purpose
The stress state at critical locations in a machine
component is required to evaluate whether the
component will satisfy strength design
requirements.

The purpose of this class is to review the concepts

and equations used to evaluate the state of stress at
a point.
 2D Cartesian Stress
σ Components
yy
τ yx
Notation
τ xy
σ Normal Stress
σ xx σ xx
τ Shear Stress
τ xy
τ xy
τ yx
Y σ yy Direction
Face

ĵ Moment equilibrium requires that

î X τ xy = τ yx
 Tensor Sign Convention
This sign convention must be
used to satisfy the differential
τ yx equilibrium equations and
Positive
Face τ xy tensor transformation
relationships.
Negative

Positive
σ xx σ xx
Face
Face

τ xy Negative
Face
Stresses acting in a positive
τ yx coordinate direction on a positive
Y σ yy face are positive.

Shear stresses acting in the

negative coordinate direction on a
ĵ negative face are positive.
î X
 2D Mohr’s Circle
Sign Convention
τ yx
τ xy The sign convention used
with the 2D Mohr’s circle
σ xx σ xx equations is slightly
different.
τ xy
τ A positive shear stress is
yx
Y σ yy one that tends to create
clockwise (CW) rotation.

ĵ
î X
 2D Mohr’s Circle
(Transformation of Axis)
y σ
τ φ
All equations for a 2-D
τ xy Mohr’s Circle are
φ derived from this figure.
ds
σ xx dy
dx ΣF in the x- and y-
x directions yields the
τ yx transformation-of-axis
σ yy equations

σ xx + σ yy σ xx − σ yy
σ= + cos(2φ ) + τ xysin (2φ )
2 2
σ xx − σ yy
τ=− sin (2φ ) + τ xycos(2φ )
2
 2D Mohr’s Circle
(Principal Stress Equations)

The transformation-of-axis equations can be used to

find planes for which the normal and shear stress are
the largest.

2
σ xx + σ yy æ σ xx − σ yy ö
σ1 , σ 2 = ± çç ÷÷ + τ 2xy
2 è 2 ø
2
æ σ xx − σ yy ö
τ1 , τ 2 = ± çç ÷÷ + τ 2xy
è 2 ø

We will use these equations extensively during this class.

2D Mohr’s Circle
(Graphical Representation)

2
σ xx + σ yy æ σ − σ yy ö
σ1 , σ 2 = ± çç xx ÷÷ + τ 2xy
2 è 2 ø
2
æ σ − σ yy ö
τ1 , τ 2 = ± çç xx ÷÷ + τ 2xy
è 2 ø

Note that the shear

stress acting on the
plane associated with
a principal stress is
always zero.

Shigley, Fig. 3.3

 Comments on Shear Stress
Sign Convention
2
τ σ xx + σ yy æ σ − σ yy ö
yx σ1 , σ 2 = ± çç xx ÷÷ + τ 2xy
2 è 2 ø
τ xy 2

σ xx σ xx æ σ − σ yy ö
τ1 , τ 2 = ± çç xx ÷÷ + τ 2xy
è 2 ø
τ xy
Tensor
τ yx
σ yy The sign convention is
important when the
τ yx
transformation-of-axis
τ xy equations are used.
σ xx σ xx
The same answer is obtained
τ xy
when computing the principal
τ 2D Mohr’s
yx stress components.
σ yy Circle
 3D Stress Components
σ yy Note that the tensor
τ yx sign convention is
used.
τ xy
τ zy σ xx There are nine
τ zx τ xz components of stress.
Moment equilibrium
σ zz can be used to reduce
the number of stress
y components to six.
τ xy = τ yx
ĵ î
x τ xz = τ zx
k̂ τ yz = τ zy
z
 Cauchy Stress Tensor

Tensor Transformation σ is known as the Cauchy stress

Equation ≈
tensor. Its Cartesian components
éσ xx τ xy τ xz ù are shown written in matrix form.
ê ú
σ = ê τ yx σ yy τ yz ú Tensors are quantities that are
≈
ê τ zx τ zy σ zz úû invariant to a coordinate
ë
transformation.

Tensor Transformation A vector is an example of a first

Equation
order tensor. It can be written with
σ mn = β mi σ ij β jn respect to many different
coordinate systems.
 y Cauchy Formula
B H
T
τ xy σzz ΣF in the x,y,and z directions
τzx n̂ yields the Cauchy Stress
σxx
τ xz P τzy Formula.
τ yz C x
τ yx
A σ yy n̂ = lî + mĵ + nk̂ éσ xx τ xy τ xz ù ì l ü ìTx ü
ê úï ï ï ï
y ê τ yx σ yy τ yz ú ím ý = íTy ý
z
B
H ê τ zx
ë τ zy σ zz úû ïî n ïþ ïîTz ïþ
T
τu
σn n̂ This equation is similar to the
τv Mohr’s circle transformation-
P C x
of-axis equation
A
z
 3D Principal Stresses
éσ xx τ xy τ xz ù ì l ü ìTx ü The shear stress on planes
ê úï ï ï ï normal to the principal
ê τ yx σ yy τ yz ú ím ý = íTy ý
stress directions are zero.
ê τ zx
ë τ zy σ zz úû ïî n ïþ ïîTz ïþ
We need to find the plane in
éσ xx τ xy τ xz ù ì l ü ìlü which the stress is in the
ê úï ï ï ï
ê τ yx σ yy τ yz ú ím ý = σ ím ý direction of the outward unit
ê τ zx
ë τ zy σ zz úû ïî n ïþ ïn ï
î þ
normal.

é(σ xx − σ ) τ xy τ xz ù ì l ü ì0ü This is a

ê
ê τ yx (σ yy − σ) τ yz
úï ï ï ï
ú ím ý = í0ý
homogeneous
linear equation.
ê τ zx
ë τ zy (σ zz − σ )úû ïî n ïþ ïî0ïþ
 3D Principal Stresses
(Eigenvalue Problem)

é(σ xx − σ ) τ xy τ xz ù ì l ü ì0ü A homogeneous linear

equation has a solution
ê
ê τ yx (σ yy − σ)
úï ï ï ï
τ yz ú ím ý = í0ý
only if the determinant of
ê τ zx
ë τ zy (σ zz − σ )úû ïî n ïþ ïî0ïþ the coefficient matrix is
equal to zero.

(σ xx − σ ) τ xy τ xz
τ yx (σ yy − σ) τ yz = 0 This is an eigenvalue
τ zx τ zy (σ zz − σ ) problem.
 3D Principal Stresses
(Characteristic Equation)

(σ xx − σ ) τ xy τ xz The determinant can be

τ yx (σ yy − σ) τ yz = 0 expanded to yield the
τ zx τ zy (σ zz − σ ) equation

I1, I2, and I3 are known as

the first, second, and third
σ3 − I1σ 2 + I 2σ − I3 = 0 invariants of the Cauchy
stress tensor.
I1 = σ xx + σ yy + σ zz
I 2 = σ xx σ yy + σ yyσ zz + σ zzσ xx − τ 2xy − τ 2yz − τ 2zx
I3 = σ xx σ yyσ zz + 2τ xy τ yz τ zx − σ xx τ 2yz − σ yy τ 2zx − σ zz τ 2xy
 3D Principal Stresses
There are three roots to the
Characteristic Equation
characteristic equation, σ1,
σ3 − I1σ 2 + I 2σ − I3 = 0 σ2, and σ3.

Each root is one of the

principal stresses.

The direction cosines can be

The direction cosines
found by substituting the
define the principal
principal stresses into the
directions or planes.
homogeneous equation and
solving.
 3D Mohr’s Circles

τ τ1,3
Note that the principal
τ1,2 stresses have been ordered
such that σ1 ≥ σ 2 ≥ σ3 .
τ2,3

σ3 σ2 σ1
σ
Maximum shear stresses

σ1 − σ 2
τ1,2 =
2
σ 2 − σ3
τ 2,3 =
2
σ1 − σ 3
τ1,3 =
2
 Octahedral Stresses

σ oct = I1 = (σ1 +σ 2 +σ3 ) = (σ xx +σ yy +σ zz )

1 1 1
3 3 3
(
2 2
) 1
τ oct = τ 1,2 + τ 2,3 + τ1,3 2
3
2 2

1
[
= (σ1 − σ 2 ) + (σ 2 − σ3 ) + (σ3 − σ1 )
3
2 2
1
2 2
]
1 é(σ xx − σ yy ) + (σ yy − σ zz ) + (σ zz − σ xx )
1
2 2 2
ù 2

= ê ú
( )
3 ê+ 6 τ 2xy + τ 2yz + τ 2xz
ë úû

Note that there eight corner planes in a cube.

Hence the name octahedral stress.
 Assignment
Read chapter 4 – Covers Mohr’s Circle in detail.
Draw a Mohr’s circle diagram properly labeled, find the principal normal and
maximum shear stresses, and determine the angle from from the x axis to σ1.
σxx=12 ksi, σyy=6 ksi, τxy=4 ksi cw.

Use the Mohr’s circle formulas to compute the principal stresses and compare
to those found using the Mohr’s circle graph.

Write the stress components given above as a Cauchy stress matrix. Use
MATLAB to compute the principal stresses. Compare the answers to those
found using Mohr’s circle. Note that tensor notation is required.

Derive the Cauchy stress formula. Hint: Ax=A l, Ay=A m, Az=A n

Verify the that the terms in the 3D characteristic equation used to compute
the principal stresses are correct.
Stress Concentration Factors and
Notch Sensitivity

Lecture 4

Engineering 473
Machine Design
 Photoelasticity

Photoelasticity is a visual method for viewing the

full field stress distribution in a photoelastic
material.
www.measurementsgroup.com
 Photoelasticity
(Continued)

When a photoelastic material is strained and viewed with

a polariscope, distinctive colored fringe patterns are
seen. Interpretation of the pattern reveals the overall
strain distribution.
www.measurementsgroup.com
Components of a Polariscope

Vishay Lecture-Aid Series, LA-101

 Radiometric Thermoelasticity

Automobile Hook and Clevis Crack Tip

Connecting Rod

When materials are stressed the change in atomic spacing

creates temperature differences in the material. Cameras
which sense differences in temperature can be used to
display the stress field in special materials.
www.stressphotonics.com
 Stress Distributions Around
Geometric Discontinuities

Photoelastic fringes in anotched Photoelastic fringes in a narrow

beam loaded in bending. plate with hole loaded in tension.
Deutschman, Fig. 5-3
Effect of Discontinuity Geometry

The discontinuity geometry has a significant effect on

the stress distribution around it. Vishay Lecture-Aid Series, LA-101
 Geometric Stress Concentration
Factors

σ max
Kt =
σ nom

F
σ nom =
A0
A 0 = (w − d )t

Geometric stress concentration factors can

be used to estimate the stress amplification
in the vicinity of a geometric discontinuity.
Shigley, Fig. 2-22
Geometric Stress Concentration
Factors
(Tension Example)

r
d Spotts, Fig. 2-8, Peterson
Geometric Stress Concentration
Factors
(Bending Example)

Spotts, Fig. 2-9, Peterson

Geometric Stress Concentration
Factors
(Torsion Example)

Spotts, Fig. 2-10, Peterson

Geometric Stress Concentration
Factors
(Tension Example)

Spotts, Fig. 2-11, Peterson

Geometric Stress Concentration
Factors
(Bending Example)

Spotts, Fig. 2-12, Peterson

Geometric Stress Concentration
Factors
(Torsion Example)

Spotts, Fig. 2-13, Peterson

 Geometric Stress Concentration
Factors
(Summary)

K t is used to relate the maximum stress at

the discontinuity to the nominal stress.

K t is used for normal stresses

K ts is used for shear stresses

K t is based on the geometry of the discontinuity

σ nom is usually computed using the minimum cross section

Rotating Beam Fatigue Tests

Spotts, Fig. 2-25

 Un-notched and Notched Fatigue
Specimens

Mc Mc
σ = Kt σ = Kf
I I

Comparisons of fatigue test results

for notched and un-notched
specimens revealed that a reduced Kt
was warranted for calculating the
fatigue life for many materials. www.stressphotonics.com
Fatigue Stress Concentration Factors

Maximum stress in notched specimen

Kf =
Stress in notch - free specimen

or

Endurance limit of a notched specimen.

Kf =
Endurance limit of a notch - free specimen.
 Notch Sensitivity Factor
The notch sensitivity of a material is a measure
of how sensitive a material is to notches or
geometric discontinuities.

Kf −1
q= 0 ≤ q ≤1
K t −1

K f = 1 + q(K t − 1) 1 ≤ Kf ≤ Kt
Notch Sensitivity Factors
(Bending Example)

Shigley, Fig. 5-16

Notch Sensitivity Factors
(Torsion Example)

Shigley, Fig. 5-17

Fatigue Stress Concentration Factors

• Kf is normally used in fatigue calculations

but is sometimes used with static stresses.

• Convenient to think of Kf as a stress

concentration factor reduced from Kt
because of lessened sensitivity to notches.

• If notch sensitivity data is not available, it is

conservative to use Kt in fatigue
calculations.
 References

Deutschmann, A.D., W.J. Michels, C.E. Wilson, Machine

Design: Theory and Practice, Macmillan, New York, 1975.

Peterson, R.E., “Design Factors for Stress Concentrations,

Parts 1 to 5,” Machine Design, February-July, 1951.

Shigley, J.E., C.R. Mischke, Mechanical Engineering Design,

5th Ed., McGraw-Hill, Inc., New York, 1989.

Spotts, M.F., Design of Machine Elements, 7th Ed., Prentice

Hall, New Jersey, 1998.

www.measurementsgroup.com

www.stressphotonics.com
 Assignment

1. Read – Sections 3-21 and 3-22

2. Find the most critically stressed

location on the stepped shaft. Note
that you will need to use the stress
concentration factors contained in
the lecture notes.
Steady Load Failure
Theories
Lecture 5

Engineering 473
Machine Design
 Steady Load Failure Theories

Ductile • Maximum-Normal-Stress Brittle

Materials • Maximum-Normal-Strain Materials
• Maximum-Shear-Stress
• Distortion-Energy
• Shear-Energy
Uniaxial • Von Mises-Hencky Multiaxial
Stress/Strain • Octahedral-Shear-Stress Stress/Strain
Field • Internal-Friction Field
• Fracture Mechanics

Many theories have been put forth – some agree

reasonably well with test data, some do not.
The Maximum-Normal-Stress Theory

Postulate: Failure occurs when one of the three

principal stresses equals the strength.

σ1, σ 2, and σ 3 are

σ1 > σ 2 > σ 3
principal stresses

Failure occurs when either

σ1 = St Tension St ≡ Strength in Tension
Sc ≡ Strength in Compression
σ 3 = −Sc Compression
Maximum-Normal-Stress Failure
Surface
(Biaxial Condition)

locus of failure states

σ2
St
According to the
Maximum-Normal-Stress
σ1
Theory, as long as stress
- Sc St state falls within the box,
the material will not fail.

- Sc
 Maximum-Normal-Stress Failure
Surface
(Three-dimensional Case)

σ2

~ St According to the
Maximum-Normal-Stress
σ1 Theory, as long as stress
state falls within the box,
~
~ the material will not fail.

σ3
- Sc
 The Maximum-Normal-Strain
Theory
(Saint-Venant’s Theory)

Postulate: Yielding occurs when the largest of the

three principal strains becomes equal to the
strain corresponding to the yield strength.

Eε1 = σ1 − ν(σ 2 + σ 3 ) = ±S y
E ≡ Young' s Modulus
Eε 2 = σ 2 − ν(σ1 + σ 3 ) = ±S y
ν ≡ Poisson' s Ratio
Eε 3 = σ 3 − ν (σ1 + σ 2 ) = ±S y
Maximum-Normal-Strain Theory
(Biaxial Condition)

locus of failure states

σ2
σ1 − νσ 2 = ±S y
Sy
σ 2 − νσ1 = ±S y
- Sy Sy σ
1

As long as the stress

state falls within the
polygon, the material
- Sy will not yield.
 Maximum-Shear-Stress Theory
(Tresca Criterion)
Postulate: Yielding begins whenever the maximum shear
stress in a part becomes equal to the maximum
shear stress in a tension test specimen that
begins to yield.

τ σ1 > σ 2 > σ 3 τ
τ1/3 = τ max τy
τ1/2
τ 2/3 σ1 = S y
σ3 σ2 σ1 σ σ 2 , σ3 σ

Stress State in Part Tensile Test Specimen

 Maximum-Shear-Stress Theory
(Continued)
Tensile Test Specimen

τ
τ max = Ss
Ss = 0.5Sy
σ1 = S y

The shear yield strength σ 2 , σ3 σ

is equal to one-half of the
tension yield strength.
 Maximum-Shear-Stress Theory
(Continued)

Stress State in Part

τ
τ1/3 = τ max σ1 − σ 2
τ1/2 τ1/2 =
2
τ 2/3
σ 2 − σ3
σ3 σ2 σ1 σ
τ 2/3 =
2
σ1 − σ 3
τ1/3 = τ max =
2
σ1 > σ 2 > σ 3
 Maximum-Shear-Stress Theory
(Continued)

Sy From Mohr’s circle for a

Ss = tensile test specimen
2
σ1 − σ 3 From Mohr’s circle for a three-
τ1/3 = τ max = dimensional stress state.
2

Sy = σ1 − σ 3
 Maximum-Shear-Stress Theory
(Hydrostatic Effect)
Principal stresses will always σ1d − σ d2
have a hydrostatic component τ1/2 =
(equal pressure) 2
σ1 = σ d1 + σ h σ d2 − σ3d
τ 2/3 =
2
σ 2 = σ d2 + σ h
σ1 − σ3
d d

σ 3 = σ d3 + σ h τ1/3 =
2
The maximum shear
σ h = 1 I1 = 1 (σ1 + σ 2 + σ 3 ) stresses are independent of
3 3 the hydrostatic stress.
d => deviatoric component
h => hydrostatic
 Maximum-Shear-Stress Theory
(Hydrostatic Effect – Continued)

Hydrostatic Stress State

If σ1d = σ d2 = σ 3d
The Maximum-Shear-Stress
Then τ max = 0, and there Theory postulates that
yielding is independent of a
is no yielding regardless hydrostatic stress.
of the magintude of the
hydrostati c stress.
 Maximum-Shear-Stress Theory
(Biaxial Representation of the Yield Surface)

Yielding will occur if For biaxial case

any of the following (plane stress)
criteria are met. σ3 = 0

± S y = σ1 − σ 2 ± Sy = σ1 − σ 2
± Sy = σ 2 − σ 3 ± Sy = σ 2
± S y = σ1 − σ 3 ± Sy = σ1

In general, all three conditions must be checked.

 Maximum-Shear-Stress Theory
(Biaxial Representation of the Yield Surface)

σ2 locus of
For biaxial case failure states
(plane stress) Sy
σ3 = 0 II
I
- Sy
± Sy = σ1 − σ 2
Sy σ1
± Sy = σ 2 III
IV
± Sy = σ1 - Sy

Note that in the I and III quadrants the Maximum-Shear-

Stress Theory and Maximum-Normal-Stress Theory are
the same for the biaxial case.
 Maximum-Shear-Stress Theory
(Three-dimensional Representation of the Yield Surface)

failure surface

Hamrock, Fig. 6.9

 Assignment
Failure Theories, Read Section 5-9.

(a) Find the bending and transverse shear stress at points A

and B in the figure. (b) Find the maximum normal stress
and maximum shear stress at both points. (c) For a yield
point of 50,000 psi, find the factor of safety based on the
maximum normal stress theory and the maximum shear
stress theory.
Steady Load Failure Theories
(Distortion Energy Theory)

Lecture 6

Engineering 473
Machine Design
 Distortion-Energy Theory

Postulate: Yielding will occur when the distortion-

energy per unit volume equals the distortion-
energy per unit volume in a uniaxial tension
specimen stressed to its yield strength.
 Strain Energy
Strain Energy σ
The strain energy per unit
volume is given by the equation σi

1 1 1
U = σ1ε1 + σ 2 ε 2 + σ 3ε 3 U
2 2 2
εi ε
Units
The strain energy in a tensile
[ ] [
[U ] = lb in 2 ⋅ [in in ] = lb − in in 3 ] test specimen is the area under
the stress-strain curve.
 Strain Energy
(Elastic Stress-Strain Relationship)

An expression for the strain energy per unit

volume in terms of stress only can be obtained
by making use of the stress-strain relationship

Algebraic Format Matrix Format

1
ε1 = (σ1 − νσ 2 − νσ3 )
E ì ε1 ü é 1 − ν − ν ù ì σ1 ü
ï ï 1ê ú ï ï
1
ε 2 = (σ 2 − νσ1 − νσ 3 ) íε 2 ý = ê− ν 1 − ν ú íσ 2 ý
E ïε ï E ê − ν − ν 1 ú ïσ ï
î 3þ ë ûî 3 þ
1
ε 3 = (σ 3 − νσ1 − νσ 2 )
E
 Strain Energy
(Stress Form of Equation)

1 1 1
U = σ1ε1 + σ 2ε 2 + σ 3ε 3
2 2 2
1 ì1 ü
= σ1 í (σ1 − νσ 2 − νσ3 )ý
2 îE þ
1 ì1 ü
+ σ 2 í (σ 2 − νσ1 − νσ3 )ý
2 îE þ
1 ì1 ü
+ σ3 í (σ 3 − νσ1 − νσ 2 )ý
2 îE þ

U=
1 2
2E
[ ]
σ1 + σ 22 + σ32 − 2ν(σ1σ 2 + σ 2σ3 + σ3σ1 )
 Distortion and Hydrostatic
Contributions to Stress State
σ2 σh σ2 − σh

σ1 σh σ1 − σ h
= +
σ3 σh σ3 − σ h
σ1 + σ 2 + σ 3
Principal Stresses σh = Distortional Stresses
Acting on Principal 3
Planes Hydrostatic Stress
The distortional stress components are often called the
deviatoric stress components.
 Physical Significance
(Hydrostatic Component)

σh The hydrostatic stress causes a

change in the volume.

σh The cube gets bigger in tension,

smaller in compression.

σh σ h = Ke
K ≡ Bulk Modulus
σ1 + σ 2 + σ 3
σh = e ≡ volumetric strain
3
 Physical Significance
(Distortional Stresses)

These unequal stresses act to

σ2 − σh deform or distort the material
element.

σ1 − σ h There is no change in volume,

but there is a change in shape.

σ3 − σ h These stresses try to elongate

or compress the material more
in one direction than in
another.
 Strain Energy Associated with the
Hydrostatic Stress

U=
1 2
2E
[
σ1 + σ2 + σ3 − 2ν(σ1σ 2 + σ 2σ3 + σ3σ1 )
2 2
]
Uh =
1 2
2E
[ ]
σ h + σ 2h + σ2h − 2ν(σ h σ h + σ h σ h + σ h σ h )

=
1
2E
[ ]
3σ 2h − 6ν ⋅ σ 2h
This term is equal to the strain
3 (1 - 2ν ) 2 energy per unit volume from the
Uh = σh
2 E hydrostatic stress components.
 Distortional Strain Energy
The distortional strain energy is equal to the difference between
the total strain energy and the hydrostatic strain energy.

Ud = U − Uh

=
1 2
2E
[ ]
σ1 + σ 22 + σ32 − 2ν(σ1σ 2 + σ 2σ3 + σ3σ1 )

3 (1 − 2ν ) (σ1 + σ 2 + σ3 )
2
−
2 E 9
=
1 2
2E
[ ]
σ1 + σ 22 + σ32 − 2ν(σ1σ 2 + σ 2σ3 + σ3σ1 )

æ σ12 + σ1σ 2 + σ1σ3 ö

1 (1 − 2ν ) ç 2 ÷
− ç + σ 2 + σ1σ 2 + σ 2σ3 ÷
2 3E ç 2 ÷
ç + σ3 + σ1σ 3 + σ 2σ3 ÷
è ø
 Distortional Strain Energy
(Continued)
Ud = U − Uh

=
1 2
2E
[ ]
σ1 + σ 22 + σ 32 − 2ν(σ1σ 2 + σ 2σ3 + σ3σ1 )

1 (1 − 2ν ) 2
−
2 3E
( )
σ1 + σ 22 + σ32 + 2(σ1σ 2 + σ 2σ3 + σ3σ1 )

Ud =
1+ ν 2
3E
[
σ1 + σ 22 + σ 32 − σ1σ 2 − σ 2 σ 3 − σ 3σ1 ]
Distortional Strain Energy in Tension
Test Specimen
Postulate: Yielding will occur when the distortion-
energy per unit volume equals the distortion-
energy per unit volume in a uniaxial tension
specimen stressed to its yield strength.

Ud =
1+ ν 2
3E
[
σ1 + σ 22 + σ 32 − σ1σ 2 − σ 2 σ 3 − σ 3σ1 ]
1+ ν 2
Ud = Sy
3E

Hamrock, Fig. 3.1

 Distortion Energy Failure Theory
Equating the distortional strain energy at the point under
consideration to the distortional strain energy in the tensile
test specimen at the yield point yields

Ud =
1+ ν 2
3E
[
σ1 + σ 22 + σ 32 − σ1σ 2 − σ 2 σ 3 − σ 3σ1 ]
1+ ν 2
= Sy
3E

S2y = σ12 + σ 22 + σ 32 − σ1σ 2 − σ 2 σ 3 − σ 3σ1

σ eff = Sy

σ eff = σ12 + σ 22 + σ 32 − σ1σ 2 − σ 2 σ 3 − σ 3σ1

Alternate Forms of Effective Stress

σ eff = σ12 + σ 22 + σ 32 − σ1σ 2 − σ 2 σ 3 − σ 3σ1 Form 1

σ eff =
(σ1 − σ 2 )2 + (σ 2 − σ 3 )2 + (σ 3 − σ1 )2 Form 2
2

The effective stress is commonly referred

to as the von Mises stress, after Dr. R. von Mises
who contributed to the theory.
 Plane Stress Condition

σ3 = 0 σ2
Sy
σ eff = σ12 + σ 22 − σ1σ 2
- Sy

σ eff =
(σ1 − σ 2 )2 + σ 22 + σ12 Sy σ1
2
- Sy

• As long as the stress state falls within the shaded area,

the material will not yield.

• The surface, blue line, at which the material just begins

to yield is called the yield surface.
 Pure Shear Condition
σ 3 = −σ1
σ3
τ1,3 Sy

σ2 - Sy
σ3 σ1
45° S σ1
y

- Sy
Mohr’s Circle
σ eff = σ12 + σ 32 − σ1σ 3
for Pure Shear
= 3σ12 = 3τ 2max = Sy

This is an important result. τ max = 0.577 ⋅ Sy = Sys

Yield Surface in 3-D Stress State

Hamrock, Fig. 6.9

 Other Names for Distortion
Energy Theory

σ eff =
(σ1 − σ 2 )2 + (σ 2 − σ 3 )2 + (σ 3 − σ1 )2
2
τ
τ1/3
People came up with the same τ1/2
equation using different starting τ 2/3
points.
σ3 σ2 σ1 σ
•Shear Energy Theory
•Von Mises-Hencky Theory
•Octahedral-Shear-Stress Theory
σ1 > σ 2 > σ 3
 Assignment
• Show that the two forms of the equation for the effective stress
are equal.
• Show that the effective stress for a hydrostatic stress state is
zero.
• Compute the effective stress at the critical location in the
stepped shaft loaded in tension (previous assignment). The
yield strength of the material is 30 ksi. Will the material yield
at the critical location?

σ eff = σ12 + σ 22 + σ 32 − σ1σ 2 − σ 2 σ 3 − σ 3σ1

σ eff =
(σ1 − σ 2 )2 + (σ 2 − σ 3 )2 + (σ 3 − σ1 )2
2
 Assignment
(Continued)
In the rear wheel suspension of the Volkswagen “Beetle” the spring motion was
provided by a torsion bar fastened to an arm on which the wheel was mounted.
See the figure for more details. The torque in the torsion bar was created by a
2500-N force acting on the wheel from the ground through a 300-mm lever
arm. Because of space limitations, the bearing holding the torsion bar was
situated 100-mm from the wheel shaft. The diameter of the torsion bar was 28-
mm. Find the von Mises stress in the torsion bar at the bearing.

Hamrock, Fig. 6.12

Steady Load Failure Theories –
Comparison with Experimental
Data

Lecture 7

Engineering 473
Machine Design
 Important Historical Studies of
Failure Theories

1864 Tresca developed Maximum Shear Stress

Theory while measuring loads required to
extrude metal through dies of various shapes.
1928 von Mises publishes the Maximum Distortion
Energy Theory
1926 Lode publishes comparison of Tresca and von
Mises Theories
1931 Repeat Lode experiments with better technique
 Experimental Test Specimen

Thinned walled cylinder loaded with an internal

pressure, axial force, and a torsional moment.

Mendelson, Fig. 6.1.1

Lode’s Data

Mendelson, Fig. 6.4.1

Taylor and Quinney Data

Mendelson, Fig. 6.4.3

Additional Test Results

Hamrock, Fig. 6-17

More Test Results

Dowling, Fig. 7-11

 Conclusions

• Both the Distortion Energy Theory and the Maximum

Shear Stress Theory provide reasonable estimates for the
onset of yielding in the case of static loading of ductile,
homogeneous, isotropic materials whose compressive and
tensile strengths are approximately the same.

• Both the Distortion Energy Theory and the Maximum

Shear Stress Theory predict that the onset of yield is
independent of the hydrostatic stress. This agrees
reasonably well with experimental data for moderate
hydrostatic pressures.
 Conclusions
(Continued)

• Both the Distortion Energy Theory and the Maximum

Shear Stress Theory under predict the strength of brittle
materials loaded in compression. Brittle materials often have
much higher compressive strengths than tensile strengths.

• The Distortion Energy Theory is slightly more accurate than

the Maximum Shear Stress Theory.Theory The Distortion Energy
Theory is the yield criteria most often used in the study of
classical plasticity. Its continuous nature makes it more
mathematically amenable.
 Industry Standards and Codes
• The American Society of Mechanical Engineers base the ASME Boiler
and Pressure Vessel Code on the Maximum Shear Stress Theory.

• The American Institute of Steel Construction does not use either in the
Manual of Steel Construction. Buildings, bridges, etc. are dominated
by normal stresses and buckling type failures.

• The American Society of Civil Engineers use the Distortion Energy

Theory in Design of Steel Transmission Pole Structures.

• There is no single standard that applies to the design of machine

components. Standard industry practice is to use either the Distortion
Energy Theory or Maximum Shear Stress Theory with an appropriate
safety factor.
 Failure Versus Yielding
• The high stresses around stress concentration factors are often
very localized, and the local yielding will cause a
redistribution of stresses to adjacent material. In many cases
the local yielding will not cause a machine component to fail
under steady load conditions.

• It is common to differentiate between local yielding and gross

yielding through the thickness of a member.

• Local yielding may lead to early fatigue failure, and stress

concentration effects must always be considered in fatigue
calculations.
 Internal Friction Theory
B τ S
s
D

σ
Sc St

Postulate: For any stress state that creates a Mohr’s circle

that is tangent to the line between points B&D,
the stresses and strengths are related by the
equation

σ1 σ 3
− = 1, where σ1 > σ 2 > σ 3 .
S t Sc
 Comparison with Maximum Shear
Stress Theory
Internal Friction Theory
σ1 σ 3
− = 1, where σ1 > σ 2 > σ 3 . Note that the IFT is a
S t Sc generalization of the
MSST. The MSST is
Maximum Shear Stress Theory limited to materials in
S t = Sc which the tensile and
compressive yield
σ1 σ 3 strengths are
− = 1, where σ1 > σ 2 > σ 3 ,
S t Sc approximately equal.
σ1 − σ 3
=1
Sy
 Plane Stress Condition

σ3 σ2 = 0
Sut

Suc IFT
σ1
Sut MSST

Suc

Whenever the stress state is within the

polygon, the material will not fail.
Comparison with Test Data

Colomb-Mohr
Theory is the IFT

Shigley, Fig. 6-28

Brittle Material Failure Summary

• Brittle materials typically have significantly different

compressive and tensile strengths.

• The Internal Friction Theory or Modified Internal

Friction Theory may be used to estimate the failure
state.

• For some materials the Modified Internal Friction

Theory may provide a slightly more accurate
estimate.
 Safety Factors
N = Safety Factor
σ eff 1
σ3
= DET
Sy N
σ1 σ 3 1
= = 1st Quadrant IFT
St S t N σ1
σ1 σ 1
= 3 = 3rd Quadrant IFT
- Sc - Sc N
σ1 σ 3 1
− − = 2nd Quadrant IFT
S t Sc N
σ1 σ 3 1
− = 4th Quadrant IFT
S t Sc N Reduced area of allowable
stress states.
 Design Margins
σ eff 1
= • For a stress state to be
Sy N acceptable, the margin must be
positive.

S y - σ eff N = 0 • A negative margin indicates

that the design objective hasn’t
been met.
S y − σ eff N
Margin ≡ M =
Sy • Provides a measure of how
close a stress state is to the
design maximum.
σ eff N
M = 1− • Design Margins are reported for
Sy
all NASA projects.
 Assignment
A hot-rolled bar has a minimum yield strength in tension
and compression of 44 kpsi. Find the factors of safety for
the MSST and DET failure theories for the following stress
states.
(a ) σ xx = 9 kpsi, σ yy = −5 kpsi
(b ) σ xx = 12 kpsi, τ xy = 3 kpsi ccw
(c) σ xx = −4 kpsi, σ yy = -9 kpsi, τ xy = 5 kpsi cw
(d ) σ xx = 11 kpsi, σ yy = 4 kpsi, τ xy = 1 kpsi cw
 Assignment
(Continued)
This problem illustrates that the
factor of safety for a machine
element depends on the particular
point selected for analysis. You are
to compute factors of safety, based
upon the distortion-energy theory, for
stress elements at A and B of the
member shown in the figure. The bar
is made of AISI 1020 cold-drawn
steel and is loaded by the forces
F=0.55 kN, P=8.0 kN, and T=30 Nm.

Shigley, Problem 6-6

 Assignment
(Continued)
The figure shows a crank loaded by a force F=300 lb which causes
twisting and bending of the 0.75 in diameter shaft fixed to a support
at the origin of the reference system. The material is hot-rolled
AISI 1020 steel. Using the maximum-shear-stress theory, find the
factor of safety based on the stress state at point A.

Shigley, Problem 6-8

Introduction to Fracture
Mechanics
Lecture 8

Engineering 473
Machine Design
 Fracture Mechanics
“…every structure contains small flaws whose size and
distribution are dependent upon the material and its
processing. These may vary from nonmetallic inclusions
and micro voids to weld defects, grinding cracks, quench
cracks, surface laps, etc.”

The objective of a Fracture Mechanics analysis is to

determine if these small flaws will grow into large enough
cracks to cause the component to fail catastrophically.

T.J. Dolan, Preclude Failure: A Philosophy for

Material Selection and Simulated Service Testing,
SESA J. Exp. Mech., Jan. 1970.
 WW II Tanker Failure

Small cracks and defects can lead to catastrophic

failure of large structural systems.
Norton, Fig. 5-13
Rocket Case Failure

Norton, Fig. 5-14

 Stress State at Plane
Crack Tip

K æ θ öé æ θ ö æ 3θ öù
σx = cosç ÷ ê1 − sin ç ÷ sin ç ÷ú + L
2π ⋅ r è 2 øë è 2 ø è 2 øû
K æ θ öé æ θ ö æ 3θ öù
σy = cosç ÷ ê1 + sin ç ÷ sin ç ÷ú + L
2π ⋅ r è 2 øë è 2 ø è 2 øû
K æ θ ö æ θ ö æ 3θ ö
τ xy = cosç ÷ sin ç ÷ sin ç ÷ + L
2π ⋅ r è 2ø è 2ø è 2 ø
σ z = 0 (Plane Stress)
σ z = ν (σ x + σ y ) (Plane Strain)
τ yz = τ zx = 0

Norton, Fig. 5-15

 Stress Intensity Factor
K ≡ Stress Intensity Factor

K = σ nom π ⋅ a
for
a << b

σ nom ≡ Stress in the absence of the crack

MPa ksi
[k] = or
m in

Norton, Fig. 5-15

Crack Tip Plastic Zone

Norton, Fig. 5-16

 Experimental Examples

Felbeck, D.K., A.G. Atkins, Strength and Fracture of

Engineering Solids, Prentice-Hall, 1984, Fig. 14-17. www.stressphotonics.com
 Crack Displacement Modes

Mode I Mode II Mode III

Opening Sliding Tearing

Hamrock, Fig. 6.8

 Fracture Toughness
K ≡ Stress Intensity Factor
As long as the stress intensity
factor K stays below a critical value
K = σ nom π ⋅ a called the fracture toughness, Kc,
the crack is considered stable.
for
a << b If K reaches K c , the crack
will propagate and lead to
σ nom ≡ Stress in the sudden failure. Propagation
absence of the crack rates can reach 1 mile/sec.

Fracture toughness is a
MPa ksi material property.
[k] = or
m in
Brittle to Ductile Transition
Temperature
Low temperatures and
high strain rates
generally promote brittle
behavior (i.e. low
fracture toughness).

Felbeck, Fig. 14-4

Transition Temperature
Examples

Felbeck, Fig. 14-5

 Temperature Sensitivity of KIC

Sailors, R.H., H.T. Corten, “Relationship Between Material Fracture Toughness Using Fracture
Mechanics & Transition Temperature Tests, Stress Analysis and Growth of Cracks,” ASTM
STP514, Am. Society of Testing Materials, 1972.
Comparison with Charpy
V-Notch Test Data

Sailors, R.H., H.T. Corten,

“Relationship Between Material
Fracture Toughness Using Fracture
Mechanics & Transition Temperature
Tests, Stress Analysis and Growth of
Cracks,” ASTM STP514, Am. Society of
Testing Materials, 1972.
Stress Intensity Factors for Different
Crack Geometries
Relationships between KIC
and other crack geometries
and loading conditions may
be found in text books and
industry publications.

K o = σ nom π ⋅ a

Shigley contains several

examples.

Shigley, Fig. 5-22

 Yield Failure Before Fracture
K IC = σ nom π ⋅ a 2024 Aluminum

K IC = Sy π ⋅ a K IC = 26 MPa m
Sy = 455 MPa
2
1 æç K IC ö÷
a=
π çè Sy ÷ø a = 0.001 m
= 1 mm = 0.04 inch

The cross section will yield before unstable

fracture for any crack less than 2 mm in
total length.
 Assignment

It is determined that a high strength alloy plate

has a ½ inch long through crack running normal
to the direction of loading. Material tests
indicate that the Mode I fracture toughness, KIC,
is 80 ksi/in1/2. A stress analysis indicates that the
plate will experience a steady stress of 100 ksi.
Will the plate experience unstable crack
propagation.
Fracture Mechanics and Steady
Load Failure Theory Summary

Lecture 9

Engineering 473
Machine Design
 Critical Crack Size

For a given crack size, there is a corresponding

stress that will cause the crack to propagate in a
catastrophic manner.
 Non-destructive Testing

Testing methods exist that can detect cracks or flaws

in metallic parts without destroying them. These
methods are called non-destructive testing (NDT).

If the flaw size can be established in a part through NDT,

and the stress state at the location of the crack is known
through analysis or test, then an analysis can be
performed to determine if the crack is close to the critical
crack size for the particular stress state.

The combination of analysis to determine the stress state

and NDT to establish the maximum flaw size are critical
components of fracture prevention programs.
 Fracture Mechanics Cases
(NDT Inspected Part)

Case 1:
1 The machine element is inspected and no
cracks are found.

All Nondestructive Testing (NDT) methods have a

minimum crack size that can be detected. In this case, the
crack length is taken to be the minimum detectable crack.

K IC
σf =
Y π⋅a

Crack geometry Minimum

factor detectable
crack length
 Fracture Mechanics Cases
(Part has been tested)

Case 2:
2 The part is tested and does not fail under a
known load.

In this case, the crack size is assumed to be slightly

smaller than the critical crack size associated with the
stress state caused by the test load.

2
1 æ K IC ö
a = çç ÷÷
π è Yσ f ø
Possible crack size
Stress caused by the test load
 Fracture Mechanics Cases
(Crack is detected)

Case 3:
3 The part is inspected and a crack is found.

The size of the crack is compared to the critical crack size

obtained from the following formula. The stress used is
that to be encountered during service.

2
1 æ K IC ö
a crit = ç ÷
π è Yσ ø
Expected service stress
 Stress-Corrosion Cracking
Parts subjected to continuous static loads in certain corrosive
environments may, over a period of time, develop cracks.

This plot shows a reduction in KIC

over time due to stress-corrosion.

Shigley, Fig. 5-27

 Non-destructive Testing
NDT is the examination of engineering materials with
technologies that do not affect the object’s future usefulness.

Common NDT Methods

X-radiography Magnetic particle
Ultrasonic Liquid penetrant
Eddy current Acoustic emission
X-radiography

Shackelford, Fig. 8-22.

Ultrasonic Testing

Schakelford, Fig. 8-23.

 Summary of Steady Load
Failure Theories

Ductile Materials Brittle Materials Fracture Mechanics

Distortion Energy Maximum Normal Linear Elastic
(von Mises) Stress Fracture Mechanics
(LEFM)
Maximum Shear Stress Internal Friction
(Tresca) (Coulomb-Mohr)

Modified Internal
Friction
 When do I apply these
failure theories?
Design Governed Design Not Governed by
by Industry Industry Design
Design Standard Standard
• Follow formulas in standard. • Choose a factor of safety
that the design is to be
• Formulas can often be derived based on.
based on a knowledge of the
failure theory incorporated in • Use appropriate failure
the standard. theory during the design of
machine elements.
• Factor of safety is included in
the standard. • Compute failure margins at
all critical locations.
Flow Chart for
Typical Analysis

Norton, Fig. 5-22

 Material Failure Mechanisms
Ductile fracture – failure that involves a significant amount of
plastic deformation prior to fracture
Brittle fracture – failure without a significant amount of
macroscopic plastic deformation prior to
fracture.
Fatigue failure – failure associated with slow crack growth due
to changing stress states.
Corrosion-fatigue failure – failure due the combined actions of
changing stress and corrosive environments.
Stress-corrosion cracking – failure in which a steady tensile stress
leads to the initiation and propagation of
fracture in a relatively mild chemical
environment.
 Material Failure Mechanisms
(Continued)

Wear failure – broad range of relatively complex, surface-

related damage phenomena.
Liquid-erosion failure – type of wear failure in which liquid is
responsible for removal of material.
Liquid-metal embrittlement – involves the material losing some
degree of ductility below its yield strength due
to its surface being wetted by a lower-melting-
point liquid metal.
Hydrogen embrittlement – notorious cause of catastrophic failure
in high strength steels exposed to hydrogen
environment which leads to lose of ductility
(few parts per million of hydrogen is enough).
 Material Failure Mechanisms
Creep and stress rupture failures – failure due to
continued strain growth under steady
load.

All of these mechanisms are associated with the

failure of the material. They do not include one
of the most important structural failure
mechanisms that must be considered in
compressive stress environments – Buckling.
Buckling
 Assignment
A high - strength steel has a yield strength of 1,460 Mpa and
a K IC of 98 Mpa m . Calculate the size of a surface crack that
will lead to catastrophic failure at an applied stress of 0.5 Sy .

An NDT inspection is used that can ensure that a structural

ceramic part will have no flaws greater that 25 µm in size.
Calculate the maximum service stress that can be used.
K IC = 9 Mpa m .