US3251220A

US3251220A - Apparatus for ultrasonic flaw testing

Info

Publication number: US3251220A
Application number: US205721A
Authority: US
Inventors: Ivan L Joy
Original assignee: Chemetron Corp
Current assignee: Chemetron Corp
Priority date: 1962-06-27
Filing date: 1962-06-27
Publication date: 1966-05-17
Anticipated expiration: 1983-05-17
Also published as: GB1055423A; DE1473384A1; ES289335A1; CH450001A; BE633687A; NL293655A

Description

'mmas SR wmzv XR 3.2511220 y 7, 1966 l. L. JOY 3,251,220

APPARATUS FOR ULTRASONIC FLAW TESTING Filed June 27. 1962 12 Sheets-Sheet 1 frzz/eizfor Iran L.

a i s May 17, 1966 1.. JOY

APPARATUS FOR ULTRASONIC FLAW TESTING 12 Sheets-Sheet '2,

Filed June 27, 2

NN k k9 llh ld WQN kQN UR HQN he m UNM MN m fmb'amfar Iva 21/ L LEM LSNLSM 3 Run 1% May 17, 1966 JOY APPARATUS FOR ULTRASONIC FLAW TESTING w mN NNA MSW BS S 5 W w t w Z w w m m kk. k v W \N v allllf n n n a M 17 n n 0/7 v 5 u n n u n n w 9 l u u W n n n u u e u m M3 Of nu M f wl 003 m. 0 l M GMQ- QUNWDA U=I Q N: 1 vi mvi 4 h 3 MS Na S May 17, 1966 JOY APPARATUS FOR ULTRASONIC FLAW TESTING l2 Sheets-Sheet 4 Filed June 27, 1962 UNM ESE um May 17, 1966 l. 1.. JOY

APPARATUS FOR ULTRASONIC FLAW TESTING l2 Sheets-Sheet 5 Filed June 27. 962

May 17, 1966 1.. JOY

APPARATUS FOR ULTRASONIC FLAW TESTING l2 Sheets-Sheet 6 Filed June 27, 1962 I. NMN

QWN

l. 1.. JOY 3,251,220

12 Sheets-Sheet 7 May 17, 1966 APPARATUS FOR ULTRASONIC FLAW TESTING Filed June 27. 1962 Q &m mm a QM F Ems. l-| mwm kkw Sv ll bwu RN a Qg May 17, 1966 JOY APPARATUS FOR ULTRASONIC FLAW TESTING l2 Sheets-Sheet 8 Filed June 27, 1962 1720622202" Ivan L.

5% 9972mm, B Mm v c/gzy'.

y 7, 1966 l. L. JOY 3,251,220

APPARATUS FOR ULTRASONIC FLAW TESTING Filed June 27, 1962 12 Sheets-Sheet 1O all Ill |1\lil 2i 1am I nw MNM. 1 1 1. y, 1 m N 90 \MM n ks. .wwhm m SNM www awn. QM W M I\ H \I\ 1 \NM kw mm M mwm u wwwm Wu. ham QNM.\\

' May 17, 1966 1. JOY 3,251,220

APPARATUS FOR ULTRASONIC FLAW TESTING HM l i l l l I I i Filed June 27, 1962 12 Sheets-Sheet 12 United States Patent 3,251 220 APPARATUS FOR ULTRASONIC FLAW TESTING Ivan L. Joy, Topeka, Kans., assignor to Chemetron Corporation, Chicago, 111., a corporation of Delaware Filed June 27, 1962, Ser. No. 205,721 8 Claims. (Cl. 7367.7)

This invention relates to ultrasonic flaw testing of solid bodies and utilizes pulsed beams of ultrasonic signal energy impinging upon an entrant surface of a solid body at an angle of incidence such that the beams are refracted along so-called fiat angles within the body; that is, angles having a direction that is almost parallel with the direction of the entrant surface of the body, and more particularly the invention is concerned with progressive ultrasonic rail fiaw testing wherein pulsed beams of ultrasonic signal energy are trans-mitted into a rail through a solid or liquid couplant (for example, water) at such angle of incidence as to trav l at fiat angles within the rail; that is, angles on the order of 75 to 89.

In ultrasonic fiaw testing of solid bodies, it has occured that particular angles of incidence are required for detecting various different types of defects due to their orientation, location, and/or configuration. Straight crystals, that is crystals that direct a beam of ultrasound along a line that is perependicular to the entrant surface of the body being tested so that the beam continues along its original direction after entering the body, have been used for detecting certain types of flaws while angle crystals operating at various angles of incidence or variable over a range of incident angles have been used for detecting other types of defects.

Ultrasonic beams that impinge at angles of incidence resulting in the production within the body of shear waves traveling at fiat angles also produce waves traveling along the entrant surface of the body. These surface waves are reflected from any minor surface discontinuity, such as scratches, surface .nicks, and the like, to produce false indications that cannot be distinguished from signal reflections of the shear wave beams from defects within the rail.

In progressive rail flaw testing with Ultrasonics, it has been proven through a number of years of experience that straight crystals are excellent for detecting rail defects such as bolt hole breaks, head web separations, and vertical and horizontal split heads but they are not able reliably to detect small size defects in the rail head or defects located under burns, shelling, or other surface discontinuities, or defects located at side edge regions of the rail head, or transverse and compound fissures. Compound fissures are of a type that grow in two directions.

In one ultrasonic rail testing arrangement that is presently being used crystals are arranged to transmit beams into the running surface of a rail at angles of incidence such that the beams are refracted along a 70 line within the rail head.

Presumably this 70 arrangement for rail flaw testing is used because at this angle substantially no surface waves are produced. Where the incident beam is transmitted through a water column and through a rubber diaphragm, the angle of incidence may be about 24. Experiments show that changing the angle of incidence to about 25 to produce an angle of 74 within the rail produces a significant amount of surface wave energy and further increases in angle result in proportionately greater quantities of surface wave energy. While the 70 angle arrangement avoids much of the surface wave problem, it does so at the expense of reflected signal strength caused by operation at angles at which scattering effects from the types of flaws to be detected result in reflected signals that are too weak. Reduced scattering effects result due either to the orientation of the flaws or to the size of the flaws or both. It is believed that residual surface wave effects are present, even at this angle of and interfere with the detection of the desired shear wave signal reflections. Moreover, spurious signals or hash arising from the phase transition at the interface are believed to aggravate the situation.

In progressive ultrasonic flaw testing, there are nu merous mechanical problems associated with the movement of the crystals in a prescribed orientation along the length of the rail. In one form disclosed herein, both angled and straight crystals are coupled to the rail through a column of coupling liquid that accommodates all necessary crystal positioning movement. In another form, the crystals press directly upon a rail contacting diaphragm that is carried in a stabilized frame that glides over rail irregularities in a gradual action that allows the diaphragm to conform immediately and maintain coupling. Flat angles are created within the rail by firing a multi-crystal array in predetermined timed sequence to create a merged or culminated beam acting at the desired angle within the rail. In still another form, angle crystals are mounted on plastic wedges that engage directly against a rail contacting diaphragm.

The principal object of the present invention is the provision of an improved ultrasonic flaw testing arrangement for detecting defects at shallow depths, defects of small size, defects located underneath surface irregularities and defects so oriented and shaped as to be incapable of producing detectable signal reflections from beams derived from straight crystals.

Another object of the present invention is the provision of an ultrasonic flaw testing system particularly useful in the progressive testing of rail for detecting transverse and compound defects, shallow depth defects particularly where located under burns, shelling or other surface irregularities, defects of small size and defects located at side edge regions of the rail head.

Still another object of the present invention is to provide an ultrasonic flaw testing arrangement utilizing an ultrasonic wave emitter oriented at an angle of about 30 to an entrant surface of a test body and coupled to the body through a liquid path to produce shear waves at angles within the body on the order of 88 or 89".

A further object of the invention is to provide an arrangement for progressive ultrasonic rail flaw testing and incorporating both straight and angled ultrasonic wave emitters driven by separate ultrasonic machines that are triggered in a predetermined cyclical sequence and that are connected to provide an integrated pictorial display the crystal to maintain it in proper working relation to the body under test.

Another object of the invention is the provision of an aligned multiple section straight crystal array triggered by a multiple stage transistorized delay line to create time delayed test signals of similar strength and appro- In the accompanying drawings forming a part of this specification and in which like numerals are employed to designate like parts throughout the same:

FIG. 1 is a diagrammatic illustration of a simplified ultrasonic flaw testing system utilizing a unique angle crystal arrangement;

FIG. 2 is a generalized circuit diagram of a progressive rail flaw testing ultrasonic testing system wherein multiple crystal arrays, each comprising vertical and angle crystals, are provided for both rails and are connected to a common memory type C.R.O. indicator tube;

FIG. 2A is a perspective miniature view illustrating an ultrasonic detector car and carriage assembly for use with the circuit equipment arrangement of FIG. 2;

FIG. 2B is a timing sequence chart illustrating a typical cycle of operation of the ultrasonic equipment of FIG. 2;

FIG. 3 is a side view partially in section illustrating the crystal array utilized for each rail;

FIG. 4 is a detailed block diagram of the circuitry utilized in the ultrasonic equipment shown in FIG. 2;

FIG. 5 is an enlarged perspective view of the carriage and coupling bag arrangement utilized in one form of this invention;

FIG. 6 is a fragmentary transverse cross-sectional view through a coupling bag for the center or vertical crystal and is taken as indicated on the line 66 of FIG. 5;

FIG. 7 is a side sectional view through an alternative form of ultrasonic coupling unit and employs a composite trough employing continuously moving liquid flow streams;

FIG. 8 is a plan view of the trough of FIG. 7;

FIG. 9 is a generalized circuit diagram like that shown in FIG. 2 and illustrating an alternative embodiment of the invention;

FIG. 10 is a diagrammatic side elevational view of the carriage and mounting details for the embodiment of FIG. 9;

FIG. 11 is a detailed block diagram of one of the ultrasonic circuits utilized for controlling a multi-section composite angle crystal such as is employed in the embodiment of FIG. 9;

FIG. 12 is a diagram of a delay circuit and its connections to a multi-section sender type angle crystal;

FIG. 13 is a diagram of a delay circuit and its connections to a multi-section receiver type angle crystal;

FIG. 14 is a generalized circuit diagram like that of FIGS. 2 and 9 and illustrating an alternative embodiment of the invention;

FIG. 15 is a lengthwise side sectional view illustrating an alternative guide shoe and trough arrangement equipped with another type of ultrasonic crystal coupling array;

FIG. 16 is a plan view of the device of FIG. 15;

FIGS. 17 and 18 are transverse sectional views taken respectively on the lines 17-17 and 18-18 of FIG. 16;

FIG. 19 is a diagrammatic illustration of the pattern of an ultrasonic beam projected lengthwise along a track rail;

FIG. 20 is a diagram of a B-scan presentation developed by angle crystal units employed in the FIG. 14 arrangement;

FIG. 21 is a plan view of a track rail and illustrates the lateral fanning of an ultrasonic beam projected lengthwise through the head of the rail;

FIG. 22 is a section through a butt weld at adjoining rail ends as is also depicted in FIG. 21;

FIG. 23 is a transverse sectional view of an alternative form of angle crystal coupling unit for use in place of the

elements

316, 316', 317 and 317 of FIG. 14;

FIG. 24 is a lengthwise sectional view through the coupling unit of FIG. 23; and

FIG. 25 is a plan View of the unit of FIG. 23.

Description 0 FIG. 1 embodiment For purposes of disclosure, the invention is illustrated and its advantages discussed with particular reference to progressive ultrasonic rail flaw testing, but it will become apparent that certain features offer important advantages in other ultrasonic flaw testing fields.

Referring now to the drawings, in FIG. 1 an ultrasonic wave emitter is illustrated at E in the form of an electromechanical transducer, such as a quartz crystal, and it is arranged to direct beams of ultrasonic energy through a couplant C (usually water through other liquids and solids, such as plastic wedges and plastic diaphragms, are also contemplated) so that the beams impinge upon an entrant surface of a body R (represented here as a track rail), with the angle of incidence I of the impinging beams being selected at a value to produce within the rail a refracted shear wave beam traveling at a so-called fiat angle P which as used herein is defined as an angle that extends in a direction that is close to being parallel to the entrant surface. For the specific case of the track rail illustrated herein, fiat angles of practical utility lie in a range of 75 to 85. Due to fanning of the beam in the rail, the center line of the beam can only be specified approximately at beam angles greater than 85". Actual operation has shown that the region of the end of the rail will produce a detectable echo at a range as great as 6 feet. Due to the width of the crystal and due to the divergent characteristic of ultrasonic beams any given beam produced in the illustrated arrangement, assuming an operating frequency of 2.5 megacycles has a fanning effect on the order of 10 as measured from the entrance point B at the interface and, therefore, a beam at a flat angle of 75 has an effective span ranging from 70 to Similarly, a beam at a flat angle of 80 has an effective span from 75 to and a beam at a fiat angle of 85 has an effective span of 80 to almost In each case, the maximum signal strength occurs at the center of the span and the signal strength at the edge regions is on the order of 20 to 30 percent of maximum. In ultrasonic rail fiaw testing, it has been found that fiat angles in the range from 75 to 85 are effective for producing detectable signal reflections from transverse and compound defects, from shallow defects even if located under burns or other irregularities, and also from defects at the side edge regions of the rail.

A transverse fissure is depicted at TF in FIG. 1 and in general transverse fissures are essentially vertical and relatively straight and smooth.

When an ultrasonic beam strikes such a relatively smooth fiat surface at a given angle, most of the energy is reflected away at an opposite equal angle and the amount of energy reflected along the angle of incidence is due only to scattering effects and becomes markedly weaker as the angle of incidence progressively departs from true perpendicularity. For this reason, the strength of shear wave beams produced by crystals operating at an angle of incidence such as is illustrated here are significantly weaker within the actual rail. Correspondingly, the shear wave beam within the rail and striking a flaw therein, is greatly weakened if it is not oriented substantially normal to the line of the flaw. For this reason, a beam operating at an angle of 70 suffers enormous losses whereas a beam operating at a Hat angle has markedly improved characteristics as respects this situation. An additional advantage for the use of a fiat angle acting substantially perpendicular to the line of a transverse fissure, is that the beam persists on the target for 8" or 9" of travel to produce a series of reflected pulses whereas a beam at a 70 angle is on target on a transverse fissure for 1" or 2" at most.

Moreover, the higher angles of 85 to 88 permit operation at even longer ranges from the defect and this affords added time on target and permits more com plete separation of the desired defect indications both from the hash generated at the liquid-metal interface by reason of the angle entry and from the surface wave reflections corresponding to surface regions near the crystal.

A defect, such as a ,4," hole, A. deep in the rail, is also shown at H in FIG. 1. Even in this situation, a beam at a flat angle can be on target for at least 3" or 4" and the hole will be detected even where it is under a surface defect such as a burn or shelling. It should be apparent that at a 70 angle, the beam can only be on target for less than one inch of travel. At a travel speed of mph. (180 inches per second) and a repetition rate of 500 per second, only one test signal is likely to find the hole and in many instances none will. Thus operation at 70 is not reliable at a speed of 10 mph. and, more importantly, it places an absolute limit on ultimate travel speed.

In a test arrangement wherein ultrasound is coupled into a steel body such as a rail through a liquid path and through a diaphragm of rubber or plastic, shear waves at an angle of 70 in the rail are produced by a crystal oriented at an angle of 24 to the entrant interface. These values have been determined empirically at the complexity of the physical relationships does not allow of a simlpe application of Snells law. Angles of more than 70 in the metal have not heretofore been employed successfully because at such increased angles a surface wave is created in addition to the shear wave that travels internally of the body. In signal reflection testing, the surface wave produces signal reflections that are indistinguishable from reflections returning from defects within the rail.

In order to produce flat angles within the rail, it is necessary to operate the crystal at an angle in a range of substantially 25 to 30(+). This range is apphcable where ultrasound is transmitted through a water couplant to enter a track rail. In this context, a sharp attenuation occurs at incident angles of slightly greater than 30 Slight variations at this upper limit can occur where a diaphragm of Teflon or rubber or other suitable material is interposed between the couplant and the rail. The range for the incident beam angle may vary appreciably in applications where the body being tested is of different characteristic and where the couplant 1s a liquid other than water or is a solid Wedge and therefore the common determinant governing all of the possible situations is best defined by the range of angles produced within the test body. This range is substantially from 75 to 85(+) for the center of the fanning shear wave beams developed within the rail In accordance with the present invention, the use of flat angles is made possible by providing for dampening the accompanying surface waves while concurrently gating out reflections returning from surface defects within about 2" of the point of entry of the ultrasound into the test body. Surface waves that are reflected from surface defects as much as 3, 4 or 5 inches or more away are more easily eliminated by damping techniques and the use of flat angles develops a range sufficient to permit gating out of as much as 2" of initial travel without appreciably sacrificing target time. Where high sensitivity is employed, 6" may be damped and gated out if desired.

In FIG. 1 the angle crystal is shown embodied in an ultrasonic flaw testing arrangement employing an ultrasonic machine UM which may be of the general type shown in Patent No. 2,949,028. The crystal E is mounted in an elongated coupling shoe which for purposes of illustrative disclosure is shown as having water entry and exit openings 16 and 17, respectively, connected in a water supply and reclaiming circuit to produce a continuous stream of water flowing towards the rail for coupling ultrasound thereinto. The lower wall 15L of the shoe constitutes a rail contact plate serving to dampen surface waves generated at the interface of the water and rail by virtue of the incident angle. This wall may be suitably extended in the forward beam direction as at 15E to increase its dampening effect.

The wall 15L is shown with a central opening so that the coupling water directly contacts the rail. The wall may be impervious so that ultrasound is transmitted through it and through a film of coupling liquid to enter the rail. Any desired plastic, metal or rubber-like material may be employed for the wall 15L. Moreover, any of the coupling arrangements shown in Patent Nos. 2,- 992,553, 3,028,751, and 3,028,753, and in my pending applications Serial Nos. 715,002 and 65,103, may also be.

employed with the angle test concept of this invention.

The ultrasonic machine UM includes a test signal generator 32 and continuously running high frequency oscillater 33 having their outputs jointly connected to the crystal E. Reflected signals detected by the crystal are supplied to a gate 34 through an IF. amplifier 35, a detector 36, and a video amplifier 37. The test signal generator 32 is periodically triggered from a rate generator RG which also controls the gate 34 through a suitable dclay circuit 38. The gated output is fed to an indicator T which may be :1 CR0. or a memory tube operated in timed relation by a control line from the rate generator.

The delay 38 may be set to enable the gate 34 after an interval of 50 microseconds to allow for a travel path through the water couplant and about 2" of the rail and the gate may be set to receive reflections from defects in a range of from 2" to 10" or more. The contact plate of the coupling shoe is effective to dampen completely any surface wave reflections from surface defects 2" or more away from the point of entry and therefore the output from the gate 34 is free from all hash at the interface and from all surface wave reflections. It is preferred,

' however, to operate at a sufficiently high sensitivity as to cause traces or shadings of has to appear on the memory tube viewing screen. This condition of operation yields easily discernable defect indications while also indicating that truly effective coupling is being maintained. The damped range is dependent upon the nature of the damping medium and the angle of entry but for typical rail testing applications, a 1 interval may be considered as the minimum permissible. Where only a liquid film on the rail is to serve as the damping medium, under some circumstances the minimum range may have to be as much.

as 3 or 4 inches.

Description of FIG. 2 embodiment While the present invention may be embodied in a number of different physical arrangements, it is disclosed in FIGS. 2, 2A and 3 in a unique embodiment of a progressive ultrasonic rail flaw testing system for inspecting and displaying a pictorial representation of intelligence from a complete lengthwise section through the rail.

In the arrangement illustrated herein for purposes of disclosure, a pair of track rails R and R are shown supporting a suitable rail vehicle which is indicated generally at 25 in FIG. 2A. The rail vehicle 25 is equipped with ultrasonic test equipment, shown schematically and in block diagram (see FIG. 4) and the vehicle may be provided with suitable suspension mechanism for mounting coupling troughs 27 and 27' in suitable position on the rails. Additional features of the trough construction and suspension are illustrated in detail in FIGS. 3', 5 and 6, and each is arranged to receive an array of i3 crystals.

Each crystal array is identical and for convenience of disclosure, the array for rail R is described and the corresponding array for rail R is given the same reference numbers followed by a prime. The crystal array includes a straight or vertical crystal V at its center flanked by oppositely inclined angle crystals 208 which, in turn, are flanked by oppositely inclined lengthwise spaced angle crystals 20R and these, in turn, are flanked by pairs of oppositely inclined side-by-side angle crystals 30S and finally, these latter are flanked by pairs of oppositely inclined sideby-side angle crystals 30R. The vertical crystal V is connected as both a sender and a receiver, while the crystals 208 are connected together and operate only as sending crystals with the crystals 20R also being connected together and operating only as receiver crystals for the crystals 205. Finally, the four crystals 308 are interconnected electrically and operate as senders while the four crystals 30R are interconnected electrically and operate as receivers for the crystals 308. In the case of each set of angle crystals, the receiver crystal is to detect only reverse path signal reflections; thus the crystal 20R at the left in FIGS. 2 and 3 receives signals from the crystal 203 that is adjacent to it and not from the crystal 205 that is at the right in FIGS. 2 and 3.

The ultarsonic gear as seen in FIGS. 2 and 4 is connected to both arrays of crystals to trigger the crystals in a predetermined timed sequence and consists of a set of six ultrasonic machines U1 to U6 serially interconnected and driven by a single rate generator RG to operate in full sequence from U1 through U6 each time the rate generator supplies a trigger pulse. Each ultrasonic machine may be like that of FIG. 1 and corresponding components bear identical reference characters. Each ultrasonic machine, except the first, also has an input delay circuit 31 that is adjustable to adapt the firing times of the machines to the physical conditions of the test. Thus the rate generator RG will trigger ultrasonic machine U1 by direct connection and triggering of machines U2 to U6 is successively accomplished over series connection lines 31C extending in sequence between the respective input delay circuits.

Ultrasonic machine U1 is connected to the vertical sender and receiver crystal V for rail R and ultrasonic machine U2 is connected to the vertical sender and receiver crystal V of rail R. Correspondingly, ultrasonic machine U3 is connected to the four sender crystals 30S and by a separate connection receives reflected intelligence from the four receiver crystals 30R. Ultrasonic machine U4 is connected to the set of four crystals 30S and receives reflected intelligence from the set of four crystals 30R.

Finally, ultrasonic machine U5 is connected to trigger the set of two sending crystals S and receives reflected intelligence from the set of two receiver crystals 20R, while ultrasonic machine U6 is connected to trigger the set of two crystals 20S and receives reflected intelligence from the two receiver crystals 20R.

An indicating device such as a memory type display tube T as shown in FIG. 2 is connected to receive gated output from parallel connected output lines 39 from each of the ultrasonic machines U1 to U6. Each machine supplies output through its individual gate circuit 34 suitably delayed and ranged in accordance with the water path and metal path conditions for the particular beams produced by its crystals. The single memory tube T is connected to present a two-dimensional pictorial display of all reflected signal intelligence from both rails and for this purpose the output from all the ultrasonic machines is applied to the control grid for the writing gun 40 of the memory tube. As represented in FIG. 2 the horizontal deflection facilities 41 are supplied with a scanning signal from a composite erase pulse generator and scanning signal generator 42 which also supplies erase pulses to the storage mesh 43. The vertical deflection plate facilities 44 are supplied with saw tooth sweep signals from a sweep generator 45 which is triggered separately by input from each of ultrasonic machines U1, U3 and US which are shown connected to the sweep generator through separate delay circuits 46.

Operation of FIG. 2

The basic sequence of operation is described with refence to the schematic diagram of FIG. 2 and the timing chart of FIG. 2B.

The rate generator RG supplies a trigger pulse to ultrasonic machine U1 at time, T=0, and substantially simultaneously, the vertical crystal V is pulsed to direct an ultrasonic test signal in a vertical water path leading towards the rail R. A connection from the input to ultrasonic machine U1 through the first delay circuit 46 initiates flyback of the sweep generator 45 which undergoes a rapid retrace to then generate a first sweep signal trace 50 (see FIG. 2B) beginning at time T=50 microseconds. Measuring from actuation of machine U1 at time T=O, a delay of approximately 90 microseconds corresponding to the round trip travel time of the signal through the water path, elapses before signals'reflected from the rail head are received. Thus the trace has progressed for 40 microseconds before signal reflections come in.

As the test signal from crystal V enters the rail R, it produces separate reflections from the head or entrant surface, from internal defects and/or from the base to produce an A-scan signal reflection pattern that is fed to the control grid of the writing gun 40 for the memory tube with all earlier and later signal reflections being removed by the gate action. The sweep generator 45 applies the vertical sweep signal to the memory tube in predetermined time relation with the signal reflection patterns from the gate of machine U1 and as this procedure is repeated in subsequent cycles, lines H and B are gradually painted in a horizontal direction on the display screen of the memory tube at a rate determined by the scanning signals from the scanning signal generator 42. It will be appreciated that the scanning signal generator 42 operates at a low repetition rate and at a slow scan speed which may be integrated directly with the travel of the transducers along the rail (in which case the scale of the representations on the screen would be constant), or which may be broadly chosen to correlate generally with the expected travel speed of the transducers along the rail, in which case the scale would be expanded during unusually slow travel or contracted during unusually high speed travel of the transducers. The sweep circuit 45 operates at a high scan speed many times faster than that of the scanning signal generator 42 and the sweep 45 operates at a repetition rate that is also very high so that several cycles of the sweep 45 can be generated with little, if any, apparent scanning motion being produced from the slow scanning signal generator 42. The basic repetition rate of the sweep generator 45 is determined by the rate generator RG, but actually the sweep 45 operates at two or three times the repetition rate of the rate generator RG. This occurs because the delay circuits 46 which are connected to certain of the ultrasonic machines are each effective to reset the sweep 45 during a single overall actuating cycle of the rate generator RG. Thus, in FIGURE 2, where three delay circuits 46 appear, the sweep 45 is reset three times during an overall actuating cycle generated by'the rate generator RG. Each time the beam traverses the screen under the control of the sweep 45, the concurrently existing signal reflection patterns cause it to produce dots of illumination. The fact that this sweep is repeated many times during even slight variations in the beam position along the other coordinate direction (due to the slow speed of scanning generator 42) causes the individual illumination dots to merge into individual continuous traces. This general display technique is described and illustrated more completely in Joy Patent No. 3,156,111. Characteristic marks corresponding to bolt hole or internal defect reflections are also painted on the tube T intermediately of the lines B and H. The A-scan pattern for producing the pictorial representation of the rail R requires from about 60 to microseconds during each cycle.

At time T=450 microseconds ultrasonic machine U2 is fired and again allowing for about microseconds of water path travel time the A-scan signal rellection pattern from the vertical crystal V over rail R comes in from about 540 to 620 microseconds and upon successive cycles of operation the lines for the head and base of the rail are painted in as indicated at H and B in FIGS. 2 and 2A. The sweep generator continues to run until time T==625 microseconds when it is reset by a signal from the delay circuit 46 that is connected to the input to ultrasonic machine U3 and then undergoes a rapid flyback to initiate a second sweepsignal trace at time T =650 microseconds.

Ultrasonic machine U3 is timed, through its internal delay circuit 31, to fire the set of four crystals 308' at T =750 microseconds and allowing for 50 microseconds delay for gating out the relatively short water path and about 2" of travel in the metal rail. The signal reflection pattern comes in from T=800 microseconds to T=1000 microseconds. This allows for signal reflections in a range of from 2 to 8" from the point of entry and they are separately displayed beneath the line B on the viewing screen.

At time T=1215 microseconds, ultrasonic machine U4 is triggered to fire the set of four crystals 30S and again allowing for about 50 microseconds for gating out the water path and 2" of rail travel, the signal reflection pattern comes in from T 1265 microseconds to T=1465 microseconds.

At time T=l475 microseconds, the sweep generator is reset by a signal from the delay circuit 46 of ultrasonic machine U4 and it undergoes a second rapid flyback and, at time T=1500 microseconds, a third sweep signal trace is initiated.

Ultrasonic machine U5 is timed through its internal delay circuit 31 to fire at time T=l810 microseconds and allowing a delay of 50 microseconds to gate out travel through the water path and about 2" of metal, signal reflection patterns come in from T=1860 microseconds to T=1960 microseconds, this range giving about 3" of travel through the metal.

Finally, ultrasonic machine U6 is fired through its internal delay circuit at time T=2275 microseconds and again allowing 50 microseconds delay the signal reflection pattern comes in from time T :2325 microseconds to T=2425 microseconds. The rate generator then fires for the second time at T =2450 microseconds and simultaneously fires ultrasonic machine U1 and another full three sweep cycle is originated. Thus the overall time of a full three sweep cycle runs a total of 2450 microseconds and on this basis the rate generator should have a repetition of approximately 400 per second.

It is contemplated that the rate generator should be operated at least at a repetition rate of 400 per second and preferably higher. In the described arrangement the rate is limited by virtue of providing physically separated areas of the display screen for presenting the intelligence from the several different types of crystals associated with both rails. This technique offers great benefits from the standpoint of facilitating interpretation of the recorded intelligence.

The repetition rate could also be increased by use of sweep signals of complex configuration but in general the cost and technical considerations would not presently justify this.

The repetition rate may readily be doubled either by presenting the intelligence from each rail on separate memory tubes or by utilizing two writing guns in a single memory tube and with each such arrangement the ultrasonic machines for each rail are arranged in a series of three. Other rearrangements for increasing the repetition rate are contemplated wherein the intelligence from the straight crystals V and V' and from the crystals S and 205 are presented on one memory tube and the intelligence from the crystals S and 305 are presented on a separate memory tube in which case the ultrasonic machines for the two memory tubes are correspondingly regrouped.

In the arrangement illustrated in FIG. 2, each of the defects are painted in easily discernible and well known characteristic patterns. I

The

crystals

20S and 205 are oriented at an angle of incidence of about 20 and create shear waves traveling in the rail at an angle of about or degrees and these crystals are intended to detect defects in the lower portion of the head and in about the top half of the web of the rail. the display screen of the memory tube and the defect indications occur in readily distinguishable patterns.

Finally and most importantly, the crystals 30S and 305' are oriented at an angle of incidence on the order of 30 though as mentioned previously this might range from 25 up to about 30. For the 30 angle of incidence a shear wave traveling at a flat angle on the order of or 86 degrees and spanning a range from about 82 to 89 degrees is created to travel within the head of the rail for detecting shallow defects and defects of relatively small size such as Vs" holes. The 30 crystals are effective also for detecting defects located underneath surface irregularities such as burns and for detecting defects which, due to their orientation and shape, are incapable of producing detectable signal reflections from the other crystals. The problems created by the surface waves. yvhich are necessarily generated by the 30 crystals are solved by gating out reflections from defects within a range of two inches of the point of entry and by providing for damping out the surface waves themselves.

In the trough arrangement diagrammatically illustrated in FIG. 3 and shown in detail in FIGS. 5 and 6, a lower wall in the form of a thin sheet 55 of rubber or plastic rides in contact with the rail surface and acts to dampen the surface waves. Since two inches of travel range are gated out, surface waves reflecting from closely adjacent surface flaws are of no concern and in view of the difficulty of perfectly dampening out such reflections this gating arrangement is important to practical opera tion. The longer range surface reflections are etficiently damped by the diaphragm sheet 55.

It is important to note that tests have also shown that the presence of a solid body or a body of water on the top of the rail is also effective as a dampening means and, therefore, the 30 angle crystal is workable even where the diaphragm sheet is omitted. However, the diaphragm sheet is preferred as it takes up the principal mechanical wear and is easily replaced. Tests have also shown that the presence of a film or layer of water on the rail as distinguished from a body or column of water is also effective for dampening the surface wave reflections though in some instances the gating range I would have to be extended from 2" to as much as 4" to achieve proper dampening. In FIG. 2A facilities 51 are shown at the forward end of the car body for-applying a prewetting film of water as this also serves to provide high efficiency of coupling.

It may also be noted that the crystals 30S and 305' are arranged in pairs in side-by-side relation crosswise of the rail head and are located so that the outer edge of the crystal is adjacent or aligned with the corresponding side edge of the rail. It has been found that in operating a crystal for producing flat angled shear wavcs within the rail, when the crystal is located towards the side edge of the rail, a gain of sensitivity of as much as 20 is obtained as compared to the sensitivity of a crystal located at the center of the rail. In practical effect, therefore, it is possible to get an echo from a rail end at a distance of as great as 6 feet where the crystal is located at the side edge whereas where a crystal is located over the center of the rail the maximum range is on the order of 4 /2 feet. This arrangement is particularly useful for Here again, separate presentations are made on detecting compound defects under shelly and it is believed that the improved performance may be explained on the basis that the fiat angled shear wave beams within the rail head prefer to follow the side of the rail and when at that location are not attenuated. A shear wave generated from the center point of the rail tends to fan out and becomes dissipated with a consequent loss in sensitivity.

The

receiver crystals

30R and 30R preferably are oriented at the sameangle of incidence as the sender crystals 30S and 305'. However, the receiver crystals may operate at other angles as they are in no way as critical as are the sender crystals. Additional advantages accrue from the illustrated arrangement of separate sender and receiver crystals. Where the receiver crystal is spaced from the sender in a direction generally opposite of the incident beams travel path the receiver crystal is more effectively isolated from surface waves and from various other spurious effects which can arise in a progressive testing system using angle crystals and water couplings and the like.

Description of carriage of FIGS. and 6 In FIG. 2A the car is shown with a

separate carriage

100 and 100 for coupling to each rail. These carriages are located in trailing relation to the car and are mounted to the car through a common suspension 101 constructed in accordance with the principles illustrated in FIGS. 1 and 2 of my Patent No. 3,028,751, having an issue date of April 10, 1962. In general, the present suspension 101 includes support arms 102 for connection to the rear axle of the car through rubber joints 102R that accommodate both vertical and lateral swinging movement of these arms.

Each carriage comprises fore and aft spaced rail contacting

guide shoes

103 and 104, respectively, rigidly interconnected by a trough frame 105 (FIG. 5).

Front and rear

telescoping bar units

106 and 107 are suspended in crosswise relation from the support arms and are connected to the fore and aft rail contacting

guide shoes

103 and 104, respectively, of the carriages to establish a reference plane relative to the rail heads.

Each telescoping bar unit is spring biased towards an elongated relation and is fitted with a motor driven cable 108 to cause each bar unit to contract progressively during elevation of the carriages by these cables and to expand progressively during lowering of the carriages to their rail engaging position.

To provide positioning adjustment for the straight crystals each carriage is equipped with a set of lateral control cables 109 operable for shifting the trough frame 105 laterally relative to the guide shoes 103 and 104 and is equipped with a vertical tipping control cable 110 connected only to control the straight crystal V or V' for adjusting its vertical lateral angle relative to the reference plane established by the guide shoes 103 and 104. The principles of crystal positioning are described in detail in my aforesaid Patent No. 3,028,751.

The frame 105 which provides the coupling trough for the crystals of each rail comprises upper and lower elongated loop shaped

clam p bands

111 and 112, respectively, (see FIG. 6) that grip the marginal edge of a generally rectangular diaphragm 55 that is preferably of rubber or rubber-like material through plastic sheeting such as Teflon may also be employed. The lower clamp band 112 establishes a rigid connection to each shoe through the positioning mechanism 113 located thereon and the upper clamp band 111 is mounted to the lower band through releasable fasteners 114 to facilitate replacement of the diaphragm.

The diaphragm 55 holds a slight pool of coupling liquid and the various crystals are mounted in a series of coupling bag units spaced along the trough in an array like that shown in FIG. 2. The central coupling bag unit 115 holds the single straight crystal V that acts as a sender and receiver. Each of the adjacent coupling bag units 116 hold a set of 20 individual sender and receiver crystals and finally each end coupling bag unit 117 holds a set comprised of two 30 sending crystals and two 30 receiving crystals.

The elongated frame and diaphragm assembly allows the diaphragm 55 to travel smoothly over rail ends and at other rails irregularities it permits the diaphragm to conform locally as required. Moreover, the coupling bags are carried from the frame in an arrangement that avoids the application to the bags of abrupt lifting or tipping forces such as would interrupt the coupling established through the bag units.

In the embodiment of FIG. 5, each crystal array and coupling trough arrangement is identical. In FIG. 6 there is shown the single straight crystal V that is located in the central coupling bag unit 115. Each coupling bag unit comprises a rubber bag 120 clamped tightly about a Bakelite crystal holder block 121. Each bag 120 is filled with coupling liquid that is maintained under a two foot pressure head from a distributor block 122 that is fed through a flexible conduit 123 leading from a reservoir bottle 124 carried on the car.

The effect of head pressure acting within the coupling bag is to bulge the bag and since the bag mounting is fixed in relation to the frame, this biases the bag surface against the diaphragm to conform the diaphragm to the rail surface. The bag cross section is not substantially greater than the crystal area in order to minimize the U volume of liquid that is displaced upon deformation of the bag. Such deformation is caused by the head pressure whenever the bag encounters sharp variations in surface contour. Due to the travel speeds involved in progressive rail testing the bag must deform instantaneously to conform the diaphragm to all such changes. This requires that the volume displacement of liquid be minimized if the relatively low head pressure is to be effective. A head of 1 p.s.i. can displace 1 cubic inch of water in 0.031 second and by maintaining a narrow bag cross section the normal bag deformation need not require displacement of more than 1 cubic inch in adapting to the normal surface contour changes.

Since adjustable positioning is not required for the angle crystals, their coupling bags are mounted rigidly to the frame. A bridge structure 125 spans the frame adjacent each angle crystal coupling bag and it is fixedly secured to the crystal holder block. The frame is equipped with cross-straps 126 in flanking relation to the coupling bag 115 for the straight crystal. The cross-straps 126 carry bearings 126B that pivotally mount a swing frame 127 that carries the coupling bag 115. The swing frame 127 may be tiltably adjusted by the vertical tipping control cable 110.

Description of trough of FIGS. 7 and 8 A composite coupling liquid trough arrangement is shown in FIGS. 7 and 8 wherein an elongated box shaped hood structure 130 mounts a set of crystal holders 131 arranged in an array like that shown in FIG. 2. The hood is open at the bottom and is provided with a diaphragm sheet 132 and the hood has fore and

aft connection arms

133 and 134, respectively, for attachment to rail contacting

guide shoes

103 and 104 in order to control suspension and lateral positioning of the hood and its crystals. An elongated holder tube 135 is provided for the straight crystal at the center of the hood 130 and is carried in a pivoted frame 136 mounted for tipping movement to control the vertical lateral angle. A continuously flowing stream of coupling liquid is provided through the holder tube 135, with the feed line being shown at 137. This holder and its coupling liquid stream arrangement may be patterned after that shown in FIG. 2

of my pending application Serial No. 715,002 now Patent No. 3,055,210. The hood is provided with an internal partitioning wall 130W dividing the same into upper and 13

lower chambers

139 and 140, respectively. Large mouthed feed lines 141 located at each side of the holder tube 135 supply coupling liquid into the upper chamber 139 with the fiow rate and capacity of this input arrangement being sufficient to maintain the upper chamber substantially full of coupling liquid. The partitioning wall 130W is provided with suitable openings for each ultrasonic path and these openings are fitted with depending tubes 130T that terminate adjacent to the diaphragm 132 with approxmatelyclearance being provided. It will be apparent that a solid flow stream of coupling liquid is maintained through each depending tube 130T to complete the ultrasonic transmission path from the crystal to the diaphragm which is preferably coupled to the rail through a film of prewetting liquid. The coupling liquid is pooled in the lower chamber 140 of the hood and is withdrawn by a suction pumping system (not shown) which connects to generally rectangular exhaust passages 142 that surround each lengthwise half section of the hood. The flow streams through each of the depending tubes 130T impinge upon the diaphragm 132 and continuously cause it to conform to the surface contour of the rail.

Descrplion of FIG. 9 embodiment An alternative embodiment of a progressive rail flaw testing system is shown in FIGS. 9 and 10 wherein crystals are arranged in direct flat-faced contact with the trough diaphragm. The trough and guide shoe arrangement may be like that shown in FIGS. 3 and 5 but as shown in FIGS. 9 and 10 different techniques are employed for coupling the crystals to the rail. With reference to the rail R, the crystal arrays illustrated in FIGS. 9 and 10 include a straight or vertical beam unit 215 at the center, comprised of separate sender and receiver crystals 2158 and 215R, respectively, mounted in a common block 215H, and flanked by composite sender crystal units 216 which in turn are flanked by composite receiver crystal units 217. In each case the receiver unit 217 is to pick up reflected signals produced from the immediately adjacent sender unit 216.

The trough frame 205 is equipped with cross braces 226 on opposite sides of the center vertical crystal unit 215 and pivot bearings 226B are provided on these cross braces to mount a swing frame 227 that carries a guide tube 227T. A guide rod 227R is fixed to the holder block 215H of the center crystal unit and is slidable vertically in the tube and is normally biased downwardly by a spring 218 seated within the tube 2271 to impress the crystals against the diaphragm sheet 232. Control of the vertical lateral angle of the center crystal unit 216 is effected by pivoting the guide tube 227T under the control of a suitable control cable 210 and since only slight angular changes (on the order of 3 on a crystal /2" wide) are required for accomplishing the desired positioning, a V thick diaphragm sheet of rubber or rubberlike material will yield sufficiently to permit such minute crystal tipping. Neoprene is suitable for the diaphragm material, particularly in View of its resistance to abrasion.

Each of the

composite crystal arrays

216 and 217 are mounted in a common holder block and each block is biased against the diaphragm 232 by a pair of vertically acting springs 220. The springs 220 seat against cross bridges 225 that are secured to opposite sides of the trough frame 205.

The connection of the ultrasonic gear for firing the crystals in a predetermined timed sequence is shown in FIG. 9. The ultrasonic gear includes a rate generator RG for controlling a set of six serially connected ultrasonic machines U1 to U6. Thus, the rate generator RG will trigger ultrasonic machine U1 by direct connection thereto and U1 is connected in turn to trigger U2 which in tu-m fires U3 and so on in sequence.

Ultrasonic machine U1 is connected to the center crystal unit 215' for rail R and ultrasonic machine U2 is connected to the forward set of composite crystal units tube.

ultrasonic machine U6 is connected to the rearward set of

composite crystal units

216 and 217 for the rail R.

A memory type display tube T is illustrated as the indicating device and it is connected to receive gated output from the parallel connected output lines 39 leading from each of the ultrasonic machines U1 to U6. The tube T is connected to present a B-scan display of all reflected signal intelligence from both rails and the output from the ultrasonic machines is therefor applied to the control grid of its writing gun 40. As indicated in FIG. 2, the horizontal deflection facilities 41 are supplied with a scanning signal from a dual unit 42 that also supplies erase pulses to the storage mesh 43. The vertical deflection plate facilities 44 of the tube T are supplied with saw tooth sweep signals from a sweep generator 45 ,which is triggered separately from connection lines that branch from the input to ultrasonic machines U1 and U5. Each of these connection lines includes a suitable delay circuit 46.

In a typical cycle of operation of the vertical deflection system controlled from sweep 45 (during which the slower acting horizontal system controlled by sweep 42 may be considered stationary), the viewing screen of the memory tube is swept only twice. On the first sweep, corresponding points along each of the head and base indication lines for rail R, as derived by machine U1, are painted at the top of the tube followed by points making up the pattern derived by machine U2 from the forward set of composite crystals on rail R then followed by corresponding points along each of the head and base lines for the vertical crystals of rail R, as derived by machine .U3, and finally followed by points making up the pattern derived by machine U4 from the forward set of composite crystals for rail R. On the second sweep, points making up the pattern for the rearward set of composite crystals on the rail R are painted in substantially exact overlapping relation to the pattern of the corresponding forward set of composite crystals and thereafter the points making up the pattern from the rearward set of composite crystals for rail R are painted in substantially in exact overlapping relation with the pattern of the corresponding set of forward composite crystals.

The arrangement of each of ultrasonic machines U2, U4, U5 and U6 for firing the composite crystals is like that of machine U2 which is shown in FIG. 11. It includes an input delay circuit 31 adjustable for appropriately setting the firing times for achieving the pattern arrangement described hereinabove. The delay circuit 31 controls the time of firing for the test signal generator 32, the output of which is coupled with that from a high frequency oscillator 33 and fed to a matched delay line and amplifier network unit 230 for simultaneously triggering the two separate composite sender crystal units 216 of the forward set. Reflected signals are detected by the two separate composite receiver crystal units 217' and are fed through matched delay line and amplifier network units 240 that supply separate receiver channels each of which includes an I.F. amplifier 35, a detector 36, and a video amplifier 37. The outputs from the video amplifiers 37 are fed to an AND/OR video gate 34VG which in turn is connected to the output gate 34. A delay generator 38 controlled in timed relation from the delay unit 31 enables the gate 34 during the desired intervals. The gated output is then fed to the writing gun of the memory The separate receiver channels avoid any phasing problems such as could lead to cancellation of signals received simultaneously by the two separate receiver crystals. This same technique may also be employed in the ultrasonic machines used in FIG. 2 in order to avoid phasing problems where separate receivers are simultaneously receiving reflected signal intelligence.

Each composite sender crystal unit may comprise a set of 10 crystals, each of 1 x cross section, arranged in a common block in closely spaced side-by-side relation along the length of the rail to present an approximately 1" x 1" composite unit. Barium titanate and lead titanate crystals have high capacity relative to cross sectional area to make them suitable for such an application. These crystal elements are arranged in straight or fiatfaced relation to the rail surface and they are actuated in a closely time spaced ordered sequence to develop a composite action for generating a wave within the rail directed at a flat angle. The general principles of this technique are described in my copending application Serial No. 855,150, filed November 24, 1959, now Patent No. 3,166,731.

In order to develop a fiat angled beam having a direction axis at an angle of 85, the time delay from one end of the crystal to the other should be closely related to the travel speed of the ultrasound within the rail. The details of the delay network 230 for the sending crystal unit 216 are shown in FIG. 12. The left-most crystal element in FIG. 12 generates a pattern that sprays and fans out in the fashion diagrammed therein. This pattern includes components directed at all angles and is characteristic of the essentially point-like application of the ultrasonic pulse resulting from the sliver-like crystal configuration that is employed. Each successive crystal along the line will generate a similar pattern and when the timing of the occurrence of these patterns is appropriately selected, the second crystal will generate a component aligned with and in phase with the 85 component from the first crystal. Thereafter the third and successive crystals will do likewise until the 85 angle component becomes sutficiently reinforced and dominates the internal signal pattern set up in the rail. Assuming the ultrasound within the rail travels 1" in 4 microseconds and assuming the composite crystal has a 1" dimension lenghtwise along the rail, the delay line for controlling the firing times of the crystals should have an over-all delay interval slightly less than 4 microseconds. Proper precision adjustment should be carried out individually on each unit. The effectiveness of delay line control over composite crystal units is subject to sharp attenuation of the applied signal and the delay line arrangement of FIG. 12 includes a stage of amplification for each delay line section. Each delay line section comprises a series coil element L and a shunt capacitor C and the signal is sharply attenuated as it passes through each such section. Accordingly, a transistor A is connected between each set of successive delay line sections and is arranged to restore the applied signal to its original strength. In the disclosed arrangement a PNP transistor is connected in a common emitter configuration. The base of the transistor A is connected to the juncture of the coil L and the capacitor C in the first delay line section, the emitter is connected to the grounded terminal of the capacitor and the collector is connected to the input to the second delay line section and is also connected to the second crystal. This arrangement is unusually simple and effective because barium titanate and lead titanate crystals have very low impedance and this matches the low input impedance of the common emitter transistor amplifier stage.

Similar provisions are made for the delay network 240, for the receiver crystal unit 217' and the details are il lustrated in FIG. 13. A reflected signal is shown returning on the line indicated at RS and at first actuates the right-most crystal element in FIG. 13 to develop a signal that is applied to the coil L of the right-most section of the delay line and this section includes a shunt capacitor C. Since this right-most delay line section comprising elements L and C causes signal attenuation, an amplification stage is provided for coupling the signal to the next delay line section. The amplification stage utilizes a transistor A connected in a common emitter configuration to establish a good impedance match with the rightmost crystal which is in its input circuit. The advancing wave SR in the rail successively approaches and actuates the crystals in sequence and the timing of the delay line sections is related to the travel speed of the ultrasound in the metal to cause a successive signal build-up along the delay line. Proper timing causes the multi-seetion crystal to be selective with reference to the angle of approach of the signal to the delay line array.

Description of FIG. 14 embodiment Another embodiment of a progressive rail flaw testing system is shown in FIG. 14 wherein each crystal array has solid coupling relation to a trough diaphragm 55. The trough and guide shoe arrangement is generally similar to that shown in FIGS. 9 and 10 but a different coupling technique is employed for the angle crystal units.

With reference to the rail R, the corresponding crystal array, as illustrated in FIGS. 14, 15 and 16, includes a straight or vertical beam unit 315 at the center comprised of separate sender and

receiver crystals

3155 and 315R, respectively, mounted in throughbores in a common holder block 3151-1. The vertical beam unit 315 is flanked by fore and aft

angle beam units

316 and 317, respectively. Angle beam unit 316 has separate sender and receiver crystals 3168 and 316R, respectively, and angle beam unit 317 has separate sender and

receiver crystals

3175 and 317R, respectively.

The trough frame 305 employs upper and lower elongated rectangularly shaped clamping

bands

311 and 312, respectively (see FIG. 17), that grip the margin edge of the generally rectangular diaphragm sheet 55. Vertically oriented studs 3128 are welded or otherwise secured to the lower clamp band 312 at points spaced about the periphery thereof. These studs 3128 project through locking pad portions 311? provided on the upper clamp band 311 and wing nuts 313 secure the parts in releasable relation. Once again the trough frame is capable of holding a small pool of coupling liquid to enhance ultrasonic coupling to and through the diaphragm 55. The central region of the diaphragm is free to flex and distort as necessary to conform to changes in rail surface contour and the elongated configuration of the trough and of the diaphragm allows smooth travel thereof over irregularities so that abrupt lifting or tipping forces are avoided.

The trough frame 305 is equipped with a lengthwise overhead brace or support 326 secured between upstanding standards 3263 provided on the guide shoes 303. Bias springs 327 are anchored to the lengthwise brace 326 by adjustable anchor screws 327S threaded therein and are engaged to the crystal units to normally bias the same downwardly and impress the crystal holder blocks continuously against the diaphragm sheet.

The holder block 3151-1 is of aluminum and is A" thick, 2 /2" long, and /4" wide. The throughbores provided in the holder block are arranged to receive the

crystals

3155 and 315R in directly cemented relation therein. For each of the angle beam units a plastic wedge is provided and is cemented within a rectangular aluminum frame 307. For the arrangement illustrated herein both the sender and the receiver crystals are mounted on One of the exposed faces of the wedge W and a dab of putty P is applied to the other exposed face of the wedge to damp internal reverberations. For the illustrated arrangement, the crystals are oriented on an incident line at an angle of 54 to the running surface of the rail so that so-called fiat angled beams are developed within the rail. The bias springs 327 are arranged in point contact with front and rear regions of the frame 307 and preferably act along the lengthwise centerline of the frame so that the frame is capable of undergoing continuous rail contour following movement

Claims

6. APPARATUS FOR SIMULTANEOUSLY PROVIDING A SEPARATE LENGTHWISE PROGRESSIVE PICTORICAL REPRESENTATION OF THE INTERNAL STRUCTURAL CHARACTERISTICS OF EACH OF A NUMBER OF DEPTH RANGE REGIONS THAT EXTEND LENGTHWISE THROUGH A TEST BODY AND COMPRISING A MULTI-ELEMENT ASSEMBLY HAVING A SEPARATE ELECTRO-MECHANICAL TRANSDUCER DEVICE CORRESPONDING TO EACH DEPTH RANGE REGION AND ALL MECHANICALLY INTERCONNECTED FOR SIMULTANEOUS MOVEMENT, EACH TRANSDUCER DEVICE BEING ARRANGED UPON ACTUATION TO EMIT A TEST SIGNAL PULSE AND TO RECEIVE ECHOES OF SUCH PULSE, MEANS FOR MOVING SAID ASSEMBLY RELATIVE TO SAID BODY TO PROGRESS EACH SAID TRANSDUCER DEVICE LENGTHWISE ALONG THE BODY AND ENABLE EACH TRANSDUCER DEVICE TO SCAN THE LENGTHWISE SUCCESSIVE PORTIONS OF THE DEPTH RANGE REGION OF THE BODY THAT CORRESPONDS TO EACH SUCH TRANSDUCER DEVICE, A MEMORY TYPE CATHODE RAY TUBE HAVING A FLUORESCENT SCREEN, MEANS FOR PROJECTING A BEAM OF CATHODE RAYS ONTO SAID SCREEN, TWO COORDINATE SCANNING MEANS FOR FOR SAID TUBE AND INCLUDING FIRST MEANS OPERABLE DURING LENGTHWISE MOVEMENT OF THE ASSEMBLY RELATIVE TO THE BODY TO ACT AT A LOW REPETITION RATE AND AT A SLOW SCAN SPEED FOR REPEATEDLY DEFLECTING SAID BEAM ALONG ONE COORDINATE DIRECTION FOR A LONG SCAN TIME, A SWEEP CIRCUIT OPERABLE AT A SCAN SPEED MANY TIMES FASTER THAN SAID SLOW SCAN SPEED FOR DEFLECTING SAID BEAM ALONG ANOTHER COORDINATE DIRECTION, MEANS PERIODICALLY OPERABLE AT A HIGH REPETITION RATE MANY TIMES FASTER THAN SAID LOW REPETITION RATE TO GENERATE AN ACTUATING CYCLE, THE LAST-NAMED PERIODIC MEANS INCLUDING MEANS OPERABLE DURING EACH SAID ACTUATING CYCLE TO PULSE SAID TRANSDUCER DEVICES IN A PREDETERMINED TIME SPACED RELATION DETERMINED BY THE ROUND TRIP PULSE TRAVEL TIME FOR EACH CORRESPONDING DEPTH RANGE REGION TO PRODUCE FROM EACH TRANSDUCER DEVICE DURING EACH CYCLE AN INDIVIDUALLY OCCURRING TIME SEPARATED OUTPUT COMPRISED OF AN A-SCAN PULSE ECHO SIGNAL PATTERN EACH REPRESENTATIVE OF A LENGTHWISE SUCCESSIVE PORTION OF THE PARTICULAR DEPTH RANGE REGION THAT CORRESPONDS TO SUCH TRANSDUCER DEVICE, MEANS FOR APPLYING THE OUTPUT OF SAID TRANSDUCER DEVICES FOR VARYING THE INTENSITY OF THE BEAM IN ACCORDANCE WITH THE INTENSITY OF THE PULSE ECHO SIGNAL PATTERNS, THE SAID PERIODIC MEANS INCLUDING MEANS OPERABLE DURING EACH CYCLE TO RESET THE SWEEP CIRCUIT ONCE FOR EACH TRANSDUCER DEVICE AND IN A PREDETERMINED TIMED RELATION TO THE TIME OF OCCURRENCE OF THE OUTPUT OF SAID TRANSDUCER DEVICES TO ENABLE PROGRESSIVE ILLUMINATION SUBSTANTIALLY SIMULTANEOUSLY OF A SEPARATE LENGTHWISE EXTENDING REGION OF THE SCREEN CORRESPONDING TO EACH DEPTH RANGE REGION OF THE BODY.