BOLTED JOINT DESIGN AND ANALYSIS

Primarily from:
Bickford, John H., An Introduction to the Design and Behavior of Bolted Joints, Marcel
Dekker, Inc., NY, 1990.
Adapted to Mathcad by Dan Frey.
NOTE: in many instances, a single variable can be calculated by one of several equations.
Read the acccompanying text, "toggle" on the appropriate equation, and "toggle" off the
alternative equations using the Math menu. A box will appear to the right of equations that are
off.
This Mathcad sheet is one integrated model. If you cut out some portions to paste them
elsewhere, be aware of which variables have been defined in other parts of the sheet.
Table of Contents
Preliminary Definitions
Derived Units
Material Properties
Behavior of Bolt Material
Bolt Terminology and Notation
Define Bolt Geometry
Strength
Stress Area of Bolts
Strength of Bolts
Strength of the Threads
Stiffness
Stiffness of the Bolt
Stiffness of the Joint Material
Total Joint Stiffness
Slocum's Approach
Tightening Bolted Joints
Torque to set Preload
Turn to set Preload
Behavior of Joints in Service
Externally Applied Forces
The Joint Diagram
Thermal Expansion
Stress Corrosion Cracking
Strain at failure
f
12 % :
Stress at failure
f
70 ksi :
check on the last three Strain at peak stress
p
6 % :
Ultimate strength
u
74 ksi :

y
E
:
Yield strength

y
57 ksi :
E E
steel
:
e.g. J429 grade 2 steel (pg.54) Define behavior of bolt material:

alum
0.3 : Young's modulus of aluminum E
alum
10 10
6
psi :

steel
0.3 : Young's modulus of steel E
steel
30 10
6
psi :
Tabulate some common material properties:
mm 10
3
m : MPa 10
6
Pa : ksi 1000 psi : psi
lbf
in
2
:
Define some useful derived units
Preliminary Definitions
The function below is a piece wise linear approximation of the stress strain curve as defined
above.
vx 0
y

y

p

y

4
+
p

p

f

p

2
+
f

,
T
:
vy 0 ksi
y

y

u

y

1.5
+
u

f

u

f

1.5
+
f

,
T
:
( ) linterp vx vy , , ( ) : 0

f
20
,
f
.. :
0 0.05 0.1
0
2
.
10
8
4
.
10
8
( )

This plot shows the

stress/strain behavior of
the bolt material.
Bolt Terminology / Notation
L
body
L
thread
L
head
L
grip
L
nut
D
E
Smin
Lgrip -- Grip length
Lnut -- Length of the nut
Esmin -- Minimum pitch diameter of the threads
D -- Nominal diameter of the threads
Lhead -- Lenngth of the head of the bolt
Lbody -- Length of the unthreaded portion of the shank
Lthread -- Length of the threaded portion of the bolt
L
grip
L
body
1
2
L
thread
+ :
L
body
50 mm :
L
nut
24 mm : L
head
24 mm : L
thread
50 mm :
A
nut
A
head
: A
head
2 A
body
: A
body

4
D
2
:
n 152.4 ft
-1
n
1
p
:
Threads per inch
p 2 mm : Thread pitch
D 16 mm :
Nominal Diameter
Define Bolt Geometry
Strength
Stress Area of Bolts
The area of the threaded portion of the bolt that sees the stress (sometimes called the
stress area) is critical. There are several ways to calculate it. Select the appropriate
equation below and ensure that it is the only active equation (use "Toggle Equation" in the
Math menu).
English Units
(Bickford, pg. 23) - Based on the mean of the
pitch and root diameters for 60 degree threads.
A
s
0.785 D
0.985
n

,
2
:
E
Smin
.89 in : Minimum pitch diameter of the threads.
A
s

E
Smin
2
0.16238
n

,
: Recommended for bolt materials with yeild strengths
>100,000psi. Use the minimum pitch diameter of the
threads.
A
s
0.7854 D
1.3
n

,
2
: Root area. More Conservative.
A
s
0.7854 D
1.3
n

,
2
: ASME Boiler and pressure vessel code.
Metric
p 2 mm Pitch of the threads
(Bickford, pg. 25) - For metric threads.
A
s
0.7854 D 0.938 p ( )
2
:
E
Smin
10 mm : Minimum pitch diameter of the threads.
A
s
0.7854 E
Smin
0.268867 p ( )
2
: Recommended for bolt materials with yeild strengths
>100,000psi. Use the minimum pitch diameter of the
threads.
A
s
0.7854 D 1.22687 p ( )
2
: Root area. More Conservative.
Strength of the Bolt
Define the approximate stress distribution in a bolt
F
p
50 lbf : Tension in the bolt
vx
0 in
L
head
L
head
L
body
+
L
head
L
body
+ ( ) 1.01
L
head
L
grip
+
L
head
L
grip
+ L
nut
+
L
head
L
body
+ L
thread
+

1
1
1
1
1
1
1
1
1
1
]
: vy
0 psi
F
p
A
body
F
p
A
body
F
p
A
s
F
p
A
s
0 psi
0 psi

,
:
bolt
l ( ) linterp vx vy , l , ( ) :
Graph the stress distribution in a bolt.
l 0 in
L
head
5
, L
head
L
body
+ L
thread
+ .. :
This is an approximate stress distribution across the length of a bolt. Left is the
head of the bolt, right is the nut. The most highly stressed portion is the
threaded area of the bolt under tension. This area therefore determines the
strength of the bolt. The actual stress distribution is more complex.
0 0.1 0.2 0.3 0.4
0
5
.
10
5
1
.
10
6

bolt
l ( )
l
The force (F) that a bolt can support before the shank (as opposed to the threads)
fails is:
F
ult

u
A
s
: F
ult
1.772 10
4
lbf
F
y

y
A
s
: F
y
1.365 10
4
lbf
L
e
D
5.954
Which is a fairly typical ratio of thread engagement length and nominal
diameter.
L
e
3.751 in
A
s
0.24 in in
Where As is the stress area (computed in the bolt strength setion)
L
e
2 A
s

n K
nmax

1
2 n
0.57735 E
Smin
K
nmax
( ) +

1
1
]

:
Rearranging, one can calculate the minimum thread engagement required to ensure that
the bolt fails rather than the threads.
A
s
0.24 in
2
A
TS
0.026 in
2

A
TS
n L
e
K
nmax

1
2 n
0.57735 E
Smin
K
nmax
( ) +

1
1
]
:
Shear Area
Threads per inch n 20 in
1
:
Minimum PD of bolt E
Smin
0.2 in :
Maximum ID of nut K
nmax
0.257 in :
This section continues to use a
1/4-20 bolt as an example.
Length of thread engagement L
e
.2 in :
Bolt threads typically fail at the root. The total cross sectional area at that point is needed
for bolt strength calculations.
Nut material stronger than the bolt material
Strength of Threads
L
e
2.582 in
A
s
0.24 in in
Where As is the stress area (computed in the bolt strength setion)
L
e
S
st
2 A
s
( )
S
nt
n K
nmax

1
2 n
0.57735 D
Smin
K
nmax
( ) +

1
1
]

:
Rearranging, one can calculate the minimum thread engagement required to ensure that
the bolt fails rather than the nut threads.
A
s
0.24 in
2
A
TS
0.072 in
2

A
TS
n L
e
D
Smin

1
2 n
0.57735 D
Smin
E
nmax
( ) +

1
1
]
:
Shear Area
Tensile strength of nut material S
nt

u
2
:
Tensile strength of bolt material S
st

u
:
Threads per inch n 10 in
1
:
Minimum OD of bolt threads D
Smin
0.25 in :
Maximum PD of nut E
nmax
0.257 in :
This section continues to use a
1/4-20 bolt as an example.
Length of thread engagement L
e
.2 in :
If threads are tapped into a weak material (Aluminum, plastic), the nut threads may fail first
even though the shear area is greater.
Nut material weaker than the bolt material
This is not much thread engagement. It is advisable to add a factor of
safety to this figure (in my opinion).
L
e
D
1.22
L
e
0.305 in
L
e
A
s
D
:
Slightly greater than for the case of nut material weaker. A
TS
0.079 in
2

A
TS
D
L
e
2
:
Length of thread engagement L
e
.2 in :
Nominal bolt diameter D
1
4
in :
Nut and bolt the same strength material
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0
1
2
3
4
5
6
7
8
9
10
Snut/Sbolt
L
e
/
D

S
nt
1
5
S
st

1.1
5
S
st
, 2 S
st
.. :
L
e
D
4.099
The weaker the nut material, the more threads must be engaged.
Stiffness
Stiffness of the Bolt
Using the stress vs length graph from the strength section above, total deflection of the
bolt under load can be estimated as:
E 3 10
7
psi
L
bolt
0
L
head
L
body
+ L
thread
+
l

bolt
l ( )
E

(
(

d :
L
bolt
2.32 10
5
in
This means that the spring constant of the bolt (and the nut) is:
K
bolt
F
p
L
bolt
: K
bolt
2.155 10
6

lbf
in

Stiffness of the Joint Material

E
joint
30 10
6
psi :
T L
grip
: Thickness of the joint material
The area of an equivalent cylinder of material that is placed in compression as the bolt is
loaded in tesion is computed below. This model assumes:
1) Elastic material behavior.
2) Concentric joint - The bolt goes through the center of the joint material.
3) The load is applied along the joint axis.
D
j
1.5 in : Outside diameter of the joint material
D
b
1.5 D : Nominal diameter of the bolt head (or washer)
D
h
1.01 D : Diameter of the hole the bolt goes through
D
j
D
b
D
h
If the thickness of the upper and lower joint layers are equal:
A
c

4
D
j
2
D
h
2

_
,
D
b
D
j
if

4
D
j
2
D
h
2

_
,

8
D
j
D
b
1

D
b
T
5
T
2
100
+

,
+ D
b
D
j
< 3 D
b
if

4
D
b
T
10
+

,
2
D
h
2

1
1
]
otherwise
:
Bickford (pg. 111) also indicates that the two last cases apply only when T<8D.
He doesn't say what to do if T>8D.
Finally, the stiffness of the joint material in compression is given by:
K
bolt
2.155 10
6

lbf
in

K
Jc
E
joint
A
c

T
: K
Jc
3.076 10
6

lbf
in

K
j
K
Jc
: If the joint is concentric.
Ecccentric joints:
The stiffness Kjc above applies only to concentric joints. For eccentric joints:
s
a
W
t
b
L
x
L
x
L
x
L
x
T
min
K
j
3.076 10
6

lbf
in
K
j
1
1
K
Jc
1
s a A
c

R
G
2
A
j

1
1
1
]
:
If the load is off center with the bolt:
K
j
1
1
K
Jc
1
s
2
A
c

R
G
2
A
j

1
1
1
]
:
If the load is on center with the bolt (ie. s=a):
A
j
b W W D
b
T
min
+ if
b D
b
T
min
+ ( ) otherwise
:
b t t D
b
T
min
+ ( ) if
D
b
T
min
+ otherwise
:
Total width of the joint W 5 in :
Thickness of the thinner joint cross section. T
min
.2 in :
Distance between bolts t 3 in :
Radius of gyration if joint is circular viewed down the bolt. R
G
D
j
2
:
Radius of gyration if joint is rectanugular viewed down the bolt. Dj is
the length of the shorter side.
R
G
0.209 D
j
:
Distance the load is off center a 2 in :
Distance the bolt is off center. s 1 in :
K
joint
1.197 10
6

lbf
in

K
joint
1
1
K
bolt
1
K
washer
+
1
K
j
+
1
K
g
+

,
:
K
washer
10 K
bolt
: K
j
3.076 10
6

lbf
in

K
g
1 10
100

lbf
in

K
bolt
2.155 10
6

lbf
in

Individual component stiffnesses behave as springs in series. Therefore they are combined
by inverse sum of inverses (as if they were resistances in parallel).
Total Joint Stiffness
If there is no gasket K
g
10
100 lbf
in
:
Compressed asbestos, 0.125 mm thick. K
g
35
MPa
mm
A
g
:
Area of the gasket viewed looking down the bolt
A
g
0.5 in
2
:
If the joint contains a gasket, the gasket stiffness may dominate the stiffness of the
joint. Gasket material stiffness values are tabulated in Bickford pp.121-2. Use these
values with care as gasket stiffness is often highly non-linear and hysteretic.
Gaskets
K
sum
1
1
K
flange_comp
1
K
flange_shear
+
1
K
bed_shear
+
1
K
bolt
+

,
:
K
bolt
K
joint
0.4
K
bolt
E
bolt
D
2

4
D
2
L
grip
+

:
E
bolt
E
steel
:
K
bed_shear
K
j
8.5
K
bed_shear
2.615 10
7

lbf
in

K
bed_shear
D E
nut

1 + ( ) ln 2 ( )
:
E
nut
E
steel
:
K
flange_shear
K
j
100.394
K
flange_shear
3.088 10
8

lbf
in

K
flange_shear
L
grip
E
joint

1 + ( ) ln 2 ( )
:
Poisson's Ratio of joint material 0.3 :
K
flange_comp
K
j
4.981
K
flange_comp
1.532 10
7

lbf
in

K
flange_comp
E
joint
D
h

ln
D
h
D
b
2 L
grip
( ) D
h
D
b
+ ( )
D
h
D
b
+ 2 L
grip
+ ( ) D
h
D
b
( )

1
]

1
1
]
:
According to Slocum, if the bolt produces a 45 deg cone of influence:
Slocum offers an alternative to the methods given above.
Slocum's Method
K
interface
5 K
sum
: Stiffness of the interface between the two joint material faces
(e.g., the bed and the rail)
A typical value. This ususally must be determined
empirically.
K
one_bolted_joint
1
1
K
interface
1
K
sum
+
:
Comparing Slocum's result to Bickford's
K
one_bolted_joint
K
joint
0.317
Which a a fairly good agreement (mostly
because the bolt stiffness is in good agreement
and tends to dominate).
For a system including a part bolted to a bed at many points:
K
part
10
6 lbf
in
:
N
bolts
8 :
K
system
1
1
K
part
N
bolts
K
one_bolted_joint
+

,
:
One can vary the number of bolts and bolt diameter to find different bolted
joint designs with the same stiffness.
You can see there is a
reasonable agreement between
the two estimates.
T
in
682.594 in lbf T
in
F
p
K
nut
D :
steel on steel K
nut
0.2 :
One can instead rely on an experimental constant, the nut factor (Knut) that combines all
the terms above. Knut values are tabulated in Bickford (pp. 141-143).
The first term is inclined plane action.
The second is thread friction.
The third is friction acting on the face of the nut.
T
in
553.205 in lbf T
in
F
p
p
2

t
r
t

cos ( )
+
n
r
n
+

,
:
Effective contact radius of the contact between the nut and
joint surface.
r
n
0.6 D :
Coefficient of friction between the face of the nut and the upper surface
of the joint (or the washer).

n
0.1 :
Half angle of the threads. 30 deg :
Effective contact radius of the threads r
t
D E
Smin
+
4
:
Coefficient of friction between the nut and the bolt threads
t
0.1 :
Thread pitch p 0.079 in
Desired preload in the bolt. Equals Fy as defined in strength section
above if tightening to yield.
F
p
F
y
:
Calculating torque (Tin) required to generate a desired preload level
Tightening Bolted Joints
Setting preload with turn angle
Often, preload can be set more accurately by controlling the number of turns rather than
input torque. Typical preload vs turn behavior is depicted in the figure below. The behavior
is soft at first as the threads embed. Then there is a linear portion of the curve. As the bolt
begins to yield, the behavior becomes non-linear again.
F
p
Turn
To estimate the needed turn angle:
F
p
1.365 10
4
lbf
K
bolt
4.785 10
5

lbf
in

K
j
3.076 10
6

lbf
in

R
F
p
360
p

K
bolt
K
j
+
K
bolt
K
j

,
:
R
8.637 10
3
deg Turn angle to apply a
preload of Fp.
F
ext
8000 lbf :
Externally applied force. Applies tension to the
base of the bolt head and nut.
F
ext
F
ext
Compression of the joint due to
preload.
OL
j
3.328 10
3
in OL
j
F
p
K
j
:
Extension of the bolt due to
preload.
OL
bolt
0.021 in OL
bolt
F
p
K
bolt
:
Preload Force F
p
1.024 10
4
lbf F
p
75 % F
y
:
K
joint
1.197 10
6

lbf
in

Stiffness of the whole joint

Stiffness of the joint materials K
j
3.076 10
6

lbf
in

Stiffness of the bolt K

bolt
4.785 10
5

lbf
in

Externally Applied Forces

Behavior of Joints in Service
F
crit
F
p
1.156
Also note that the critical load of the joint is always higher
than the bolt preload. The ratio is higher when the joint is
less stiff compared to the bolt. This can be understood better
by studying the loint diagram below.
Since external loads are significantly attenuated by this
effect, to maximize fatigue life preloads should be set high
enough to ensure that Fcrit is not exceeded in service.
F
bolt
F
ext
0.135
This will be a low number especially if the joint alone is
much stiffer than the bolt (Kj>>Kbolt).
If the applied force > Fcrit then all of the additional applied force is borne by the bolt alone.
This is critical because while there is some clamping force the ratio additional load seen by
the bolt due to applied load is:
F
crit
1.183 10
4
lbf F
crit
F
p
K
bolt
OL
j
+ :
The force at which the
clamping force goes to zero.
F L ( ) K
bolt
L L OL
bolt
L
bolt
+ if
F
p
K
j
L OL
bolt
( ) otherwise
:
F
bolt
1.077 10
3
lbf F
bolt
K
bolt
L
bolt
:
L
bolt
L
j
: L
j
F
j
K
j
:
F
j
6.923 10
3
lbf F
j
F
ext
K
j
K
j
K
bolt
+

,
:
The Joint Diagram
F
o
L ( ) K
bolt
L L OL
bolt
if
F
p
K
j
L OL
bolt
( ) otherwise
:
L 0
OL
bolt
OL
j
+
200
, OL
bolt
OL
j
+ .. :
0 5
.
10
4
0.001 0.0015 0.002
0
2
.
10
5
4
.
10
5
F L ( )
F
o
L ( )
F
crit
F
p
L
To better
understand this
diagram, see the
labeled figure
below.
This figure is helpful for understanding the behavior of bolted joints under applied loads.
Bickford explains the concept well on pp.354-360. This sheet allows one to see how
the figure applies to different bolted joint geometries.
OL
bolt
OL
j
L
bolt
L
j
F
b
F
bolt
F
j
F
p
F
j
F
T
F
p
8.359 10
3

Fraction of preload.
For steel / aluminum combination.
F
T
85.585 lbf F
T
K
bolt
K
j

K
bolt
K
j
+
L
j
L
bolt
( ) :
L
j

j
T L
grip
:
Thermal expansion of the bolt and joint
L
bolt

bolt
T L
grip
:
deg F T 10 :
Aluminum per deg F
j
13 10
6
:
Carbon steel per deg F
bolt
6 10
6
:
Tension in a bolt due to differential thermal expansion
Stress Corrosion Cracking
C 1.5 : Shape factor (1.5 for threads)

max

bolt
L
head
L
body
+
L
thread
2
+

,
: Stress at point of interest (in this case the
threaded portion of the bolt.
a .001 in : Crack depth
K
ISCC
C
max
a : Threshold stress intensity factor for SCC
KISCC is material dependant and must be tabulated. See Bickford pp. 560.
If stress exceeds KISCC, then crack growth will be accelerated by corrosion.
Recommended for bolt materials with yeild strengths
>100,000psi. Use the minimum pitch diameter of the
input torque. Typical preload vs turn behavior is depicted in the figure below. The behavior
is soft at first as the threads embed. Then there is a linear portion of the curve. As the bolt
F
ext