Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

Mechanical testing - notched bar or impact testing


Before looking at impact testing let us first define what is meant by 'toughness' since the impact test is only one method by which this material property is measured.

Toughness is, broadly, a measure of the amount of energy required to cause an item - a test piece or a bridge or a pressure vessel - to fracture and fail. The more energy that is required then the tougher the material.


 Click here to see our latest technical engineering podcasts on YouTube.


The area beneath a stress/strain curve produced from a tensile test is a measure of the toughness of the test piece under slow loading conditions. However, in the context of an impact test we are looking at notch toughness, a measure of the metal's resistance to brittle or fast fracture in the presence of a flaw or notch and fast loading conditions.

It was during World War II that attention was focused on this property of 'notch toughness' due to the brittle fracture of all-welded Liberty ships, then being built in the USA. From this work the science of fracture toughness developed and gave rise to a range of tests used to characterise 'notch toughness' of which the Charpy-V test described in this article is one.

There are two main forms of impact test, the Izod and the Charpy test.

Both involve striking a standard specimen with a controlled weight pendulum travelling at a set speed. The amount of energy absorbed in fracturing the test piece is measured and this gives an indication of the notch toughness of the test material.

These tests show that metals can be classified as being either 'brittle' or 'ductile'. A brittle metal will absorb a small amount of energy when impact tested, a tough ductile metal a large amount of energy.

It should be emphasised that these tests are qualitative, the results can only be compared with each other or with a requirement in a specification - they cannot be used to calculate the fracture toughness of a weld or parent metal, such as would be needed to perform a fitness for service assessment. Fracture toughness tests that can be used in this way are covered in other Job Knowledge articles (Job Knowledge 76 & 77). The Izod test is rarely used these days for weld testing having been replaced by the Charpy test and will not be discussed further in this article.

The Charpy specimen may be used with one of three different types of notch, a 'keyhole', a 'U' and a 'V'. The keyhole and U-notch are used for the testing of brittle materials such as cast iron and for the testing of plastics. The V-notch specimen is the specimen of choice for weld testing and is the one discussed here.

The current British Standard for Charpy testing is BS EN ISO 148-1:2009 and the American Standard is ASTM E23. The standards differ only in the details of the strikers used. The standard Charpy-V specimen, illustrated in Fig.1. is 55mm long, 10mm square and has a 2mm deep notch with a tip radius of 0.25mm machined on one face.

Fig.1. Standard Charpy-V notch specimen
Fig.1. Standard Charpy-V notch specimen

To carry out the test the standard specimen is supported at its two ends on an anvil and struck on the opposite face to the notch by a pendulum as shown in Fig.2. The specimen is fractured and the pendulum swings through, the height of the swing being a measure of the amount of energy absorbed in fracturing the specimen. Conventionally three specimens are tested at any one temperature, see Fig.3, and the results averaged.

Fig.2. Charpy testing machine
Fig.2. Charpy testing machine
Fig.3. Schematic Charpy-V energy and % age crystallinity curves
Fig.3. Schematic Charpy-V energy and % age crystallinity curves

A characteristic of carbon and low alloy steels is that they exhibit a change in fracture behaviour as the temperature falls with the failure mode changing from ductile to brittle.

If impact testing is carried out over a range of temperatures the results of energy absorbed versus temperature can be plotted to give the 'S' curve illustrated in Fig.3.

This shows that the fracture of these types of steels changes from being ductile on the upper shelf to brittle on the lower shelf as the temperature falls, passing through a transition region where the fracture will be mixed.


Many specifications talk of a transition temperature, a temperature at which the fracture behaviour changes from ductile to brittle. This temperature is often determined by selecting, quite arbitrarily, the temperature at which the metal achieves an impact value of 27 Joules - see, for example the impact test requirements of EN 10028 Part 2 Steel for Pressure Purposes.

What the curve shows is that a ductile fracture absorbs a greater amount of energy than a brittle fracture in the same material. Knowing the temperature at which the fracture behaviour changes is therefore of crucial importance when the service temperature of a structure is considered - ideally in service a structure should operate at upper shelf temperatures.

The shape of the S curve and the positions of the upper and lower shelves are all affected by composition, heat treatment condition, whether or not the steel has been welded, welding heat input, welding consumable and a number of additional factors. All the factors must be controlled if good notch toughness is required. This means that close control of the welding parameters is essential if impact testing is a specification requirement. Austenitic stainless steels, nickel and aluminium alloys do not show this change in fracture behaviour, the fracture remaining ductile even to very low temperatures. This is one reason why these types of alloys are used in cryogenic applications.

In addition to the impact energy there are two further features that can be measured and may be found as a requirement in some specifications. These are percentage crystallinity and lateral expansion.

The appearance of a fracture surface gives information about the type of fracture that has occurred - a brittle fracture is bright and crystalline, a ductile fracture is dull and fibrous.

Percentage crystallinity is therefore a measure of the amount of brittle fracture, determined by making a judgement of the amount of crystalline or brittle fracture on the surface of the broken specimen.

Lateral expansion is a measure of the ductility of the specimen. When a ductile metal is broken the test piece deforms before breaking, a pair of 'ears' being squeezed out on the side of the compression face of the specimen, as illustrated in Fig 4. The amount by which the specimen deforms is measured and expressed as millimetres of lateral expansion. ASME B31.3 for example requires a lateral expansion of 0.38mm for bolting materials and steels with a UTS exceeding 656N/mm2, rather than specifying an impact value.


Fig.4 Lateral expansion
Fig.4 Lateral expansion

The next article in this series will look at the testing of welds, how the impact strength can be affected by composition and microstructure and some of its limitations and disadvantages.

For information on this article please contact us.

Mechanical testing - Notched bar or impact testing - Part II

For more information please email: