US2869073A - Method and apparatus for distinguishing harmless surface flaws from dangerous fissures in magnetizable bodies - Google Patents

Description

Jan. 13, 1959 c. w. MCKEE ETAL METHOD AND APPARATUS FOR DISTINGUISHING HARMLE SURFACE FLAWS FROM DANGEROUS FISSURES IN MAGNETIZABLE BODIES- Filed March 9, '19s:

4 Sheets-Sheet 1 J l I I {RA/L BALL z W aim INVEI:%)5%e V Jan. 13, 1959 c. w. MOKEE ET AL 2,869,073

METHOD AND APPARATUS FOR DISTINGUISHING HARMLESS SURFACE FLAWS FROM DANGEROUS FISSURES IN MAGNETIZABLE BODIES- Filed March 9, 1953 v 4 Sheets-Sheet 2 GAUGE 66 68 22 711 N81 39 A n n n V U U U 1' 4;.- piaa J85 230 178 17g 180 BELL D E y vpu U" STAVGEQ J 5TA6'E *3 255112 V OUTPUT fi iag I J86 1A5 J75 172 179 INVENTORS.

c. w. M KEE ET AL 2,869,073

G HARMLESS SURES Jan. 13, 1959 METHOD AND APPARATUS F OR DISTINGUISHIN SURFACE FLAWS FROM DANGEROUS FIS IN MAGNETIZABLE BODIES Om 9% 02 mm QQW Q2 \M Q2 Q2 1 OZ NQ l m9 v WNX a2 NW 02 09 WWX l 3 QT OZ ESL um mwmwi mew 3595 Q3 mp3 Nu on a RA INVENTORS 01/ Jan. 13, 1959 c. w. MGKEE ET AL 2,369,073

1 METHOD AND APPARATUS FOR DISTINGUISHING HARMLESS SURFACE FLAWS FROM DANGEROUS FISSURES IN MAGNETIZABLE BODIES Filed March 9, 1953 4 Sheets-Sheet 4 GAU i -"if I h. \EXTENT OF F/SSURE FIELD LEAD TURNS 0F COIL 198- 206 A REAR- TURN5 0F can. 198* 212 LEAD TURNS 0F COIL 200i A 210 216 REAR TURNS 0F can. 200 1 22? Campos/TE SIGNAL 1 RECEIVED AT FIRST --V- M STAGE OF AMPLIFIER 220 INVENTOR C 6 66 228 BY A United States Patent Chester W. McKee, Hoinewood, and Richard W. McKee, .Highland Park, llh, assignors, by mesne assignments, to Teleweld, Inc., a corporation of Idaho Appiication March 9, 1953, Serial No. 340,944 4 Claims. (Cl. 324-37) This invention relates to an improved method and apparatus for distinguishing harmless surface flaws from dangerous defects in magnetizable bodies. It will be particularly useful in segregating signals derived from hardened portions of a magnetic body. Its success is closely dependent upon the use of a sustained magnetic field in which a flux-responsive means is operating. Inasmuch as the invention stems from an observation made during rail testing, applicants will describe the figures of the drawings, and then describe the experiment.

Fig. l is a schematic side elevation of a Teledetector fissure detector car;

Fig. 2 is a side elevation of a rail illustrating applicants sustained flux field pattern;

Fig. 3 is a side elevation of a rail indicating flux patterns from a shell and a fissure in a residually magnetized rail;

Fig. 4 is a side elevation of a rail ball indicating flux patterns from a shell and a fissure in a sustained magnetic field;

Fig. 5 is a plan view of the top of a portion of a rail showing schematically the positioning of applicants coils thereabove;

Fig. 6 shows typical graphs of a fissure potential signal and a shell potential signal derived by applicants apparatus;

Fig. 7 sets forth two hypothetical tapes and compares the results obtained by full wave rectification as compared with half wave rectification in conjunction with a sustained field;

Fig. 8 is a wiring diagram of applicants pickup circuit and amplifier;

Fig. 9 is a composite arrangement of a side elevation of a rail, a plan View of a rail, and graphs of potential signals for the purpose of suggesting a reason why applicants invention functions;

Fig. 10 is a graph of a fissure potential signal seen on an oscilloscope connected to the Teledetector apparatus; and,

Fig. 11 is a graph of a shell potential signal seen on an oscilloscope connected to the Teledetector apparatus.

In detecting dangerous defects, whether internal or surface, in magnetizable bodies, the first step is to explore a magnetic field adjacent a body with a flux-responsive means as rapidly as possible. For example, in testing a new rail in a rolling mill, the flux-responsive means will be moved along a rail in either a sustained or a residual field and if it produces a potential of a selected amplitude, a check of that portion of the rail producing the. magnetic field will be made. Because the rail is new and does not have in its surface shells, flows and wheel burns, the detecting apparatus can be adjusted to a high degree of efficiency. The surface contour of the new rail ball is reasonably uniform with the result that the detection apparatus can be made responsive almost exclusively to the type of signal derived from a dangerous defect. The same holds true inthe checking of new pipe and other factory-madearticles of steel. An im- 2,869,073 patented Jan. is, 1959 portant field for fissure detection work, however, is in checking magnetizable bodies which have been used. In the case of rails in track, the uniform contour of the ball no longer exists. ,It is broken by rail joints, wheel burns, shells and flows, all of which are harmless, but all of which produce distortions in the magnetic field adjacent the rail which in turn produce potential signals in the flux-responsive means. The same is true of such material as oil well drill pipe, which upon removal from the ground may have slight indentations due to blows or the teeth of a wrench or a collet of a rotary drill.

These surface. defects render what is called the exploratory step of fissure detection work inaccurate. A potential signal derived from a distorted magnetic field created by a shell may have the same amplitude as a potential signal derived from a flux field created by a fissure, and consequently, the visual portrayals of these potential signals will not have such distinctive significanoe as to enable an observer to separate dangerous fissures from harmless surface defects This is a very old problem infissure detection work and hundreds of patents have issued on pickup coils, positioning the pickupcoils, connecting the pickup coils, amplifier circuits, suppressors and visual presentation means, all directed toward. assisting the operator in distinguishing the harmless from the .dangerous flaws, in the first instance. Whereheis unable to inake this distinction with confide-ncein the first instance, he is obliged to make close visual examination or a hand check of that portion of the magnetizable bodyfrom which the potential signal was derived. The hand checking of harmless flaws, or more commonly the visual examination of the exterior of a mag netizable body to confirm. that a visual signal came from a harmless defect, is a major item in fissure detection service. In: the case of testingrail in track, each stop of the exploratory car consumes several minutes. Inasmuch as these cars must test about thirty miles of track in .a seven-hour day in order to be commercially practical, and inasmuch as there is only about one dangerous fissure in every thirty or fortymiles of. track which is checked at intervals of three to six months, it is evident that the ability to distinguish the harmless from the dangerouswfissures in the first instance, that is, when the detector car is moving at ten to twelve miles an .hour along the track, is of crucial importance.

Applicants are the designers of what is known in the trade asthe Teledetector fissure detector car. A study of the tapes made by this car on the north rail of the Pocatello, Idaho, test track of the Union Pacific Railroad resulted in an. observation which had never been made before, although the applicants had been working on, the present Teledetector sustained field system for several years. In order toappreciate the observation, it is necessary to understand rather specifically the Teledetector apparatus in use at the time that the experiment occurred,

TheTeledetector car, referring to Fig. 1 is a 60- to 70- foot motor car 10 mounted on a forward motor truck 12 and a trailing truck 14. In the forward end, is a drivers. cab 16 and at the rear is a fissure detector operators cab 18. Generatorsprovide power fordriving, the car along the rail and functioning the magnets, amplifiers and visual presentation; means. The latter usually consists of two, batteries of pens, one for each rail, writing on a tape moving approximately two inches for each 39 feet of rail traversed. Beneath the car are three magnet trucks 20, .22 a;nd124, each having vertically positionedelectrically energized magnets. While the car is moving, there is a sustained magnetic field following .the dotted lines 26. At 1the point where the flux lines leave therail at substantially right angles to the ball of. the rail, there lis nositio e a fiu spqns m an '2 w c Campinas.

a plurality of pairs of coils having non-magnetic cores.

This is the point where a dip needle stands somewhat vertically and it occurs between the two wheels primarily because of the spacing of the rear magnetic truck 24 from them. The output of each pair of coils of the fluxresponsive means 28 is carried to a separate amplifier 30 which in turn functions a separate pen unit 32. Whenever a coil is moving through a field in which the number of force lines cut is increasing, a positive potential will be induced in the coil; and conversely, whenever it is moving through a field in which the number of lines of flux is decreasing, a negative potential will be induced in the coil.

Theorizing on the functioning of fissure detection apparatus has not proved very satisfactory in the past. However, it is likely that the observation which applicants made would be true only of a flux-responsive means operating at a fixed position in a moving sustained field and hence the general nature of such a field should be understood. In Fig. 2, applicants show the rear portion of the magnet carried by the truck 24. This magnets lower pole has a north polarity adjacent to the rail. When energized, it pushes out into the rail lines of flux which leave the rail along the dotted lines such as 36. One pair of opposed connected, longitudinally aligned coils, in cross section (much enlarged), is indicated by the numeral 38 positioned in that part of the field where a dip needle 40 stands substantially vertically. The fluxresponsive means 38 is at a fixed distance behind the magnets. Assuming for a moment that one has a perfect rail and that the assembly is moved from right to left, the flux pattern in the air remains exactly the same. It is just as if nothing moved. Now, assuming that there is a shell, which is a surface defect 42, as the flux field moves to the left, the pattern of the lines of flux immediately adjacent that defect is altered. The density and the direction of the flux is different from that existing over the normal rail ball. Inasmuch as the coils 38 generate only a potential when there is a change in the quantity of the flux or in the direction of the flux passing through them, it is evident firstly that the coils 38 will generate some kind of a potential signal as they move over the defect. All that the coils 38 check are changes in the quantity and direction of the lines of flux coming out of the rail immediately beneath the coils, that is, in the area between the two arrows 44 and 46.

Further as to the nature of the flux field above a rail, in Fig. 3 the rail has been residually magnetized, that is, a magnetizing force has been moved along the rail ball to a point where it no longer exerts any appreciable effect on the field around the rail. The numeral 48 indicates a shell, above which is suggested the closed lines of flux 50 as they would appear in a residual field. A shell is a portion of a rail which has become hardened.- The steel in a shell has a higher reluctance than a normal portion of a rail. Its color is dark. It may extend along the gauge side wall of the ball for several inches and occasionally extends on the top of the ball as far as the median line thereof. The numeral 52 identifies an internal fissure. The faces of the internal fissure in the rail have opposite polarities so that if the left side 54 is north, the right side 56 is south. The residual flux field in the air above the fissure will be as indicated by the dotted lines 58.

In a sustained field, however, the concepts of the flux fieldsmodified by defects are quite different. Here the powerful magnet 24, see Fig. 4, having a north pole adjacent to the rail is lining up the molecules and holding them as indicated by the arrows, wherein the head of each arrow indicates a north pole. When this sustained field flux pattern encounters a shell 49, the lines of flux are concentrated around the perimeter of the shell. In the center ofthe shell, the number of lines of flux are comparatively few. When this sustained field flux pattern encounters'an internal fissure 63, there isa concentration of flux 66 on the lead side of the fissure, and 'a.

to traverse a field 67 of subnermal intensity.

selected amplitude is received.

' into the last stage is positive.

less than normal quantity of flux 67 on the trailing side of the fissure. N

Continuing to refer to Fig. 4, the pickup 64, which is moving to the left, leaves a field of normal intensity It next threads the much denser field 66, after which it re-enters the normal flux field 61. The potential signal induced in the coils by the fissure sustained flux field will therefore be negative, strong positive, and negative. In traversing the sheel field 66, on the other hand, the pickup 64 moves from a normal flux field into a more dense, next into a very subnormal 69 in the center of the shell, and then back into a normal field. The potential signal induced in the coil by the shell field will therefore be first positive, and then strongly negative, and then positive. In traversing the shell field, there may be many reversals in the direction of the lines of flux, but the important fact is that the low density sustained flux field above the shell produces a high amplitude negative component.

The pickup coils of the Teledetector car on this particular test were three effective pairs. See copending application Serial No. 256,502, filed November 15, 1951, now Patent No. 2,766,425, of which this application is a continuation in part. Their full size and position over the rail is shown in Fig. 5. The numeral 66 identifies the ball of the rail, having a gauge side 63 and a field side 70. The flux-responsive means is a non-magnetic block 72 having vertical, cylindrical cavities into which are dropped pairs of coils connected in opposed relationship. The gauge pair of coils 74, the center pair 76, and the field pair 78 are all identical. The two dotted lines 80 and 82 indicate the outer and inner limits of the windings on each coil. Each pair of coils has one lead grounded and the other is connected to a separate amplifier which in turn functions a separate pen unit. The three pen units write on the same tape 132, see Fig. 7, the gauge pen being designated No. l; the center pen, No. 2; and the field pend, No. 3.

The Teledetector car checked the north rail of the Pocatella test track. This test track contains ten fissures ranging in size from eight percent to fifteen percent and approximately twenty shells. The tape recorded eight of the fissures and thirteen shells. The applicants connected one amplifier to an oscilloscope and they studied the potential signal delineation of all thirteen shell signals and eight fissure signals. In Fig. 6, applicants sketch side by side a shell signal 83 and a fissure signal 81, typical of those studied. While the amplitude of the eight fissure signals differed and the duration of the signal differed slightly, all of the fissure signals had one common characteristic, namely, that the high amplitude potential generated was positive. In the case of the shell potential signals, they had one common characteristic and that was that the high amplitude signal was always negative.

In the Teledetector apparatus at that time, there was a twin diode or full wave rectification stage, the output of which converts all potential signals received by that stage, whether negative or positive, into positive signals. The output stage of the amplifier functions a relay which in turn functions the pen unit whenever a signal of a importantly, the Teledetector apparatus has its stages so arranged that the signal Upon observing that the high potential signal derived from the fissure field was of a polarity opposite to the high potential signal of the shell, the applicants removed the full wave rectification stage in the amplifier and amplified only signals of that polarity which was the same as that of the high amplitude component of a fissure potential signal.

With this change, the rail was again checked. The results were outstandingly successful. One fissure field wrote on all three pens, five wrote on Nos. 1 and 2- pens, and four wrote on No. 1 pen. Every fissure was caught. Not a single shell field wrote on both Nos. 1

The No. l pen is connected to the gauge coil, the No. 2 to the center coil, and the No. 3 tothe field coil. Inasmuch as the shellsare almost always physically beneath the gauge and center coils, the operator was saved at least eight shell inspections. I In order that the advantage of this arrangement may be more clearly perceived, applicants have made up two hypothetical tapes positioned next to a hypothetical rail containing threefissures and six shells. This is shown in Fig. 7. The rail ball 84, shown out ofproportion, is illustrated as part ofa track in which joint bars 86 connect the rail 84 to rails 88 and 90. The three dotted lines 92, 94 and 96 indicate difierent size internal fissures in the rail. The groups of dots 98, 100, 102, 104 and 106 indicate different size shells on the gaugeside 108 of the rail. The flux-responsive means 110 contains the three pairs of coils, the gauge pair 112 connected through an amplifier 114 to No. 1 or gauge pen 116; the center pair 118 connected through an amplifier 120 to the No. 2 pen 122; and the field pair 124 connected through the amplifier 126 to the N0. 3 pen 128. Assuming that the flux-responsive means .124 is moving inthedirection of the arrow 130, the pens typically would have written the tape 132 when the Teledetector car was employing full wave rectification. At the joints, all three pens write 134. The first shell wrote on two pens 136. The first fissure wrote on the same two pens 138. The small shell 1'00 wrote on No. l pen 140. The shell-102 wrote on two pens 142. The small fissure 94 wrote on none of the pens. The large fissure 96 wrote on all three pens. The shell 104 wrote on No. l pen only, 144. The shell 105 wrote on no pens. The large shell 106' wrote on all pens 146.

In the column to the right of thetape 132 headed Is a Stop Indicated, applicants have indicated the correct and 2 pens.

answer in view of the tape, assuming that the detector operator was not looking out of his window to observe the presence of a shell on the rail. The correct answers call for five stops. From the standpoint of dangerous fissures, only two of these five stops should have been made and one stop, that is, for fissure 94, was not re quired.

The tape 148 is a hypothetical tape showing the results of running the car Without full wave rectification over the same rail. This tape is comparable to the long tape made on the test track. On this tape, there are only four signals that came through. It should be borne in mind that the amplification had been slightly stepped up on the second running of the track in order to catch the small fissure 94. The signals 150 and 152 indicated to the operator clearly that a stop must be made. The other two signals 154 from thesmall fissure 94 and 156 from the large shell 106 are substantially the same. Assuming that the operator could not see the track, he would have to stop both times. In practice, however, the operator is looking out the car window and the large shell is clearly visible. When he received the fissure signal 154 and saw no deformation of the rail, he would give the signal to stop. When he received the signal 156 and then saw the large shell, he would not give a signal to stop.

Comparing the two tapes, on the tape 132, the operator would have stopped twice for the fissures 92 and 96. And he probably woud have passed the shells 98, 102 and 106 because he could see them on the rail. .Had he increased the amplification so as to bring upthe small fissure 94, the shells 98 and 100 would probably have recorded on the third pen. The tape 148 .is vastly superior. It makes it possible to maintain the amplification at a sufficiently high point so that the smallest fissure will produce a signal in atleast one pen while at the same time reducing the number of shell signals which actuate the pen. The cleanness of the tape is important firstly because it is not confusing to the operator, but also because the tape is delivered to the railroad which will check it against a subsequent service failure.

Describing now the specific hookup employed by the applicants, and referring to Fig. 8, 158 is a section throughthe north pole of a vertically positioned magnet immediately adjacent a rail 160. Positioned above the rail rearwardly of the magnet are two coils wound on vertically positioned non-magnetic cores. The coils, one-half to three-quarters of an inch in diameter, are Wound identically the same. The inside lead of the leading coil 162 is connected by a conductor 164 to the first stage 166 of applicants amplifier. The outer lead of the leading coil 162 is connected by a conductor 168 to the outer. lead of the second coil 170. The inner lead of the coil 170 is grounded. Commencing. with the conductor 164, the circuits are diagrammatic only. Actually, the circuits are quite complex and vary for particular installations. The signal is carried through several amplification stages to a differentiating stage which is the fourth stage and bears the numeral 172. Here it is again phase inverted and with the high amplitude component of the fissure potential signal now positive, 180, the signal enters the output stage. The tube in this stage 182 is so biased (standard practice) that itbecomes conductive only upon receiving a positive potential signal. Negative components 179 do not affect this output stage. They are lost. Only the positive signal 180 actuates the pen unit 184 or the bell .185.

Applicants early'in this disclosure indicated that the. polarity of the high potential component in a shell potential signal was opposite to the high potential component of a fissure potential signal because the number of lines of flux within the sustained field passing through the shell and back up to the magnet were comparatively few. Establishing the truth of this simple explanation is difficult, in part because of the nature of the fluxrespon sive means. Applicants two coils on vertical axes positioned one behind the other are connected in opposed relationship and the lead turns on one coil. will produce apotential signal of the opposite polarity to the trailing turns of the same coil. Nevertheless, applicants have laid out Fig. 9 for the purpose of explaining What actually occurs.

Referring to Fig. 9, a side elevation of the ball 192 of a rail 188 is shown above and therebelow the top of the same rail ball. A fissure 194 isi ndicated by dash lines. The magnet is to the right and is moving from left to right. The pickup coil is to the left of the fissure 194 and consists of the two coils 198 and 2&0. The spacing of the lines offiux bracketed by the numerals 189 and 191 indicates the density of the flux flowing from perfect rail. The fissure 194 constitutes an obstacle to the flow of flux from the magnets and there is a concentration of flux 1% in the magnetside of the rail. Resultingly,

there is substantially les-flux flowing from the rail on the pickup side of the fissure, the lines being bracketed and indicated by the numeral 193. The shell 224, being composed of harder steel, presents an obstacle to the flow of flux at the surface of the rail and much of the flux that normally would leave that portion of the rail is pushed over to theedges of the shell. The arrows 226, in the side elevation of the ball 192., show a heavy density of flux. However, referring to the plan view, the dots indicate lines of flux and it will be seen that the lines of flux at 226 are denser than in the normal portion of the rail 191 and that the lines of fiux within the shell itself are much fewer than the number in the normal portion of the rail 191.

it is evident that as these coils leave the normal field 189 and enter the less dense field 1%, their overall output will be a negative potential. As they traverse the dense field 196, their.over-all potential will be strongly positive. Upon leaving this dense filed and returning tothe normal field 191, the cells will, produce a negative potential. 1 t t 7 The coils 198 and 200 probably would pass the shell 224 without producing an appreciable signal. However, the gauge coils, suggested by the coil 203, would pass over the shell. These coils 203 would produce a typical fissure signal-minus, strongly plus, minusin traversing the field above the fissure 194. As they leave the normal flux field 191, they would first encounter an increase in the number of lines of flux, then a decrease in the number of lines of flux cut, and finally an increase. The signal would be plus, strongly minus, then plus.

At the bottom of Fig. 9, applicants have indicated the composite signal received at the first stage of the amplifier 166 by the pickup coils as they traverse first the fissure, the signal bearing the numeral 220, and then the shell, the signal bearing the numeral 228. The high rate of change of the shell signal is indicated by the portion 226 of the signal 228. This also is a composite of impulses received by the pickup coils.

Continuing to refer to Fig. 9, the applicants have suggested how the lead and rear turns of the coils may reinforce each other. The diagram indicates how the lead turns at first produce. a negative signal 204 followed by a positive signal 206; how the rear turns of the lead coil 198 produce a positive signal 208 which partly overlaps the positive signal of the lead turns of the same coil. The lead turns of the coil 200 which is connected in opposed relationship with the coil 198 then produce a positive impulse which overlap in point of time the impulses induced by the other parts of the pickup circuit. The signals 212, 214, 216 and 218 occur in point of time so that they do not overlap. Signals 206, 208 and 210 may be additive to produce the large signal 230. This same effect undoubtedly occurs in the case of-the shell. The spacing of the lead and rear turns of'each coil and the spacing of the coils from each other Will affect the potential signal that reaches the first stage of the amplifier.

In making the connections between the pickup coils and the amplifier, two things are important. Firstly, the polarity of the large component of the fissure detector signal such as 230 must be known. Secondly, that polarity must be positive when it reaches the output stage 182. This latter can readily be accomplished by selecting the appropriate number of stages in the amplifier. Ordinarily, the polarity of a signal is reversed as it passes through each stage. The graphs of the signal as it moves through the amplifier are schematically shown between the stages in Fig. 8. The right result is obtained because there are five stages in the amplifier and the large component 230 is positive as it enters the first stage. If the connections to the pair of pickup coils are reversed, that is, the dot-dash connections 184 and 186 were used, the proper result would be obtained by adding or subtracting one stage in the amplifier. It is evident that the critical element is the polarity of the large component of the fissure detector signal, for in applicants sustained field system of detection, that polarity will always he opposite to the polarity of the large component of a surface defect signal.

The operability of applicants method and apparatus is dependent upon having the positive signal enter the output stage only because of the nature of that stage. That stage contains a tube which becomes conductive so as to operate a pen relay only when a signal of positive polarity is received by the stage. It is possible to design a stage in which a negative signal received at the input of the stage would alone establish an operating circuit through the output conductor of that stage.

A detailed wiring diagram of the amplifier has not been included because it bears no important relationship to the invention here disclosed. There have been slight adjust rnents of otentiometers and the insertion or withdrawal of condensers, all of which have improved the function ing of the device. But the basic idea resides in the elimination of full wave rectification in a flux-responsive 55 means testing or threading a uni-directional sustained field Applicants uni-directional trailing sustained field induces into the pickup coils a potential signal from a fissure field having a large component of a polarity opposite to the polarity of the large component of a signal derived from a shell. I g

It is important that too much weight not be given to the fact that the component of the fissure signal having the high amplitude appears as the second component in the schematic illustration in Fig. 9. The applicants are not certain that the large component has the same or the opposite polarity of the initial component in the fissure signal. As illustrated in Figs. 10' and 11, some tests indicate that the large component of either the fissure or the shell is the third and not the second component. The important thing is that the polarity of the large component of the fissure field is opposite to the polarity of the large componentof the shell, burn, flow or other surface defect creating closed fields.

The high speed in performing the exploratory step is becoming increasingly important in fissure detection work because of the increasing amount of controlled cooled rail in track. Controlled cooled rail does not develop fissures generated by shatter cracks within the ball. This had always been the primary source of internal fissures. Prior to the controlled cooled rail, the rails were cooled at the mill in air. The surface cooled first and established stress locks. As the interior of the rail cooled, it established its own stress locks. The result was stresses in a rail ball which repeated blows from wheels would release so as to start an internal fissure. As controlled cooled rails leave a mill, they are stacked white hot in an insulated railroad gondola and covered. Cooling requires from two days to a week. The result is that stress locks have been nearly eliminated. In these controlled cooled rails, internal fissures develop primarily from slag inclusions. At any event, the number of fissures developed is decreasing. The result is that many more miles of track must be tested for each fissure found. This makes it necessary to perform the exploratory step faster and more accurately.

Having thus described their claim:

1. The method of operating that type of apparatus for segregating internal fissures from harmless surface defects in magnetizable objects which equipment includes (a) a magnet having one pole positioned close to the magnetizable object with the other pole spaced vertically thereabove so as to create a flux field of lines of force leaving the object at an acute angle, said field being formed by flux moving from the first pole through a portion of the object and then outwardly through its surface and the field to the other pole; (b) a flux-responsive means sufficiently short so that in moving through said flux field adjacent the surface of the object and distorted by an internal fissure major changes in flux density .will generate in a lead to the flux-responsive means potentials of opposite polarity; (c) signal-producing means apprehendible by the human senses and capable ofdistinguishing between signals of opposite polarity; and (0!) means for rendering the apparatus responsive to signalsof just one polarity, which comprises the steps of magnetizing the objects so that surface flux fields aifected by internal fissures will have characteristically difierent flux patterns from those from surface defects, of moving the flux-responsive means through these fields and determining the polarity of the major potential received from a flux field modified by a known internal fissure, and of setting the apparatus to render, apprehendible signals of that polarity and that polarity only.

2. The method of operating that type of apparatus for segregating internal fissures from harmless surface defects in magnetizable objects which equipment includes (a) a magnet having one pole positioned close to the magnetizable object with the other pole spaced vertically invention, applicants thereabove so as to create a flux field of lines of force leaving the object at an acute angle, said field being formed by flux moving from the first pole through a portion of the object, and then outwardly through its surface and the field to the other pole; (b) a coil disposed on a vertical axis and suificiently short so that in moving through said flux field adjacent the surface of the object and distorted by an internal fissure major changes in flux density will generate in a lead to the coil potentials of opposite polarity; signal-producing means apprehendible by the human senses and capable of distinguishing between signals of opposite polarity; and (d) means for rendering the apparatus responsive to signals of just one polarity, which comprises the steps of magnetizing the object so that surface flux fields aifected by internal fissures will have characteristically different fiux patterns from thosefrom surface defects, of moving the coil through these fields and determining the polarity of the major potential received from a flux field modified by a known internal fissure, and of connecting the leads of the coil to thesignal-producing means so that only signals of the polarity of the major potential of the flux field modified by the internal fissure actuate the signal producing means.

3. The method of operating that type of apparatus for segregating internal fissures from harmless surface defects in magnetizable objects which equipment includes (a) a magnet having one pole positioned close to the magnetizable object with the other pole spaced vertically thereabove so as to create a flux field of lines of force leavingthe object at an acute angle, said field being formed by flux moving from the first pole through a portion of the object, and then outwardly through its surface and the field to the other pole; (b) a flux-responsive means sutficiently short so that in moving through said flux field adjacent the surface of the object and distorted by an internal fissure major changes in flux density will generate in a lead to the flux-responsive means potentials v of opposite polarity; (c) signal-producing means appre hendible by the human senses and capable of distinguishing between signals of opposite polarity; and (d) means for rendering the apparatus responsive to signals of just one polarity, which comprises the steps of magnetizing the object so that unidirectional flux flows outwardly into the space through which the flux-responsive means is to pass thereby creating flux fields adjacent internal fissures characteristically different from those adjacent surface defects, of moving the flux-responsive means through these fields and determining the polarity of the major potential received from a flux field modified by a known internal fissure, and of setting the apparatus to render apprehendible signals of that polarity and that polarity only.

4. The method of operating that type of apparatus for segregating internal fissures from harmless surface defects in magnetizable objects which equipment includes (a) a magnet having one pole positioned close to the magnetizable object with the other pole spaced vertically thereabove so as to create a flux field of lines of force leaving the object at an acute angle, said field being formed by flux moving from the first pole through a portion of the object, and then outwardly through its surface and the field to the other pole; (b) a coil disposed on a vertical axis and sufiiciently short so that in moving through said flux field adjacent the surface of the object and distorted by an internal fissure major changes in flux density will generate in a lead to the coil potentials of opposite polarity; (c) signal-producing means apprehendible by the human senses and capable of distinguishing between signals of opposite polarity; and (d) means for rendering the apparatus responsive to signals of just one polarity, which comprises the steps of magnetizing the object so that unidirectional flux flows outwardly into the space through which the coil is to pass therebyvcreating flux fields adjacent internal fissures characteristically difierent from those adjacent surface defects, of moving the coil through these fields and determining the polarity of the major potential received from a flux field modified by a known internal fissure, and connecting the leads of the coil to the signal-producing means so that only signals of the polarity of the major potential of the flux field modified by the internal fissure actuate the signalproducing means.

References Cited in the file of this patent UNITED STATES PATENTS 2,461,252 Barnes et a1 Feb. 8, 1949 2,461,253 Barnes et al.- Feb. 8, 1949 2,571,998 Barnes et al. Oct. 23, 1951 2,571,999 Barnes et al Oct. 23, 1951 2,602,108 Dionne July 1, 1952 2,614,154 Dionne Oct. 14, 1952 2,624,779 Keaton et a1. Jan. 6, 1953 2,639,316 McKee et a1 May 19, 1953 2,671,197 Barnes et a1. Mar. 2, 1954 2,729,785 Keevil Jan. 3, 1956

Images