What was the load on the component at the time of failure? Was the correct material used and manufacturing/processing sound? Was the part designed properly? Did the environment influence the failure? Of course, many situations may involve thin sections and/or very ductile materials, where the conditional constraint for LEFM may not apply. In this case, the measure of toughness is plane-stress fracture toughness(K)and requires the use of elastic-plastic fracture mechanics(EPFM), as the process of unstable fracture involves some plasticity. Plane-stress fracture toughness(Ke)is higher than plane-strain fracture toughness(Kl), but when thinner sections and more ductile materials are involved net-section instability becomes a factor. See the appendix in the article Fracture Appearances and Mechanisms of Deformation and fracture for more information on the use of fracture mechanics in failure analysis Metallographic examination Metallographic examination of polished and of polished-and-etched sections by optical microscopy and by electron- optical techniques is a vital part of failure investigation and should be carried out as a routine procedure when possible Metallographic examination provides the investigator with a good indication of the class of material involved and its structure. If abnormalities are present, these may be associated with undesirable characteristics that predispose to early failure. It is sometimes possible to relate them to an unsuitable composition or to the effects of service, such as aging in low-carbon steel that has caused precipitation of iron nitride, or gassing in copper. Microstructural examination may also provide information as to the method of manufacture of the part under investigation. It can reveal the heat treatment and possible deficiencies in heat treatment such as decarburization at the surface. Microstructural inspection can also reveal possible overheating through coarsening of carbides of superalloys and solution and precipitation of manganese sulfide Other service effects, such as corrosion, oxidation, and severe work hardening of surfaces, also are revealed, and their extent can be investigated. The topographical characteristics of any cracks, particularly their mode of propagation, can be determined, for example, transgranular or intergranular. This provides information that can be helpful in distinguishing between different modes of failure. For example, fatigue cracks always propagate perpendicular to the maximum cyclic tensile stress while stress-corrosion cracking may propagate along grain boundaries Only a few general directions can be given as to the best location from which to take specimens for microscopic examination, because almost every failure has individual features to be taken into account. In most examinations however, it must be determined whether the structure of a specimen taken adjacent to a fracture surface or a region at which a service defect has developed is representative of the component as a whole. This can be done only by the examination of specimens taken from the failure region and specimens taken from other locations. For instance, in the case of ruptured or bulged boiler tubes in which failure is usually restricted to one portion only, it is desirable to examine specimens taken from both sides of the fracture, from a location opposite the affected zone, and also from an area as remote from the failure as the size of the sample permits in order to determine whether the failure has been due to a material defect or to overheating-and, if the latter, whether this was of a general or localized nature. In investigations involving general overheating, sometimes the original condition of the material can be ascertained only from a sample cut from a part of the tube many feet away from the affected zone Metallographic specimens should be taken perpendicular to the fracture surface, showing the fracture surface in edge w. In cases where metal cleanliness may be an issue, the specimen orientation must be selected properly to determine inclusion density and morphology. This type of examination must be performed on the unetched metallographic In the investigation of fatigue cracks, it may be desirable to take a specimen from the region where the fracture originated to ascertain if the initial development was associated with an abnormality such as a weld defect, a decarburized surface, a zone rich in inclusions, or, in castings, a zone containing severe porosity. Multiple fatigue-crack initiation is very typical of both fretting and corrosion fatigue and may form in areas where there is constant stress across a section. Similarly, with surface marks, where the origin cannot be identified from outward appearances, a microscopic examination will show whether they occurred in rolling or arose from ingot defects, such as scabs, laps, or seams. In brittle fractures, it is useful to examine a specimen cut from where the failure originated, if this can be located with certainty. Failures by brittle fracture may be associated with locally work-hardened surfaces, arc strikes, local untempered martensite, and so forth For good edge retention when looking at a fracture surface, it is usually best to plate the surface of a specimen with a Ei lishing and can be included in the examination. Alternative means include hard metal or nonmetal particles embedded the mount adjacent to the edge Analysis of Metallographic Sections. As with hardness testing and macroscopic examination, the examination of metallographic sections with a microscope is standard practice in most failure analyses, because of the outstanding capability of the microscope to reveal material imperfections caused during processing and of detecting the results of a variety of in-service operating conditions and environments that may have contributed to failure. Inclusions, Thefileisdownloadedfromwww.bzfxw.com
· What was the load on the component at the time of failure? · Was the correct material used and manufacturing/processing sound? · Was the part designed properly? · Did the environment influence the failure? Of course, many situations may involve thin sections and/or very ductile materials, where the conditional constraint for LEFM may not apply. In this case, the measure of toughness is plane-stress fracture toughness (Kc) and requires the use of elastic-plastic fracture mechanics (EPFM), as the process of unstable fracture involves some plasticity. Plane-stress fracture toughness (Kc) is higher than plane-strain fracture toughness (KIc), but when thinner sections and more ductile materials are involved net-section instability becomes a factor. See the appendix in the article “Fracture Appearances and Mechanisms of Deformation and Fracture” for more information on the use of fracture mechanics in failure analysis. Metallographic Examination Metallographic examination of polished and of polished-and-etched sections by optical microscopy and by electronoptical techniques is a vital part of failure investigation and should be carried out as a routine procedure when possible. Metallographic examination provides the investigator with a good indication of the class of material involved and its structure. If abnormalities are present, these may be associated with undesirable characteristics that predispose to early failure. It is sometimes possible to relate them to an unsuitable composition or to the effects of service, such as aging in low-carbon steel that has caused precipitation of iron nitride, or gassing in copper. Microstructural examination may also provide information as to the method of manufacture of the part under investigation. It can reveal the heat treatment and possible deficiencies in heat treatment such as decarburization at the surface. Microstructural inspection can also reveal possible overheating through coarsening of carbides of superalloys and solution and precipitation of manganese sulfide. Other service effects, such as corrosion, oxidation, and severe work hardening of surfaces, also are revealed, and their extent can be investigated. The topographical characteristics of any cracks, particularly their mode of propagation, can be determined, for example, transgranular or intergranular. This provides information that can be helpful in distinguishing between different modes of failure. For example, fatigue cracks always propagate perpendicular to the maximum cyclic tensile stress while stress-corrosion cracking may propagate along grain boundaries. Only a few general directions can be given as to the best location from which to take specimens for microscopic examination, because almost every failure has individual features to be taken into account. In most examinations, however, it must be determined whether the structure of a specimen taken adjacent to a fracture surface or a region at which a service defect has developed is representative of the component as a whole. This can be done only by the examination of specimens taken from the failure region and specimens taken from other locations. For instance, in the case of ruptured or bulged boiler tubes in which failure is usually restricted to one portion only, it is desirable to examine specimens taken from both sides of the fracture, from a location opposite the affected zone, and also from an area as remote from the failure as the size of the sample permits in order to determine whether the failure has been due to a material defect or to overheating—and, if the latter, whether this was of a general or localized nature. In investigations involving general overheating, sometimes the original condition of the material can be ascertained only from a sample cut from a part of the tube many feet away from the affected zone. Metallographic specimens should be taken perpendicular to the fracture surface, showing the fracture surface in edge view. In cases where metal cleanliness may be an issue, the specimen orientation must be selected properly to determine inclusion density and morphology. This type of examination must be performed on the unetched metallographic specimen. In the investigation of fatigue cracks, it may be desirable to take a specimen from the region where the fracture originated to ascertain if the initial development was associated with an abnormality, such as a weld defect, a decarburized surface, a zone rich in inclusions, or, in castings, a zone containing severe porosity. Multiple fatigue-crack initiation is very typical of both fretting and corrosion fatigue and may form in areas where there is constant stress across a section. Similarly, with surface marks, where the origin cannot be identified from outward appearances, a microscopic examination will show whether they occurred in rolling or arose from ingot defects, such as scabs, laps, or seams. In brittle fractures, it is useful to examine a specimen cut from where the failure originated, if this can be located with certainty. Failures by brittle fracture may be associated with locally work-hardened surfaces, arc strikes, local untempered martensite, and so forth. For good edge retention when looking at a fracture surface, it is usually best to plate the surface of a specimen with a metal, such as nickel, prior to mounting and sectioning, so that the fracture edge is supported during grinding and polishing and can be included in the examination. Alternative means include hard metal or nonmetal particles embedded in the mount adjacent to the edges. Analysis of Metallographic Sections. As with hardness testing and macroscopic examination, the examination of metallographic sections with a microscope is standard practice in most failure analyses, because of the outstanding capability of the microscope to reveal material imperfections caused during processing and of detecting the results of a variety of in-service operating conditions and environments that may have contributed to failure. Inclusions, The file is downloaded from www.bzfxw.com
microstructural segregation, decarburization, carbon pickup, improper heat treatment, untempered white martensite, econd phases such as y in nickel-base superalloys and intergranular corrosion are among the many metallurgical imperfections and undesirable conditions that can be detected and analyzed by microscopic examination of metallographic sections Even in the absence of a specific metallurgical imperfection, examination of metallographic sections is invaluable to the investigator in the measurement of microstructural parameters such as case depth, grain size, thickness of plated coatings and heat-affected zone size(hazhall of which may have a bearing on the cause of failure Mechanical testing Hardness testing is the simplest of the mechanical tests and is often the most versatile tool available to the failure analyst Among its many applications, hardness testing can be used to assist in evaluating heat treatment(comparing the hardness of the failed component with that prescribed by specification), to provide an estimate of the tensile strength of steel, to detect work hardening, or to detect softening or hardening caused by overheating, decarburization, or by carbon or nitrogen pickup. Hardness testing is also essentially nondestructive except when preparation of a special hardness test specimen is required, as in microhardness testing. Portable hardness testers are useful for field examination, but the type of hardness test must be appropriate for the sample. For example, Brinell is preferred over Rockwell for a gray cast iron part One must ensure the proper load is used for the test specimen thickness Other mechanical tests are useful in confirming that the failed component conforms to specification or in evaluating the effects of surface conditions on mechanical properties. Where appropriate, tensile and impact tests should be carried out, provided sufficient material for the fabrication of test specimens is available. After all photography, fractography, and nondestructive testing have been performed, the tensile properties of the failed component(s) may be tested. (It may be necessary to obtain permission for this destructive testing if litigation is involved. This involves sectioning the component and machining test specimens. (In all phases, from rough sectioning to the application of load and measurement of the final dimensions, it is advisable to photodocument the process. It is especially important to understand that the failed component may have been exposed to environmental conditions not experienced by exemplar components. For example, if the component was involved in a crash, there may have been a fire that exposed the component to temperatures that would have altered the mechanical properties. Likewise, the forces that acted to produce an overload failure in one area of the component may have also plastically deformed the component in other areas. To determine the mechanical properties of the component material, a location for testing must be chosen that has not been exposed to detrimental conditions. Here again, exemplar testing is a valuable tool for the failure analyst and should be employed when data from a failed component is skewed by environmental conditions Exemplar components are components that match the failed component (i.e, the same part number and hopefully the ame lot or batch). The use of exemplars can range from simply being a visual reference in a demonstration to being used for testing specimens to provide mechanical property data about a particular part or batch of parts. The closer the exemplars are to the failed component(geometrically and chronologically), the more reliable the comparison of test data If a raw material discrepancy is suspected, it may be that other components of the same batch or lot have the same properties. It may also be useful to have exemplars from other batches for The failure analyst should exercise care in interpreting mechanical test results. If a material has a tensile strength 5 to 10% below the minimum specified, this does not mean that low hardness or strength is the cause of its failure in service. Also it should be understood that laboratory tests on small specimens may not adequately represent the behavior of a much larger structure or component in service. For instance, it is possible for a brittle fracture of a large structure to occur at or near ambient temperature, while subsequent laboratory tests of Charpy or Izod specimens show a transition temperature well below-18C(0F). The effects of size in fatigue, stress-corrosion, and hydrogen-embrittlement testing are not well understood. However, on the basis of the limited evidence available, it appears that resistance to these failure processes decreases as specimen size increases. Several investigators have found correlation problems of transition-temperature type impact tests with service performance Occasionally, the mechanical properties may be acceptable over most of the component, but may vary at a bend or other discontinuity. Castings can have significant variations in properties from one location to another depending on the solidification practice for the casting. Thus, the location of the test specimen within the component can also be significant Mechanical property tests for cast components are frequently performed on coupons separately cast for this purpose Other factors may affect material properties results. Material may have been tested prior to forming or other deformation Subsequent coatings or case hardening may have improved or degraded mechanical properties. Variations from one location to another due to local material processing variations may help explain differences in mechanical properties between the bulk material and the failure origin Tensile tests, in many failure analysis investigations, do not provide enough useful information because relatively few failures result from metal that is deficient in tensile strength. Furthermore, samples cut from components that have failed in a brittle manner generally show adequate ductility under the conditions imposed during a tensile test
microstructural segregation, decarburization, carbon pickup, improper heat treatment, untempered white martensite, second phases such as γ′ in nickel-base superalloys and intergranular corrosion are among the many metallurgical imperfections and undesirable conditions that can be detected and analyzed by microscopic examination of metallographic sections. Even in the absence of a specific metallurgical imperfection, examination of metallographic sections is invaluable to the investigator in the measurement of microstructural parameters such as case depth, grain size, thickness of plated coatings, and heat-affected zone size (HAZ)—all of which may have a bearing on the cause of failure. Mechanical Testing Hardness testing is the simplest of the mechanical tests and is often the most versatile tool available to the failure analyst. Among its many applications, hardness testing can be used to assist in evaluating heat treatment (comparing the hardness of the failed component with that prescribed by specification), to provide an estimate of the tensile strength of steel, to detect work hardening, or to detect softening or hardening caused by overheating, decarburization, or by carbon or nitrogen pickup. Hardness testing is also essentially nondestructive except when preparation of a special hardness test specimen is required, as in microhardness testing. Portable hardness testers are useful for field examination, but the type of hardness test must be appropriate for the sample. For example, Brinell is preferred over Rockwell for a gray cast iron part. One must ensure the proper load is used for the test specimen thickness. Other mechanical tests are useful in confirming that the failed component conforms to specification or in evaluating the effects of surface conditions on mechanical properties. Where appropriate, tensile and impact tests should be carried out, provided sufficient material for the fabrication of test specimens is available. After all photography, fractography, and nondestructive testing have been performed, the tensile properties of the failed component(s) may be tested. (It may be necessary to obtain permission for this destructive testing if litigation is involved.) This involves sectioning the component and machining test specimens. (In all phases, from rough sectioning to the application of load and measurement of the final dimensions, it is advisable to photodocument the process.) It is especially important to understand that the failed component may have been exposed to environmental conditions not experienced by exemplar components. For example, if the component was involved in a crash, there may have been a fire that exposed the component to temperatures that would have altered the mechanical properties. Likewise, the forces that acted to produce an overload failure in one area of the component may have also plastically deformed the component in other areas. To determine the mechanical properties of the component material, a location for testing must be chosen that has not been exposed to detrimental conditions. Here again, exemplar testing is a valuable tool for the failure analyst and should be employed when data from a failed component is skewed by environmental conditions. Exemplar components are components that match the failed component (i.e., the same part number and hopefully the same lot or batch). The use of exemplars can range from simply being a visual reference in a demonstration to being used for testing specimens to provide mechanical property data about a particular part or batch of parts. The closer the exemplars are to the failed component (geometrically and chronologically), the more reliable the comparison of test data. If a raw material discrepancy is suspected, it may be that other components of the same batch or lot have the same properties. It may also be useful to have exemplars from other batches for comparison testing. The failure analyst should exercise care in interpreting mechanical test results. If a material has a tensile strength 5 to 10% below the minimum specified, this does not mean that low hardness or strength is the cause of its failure in service. Also, it should be understood that laboratory tests on small specimens may not adequately represent the behavior of a much larger structure or component in service. For instance, it is possible for a brittle fracture of a large structure to occur at or near ambient temperature, while subsequent laboratory tests of Charpy or Izod specimens show a transition temperature well below -18 °C (0 °F). The effects of size in fatigue, stress-corrosion, and hydrogen-embrittlement testing are not well understood. However, on the basis of the limited evidence available, it appears that resistance to these failure processes decreases as specimen size increases. Several investigators have found correlation problems of transition-temperaturetype impact tests with service performance. Occasionally, the mechanical properties may be acceptable over most of the component, but may vary at a bend or other discontinuity. Castings can have significant variations in properties from one location to another depending on the solidification practice for the casting. Thus, the location of the test specimen within the component can also be significant. Mechanical property tests for cast components are frequently performed on coupons separately cast for this purpose. Therefore, results from samples from the casting itself may not be directly comparable. Other factors may affect material properties results. Material may have been tested prior to forming or other deformation. Subsequent coatings or case hardening may have improved or degraded mechanical properties. Variations from one location to another due to local material processing variations may help explain differences in mechanical properties between the bulk material and the failure origin. Tensile tests, in many failure analysis investigations, do not provide enough useful information because relatively few failures result from metal that is deficient in tensile strength. Furthermore, samples cut from components that have failed in a brittle manner generally show adequate ductility under the conditions imposed during a tensile test
Sometimes, however, there is justification for tensile testing of failed components to eliminate poor-quality material as a possible cause of failure. Often, these tensile tests for determining material quality are carried out by manufacturers and suppliers when examining components that have been returned to them for analysis The role of directionality in tensile testing of wrought metals should also be considered Specimens cut transversely to the longitudinal axis of a component(such as a shaft, plate, or sheet )usually give lower tensile and ductility values than those It along the longitudinal axis. This is due to the marked directionality and the resulting anisotropy produced during rolling or forging. When sectioning tension-test specimens from the failed component, special attention should be paid to the orientation of the specimen. Some components may have quality-assurance notes on the drawings that indicate where tensile specimens are to be taken. This is especially true of critical aircraft components. Anisotropic materials have properties that vary with test specimen orientation. Tensile-strength and yield-strength specifications are usually given in the longitudinal and transverse directions. Typically, and unless otherwise specified, the tensile specimen should be taken with its major axis parallel to the direction of grain flow; however, test specimens are typically taken in two of the three directions: longitudinal and long transverse.( The short transverse direction is typically not tested since it is difficult to obtain specimens of sufficient length in that direction. It may be necessary to lightly polish the component surface and parent component by first cutting a rough specimen shape, then final machining to the specified form. emoved from the use a macroetchant for the material being worked with to determine the grain flow. The specimens are removed from the Residual stresses in the component may result in warped specimens or pinched cutting tools during rough machining Care should be taken to document the residual stress observations, so that comparison to exemplar parts sectioned in the ame manner can be made. Hopefully, final machining of the testpiece will correct the warped shape while still allowing the specimen to run parallel to the grain flow. Straightening of a warped tensile specimen during the test will result in a nonlinear indication on the initial portion of the stress-strain curve. This can be corrected to a line by using curve-fitting ftware or by drawing a line by eye back to the abscissa as shown in Fig. 5. There may also be nonuniformity in the specimen microstructure that affects the properties of the material, for example, a change in grain size due to cold work or changes in temper due to the haz of a weld nside of a curved specimen Extensometer located on the outside of a curved specimen to zoro stress Foot correction Fig. 5 Examples of stress-strain curves requiring foot correction. D, point where the extension of the straight(elastic)part diverges from the stress-strain curve. Source: Ref 1 It may be necessary to make some tests either at slightly elevated or at low temperatures to simulate service conditions Also, it may be helpful to test specimens after they have been subjected to particular heat treatments simulating those of the failed component in service to determine how this treatment has modified mechanical properties. For example treating a steel at a temperature in the embrittling range for about I h prior to impact testing will indicate any tendency to strain-age embrittlement. The determination of the ductile-to-brittle transition temperature may be useful in investigating brittle fracture of a low-carbon steel Component Proof Testing Critical components are sometimes proof tested after manufacture. This is especially true of components that will be subjected to large loads. Proof testing involves loading a compe onent commended operating limits, and possibly slightly into the yield zone. This helps to ensure there are no manufacturing or materials defects that would cause premature failure. Proof loads are usually determined on a case-by-case basis and may be expressed in several forms, such as proof load to twice the operating load or proof load to 90% of the determined yield strength. Extreme care should be taken not to damage the component in the specification of a proof load. For example, stretching a component may alter a designed preload by relieving residual compressive stresses introduced during manufacture. There are usually documents with details of specific methods, tools, and loads used in proof testing and results of tests on critical components. If performing proof testing on exemplar parts from a given manufacturer, will lend credibility to the results if the manufacturer's test procedures are followed. If the proof tests cite standard methods, such as ASTM, ISO, or other industry-accepted procedure, acquire a copy of the method and follow it closely. If results vary from those reported by the manufacturer, it may be necessary to prove the results are valid Thefileisdownloadedfromwww.bzfxw.com
Sometimes, however, there is justification for tensile testing of failed components to eliminate poor-quality material as a possible cause of failure. Often, these tensile tests for determining material quality are carried out by manufacturers and suppliers when examining components that have been returned to them for analysis. The role of directionality in tensile testing of wrought metals should also be considered. Specimens cut transversely to the longitudinal axis of a component (such as a shaft, plate, or sheet) usually give lower tensile and ductility values than those cut along the longitudinal axis. This is due to the marked directionality and the resulting anisotropy produced during rolling or forging. When sectioning tension-test specimens from the failed component, special attention should be paid to the orientation of the specimen. Some components may have quality-assurance notes on the drawings that indicate where tensile specimens are to be taken. This is especially true of critical aircraft components. Anisotropic materials have properties that vary with test specimen orientation. Tensile-strength and yield-strength specifications are usually given in the longitudinal and transverse directions. Typically, and unless otherwise specified, the tensile specimen should be taken with its major axis parallel to the direction of grain flow; however, test specimens are typically taken in two of the three directions: longitudinal and long transverse. (The short transverse direction is typically not tested since it is difficult to obtain specimens of sufficient length in that direction.) It may be necessary to lightly polish the component surface and use a macroetchant for the material being worked with to determine the grain flow. The specimens are removed from the parent component by first cutting a rough specimen shape, then final machining to the specified form. Residual stresses in the component may result in warped specimens or pinched cutting tools during rough machining. Care should be taken to document the residual stress observations, so that comparison to exemplar parts sectioned in the same manner can be made. Hopefully, final machining of the testpiece will correct the warped shape while still allowing the specimen to run parallel to the grain flow. Straightening of a warped tensile specimen during the test will result in a nonlinear indication on the initial portion of the stress-strain curve. This can be corrected to a line by using curve-fitting software or by drawing a line by eye back to the abscissa as shown in Fig. 5. There may also be nonuniformity in the specimen microstructure that affects the properties of the material, for example, a change in grain size due to cold work or changes in temper due to the HAZ of a weld. Fig. 5 Examples of stress-strain curves requiring foot correction. D, point where the extension of the straight (elastic) part diverges from the stress-strain curve. Source: Ref 1 It may be necessary to make some tests either at slightly elevated or at low temperatures to simulate service conditions. Also, it may be helpful to test specimens after they have been subjected to particular heat treatments simulating those of the failed component in service to determine how this treatment has modified mechanical properties. For example, treating a steel at a temperature in the embrittling range for about 1 h prior to impact testing will indicate any tendency to strain-age embrittlement. The determination of the ductile-to-brittle transition temperature may be useful in investigating brittle fracture of a low-carbon steel. Component Proof Testing. Critical components are sometimes proof tested after manufacture. This is especially true of critical components that will be subjected to large loads. Proof testing involves loading a component past the recommended operating limits, and possibly slightly into the yield zone. This helps to ensure there are no manufacturing or materials defects that would cause premature failure. Proof loads are usually determined on a case-by-case basis and may be expressed in several forms, such as proof load to twice the operating load or proof load to 90% of the determined yield strength. Extreme care should be taken not to damage the component in the specification of a proof load. For example, stretching a component may alter a designed preload by relieving residual compressive stresses introduced during manufacture. There are usually documents with details of specific methods, tools, and loads used in proof testing and results of tests on critical components. If performing proof testing on exemplar parts from a given manufacturer, it will lend credibility to the results if the manufacturer's test procedures are followed. If the proof tests cite standard methods, such as ASTM, ISO, or other industry-accepted procedure, acquire a copy of the method and follow it closely. If results vary from those reported by the manufacturer, it may be necessary to prove the results are valid. The file is downloaded from www.bzfxw.com
Conductivity of Aluminum Alloys. During a failure analysis investigation of a heat treated aluminum alloy, conductivity testing may be performed to evaluate proper heat treat condition, to assess areas for heat damage, or to estimate tensile rength. Conductivity is the reciprocal of electrical resistivity and is directly proportional to the mean free path of an electron in the crystal structure of the material. The mean free path is affected by the microstructure of the material, which is affected by heat treat condition Solution-annealed structures have soluble constituents that have precipitated out of solution, providing for a material with high conductivity. Aging causes fine precipitation of a second phase and thus decreases the conductivity of the material Natural aging yields the lowest conductivity and intermediate strength, while artificial aging results in intermediate conductivity and highest hardness/strength. Overaging of the alloy causes the conductivity and hardness values te approach those of the solution-annealed condition The manufacturer does not typically specify conductivity values of an aluminum alloy component. However, with proper design and material selection, the alloy and temper of the component are specified, and corresponding values of conductivity may be found in reference data such as specification SAE-MIL-H-6088. Conductivity measurements must be paired with hardness measurements to determine if the failed component meets alloy and heat treatment requirements. An aluminum alloy and its temper can be determined by measuring hardness and conductivity and verifying against reference Chemical Analysis In a failure investigation, routine analysis of the material is usually recommended. Often it is done last because an analysis usually involves destroying a certain amount of material. There are instances where the wrong material was used, under which conditions the material might be the major cause of failure. In many cases, however, the difficulties are caused by factors other than material composition In most instances, slight deviations from specified compositions are not likely to be of major importance in failure analysis. However, small deviations in aluminum content can lead to strain aging in steel, and small quantities of impurities can lead to temper embrittlement. In specific investigations, particularly where corrosion and stress corrosion are involved, chemical analysis of any deposit, scale, or corrosion product, or a substance with which the affected material has been in contact, is required to assist in establishing the primary cause of failure here analysis shows that the content of a particular element is slightly greater than that required in the specifications, it should not be inferred that such deviation is responsible for the failure. Often, it is doubtful whether such a deviation has played even a contributory part in the failure. For example, sulfur and phosphorus in structural steels are limited to 0.04% in many specifications, but rarely can a failure in service be attributed to sulfur content slightly in excess of 0.04% Within limits, the distribution of the microstructural constituents in a material is of more importance than their exact proportions. An analysis(except a spectrographic analysis restricted to a limited region of the surface)is usually made on drillings representing a considerable volume of material and therefore provides no indication of possible local deviation due to segregation and similar effects Also, certain gaseous elements, or interstitials, normally not reported in a chemical analysis, have profound effects on the mechanical properties of metals. In steel, for example, the effects of oxygen, nitrogen, and hydrogen are of major ortance.Oxygen and nitrogen may give rise to strain aging and quench aging. Hydrogen may induce brittleness icularly when absorbed during welding, cathodic cleaning, electroplating, or pickling. Hydrogen is also responsible for the characteristic halos or fisheyes on the fracture surfaces of welds in steels, in which instance the presence of hydrogen often is due to the use of damp electrodes. These halos are indications of local rupture that has taken place under the bursting microstresses induced by the molecular hydrogen, which diffuses through the metal in the atomic state and collects under pressure in pores and other discontinuities. Various effects due to gas absorption are found in other metals and alloys. For example, excessive levels of nitrogen in superalloys can lead to brittle nitride phases that cause failures of highly stressed parts Various analytical techniques can be used to determine elemental concentrations and to identify compounds in alloys bulky deposits, and samples of environmental fluids, lubricants, and suspensions. Semiquantitative emission petrography spectrophotometry, and atomic-absorption spectroscopy can be used to determine dissolved metals( as in nalysis of an alloy with wet chemical methods used where greater accuracy is needed to determine the concentration of metals. Combustion methods ordinarily are used for determining the concentration of carbon, sulfur, nitrogen, hydrogen and oxygen Wet chemical analysis methods are employed for determining the presence and concentration of anions such as Cl, NO and S. These methods are very sensitive X-ray diffraction identifies crystalline compounds either on the metal surface or as a mass of particles and can be used to analyze corrosion products and other surface deposits. Minor and trace elements capable of being dissolved can be determined by atomic-absorption spectroscopy of the solution X-ray fluorescence spectrography can be used to analyze both crystalline and amorphous solids, as well as liquids and
Conductivity of Aluminum Alloys. During a failure analysis investigation of a heat treated aluminum alloy, conductivity testing may be performed to evaluate proper heat treat condition, to assess areas for heat damage, or to estimate tensile strength. Conductivity is the reciprocal of electrical resistivity and is directly proportional to the mean free path of an electron in the crystal structure of the material. The mean free path is affected by the microstructure of the material, which is affected by heat treat condition. Solution-annealed structures have soluble constituents that have precipitated out of solution, providing for a material with high conductivity. Aging causes fine precipitation of a second phase and thus decreases the conductivity of the material. Natural aging yields the lowest conductivity and intermediate strength, while artificial aging results in intermediate conductivity and highest hardness/strength. Overaging of the alloy causes the conductivity and hardness values to approach those of the solution-annealed condition. The manufacturer does not typically specify conductivity values of an aluminum alloy component. However, with proper design and material selection, the alloy and temper of the component are specified, and corresponding values of conductivity may be found in reference data such as specification SAE-MIL-H-6088. Conductivity measurements must be paired with hardness measurements to determine if the failed component meets alloy and heat treatment requirements. An aluminum alloy and its temper can be determined by measuring hardness and conductivity and verifying against reference data. Chemical Analysis In a failure investigation, routine analysis of the material is usually recommended. Often it is done last because an analysis usually involves destroying a certain amount of material. There are instances where the wrong material was used, under which conditions the material might be the major cause of failure. In many cases, however, the difficulties are caused by factors other than material composition. In most instances, slight deviations from specified compositions are not likely to be of major importance in failure analysis. However, small deviations in aluminum content can lead to strain aging in steel, and small quantities of impurities can lead to temper embrittlement. In specific investigations, particularly where corrosion and stress corrosion are involved, chemical analysis of any deposit, scale, or corrosion product, or a substance with which the affected material has been in contact, is required to assist in establishing the primary cause of failure. Where analysis shows that the content of a particular element is slightly greater than that required in the specifications, it should not be inferred that such deviation is responsible for the failure. Often, it is doubtful whether such a deviation has played even a contributory part in the failure. For example, sulfur and phosphorus in structural steels are limited to 0.04% in many specifications, but rarely can a failure in service be attributed to sulfur content slightly in excess of 0.04%. Within limits, the distribution of the microstructural constituents in a material is of more importance than their exact proportions. An analysis (except a spectrographic analysis restricted to a limited region of the surface) is usually made on drillings representing a considerable volume of material and therefore provides no indication of possible local deviation due to segregation and similar effects. Also, certain gaseous elements, or interstitials, normally not reported in a chemical analysis, have profound effects on the mechanical properties of metals. In steel, for example, the effects of oxygen, nitrogen, and hydrogen are of major importance. Oxygen and nitrogen may give rise to strain aging and quench aging. Hydrogen may induce brittleness, particularly when absorbed during welding, cathodic cleaning, electroplating, or pickling. Hydrogen is also responsible for the characteristic halos or fisheyes on the fracture surfaces of welds in steels, in which instance the presence of hydrogen often is due to the use of damp electrodes. These halos are indications of local rupture that has taken place under the bursting microstresses induced by the molecular hydrogen, which diffuses through the metal in the atomic state and collects under pressure in pores and other discontinuities. Various effects due to gas absorption are found in other metals and alloys. For example, excessive levels of nitrogen in superalloys can lead to brittle nitride phases that cause failures of highly stressed parts. Various analytical techniques can be used to determine elemental concentrations and to identify compounds in alloys, bulky deposits, and samples of environmental fluids, lubricants, and suspensions. Semiquantitative emission spectrography, spectrophotometry, and atomic-absorption spectroscopy can be used to determine dissolved metals (as in analysis of an alloy) with wet chemical methods used where greater accuracy is needed to determine the concentration of metals. Combustion methods ordinarily are used for determining the concentration of carbon, sulfur, nitrogen, hydrogen, and oxygen. Wet chemical analysis methods are employed for determining the presence and concentration of anions such as Cl- , NO3 - , and S- . These methods are very sensitive. X-ray diffraction identifies crystalline compounds either on the metal surface or as a mass of particles and can be used to analyze corrosion products and other surface deposits. Minor and trace elements capable of being dissolved can be determined by atomic-absorption spectroscopy of the solution. X-ray fluorescence spectrography can be used to analyze both crystalline and amorphous solids, as well as liquids and gases
Infrared and ultraviolet spectroscopy are used in analyzing organic materials. When the organic materials are present in a complex mixture(such as, for example, solvents, oils, greases, rubber, and plastics), the mixture is first separated into its components by gas chromatography Analysis of Surfaces and Deposits. Wavelength-dispersive x-ray spectrometers(WDSs) and edSs are frequently used for providing information regarding the chemical composition of surface constituents. They are employed as accessories for SEMs and permit simultaneous viewing and chemical analysis of a surface. The Auger electron spectrometer is useful for detecting the elements in extremely thin surface layers. The Auger electron spectrometer can provide semiquantitative determinations of elements with atomic numbers down to 3 (lithium). The size of the area examined varies greatly with the test conditions; it may be from I to 50 um in diameter For chemical analysis of surface areas as small as I um in diameter, the electron-microprobe analyzer is widely used. This instrument can determine the concentration of all but the low atomic number elements with a limit of detection below 0. 1%. The area examined with the ion-microprobe analyzer is a few microns in diameter larger than that examined with the electron-microprobe analyzer. The ion-microprobe analyzer has the advantage of being able to detect nearly all elements(including those of low atomic weights) in concentrations as low as 100 ppm. It is sometimes used to volatilize materials, which are then passed through a mass spectrometer Electron microprobes and other modern analytical instruments are described in greater detail in Materials Characterization. Volume 10 of the ASM Handbook. The instruments discussed previously are used for direct analysis of surfaces other techniques can be used for analyzing material that has been removed from the surface. For example, if material is removed in a replica(perhaps chemically extracted), it can be analyzed structurally by x-ray diffraction or electron diffraction. Also, depending on the quantity of material extracted, many of the routine chemical analysis techniques may be applicable Spot testing uses chemical tests to identify the metal, the alloying elements present, deposits, corrosion products, and soil Spot tests can be performed both in the laboratory and in the field; they do not require extensive training in analytical chemistry. The only requirement is that the substance be dissolvable, hydrochloric acid or even aqua regia may be used to dissolve the material Spot tests for metallic elements such as chromium, nickel, cobalt, iron, and molybdenum are usually done by dissolving a small amount of the alloy in acid and mixing a drop of the resulting solution with a drop of a specific reagent on absorbent paper or a porcelain plate. Spot colorings produced in this way indicate the presence or absence of the metallic radical under test. Samples may be removed from gross surfaces by spotting the specimen with a suitable acid, allowing time for solution and collecting the acid spot with an eyedropper Simulated-Service Testing During the concluding stages of an investigation, it may be necessary to conduct tests that simulate the conditions under which failure is believed to have occurred. Often, simulated-service testing is not practical because elaborate equipment is required, and even where practical it is possible that not all of the service conditions are fully known or understood Corrosion failures, for example, are difficult to reproduce in a laboratory, and some attempts to reproduce them have iven misleading results. Serious errors can arise when attempts are made to reduce the time required for a test by artificially increasing the severity of one of the factorssuch as the corrosive medium or the operating temperature Similar problems are encountered in wear testing On the other hand, when its limitations are clearly understood, the simulated testing and statistical experimental design analysis of the effects of certain selected variables encountered in service may be helpful in planning corrective action or at least, may extend service life. The evaluation of the efficacy of special additives to lubricants is an example of the uccessful application of simulated-service testing. The aircraft industry has made successful use of devices such as the wind tunnel to simulate some of the conditions encountered in flight, and naval architects have employed tank tests to evaluate hull modifications, power requirements, steerage, and other variables that might forestall component failure or promote safety at sea Taken singly, most of the metallurgical phenomena involved in failures can be satisfactorily reproduced on a laboratory scale, and the information derived from such experiments can be helpful to the investigator, provided the limitations of the tests are fully recognized. However, many company managers prefer to conduct trials to verify improvements before major conversions are approved. This is a more conservative approach, but it only takes one improper recommendation that results in an adverse result to justify trials of major changes Reference cited in this section P.M. Mumford, Test Methodology and Data Analysis, Tensile Testing, P. Han, Ed, ASM International, 1992, Thefileisdownloadedfromwww.bzfxw.com
Infrared and ultraviolet spectroscopy are used in analyzing organic materials. When the organic materials are present in a complex mixture (such as, for example, solvents, oils, greases, rubber, and plastics), the mixture is first separated into its components by gas chromatography. Analysis of Surfaces and Deposits. Wavelength-dispersive x-ray spectrometers (WDSs) and EDSs are frequently used for providing information regarding the chemical composition of surface constituents. They are employed as accessories for SEMs and permit simultaneous viewing and chemical analysis of a surface. The Auger electron spectrometer is useful for detecting the elements in extremely thin surface layers. The Auger electron spectrometer can provide semiquantitative determinations of elements with atomic numbers down to 3 (lithium). The size of the area examined varies greatly with the test conditions; it may be from 1 to 50 μm in diameter. For chemical analysis of surface areas as small as 1 μm in diameter, the electron-microprobe analyzer is widely used. This instrument can determine the concentration of all but the low atomic number elements, with a limit of detection below 0.1%. The area examined with the ion-microprobe analyzer is a few microns in diameter larger than that examined with the electron-microprobe analyzer. The ion-microprobe analyzer has the advantage of being able to detect nearly all elements (including those of low atomic weights) in concentrations as low as 100 ppm. It is sometimes used to volatilize materials, which are then passed through a mass spectrometer. Electron microprobes and other modern analytical instruments are described in greater detail in Materials Characterization, Volume 10 of the ASM Handbook. The instruments discussed previously are used for direct analysis of surfaces; other techniques can be used for analyzing material that has been removed from the surface. For example, if material is removed in a replica (perhaps chemically extracted), it can be analyzed structurally by x-ray diffraction or electron diffraction. Also, depending on the quantity of material extracted, many of the routine chemical analysis techniques may be applicable. Spot testing uses chemical tests to identify the metal, the alloying elements present, deposits, corrosion products, and soil. Spot tests can be performed both in the laboratory and in the field; they do not require extensive training in analytical chemistry. The only requirement is that the substance be dissolvable; hydrochloric acid or even aqua regia may be used to dissolve the material. Spot tests for metallic elements such as chromium, nickel, cobalt, iron, and molybdenum are usually done by dissolving a small amount of the alloy in acid and mixing a drop of the resulting solution with a drop of a specific reagent on absorbent paper or a porcelain plate. Spot colorings produced in this way indicate the presence or absence of the metallic radical under test. Samples may be removed from gross surfaces by spotting the specimen with a suitable acid, allowing time for solution and collecting the acid spot with an eyedropper. Simulated-Service Testing During the concluding stages of an investigation, it may be necessary to conduct tests that simulate the conditions under which failure is believed to have occurred. Often, simulated-service testing is not practical because elaborate equipment is required, and even where practical it is possible that not all of the service conditions are fully known or understood. Corrosion failures, for example, are difficult to reproduce in a laboratory, and some attempts to reproduce them have given misleading results. Serious errors can arise when attempts are made to reduce the time required for a test by artificially increasing the severity of one of the factors—such as the corrosive medium or the operating temperature. Similar problems are encountered in wear testing. On the other hand, when its limitations are clearly understood, the simulated testing and statistical experimental design analysis of the effects of certain selected variables encountered in service may be helpful in planning corrective action or, at least, may extend service life. The evaluation of the efficacy of special additives to lubricants is an example of the successful application of simulated-service testing. The aircraft industry has made successful use of devices such as the wind tunnel to simulate some of the conditions encountered in flight, and naval architects have employed tank tests to evaluate hull modifications, power requirements, steerage, and other variables that might forestall component failure or promote safety at sea. Taken singly, most of the metallurgical phenomena involved in failures can be satisfactorily reproduced on a laboratory scale, and the information derived from such experiments can be helpful to the investigator, provided the limitations of the tests are fully recognized. However, many company managers prefer to conduct trials to verify improvements before major conversions are approved. This is a more conservative approach, but it only takes one improper recommendation that results in an adverse result to justify trials of major changes. Reference cited in this section 1. P.M. Mumford, Test Methodology and Data Analysis, Tensile Testing, P. Han, Ed., ASM International, 1992, p 55 The file is downloaded from www.bzfxw.com