FR2775787A1 - METHOD AND DEVICE FOR NON-DESTRUCTIVE ULTRASONIC TESTING OF A JOINT WELD BETWEEN TWO PARTS - Google Patents

Abstract

The invention concerns a method which consists in emitting simultaneously, from an ultrasonic wave emission source (1), a first ultrasonic wave beam (14) towards the weld seam (9) between the two parts (7, 8) forming a first angle with the normal line at the surface (7a) of the parts (7, 8) and a second ultrasonic wave beam (15) forming a second angle with the normal line at the surface (7a) of the two parts (7, 8). When there occurs an echo due to a fault (16) of the weld seam (9), the first and second ultrasonic wave beams are successively emitted to determine from which of the first or second beams originates the echo representing the fault (16). The fault (16) characteristics are determined from the beam from which the echo originates. The control device comprises a first housing (1a) containing a first transducer and a second housing (1b) containing a sound transducer connected by flexible links, an electronic box (3) powering the transducers and recovering the ultrasonic signals whereon is placed a switch (5) and a unit (13) generating pulses and processing the ultrasonic signals received by the transducers.

Description

The invention relates to a method of non-destructive testing by ultra-

  sounds of a junction weld between two metal parts.

  In the context of mechanical construction and for example in the context of the construction of nuclear reactors, connections are made by welding between steel parts such as flat sheets or even annular ferrules. These welds can in particular be butt welds of two pieces such as two flat pieces or two ferrules. The junction welds extend over the entire thickness of the parts and in a direction of the parts which will be designated subsequently as the longitudinal direction of the weld. For example, in the case of flat parts assembled end to end, the longitudinal direction of the weld is a straight direction parallel to the edges of the parts between which

welds.

  In the case of ring-shaped ferrules, the longitudinal direction

  along which the weld extends is a circular line parallel to the ex-

  annular hoppers of the parts between which the welding is carried out.

  In all cases, in particular when the parts are thick parts of a nuclear reactor, it is necessary to control the weld

  between the parts, so as to be able to detect the faults that could pre-

  feel this weld. Defects must also be characterized in that

  which concerns their position according to the thickness of the pieces and their dimensions

  to determine if repair is necessary.

  The joint welds of certain nuclear reactor components

  Indeed, there must be no defect or at least no defect.

  can evolve during the operation of the nuclear reactor.

  To check the weld faults between two metal parts, it is known to use a non-destructive test method by

  ultrasound consisting in moving an emission source by at least one beam

  ultrasonic skin on an accessible surface of the parts, in the vicinity of the weld, so that the source of ultrasonic emission directs a beam

  ultrasound towards the weld, in the thickness of the parts.

  We collect and record the echo signals which are due to

welding faults.

  Generally, as a source of ultrasound emission, a

  transmitter-receiver transducer which generates the ultrasonic beam

  our incident and recover the echo signals that can be recorded

  very and analyzed by electronic means associated with the transmitter-receiver transducer. To scan the weld along its entire length and along its entire thickness, it is necessary to move the transducer, in contact with the surface of the parts, in the longitudinal direction and in a

  transverse direction perpendicular to the longitudinal direction, general-

alternatively.

  In order to improve the detection of faults whatever their orientation

  tion, in all areas of the weld, and analyze the defects for

  determine their characteristics more precisely, we use gener-

  ralement, successively, a first transducer which is moved in contact with the surface of the parts so as to emit a beam of ultrasound at a first angle with the normal to the surface of the parts

  and a second transducer which is moved on the surface of the parts, ma-

  to emit, in the direction of welding, a second beam of ultrasound making a second angle with the normal to the surface of the parts on which

the transducer is moved.

  Generally, the first transducer comprises a piezoelectric element.

  electric arranged so as to emit a beam of ultrasound making an angle close to 45 with the normal to the parts and the second transducer, a beam making an angle close to 60 with the normal to the surface of the

rooms.

  The control operations which require a complete sweep of the weld using the first and the second transducers successively can require a long execution time, in particular in the case

  very large rooms. In addition, operations using the first and the second

  cond transducers being distinct and successive, it is very difficult to re-

  manage and compare the records relating to the same defect,

  to specify its characteristics.

  The object of the invention is therefore to propose a method of control not

  destructive by ultrasound of a junction weld between two metal parts

  ques extending in a direction of parts called longitudinal direction

  nal, by moving parts over a surface in the longitudinal direction

  in a transverse direction, substantially perpendicular to the longitudinal direction, of a source of emission of at least one ultrasound beam in the direction of the weld and recording of the echoes due to faults in the weld, this process making it possible to carry out the control of

  quickly with good characterization of faults.

  For this purpose, a first beam is emitted simultaneously from the source of emission of ultrasonic waves in the direction of the weld making

  a first angle with the normal to the surface of the parts and a second

  ultrasonic waves, making a second angle with the normal to the

  surface of the parts and, when an echo due to a defect in the weld is received

  recorded, the first and second beams are successively transmitted to

  finish which of the first and second beams is the incident beam

  having provided an echo representative of a defect and the characteristics are determined

  fault ticks from the echo of the incident beam.

  In order to clearly understand the invention, a description will now be given by way of example, with reference to the attached figures, of a method of

  carrying out the welding control method according to the invention and a dis-

  positive to perform this check.

  Figure 1 is a perspective view of a device for performing the

  non-destructive testing of a weld with two ultrasonic beams followed

with different inclinations.

  Figure 2 is a sectional and elevational view of the connection device.

  hole in service position for checking a weld between two parts

these.

  FIG. 3 is a top view along 3 in FIG. 2.

  FIG. 4 is a diagram showing the amplitude of echoes represented

  sensed by welding defects, depending on the beam path

  inside the rooms.

  In FIG. 1, a control device according to the invention is seen.

  generally signed by mark 1.

  The control device 1 comprises a first housing enclosing it a first ultrasonic transducer and a second housing 1 b enclosing a second ultrasonic transducer, the housings 1a and 1b being interconnected by flexible connection devices such as 2 which may be formed

  for example by rubber elements.

  The lower faces of the first and second housings each have a support sole intended to come into contact with a surface of a part in which it is desired to control a weld. Soles

  of support boxes l a and 1 b are protected by anti-wear pins

  managing the soles, when moving the boxes in contact with the

  part to carry out the welding control.

  The piezoelectric elements of the ultrasonic transducers available

  its in the housing 1a and in the housing 1b are arranged in the vicinity of the lower bearing surface of the housing, so as to emit beams

  ultrasonic inside the metal of the part on which the device is moved

  ultrasonic control positive 1. The piezoelectric elements of the first transducer disposed in the housing 1a and of the second transducer disposed in the housing 1b are inclined differently relative to the underside of the housing intended to come to bear on the workpiece, so as to emit ultrasonic beams with a different incidence compared to the surface of the part. The piezoelectric element of the first transducer disposed in the first housing 1a is inclined so as to emit a beam whose mean direction makes an angle of 45 with the normal to the part on

which rests the housing.

  The piezoelectric element of the transducer arranged in the second box 1b is oriented so as to emit an ultrasonic beam whose average direction makes an angle of 60 with the normal to the surface of the part on which the second box 1b rests. Flexible connections such as

  that 2 between the boxes la and lb allow to apply the sur-

  face of the support sole of the housings 1a and 1b on a part, even if the latter has discontinuities or parts having inclinations

slightly different.

  On the housing 1a, is fixed a housing 3 containing electro-

  feeding the transducers contained in the boxes l a and 1 b and recovering and processing the signals supplied by the transducers

  which are transceiver transducers.

  The housing 3 has, on one of its faces, connection elements

  electrical connections allowing the electronic box 3 to be connected and trans-

  ductors to a supply and processing unit of a type used usually

  tually for ultrasonic testing.

  A control button is also mounted on the electronic unit 3.

  mutation 5 which can be placed in one of three positions 6a, 6b and 6c which

  correspond respectively to the power supply of the first transducer

  placed in the housing la, to supply the two simultaneous transducers

  ment and to the power supply of the second transducer placed in the second box.

tier 1b, respectively.

  It is thus possible, by placing the switch 5 in one of the three positions

  tions 6a, 6b and 6c, perform the emission of an ultrasonic beam at 45 uni-

  only, the emission of a beam at 45 and a beam at 60 simultaneously-

  ment or the emission of a beam at 60 only.

  In Figure 2, there is shown a first flat part 7 and a second flat part 8 which are fixed end to end along one of their edges by a junction weld 9. The weld 9 extends along the entire thickness flat parts 7 and 8 which may have a fairly large thickness, for example of the order of 100 mm. In addition, as can be seen in FIG. 3, the weld 9 extends in a direction 10, called the longitudinal direction, which is parallel to the edges of the plates 7 and 8 which are joined.

  end to end and connected by solder 9.

  To carry out the control of the weld 9, a control device 1 similar to the device represented in FIG. 1 is used. The bearing faces of the housings la and lb of the device 1 are applied against the upper surface of one of the parts , for example the upper face 7a of the part 7 and the device 1 is moved over the surface of the part, so as to

  scan the weld 9 in the longitudinal direction 10 and follow

the thickness of the weld.

  To carry out the scanning along the thickness of the weld, taking into account the inclination of the ultrasonic beam (s) emitted by the transducer of the control device 1, the control device 1 as a whole is moved, in contact with the upper surface 7a of the part 7, in a

  transverse direction substantially perpendicular to the longitudinal direction

nale 10 of the weld 9.

  The scanning in the di-

  longitudinal section by the double arrow 11 in FIG. 3 and the sweep

  in the transverse direction by the double arrow 12 in FIG. 2.

  The electronic unit 3 of the ultrasonic control device 1 is connected to a unit 13 for supplying the transducers and for recovering and

  for processing the ultrasonic signals received back by the transducers

  teurs. Unit 13 is of a usual type used in the case of ultra-

  and in particular comprises means for supplying electrical power to the piezoelectric elements of the transducers and a screen for displaying the

  ultrasonic echoes received by each of the transducers.

  When the piezoelectric elements of the transducers of the device

  control tif 1 are excited by an electric current, the first transducer

  tor placed in the first housing emits an ultrasonic beam whose mean direction 14 makes an angle of 45 with the normal to the surface 7a

  of the part at the level of the housing.

  The second transducer arranged in the second housing 1 b emits an ultrasonic beam whose mean direction 15 makes an angle of 60 with

  normal to the surface 7a of the part 7, at the level of the second housing 1b.

  Of course, it is also possible to carry out the scanning of the weld by placing the control device on the upper surface of the second part 8, so as to direct the ultrasonic beams towards the

solder 9.

  In the case where the weld 9 has a defect 16 at a depth P below the upper surface of the parts, such a defect must be detected during the ultrasonic control and its characteristics must be determined, in

  in particular the depth P of the defect under the surface of the parts and the length

  length L of the defect in the longitudinal direction 10 of the weld (see

figure 3).

  To carry out the control by the method according to the invention, firstly, the switch 5 of the control device is placed in its central position 6b and the transducers are supplied, so as to simultaneously emit a beam of ultrasonic waves directed at 45 and a beam of ultrasonic waves directed at 60, with respect to the normal to the surface of the parts. The transducer is moved on one of the surfaces of the parts, in the vicinity of the weld 9 (for example on the upper surface 7a of the part 7), so as to scan the weld in the longitudinal direction (longitudinal scanning shown by the double arrow 11) and a scan of the weld according to its thickness (scan in the transverse direction 12). During the scanning, the beams of ultrasonic waves of medium direction 14 and 15 carry out a scanning of at least one of the

  weld junction surfaces 9.

  In the event that a fault 16 is present in an area of the support

  lasts, at a certain point in the scan, as shown in FIG. 2, one of the beams (for example the beam of medium direction 15 to 60) is incident on the zone of the weld comprising the defect 16. The signal of echo representative of the fault is then emitted and displayed on the unit screen

  13 of ultrasonic control.

  The scanning is stopped in the position allowing the signal to be obtained

  echo at its maximum amplitude. Then place the successive switch

  in positions 6a and 6c, so as to identify the incident beam

  having given rise to an echo signal representative of the fault.

  In the case shown in FIG. 2, the switching in the posi-

  tion 6c makes it possible to obtain the echo signal corresponding to the incident beam at 60. We can then calculate the depth P of the defect and determine the length L of this defect by longitudinal scanning, the switch being

held in position 6c.

  In order to improve the determination of the fault, the device is moved

  ultrasonic control positive 1 in the direction of transverse scanning, up to position 1 ′ shown in broken lines in FIG. 2, for which the incident beam of ultrasound at 45 provides a representative echo

of the fault 16.

  The characteristic values of the fault supplied from the beam at 45 are verified by calculation. FIG. 4 shows the echo 17 obtained from the beam at and the echo 18 obtained from the incident beam at 45, for the same

default 16.

  Because the length of the path of the ultrasound beam inside the part 7 is shorter in the case of the incident beam at 45, this length being less than 70 mm, the amplitude of the echo at 45 ( 80% of the amplitude of the emitted beam) is greater than the amplitude of the echo at 60 (60% of the amplitude of emission), the path of the ultrasound beams in

  the part 7 then being greater than 90 mm.

  The characteristics of the fault which can be determined from the ultrasonic measurements are in particular the depth P and the length L

  of the defect and the amplitude of the echo.

  After detecting and characterizing a fault, the scanning of the weld is continued using the ultrasonic wave beams simultaneously

  45 and 60 by replacing the switch in the central position 6b.

  For the implementation of the method according to the invention, it is possible to use a device whose housings have a length in the longitudinal direction

  at most equal to 80 mm, a height of 40 mm and a thick-

  in the transverse direction of the order of 28 mm for a width of

  contact of the housing in the transverse direction of the order of 26 mm.

  Ultrasonic transducers have an emission frequency of 2 +

0.2 MHz.

  The propagation angle of the ultrasonic wave beam must be

  finished to within +2. The bigle angle of the transducer, i.e. the angle of the average direction of the beam with the normal to the surface of the transducer

must be zero to within + 2.

  The piezoelectric elements of the transducers are constituted by square or rectangular pads having dimensions close to mm.

  Before implementing the device to implement the program

  transferred according to the invention, the sensitivity of the transducers is checked, which must make it possible to obtain an echo having an amplitude greater than 25% for a defect constituted in the form of a cylindrical hole with a diameter of 2 mm situated at a depth of 150 mm inside the rooms, the gain adjustment by the electronic unit being such that the echo signal has an amplitude of 80%

  for a defect consisting of a cylindrical hole from 2 mm to 25 mm deep

  founder. The transducers and the measurement processing means are such that the signal / noise S / N ratio is greater than or equal to 20 dB, the transducer being excited at a voltage of 100 volts with a gain in reception of 50

dB.

  The method and the device according to the invention therefore make it possible to

  read more quickly the control of a weld by scanning the weld along its length and according to its thickness inside parts

  whose weld joins.

  The method also makes it possible to characterize the defects detected,

precisely.

  The invention is not limited to the embodiment which has been described.

  Thus it is possible to use a control device which can simultaneously carry out the emission of a first and a second beam whose angles with the normal to the surface of the parts are different from 45 and

600,.

  Preferably, a first transducer emitting a pre-

  mier beam having an angle of incidence with the normal to the surface of the parts between 35 and 50 and generally close to 45 and a second

  transducer emitting a second beam with an angle of incidence

  taken between 500 and 80 and generally close to 60.

  The control can be carried out by simply carrying out the detection of the

  fault using simultaneously the two beams of different orientations

  tes, then the identification of the incident beam having given rise to an echo relating to a fault, without seeking a second position of the control device making it possible to locate the fault with the beam having the second orientation. The method according to the invention can be applied not only to the control of welds of flat parts but also to the control of welds

  end to end of annular parts. In this case, the bearing surfaces of the boxes

  third of the piezoelectric transducers and emission elements are adapted to the shape of the cylindrical surface of the parts on which

  moves the ultrasonic testing device.

  The invention applies not only in the case of control by ul-

  welds of nuclear reactor components, but also

  ment in the case of any component of mechanical assemblies comprising

  welds between thick pieces.

Claims (6)

  1 - Method for non-destructive ultrasonic testing of a weld (9)   of junction between two metal parts (7, 8), extending in a direction   tion (10) of the parts, called longitudinal direction, by displacement on a surface (7a) of the parts (7, 8), in the longitudinal direction (10) and in a transverse direction substantially perpendicular to the longitudinal direction (10), d '' a source (1) of emission of at least one beam of ultrasound in the direction of the weld (9) and recording of the echoes (17, 18) due to faults (16) of the weld (9), characterized by the fact that a first beam of ultrasonic waves (14) is emitted simultaneously from the ultrasonic wave emission source (1) in the direction of the weld (9), making a first angle with the normal to the surface (7a) of the parts (7, 8) and a second beam of ultrasonic waves (15), making a second angle with the normal to the surface (7a) of the parts (7, 8) and that, when 'an echo   (17, 18) due to a defect (16) of the weld (9) is recorded, success is emitted   the first and second beams (14, 15) repeatedly to determine the   which of the first and second beams (14, 15) is the incident beam having provided an echo (17, 18) representative of the defect (16) and the   characteristics of the fault, from the echo (17, 18) of the incident beam.   2.- Method according to claim 1, characterized in that after having determined the characteristics of the fault (16) from the echo (17) of the incident beam, the emission source (1) of the beam is moved of ultrasound in the transverse direction, from the first position   for which we obtained the first echo (17) provided by the beam inci-   tooth, up to a second position for which a second echo (18) is obtained provided by one of the second and first ultrasound beams which was not the incident beam in the first position of the emission source and we check the characteristics of the fault (16) obtained from the second echo (18).   3.- Method according to any one of claims 1 and 2, ca-   characterized by the fact that at least one of the following characteristics is determined   defaults (16): depth of the defect below the surface (7a) of the parts (7, 8), length (L) of the defect in the longitudinal direction (10), amplitude of the echo signal (17, 18).   4.- Method according to any one of claims 1 to 3, ca-   characterized by the fact that the first angle of inclination of the first beam (14) is between 35 and 50 and generally close to 45 and that the second angle of inclination of the second beam (15) is between 50 and   and generally around 60.   5.- Device for non-destructive ultrasonic testing of a weld   (9) joining between two metal parts (7, 8) extending in a di-   rection of the parts (7, 8), called longitudinal direction (10), by displacement-   lying on a surface (7a) of the parts (7, 8) in the longitudinal direction   (10) and in a transverse direction perpendicular to the longitudinal direction   line (10) of an emission source (1) of at least one beam of ultrasound in the direction of the weld (9) and recording of the echoes (17, 18) due to faults (16) in the weld (9), characterized by the fact that it comprises: - a first and a second casing (la, 1 b) respectively containing a first and a second ultrasound transducer connected together by flexible connecting elements (2) and comprising bearing surfaces on the surface (7a) of the parts (7, 8), the flexible elements (2) making it possible to vary the relative orientation of the bearing surfaces of the first and second housings (la, lb) , - an electronic unit (3) for supplying the transducers, for the production of ultrasonic beams and for the recovery and processing of the ultrasonic signals received by the transducers, fixed to one of the housings (la, lb), - a switch (5) mounted on the electronic unit (3) comprising   three positions (6a, 6b, 6c) corresponding to the supply of a first trans-   conductor, the simultaneous supply of a first and a second transducer and the supply of the second transducer only, respectively, and - a unit (13) for supplying the ultrasonic transducers and for   processing of ultrasonic signals received by electrically connected transducers   tracing to the electronic unit (3).   6.- Device according to claim 5, characterized in that the first transducer disposed in the first housing (la) of the control device is oriented so as to emit a beam of ultrasound whose average direction makes an angle close to 45 with the normal to the bearing surface of the first housing (la) and the second transducer arranged in the second housing (lb) is arranged so as to emit a beam of ultrasonic waves whose average direction makes an angle close to 60 with the   normal to the support surface of the second housing (lb).