Search

Abstract Comparative tests related to locating acoustic emission (AE) signals due to shock impacts on a T700 carbon fiber sample were carried out. Piezoelectric acoustic emission transducers (AETs) and fiber-optic sensors (FOSs) were installed on the sample, forming rectangular location antennas measuring $$360 \times 280$$ mm. Strikes were delivered with balls weighing 10 and 18.5 g. Antennas consisting of four AET sensors and four FOS sensors and an antenna consisting of two AET sensors and two FOS sensors were organized. When using the antenna containing four FOS sensors, the impact on the sample was produced by a load weighing 530 g dropped from a height of 400 mm. AE signals were recorded by the SCAD-16.10 system with “floating” selection thresholds when the ball was dropped and during its repeated bounces. Then AE signal clusters were formed and recorded during the impact of loads. The arrival times of AE signals to the antenna sensors were calculated using the threshold method, the root mean square (RMS) deviation method and the two-interval method. It is shown that the maximum error in locating AE signals is observed when a steel ball with a diameter of 16 mm is dropped from a height of 300 mm and the minimum error is when using an electronic simulator.

The results of experimental studies of a nondestructive testing method for assessing the loading of bearing rings on an axle journal using the acoustoelasticity effect based on measuring the characteristics of elastic surface waves are presented. Reference blocks for loading with a difference in pilot diameters in the range from 38 to 119 $$\mu {\text{m}}$$ with an error of no more than 7 µm have been developed and manufactured. A hydraulic testing bench for mechanical loading of bearing rings simulating their press fitting on an axle with loading of up to 46 microns was certified. The acoustoelastic coefficients $$\alpha = (5.48 \pm 0.14)$$ $${\text{TP}}{{{\text{a}}}^{{-1}}}$$ were experimentally determined for surface waves in steel ShKh15 based on the delay in the time of recording pulses that repeatedly (2–5 times) circled the ring. The correlation method is optimized for determining the delay time of a wave in a loaded ring relative to an unloaded one according to the condition of minimizing the uncertainty associated with a change in the pulse shape. The main reasons for errors and uncertainties in measuring the absolute time of propagation of a surface wave are established: the effect of waves reflected from the boundaries, surface curvature, and attenuation. The bearing rings loading was measured on loading reference blocks using the acoustoelasticity effect, and the estimation of the stress measurement uncertainty was made.

Static tests of the samples of Torayca T700, T800 carbon fiber reinforced plastics consisting of nine monolayers with $$\left[ { \pm 45{\text{/}}90{\text{/}}{{0}_{3}}{\text{/}}90{\text{/}} \pm 45} \right]$$ stacking and geometric dimensions $$600 \times 100 \times 0.9$$ mm and $$500 \times 100 \times 1.4$$ mm were carried out. In the process of static and thermal loading, defects were monitored by the acoustic emission (AE) method. Two samples were statically loaded and simultaneously exposed to temperature $$T = 100$$ °C. Six samples were statically loaded at temperature $$T = 20$$ °C. The sources of AE signals corresponding to sample material destruction were localized in real time. For these signals, thin polished sections were prepared from location zones in the sample, and the fractography of the CFRP material was carried out. The analysis of the recorded AE information has been carried out. The main informative parameters of the AE signals localized in the working area of the samples are considered. Using cluster analysis by digital waveform, three types of clusters are identified for the AE signals. For the first type of clusters, the frequency of AE signals exceeded 175 kHz, and a relatively high level of the MARSE energy parameter was noted. In clusters of the second type, AE signals recorded under static loading and sample heating had lower MARSE values and a median frequency not exceeding 170 kHz. When testing samples with no heating, the median frequency did not exceed 140 kHz. The third type of clusters was formed by AE signals with the frequency not exceeding 400 kHz. Such clusters were typical only for samples tested at temperature $$T = 20$$ °C.

Static tests of a two-spar T800 CFRP wing box have been carried out. The object was tested when loaded using the methods of acoustic emission (AE) and strain gaging. Strain in the material was determined in real time in the areas where strain gages were bonded (the panels of wing box, the walls of side members, and the shelves of stringers). The elements of the wing box have been established in which nonlinear strain change is observed and residual deformations are recorded after unloading. The AE method has been used to localize the sources of signals. The coordinates of the sources corresponded to the location of the third rib. Clustering recorded AE signals based on their digitized form has made it possible to group the signals and assign them to sources that correspond to structure failure. It is shown that in the process of loading the wing box, an increase in the structural coefficient of AE signals corresponding to CFRP delamination in the location zone is observed.

A procedure is proposed for rejecting girth welds based on an analysis of the main parameters of acoustic emission (AE) signals recorded during multiple-pass welding. The accuracy of locating AE signals recorded from defects and their filtering from noise and interference were ensured by the use of a modified tabular method and rejection grid. Normalized coefficients were calculated from the energy and amplitude of localized AE signals and used for in-process assessment of weld quality. Threshold values were determined for the normalized coefficients above which the weld is rejected. Defects in girth welds were identified using this procedure, and the possibility of determining defect hazardousness in real time was demonstrated.

The results of testing samples made of T700 carbon fiber reinforced plastic with a stress concentrator in the form of a hole 14 mm in diameter are considered. Some of the samples were heated to a temperature of +100°C, followed by exposure to a static tensile load, with the remaining samples statically loaded at a temperature of +20°C. In the process of loading, information was recorded using the method of acoustic emission. At first, the samples were loaded statically up to 50 kN, which was 50% of the average breaking load. Then the load was increased in increments ΔP = 10 kN until sample failure. It was observed that the simultaneous exposure of samples to static and heat loads decreased the load capacity of the samples.

Results of the strength tests of an aircraft wing spar made of Toray T1000 carbon fiber-reinforced plastic are presented. Construction failure zones have been inspected by the methods of acoustic emission (AE) and tensometry. Loading-induced changes in the stress-strain state (SSS) of wing-spar material have been analyzed. The coordinates of AE signal sources that correspond to the separation of the upper spar cap from the spar web have been determined. Algorithms for AE-data processing using wavelet transforms are considered. Connections have been established between changes in stresses emerging in CFRP material, AE signal structure, and the occurrence and propagation of spar fracture.

The acoustic-emission method for inspecting flaws that form during laser bonding with the use of a CO2 continuous-wave laser with a power of up to 8 kW and characteristics that can be preadjusted prior to welding is considered. The structure of acoustic-emission signals registered in a specimen’s flawed domain has been analyzed using the wavelet transform. It has been shown that the structural coefficients of the acoustic-emission signals localized in specimen’s flawed sections become identical over time.

Relationship between the main informative parameters of acoustic-emission (AE) signals recorded during static loading of a CFRP sample and the process of material fracture has been investigated. Clusterization of AE signals by digitized shape has made it possible to locate a zone they originate from. Material was then cut out from this zone to prepare thin sections that were subjected to fractography. Fractography revealed fiber fracture, brittle cracking of the matrix, its exfoliation from the carbon fibers, etc.

Ni(Mg)Al-layered double hydroxides (LDH) were synthesized by coprecipitation. Structural properties of LDH and mixed oxides Ni(Mg)AlOx, obtained by LDH calcination, were studied by XRD and SEM. Textural properties of Ni(Mg)AlOx were studied by the nitrogen adsorption-desorption method. It was found that the formation of the single-phase LDH was occurred for any nickel content in the samples. The mixed oxides have a high specific surface area, the value of which decreases with increasing nickel content in the samples. Ni(Mg)AlOx are characterized by high activity in furfural hydrogenation. Selectivity of the formation of different reaction products depends on the nickel content in the catalysts. The full furfural conversion with the furfuryl alcohol selectivity no more than 65% was observed for the catalyst with a low nickel content. At the same time an increase in selectivity for furfuryl alcohol up to 84% occurs for the NiAl-catalyst with a furfural conversion not exceeding 43%.

Titanium and duralumin inserts were introduced into a weld to imitate welding flaws in St3 steel samples. Analysis of the accuracy of locating the introduced flaws with different clustering methods (by digitized shape, main informative parameters, the rise rate of leading-edge envelope) has shown that these methods involve considerable time expenditures. The developed dynamic clustering has been combined with the clustering by main informative parameters and allowed in situ location of flaws during welding, while the weld has not been completed yet.

A modified tabular method for locating acoustic-emission signals, which is intended for determining the coordinates of flaws when inspecting a multiple-run welding process with a lower error rate, and the result of using this method for determining the coordinates of flaws when welding a circular contour using a piezoelectric antenna, which consists of four acoustic-emission transducers, are considered. It is shown experimentally that the application of the modified tabular method allowed an increase in the accuracy of determining the coordinates of introduced flaws when using the acoustic-emission testing method.

The results of static strength tests of samples made from T700 carbon plastic with various thicknesses and packings of monolayers are given. The coordinates of the destruction range were determined and the loading values were chosen at which the destruction of samples is initiated. During the loading of samples, their state was monitored using the AE method and strain measurement.

Four passes were performed in the welding process of a 700-mm diameter pipe and 1800 × 1800-mm size steel sheet with a hole for the pipe. Welding defects were simulated by embedding artificial defects at the root of the welded seam (pores, cracks, slag, faulty fusions, and titanium inserts) and violating the welding procedure. The quality of the welded connection during the welding process was evaluated by the acoustic-emission method. After the completion of the welding the specimen underwent ultrasonic and radiographic inspection.

The results of static tests of specimens that were made from carbon fiber reinforced plastic with a honeycomb filler according to acoustic emission (AE) and strain-gauge measurement are given. When using the two-lag method, the stable locations of AE signals are obtained, which indicates the development of fatigue cracks. The dependences between the main informative parameters of AE signals and features of the destruction of a material are analyzed. According to strain-gauge measurement data, the values of the failure stress are determined.

During quality control of welding, rejection levels were determined using the comparison of the distribution of the parameters of acoustic-emission (AE) signals with introduced defects represented by titanium and tungsten inserts. The MARSE energy parameter, the mean number of oscillation along the leading edge of an AE signal, and the two-interval coefficient were used as the principal informative parameters that are most sensitive to welding defects. The Student-t and Kolmogorov-Smirnov criteria with the calculation of rejection levels and their localization allow the location of defects in the course of multiple-pass welding.

The main parameters of acoustic emission (AE) signals, viz., the energy, the total number, and the activity, were studied under a static and a subsequent cyclic loads during strength testing of cast parts (solebars and bolsters) of a freight car truck. It was shown that loading with a small number of cycles made it possible to obtain additional data, which extended the information on the state of the parts under study.

In this article, the results of endurance testing of aircraft using the acoustic-emission method (AE) are given. Measurement information was recorded by certified distributed and combined systems. A total of 62 loading blocks were performed; each included nonflight, take-off, and flight modes. Arrival of recorded AE signals was combined with the onset of the program cycle. Examples of the localization of AE signals were given from the zones under study with the development of fatigue fractures along the left and right sides of the aircraft. The results of comparative testing on an aircraft panel with a size of 1500 × 1800 mm were analyzed using distributed and combined AE systems.

The results are presented of the investigation of fatigue crack growth in 200 × 100 × 10 mm specimens made from the upper chord of rejected solebars with boxlike cross sections. The specimens were placed in an MTS-50 electrohydraulic machine and alternately loaded with cyclic and static loads from the instant of fatigue-crack nucleation until specimen damage occurred. The data on nucleation and the development of fatigue cracks in the specimens were recorded using the method of acoustic-emission (AE) strain measurements. A piezo antenna consisting of four acoustic emission transducers (AET) was mounted on each specimen. Four wire strain sensors were glued near the saw cut that served as a stress concentrator. After specimen damage, chemical analysis of the material was performed and nonmetallic inclusions and the degree of impurity were evaluated. Micro- and macrofractographic analyses, which allowed determination of the character of damage, were performed.

The problems of the localization of sources of acoustic-emission (AE) signals for objects with small geometric dimensions are considered. The dependence of the localization error for a flaw inside a piezoelectric antenna on the error in determining the difference in the times of signal arrival (DTA) at acoustic transducers was obtained. A method for determining the DTA using wavelet scalograms of AE signals was studied. Experimental studies showed the advantages of this method over the method for determining the rms deviation of AE-signal amplitudes within a “time window.”

The results of acoustic-emission (AE) testing of flaws in steel specimens during welding and subsequent loading until destruction in MTS-50 and MTS-250 electrohydraulic machines are presented. Flaws were simulated via faulty fusions or the introduction of titanium and duralumin inserts. During loading, some specimens were broken along the weld seam and the other specimens were broken in the near-seam zone. A comparative analysis of the distributions of the informative parameters of AE signals, which are recorded during different types of specimen fracture, is presented.

A two-stage method for the clustering of acoustic-emission (AE) signals using their parameters and digitized shapes was considered. AE signals were initially clustered according to a set of their informative parameters; the primary clusters that were obtained using these parameters were then subjected to clustering using their digitized AE signal shape. The use of two-stage clustering for the processing of AE signals that were recorded at the moment of welding allowed the calculation time to be reduced without loss of localization accuracy.

The changes in the design of the 18–578 solebar with a boxlike cross section required the development of a new scheme of arrangement of acoustic-emission (AE) transducers and improvement of both technique and software. All sources that were detected during AE testing were divided according to their degree of hazard into passive and critically active ones. Location grids that specified the rejection levels were imposed on the tested zones under study. In the refined technique, linear location was used, which made it possible to preserve the reliability of the investigation results and reduce the labor intensiveness of the AE testing.

A method for detecting hazardous sources of acoustic-emission (AE) signals by the level of their energy activity using clustering by digitized signal shape was considered. The influence of the distance between a source and an AE transducer on the energy of recorded AE signals was analyzed. The energy distribution median of signals in a cluster was proposed for evaluating the hazard of an AE signal source. The application of the developed method was tested based on the example of the analysis of AE inspection data on the welding of steel specimens with incomplete fusion and a crack at the root of a weld seam. This testing confirms the results of the fractographic inspection of a defective weld seam.

The experimental dependences of changes in the output signals of standard and high-temperature acoustic-emission transducers (AETs) during heating and cooling are considered. It is shown that high-temperature AETs have a hysteresis that must be taken into account in high-temperature tests of various materials. The errors in localizing acoustic-emission signals from a simulator during heating and cooling of high-temperature and standard AETs are analyzed using the two-interval method.

The regularities of the propagation of elastic vibrations in rails under buffing loading were studied for developing the technique for testing wheel pairs of freight cars in motion. Using strain gauges glued on both sides of a rail web and a fast Dinamika-1 strain-gauging system, vertical and horizontal deformations were measured. The parameters of elastic vibrations of a rail have been experimentally determined. The presence of two rail vibration modes (symmetric and antisymmetric) and the dependence of their amplitude on the misalignment of the buffing loading point have been established.

The technique of clustering by a set of parameters of acoustic emission (AE) signals, which allows reducing the time of processing of the recorded data, is considered. A comparative analysis of the stability of three clustering techniques (by shape, leading edge rise rate, and a set of AE signal parameters) with respect to the effect of random noise distributed according to a normal law is performed. A possibility of using the considered clustering techniques for distinguishing different stages of duralumin specimen fractures and during welding of steel specimens is experimentally demonstrated.

The techniques for acoustic-emission (AE) testing of cast parts (side frames and bogie bolsters) of a freight-car truck, which improve the reliability of evaluating their engineering state at different stages of the working cycle and decrease their repeated rejection, are considered. It is proposed to determine the degree of damage and the service life of cast parts on the basis of the analysis of AE-signal sources, which have the highest integral characteristics.

The results of testing the bearing rings (BRs) of a locomotive axle box by the acoustic-emission (AE) method are presented. The small size of the localization zone of BRs leads to a significant spread of the coordinates of flaws even at minimal errors of the time of arrival of AE signals at piezoelectric-antenna probes. To refine the testing technique, the maximum testing load necessary for initializing AE signals is determined. Deformations in BRs are measured with a strain-gauging system related to the AE system. During recording of AE signals, the mechanical stresses are measured, thus allowing rejection of parasitic signals and an increase in the localization accuracy.

A method of localizing acoustic-emission signals from flaws of a welded joint during its cooling is considered. Aluminum and titanium inserts were used to simulate flaws of the welded joint. The developed technique of clustering made it possible to exclude noise accompanying the processes of welding and cooling the welded joint. The welded specimens were tested on an MTS-50 loading machine. At the finishing stage, specimens with double-sided welded joints containing titanium inserts were subjected to metallographic analysis.

A technique for determining the times of arrival of acoustic-emission (AE) signals at the probes of a piezoelectric antenna with simultaneous estimation of their errors is considered. To increase the localization accuracy, it is proposed to refine the times of arrival by linearly approximating the leading edge of an AE signal taking into account the influence of noise. The slope of the leading edge is calculated according to the criterion of the minimum rms deviation from the signal envelope using the moving-window method. Examples of use of the developed method for calculating the coordinates of flaws in strength tests of structural elements are considered.

The crack resistance of specimens made of a fiberglass plastic composite material has been studied. All specimens had approximately the same geometric dimensions. At the center of the specimens, there were 25-mm-long notches on both sides. The specimens were loaded with a static or cyclic load at a frequency f = 5 Hz. The tests were performed using acoustic-emission (AE) and strain gage measurements. The AE technique allowed stable localization of a flaw at an early stage of growth and made it possible to automate the measurement process. The strain gage measuring system was used to find deformations in the zones of localization of AE signals.

The technique of acoustic-emission testing of curvilinear fuselage panels of an RRJ airplane in life tests is considered. Specimens are loaded with a distributed mechanical load and pressurization with an air overpressure.

The use of acoustic-emission (AE), optical, and strain-gaging methods in cyclic tests of an airplane has been considered. During tests, before the failure of an airplane, three zones were inspected. For this purpose, saw cuts were made in the lower planes of the left and right wings and in the load-carrying structure on the right side. The AE method allowed detection of growing flaws at a 10% load relative to the rated load, when the length of a fatigue crack was at most 1 mm and was not detected via the optical method.

A technique for continuous strain gaging during lifetime tests of structures with the use of static strain-gaging systems is developed. The performed studies allowed automatic monitoring of the readings of strain gages and the analysis of the fall of these readings beyond the tolerable limits. When the readings of a strain gage exceed the tolerable value, the loading system is stopped automatically.

A testing technique developed for determining the acceptance levels of different inspected zones of a wheel pair is considered. It has been shown that the acceptance levels are individual for different inspected zones and are set on the basis of the statistical data obtained during the tests of wheel pairs containing flaws detected beforehand. A test record sheet containing coordinates of flaws and comments on their hazard is issued during automated acoustic-emission testing of the wheel pairs of a freight-car carriage.

The results of cyclic tests of structural elements (SEs) with a stringer, which are made of the composite material (CM) Organit-10T and are parts of an aileron from an e-80Ee aircraft, are considered. As SEs are broken, information is obtained simultaneously from acoustic-emission (AE) and tensometric equipment and measurements of the fatigue crack’s opening are carried out. An insignificant increase (by 0.25 µm) of the crack opening generates a considerable flow of AE signals. These signals are processed using cluster analysis, which makes it possible to separate the signals emitted by a propagating fatigue crack in the stringer region from the signals arriving from the region of the fracturing gripping jaw.

The results obtained by acoustic-emission (AE) testing and continuous tensometry in endurance testing of half-axles of maneuverable aircraft’s stabilizers are described. In order to ensure the reliability of the tests, fractography of the fractures obtained is carried out along with metallographic studies of the material’s structure in the zone of fracture nucleation and propagation.

Experimental results on the localization of acoustic emission (AE) signals obtained in stress-cycling tests of specimens made from the Organit-10T composite material (CM) are considered. The technique developed for collecting prompt information on the changes in the shape of the AE amplitude distribution is analyzed. This distribution characterizes the flow of arriving signals and allows one to determine the beginning of destruction for CM specimens.

A multimodule strain-gauging measuring and data-processing system equipped with microprocessors in which 4096 channels are controlled by a single computer is designed. The introduction of a Host-controller makes it possible to carry out the parallel interrogation of the measurement channels. This increases the general productivity of the system and shortens the length of the connected cables by more than a factor of 10.

The paper reports on basic underlying principles of methods for evaluating the danger degree of fatigue cracks in the acoustic emission (AE) testing of cast pieces of freight-car trucks on the basis of statistical dependences of the AE signal parameters in the aw development stage.

The paper describes a new approach to designing multichannel AE diagnostic systems using signal processors. A new processing algorithm is suggested to eliminate isolated spurious single pulses in processed AE signals. Practical results are reported which were obtained with the help of the microprocessor system in strength tests of a panel of a heavy transport plane.

Five rejection criteria for acoustic-emission (AE) testing of metallic structures are discussed. The paper contains analytical expressions for calculating rejection levels and a practical example of a calculation of the degree of hazard caused by a aw detected by the AE method in an oil-gasoline tank.

Results of cyclic stress tests of planar samples from the D16T duralumin 2 and 6 mm thick performed with the help of the methods of acoustic emission (AE) and strain measurements are reported. The variables recorded in the tests were the applied load, number of loading cycles, length of a fatigue crack, strain in sample material, and parameters of AE signals. In addition, in tests of samples 3 mm thick the stress intensity factor was maintained constant as fatigue cracks grew over fixed intervals of lengths, and information was derived from total counts of AE pulses, their energy distributions and spectral characteristics.

Results of practical hydraulic tests of railway tanks owned by Altaivagon Corporation by the method of acoustic emission (AE) are given. A new technique for localization of AE signals and an appropriate software package have been developed to detect with a high certainty both internal and penetrating flaws in tanks under a load of 0.5 atm. A technique for estimating the hazard presented by the flaws in railway tanks based on detection of AE signals is discussed.

The structure of a 64-channel microprocessor strain-measuring system designed for operation with strain gauges and thermal transducers is discussed. The system measures strain and temperature during strength and thermal-strength tests of structures of machines and buildings. The computer codes have been designed in the Borland C++ Builder medium for operations with isolated transducers (1/4-bridge mode), and in the modes “1/2-bridge,” “bridge,” and “thermal transducer.” The system has been certified. The paper describes examples of its practical utilization in strength tests of struts of the Tupolev-154B airliner and of the undercarriage foreleg of an aerobatic airplane.

A technique for calculating coordinates of acoustic emission (AE) sources in metallic tested objects with the help of empirical (tabulated) data is discussed. An example of practical application of the method to strength tests of sickle-shaped suspensions and Cardan shafts of electric locomotives in conditions of car barns is presented.