The hardness of a material can have a number of meanings depending upon the context, which in the case of metals generally means the resistance to indentation. There are a number of test methods of which only the Brinell, Vickers and portable hardness testing will be covered in this article.
Brinell Hardness Test
The Brinell test was devised by a Swedish researcher at the beginning of the 20th century. The test comprises forcing a hardened steel ball indentor into the surface of the sample using a standard load as shown in Fig.1(a). The diameter/load ratio is selected to provide an impression of an acceptable diameter. The ball may be 10, 5 or 1mm in diameter, the load may be 3000, 750 or 30kgf, The load, P, is related to the diameter, D by the relationship P/D2 and this ratio has been standardised for different metals in order that test results are accurate and reproducible. For steel the ratio is 30:1 - for example a 10mm ball can be used with a 3000kgf load or a 1mm ball with a 30kgf load. For aluminium alloys the ratio is 5:1. The load is applied for a fixed length of time, usually 30 seconds. When the indentor is retracted two diameters of the impression, d1 and d2 , are measured using a microscope with a calibrated graticule and then averaged as shown in Fig.1(b).
The Brinell hardness number (BHN) is found by dividing the load by the surface area of the impression. There is a somewhat tedious calculation that can be carried out to determine the hardness number but it is more usual and far simpler to refer to a set of standard tables from which the Brinell hardness number can be read directly.
The Brinell test is generally used for bulk metal hardness measurements - the impression is larger than that of the Vickers test and this is useful as it averages out any local heterogeneity and is affected less by surface roughness. However, because of the large ball diameter the test cannot be used to determine the hardness variations in a welded joint for which the Vickers test is preferred. Very hard metals, over 450BHN may also cause the ball to deform resulting in an inaccurate reading. To overcome this limitation a tungsten carbide ball is used instead of the hardened steel ball but there is also a hardness limit of 600BHN with this indentor.
Vickers Hardness Test
The Vickers hardness test operates on similar principles to the Brinell test, the major difference being the use of a square based pyramidal diamond indentor rather than a hardened steel ball. Also, unlike the Brinell test, the depth of the impression does not affect the accuracy of the reading so the P/D2 ratio is not important. The diamond does not deform at high loads so the results on very hard materials are more reliable. The load may range from 1 to 120kgf and is applied for between 10 and 15 seconds.
The basic principles of operation of the Vickers hardness test are illustrated in Fig.2 where it can be seen that the load is applied to the indentor by a simple weighted lever. In older machines an an oil filled dash pot is used as a timing mechanism - on more modern equipment this is done electronically.
As illustrated in Fig.3(b) two diagonals, d1 and d2 , are measured, averaged and the surface area calculated then divided into the load applied. As with the Brinell test the diagonal measurement is converted to a hardness figure by referring to a set of tables. The hardness may be reported as Vickers Hardness number (VHN), Diamond Pyramid Number (DPN) or, most commonly, Hvxx where 'xx' represents the load used during the test.
As mentioned earlier, the Vickers indentation is smaller than the Brinell impression and thus far smaller areas can be tested, making it possible to carry out a survey across a welded joint, including individual runs and the heat affected zones. The small impression also means that the surface must be flat and perpendicular to the indentor and should have a better than 300 grit finish.
Errors in Hardness Testing
There are many factors that can affect the accuracy of the hardness test. Some of these such as flatness and surface finish have already been mentioned above but it is worth re-emphasising the point that flatness is most important - a maximum angle of approximately ± 1° would be regarded as acceptable.
To achieve the required flatness tolerance and surface finish surface grinding or machining may be necessary. The correct load must be applied and to achieve this there must be no friction in the loading system otherwise the impression will be smaller than expected - regular maintenance and calibration of the machine is therefore essential. The condition of the indentor is crucial - whilst the Vickers diamond is unlikely to deteriorate with use unless it is damaged or loosened in its mounting by clumsy handling, the Brinell ball will deform over a period of time and inaccurate readings will result. This deterioration will be accelerated if a large proportion of the work is on hard materials. The length of time that the load is applied is important and must be controlled.
The specimen dimensions are important - if the test piece is too thin the hardness of the specimen table will affect the result. As a rule of thumb the specimen thickness should be ten times the depth of the impression for the Brinell test and twice that of the Vickers diagonal. Similarly, if the impression is too close to the specimen edge then low hardness values will be recorded - again as a rule the impression should be some 4 to 5 times the impression diameter from any free edge. Performing hardness testing on cylindrical surfaces eg pipes and tubes, the radius of curvature will affect the indentation shape and can lead to errors. It may be necessary to apply a correction factor - this is covered in an ISO specification, ISO 6507 Part 1.
The specimen table should be rigidly supported and must be in good condition - burrs or raised edges beneath the sample will give low readings. Impact loading must be avoided. It is very easy to force the indentor into the specimen surface when raising the table into position. This can strain the equipment and damage the indentor. Operator training is crucial and regular validation or calibration is essential if hardness rest results are to be accurate and reproducible.
This article was written by Gene Mathers.