Sie sind auf Seite 1von 28

CHAPTER 10

Mechanical Springs
Flexibility is sometimes needed and is often provided by metal bodies with cleverly
controlled geometry
Such flexibility can be linear or nonlinear in relating deflection to load.
These devices allow controlled application of force or torque; the storing and release of
energy can be another purpose
In general, spring may be classified as:
1. Wire springs such as helical springs of round or square wire, made to resist and
deflect under tensile, compressive, and torsional loads.
2. Flat springs which includes cantilever and elliptical types, wound motor-or clocktype power springs, a flat spring washers, usually called Belleville springs.
3. Special-shaped springs

10.1 Stresses in Helical Springs

A round-wire helical compression spring loaded by the axial force F is shown in figure
10-1a.

101

Two important parameters in spring design:

1. The mean coil diameter D
2. The wire diameter d.
If the spring is cut at some point, the effect of the removed portion replaced by the net
internal reactions. See figure 10-1b.
Using the equation of equilibrium, the cut portion would contain a direct shear force F
and a torsion T=FD/2
The maximum stress in the wire may be computed by superposition of the direct shear
stress and the torsional shear stress
max =

Tr F
+
J
A

=

8 FD 4 F
+
d 3 d 2

(10-1)

D
d

(10-2)

Defined the spring index as:

C=
Equation (10-2) measures the coil curvature.
Substitute equation (10-2) into equation (10-1), we have:
= KS

8 FD
d 3

(10-3)

KS =

2C + 1
2C

(10-4)

For most springs the range for C is: 6 C 12.

Note that equation (10-3) is a general formula and is applied for both static and fatigue

102

10.2 The Curvature Effect

Equation (10-1) is based on the wire being straight. But, the wire has a curvature.
The curvature of the wire increases the stress on the inside of the spring but decreases it
only slightly on the outside.
In static load: these stresses can normally be neglected because it will be relieved by local
yielding with first application of a load.
In fatigue load: It is important to include the curvature stress because of its high
localized.
To include the curvature in equation (10-3), the factor KS is need to be modified.
Two different curvature factors known as:
Wahl factor:
KW =

4C 1 0.615
+
4C 4
C

(10-5)

4C + 2
4C 3

(10-6)

Bergstrasser factor:
KB =

The different between the two equations is less than one percent which makes equation
(10-6) is preferred to used

The curvature correction factor can now be obtained by canceling out the effect of the
direct shear form KB, thus
KC =

KB
2C (4C + 2)
=
K S (4C 3)(2C + 1)

(10-7)

= KB

8 FD
d 3

10.3 Deflection of Helical Springs

Using the strain energy method to include both the torsional and shear components, thus

103

U=

T 2l
F 2l
+
2GJ 2 AG

Substituting T = FD/2, l=DN, J=d4/32, A=d2/4, in the previous equation we have:

4 F 2 D 3 N 2 F 2 DN
U=
+
d 4G
d 2G
where N =Na = number of active coils
The total deflection y can now be calculated by:
U 8FD 3 N 4 FDN
y=
=
+ 2
F
d 4G
d G
Since, C=D/d, the deflection y becomes:
8 FD 3 N
d 4G

(10-8)

F
d 4G
=
y 8D 3 N

(10-9)

y
The spring rate can be calculated by
k=

10.4 Compression Springs

There are four standard types on helical compression springs. They are plain end, squared
end, plain-ground end, and squared-ground as illustrated in figure 10-2.

104

A spring with plain ends has a noninterrupted helicoids; the ends are the same as if a long
spring had been cut into sections.
A spring with plain ends that are squared or closed is obtained by deforming the ends to a
zero-degree helix angle.
Springs should always be both squared and ground for important applications, because a
better transfer of the load is obtained.
A spring with squared and ground ends compressed between rigid plates can be
considered to have fixed ends.
The type of end used affects the number of active coils Na and the solid height of the
spring.
Square ends effectively decrease the number of total coils Nt by approximately two:
Nt = Na+2
Table 10-1 shows how the type of end used affects the number of coils and the spring length.
Types of Springs Ends

Squared or
Closed
2

Squared and
Ground
2

Na

Na+1

Na+2

Na+2

Free length, L0

pNa+d

p(Na+1)

pNa+3d

pNa+2d

Solid length, Ls

d(Nt+1)

dNt

d(Nt+1)

dNt

Pitch, p

(L0-d)/ Na

L0/ (Na+1)

(L0-3d)/ Na

(L0-2d)/ Na

Term

Plain

Plain and ground

End coils, Ne

Total coils, Nt

Forys gives an expression for calculating the solid length of squared and ground ends
Ls=(Nt - a)d
Where a varies, with an average of 0.75 which means in this case that the entry dNt in
table 10-1may be overstated.
The way to check these variations is to take a spring and count the wire diameters in the
solid stack

105

Set removal or presetting:

A process used to induce (make) useful residual stresses.
It is done by making the spring longer than needed and then compressing it to its solid
height Ls.
This operation sets the spring to the required final free length L0 and, since the torsional
yield strength has been exceeded, induces (makes) residual stresses opposite in the
direction to those induced in service.
Set removed increases the strength of the spring and so is especially useful when the
spring is used for energy-storage purposes. But, set removal should not be used when
springs are subjects to fatigue

10.5 Stability
Compression coil springs may buckle when the deflection becomes too large.
The critical deflection is given by the equation:

C'
y cr = L0 C 1 1 22
eff

'
1

1/ 2

(10-10)

where
ycr is the deflection corresponding to the onest of instability
eff is the effective slenderness ratio and is given by:

eff =
the end condition constant

L0
D

(10-11)

C1' & C 2' are elastic constants defined by the equations:

C1' =
C 2' =

E
2(E G )

2 2 (E G )
2G + E

The end condition constant depends upon how the ends of the spring are support.

106

Table 10-2 gives values of for usual end conditions.

End condition
Spring supported between flat parallel surfaces (fixed ends)
One end supported by flat surface perpendicular to spring axis (fixed);
other end pivoted (hinged)
Both ends pivoted (hinged)
One end clamped; other end free

Constant
0.5
0.707
1
2

Absolute stability occurs when equation the term C 2' / 2eff in equation (10-10) is greater
than unity. Thus, for stability we have:
L0 <

D 2(E G )
2G + E

1/ 2

(10-12)

For steels, equation 10-12 becomes: L0 < 2.63D/

For squared and ground ends =0.5 and L0 < 5.26D
10.6 Spring Materials
A great variety of spring materials are available to the designer, including plain carbon
steels, alloy steels, and corrosion-resisting steels, as well as nonferrous materials such as
spring brass, and various nickel alloys.
Most common used steels are found in table 10-3
The UNS steel listed in appendix A should be used in designing hot-worked, heavy-coil
springs, as well as flat springs, leaf springs, and torsion bars.
Spring materials may be compared by an examination of their tensile strength; these vary
so much with wire size that they cannot be specified until the wire size is known
The graph of tensile strength versus wire diameter is almost a straight line for some
materials
This can be expressed as a linear function which can be written as:
S ut =

A
dm

(10-14)

where A and m are constants and are given in table 10-4.

Note:
If d measured in mm, then the unit of A is MPa.mmm.
If d measured in inches, then the unit of A is kpsi.inm.
107

A very rough estimate of the torsional yield strength can be obtained by assuming that the
tensile yield strength is between 60 and 90 percent of the tensile strength.
0.6 S ut < S y < 0.9 S ut
Then, the torsional yield strength can be found by Sys=0.577Sy. Therefore, for steel:
0.577(0.6) S ut < S ys < 0.577(0.9) S ut
0.35S ut < S ys < 0.52 S ut

(10-15)

For wires listed in table 10-5, the maximum allowable shear stress in a spring can be seen
in column 3.
Joerres uses the maximum allowable torsional stress for static applications and it is given
in table 10-6.
Samonov shows that Ssy=all = 0.56Sut for high-tensile spring steels, which is close to the
value given by Joerres for hardened alloy steels.
Example
A helical compression spring is made of hard drawn carbon A227 steel has a wire
diameter of 0.94mm. The outside diameter of the spring is 11.11mm. The ends are
squared and the there are 12.5 total turns.
a) Estimate the torsional yield strength of the wire.
b) Estimate the static load corresponding to the yield strength.
c) Estimate the scale of the spring (the spring constant).
d) Estimate the deflection that would be caused by the load in part (b)
e) Estimate the solid length of the spring
f) What length should the spring be to ensure that when it is compressed solid and
then released, there will be no permanent change in the free length?
g) Given the length found in part (f), is buckling a possibility?
h) What is the pitch of the body coil?
Solution
a) The torsional yield strength of the wire:
d = 0.94mm
From table 10-4, A=1783MPa.mmm and m=0.19
From equation 10-14: Sut = A/dm = 1783/(0.94)0.19 =1804MPa
From table 10-6: Ssy = 0.50Sut = 902MPa
b) The static load corresponding to the yield strength:
The main spring coil diameter is D=Dout-d=11.11-0.94=10.17mm
Thus, the spring index is: C = D/d = 10.17/0.94 =10.82
Therefore, from equation 10-3 replacing Ks and with KB and Ssy, Thus:

108

= S sy = K B
F=

8 FD
d 3

d 3 S sy
8K B D

; KB =

4C + 2 4(10.82) + 2
=
= 1.124
4C 3 4(10.82) 3

F = 25.734 N
c) the scale of the spring:
d 4G
k=
8D 3 N a
From table 10-5, d = 0.94mm = 0.037in G=80GPa
Na for both ends squared: Nt = Na +2 Na = 12.5-2=10.5 turns
d 4G
k=
= 707 N / m
8D 3 N a
d) The deflection that would be caused by the load in part (b):
F 25.734
y= =
= 0.0364m = 36.4mm
k
707
e) The solid length of the spring, from table 10-1:
Ls = d ( N t + 1) = 0.94(12.5 + 1) = 12.69mm
f) The free length of the spring suppose to have to be sure when it is compressed and
then released, there will be no permanent change in it is free length:
L0 = Ls + y = 12.69 + 36.4 = 49.09mm
g) To Check if buckling is possible:
For steel and squared ground: L0 < 5.26D L0 < 53.94mm, since L0 = 49.9mm,
therefore there will be no buckling.
h) The pitch of the body coil is:
(L 3d ) (49.09 3(0.94)
p= 0
=
= 4.41
Na
10.5

10.7 Helical Compression Spring Design for Static Service

The design of a new spring involves the following considerations:

Space into which the spring must fit and operate

Values of working forces and deflections
Accuracy and reliability needed
Tolerances and permissible variations in specifications

109

Environmental conditions such as temperature and presence of a corrosive

atmosphere
Cost and quantities needed

Designers use these factors to select a material and specify suitable values for wire size,
the number of turns, the diameter and free length, the type of ends, and the spring rate
needed to satisfy the working force-deflection requirements.
Some important limits:

The range index is 4 C 12, Lower indexes being more difficult to form (because
of the danger of surface cracking). Higher indexes tending to tangle (knot) often
enough to require individual packing.

To maintain linearity when a spring is about to close, it is necessary to avoid the

gradual touching of coils.

The characteristic of the force-deflection for the helical coil spring is nearly linear.

The spring force is not reproducible for very small deflections and near closure,
nonlinear behavior begins as the number of active turns diminishes as coils begin to
touch.
The designer confines (limits) the springs operating point to the central 75% of the
curve between no load F=0 and closure (end) F=FS

Fmax

7
FS
8

FS = (1 + )Fmax

(10-17)

FS
7
FS
1+ 8
Solve for gets:

1
7

110

Summery of the recommended design conditions:

4 C 12

(10-18)

3 N a 15

(10-19)

1
7

n s 1.2

(10-20)
(10-21)

Figure of merit:
1. The cost of wire from which the spring is wound (coiled)
2. Help in making the decision for the optimal spring design.
2 d 2 N t D
3. Formula:
fom = (RMC )
4

(10-22)

Spring design is an open-ended process

There are many decisions to be made, and may possible solution paths as well as
solutions.
One possible approach for design spring coil:
1. Make a priori decisions, with hard-drawn steel wire the first choice (relative
material cost =1)
2. Choose a wire size d
3. generate a column of parameters: d, D, OD or ID, Na, Ls, Lo, (Lo)cr, ns, and fom

By incrementing wire sizes available, a table of parameters will be generated. Then, the
design recommendation conditions are applied to choose the good design.
After wire sizes are eliminated, choose the spring design with the highest figure of merit.
This will give the optimal design despite the presence of a discrete design variable d and
aggregation of equality and inequality constraints.
The column vector of information can be generated by using the flowchart displayed in
figure 10-3.

111

112

As wound and set-removed springs, operating over a rod, or in a hole free of rod or hole.
In as-wound springs the controlling equation must be solved for the spring index as
follows:
From equation 10-3, with: =Ssy/ns, C = D/d, KB
S sy

= KB

ns
With

KB =

8 FS D
d 3

4C + 2
and FS = (1 + )Fmax
4C 3

Therefore:
S sy 4C + 2 8(1 + )Fmax C
=

4C 3
ns
d 2

Let
=

S sy
ns

& =

8(1 + )Fmax
d 2

Substituting and into the torsion equation and simplifying yields a quadratic
equations in C. The larger of the two solutions will yield the spring index
2 2
C=
+
4
4

3

4

1/ 2

(10-23)

Example
An A228 wire helical compression spring is needed to support a 20-Ibf load after being
compressed 2 in. Because of assembly considerations the solid height cannot exceed 1 in
and the free length cannot be more than 4 in. Design the spring.
Solution:
For A228 wire helical spring: From table, A = 201 kpsi-inm and m = 0.145
From table 10-5, E = 28.5Mpsi, G =11.75Mpsi (choosing d > 0.064 in)
Ends squared and ground
Function: Fmax=20 Ibf, ymax=2 in
Safety: use design factor at solid height of (ns)d = 1.2
Select robust linearity: = 0.15
Use as wound spring, Ssy=0.45Sut from table 10-6
Decision variable: d = 0.08 in.
From fig. 10-3 and table 10-6

113

S sy = 0.45
=
=

S sy
ns

A
= 130.46kpsi
dm

= 108.713kpsi

8(1 + )Fmax
= 9.15kpsi
d 2

2
2 2
3

+
C=
4
4
4

D = Cd = 0.8424
K B = (4C + 2) /( 4C 3) = 1.128

1/ 2

= 10.53

s = K B 8(1 + ) Fmax D /(d 3 ) = 108.7 kpsi

n s = S sy / s = 1.2
OD = D + d = 0.923in
N a = Gd 4 y max /(8 D 3 Fmax ) = 10.05turns
N t = 10.05 + 2 = 12.05turns
Ls = dN t = 0.964in
Lo = Ls + y s = Ls + (1 + ) y max = 3.264in

(Lo )cr

= 2.63D / = 4.43in

fom = 2.6 2 d 2 N t D / 4 = 0.417

Repeat the above analysis for other diameters and form a table to select the best spring
design:
d
0.063
0.067
0.071
0.075
0.08
0.085
0.09
0.095
0.391
0.479
0.578
0.688
0.843
1.017
1.211
1.427
D
6.205
7.153
8.143
9.178
10.53
11.96
13.46
15.02
C
0.454
0.546
0.649
0.763
0.923
1.102
1.301
1.522
OD
39.1
26.9
19.3
14.2
10.1
7.3
5.4
4.1
Na
2.587
1.936
1.513
1.219
0.964
0.790
0.668
0.581
Ls
4.887
4.236
3.813
3.519
3.264
3.090
2.968
2.881
Lo
2.06
2.52
3.04
3.62
4.43
5.35
6.37
7.51
(Lo)cr
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
ns
-0.409
-0.399
-0.398
-0.404
-0.417
-0.438
-0.467
-0.505
fom
Examine the table and perform the adequacy assessment.
The constraint 3 Na 15 cancel wire diameters less than 0.08 in
The constraint 4 C 12 cancel diameters larger than 0.085 in.
The constraint Ls > 1 in cancel diameters less than 0.080 in.
The constraint Lo > 4 in cancel diameters less than 0.071 in.
The buckling criterion cancel free length longer than (Lo)cr, which cancel diameters less
than0.075in.

114

The result is that there are only two springs in the feasible domain.
The figure of merit decides that the wire diameter is 0.08in.

10.8 Critical Frequency of Helical Springs

When helical springs are used in applications requiring a rapid reciprocating motion, the
designer must be certain that the physical dimensions of the spring are not such as to
create a natural vibratory frequency close to the frequency of the applied force; otherwise
resonance may occur, resulting in damaging stresses.
The governing equation for the translational vibration of a spring is the wave equation:
2u
W 2u
=
x 2 kgl 2 t 2

(10-24)

where
k = spring rate
g = acceleration due to gravity
l = length of spring
W= weight of spring
x = coordinate along length of spring
u = motion of any particle at distance x
Equation (10-24) has a harmonic solution and it depends on:
1. Given physical properties
2. End conditions of the spring
The natural frequencies for a spring placed between two flat and parallel plates:
kg
W

m = 1,2,3,...

1 kg
2m W

m = 1,2,3,...

= m
Since =2f, thus
f =

If m =1, it is called the fundamental frequency and it is equal to:

f =

1 kg
2 W

(10-25)

115

For spring has one end against a flat plate and other end free, the frequency is
f =

1 kg
4 W

(10-26)

W can be calculated as:

W = AL =

d 2
(DN a ) = 1 2 d 2 DN a
4
4

where is the specific weight

To avoid resonance with the harmonic it is required that the fundamental critical
frequency is 15~20 the frequency of the force or motion of the spring.
If it is not enough, the spring should be redesign to increase k or decrease W

10.9 Fatigue Loading of Helical Compression Springs

Helical springs are never used as both compression and extension springs. This is because
they are usually assembled with a preload so that the working load is additional. Thus,
the spring application fall under the condition of fluctuating loads. Thus,
Fmax Fmin
2
(10-31)
Fmax + Fmin
Fm =
= Fmin + Fa
2

Fa =

Therefore, the amplitude and midrange shear stresses respectively can be written as:
8 Fa D
d 3
8F D
m = K B m3
d
a = KB

(10-32,33)

Endurance limits for infinite life were found to be for unpeened and peened springs:
Unpeened:
Peened:

Ssa=35.0 kpsi (241MPa)

Ssa=57.5 kpsi (398MPa)

Ssm=55.0kpsi (379MPa)
Ssm=77.5kpsi (534MPa)

116

Then Sse can be calculated using Goodman theory:

S sa
S se =
where Ssu=0.67Sut
1 (S sm / S su )
Example
A helical compression spring, made of A228 wire, has a wire size of 0.092 in, an outside
diameter of 0.5625 in, a free length of 4.125 in, 21 active coils, and both ends squared
and ground. The spring is to be assembled with a preload of 5 lb and will operate to a
maximum load of 35 lb during use. Knowing that the spring is unpeened type.
1. Find the factor of safety guarding against a fatigue failure.
2. Find the critical operating frequency
Solution
Given OD=0.5625 in, d=0.092 in, Na=21
A228 spring type material, G=11.75GPsi
Fmax=35lb, Fmin=5lb
Both ends squared and ground, and unpeened.
1. The fatigue factor of safety:
D = OD d = 0.4705in
C = D / d = 5.11
K B = (4C + 2) /( 4C 3) = 1.287
Fa = (35 5) / 2 = 15lb

Fm = (35 + 5) / 2 = 20lb

8Fa D
= 29.7 kpsi
d 3
8F D
m = K B m 3 = 33.8kpsi
d
a = KB

From table 10-4, A=201kpsi, m=0.145, therefore

A
= 284.1kpsi
dm
= 0.67 S ut = 190.347 kpsi

S ut =
S su

S sa = 35kpsi
S se =

S sm = 55kpsi

S sa
= 49.22kpsi
1 (S sm / S su )

117

The factor of safety guarding against failure to be

a m
S se S su
1
+
=
nf =
= 1.28
S se S su n f
a S su + m S se
2. The critical frequency:
d 4G
k=
= 48.1lb / in
8D 3 N a
1
W = 2 d 2 DN a ; = 0.284lb / in 3
4
W = 0.0586lbf
f =

1 kg 1 48.1(386)
=
= 281Hz
0.586
2 W
2

f
f
For good design
20 f op
f op 14.1
f op
20

10.10 Helical Compression Spring Design for Fatigue Loading

Example
A helical compression spring, made of A228 wire, with infinite life is needed to resist a
dynamic load that varies from 5 to 20 lbf at 5 Hz while the end deflection varies from 0.5
to 2 in. Because of assembly considerations, the solid height cannot exceed 1.2 in and the
free length cannot be more than 4 in. The springmaker has the following wire sizes in
stock: 0.069, 0.071, 0.080, 0.085, 0.090, 0.095, 0.105, and 0.112 in.
Solution
From table 10-4 for A228: A=201 kpsi.inm, m=0.145, G =11.75Mpsi, relative cost of
wire=2.6
Surface treatment: unpeended
End treatment: squqred and ground
Select robust linearity: = 0.15 and fop = 5Hz
Fatigue safety: nf =1.5 using the Sines-Zimmerli fatigue-failure criterion
Use as wound spring, Ssy = 0.45Sut from table 10-6
Fmin = 5lbf, Fmax= 20lbf, ymin= 0.5 in, ymax = 2 in, spring operates free ( no rod or hole)
Decision variable: wire size d
For d = 0.112 in

118

20 5
20 + 5
Fm =
= 7.5lbf
= 12.5lbf
2
2
F
20
k = max =
= 10lbf / in
y max
2
Fa =

201
= 276.1kpsi
0.112 0.145
= 0.67 S ut = 185.0kpsi

S ut =
S su

S sy = 0.45S ut = 124.2kpsi
For Unpeened spring form equation (10-28): Ssa=35.0 kpsi Ssm=55.0kpsi
In Sines failure criterion, the terms Ssm ignores, thus
S sa
35
S se =
=
= 35kpsi
1 (S sm / S su ) 1 0
To find C, we replace Ssy by Sse, ns by nf, and (1+)Fmax by Fa, thus:
S
= se = 23.333kpsi
nf
=

8Fa
= 1.523kpsi
d 2

2 2
C=
+
4
4

D = Cd = 1.569in
Fs = (1 + )Fmax = 23lbf

3

4

1/ 2

= 14.005

N a = Gd 4 /(8D 3 k ) = 5.98turns
N t = 5.98 + 2 = 7.98turns
Ls = dN t = 0.894in

Lo = Ls + y s = Ls + (Fs / k ) = 3.194in
ID = D d = 1.457in
OD = D + d = 1.681in
y s = Lo Ls = 2.3in

(Lo )cr

= 2.63D / = 8.253in

2 d 2 DN a
= 0.0825lbf
4
386k
f n = 0.5
= 108Hz
W

W=

119

K B = (4C + 2) /( 4C 3) = 1.094
8 Fa D
= 23.334kpsi
d 3
8F D
m = K B m 3 = 38.89kpsi
d
8F D
s = K B s 3 = 71.56kpsi
d
n f = S sa / a = 1.5
a = KB

n s = S sy / s = 1.74
fom = 2.6 2 d 2 N t D / 4 = 1.01
Repeat the above analysis for other wire diameters and form a table to select the best
spring design:
d
0.069
0.071
0.080
0.085
0.090
0.095
0.105
0.112
0.297
0.332
0.512
0.632
0.767
0.919
1.274
1.569
D
0.228
0.261
0.432
0.547
0.677
0.824
1.169
1.457
ID
0.366
0.403
0.592
0.717
0.857
1.014
1.379
1.681
OD
14.00
4.33
4.67
6.40
7.44
8.53
9.67
12.14
C
127.2
102.4
44.8
30.5
21.3
15.4
8.63
6.0
Na
8.916
7.414
3.74
2.75
2.10
1.655
1.116
0.895
Ls
11.216
9.714
6.040
5.05
4.40
3.955
3.416
3.195
Lo
1.562
1.744
2.964
3.325
4.036
4.833
6.703
8.25
(Lo)cr
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
nf
1.86
1.85
1.82
1.81
1.79
1.78
1.75
1.74
ns
87.5
89.7
96.9
99.7
101.9
103.8
106.6
108
fn
-1.17
-1.12
-0.983
-0.948
-0.930
-0.927
-0.958
-1.01
fom
Examine the table and perform the adequacy assessment.

General Constraints:
o The constraint 3 Na 15 cancel wire diameters less than 0.105 in
o The constraint 4 C 12 cancel diameters larger than 0.105 in.

Problem Constraints:
o The constraint Ls 1.2 in cancel diameters less than 0.1050 in
o The constraint Lo 4 in cancel diameters less than 0.095 in
o fn 20fop fn 100Hz cancel diameters less than 0.090 in
o The buckling criterion cancel free length longer than (Lo)cr, which cancel
diameters less than0.075in.

120

10.11 Extension Springs

Most of the preceding discussion of compression springs applied equally to helical
extension springs
The natural frequency of a helical extension spring with both ends fixed against axial
deflection is the same as that for a helical spring in compression
But, In extension spring, the coils are usually close wound so that there is an initial
tension or so termed preload F. Therefore, no deflection occurs until the initial tension
built into the spring is overcome. That is: the applied load F becomes larger than initial
tension (F > Fi)
The load transfer can be done with: a threaded plug or a swivel hook
Figure 10-6 shows types of ends used on extension springs

Stresses in the body of the extension spring are handled the same as compression springs.
In designing a spring with a hook end, bending and torsion in the hook must be included
in the analysis.
Figure 10-7 shows two common used method of designing the end. c and d shows an
improved design due to a reduced coil diameter, not to elimination of stress
concentration. The reduced coil diameter results in a lower stress because of the shorter
moment arm.
No stress-concentration factor is needed for the axial component of the load

121

The maximum tensile stress at A, due to bending and axial loading, is given by:
16 D
4

A = F ( K ) A
+ 2
3
d
d

(10-34)

Where (K)A is a bending stress correction factor for curvature, given by:

(K ) A =

4C12 C1 1
4C1 (C1 1)

C1 =

2r1
d

(10-35)

The maximum torsional stress at point B is given by:

B = (K ) B

8 FD
d 3

(10-36)

where (K)B is the stress correction factor for curvature and it is given by:

(K )B

4C 2 1
4C 2 4

C2 =

2r2
d

(10-37)

122

If coils in contact with one another, the springs is known as close-wound.

To have a more accurate free length, it is preferred that some initial tension in close
wound springs.
Figure 10-8a shows the load-deflection curve, where y is the extension beyond the free
length Lo and Fi is the initial tension in the spring that must be exceeded before the spring
deflects
Thus, the load-deflection relation can be written as:
F=Fi+ky
Where k is the spring rate

The free length Lo of a spring measured inside the end loops or hooks as shown in figure
10-8b can be expressed as
Lo= 2(D-d) + (Nb+1)d = (2C-1+Nb)d
D mean coil diameter
Nb number of body coils
C the spring index

The equivalent number of active helical turns Na for use in equation 10-9 is
Na = Nb + G/E
where G and E are the shear the tensile modulus of elasticity

123

The amount of initial tension that a springmaker can routinely incorporate is as shown in
figure 10-8c

The preferred range can be expressed in terms of the uncorrected torsional stress i as
i =

33500
C 3

1000 4
psi
exp(0.105C )
6.5

(10-41)

The maximum allowable corrected stresses (KW or KB) for static applications of extension
springs are given in table 10-7
Percent of Tensile Strength
Materials
Patented, cold-drawn or hardened and tempered
carbon and low-alloy steels
Austenitic stainless steel and nonferrous alloys

In Torsion
body
end
45-50
40
35

30

In Bending
end
75
55

124

Example
A hard drawn steel wire extension spring has a wire diameter of 0.035 in, an outside coil
diameter of 0.248 in, hook radii of r1= 0.106 in and r2 = 0.089in, and an initial tension of
1.19 lbf. The number of body turns is 12.17. From the given information:
1. Determine the physical parameters of the spring(D, C, KB, Na, k, Lo, ymax)
2. Check the initial preload stress conditions
3. Find the factors of safety under a static 5.25 lbf load.
Solution
d = 0.035 in, OD = 0.248 in, r1 = 0.106 in, r2 = 0.089 in, Nb = 12.17
Fi = 1.19 lbf, Fmax=5.25
HD steel
Form table 10-5, with 0.033 < d < 0.063 E = 28.7Mpsi, G = 11.6Mpsi
From table 10-4, A = 140kpsi-inm, m = 0.19
1. The physical parameters
D = OD d = 0.248 0.035 = 0.213
C = D / d = 6.086
4C + 2
KB =
= 1.234
4C 3
N a = N B + G / E = 12.17 + 11.6 / 28.7 = 12.57turns
k=

d 4G
= 17.76lbf / in
8D 3 N a

Lo = (2C 1 + N b )d = 0.817in
The deflection under the service load is:
(F Fi ) = 5.25 1.19 = 0.229in
y max = max
k
17.76
Therefore the maximum spring length is:
L = Lo + ymax =0.817+0.229 = 1.046 in
2. Initial preload stress condition:
The uncorrected initial stress is given by equation (10-3) without the correction factor:
( i )uncorr = 8Fi D3 = 15.1kpsi
d
The preferred range is given by equation (10-41):
( i ) pref = 33500 1000 4 C 3 = 17681 3525 = 14.2kpsi or
exp(0.105C )
6.5

21.2kpsi

125

3. The factor of safety under static load

We need to check three positions:

The shear stress under the service load

The bending at the end hook which is represented by point A
The torsion at the end hook which is represented by point B

1. For the shear stress under the service load

S sy
8F D
; S sy = 0.45S ut & max = K B max3
n=
max
d
140
A
= 264.7 kpsi
S ut = m =
0.035 0.19
d
8(5.25)(0.213)
= 82kpsi
max = 1.234
0.035 3
0.45(264.7 )
= 1.45
n=
82

2. The bending at the end hook which is represented by point A

Sy
; S y = 0.75S ut = 0.75(264.7 ) = 198.5kpsi
nA =
A
16 D
4

+ 2 ;
A = Fmax (K ) A
3
d
d

(K ) A = 1.14 A = 156.9kpsi
nA =

C1 1
4C1 (C1 1)
2

(K ) A = 4C1

C1 = 2r1 / d = 6.057

198.5
= 1.27
156.9

3. The torsion at the end hook which is represented by point B

S sy
nB =
; S sy = 0.4 S ut = 0.4(264.7 ) = 105.88kpsi
B
8Fmax D
4C 2 1
; ( K )B =
3
4C 2 4
d
= 1.18 B = 78.4kpsi

B = (K ) B

(K )B
nA =

C 2 = 2r2 / d = 5.086

105.88
= 1.35
78.4

Note that Ssy=0.4Sut from table 10-7 under torsion for the end part.
From all three calculations, the yield will first occur due to the bending of the end hook.

126

Example: Fatigue
The helical extension spring of the pervious example is subjected to a dynamic loading
from 1.5 lbf to 5 lbf. Estimate the factors of safety using Goodman failure criterion for
1. The coil fatigue
2. The coil yield
3. The end hook bending fatigue at point A
4. The end hook torsion fatigue at point B
Solution
d = 0.035 in, D = 0.213 in, r1= 0.106 in, r2 = 0.089 in, Nb = 12.17
Fi = 1.19 lbf, Fmin=1.5, Fmax=5,
From the pervious example we have:
C=6.086, Lo=0.817, k=17.76 lbf/in
KB=1.234, (K)A=1.14, (K)B=1.18, (i)uncorr=15.1 kpsi
Sut = 264.7kpsi, Ssu = 0.67Sut= 177.3 kpsi, Sy =198.5 kpsi, Ssy =119.1 kpsi (shear in body)
1. The fatigue in the body coil:
Fa = (5 1.5) / 2 = 1.75lbf
Fm = (5 + 1.5) / 2 = 3.25lbf

8Fa D
= 27.3kpsi
d 3
8F D
m = K B m 3 = 50.7kpsi
d
a = KB

S sa = 35kpsi
S sm = 55kpsi
S se =

S sa
= 50.74kpsi
1 (S sm / S su )

The factor of safety guarding against failure to be

a m
S se S su
1
+
=
nf =
= 1.214
S se S su n f
a S su + m S se
2. The coil yield:
S
n y = sa
a
To find the values of Ssa:
a
r=
(1)
m i
S sa S sm
+
= 1 (2)
S sy S sy

Ssy

S sa S sm
+
=1
S sy S sy
r

Ssa

Ssm

Ssy
127

The intersection between 1 & 2 gives:

r
(S sy i )
S sa =
r +1
Therefore:
F
i a
1.19
=
i = i a =
27.3 = 18.6kpsi
Fi Fa
1.75
Fa
27.3
= 0.85
r=
50.7 18.6
S sa = 46.2kpsi
Thus :
46.2
= 1.69
ny =
27.3
3. The end hook bending fatigue at point A
n f = 1 / ( a / S e + m / S ut ) ;
16 D
4

a = Fa (K ) A
+ 2 = 52.3kpsi ;
3
d
d

F
m = m a = 97.1kpsi
Fa
S se
= 67.1kpsi
0.577
(n f )A = 0.87
Se =

4. The end hook torsion fatigue at point B

n f = 1 / ( sa / S se + sm / S su )
8 Fa D
= 26.1kpsi
d 3
( m )B = (K )B 8Fm 3D = 48.5kpsi
d
(n f )B = 1.27

( a )B = (K )B

128