UNIVERSAL TESTING MACHINE
HISTORY
Shortly after the end of the Civil War, with industrialization gaining momentum in the
United States and new construction sprouting up everywhere, there was an increasing need for
better ways to test the materials used in machinery, in buildings, and elsewhere. Engineers
decried the lack of reliable information on the properties of materials. “The art of
construction….is involved in mystery and obscurity,” noted Van Nostrand’s Engineering
Magazine. “The knowledge of materials is at present….absolutely empirical….Before the
constructor makes use of either a new material, or an old one in a new form, the only safe method
is to experiment.”
Without some means of testing the strength of materials before their use, however, these
experimental results were likely to take the form of explosions or building collapses. From the
1850’s on, various devices for testing materials had been developed, but the goal of a truly
universal testing machine proved elusive until 1880, when Philadelphia engineer Tinius Olsen, a
Norwegian immigrant who had just lost his job, devised and patented what became known as
the Little Giant. Here at last was a machine for tensile, transverse, and compression testing united
in a single instrument. Olsen’s mechanism was to become the ancestor of all testing machines
subsequently produced around the world, while the company Olsen set up to market his
invention continues in the testing machine business to this day. This year Tinius Olsen Inc.
celebrates its 125th year of continuous operations.
STRESS, STRAIN AND ELASTIC MODULI
The rigid body is a useful idealized model, but the stretching, squeezing, and twisting of
real bodies when forces are applied are often too important to ignore. Figure 1 shows three
examples. We want to study the relationship between the forces and deformations for each case.
For each kind of deformation we will introduce a quantity called stress that characterizes the
strength of the forces causing the deformation, on a “force per unit area” basis. Another quantity,
strain, describes the resulting deformation. When the stress and strain are small enough, we
often find that the two are directly proportional, and we call the proportionality constant an
JAT Page 1 of 11
elastic modulus. The harder you pull on something, the more it stretches; the more you squeeze
it, the more it compresses. In equation form, this says
𝑆𝑡𝑟𝑒𝑠𝑠
= 𝐸𝑙𝑎𝑠𝑡𝑖𝑐 𝑚𝑜𝑑𝑢𝑙𝑢𝑠 (𝐻𝑜𝑜𝑘𝑒 ′ 𝑠𝐿𝑎𝑤 ) 𝐸𝑞 1
𝑆𝑡𝑟𝑎𝑖𝑛
The proportionality of stress and strain (under certain conditions) is called Hooke’s law,
after Robert Hooke (1635–1703).
Figure 1. Three types of stress. (a) Bridge cables under tensile stress, being stretched by
forces acting at their ends. (b) A diver under bulk stress, being squeezed from all sides by forces
due to water pressure. (c) A ribbon under shear stress, being deformed and eventually cut by
forces exerted by the scissors.
TENSILE AND COMPRESSIVE STRESS AND STRAIN
The simplest elastic behavior to understand is the stretching of a bar, rod, or wire when
its ends are pulled (Fig. 1(a)). Figure 2 shows an object that initially has uniform cross-sectional
area A and length l0.
Figure 2. An object in tension. The net force on the
object is zero, but the object deforms. The tensile
stress (the ratio of the force to the cross-sectional
area) produces a tensile strain (the elongation
divided by the initial length).
JAT Page 2 of 11
We then apply forces of equal magnitude Fꓕ but opposite directions at the ends (this
ensures that the object has no tendency to move left or right). We say that the object is in
tension. We’ve already talked a lot about tension in ropes and strings; it’s the same concept here.
The subscript ꓕ is a reminder that the forces act perpendicular to the cross section.
We define the tensile stress at the cross section as the ratio of the force to the cross-
sectional area A:
𝐹ꓕ
𝑇𝑒𝑛𝑠𝑖𝑙𝑒 𝑠𝑡𝑟𝑒𝑠𝑠 = 𝐸𝑞 2
𝐴
This is a scalar quantity because Fꓕ is the magnitude of the force. The SI unit of stress is
the pascal (abbreviated Pa and named for the 17th-century French scientist and philosopher
Blaise Pascal). Equation (2) shows that 1 pascal equals 1 newton per square meter (N/m 2):
𝑁
1 𝑝𝑎𝑠𝑐𝑎𝑙 = 1 𝑃𝑎 = 1
𝑚2
In the British system the logical unit of stress would be the pound per square foot, but
the pound per square inch (lb/in2 or psi) is more commonly used. The conversion factors are
1 𝑝𝑠𝑖 = 6895 𝑃𝑎 𝑎𝑛𝑑 1 𝑃𝑎 = 1.450 𝑥 10−4 𝑝𝑠𝑖
The units of stress are the same as those of pressure. Air pressure in automobile tires is
typically around 3 x 105 Pa = 300 kPa, and steel cables are commonly required to withstand tensile
stresses of the order of 108 Pa.
The object shown in Fig. 2 stretches to a length l = l0 + ∆l when under tension. The
elongation ∆l does not occur only at the ends; every part of the bar stretches in the same
proportion. The tensile strain of the object is equal to the fractional change in length, which is
the ratio of the elongation ∆l to the original length l0.
𝑙 − 𝑙0 ∆𝑙
𝑇𝑒𝑛𝑠𝑖𝑙𝑒 𝑠𝑡𝑟𝑎𝑖𝑛 = = 𝐸𝑞 3
𝑙0 𝑙0
Tensile strain is stretch per unit length. It is a ratio of two lengths, always measured in the
same units, and so is a pure (dimensionless) number with no units.
JAT Page 3 of 11
Experiment shows that for a sufficiently small tensile stress, stress and strain are
proportional, as in Eq. (1). The corresponding elastic modulus is called Young’s modulus, denoted
by Y:
𝐹ꓕ
𝑇𝑒𝑛𝑠𝑖𝑙𝑒 𝑠𝑡𝑟𝑒𝑠𝑠 𝐹ꓕ 𝑥 𝑙0
𝑌= = 𝐴 = 𝐸𝑞 4
𝑇𝑒𝑛𝑠𝑖𝑙𝑒 𝑠𝑡𝑟𝑎𝑖𝑛 ∆𝑙 𝐴 𝑥 ∆𝑙
𝑙0
Since strain is a pure number, the units of Young’s modulus are the same as those of stress:
force per unit area. Some typical values are listed in Table 1.
TABLE 1 APPROXIMATE ELASTIC MODULI
Young’s Modulus, Y Bulk Modulus, B Shear Modulus, S
Material
(Pascal) (Pascal) (Pascal)
Aluminum 7.0 x 1010 7.5 x 1010 2.5 x 1010
Brass 9.0 x 1010 6.0 x 1010 3.5 x 1010
Copper 11 x 1010 14 x 1010 4.4 x 1010
Crown glass 6.0 x 1010 5.0 x 1010 2.5 x 1010
Iron 21 x 1010 16 x 1010 7.7 x 1010
Lead 1.6 x 1010 4.1 x 1010 0.6 x 1010
Nickel 21 x 1010 17 x 1010 7.8 x 1010
Steel 20 x 1010 16 x 1010 7.5 x 1010
A material with a large value of Y is relatively unstretchable; a large stress is required for
a given strain. For example, the value of Y for cast steel is (2 x 1011 Pa) much larger than that for
rubber (5 x 108 Pa).
Figure 3. An object in compression. The compressive
stress and compressive strain are defined in the same
way as tensile stress except that ∆l now denotes the
distance that the object contracts.
JAT Page 4 of 11
When the forces on the ends of a bar are pushes rather than pulls (Fig. 3), the bar is in
compression and the stress is a compressive stress. The compressive strain of an object in
compression is defined in the same way as the tensile strain, but ∆𝑙 has the opposite direction.
Hooke’s law and Eq. (4) are valid for compression as well as tension if the compressive stress is
not too great. For many materials, Young’s modulus has the same value for both tensile and
compressive stresses. Composite materials such as concrete and stone are an exception; they
can withstand compressive stresses but fail under comparable tensile stresses. Stone was the
primary building material used by ancient civilizations such as the Babylonians, Assyrians, and
Romans, so their structures had to be designed to avoid tensile stresses. Hence they used arches
in doorways and bridges, where the weight of the overlying material compresses the stones of
the arch together and does not place them under tension.
In many situations, bodies can experience both tensile and compressive stresses at the
same time. As an example, a horizontal beam supported at each end sags under its own weight.
(a) (b)
Figure 4. (a) A beam supported at both ends is under both compression and tension. (b) The
cross-sectional shape of an I-beam minimizes both stress and weight.
As a result, the top of the beam is under compression, while the bottom of the beam is under
tension (Fig. 4(a)). To minimize the stress and hence the bending strain, the top and bottom of
the beam are given a large cross-sectional area. There is neither compression nor tension along
JAT Page 5 of 11
the centerline of the beam, so this part can have a small cross section; this helps to keep the
weight of the bar to a minimum and further helps to reduce the stress. The result is an I-beam of
the familiar shape used in building construction (Fig. 4(b)).
EXAMPLE 1
A steel rod 2.0 m long has a cross-sectional area of 0.30 cm2. It is hung by one end from a support,
and a 550-kg milling machine is hung from its other end. Determine the stress on the rod and the
resulting strain and elongation.
BULK STRESS AND STRAIN
When a scuba diver plunges deep into the ocean, the water exerts nearly uniform
pressure everywhere on his surface and squeezes him to a slightly smaller volume (see Fig. 1(b)).
This is a different situation from the tensile and compressive stresses and strains we have
discussed. The stress is now a uniform pressure on all sides, and the resulting deformation is a
volume change. We use the terms bulk stress (or volume stress) and bulk strain (or volume
strain) to describe these quantities.
JAT Page 6 of 11
If an object is immersed in a fluid (liquid or gas) at rest, the fluid exerts a force on any part
of the object’s surface; this force is perpendicular to the surface. (If we tried to make the fluid
exert a force parallel to the surface, the fluid would slip sideways to counteract the effort.) The
force 𝐹ꓕ per unit area that the fluid exerts on the surface of an immersed object is called the
pressure p in the fluid:
𝐹ꓕ
𝑝= (𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 𝑖𝑛 𝑎 𝑓𝑙𝑢𝑖𝑑 ) 𝐸𝑞 5
𝐴
The pressure in a fluid increases with depth. For example, the pressure of the air is about
21% greater at sea level than in Denver (at an elevation of 1.6 km, or 1.0 mi). If an immersed
object is relatively small, however, we can ignore pressure differences due to depth for the
purpose of calculating bulk stress. Hence we will treat the pressure as having the same value at
all points on an immersed object’s surface.
Pressure has the same units as stress; commonly used units include 1 Pa (1 N/m2) and
1lb/in2 (1 psi). Also in common use is the atmosphere, abbreviated atm. One atmosphere is the
approximate average pressure of the earth’s atmosphere at sea level:
14.7𝑙𝑏
1 𝑎𝑡𝑚𝑜𝑠𝑝ℎ𝑒𝑟𝑒 = 1 𝑎𝑡𝑚 = 1.013 𝑥 105 𝑃𝑎 =
𝑖𝑛2
CAUTION Pressure vs. force Unlike force, pressure has no intrinsic direction: The pressure on the
surface of an immersed object is the same no matter how the surface is oriented. Hence pressure
is a scalar quantity, not a vector quantity.
Figure 5. An object under bulk stress. Without the stress,
the cube has a volume V0; when the stress is applied,
the cube has a smaller volume V. The volume change ∆V
is exaggerated for clarity.
JAT Page 7 of 11
Pressure plays the role of stress in a volume deformation. The corresponding strain is the
fractional change in volume (Fig. 5)—that is, the ratio of the volume change ∆V to the original
volume V0:
∆𝑉
𝐵𝑢𝑙𝑘 (𝑣𝑜𝑙𝑢𝑚𝑒)𝑠𝑡𝑟𝑎𝑖𝑛 = 𝐸𝑞 6
𝑉0
Volume strain is the change in volume per unit volume. Like tensile or compressive strain,
it is a pure number, without units.
When Hooke’s law is obeyed, an increase in pressure (bulk stress) produces a proportional
bulk strain (fractional change in volume). The corresponding elastic modulus (ratio of stress to
strain) is called the bulk modulus, denoted by B. When the pressure on a body changes by a small
amount ∆p, from p0 to p0 + ∆p, and the resulting bulk strain is ∆V/V0, Hooke’s law takes the form
𝐵𝑢𝑙𝑘 𝑠𝑡𝑟𝑒𝑠𝑠 ∆𝑝
𝐵= = − (𝑏𝑢𝑙𝑘 𝑚𝑜𝑑𝑢𝑙𝑢𝑠) 𝐸𝑞 7
𝐵𝑢𝑙𝑘 𝑠𝑡𝑟𝑎𝑖𝑛 ∆𝑉
𝑉0
We include a minus sign in this equation because an increase of pressure always causes a
decrease in volume. In other words, if ∆p is positive, is ∆V negative. The bulk modulus B itself is
a positive quantity.
For small pressure changes in a solid or a liquid, we consider B to be constant. The bulk
modulus of a gas, however, depends on the initial pressure p0. Table 1 includes values of the bulk
modulus for several solid materials. Its units, force per unit area, are the same as those of
pressure (and of tensile or compressive stress).
The reciprocal of the bulk modulus is called the compressibility and is denoted by k. From
Eq. (7),
∆𝑉
1 𝑉 ∆𝑉
𝑘= = − 0 = − (𝑐𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑖𝑏𝑖𝑙𝑖𝑡𝑦) 𝐸𝑞 8
𝐵 ∆𝑝 𝑉0 𝑥 ∆𝑝
Compressibility is the fractional decrease in volume, -∆V/V0, per unit increase ∆p, in pressure.
The units of compressibility are those of reciprocal pressure, Pa-1 or atm-1.
Table 2 lists the values of compressibility k for several liquids. For example, the
compressibility of water is 46.4 x 10-6 atm-1 which means that the volume of water decreases by
JAT Page 8 of 11
46.4 parts per million for each 1-atmosphere increase in pressure. Materials with small bulk
modulus and large compressibility are easier to compress.
TABLE 2 COMPRESSIBILITIES OF LIQUIDS
Compressibility, k
LIQUID Pa-1 atm-1
Carbon disulfide 93 x 10-11 94 x 10-6
Ethyl alcohol 110 x 10-11 111 x 10-6
Glycerine 21 x 10-11 21 x 10-6
Mercury 3.7 x 10-11 3.8 x 10-6
Water 45.8 x 10-11 46.4 x 10-6
EXAMPLE 2
A hydraulic press contains 0.25m3 (250 L) of oil. Find the decrease in the volume of the oil when
it is subjected to a pressure increase ∆p = 1.6 x 107 Pa (about 160 atm or 2300 psi). The bulk
modulus of the oil is B = 5.0 x 109 Pa (about 5.0 x 104 atm) and its compressibility is k = 1/B = 20
x 10-6 atm-1.
JAT Page 9 of 11
SHEAR STRESS AND STRAIN
The third kind of stress-strain situation is called shear. The ribbon in Fig. 1c is under shear stress:
One part of the ribbon is being pushed up while an adjacent part is being pushed down, producing
a deformation of the ribbon. Figure 6 shows a body being deformed by a shear stress.
Figure 6. An object under shear stress. Forces are
applied tangent to opposite surfaces of the object
(in contrast to Fig 2, in which the forces act
perpendicular to the surfaces).
In the figure, forces of equal magnitude but opposite direction act tangent to the surfaces of
opposite ends of the object. We define the shear stress as the force F│ acting tangent to the
surface divided by the area A on which it acts:
𝐹│
𝑆ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑒𝑠𝑠 = 𝐸𝑞 9
𝐴
Shear stress, like the other two types of stress, is a force per unit area.
Figure 6 shows that one face of the object under shear stress is displaced by a distance x
relative to the opposite face. We define shear strain as the ratio of the displacement x to the
transverse dimension h:
𝑥
𝑆ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑎𝑖𝑛 = 𝐸𝑞 10
ℎ
In real-life situations, x is nearly always much smaller than h. Like all strains, shear strain
is a dimensionless number; it is a ratio of two lengths.
JAT Page 10 of 11
If the forces are small enough that Hooke’s law is obeyed, the shear strain is proportional
to the shear stress. The corresponding elastic modulus (ratio of shear stress to shear strain) is
called the shear modulus, denoted by S:
𝐹│
𝑆ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑒𝑠𝑠 𝐹│ ℎ
𝑆= = 𝐴
𝑥 = (𝑠ℎ𝑒𝑎𝑟 𝑚𝑜𝑑𝑢𝑙𝑢𝑠) 𝐸𝑞 11
𝑆ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑎𝑖𝑛 𝐴 𝑥
ℎ
Table 11.1 gives several values of shear modulus. For a given material, S is usually one-
third to one-half as large as Young’s modulus Y for tensile stress. Keep in mind that the concepts
of shear stress, shear strain, and shear modulus apply to solid materials only. The reason is that
shear refers to deforming an object that has a definite shape (see Fig. 6). This concept doesn’t
apply to gases and liquids, which do not have definite shapes.
EXAMPLE 3
Suppose the object in Fig. 6 is the brass base plate of an outdoor sculpture that experiences shear
forces in an earthquake. The plate is 0.80 m square and 0.50 cm thick. What is the force exerted
on each of its edges if the resulting displacement x is 0.16 mm?
JAT Page 11 of 11
