Metal fatigue is caused by repeated cycling of the load. It is a progressive localized damage due to fluctuating stresses and strains on the material. Metal fatigue cracks initiate and propagate in regions where the strain is most severe. This cyclic loading and crack initiation is represented using S-N curves. The Fig 1.1 consists of constant cyclic stress amplitude(S) which is applied to a specimen and the number of loading cycles (N) until the specimen fails is determined. The process fatigue failure is consists of three main stages. The first stage consists of initial crack initiation. The second stage consists of progressive crack growth across the part and the third and final stage consists of sudden fracture of the remaining cross section.
The fatigue strength is the stress at which failure occurs for a given number of cycles, whereas the fatigue life is the number of cycles required material to fail. The most important concept of the S-N diagram is shown in Fig 1.1. This figure consists of S-N curves for Steel and Aluminium. (By Shawn M. Kelly)
http://www.efunda.com/formulae/solid_mechanics/fatigue/images/fatigue_SN_01.gif
Figure 1.1 S-N curves for steel and aluminium.
The subject of fatigue testing is extensive, and is complicated by the important factors like the surface conditions of the specimen, the type of the stress variation, and the influence of the shape of the specimen on the stress flow. As it is known as that the highly polished specimens withstand better fatigue than the normal fatigue ones. The most damaging type of stress variation is the complete reversal, which is between the limit ±Ïƒ for which the stress range is 2σ. Fluctuating stresses are less damaging, the standard case is between the limit 0 and +σ. For some materials such as aluminium, no endurance limit exists and therefore it should be planned lifetime of the structure to be less than the failure point.
http://htmlimg1.scribdassets.com/izqlx4lamohzwzk/images/10-d0617ea942/000.jpg
Figure 1.2 Fluctuating Stress Cycle.
The above figure illustrates repeated stress cycle in which σmax (Rmax) is the maximum stress and σmin (Rmim) is the minimum stress and both are not equal. Here t is the time and σa is the stress amplitude and σm is the mean stress.
In low fatigue cycle region (N<104 or 105 cycles), tests are conducted with controlled cycles of plastic and elastic strain instead of stress cycle. The basic procedure of determining an S-N curve is to test the first specimen at a high stress where failure is expected within a less number of cycles, e.g., at around two-thirds the static tensile strength of the material. The applied load/stress is decreased for each succeeding specimen until one or two specimens do not fail under the specified number of cycles, which is at least 107 cycles. The highest stress or load at which the specimen runout (non-failure) that point or limit is taken as a fatigue limit of the material. The materials without a fatigue limit the test is usually terminated for practical consideration at low stress at about 108 or 5x108 cycles. (www.key-to-metals.com.cn)
In this experiment fatigue test for aluminium alloys of series 2000 have been conducted and described.
S-N Curve Experiment for 6000 and 2000 Aluminium Alloys Series:
The fatigue failure experiment is carried out for two different types of aluminium alloys i.e. 6082 and 2011 specimens. These experiments are carried in two different groups.
A typical standard specimen is shown in Fig 1.3 as below. It is recommended to test at least 10 specimens of each type and they all must be cut from one length of the material.
http://static.tecquipment.com/Products/RF1020_ALUMINIUM-FATIGUE-SPECIMEN.jpg
Figure 1.3 Test Specimen.
A set of bending stresses from 0.9 of the yield or proof stress to 0.4 of the ultimate strength is selected to match the number of the test specimens for the complete experiment. The setting up of the specimen on the machine is a reasonably simple operation which is done in proper methodology. The main object is to align the specimen and loading arm with the axis of rotation to eliminate stresses due to eccentric whirling of the specimen.
Both in drive shaft and the loading arm chucks, loose collet grip is inserted. These inserts 9mm diameter ends of the test specimen are slid as shown in Fig. 1.4.
http://www.twi.co.uk/twiimages/jk78f1.jpg
Figure 1.4 Setting up of machine. ( by http://www.twi.co.uk/content/jk78.html)
The collet is first tighten on the drive shaft chuck so that so that about 1 mm shoulder shows between the start of the neck and the face of the collet of the specimen. Then the loading arm is pushed on to the end of the specimen and adjusts the collet to give a sliding fit. The position of loading the loading arm is in such a way that the dimension of 109.5 mm is attained from the rear face of the bearing housing to the adjacent end of the neck of the specimen as shown in Fig 1.4 and finally tight the collet with the spanner. The specimen is rotated to check that the end of the cantilever run axially otherwise the specimen must get bend and can be discarded.
Bearing
Drive shaft and bearing
Electric motor
Chuck in which specimen is fitted.
ON/OFF SwitchC:UsersasimDesktopall folderpicsmaterialsimagesattachments_16_12_2010DSC01501.JPG
Figure 1.4 Rotating Fatigue Machine
The counterbalance and load hangers should be ensured are in place. Switch motor ON and OFF to verify smooth running. The bending stress for the test is selected and required load or weight is applied on the load hanger. The revolution counter is set to zero before starting the machine and safety guard is used over the apparatus. The fracture time which might occur is estimated and noted.
Endurance Limit:
The stress value below which the material will withstand many number of load cycles. It is also known as fatigue limit. “The stress level below which a specimen will withstand cyclic stress indefinitely without exhibiting fatigue failure. Rigid, elastic, low damping materials such as thermosetting plastics and some crystalline thermoplastics do not exhibit an endurance limit. Also known as FATIGUE LIMIT.”
(CRC Press LLC 1989)
Ultimate stress:
It is defined as the maximum/ultimate load under which a specimen breaks or fails. Stress corresponding to ultimate load is ultimate stress.
Mean Stress:
It is defined as the algebraic sum of maximum and minimum stress divided by 2.
Ultimate Tensile Strength:
In the given experiment ultimate tensile strength is calculated using the following formula
p. Where p is the load applied to the material and is the stress.
Recorded Data and Graph for 6000 series:
Sample No.
Load (N)
Bending Stress (MPa)
Result No. of Cycle
Results (Fail/No- Fail)
1
11.25
225
21400
fail
2
10.6
212
114800
fail
3
9.95
199
115300
fail
4
9.3
186
293800
fail
5
8.65
173
161000
fail
6
8
160
184700
fail
7
7.35
147
905100
fail
8
6.7
134
2411100
fail
9
6.05
121
2765800
fail
10
5.9
118
3156700
fail
Recorded Data and Graph for 2000 series:
Sample No.
Load
(N)
Bending Stress (MPa)
Result No. of cycles (x100)
Result (Fail/Not-fail)
1
13.5
270
95
Fail
2
12
240
407
Fail
3
10.5
210
482
Fail
4
9
180
1948
Fail
5
8.25
165
1781
Fail
6
7.5
150
2662
Fail
7
7.0
140
2165
Fail
8
6.0
120
4916
Fail
9
5.6
112
19970
Fail
10
5.2
104
More than 107
Not-fail
11
5
100
10 533
Fail
12
4.8
96
More than 107
Not-fail
13
4.6
92
14
4.6
92
More than 107
Not-fail
15
4
80
More than 107
Not-fail
2. Non-Destructive Testing Methods:
2.1 Introduction:
Non-destructive testing (NDT) is a wide range of analysis technique used in science and industries to evaluate the properties of material or component and to detect the flaws in the material without causing damage. The non-destructive testing is the testing of the materials used to find surface or internal flaws or metallurgical conditions without interfering in the integrity of the materials. The flaw includes cracks or inclusions in welds and castings, or variations in structural properties which may lead to loss of strength and finally failure of materials. Non-destructive testing is used for measurement of components and spacing and for the measurement of physical properties such as internal stress and hardness. It is also used for in-service inspection and for conditions monitoring of operating plants. It is also used to look for sign of wear or internal changes on airplanes in aircraft industries. The NDT method is also a function part of quality control which is based on sampling analysis, this method is not just for rejecting the substandard material but gives assurance that it is good.
The common types of Non-Destructive Testing are stated as follows:
Magnetic Particle Inspection.
Figure 2.1 Magnetic Particle Testing http://www.azom.com/work/8is7fjkADJ5v0JQByKTw_files/image003.gif
Radiography Inspection.
Figure 2.2 Radiography Testing.
Ultrasonic Testing.
Figure 2.3 Ultrasonic Testing.
Liquid Penetrant Testing.
Figure 2.4 Liquid Penetrant Testing.
http://www.twi.co.uk/twiimages/ksijm001f1.gif
Eddy Current Testing and Electro Magnetic Testing.
Figure 2.5 Electro Magnetic Testing.http://www.eurondt.com/index_2.gif
There are different types of non-destructive testing used for removing flaws as shown above but two main types of NDT on which experiments are perform are Magnetic particle testing and Ultrasonic Testing.
2.2 Magnetic Particle Testing:
Introduction:
Magnetic particle testing is type of non-destructive testing which is used for the detection of surface and near-surface flaws in the ferromagnetic materials and it is basically used for crack detection. Such flaws present in the magnetized part will cause a magnetic field, i.e. flux, to leave the part. It is however same as if there is a surface-breaking flaw in the specimen, the magnetic field is distorted, causing local magnetic flux leakage around the flaw. If the magnetic particles are applied to the surface of this specimen, the surface is covered by very fine iron particles and they will be held in the place by the flux leakage to give a visual indication.
Figure 2.6 Deflection in the magnetic flux.http://www.ndt-ed.org/GeneralResources/MethodSummary/MT1.jpg http://www.milinc.com/images/magpartimgs/magpartdiag1.gif
Thus a crack is indicated as a line or iron powder particles on the surface. The method of MPT is applicable to all metals which can be strongly magnetised such as ferritic steels, irons and some other alloys but not generally to austenitic steels. The modern equipments generate the magnetic field electrically either directly or indirectly. In direct method high ampere of current is passed through the specimen and magnetic flux is generated at right angle to the current flow. Therefore current flow is in the same direction of suspect defect. If this method is not possible to carry out because of the orientation of the defect, then the indirect method is used. This consists of two forms:
Passing high current through a coil that encircles the specimen.
Making the test piece form part of a yoke, this is surrounded by a current carrying coil. The effect is to pass magnetic flux along the part to reveal transverse and circumferential defects.
Flux:
It is a term which is used to refer the amount of magnetic field that exist at specific point within that field. It is measured in Webers.
Flux Density:
It is an indication of the strength of the magnetic field. It is represented by the lines of forces which are surrounded around the magnetic circuit, where the lines are closer to the flux. It is measured in Webbers/M2 or Telsa.
Magnetic lines of force:
These lines of force are imaginary lines which describes the path a free north pole would take in a magnetic field. These lines can be plotted using a compass.
Experimental Procedure:
Pump switch.The equipment used for performing the experiment is known as Johnson Allen NDT (SBU 2000). It should be sited on a firm concrete allowing access for servicing. The specimen provided contains five holes in it. The equipment should be connected to a 230 volt AC, 50Hz, 16 Amp power supply and “Quick blow” fuses should not be used. Connect Footswitch, Pump, and UV Light via socket to the equipment. Turn main supply on at the wall isolator. Turn the UV light ON and allow 10 mins to warm up. Slide out the ink hopper to within 75mm of the top flange with the Fluorescent Magnetic Ink and slide in the hopper gently. Turn the agitation ON and dispensing pump using the twist switch on the right of equipment.
Figure 2.7 Magnetic Particles Testing Machine.
Headstocks.
Ink spray.
Clamping knobs.
C.F. control knob.
Ammeter
ON/OFF Switch.C:UsersasimDesktopall folderpicsmaterialsimagesIMG_0049.JPG
Headstocks can be adjusted using the Clamping Knobs at the base. Once specimen is inserted and left side headstock is fixed then the specimen is ready to be tested. AC is current is used for surface defect and HWDC is used for all defect, so select the HWDC. Check the current control dial is turned to minimum.
Experimental Precautions:
The hole inside the given specimen should be clear otherwise the crack line does not come properly.
Headstocks should be regularly checked.
The specimen should be cleaned properly to prevent arcing and pitting on the surface.
Circuits should be check regularly before performing any operation.
Experimental Results:
SpecimenThe given specimen contains 5 holes in it. The specimen is adjusted and fixed between the headstocks as shown in the figure below. Before performing the experiment the specimen should be clean properly other the line is not visible. As the hole was not clear and contains ink in it so the defect line is not obtained on the specimen. This can be seen in the figure below.
Figure 2.8 Specimen without defect line
Headstocks
Ink spray.C:UsersasimDesktopall folderpicsmaterialsimagesattachments_16_12_2010DSC01541.JPG
Then the holes are clean, and the experiment is performed again. When it is clean, the two defect lines are visible on the specimen. Hence, as the lines are seen on the specimen it can considered that it contains flaws in it. This works under the principle of right hand rule and so specimen has to be adjusted perpendicular to the headstocks so flux travel properly from the specimen and flaw.
Testing Specimen.
Defect lines.
Figure 2.8 Specimen with visible defect line.C:UsersasimDesktopall folderpicsmaterialsimagesIMG_0060.JPG
Advantages of Magnetic Particles Testing:
It gives instant result and rapid inspection can be done for large surfaces.
It is simple and easy to conduct.
It detects a variety of surface and sub surface flaws, such as crack, porosity, inclusions, shrinkage, laps etc.
Surface preparation is less critical than it is penetrant inspection.
Sensitivity of testing can be specified and checked.
It is economically cheap.
Disadvantages of Magnetic Particles Testing:
High currents applied to the components may cause damage.
Smooth surface is required for application of this method.
Materials which are tested must be ferromagnetic.
Deep cracks or flaws are not detected.
Materials may need to be demagnetized.
Equipment is bulky and heavy.
Material or part permeability may affect results.
High power supply is needed for low surface.
2.3 Ultrasonic Testing:
Introduction:
Ultrasonic testing is a type of non-destructive testing method that uses high frequency sound waves (ultrasonic) which are above the range of human hearing and they are used to measure the geometric and physical properties of the materials. Ultrasonic waves travels at different velocities in different materials. There are different ways of sound travels through the material. One type of sound wave is called as compression or longitudinal travels which is at about 330m/s in air and 6400m/s in aluminium and approximately 5960m/s for steel.
Figure 2.9 Ultrasonic Testing.http://www.energyworkforce.net/wp-content/uploads/ut1.jpg
The pulsed beams of the ultrasound are used in a simplest instrument and a single probe (transducer) which is hand-held is paced on the surface of the specimen. An oscilloscope is connected to the probe which displays with the time-base the time that requires for an ultrasonic pulse to travel through the reflector which can be flaw, black surface etc. The height of the reflected pulse is related to the flaw size as seen from the transmitter probe which is displayed on oscilloscope screen. A single probe acts as both transmitter and receiver and hence the inspection can be done from one side of the specimen. Large grain material such as austenitic steel welding, copper casting etc., produce severe attenuation and are difficult to test but fine grain material such as forged material can be tested easily. Using the indication on the oscilloscope the size of the flaw can be determined. Ultrasonic attenuation and ultrasonic velocity measurements are used to study various material properties.
Experimental Procedure:
A typical ultrasonic testing system consists of several functional units such as:
Pulser/receiver.
Transducer.
Oscilloscope.
Connecting wires.
Couplant.
Mild steel material (specimen).
A pulser is an electronic device that can produce high voltage electrical pulses. From the high voltage pulses, transducer generates high frequency ultrasonic energy. This sound energy propagates through the material in the form of waves.
Couplant.
Transducer.
Oscilloscope
Figure 2.10 Setting up of Ultrasonic Testing.C:UsersasimDesktopall folderpicsmaterialsimagesattachments_16_12_2010DSC01547.JPGC:UsersasimDesktopall folderpicsmaterialsimagesattachments_16_12_2010DSC01545.JPG
Mild steel specimen.
Defects
Display Screen
The left side figure shows the cathode ray oscilloscope and the right side figure shows the mild steel material on the ultrasonic test is conducted. The given specimen was of mild steel with five holes in it. Before starting the experiment, the specimen should be properly clean. Then apply a couplant on it. Start the cathode ray oscilloscope. There are different button on oscilloscope such as gain, light, velocity, angle, dialog etc. as shown in figure below.
Figure 2.11 various buttons on oscilloscope.C:UsersasimDesktopall folderpicsmaterialsimagesultrasonic 8122010(002).jpg
So it is adjusted in such a way that the deflection should be visible. The transducer is connected to the oscilloscope with the help of connecting wires. Then this transducer is placed on the specimen for testing the flaws in it.
Experimental Observation:
When the transducer is placed on the specimen it emits ultrasonic waves from the material (specimen). When the probe is placed on specimen it gives a very nice and high peak on the oscilloscope as shown in the below figure. It means that the ultrasonic waves travel easily from that part of the specimen and hence it does not have any flaw in that part.
Figure 2.12 High peak signal in nice surface.C:UsersasimDesktopall folderpicsmaterialsimagesIMG_0065.JPG C:UsersasimDesktopall folderpicsmaterialsimagesultrasonicP081210_11.230001.jpg
Similarly, when the probe is moved further on it can be seen in the oscilloscope that the highest peak is coming down and it nearly become nil or very low. It means that the high frequency waves are reflected. Hence it can be observed that, there is a flaw in that part of the specimen. It is the same part under which a big hole is there.
Figure 2.13 Low peak signal in defective part.C:UsersasimDesktopall folderpicsmaterialsimagesIMG_0081.JPG C:UsersasimDesktopall folderpicsmaterialsimagesattachments_16_12_2010DSC01545.JPG
Hence, from the above two figure it can be concluded that when there is high peak in oscilloscope there is no flaw at that part and similarly when the high peak decreases to low it means that part contains some flaw in it.
Advantages of Ultrasonic Testing:
The measurement is superior to other NDT methods.
It is sensitive to both surface and subsurface discontinuities.
It requires very minimal part preparations.
It is highly accurate in determining reflector position and estimating size and shape.
As the electronic equipment as used it give instantaneous results.
It requires only single-sided access when the pulse-echo technique is used.
It can also be used for other purposes such as thickness measurement etc.
Disadvantages of Ultrasonic Testing:
It normally requires a coupling medium to promote the transfer of sound energy into the test specimen.
Those materials which are rough, very small, irregular in shape, thin or not homogeneous are very difficult to test.
Surface must be accessible to transmit ultrasound.
Cast iron and other grained materials are difficult to test due to their low sound transmission and high signal noise.
Linear defects oriented parallel to the sound beam may go undetected.
Feedback and Evaluation:
S-N curve:
The experiment S-N curve is used to determine the endurance fatigue limit of the material used in automotive industries. The given specimens were of aluminium alloys. There were two series of aluminium alloys i.e. 6000 series and 2000 series on which experiments are performed. Ten samples of each specimen were given. This experiment was performed in two groups. One group perform the experiment on 6082 specimen and other group perform the experiment on 2011 specimen. In this experiment weights are taken in Newton and the number of cycles is shown on the machine. Corresponding to this stress/load (S) and number of cycles (N), S-N curve is plotted. When the load is high it number of cycles is low and when the load is decreased its number of cycles increases. This is shown in graphs below.
Find Out How UKEssays.com Can Help You!
Our academic experts are ready and waiting to assist with any writing project you may have. From simple essay plans, through to full dissertations, you can guarantee we have a service perfectly matched to your needs.
View our academic writing services
This is a S-N curve graph for 6011 aluminium alloy (specimen). From the above graph it can be seen that at high stress/load i.e. 225 MPa or 11.25N. The number of cycles is 21400. As the stress goes on increasing the number of cycles goes on decreasing. Hence the endurance fatigue limit varies at different load for the same specimen. When the specimen fails it is taken out from the chuck and it can be checked microscope and the cracks can be seen properly. From this the physical properties of the material can be checked and the fatigue limit can be obtained.
There are some microscopic images of specimens, they are as follows:
Sample 1 at 11.25N load.
C:UsersasimDesktopall folderpicsmaterialsimagesultrasonicsamplemicroscopicimages- sncurveSample1_1.JPGC:UsersasimDesktopall folderpicsmaterialsimagesultrasonicsamplemicroscopicimages- sncurveSample1_2.JPG
Sample 2 at 10.6N load.
C:UsersasimDesktopall folderpicsmaterialsimagesultrasonicsamplemicroscopicimages- sncurveSample2_1.JPGC:UsersasimDesktopall folderpicsmaterialsimagesultrasonicsamplemicroscopicimages- sncurveSample2_2.JPG
There are many more microscopic images which look similar to the above images and it fails at different loads. The below figure is of the specimen before experiment.
http://static.tecquipment.com/Products/RF1020_ALUMINIUM-FATIGUE-SPECIMEN.jpg
This specimen is fixed between the chucks and when the experiment is performed at different loads the specimen which is fixed between the chuck breaks. This is shown in the figure below.
C:UsersasimDesktopall folderpicsmaterialsimagesIMG_0040.JPG
It can be clearly seen from the previous figure that the specimen breaks from its middle. First the crack propagates in it and then its finally fails. The graph below is the S-N curve graph for 2011 aluminium alloy specimen which is experimented by other group.
This graph start from 270Mpa and the applied on the specimen is 13.5 N and therefore the number of cycles perform by the specimen is less i.e. 9500. These two specimens cannot be compared as both the group has taken different values of stress and therefore the number of cycles are different both the specimens. From the given ten samples, the discussion of the fatigue limit is nearly impossible as we find the scatter point graph not a perfect curve of S-N. For establishing the endurance fatigue limit of both the specimens more experiments must be conducted on it. Also the specimen should be properly fixed in the chuck otherwise it fails in somewhere before the expected value.
Magnetic Particles Testing:
This experiment is performed to check the flaws i.e. cracks, holes, black surface of the materials. This experiment comes under non-destructive testing. A specimen of mild steel is given and it contains five holes in it. This specimen is fixed in Johnson Allen NDT machine. This specimen is fixed between the headstocks of the machine. The specimen is fixed perpendicular to the headstocks. The perpendicular arrangement of headstock and specimen is done because this machine works under the principle of right hand rule and it generates flux into the specimen. This flux passes through the material. When there is any flaw into the material the flux gets affected and a black line of iron particles is seen on the surface of the material. In this experiment also the specimen contains five holes in it and a black line of iron particles is seen on the surface of the material. The current is control with the help of C.F. control switch. The black defect line on the specimen can be seen in figure below.
The amount of flux running through the specimen can be measured with the help of ammeter. But the ammeter in the machine under which this testing is performed was not working. So the exact amount of flux generated in specimen cannot be measured. The ammeter and C.F. control switch is shown in the figure below.
C.F. control switch.
AmmeterC:UsersasimDesktopall folderpicsmaterialsimagesIMG_0045.JPG
Ultrasonic testing:
This type of testing is also used to detect flaws and crack inside the material without damaging it. The specimen provided is a mild steel slab and it contains holes in it. It can be seen that when there is no flaw inside the specimen it shows high peak on the oscilloscope and similarly when there is any flaw the ultrasonic waves get reflected and the high peak decreases to low peak. This is clearly in figures below.
C:UsersasimDesktopall folderpicsmaterialsimagesIMG_0065.JPGC:UsersasimDesktopall folderpicsmaterialsimagesIMG_0081.JPG
Conclusion:
From the above experiments we can conclude that S-N curve is best method to determine the fatigue limit but it can’t be done using ten samples as it does not gives the proper curve graph for the specimen. Magnetic Particle Testing is a good way of finding flaws from the materials without damaging it but the machine is not that comfortable and more precautions must be used before using it otherwise the flaws are not detected. Similarly Ultrasonic Testing is a good way cracks and other flaws detecting and these are used in aircraft industries.
Referencing:
Mark Wilcox and George Downes, A brief description of NDT techniques Available at: http://www.turkndt.org/sub/makale/ornek/a%20brief%20description%20of%20NDT.pdf [Accessed on:3 Dec 2010]
www.ndted.org.com NDT Method Summary Available at: http://www.ndt-ed.org/GeneralResources/MethodSummary/MethodSummary.htm [Accessed on:12 Dec 2010]
www.insight-ndt.com (2007). ‘Qualiron’ ductile iron metal quality tester. Available: http://www.insight-ndt.com/ultrasonic/qualiron.html. [Last accessed 3rd dec 2010.]
www.NDT.net Non-destructive Material Testing with Ultrasonic’s
– Introduction to the Basic Principles – Available at: http://www.ndt.net/article/v05n09/berke/berke1.htm [Accessed on:15 Dec 2010]
www.materialsengineer.com (2007) Metallurgical Fatigue Failures Available at
http://www.materialsengineer.com/CA-fatigue.htm [Last accessed on 3rd Dec 2010]
www.sv.vt.edu (1997). By Shawn M. Kelly, Fatigue, Available at http://www.sv.vt.edu/classes/MSE2094_NoteBook/97ClassProj/anal/kelly/fatigue.html [Accessed on 6 Dec 2010].
www.scribd.com Fatigue Failure Available at
http://www.scribd.com/doc/29476995/fatigue-failure [Accessed on 8 Dec].
www.key-to-metals.com.cn (99-2000) Fatigue of metals (stress cycles) Available at http://www.key-to-metals.com.cn/page.aspx?ID=CheckArticle&site=kts&NM=281 [Accessed on 8 Dec].
www.twi.co.uk (2005) Fatigue Testing Available at http://www.twi.co.uk/content/jk78.html [Accessed on 12 Dec].
www.azom.com (2010) Non-destructive Testing-Surface Examination Techniques. http://www.azom.com/Details.asp?ArticleID=522 [Accessed on 13 Dec].
www.energyworkforce.net (2010) Principles of Ultrasonic Testing. Available at http://www.energyworkforce.net/?p=126 [Accessed on 16 Dec].
Cite This Work
To export a reference to this article please select a referencing style below:
