US2441380A - Magnetic analysis - Google Patents

Description

M y 11, 1948- T. zUswLAG 2,44 ,3

MAGNETIC ANALYSIS ori in-a1 Filed April 8 1941 FIG. 4. .27 FIG. 6.

F 26a 2267 /26@ Lo 16' 33 I "Fr .15

iji 55' J6 INVENTOR THEODOR ZUSCHLAG v BY M h /m' w mm.

I ATTORNEYS Patented May 11, 1948 MAGNETIC ANALYSIS Theodor "Zuschlag, Welt Engiewood, N. 1., as-

signor to Magnetic Analysis Corporation, Long Island City, N. Y., a corporation of New York Original application April 8, 1941, Serial No.

387,436. Divided and this application November 30, 1944, Serial No. 565,838

This invention is concerned with magnetic analysis and particularly with testing ferromagnetic objects by magnetic means to determine variations in dimensions or metallurgical characteristics or both. In its preferred form, the

apparatus of my invention may be employed for 3 Claims. .(Cl. 175-483) I men. These changes in flux are indicative of simultaneously determining dimensional variations and the existence of flaws in steel and the like. The apparatus is capable of detecting and differentiating between deep-seated flaws and those which affect the surface characteristics of the objects. The present application is a division of my co-pending application Serial No. 387,436, filed April 8, 1941, now Patent No. 2,398,488, granted April 16, 1946.

I have discovered that variations in dimensions of ferromagnetic specimens and in the surface characteristics of such specimens may be detected and meansured with great accuracy by means of a ferromagnetic core energized by an alternating electromagnetic field with one end of the core disposed adjacent a ferromagnetic end piece but separated therefrom by a small non-magnetic gap of constant dimensions, and the other end of the core disposed adjacent the ferromagnetic specimen to be tested but separated therefrom by a small non-magnetic gap which varies withdimensional varia- =.tions in the specimens being tested. The apparatus is also provided with means for determining changes in flux in the end portion of the core adjacent the specimen. These changes in flux are indicative of variations in the dimensions or surface characteristics or both of specimens undergoing test.

For the determination of deep-seated flaws and other defects in magnetizable specimens, 1 have found that the core should be of magnetic material capable of becoming saturated at relatively low flux densities as compared with iron and the like. Such a core need not be provided with the ferromagnetic end piece described hereinbefore but should be made of a magnetic material that is easily saturated such, for example, as a nickel-chromium alloy of the nature below decribed. The core is disposed with one end in magnetic relationship with a ferromagnetic specimen to be tested, but is separated therefrom by a small non-magnetic gap (which is preferably but not necessarily constant in dimensions). Means are provided for creating a unidirectional magnetic field in the specimen. for creating an alternating electro-magnetic field in the core, and for determining changes in flux in the end portion of the core adjacent the specifor use in the detection of flaws as manifested deep-seated variations in the structure of the specimen undergoing test.

It will be apparent that the present invention may be employed in the testing of any magnetic or magnetizable material, and that the term ferromagnetic" is used broadly in the specification and claims to mean any magnetic material whether or not it contains iron.

In the preferred form of my apparatus both types of cores and their appurtenant equipment are provided so that variations in both external and internal characteristics of the specimens undergoing tests can be determined simultaneously.

These and other features of my invention will be more thoroughly understood in the light of .the following detailed description taken in conjunction vwith the accompanying drawings in which:

Fig. l is a wiring diagram of a preferred type of the apparatus of my invention provided with two test cores and means for creating a unidirectional magnetic field in the specimen undergoing test;

Fig. 2 illustrates an alternative circuit arrangement which may be employed with either of the two types of cores of my invention;

Fig. 3 is a further alternative circuit which can be employed with either type of core;

Fig. 4 is an elevation of a means for holding the first type of core in appropriate relationship to a specimen being tested, for example, for variations in thickness;

Fig. 5 is an elevation of a type of core-holding device of my invention particularly adapted for use in determination of dimensional variations in cylindrical bar stock or the like; and

Fig. 6 illustrates another type 'of core-holding device of my invention particularly adapted for holding a core in fixed magnetic relationship to a sample. of bar stock orthe like being tested for major variations in shape or dimension, or for slight variations in surface characteristics. The apparatus of Fig. 6 is particularly adapted by slight surface variations. I

Referring to F18. 1, it will be observed that an alternating-current supply l0, which may be a conventional 110 volt, cycle source, is connected to the input side of a step-down transformer I2 provided with a pair of substantially identical secondary coils l3, M. These secondary coils are connected respectively to two symmetrical bridge networks each provided with center tapped energizing coils l5, l8 shunted respectively by potentiometers l1, l8.- Each of these coils is wound in two portions and these portions are arranged and connected in series aiding relation, as shown. The taps oi. the respective pairs oi! energizing coils and the sliders oi the potentiometers are connected in bridge relationship across the primaries of two signal transformers i9, 20 respectively. The secondaries oi the signal transformers are connected respectively through amplifier combinations 2|, 22 to meters 23, 24.

The taps or sliders oi! the potentiometers are adjustable for initial balancing of the bridge circuits. The two bridges are substantially identical in all respects except as to the nature of the cores in the center-tapped energizing coils.

The energizing coils l5, 06 may be of any appropriate dimensions. Conveniently they are about 3" long and 1 /2" in diameter, and consist of several hundred turns of insulated wire of low ohmic resistance. As shown in the drawing, they are preferably wound in two equal consecutive portions positioned coaxially on an open magnetic core.

The multi-layered energizing coil i5 is provided with a magnetic core 25 which preferably is a single substantially straight piece 01 magnetize able wire of small gauge slightly longer than the coil. The core should be of small diameter (preferably less than about one-tenth of a centimeter) in order that the fiux densities in the core will be sufiicient to approach the saturation point of the material or which the core is made.

Core material that iscapable of being saturated at relatively low flux densities as compared with iron is much to be preferred. Alloys of nickel and chromium are, in general, suitable. Thus, a wire or about #22 gauge comprising an alloy containing approximately 60 to 80% nickel, to 20% chromium, and 0 to 25% iron and capable of becoming saturated at relatively low flux densities is preferable.

One end, viz., the adjacent end, of the core 25 is disposed adjacent a magnetizable specimen 28 to be tested. The axis of the core is perpendicular to that of the specimen, so that the other end of the open core is magnetically remote from the specimen. By magnetically remote" is meant that the other or remote end of the core is effectively spaced sumciently far away from the specimen so that in operation the magnetic influence of variations in the specimen on the end of the core remote from the specimen is substantially less than the influence on the adjacent end. In Fig. 1, the specimen is represented as a piece of cylindrical steel bar stock. The core 25 is supported so that the end adjacent the specimen is adapted to be spiraled" past the core by means of rollers (not shown), or other known supporting means for the specimen or material under test. To accomplish-this the specimen is rotated in a clockwise direction (as indicated by the arrow) while also being moved longitudinally.

Disposed immediately below the specimen as shown in Fig. 1 is a magnet 29 which has its axis transverse to that of the specimen. This magnet creates a unidirectional non-fluctuating magnetic field in the specimen. The magnet may be a permanent magnet or may be an electromagnet provided with a solenoid 29a energized by direct current source 2%. The reliability of the analysis often may be increased by demagnetizing thespecimen before testing it, for in- States Patent No. 2,207,392. If the specimen 28 under test is of thin-wall material such as tubing. the magnet 29 should be placed inside to oppose the core 25 so that the procedure previously described may be followed. In this event variations in thickness of the wall will be indicated.

Other means for creating a unidirectional magnetic field (i. e., a field such as that produced by non-fluctuating direct current) in the specimen undergoing test may be employed. For example, a direct current may be passed through the section or the specimen undergoing test from brushes in contact therewith, or other forms of direct current electromagnets may be employed and disposed at other angular relationships to the specimen.

For regulating the flux in cores 25 and 26 during initial adjustment of the apparatus, means such as rheostats may be employed. Although such regulation of the alternating current in the energizing coils is seldom necessary, it is advisable to provide for moving magnet is with respect to the specimen, or for varying the current in solenoid 29a if magnet 29 is an electromagnet, in order to adjust the field for maximum sensitivity of the testing core and coil arrangement.

The other energizing coil is of the apparatus oi Fig. l. is provided with a ferromagnetic core 2% which conveniently is a piece of quarter inch diameter substantially straight iron rod slightly longer than the winding of this coil. Such a core has a relatively large cross section, as compared with that of the core 25 used in exploring for deep-seated flaws, etc., and is capable of being saturated only at flux densities higher than those required in the case of the core 25. However, if desired this core 26 may also be made of an alloy of iron, nickel and chromium or the like provided it has a relatively large cross-section. The core 26 is disposed perpendicularly to the axis of the specimen undergoing test. The end of the core adjacent the specimen, via, the adjacent end, is separated from the specimen by a small air gap which varies during the rotation of the specimen in accordance with dimensional variations in the specimen. Thus, if the specimen is supported on rollers, the gap between the specimen and the end of the core will change as the diameter of the specimen changes. The gap between the core and the specimen preferably is very small, for example, only a few thousandths or an inch.

The other or remote end of the open core 26 is magnetically remote from the specimen, viz., is spaced sufficiently far away from the specimen so that in operation the magnetic influence of variations in the specimen on the remote end of the core is substantially less than the influence on the adjacent end of the core. This remote end is provided with a ferromagnetic end piece or counterweight, such as a disk of iron 21, disposed in any suitable manner with a constant non-magnetic gap 26a of suitable dimension between it and the end of core 26. The cross-section of the end-piece should be the same as, or larger than, that of the material under test. The end piece may be supported in such relation to core 26 by means of supports independent of the core, or it may be fastened to the end of the core by means of small bolts or screws of nonmagnetic material and separated from the core as by small non-magnetic washers, leaving a constant gap of air except for the space occupied by bolt and washer by which the dimension of stance, in the manner described in my United the gap can be adjusted. Alternatively a suitable thickness of insulating cloth may be interposed between the core and the end piece. From the foregoing it will be clear that the dimension of the air gapand the size of the end piece together comprise two variables with which the magnetic effect on the core due to the material under test may be counterbalanced by that of the end piece-hence the term counterweight." If the specimen 28 is supported from below on rollers (not shown) and is rotated and simultaneously advanced longitudinally, the air gap between the magnet 23 and the specimen will remain substantially constant, while the air gaps between the specimen and the two cores will vary with changes in cross-section of this specimen.

In the operation of the apparatus of Fig. 1. the bridge associated with the core 26 is balanced with a specimen 28 of standard dimension in the apparatus. The air gap between the specimen and the end of the core 26 should be fairly small, because the sensitivity of the apparatus increases markedly with a decrease in this air gap; By keeping the air gap small and employing high amplification, the meter 24 will register reliably variations in the air gap as low as one ton thousandth of an inch. The gap 26a between the other end of the core and the counterweight should also be small.

In thepreferred form of my apparatus, the size of the gap 26a between the counterweight and the core is adjustable. The bridge associated with the core 28 isbalanced roughly by adjusting this gap with a standard test specimen of known and desirable dimension placed in the apparatus. It is desirable to adjust this gap so that leakage flux linking only one side of center-tapped energizing coil l6 encounters a flux path whose reluctance is about the same as that encountered by leakage flux linking only the other side of coil l6, 1. e., so that about equal voltages appeal-across each half of the coil. Thereafter, the standard specimen is replaced by a specimen to be investigated. Any variations between the dimensions of the standard specimen and of the specimen being tested will unbalance the bridge, causing current to flow in transformer 20 and indicating the variation at once on the meter 24.

be discovered by exploring the surface t1. the

bars with finger tips. but this procedure is neither safe. nor reliable. However, in the apparatus of my invention, particularly with an energizing coil equipped with a core 26 having a magnetically balanced end .piece or counterweightjf minute changes in smoothness of the bar surface are detected reliably. When both typ s of cores are The bridge associated with the other test coil ii is brought into a balanced condition by adlusting the slider on the potentiometer I! while It will be apparent that it is unnecessary to I employ both types of cores in a single test apparatus. However, it is frequently desirable to do so. because the-core 25 'is sensitive to deepseated flaws and relatively insensitive to changes in dimensions or in surface characteristics of the specimen, while the core 26 is relatively insensitive to deep-seated flaws but highly sensitive to changes in. cross-sectional dimensions of the specimen being tested, and also to changes in surface characteristics. Certain types of defects in; steel samples, for example cracks, are manitested by a slight change in the smoothness of the surface of the bar. Sometimes cracks may used simultaneously, certain defects which may bemissed by one indicating system will be indicated by the other.

Referring now to Fig. 4, it will be observed that, in this instance, th core 26 is mounted on a bracket 3|, the height of which is adjustable on a standard or upright 32 extending upwardly from a base. The base carries a support 34 of non-magnetic material, for example, copper, brass, or insulating material, upon which a ferromagnetic specimen 33 to be tested is disposed. This specimen may be a thin piece of ferrous metal foil, which is carried over the support so that the gap between the end of the core 26 and the foil varies with changes in the thickness'of the foil. There is no contact between the core and the foil. The core 26 is provided with the energizing coil l6, non-magnetic gap 260:, and end piece 21', and is connected in the circuit as shown in Fig. 1.

In summary, the apparatus of Fig. 1 and Fig. 4 comprises a rigidly supported ferromagnetic core, coil means for creating an alternating electromagnetic field in the core, a ferromagnetic end piece disposed adjacent one end of the core and separated therefrom by a small non-magneticgap of constant dimensions, during test on material of given dimensions, non-magnetic means for supporting in sequence a succession of generally similar ferromagnetic specimens, or contiguous sections of the same specimen, adjacent the opposite and magnetically remote end of the core in magnetic relationship therewith but separated therefrom by a small non-magnetic gap (usually an air gap) which varies with the dimensional variations of the specimens, and means for determining changes (due to dimensional variations of the'specimens) in the flux'linking the portion of the coil means near the end of the core adjacent the specimen.

When round material is tested, it is preferable to employ the apparatus shown in Fig. 5. In this apparatus a plurality of steel balls 35, rid- 'ing on a cylindrical specimen 36 being tested, support a framework 31 on the underside of which is fastened the detecting coil l6 with its core 26 and counterweight 21. After adjustment of the air gap between the bar 36 and the core 26 any change in the diameter of the bar will be immediateiy indicated by the meter 24 connected in the bridge network.

The apparatus of Fig. 5, like that of Fig. 4, is adapted to be employed in conjunction with either type of test coil, and may be employed with the bridges as shown in Fig. l.

Ordinarily, it is more important to detect the presence of flaws in bar stock and the like than it is to detect changes in dimensions. Hence, in testing cylindrical bar stock for the presence of flaws which are manifested by changes in the superficial characteristics of the bar, it is preferable to employ the apparatus of Fig. 6. In this I apparatus, the energizing coil I6 is mounted upon the apparatus becomes sensitive primarily to the changes in the smoothness of the bar surface adiacent the end of the core, although. of course. very large variations in diameter or shape of the test specimen will change the gap.

The apparatus of Fig. 6-15 particularly suitable for use as a portable unit for the detection of cracks or other defects in cylindrical bar stock. which defects are characterized by superficial irregularities. For this purpose it is important that the gap between core and specimen be kept constant except for such superficial irregularities, and changes in the gap due to changes in diameter are undesirable in this use of testing core 26. It may be noted that the supporting arrangements illustrated in Figs. 1, and 6 may be adapted for use with non-cylindrical stock of any generally uniform cross-section, or for the inspection of parts such as races. In this case the specimen cannot be rotated, and is merely displaced longitudinally.

The supporting means of Fig. 6 (like the supporting means of Figs. 4 and 5) may also be employed with the core 25 and its appurtenant apparatus for the detection of deep-seated flaws. The apparatus of Fig. 6 is desirable for such use because it tends to minimize the variations in the air gap between the end of the core and the specimen. Although the core 25 is not particularly sensitive to changes in air gap, it will be preferable to keep these at a minimum.

Instead of the bridge arrangements illustrated in Fig. 1, other testing circuits, such as those illustrated in Figs. 2 and 3, may be employed. In the apparatus illustrated in these figures, instead of measuring changes in the voltages across two halves 38, 38' of the coil carrying the energizing alternating current, additional pick-up coils 39, 39' are place in inductive relationship with each half of the energizing coil, and the voltages induced in these pick-up coils are measured.

Changes in the flux linking one end portion only of the core will cause changes in the voltage induced in the pick-up coil near that end of the core.

In the particular arrangement illustrated in Fig. 2, the energizing coil is separated into two halves 38, 38' wound on core 26, and the center tap is omitted. The pick-up coils 38, 39 of like construction are connected so that the voltages induced in them buck each other, i. e., the coils are connected in series opposition, the relation of the coil assembly thus being symmetrical. When using a core 26 having an end piece and a fixed gap (26a) adjusted for magnetic balance as described above in connection with Fig. '1, the two voltages will tend to cancel each other completely, and a change in the flux linking the pick-up coil nearest the specimen being tested will cause a relatively great change in the voltage induced across the two coils. This voltage is impressed on the primary of a signal transformer ll, caus-- ing corresponding currents to appear in the secondary of the signal transformer. These currents are fed to an amplifier l2, and the resulting variations may be observed on an output meter 43 associated therewith. The coils 39, '38 may be constructed in any manner so as to be inductively coupled to the corresponding sections 38, 38' of the energizing coil, and may suitably be concentric with these sections. As a source of the alternating electromagnetic field in the core, the energizing coil may be connected to the secondary or a power transformer l0 similar to one side of the transformer 12 of F18. 1 and itseli energized from the line "I.

A similar core and coil arrangement is illustrated in Fig. 3,'in which, however, an electronic oscillator is used as a source of alternating current for the energizing coils 38, 38'. To accomplish this result. a vacuum tube 45 is employed with a source of plate power 49 and filamerit supply (not shown). The plate circuit con-- tains a tuned circuit comprising a variable condenser 4B shunted by a plate inductance M in series with the energizing coils 3B, 33' and a meter 50. The grid circuit contains a variable condenser 41 shunted by a feed-back coil 48, coupled to the coil 44 and in series with the pick-up but will also cause relatively great changes in the voltage appearing across the two pick-up coils. This voltage, being in series with the feed-back voltage appearing across coil 48, causes changes in the oscillations. Meter 50 will measure these changes and thus indicate the original changes in the flux caused by variations in the specimen.

It should be observed that the core 25 (Fig. 1) may be employed in place of the core 26 in the apparatus of Figs. 2 and 3, and that .the circuits of these figures may be employed in pairs (with both types of core) as illustrated in Fig. l.

Ordinarily energizing coils 38 and 38 of Figs. 2 and 3 are connected in series-aiding relationship, but they may also be connected in seriesopposing relationship. In either event the polarity of coils 39 and 39' should be opposite to that of coils 38 and 38', respectively, so that if coils 38 and 38'v are aiding, coils 3,9, 38 should be opposing, and vice .versa.

It will be apparent that marking relays, and indicators of various types may be employed in place of the current meters 23, 24, 43 and 50.

The apparatus of my invention, and particularly that form in which both types 25, 28 of cores are employed, is especially suited for production testing and permits more rapid and more reliable inspection than has been possible heretoforc. Moreover, more information can be obtained with respect to the material tested, since it is possible not only to inspect in one operation for both flaws and dimensional variations, but also to differentiate between different types of flaws.

What is claimed is: I

1. In magnetic analysis apparatus, the combination which comprises an open ferromagnetic core, a ferromagnetic end piece held adjacent one end of the core and separated therefrom by a. small non-magnetic gap, means for supporting a magnetizable specimen adjacent the opposite end 0! the core in magnetic relationship therewith but separated therefrom by an air gap, said end piece being proportioned approximately to magnetically balance the magnetic effect of said specimen on said core, a vacuum tube having grid and plate circuits, the plate circuit thereof being parallel-tuned and comprising a condenser shunted by a plate inductance in series with an energizing coil wound in two like sections in series core having two ends, means for disposing a ferromagnetic specimen to be analyzed adjacent one end of said core in magnetic relation thereto but separated therefrom by a small non-magnetic gap, a ferromagnetic end piece disposed adjacent the other end of said open core proportioned approximately to magnetically balance the magnetic effect of said specimen on said core, said end piece being separated from said core by a small non-magnetic gap, an energizing coil wound on said core and effectively connected to a source of alternating current whereby to create an alternating-current field in said core, two pick-up coils each of which is electromagnetically coupled to a portion of said energizing coil, said pick-up coils being connected in series opposition, said energizing coil and said pick-up coils being symmetrically arranged with respect to each other and said core, a circuit including said pick-up coils, and an indicating device efiectively coupled to said circuit for indicating a change in magnetic flux in said core due to a variation in said specimen.

3. In magnetic analysis apparatus, the combination which comprises, a substantially straight ferromagnetic open core having two ends, means for disposing a ferromagnetic specimen to be plate circuit thereof being parallel-tuned and comprising a condenser shunted by a plate inductance in series with an energizing coil wound in two like sections in series aiding relationship, one section being wound on each half of said core, a feedback coil connected in said grid circuit of the tube and coupled to said plate inductance, a pick-up coil connected in series with said feedback coil comprising two like sections in series opposing relationship, each section being wound concentrically with and coupled inductively to a section of the energizing coil, and means connected in the plate circuit of said tube to indicate current variations.

THEODOR ZUSCHLAG.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Date

Images