METHOD AND APPARATUS FOR DETERMINING MATERIAL FATIGUE BY MEASURING DAMPING FACTORS
Background and Summary of the Invention
There are many applications wherein metallic materials are used for their strength and endurance and are therefore subjected to loads, stresses, strains, and other forces which, over time, may have a tendency to fatigue the material and create a risk of catastrophic failure. It would be highly desirable to be able to test these discrete metallic parts in situ for their state of fatigue such that they might be replaced or renewed prior to any such catastrophic failure. In still other in¬ stances, and especially for critical applications involv¬ ing health and safety, standards have been established for the routine testing of certain metallic parts prior to their being placed in service to ensure against any such failure of the part. In those applications, tech¬ niques have been developed and are available in the prior art to achieve such testing. These include such things as x-ray, destructive testing of selected parts from a lot, etc., all as well known in the art. However, these
techniques are all subject to certain drawbacks such as expense, inconvenience, and in some cases failure to entirely eliminate the possibility of premature failure of the part. Still another situation in which these kinds of tests for fatigue are conducted involve many instances where materials or parts are welded and the integrity ovfthe weld must be verified prior to the equipment being placed in service. One particular appli¬ cation, from amongst many, involves the federal safety standards which govern the construction of nuclear power plants. Certain welds in certain critical equipment contained within the plant are subjected to x-ray and other kinds of testing in order to verify their integrity prior to the plant being placed in service. A nuclear power plant presents perhaps an extreme example of the potential harm which might befall not only the people involved but the public at large should a critical piece of equipment suffer a premature failure. There are a myriad of other applications perhaps considered not as critical but which also are important to the health and safety of many people, including the public at large.
Despite the fact that testing for fatigue has been utilized for some time, and the relationship of damping to fatigue has been well known for some time, the inven- tor is not aware of any other efforts in the prior art to utilize the relationship of damping to fatigue in the arena of fatigue testing. For example, in a paper pre¬ sented at a colloquium on structural damping at the ASME Annual Meeting in Atlantic City, New Jersey in December of 1959, the phenomenon of "plastic strain" was analyzed. In particular, damping was used as a parameter for deter¬ mining the interrelationship between stress history and stress amplitude as mechanisms for affecting plastic strain in a material. As concluded in the paper, at low stresses and intermediate stresses, within 1-50% of a fatigue limit, damping was not seen to be affected by the
stress history of the material. On the other hand, at high stresses, typically above 50% fatigue limit, where large plastic strain damping may be observed, stress history played a part in affecting plastic strain, as measured by the damping factor. Stated differently, data were presented indicating that at low and intermediate stress, the damping factor does not change with the num¬ ber of fatigue cycles. However, above a critical stress, damping increases with the number of fatigue cycles thereby indicating that stress history plays a part in plastic strain under these conditions. Although this article treated the interrelationship between stress history and stress amplitude, and their effect on damping (plastic strain), there was no disclosure or suggestion of utilizing a measured damping factor as an indicator of the state of fatigue of a material. As stated therein, the article focused on how stress history and amplitude might produce a particular damping factor but not how a measured damping factor could be used as a predictor of relative fatigue in a part. See Structural Damping edit¬ ed by Jerome E. Ruzicka, ASME Proceedings, 1959,
In order to solve these and other problems in the prior art, and as a departure from the teachings in the prior art, the inventor herein has succeeded in develop- ing the technique of measuring the damping factor of a discrete piece of metallic material, such as a part in an assembly or the like, and using that damping factor for determining the fatigue integrity of that part either by comparing it with a standardized damping factor or with previously measured damping factors for the same part.
The part might be a single piece of material, or it might be a welded or otherwise joined piece of material and the test may be one for integrity, i.e. cracking, voids, or the like, as might be required for a new part, or the test might be conducted for determining the fatigue in the part after having been installed and used over time.
For new part testing, it is anticipated that standardized damping factors may be determined and available for com¬ parison with the measured damping factor for the new part. Alternately, the damping factor of a series of identical new parts might be measured and used to cull out those new parts which evidence signs of early fatigue and failure, or cracks, voids, or other defects in manu¬ facture. After a part has been installed and used over a period of time, a damping factor measurement may be made periodically to determine the part's increasing fatigue. This technique may be used to identify parts which are need of replacement prior to any chance of catastrophic failure. There are other applications and situations in which the damping factor measurement of a discrete piece of metallic material might be used to good advantage. These particular examples are being given as exemplary.
In making the damping factor measurement, the inventor herein has also succeeded in developing a simple but effective and accurate technique for measuring the damping factor in either of two methods. Utilizing a first method, an impulse of energy may be applied to the part, such as by striking it with a blunt object or the like, and the induced vibration in the part measured by a transducer which converts the vibration into an electri- cal signal for input to a computer. A computer may then easily make the appropriate calculation from the induced vibration to determine the damping factor. Generally, as is known in the art, the damping factor of a part vibrat¬ ing at its natural frequency may be determined by compar- ing peak amplitudes of successive cycles of the vibra¬ tion. In an alternative method, a continuous input of energy may be provided to the part instead of an impulse of energy. In a preferred embodiment, a frequency gener¬ ator may be coupled to a transducer, such as a speaker, shaker, or other such device, and the frequency generator tuned or adjusted so as to sweep through the range of the
lowest natural frequencies of the part. As the input of energy remains constant, the part would continue to vi¬ brate at its natural frequency such that the damping factor may be readily calculated by measuring the half- power bandwidth of a cycle and dividing it by the center frequency, as is well known in the art. Using either of these methods, a vibration is induced in the part and the response thereto is measured from which the damping fac¬ tor is determined. One of the advantages of using the inventor's method of inducing a vibration in the part is that it is believed that the part need not be isolated and may be tested in place. This eliminates disassembly of the part from any larger assemblage which dramatically reduces any costs involved in using the present method in determining the damping factor. This provides great advantages over other prior art methods which require disassembly and isolation of the part to be tested, such as in the x-ray method. Furthermore, the device used to implement the method disclosed herein may be relatively compact, readi¬ ly portable, and sufficiently small such that the testing of many differently sized parts which might be otherwise relatively difficult to access may be readily tested.
While the principal advantages and features of the present invention have been described above, a more com¬ plete and thorough understanding of the invention may be attained by referring to the drawings and description of the preferred embodiment which follow. Brief Description of the Drawings Figure 1 is a diagrammatic view of the inventor's first technique for measuring material integrity utiliz¬ ing an impulse of energy input to induce a vibration into the material;
Figure 2 is a graph of the harmonic response of vibrations induced in the material through the technique shown in Figure 1;
Figure 3 is a diagrammatic view of the inventor's second technique for measuring material integrity through the coupling of a continuous energy source to the materi¬ al; and Figure 4 is a graph of the vibrational response induced in the material using the technique of Figure 3. Detailed Description of the Preferred Embodiment
As shown in Figures 1 and 2, the inventor's first technique for measuring relative fatigue in a part in- eludes the step of inducing a vibration in the part de¬ sired to be measured, such as by striking the part 20 with a blunt instrument such as a rod 22 to thereby in¬ duce vibrations in the part 20. For convenience, the opposite ends of the part 20 may be supported by a pair of supports 24, 26, although this is not believed to be necessary. A transducer 28 measures the induced vibra¬ tion and produces an electrical output which is amplified by an amplifier 30 and then input to a computer 32 for calculation of the damping factor. As shown in Figure 2, the vibration induced by the input of an impulse of ener¬ gy into the part 20 may have a varying amplitude or force level. The damping factor, as is well known, may be readily calculated by comparing the amplitudes of succes¬ sive cycles of vibration induced by any one of these force levels. As shown in Figure 2, the intensity of the blow to the part does not affect the measurement of the damping factor as the damping factor is determined by comparing two successive peak amplitudes, regardless of the size thereof. Whether the initial amplitude has an intensity of a, b, or c, there is no variation in the measured damping factor. Instead, the damping factor is determined solely by the characteristics of the part 20.
As shown in Figure 1, the part 20 may in actual fact be comprised of a pair of elements 34, 36 which are joined by a weld 38 or the like. If that is the case, then the integrity of the weld 38 may be readily deter-
mined by the measurement of the damping factor. Similar¬ ly, the joint, shown in Figure 1 as weld 38, may be any other joint or connection and its integrity similarly measured through the methodology disclosed herein. As shown in Figures 3 and 4, an alternate tech¬ nique for measuring the damping factor and, hence, mate¬ rial fatigue, may be used. As before, the vibration in the part 20 is picked up by a transducer 28 for convert¬ ing the sensed vibrational signals to an electrical sig- nal which is then amplified by an amplifier 30 for input to a computer 32. However, the initial energy input to the part 20 is achieved by way of a frequency generator 40 which produces an electrical output at a particular frequency which is then amplified by a power amplifier 42 and fed to a second transducer 44, which may be a speaker or shaker or other such device, which is coupled to the part 20. The frequency generator 40 is then tuned to frequencies sweeping through the range of the lowest natural frequencies of the part 20 to thereby produce a continuous vibrational response therein as shown in Fig¬ ure 4. A peak amplitude F c of one of the several natural harmonics induced in the part 20 is chosen for measure¬ ment of the damping factor. As is well known in the art, the damping factor is equal to the half power bandwidth ΔF, or F2 - Fl, divided by the center frequency F c . Fl and F2, the half power frequencies, are those frequencies at which the amplitude is .707 times the maximum ampli¬ tude. As shown in Figure 3, the existence of a crack 46 would affect the vibrational response of the part 20 and, hence, the damping factor measured with the inventor's technique, thereby becoming detected for suitable correc¬ tion thereto.
Either or both of these techniques may be conve¬ niently used to determine the damping factor of a partic- ular part. The damping factor may be periodically mea-
sured for a particular part to develop a history thereof and thereby be used to monitor the developing fatigue in the part as an aid to deciding when it should be replaced or repaired. Alternately, the damping factor measured by the techniques disclosed and claimed herein could be used by comparing them with standardized damping factors for similar kinds of metals and parts.
In the preferred embodiment, and as shown in the drawings, stand alone personal computers are depicted. However, as is well known to those of ordinary skill in the art, portable PC's are well known and readily avail¬ able commercially such that a suitable device for porta¬ ble use and application could readily be achieved. Fur¬ thermore, although the inventor has not developed any such device, a custom made "damping factor detector" could readily be designed utilizing only those computer elements required including, e.g., a computer chip, a custom readout, and keyboard or other data entry means. Additionally, hard copy readout could readily be provid¬ ed. It is intended that all of these alternatives be included within the scope of the present invention.
There are various changes and modifications which may be made to the invention as would be apparent to those skilled in the art. However, these changes or modifications are included in the teaching of the disclo¬ sure, and it is intended that the invention be limited only by the scope of the claims appended hereto.
