Fracture mechanics requires the evaluation of defects in structural elements by nondestructive test methods. It is necessary to know the geometry of the defects, the microstructure of the material, and the stresses near the defects. Acoustic emission is used to investigate the crack growth. We also give a brief overview of recent advances in this field.

A new method for automatic tool wear monitoring at turning has been developed based on the analysis of the continuous acoustic emission (machining noise) generated by the tool during machining. Different wear types (wear of tool flank face and tool chipping) result in changes in the different characteristic values of the noise signal. In case of a uniform abrasion of the insert, e.g., flank face or crater wear, an increased mean signal level is observed, whereas for microbreakage at the edge, an increase of the crest factor with nearly constant mean signal level is found. The burst-like signals from collision between chip and tool and from chip breakage have to be eliminated from analysis to avoid the distortion of the signal parameters of the continuous acoustic emission. This method should be well suited especially for monitoring of finishing processes (small depth of cut).

In the last years many hydrotests, cyclic fatigue tests and thermal shock tests have been carried out on reactor pressure vessels, which exhibited crack-like defects, with the aim to develop the conditions of detectability of growing and non-growing cracks by acoustic emission (AE). The AE generated during these experiments has been recorded and analyzed. Techniques for evaluation of AE-data have been developed which allowed the separation of AE-events from crack growth and crack surface friction by risetime. Crack growth can be detected with high sensitivity at thermal shock and fatigue tests, whereas the detectability of crack surface friction increases, as the amount of compressive stresses between the crack surface grows. At hydrotest only a low level AE is produced by non-growing cracks due to the lack of compressive stresses.

Usually burst-like acoustic emission (AE) is localized by triangulation. For continuous AE, e.g. from leakages, this method is not feasible. Therefore a new method for localization of continuous AE has been developed. It is based on a phase-controlled probe array which consists of many single sensor elements. The AE signals received by the different sensor elements are delayed according to their time-of-flight differences from the source to the single elements of the receiver array. By choosing special combinations of time differences between the array elements the directivity pattern of the sensitivity of the array can be changed, e.g. rotated in the plane of a large plate. Thus, the source direction can be determined by one array. Some preliminary experiments with an artificial noise source, positioned on a large steel plate, have been performed and have demonstrated the feasibility of this approach.

To demonstrate and to extend the performance of acoustic emission testing as a method of detecting and classifying flaws, six institutes conducted acoustic emission measurements in the course of various loading tests on a medium-sized, thick-walled vessel (model of a reactor pressure vessel) containing natural flaws. This paper will present a description of the vessel and the preparation of the flaw patch with 14 natural flaws, the performance of the loading tests to simulate pressure test and operating conditions existing in the primary systems of pressurized water reactors. and especially the conduction of acoustic emission measurements by using various monitoring systems.In the pressure tests conducted with slowly rising pressures, only one flaw was detected unequivocally by the acoustic emission monitoring, although several flaws had grown in the test phases between the pressure tests. Cyclic loading over prolonged periods of time produced clear signals of larger flaws, which calculations and subsequent destructive investigations showed to have grown. The small flaws which, most probably, had not changed, could not be detected.

The examination of the crack growth processes during thermal shock experiments performed with a corrosive medium on a cladded nozzle of the HDR-RPV was carried out by an acoustic emission and analysis system, that is able to automatically identify acoustic emission events as crack growth or as other processes (such as crack friction). This classification is based on a statistical model that considers different signal parameters (rise time, endurance, energy of the signal) for each acoustic event. More often, additional signal parameters are the position and time of starting. Using an evaluation system it is possible to pick out the events with certain parameters, using many different combinations, and plot and compare them against other results obtained from measurements or calculations. In this way the acoustic emission location method is confirmed by a general comparison with a nondestructive test result. In contrast, the timing of the signal start compared with the cyclic crack loading was unexpected. Most of the acoustic signals did not occur when the thermal load was at a maximum, but during an intervening period during the relief phase which is caused by reheating. In order to find the reasons for this, the signals are separated according to fracture mechanics models. From this it was discovered that the acoustic emission model and the fracture mechanics model are partly in agreement: that is, for the friction aspects. Finally the acoustic emission results suggest the following important and new conclusion regarding the mechanism of crack propagation: in the particular case of the nozzle experiments the crack growth is not primarily controlled by the load parameter delta-K, but occurs only during a short period of adverse temperature. This conclusion should be proved by means of further investigations on the HDR-RPV. The results are conclusive especially when applying the results from samples to components. This paper demonstrates that current ideas on fracture mechanics processes are only possible: • - through the use of acoustic emission testing, • - through a powerful evaluation and presentation system for the measured results which include acoustic emission, • - and not least through the simulation of the real-life conditions found in a nuclear power station (HDR).

AE measurements are presented which were performed at pressure vessel components during different loadings: thermal shocking, hydrotest, cyclic loading. The tests were executed with the IzfP AE system. With this it is possible to locate the AE sources and to identify two source mechanisms: crack growth and crack flank friction. The AE results are discussed and compared with the results of calculations about the residual stress in a cladded vessel wall: • — AE signals originating from crack growth processes are detected very sensitively. • — During thermal shock a lot of crack border friction occurs in consequence of compressive stresses on the crack flanks. • — However, during the hydrotests no crack border friction appears, because the crack flanks are mainly under tensile stresses.

The investigations were aimed at demonstrating the state of the art of acoustic emission testing (AET) of reactor pressure vessels. The object under investigation was the large reactor pressure vessel of the MPA in Stuttgart, a boiling-water reactor pressure vessel, which was provided with a multitude of flaws in weld seams and in the base material. Six hydrostatic tests approximately up to the working pressure of a boiling-water reactor (71 bar) were carried out. In addition to the global multichannel locating technique, also local monitoring techniques were applied. Global location permitted a large number of different indications to be detected simultaneously. Not all of the known flaws did, however, show the expected number of AE events. On the other hand, it was possible to detect flaws previously unknown to the AE staff in some weld seams; these indications were confirmed by nondestructive testing. It was demonstrated that the locating accuracy of local monitoring using signal analysis was improved by a factor of 20 to 30 compared to global monitoring.

In the framework of a reactor safety research programme, thermal shocking of a clad feed water nozzle has been carried out during simulated operating conditions (temperature = 300°C, pressure = 11 MPa). The objective of these trials is to contribute to the understanding of crack initation and propagation under realistic thermal shock conditions, and to monitor crack initiation and growth by NDT on-line as well as off-line. Up to now the nozzle corner has been shocked 5200 times. The shocked area was monitored by acoustic emission (AE). During the last 800 thermal shocks new AE equipment with optimized capabilities for detection, location of AE sources and interpretation of AE signals has been used. This AE system and the results obtained during the last 800 shocks will be described. One important result is that it is possible to separate crack-growth signals from friction noise of the cracks by signal analysis when the AE signals are received with broad-band transducers.

Cyclic thermal shock tests have been performed on a RPV nozzle (HDR) with cracks in order to apply and to assess technics available for the description of real cracks (non-destructive testing, fracture mechanics calculation). In this paper the application of dye penetrant testing, potential drop testing and acoustic emission measurements on the nozzle is considered. Some experimental results are described in detail; assessments have been possible on the basis of destructive testing (fractography). The determination of crack parameters for correlating these NDT results with fracture mechanics data are discussed such as crack depth, a, cracklength, crack area and crack growth. Crack growth is considered over a period of cycles, a(N), per cycle, da/dN, and during a cycle, da/dt. A selection of numerical NDT findings is presented and compared with the results of fracture mechanics calculations. The investigation leads to the following conclusions: 1.1. The crack dimensions found by NDT are not quite conservative (fractography).2.2. As the performed fracture mechanics calculations overestimated the crack growth advanced NDT data like da/dN indicate starting points for improvements of the calculations (influence of mean stress).3.3. Although acoustic emission measurements give no quantitative fracture mechanics parameters qualitative correlations yield decisive understanding of dominating crack growth mechanisms (corrosion).4.4. Reversely different comparisons between acoustic emission and fracture mechanics results seem to confirm the classification of acoustic emission signals (crack surface friction, crack growth).