CA1194977A - Process for detecting an acoustic source and an acoustic emission monitoring apparatus - Google Patents
Process for detecting an acoustic source and an acoustic emission monitoring apparatusInfo
- Publication number
- CA1194977A CA1194977A CA000394000A CA394000A CA1194977A CA 1194977 A CA1194977 A CA 1194977A CA 000394000 A CA000394000 A CA 000394000A CA 394000 A CA394000 A CA 394000A CA 1194977 A CA1194977 A CA 1194977A
- Authority
- CA
- Canada
- Prior art keywords
- acoustic source
- difference
- acoustic
- waves
- source
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H1/00—Measuring characteristics of vibrations in solids by using direct conduction to the detector
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R7/00—Diaphragms for electromechanical transducers; Cones
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- Signal Processing (AREA)
- General Physics & Mathematics (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Length Measuring Devices Characterised By Use Of Acoustic Means (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
A PROCESS FOR DETECTING AN ACOUSTIC SOURCE AND AN ACOUSTIC EMISSION
MONITORING APPARATUS
ABSTRACT OF THE DISCLOSURE
The invention provides a process and an apparatus which by detection of at least two propagation or polarization modes of the same acoustic wave allows a signal to be elaborated whose duration is representative of the difference in detection time between the two modes. The result allows the acoustic source to be located and the nature thereof to be given.
MONITORING APPARATUS
ABSTRACT OF THE DISCLOSURE
The invention provides a process and an apparatus which by detection of at least two propagation or polarization modes of the same acoustic wave allows a signal to be elaborated whose duration is representative of the difference in detection time between the two modes. The result allows the acoustic source to be located and the nature thereof to be given.
Description
7t7 BACI'GROUND OF THE INVENTION
The present invention relates to a process for detecting an acoustic source and an acoustic emission monitoring device. It relates then to industrial acoustic emission monitoring.
In an industrial installation a certain number of p~enomena are gen-erators of noise. These phenomena are o~ two types :
- Thermal or other actions cause local mechanical stress zones. These stresses, when they exceed a certain resistance rate of the material, may cause cracking or breaks.
- Mechanical actions such as the movement of fluids in piping (turbulences, hammering, etc.) or else projection impacts may also in the long run cause destruction of the equipment.
In order to avert these accidents, it has seemed necessary to perma-- nently monitor the zones,damage to which might cause serious disadvantages (high-pressure piping, nuclear reactor vessel, turbo-alternator support platform, etc.).
In the prior art constructions, acoustic emission monitoring systems were provided comprising a plurality of sensors, each connected to a receiv-ing chain which detects the acoustic wave which is propagated in the mat-erial at a speed characteristic of this latter. By testing this latter can be known. By measuring the time intervals separating the reception of the wave by each sensor, the problem of locating the p~npoint source may be re-solved by hyperbolic triangulation with the help of a computer or else an approximative zone where the source is situated may be determined. Alarms may be disposed so as to take into account the acoustic emission sources with high breakdown risk.
BREIF SUMMARY OF THE INVENTION
These constructions require the use of substantial calculating means.
To reduce their complexity, the present invention proposes a process for 3 detecting an acoustic source based on the measurement of the detection time '~';~
differences of the different wave modes emitted by the acoustic source.
Another object of the invention is an acoustic emission monitoring apparatus which uses the process of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention will be better understood from the descrip-tion and accompanying drawings which show :
Figure 1, an industrial structure monitored with two sensors;
Figure 2, an industrial structure monitored with one sensor;
Figure 3 d a unidimensional location method;
Figure 3b, a two-dimensional method of the extension thereof;
- Figure 4, a mode-distinction timing diagram;
Figure 5, a diagram of a detection chain and a processing chain;
Figure 6, a diagram of another detection chain and another process-ing chain;
Figure 7, an analog locating apparatus; and Figure 8, a digital locating apparatus.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
In an industrial structure are to be found different elements such as tanks, blocks, pipes, valves, ducts, doors, etc. A number of these elements are subjected to stresses which may locally form acoustic emission sources~ The appearance of these sources are unpredictable in time, in space and are of variable duration. A f`ew of the causes thereof have been described. The source is then formed by a digital-stress gradient which persists for a short time in a small zone of the material. The gradient
The present invention relates to a process for detecting an acoustic source and an acoustic emission monitoring device. It relates then to industrial acoustic emission monitoring.
In an industrial installation a certain number of p~enomena are gen-erators of noise. These phenomena are o~ two types :
- Thermal or other actions cause local mechanical stress zones. These stresses, when they exceed a certain resistance rate of the material, may cause cracking or breaks.
- Mechanical actions such as the movement of fluids in piping (turbulences, hammering, etc.) or else projection impacts may also in the long run cause destruction of the equipment.
In order to avert these accidents, it has seemed necessary to perma-- nently monitor the zones,damage to which might cause serious disadvantages (high-pressure piping, nuclear reactor vessel, turbo-alternator support platform, etc.).
In the prior art constructions, acoustic emission monitoring systems were provided comprising a plurality of sensors, each connected to a receiv-ing chain which detects the acoustic wave which is propagated in the mat-erial at a speed characteristic of this latter. By testing this latter can be known. By measuring the time intervals separating the reception of the wave by each sensor, the problem of locating the p~npoint source may be re-solved by hyperbolic triangulation with the help of a computer or else an approximative zone where the source is situated may be determined. Alarms may be disposed so as to take into account the acoustic emission sources with high breakdown risk.
BREIF SUMMARY OF THE INVENTION
These constructions require the use of substantial calculating means.
To reduce their complexity, the present invention proposes a process for 3 detecting an acoustic source based on the measurement of the detection time '~';~
differences of the different wave modes emitted by the acoustic source.
Another object of the invention is an acoustic emission monitoring apparatus which uses the process of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention will be better understood from the descrip-tion and accompanying drawings which show :
Figure 1, an industrial structure monitored with two sensors;
Figure 2, an industrial structure monitored with one sensor;
Figure 3 d a unidimensional location method;
Figure 3b, a two-dimensional method of the extension thereof;
- Figure 4, a mode-distinction timing diagram;
Figure 5, a diagram of a detection chain and a processing chain;
Figure 6, a diagram of another detection chain and another process-ing chain;
Figure 7, an analog locating apparatus; and Figure 8, a digital locating apparatus.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
In an industrial structure are to be found different elements such as tanks, blocks, pipes, valves, ducts, doors, etc. A number of these elements are subjected to stresses which may locally form acoustic emission sources~ The appearance of these sources are unpredictable in time, in space and are of variable duration. A f`ew of the causes thereof have been described. The source is then formed by a digital-stress gradient which persists for a short time in a small zone of the material. The gradient
2~ is reduced to zero while emitting its stress energy outwardly of the zone.
The wave, generating any modes whatsoever, is then propagated in different media. In a given medium, the pressure wave may, depending on the shape and the nature of the element of the industrial structure considered, be propagated in accordance with different modes which coexist in propagation or polarization modes. Different wave polarization modes are :
7~
- surface-wave mode;
- transverse-wave mode;
- longitudinal-wave mode, etc.
Each mode transports a greater or lesser fraction of the stress energy depending on its preponderance in the overall acoustic wave. It may be noted that each propagation mode is distinct from the others for numerous physical parameters : speed, duration, amplitude, spectral band, etc.
Measurements may then allow each mode to be selected by distinction.
Furthermore, when two media are acoustically coupled, a fraction of the cverall incident wave passes through the following medium. But the characteristics of the wave transmitted in the following medium, with res-pect to those of the initial wave, are modified. This is the case of the propagation of the wave emitted by a stress in a tank and its transmission in the ambient air. It is therefore possible to obtain information about the acoustic source from several propagation media.
In Figure 1, a cylinder represents a duct portion 1. An acoustic source 2 appears momentarily. In accordance with the invention, two sensors
The wave, generating any modes whatsoever, is then propagated in different media. In a given medium, the pressure wave may, depending on the shape and the nature of the element of the industrial structure considered, be propagated in accordance with different modes which coexist in propagation or polarization modes. Different wave polarization modes are :
7~
- surface-wave mode;
- transverse-wave mode;
- longitudinal-wave mode, etc.
Each mode transports a greater or lesser fraction of the stress energy depending on its preponderance in the overall acoustic wave. It may be noted that each propagation mode is distinct from the others for numerous physical parameters : speed, duration, amplitude, spectral band, etc.
Measurements may then allow each mode to be selected by distinction.
Furthermore, when two media are acoustically coupled, a fraction of the cverall incident wave passes through the following medium. But the characteristics of the wave transmitted in the following medium, with res-pect to those of the initial wave, are modified. This is the case of the propagation of the wave emitted by a stress in a tank and its transmission in the ambient air. It is therefore possible to obtain information about the acoustic source from several propagation media.
In Figure 1, a cylinder represents a duct portion 1. An acoustic source 2 appears momentarily. In accordance with the invention, two sensors
3, 4 have been previously disposed so as to monitor the duct 1. Sensor 3 is placed in the air close to duct 1. Sensor 4 is acoustically coupled to its surface. Sensor 3 receives the ambient wave 5 emitted from source 2 into the ambient air, whereas sensor 4 receives the stress wave 6.
In Figure 2, a single sensor 7 has been acoustically coupled to duct 1. It receives from source 2 an overall wave which comprises at least two propagation modes. Only the two preponderant modes have been shown for waves 8 and 6, which may be for example surface waves and longitudinal waves.
The process of the invention consists then in measuring at least one characteristic parameter knowing the characteristics of the medium or of the media. In the case of Figure 1 we may consider, in accordance with the in~
vention, measuring the time interval which separates the reception of wave 6 7~7 from that of wave 5. Designating by t1 the traveling time of wave 6 which travels over an unknown distance d1 as far as sensor 4, at the known speed v1, we have :
d1 ~ v1t1 By changing the index, the same magnitudes may be given for sensor 3 and wave d2 = V2 t2 d1, d2, t1, t2 are unknown. v1, v2 may be measured by periodically testing with a noise generator or acoustic test emitter. But the time interval ~t ~o separating the reception of wave 6 from that of wave 5 may also be measured, i.e~ t1 - t2.
Thus the four linear equation system is obtained :
d1 = v1t1 d2 = v2t2 ~ 2 ~t = t1 ~ t2 which comprise four unknowns : dl, d~, t1, t2, three known magnitudes : v1, V2 and d iinear distance between the two sensors and a magnitude supplied by measurement : ~t. This system is resolved in a conventional way.
If we contemplate the case of a bimode sensor shown in Figure 2, the preliminary measurement of d is eliminated, d being the distance between the two sensors, which ~s equal to zero, the sensor only occupying a point.
The equations are :
v1t1 - v2t2 = 0 t1 ~ t2 = ~t dl = vltl d2 = V2 t2 which are resolved in the same way, with for index 1 the magnitudes designa-ting one propagation mode and for index 2 the other propagation mode detec-ted.
9~77 The propagation equations may be chosen more complete and not inthe form d = vt without modifying the process of the invention.
The process of the invention such as described here calls forth two remar~s :
- It is written for a practi~ally linear structure having a great longitudinal dimension and a very small cross-section. This is the case for certain pipes and bars for concrete.
- The information must be complete so as to describe that the overall wave comes in one direction or in the other.
To extend the field of application of the process of the invention, there is proposed in Figure 3 an arrangement with two bimode sensors 10, 11 acoustically coupled to a curved plate 9, forming part for example of a tank wall or similar. Source 12, placed in a random way, has emitted a wave in the direction of sensor 10 and another in the direction of sensor 11.
For each of these sensors, it has been possible to trace an isodetection line 13 or 14 which corresponds to the localil;y of the acoustic source which, emitting towards the sensor considered 10 or 11, would be detected by it at the same time and this for each polarization mode of the wave.
Source 12 appears clearly as being at the intersection of the two localities.
It is remarkable that the lack of determination of the direction of propagation of the wave relative to the sensor is found again here since, in Figure 3, it can be seen that localities 13, 14 intersect each other twice. To remove this lack of determination, it is necessary to use three sensors. Each of them receives and distinguishes two waves which allow it to choose an isodetection locality per sensor. The common intersection of the three localities allows the source to be accurately located.
This method of removing the lack of determination is shown in Fig-ure 3b.
In this figure, three isodetection localities A2, B3, C2 have been plotted. Their shapes correspond to particular structures comprising 37~7 particular materials. A simple particular case, often a ~ood approxima-tion of the problem, is that of the circle centered on the sensor. Here, around each sensor A, B, C has been plotted a family of isodetection locali-ties. These localities are drawn from topogram plots on the structure itself. With the three sensors positioned an appropriate noise generator is placed at a large number of points. The points corresponding to equal detection-time differences between two modes are associated in the same lo-cality, called isodetection locality.
With the isodetection locality ~amilies known, when an acoustic emitting defect is detected, the measurement of the three intermode detection-time differences allows three localities to be selected, one per family, whose intersection provides the location of the def`ect shown at M.
In order to reduce the disadvantages of providing topograms, it is also possible to calculate directly the distance from a single previous measurement : that of the propagation speed of an acoustic wave in the structure monitored.
In certain configurations, it is possible to avoid the need of re-moving the lack of determination when a single one of the points calculated from a single measurement is physically possible, the others not correspond-ing to a point in the monitored structure or else the system being incapable of detecting them. This latter case is that of a material structure of finite extent, the sensor being placed close to the edge of the structure~
If this latter is not acoustically coupled to a neighboring structure, the only waves received by the sensor come from a single side of the space.
The invention also relates to an acoustic emission monitoring appa-ratus implementing the process of the invention. A detection chain detects waves having different characteristics l~hich are distinguished and analysed by a signal processing chain so as to supply information about the acciden-tal acoustic source emitting the detected waves.
This information may be information concerning the position or ~L9g977 nature of the source. The invention also relates to use of the '~propaga-tion time~ parameter which leads to locating the acoustic source~ In Figure 4, there is shown the elaboration of a propagation time difference signal. Signal 20 corresponds to the signal detected by a bimode sensor.
There appears a first evanescent wave 15, then another evanescent wave 16.
If the sensor is provided to detect a longitudinal-mode wave and a surface-mode wave, then the mode distinction is easy. In fact, the surface wave of great amplitude is slower than the longitudinal wave of smaller amplitude and is therefore detected after the longitudinal wave.
Signal 21 comprises two square-waves 17 and 18 triggered off respec-tively by waves 15 and 16 and of a duration fixed for example by a monostable so as to blind the system from noise and parasites. Furthermore, detec~
tion is carried out with a comparator which only takes into consideration the waves exceeding a certain amplitude.
Signal 22 is formed from signal 21 by Means of a flip-flop whose output changes state for each falling edge at its inputs. The square-wave 19 obtained is of a width T19 proportional to the detection time difference t1 ~ t2. The result is :
T19 = k.(t1 2 k is a characteristic coefficient of the detection and processing chain.
This chain supplies therefore the "detection time difference" parameter of the modes distinguished.
~ chain of this type has been shown in Figure 5. ~ bimode sensor 23 is connected to an amplifier 24 for reshaping the signals delivered by sensor 23. The amplified signal is fed to an envelope detector whose out-put is connected to the clock input 260 of a flip-flop 26. This latter is initialized manually by means of the RAZ control 261 through a switch.
The output 263 of the fl:ip-flop delivers a square-wave of duration T19.
The operation of the chain may be follGwed in Figure 5. Sensor 23 receives two energizations corresponding to the two propagation modes of the acoustic wave emitted by the defect. Sensor 23 delivers a signal 230. At the output of the thresholA detector 25, adjusted by its reference input 250, a signal is obtained with two square-waves 251. Since flip-flop 26 changes state at each rising transition at its clock input, a square-5 wave 262 is obtained the duration of which represents the mode-detection time interval ~t.
In Figure 6 there is shown an analog variation with two sensors 30 and 31. This is the case for the embodiment of Figure 1. Each sensor is connected to a detection and shaping chain which comprises an amplifier 32 (or 33), envelope detector 34 (or 35), an ampl:ifier with monostable 36 and flip-flop 37. The functions and operations thereof are described in Fig-ure 5.
For using square-wave 19 of Figure 4, representing the "detection time difference", two possible solutions are presented. The analog solu-tion described in Figure 7 provides rapid location for a substantially uni-dimensional case. Square-wave 19, supplied for example by flip-flop 26 of Figure 5, is fed to a voltage-ramp generator 40 supplied with DC voltage 400.
At the output 401, a voltage ramp 402 is obtained which begins at the rising edge of square-wave ~9 and remains stationary 404 from the falling edge.
20 rrhis signal is applied to the horizontal (or vertical) deflection plates 410 of a cathode-ray tube 41.
Square-wave 19 is also fed to a falling-edge detector 42 which, at its output, emits a pulse 420 when square-wave 19 falls back to the inac-tive level. This pulse 420 is fed to a control 43 controlling the firing of the electron gun 411 of tube 41.
The screen, graduated in distance, by means of a scale factor taking into account the propagation speed of the acoustic wave and the slope of ramp 403, will bear a light trace 412 which will represent the position of the located defect~ The persistence of this trace may be ensured by means of an afterglow screen, by taking a photograph of screen 41 (for example by ./
'7~7 means of an additional detection of the rising edge of square-wave 19) or else by means for repeating the electronic image.
The handling means of the processing may follow another digital-type solution. It consists in providing an array of bimode sensors. ~ach bimode sensor is associated with a chain for detecting the signal and a chain for processing the signal. At the output of this latter, there is obtained in t~le example described here, the detection time difference bet-ween the two distinguished modes of-the acoustic wave. For a given sensor, there exists a locality and a single one of the equal-detection points, called isodetection locality, corresponding to a given value of the detection time difference. We saw above that the source was the common point between the three localities associated with the three sensors of the two-dimension-al locating apparatus.
In Figure 8, there has then been shown a two-dimensional acoustic source locating apparatus of a digital type. Three bimode sensors 44-46 are coupled acoustically to the monitored s~face, at given fixed points.
These points are the poles of the isodetection localities previously measured for example by the method already described. Sensors 44-46 are connected to chains 47-49 for detecting the signal and to chains 50-52 for processing the signal, of the type for example described with reference to Figure 3.
The data available at each of the outputs of the three chains 50-52 is a square-wave 19, fed to an isodetection-locality selector 53-55.
This digital-type selector has been drawn in detail in Figure 8 for selector 53, selectors 54, 55 being structurally identical therewith.
The square-wave 19 is fed to a clock 53t). This clock 530 is initialized by the rising edge of square-wave 19 and inhibited by its falling edge. The binary word present at the output of clock 530 is then represen-tative of the value of the difference in detection time between the two propagation modes distinguished by the acoustic sensor 44. It allows the generation, through a reading-order generator 531, of a sequence which 7~
comprises the selection of an address zone in a prerecorded memory 532 to which generator 531 i~ connected and the transfer orders from this memory zoné to an isodetection locality register 533 to which the memory 5~Z of the isodetection localities is connected.
The apparatus described also comprises a coordinate data comparator 56 which receives at inputs 560-562 the coordinate data from the isodetector locality registers such as register 533 of selectors 53-55 which are sequen-tially cleared of their respective contents by a control 563 so as to select the coordinate data corresponding to the position of the acoustic source.
This data 564 is then transferred to a display 57 or else a graph which allows the position to be immediately discerned on a screen or else by means of an alphanumeric display in coded form.
In any case, a much simpler solution by circular triangulation may be adopted as was described above. The single data relating to the radius is sufficient for the coordinate data comparator 56 to find ~he angular po-sition, in an isodetection locality circle, of the acoustic source to be located.
It is possible to add to the position information of the emitting source an indication of the nature of the defect. By using a means repeat-ing the detection, at the end of the operation, at a rate compatible withthe geometric growth of the acoustic source, for example 32 interrogations per second, it is possible to pick up the acoustic wave several times before it dies away~ Moreover, should the defect move, the locating points of the defect are not identical but allow for example the image of the crack to be obtained in real time and its progression. It is possible to connect~
at the first intervention of the sensors, a cinematographic recording appa-ratus which allows the image to be reconstructed in the time of growth of the crack, which may allow it to be studied.
Other parameters may also be considered other ~han the "detection time difference~, such as the spectrum per mode so as to obtain other :~94~77 representations of the defect.
With respect to a conventional detection chain operating for example with the hyperbolic triangulation process, it can be noted that the number of sensors required for locating the acoustic source defect is reduced by one sensor. This is then made possible because of the increase in the num-ber of parameters detected, since two modes are distinguished instead of a sin~le wave as in the conventional case.
1~
In Figure 2, a single sensor 7 has been acoustically coupled to duct 1. It receives from source 2 an overall wave which comprises at least two propagation modes. Only the two preponderant modes have been shown for waves 8 and 6, which may be for example surface waves and longitudinal waves.
The process of the invention consists then in measuring at least one characteristic parameter knowing the characteristics of the medium or of the media. In the case of Figure 1 we may consider, in accordance with the in~
vention, measuring the time interval which separates the reception of wave 6 7~7 from that of wave 5. Designating by t1 the traveling time of wave 6 which travels over an unknown distance d1 as far as sensor 4, at the known speed v1, we have :
d1 ~ v1t1 By changing the index, the same magnitudes may be given for sensor 3 and wave d2 = V2 t2 d1, d2, t1, t2 are unknown. v1, v2 may be measured by periodically testing with a noise generator or acoustic test emitter. But the time interval ~t ~o separating the reception of wave 6 from that of wave 5 may also be measured, i.e~ t1 - t2.
Thus the four linear equation system is obtained :
d1 = v1t1 d2 = v2t2 ~ 2 ~t = t1 ~ t2 which comprise four unknowns : dl, d~, t1, t2, three known magnitudes : v1, V2 and d iinear distance between the two sensors and a magnitude supplied by measurement : ~t. This system is resolved in a conventional way.
If we contemplate the case of a bimode sensor shown in Figure 2, the preliminary measurement of d is eliminated, d being the distance between the two sensors, which ~s equal to zero, the sensor only occupying a point.
The equations are :
v1t1 - v2t2 = 0 t1 ~ t2 = ~t dl = vltl d2 = V2 t2 which are resolved in the same way, with for index 1 the magnitudes designa-ting one propagation mode and for index 2 the other propagation mode detec-ted.
9~77 The propagation equations may be chosen more complete and not inthe form d = vt without modifying the process of the invention.
The process of the invention such as described here calls forth two remar~s :
- It is written for a practi~ally linear structure having a great longitudinal dimension and a very small cross-section. This is the case for certain pipes and bars for concrete.
- The information must be complete so as to describe that the overall wave comes in one direction or in the other.
To extend the field of application of the process of the invention, there is proposed in Figure 3 an arrangement with two bimode sensors 10, 11 acoustically coupled to a curved plate 9, forming part for example of a tank wall or similar. Source 12, placed in a random way, has emitted a wave in the direction of sensor 10 and another in the direction of sensor 11.
For each of these sensors, it has been possible to trace an isodetection line 13 or 14 which corresponds to the localil;y of the acoustic source which, emitting towards the sensor considered 10 or 11, would be detected by it at the same time and this for each polarization mode of the wave.
Source 12 appears clearly as being at the intersection of the two localities.
It is remarkable that the lack of determination of the direction of propagation of the wave relative to the sensor is found again here since, in Figure 3, it can be seen that localities 13, 14 intersect each other twice. To remove this lack of determination, it is necessary to use three sensors. Each of them receives and distinguishes two waves which allow it to choose an isodetection locality per sensor. The common intersection of the three localities allows the source to be accurately located.
This method of removing the lack of determination is shown in Fig-ure 3b.
In this figure, three isodetection localities A2, B3, C2 have been plotted. Their shapes correspond to particular structures comprising 37~7 particular materials. A simple particular case, often a ~ood approxima-tion of the problem, is that of the circle centered on the sensor. Here, around each sensor A, B, C has been plotted a family of isodetection locali-ties. These localities are drawn from topogram plots on the structure itself. With the three sensors positioned an appropriate noise generator is placed at a large number of points. The points corresponding to equal detection-time differences between two modes are associated in the same lo-cality, called isodetection locality.
With the isodetection locality ~amilies known, when an acoustic emitting defect is detected, the measurement of the three intermode detection-time differences allows three localities to be selected, one per family, whose intersection provides the location of the def`ect shown at M.
In order to reduce the disadvantages of providing topograms, it is also possible to calculate directly the distance from a single previous measurement : that of the propagation speed of an acoustic wave in the structure monitored.
In certain configurations, it is possible to avoid the need of re-moving the lack of determination when a single one of the points calculated from a single measurement is physically possible, the others not correspond-ing to a point in the monitored structure or else the system being incapable of detecting them. This latter case is that of a material structure of finite extent, the sensor being placed close to the edge of the structure~
If this latter is not acoustically coupled to a neighboring structure, the only waves received by the sensor come from a single side of the space.
The invention also relates to an acoustic emission monitoring appa-ratus implementing the process of the invention. A detection chain detects waves having different characteristics l~hich are distinguished and analysed by a signal processing chain so as to supply information about the acciden-tal acoustic source emitting the detected waves.
This information may be information concerning the position or ~L9g977 nature of the source. The invention also relates to use of the '~propaga-tion time~ parameter which leads to locating the acoustic source~ In Figure 4, there is shown the elaboration of a propagation time difference signal. Signal 20 corresponds to the signal detected by a bimode sensor.
There appears a first evanescent wave 15, then another evanescent wave 16.
If the sensor is provided to detect a longitudinal-mode wave and a surface-mode wave, then the mode distinction is easy. In fact, the surface wave of great amplitude is slower than the longitudinal wave of smaller amplitude and is therefore detected after the longitudinal wave.
Signal 21 comprises two square-waves 17 and 18 triggered off respec-tively by waves 15 and 16 and of a duration fixed for example by a monostable so as to blind the system from noise and parasites. Furthermore, detec~
tion is carried out with a comparator which only takes into consideration the waves exceeding a certain amplitude.
Signal 22 is formed from signal 21 by Means of a flip-flop whose output changes state for each falling edge at its inputs. The square-wave 19 obtained is of a width T19 proportional to the detection time difference t1 ~ t2. The result is :
T19 = k.(t1 2 k is a characteristic coefficient of the detection and processing chain.
This chain supplies therefore the "detection time difference" parameter of the modes distinguished.
~ chain of this type has been shown in Figure 5. ~ bimode sensor 23 is connected to an amplifier 24 for reshaping the signals delivered by sensor 23. The amplified signal is fed to an envelope detector whose out-put is connected to the clock input 260 of a flip-flop 26. This latter is initialized manually by means of the RAZ control 261 through a switch.
The output 263 of the fl:ip-flop delivers a square-wave of duration T19.
The operation of the chain may be follGwed in Figure 5. Sensor 23 receives two energizations corresponding to the two propagation modes of the acoustic wave emitted by the defect. Sensor 23 delivers a signal 230. At the output of the thresholA detector 25, adjusted by its reference input 250, a signal is obtained with two square-waves 251. Since flip-flop 26 changes state at each rising transition at its clock input, a square-5 wave 262 is obtained the duration of which represents the mode-detection time interval ~t.
In Figure 6 there is shown an analog variation with two sensors 30 and 31. This is the case for the embodiment of Figure 1. Each sensor is connected to a detection and shaping chain which comprises an amplifier 32 (or 33), envelope detector 34 (or 35), an ampl:ifier with monostable 36 and flip-flop 37. The functions and operations thereof are described in Fig-ure 5.
For using square-wave 19 of Figure 4, representing the "detection time difference", two possible solutions are presented. The analog solu-tion described in Figure 7 provides rapid location for a substantially uni-dimensional case. Square-wave 19, supplied for example by flip-flop 26 of Figure 5, is fed to a voltage-ramp generator 40 supplied with DC voltage 400.
At the output 401, a voltage ramp 402 is obtained which begins at the rising edge of square-wave ~9 and remains stationary 404 from the falling edge.
20 rrhis signal is applied to the horizontal (or vertical) deflection plates 410 of a cathode-ray tube 41.
Square-wave 19 is also fed to a falling-edge detector 42 which, at its output, emits a pulse 420 when square-wave 19 falls back to the inac-tive level. This pulse 420 is fed to a control 43 controlling the firing of the electron gun 411 of tube 41.
The screen, graduated in distance, by means of a scale factor taking into account the propagation speed of the acoustic wave and the slope of ramp 403, will bear a light trace 412 which will represent the position of the located defect~ The persistence of this trace may be ensured by means of an afterglow screen, by taking a photograph of screen 41 (for example by ./
'7~7 means of an additional detection of the rising edge of square-wave 19) or else by means for repeating the electronic image.
The handling means of the processing may follow another digital-type solution. It consists in providing an array of bimode sensors. ~ach bimode sensor is associated with a chain for detecting the signal and a chain for processing the signal. At the output of this latter, there is obtained in t~le example described here, the detection time difference bet-ween the two distinguished modes of-the acoustic wave. For a given sensor, there exists a locality and a single one of the equal-detection points, called isodetection locality, corresponding to a given value of the detection time difference. We saw above that the source was the common point between the three localities associated with the three sensors of the two-dimension-al locating apparatus.
In Figure 8, there has then been shown a two-dimensional acoustic source locating apparatus of a digital type. Three bimode sensors 44-46 are coupled acoustically to the monitored s~face, at given fixed points.
These points are the poles of the isodetection localities previously measured for example by the method already described. Sensors 44-46 are connected to chains 47-49 for detecting the signal and to chains 50-52 for processing the signal, of the type for example described with reference to Figure 3.
The data available at each of the outputs of the three chains 50-52 is a square-wave 19, fed to an isodetection-locality selector 53-55.
This digital-type selector has been drawn in detail in Figure 8 for selector 53, selectors 54, 55 being structurally identical therewith.
The square-wave 19 is fed to a clock 53t). This clock 530 is initialized by the rising edge of square-wave 19 and inhibited by its falling edge. The binary word present at the output of clock 530 is then represen-tative of the value of the difference in detection time between the two propagation modes distinguished by the acoustic sensor 44. It allows the generation, through a reading-order generator 531, of a sequence which 7~
comprises the selection of an address zone in a prerecorded memory 532 to which generator 531 i~ connected and the transfer orders from this memory zoné to an isodetection locality register 533 to which the memory 5~Z of the isodetection localities is connected.
The apparatus described also comprises a coordinate data comparator 56 which receives at inputs 560-562 the coordinate data from the isodetector locality registers such as register 533 of selectors 53-55 which are sequen-tially cleared of their respective contents by a control 563 so as to select the coordinate data corresponding to the position of the acoustic source.
This data 564 is then transferred to a display 57 or else a graph which allows the position to be immediately discerned on a screen or else by means of an alphanumeric display in coded form.
In any case, a much simpler solution by circular triangulation may be adopted as was described above. The single data relating to the radius is sufficient for the coordinate data comparator 56 to find ~he angular po-sition, in an isodetection locality circle, of the acoustic source to be located.
It is possible to add to the position information of the emitting source an indication of the nature of the defect. By using a means repeat-ing the detection, at the end of the operation, at a rate compatible withthe geometric growth of the acoustic source, for example 32 interrogations per second, it is possible to pick up the acoustic wave several times before it dies away~ Moreover, should the defect move, the locating points of the defect are not identical but allow for example the image of the crack to be obtained in real time and its progression. It is possible to connect~
at the first intervention of the sensors, a cinematographic recording appa-ratus which allows the image to be reconstructed in the time of growth of the crack, which may allow it to be studied.
Other parameters may also be considered other ~han the "detection time difference~, such as the spectrum per mode so as to obtain other :~94~77 representations of the defect.
With respect to a conventional detection chain operating for example with the hyperbolic triangulation process, it can be noted that the number of sensors required for locating the acoustic source defect is reduced by one sensor. This is then made possible because of the increase in the num-ber of parameters detected, since two modes are distinguished instead of a sin~le wave as in the conventional case.
1~
Claims (23)
1. A process for locating an acoustic source in an industrial structure, the acoustic source being a pinpoint source or of a small extent, accidental or fortuitous in origin and emitting waves admitting in said structure at least two wave modes, said process comprising the steps of:
- detecting at least two waves emitted according to distinct wave modes;
- measuring at least one difference in detection time between said waves; and - processing said difference in order to localize said acoustic source.
- detecting at least two waves emitted according to distinct wave modes;
- measuring at least one difference in detection time between said waves; and - processing said difference in order to localize said acoustic source.
2. A process as defined in claim 1 wherein a difference in detection time is measured simultaneously at each of two different locations in the structure.
3. A process as defined in claim 2 wherein the processing of said difference comprises the step of removing the lack of determination between at least two possible locations of the acoustic source.
4. A process as defined in claim 3 wherein the removing of the lack of determination is achieved by measuring at least one additional difference in detection time at another location in the industrial structure.
5. A process for locating an acoustic source in an industrial structure, the acoustic source being a pinpoint source or of a small extent, accidental or fortuitous in origin and emetting waves admitting in said structure at least two wave modes, said process comprising the steps of:
- detecting at least two waves emitted according to distinct propagation modes;
- measuring at least one difference in detection time between said waves; and - processing said difference in order to localize said acoustic source.
- detecting at least two waves emitted according to distinct propagation modes;
- measuring at least one difference in detection time between said waves; and - processing said difference in order to localize said acoustic source.
6. A process for locating an acoustic source in an industrial structure, the acoustic source being a pinpoint source or of a small extent, accidental or fortuitous in origin and emitting waves admitting in said structure at least two wave modes, said process comprising the steps of:
- detecting at least two waves emitted according to distinct polarization modes;
- measuring at least one difference in detection time between said waves; and - processing said difference in order to localize said acoustic source.
- detecting at least two waves emitted according to distinct polarization modes;
- measuring at least one difference in detection time between said waves; and - processing said difference in order to localize said acoustic source.
7. A process as defined in claims 1, 2 or 3 wherein localization of said acoustic source is achieved by intersection of isodetection localities determined from measurements of differences in detection times.
8. A process as claimed in claims 1, 2 or 3 wherein localization of the acoustic source is achieved by determining the propagation speed of waves emitted according to distinct modes.
9. A process as claimed in claims 1, 2 or 3 wherein the localization is repeated at a rate compatible with the geometric growth of said acoustic source.
10. A process for locating an acoustic source in an industrial structure, the acoustic source being a pinpoint source or of a small extent, accidental or fortuitous in origin and emitting waves admitting in said structure at least two wave modes, said process comprising the steps of:
- detecting on at least one location on said structure, waves emitted according to distinct wave modes;
- measuring the difference in detection times between said waves; and - processing said difference to localize said acoustic source.
- detecting on at least one location on said structure, waves emitted according to distinct wave modes;
- measuring the difference in detection times between said waves; and - processing said difference to localize said acoustic source.
11. A process for locating an acoustic source in a substantially unidirectional industrial structure, the acoustic source being a pinpoint source or of a small extent, accidental or fortuitous in origin and emitting waves admitting in said structure at least two wave modes, said process comprising the steps of:
- detecting at least two waves emitted according to distinct wave modes;
- measuring a single difference in detection time between said waves; and - processing said difference in order to localize said acoustic source.
- detecting at least two waves emitted according to distinct wave modes;
- measuring a single difference in detection time between said waves; and - processing said difference in order to localize said acoustic source.
12. An acoustic emission monitoring apparatus comprising:
- a detection chain for an acoustic signal comprising sensor means responsive to at least two waves emitted according to distinct wave modes;
- a processing chain operatively connected to said detection chain for establishing a difference in detection times between waves emitted according to distinct wave modes;
- means monitoring said acoustic emission by utilizing said difference.
- a detection chain for an acoustic signal comprising sensor means responsive to at least two waves emitted according to distinct wave modes;
- a processing chain operatively connected to said detection chain for establishing a difference in detection times between waves emitted according to distinct wave modes;
- means monitoring said acoustic emission by utilizing said difference.
13. An apparatus as claimed in claim 12 wherein said sensor means comprises a sensor acoustically coupled to an industrial structure and another sensor acoustically coupled to an adjacent environment.
14. An apparatus as claimed in claim 12 wherein said sensor means comprises a bimode type sensor acoustically coupled to an industrial structure.
15. An apparatus as claimed in claims 12, 13 or 14 wherein said detection chain comprises an amplifier connected to a sensor.
16. An apparatus as defined in claims 12, 13 or 14 wherein said processing chain comprises at least one envelope detector and a flip-flop which switches at each transition at its clock input.
17. The apparatus as claimed in claim 12, wherein each output of said sensor means associated to a given propagation mode is connected to an amplifier itself connected to an envelope detector connected to an amplifier, connected to a flip-flop and wherein the gain of said amplifiers is adjustable by means of an external circuit.
18. The apparatus as claimed in one of claim 12, 13 or 14, wherein there are further provided means for counting the time initialized by a change in level of the signal form the chain for processing the signal, and inhibited by another change in level of the same signal.
19. An apparatus as defined in claim 14, wherein each bimode sensor is connected to an isodetection locality selector through a chain for detecting the acoustic signal and a chain for processing the signal, said iso-detection locality selector comprises successively connected together:
- a clock initialized by a transition of the signal emitted by the processing chain and inhibited by the following transition;
- a reading-order generator;
- a prerecorded memory of the isodetection localities;
- a register of the selected isodetection locality;
and wherein each output of said isodetection locality selector is connected to the inputs of a coordinate data comparator which delivers:
- at one output the reading synchronizations of the different isodetection locality registers;
- at one output, the coordinate data corres-ponding to the location of the acoustic source to a display device.
- a clock initialized by a transition of the signal emitted by the processing chain and inhibited by the following transition;
- a reading-order generator;
- a prerecorded memory of the isodetection localities;
- a register of the selected isodetection locality;
and wherein each output of said isodetection locality selector is connected to the inputs of a coordinate data comparator which delivers:
- at one output the reading synchronizations of the different isodetection locality registers;
- at one output, the coordinate data corres-ponding to the location of the acoustic source to a display device.
20. The apparatus as claimed in claim 19, wherein said display device is a bidimensional screen system.
21. The apparatus as claimed in claim 19, wherein said display device is an alphanumeric code display.
22. The apparatus as claimed in claim 19, wherein there is further provided a ramp-voltage generator initialized by a transition of the signal emitted by the processing chain and inhibited by the next transition, connected to a deflection input of a cathode-ray tude, and wherein the signal delivered by the processing chain is also fed to a second transition detector which delivers an order to a control for firing the gun of the tube.
23. The apparatus as claimed in one of claims 12 and 13, wherein said means for monitoring said accoustic emission comprise means repeating the detection at the end of utilization at a rate compatible with the geometric growth of the acoustic source.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| FR8100776A FR2498320A1 (en) | 1981-01-16 | 1981-01-16 | ACOUSTIC SOURCE DETECTION METHOD AND ACOUSTIC TRANSMISSION CONTROL APPARATUS |
| FR8100776 | 1981-01-16 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1194977A true CA1194977A (en) | 1985-10-08 |
Family
ID=9254232
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000394000A Expired CA1194977A (en) | 1981-01-16 | 1982-01-12 | Process for detecting an acoustic source and an acoustic emission monitoring apparatus |
Country Status (9)
| Country | Link |
|---|---|
| EP (1) | EP0056553B1 (en) |
| JP (1) | JPS57172219A (en) |
| KR (1) | KR830009868A (en) |
| CA (1) | CA1194977A (en) |
| DE (1) | DE3173039D1 (en) |
| ES (1) | ES508766A0 (en) |
| FR (1) | FR2498320A1 (en) |
| YU (1) | YU7382A (en) |
| ZA (1) | ZA82113B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022118060A1 (en) * | 2020-12-04 | 2022-06-09 | 10855561 Canada Inc. | Device, system and method of monitoring a steam trap and detecting a steam trap failure |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH065272B2 (en) * | 1984-03-02 | 1994-01-19 | 株式会社東芝 | Signal source search method |
| FI93496C (en) * | 1987-01-31 | 1995-04-10 | Siemens Ag | Method of locating shocks by means of a monitoring system which uses the propagation of sound in solid bodies, in particular in closed parts of pressure plants in power plants |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2164031A5 (en) * | 1971-12-10 | 1973-07-27 | Commissariat Energie Atomique | Locating device - for source of strain waves eg in stressed pressure vessels |
| FR2262303A1 (en) * | 1974-02-22 | 1975-09-19 | Commissariat Energie Atomique | Position detector for source of strain waves - employs piezoelectric transducer network with counter and computer |
-
1981
- 1981-01-16 FR FR8100776A patent/FR2498320A1/en active Granted
- 1981-12-28 DE DE8181402081T patent/DE3173039D1/en not_active Expired
- 1981-12-28 EP EP81402081A patent/EP0056553B1/en not_active Expired
-
1982
- 1982-01-08 ZA ZA82113A patent/ZA82113B/en unknown
- 1982-01-12 CA CA000394000A patent/CA1194977A/en not_active Expired
- 1982-01-13 YU YU00073/82A patent/YU7382A/en unknown
- 1982-01-14 JP JP57004753A patent/JPS57172219A/en active Pending
- 1982-01-15 ES ES508766A patent/ES508766A0/en active Granted
- 1982-01-16 KR KR1019820000192A patent/KR830009868A/en unknown
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022118060A1 (en) * | 2020-12-04 | 2022-06-09 | 10855561 Canada Inc. | Device, system and method of monitoring a steam trap and detecting a steam trap failure |
Also Published As
| Publication number | Publication date |
|---|---|
| ES8305926A1 (en) | 1983-04-16 |
| JPS57172219A (en) | 1982-10-23 |
| ES508766A0 (en) | 1983-04-16 |
| KR830009868A (en) | 1983-12-23 |
| EP0056553B1 (en) | 1985-11-21 |
| YU7382A (en) | 1985-10-31 |
| FR2498320B1 (en) | 1983-12-09 |
| FR2498320A1 (en) | 1982-07-23 |
| DE3173039D1 (en) | 1986-01-02 |
| EP0056553A1 (en) | 1982-07-28 |
| ZA82113B (en) | 1982-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| FI74152C (en) | Portable ultrasonic inspection device on a tubular object | |
| US4592034A (en) | Acoustic emission source location on plate-like structures using a small array of transducers | |
| JP3491172B2 (en) | Method for telemetry of three-dimensional velocity of fluid such as air or water | |
| AU614212B2 (en) | Acoustic emission leak source location | |
| US5043706A (en) | System and method for detecting bubbles in a flowing fluid | |
| US4910718A (en) | Method and apparatus for acoustic emission monitoring | |
| US5528557A (en) | Acoustic emission source location by reverse ray tracing | |
| US4428236A (en) | Method of acoustic emission testing of steel vessels or pipelines, especially for nuclear reactor installations | |
| KR860001348A (en) | Ultrasonic Scanning Methods and Devices | |
| US4392214A (en) | Sound-monitoring device for an industrial installation | |
| US3744301A (en) | Ultrasonic void fraction detector | |
| CA1194977A (en) | Process for detecting an acoustic source and an acoustic emission monitoring apparatus | |
| US3715914A (en) | On site pressure vessel inspection device | |
| WO1982000718A1 (en) | Method of detecting,preferably continuously,the presence of cracks,the formation of cracks and the growth of cracks in constructions | |
| KR880001290B1 (en) | Method and apparatus for acoustically monitoring on industrial plant | |
| US5010503A (en) | Apparatus for acoustic-emission inspection of articles | |
| CN114199441B (en) | Method, processor and apparatus for determining maximum service stress of buried pipeline | |
| RU2066458C1 (en) | Method for determination of coordinates of electromagnetic field source | |
| KR101809666B1 (en) | Method for predicting defect location by triangulation sensors | |
| US3701990A (en) | Fiber optic display device | |
| Macleod et al. | Acoustic monitoring techniques for structural integrity | |
| FR2349841A1 (en) | Depth and location detection of buried pipes - uses magnetic field generated in region of pipe and measured at two heights above ground | |
| Komarov et al. | Acoustic emission method for testing oil and gas tanks | |
| RU2496117C1 (en) | Method of measuring displacement parameters of probing signal source | |
| RU2011158C1 (en) | Method of determining cross-section area of round objects |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKEX | Expiry |