Design of helical compression springs can be one such example.

For given force (F) and deflection

(y) characteristics, the parameters to be determined are outside diameter D0, inside diameter Di, wire
diameter d, free length Lf, shut or solid length Ls, number of active coils Na and spring rate k. Safety
checks are to be provided for static stresses, and buckling. Designer can select from a range of spring
index C = D/d (ratio of mean coil diameter, D = (D0 + Di)/2, to wire diameter d), a set of standard
wire diameters d, materials and their ultimate strengths Sut and shear modulus G. The material
strength is dependent on wire diameter d and design calculations are very sensitive to the wire size.
The relation between spring stiffness k and deflection y can be found in any machine element
design book (e.g. by Norton, Shigley, and others)3,4
y
FD N
Gd
k Fy
Gd
DN
a
a
=
8
,==
8
3
4
4
3 (1.14)
Maximum shear stress is given by


=
8
3K
FD
dw
where the Wahl’s correction factor for stress concentration is
K
C
wC C =
4–1
4–4
+ 0.615
(1.15)
The tensile strength Sut is related to the wire diameter d as
Sut = Adb (1.16)
Here A and b are constants depending upon the wire material and diameter. A set of typical values is
given in Table 1.25. All standard data has been taken from this reference.
A chronology of the design steps for static loading is described below, although there may be
variations in the procedure depending on the requirements.
Problem: Design a helical compression spring, which should apply a minimum force Fmin and a
maximum force Fmax over a range of deflection . The initial compression on the spring is given to
be Finitial.
Step 1: From Table 1.1(a), a suitable material is selected. For static loading, the most commonly
used, least expensive spring wire material is A227. Select a preferred wire diameter (d) from Table
1.1(b). For example, A227 is available in the diameter range from 0.70 mm to 16 mm. Select an
intermediate value so that it leaves some space for iterations later, if required.
Step 2: The spring index C=D/d is generally recommended to be in the range
12 > C > 4 (1.17)
3Dimarogonas, A. (1989) Computer Aided Machine Design, Prentice-Hall, N.Y.
4Shigley, J.E., Mischke, C.R. (2001) Mechanical Engineering Design, McGraw-Hill, Singapore.
5Associated Springs-Barnes Group (1987) Design Hand Book, Bristol, Conn.

30 COMPUTER AIDED ENGINEERING DESIGN

the origin O, shifting the line L parallel to itself to a translated position L*.
(b) Rotate L* by an angle such that it coincides with the y-axis (new position of the line is
L**, say).
(c) Reflect S about the y-axis using Eq. (2.9).
(d) Rotate L** through −to bring it back to L*.
(e) Translate L* to coincide with its original position L.
The schematic of the procedure is shown in Figure 2.8. The new image S* is the reflection of S about
L and the transformation is given by
10
01
001
cos sin 0
–sin cos 0
001
–1 0 0
010
001
cos –sin 0
sin cos 0
001
p
q
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥




×
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
10–
01–
001
p
q = TODR(–)RfyR()TDO (2.10)
x*
y
x
y
x
y
x
* fyy
1
=
–
1
=
–1 0 0
010
0011
=
1
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
R
(2.9)
Example 2.4. Consider a trapezium ABCD with A = (6, 1, 1), B = (8, 1, 1), C = (10, 4, 1) and
D = (3, 4, 1). The entity is to be reflected through the y-axis. Applying Rfy in Eq. (2.9) results in
A
B
C
D
TTT*
*
*
*
=
–1 0 0
010
001
611
811
10 4 1
341
=
–6 1 1
–8 1 1
–10 4 1
–3 4 1
⎡
⎣
⎢⎢⎢⎢⎢
⎤
⎦
⎥⎥⎥⎥⎥
⎡
⎣
⎢⎢⎢
⎤
⎦
⎥⎥⎥
⎡
⎣
⎢⎢⎢⎢⎢
⎤
⎦
⎥⎥⎥⎥⎥
⎡
⎣
⎢⎢⎢⎢⎢
⎤
⎦
⎥⎥⎥⎥⎥
The new position for the trapezium is shown as A∗B∗C∗D∗in Figure 2.7. Note that identical result
may be obtained by rotating the trapezium by
180°about the y axis. As expected there is no
distortion in the shape of the trapezium. Since
reflection results by combining translation and/
or rotation, it is a rigid body transformation.
2.2.6 Reflection About an Arbitrary Line
Let D be a point on line L and S be an object in
two-dimensional space. It is required to reflect S
about L. This reflection can be obtained as a
sequence of the following transformations:
(a) Translate point D ( p, q, 1) to coincide with
–15 –10 –5 0 5 10 15 x
AB
C
D
B* A*
D*
C*
10 y
5
0
–5
Figure 2.7 Reflection about the y-axis