US2471251A - Magnetic transducer adapted to compensate for twisting of the record wire - Google Patents

Description

May 24, 1949. H E. TOMPKINS MAGNETIC TRANS DUGER ADAPTED T0 COMPENSATE FOR TWISTING OF THE RECORD WIRE Filed Sept. 6, 1945 5 Sheets-Sheet 1 Fig. 2 /0 INVENTOR. HOWARD E. TOMPK/NS ATTORNEYS y 1949- H. E. TOMPKINS 2,471,251

MAGNETIC TRANSDUCER ADAPTED TO COMPENSATE FOR TWISTING' OF THE RECORD WIRE Filed Sept. 6, 1945 3 Sheets-Sheet 3 v INVENTOR.

' HOWARD E. TOMPK/NS A TTORNEVS Patented May 24, 1949 MAGNETIC TRANSDUCER ADAPTED TO COMPENSATE FOR TWISTING OF THE RECORD WIRE Howard E. Tompkins, Media, Pa., assignor, by mesne assignments, to Philco Corporation, Philadelphia, Pa, a corporation of Pennsylvania Application September 6, 1945, Serial No. 614,739

18 Claims. 1

This invention relates to a magnetic wire recorder, and more specifically it relates to a system for reproducing transversely magnetized recordings upon a round wire in such a manner that the reproduced signals are unaffected by twistin of the wire on the longitudinal axis.

In recording intelligence upon a round wire by the transverse method of magnetization, a difiiculty is encountered in the reproduction of the sound or other intelligence from the record on the wire because of the circular cross-section of the wire. This circular cross-section makes it possible for the wire to twist upon its longitudinal axis.

Consequently it is possible during reproduction for the wire to twist to such an extent that the magnetization of the wire is at right angles to the pick-up coil of the reproducer. Under these circumstances, the amplitude of the signal picked up may vary from full signal at no twist to no signal at a 90 twist of the wire.

Various arrangements have been suggested for overcoming such a difliculty. One such arrangement involves the use of automatic volume control in the amplifier, but this automatic volume control ceases to function when the wire has rotated a full 90 from the pick-up coil.

Another method which has been suggested for overcoming this difiiculty employs an automatic follow-up system whereby the pick-up coils are caused to rotate around the longitudinal axis of the wire in synchronism with the twisting of the wire. Such a mechanism, while accomplishing the desired result. is subject to high maintenance cost and possible failure,

In my invention I use two pick-up magnetic structures oriented at right angles with respect to one another and at substantially the same location longitudinally on the wire which is to be run through the reproducer. The use of two such coils at right angles to each other insures that there always will be a signal available from the wire for if in one coil there is zero signal because the magnetism of the wire is oriented at right angles to the pick-up axis of the coil, then the other coil will receive a full signal.

In order to obtain an output from these two pick-up heads, it is necessary to find a method of combining the two voltages in such a manner that the combined signal will be substantially independent of the orientation of the magnetic axis within the pick-up head structure. It can be shown that a simple addition of these two voltages will not sufiioe since under certain circumstances, with these two voltages added togetherdirectly, there will be a cancellation between the two and no signal will result from their combination.

I have discovered that in order to combine these two signals properly, one of the signals should be shifted in phase by with respect to the other one. When this is done, the two resultant signals can be added together vectorially and their sum does not contain the orientation angle of the magnetic axis of the wire in such a way as to vary the amplitude of the signal obtained.

This means that regardless of the orientation of the wire, the output from the system has a constant amplitude if the magnetization of the wire is constant in amplitude. In other words, as the amplitude of magnetization of the wire varies, the output amplitude varies in the same proportion independently of the orientation of themagnetic axis of the wire on which the record has been recorded,

Consequently, an object of my invention is to provide a means for reproducing transversemagnetization upon a round wire in such a manner that there will be no interference or distortion resulting from random orientation of the magnetic axis of the wire.

Another object of my invention is to provide means for cancelling out the effect of twisting of a round wire in a magnetic recording.

Other objects of my invention will become apparent upon a study of the diagrams, in which Figure 1 represents a schematic diagram of my pick-up head.

Figure 2 represents certain vector diagrams used in explaining the action of Figure 1.

Figure 3 is a block diagram of my invention.

Figure 4 represents phase shift curves obtained from the apparatus of Figure 3; and

Figure 5 is a detailed circuit diagram of the apparatus used to obtain the phase shift curves of Figure 4 and used in conjunction with my reproducing mechanism.

Referring now to Figure 1, the wire I0! is shown in cross-section. It is considered to be a long, thin, round wire whose length is perpendicular to the plane of the drawing, In order to explain the operation of this device, the wire is represented as having magnetic poles impressed upon it, the north pole being represented by the reference letter N and the south pole being 5 represented by the reference letter S.

The magnetic axis of the magnetic force impressed on the wire by the recording mechanism is at an angle 0 with respect to a horizontal reference axis. The strength of these magnetic poles varies as the position is changed longitudinally along". the'wireand' it is the'variationin the strength of-these magnetic poles which constitutes the recording upon the wire. The recording mechanism, which may be of any well known construction, is arranged toimpress. upon the. wire variations in magnetic pole strength inaccordance with desired signal recording.

The object of my invention is to satisfactorily reproduce the variations in magnetic pole strength ranged at such an angle as to carry flux-from the 5- permanent magnetism of th wire IIlI in the horizontal direction and -magnetic structure I83" is? similarly arranged: to: carry flux from the per; manent :magnetism in the vertical directions Each-1 of" these two magnetic structures is wound-, with a= pick up coil; structure IElZ being' wound :with -pick upcoil IM and structure" I03 being=;wound-with-pick-up coil I05; Variations in the magnetization of the wire lIlI cause varia l tionsiinzthe flux-inthe pick-up magnetic structunes I02.v and I03. and these' variations in'flux through .coils- I04 and I05 induce voltages inv thesercoilsr. Inworder to more fully understand the operationof my, invention, I have labeled these twovoltages: as-E1 and E2.

In Figure 2w, thereis showna simple construction-lion obtain-ing the magni-tude of these two voltagesEiandfEa Theangle a-is'plotted, as is indicated linathis diagram; andwa-line for vector IUBr-isrdraWn'.at-this' angle with respect to the" horizontal- Vector lilfi has a length-whieh is" proportional to: :the": voltage: which would be in troducedfiin-one oiathe1pick up coils-if it were oriented-:to 'be -parallelto the magneticaxis of thecwirev- That is,-if-eitherone"of the'two'pickan up coils were rotated so that the northand south" poles sof the--magnetization on the wire were-directly: opposite the I polepieces" of the magnetic structures;the -voltages picked up? in the coils wouldb'eproportional 'to the vector I 56.

In.Fig-; 1,,however, this arrangement is not carried out, and consequently thevoltages induceddn thetwdwindi'ngsare proportional to certaimfunctionspfthe angle 0.- In particular, the voltage induced-in the winding IIl l'is the m maximum. voltage" multiplied by the cosine of anglefios correspondingly, the voltage induced in coil IIl5 is :the maximum possible induced voltage multiplied by the sine 0. The construction of Figure 2' carries out this multiplication:

The magnitudepf Ei', represented by vector IIlI, is showrr'as the'verticalprojection of vector I06 andxthecmagnitude of E2-is shown as thehorizontal projection I 08 of'vector Hi6;-

Itiis to-befu-nderstood that these two vectors llt'bandlflfl are-riot vectors in the sense that they represent vector quantities oi E1 and E2. Their lengths only are associated with E1 and The anglesrinvolved are drawn from the angles onorientationro zthe magnetic structure and wire,

and consequently the angles in Figure 2a correspond to angles in the mechanical structure and not to electrical angles.

In fact, the two voltages E1 and E2 are in phase with each other, and consequently when they are added together in Figure 2b, the direct addition of these two voltages E1 and E2 is shown. Here the total voltage is the sum of the two voltages and since both E1 and E2 are in phase with each other, the vectors representing E1 and E2 are drawn as co-linear vectors.

As":the'-'ang1e"-0"is varied; the two projections IilI and mIi sh'own' in Figure-2a will vary in size. It can be shown that as these two vary, the sum oiFthe two as shown in Figure 212 will also vary in .size.. In. particular, if the angle 0 should vary betweenQD and 180, it will be observed that the voltage-E2'willube reversed in phase. As a result, in Figure 21), E2 would turn out to be oppositexir rdirectionto E1 and when the angle 0 became'fi the amplitude of E2 would equal the amplitude of'Er. Under this circumstance, thertwewouldcanceland the sum of voltages E1 and E2 would then be zero. This would not be a satisfactory'system for reproducing the recording on the wire; it would be no better than the use of?a single' piclr-up coil.

In my-invention l Fuse electrical means -to asso ciate the amplitudes E1 andE2 with the-angle 0' g in the same manner that the diagram-of Figure 227, associates the two amplitudes and obtains therefrom the vector I06. otm-y invention is shown-in Figure 2'0. According to my inventionfl cause vector E1 to be shifted in phase from vector E2 by In'carrying out this-process, both vectors may he -shifted inphase; but vector E1 is shiitediin- Consequently, in

phase 90 more than vector E2. Figure:2c,=the--vector-E1 after its phase shift, is

represented by vector )9. This vector is called Similarly, vector IIO-'"represents voltage E2.

Er; after having-been-shifted by the phase angle. This vector" I I Wis used to represent the voltage which-is voltage'E2 after its phase shift. The characteristics of the phase shift given to the two-voltages, cause E1 and E2 to be out of accordingly with the result that Figure 2c al-- ways bears an essential similarity-to Figure 2a. Consequently, vector III remains at a constant length-fora constant input amplitude independ- Infact, theangle entof the phase angle of 0. I IZin' Figure. 2c is always the same'as angle 0 1 in'Figure 2a.

A' block diagram ofthe apparatus used in ace complishing'my, invention is showniniFigure- 3.

Here, pick-up .coils I05andv I04 feed voltages .Er and-E2-into phase shifters H2 and 3.. Phasetorially and: fed into theamplifier. I14, which in The vector'diagram" Likewise, the.

Phase turn feeds the loud-speaker H5. This loudspeaker is symbolic of any reproducing device which may be desired. Thus, in order to more thoroughly explain the action of my invention, I will make reference to certain equations, numbered 1 to 10, which follow in the description below.

Let the magnetization in the wire be represented by Equation 1.

This equation can be thought of as the flux which would pass through one of the pick-up structures if the wire were so oriented that the maximum flux passed through this pick-up structure. The time variation represented by sine of is obtained by the longitudinal motion of the wire through the pick-up structure. In view of the random orientation of the wire, each of the pick-up coils will receive flux proportional to the appropriate functions of the angle 9. These two fluxes are indicated as 1 and 4m in Equations 2 and 3.

1= m sin of sin 6 (2) These are the fluxes which pass through the magnetic structures I02 and E03. From these fluxes, voltages are induced in the coils I04 and 105. Thus, in coil [05, the voltage induced is E1 given by Equation 4 and in coil I04, the voltage induced is E2 as given by Equation 5. In these equations, N is the number of turns in each coil, assuming them to be identical.

As has been mentioned before, the summation of these voltages directly is not useful in reproduction of the signals on wire 50!. This is shown by Equation 6 in which is given the sum of E1 and E2.

The last term in this equation, that is, the sine of (+45) is obtained by mathematical processes from the appropriate Equations 4 and that is, the sum of sine 6 and cosine 0. It is recognized in mat..ematics that this function, sine of (04-45"), varies between 0 and I depending upon the value of 0. Consequently, for specified values of 6, this function will be Zero and no output voltage would result. This makes such a system of direct combination of the two coils totally impractical.

The output voltages of my phase shifters, however, do make a practical system. These output voltages are given in Equations 7 and 8.

If E1 is the voltage as shifted by phase shifter H3, and E2 is the voltage as shifted by phase shifter H2, it is to be observed that the phase shifters cause the addition of the term cc and oz90 in the functions involving wt in these two equations.

With my invention, these two voltages E1 and E2 are added together. In order to make these two additions easy, it is first to be observed that the cosine curve involving of, Equation 8, is equivalent to a sine curve as is shown by Equation 9, which is an identity taken from trigonometry.

6 Consequently, the sum of E1" and E2, as given by Equation 10 is in a form which is recognized in trigonometry as being the expanded form of the sine of the sum of two angles.

When this Equation 10 is collected into a more concise form by trigonometric transformation, Equation 11 results, in which the sum of the two angles E1 and E2 is given as the sine of wia+6.

Inspection of this equation reveals that the angle 0 enters in the equation only as a phase angle in the resultant voltages. It does not appear in such a way as to alter the amplitude of the resultant voltage. This is precisely the effect which I desired to obtain when I made my invention. It makes the invention a satisfactory one for reproducing sound from a recording made transversely on a round wire since the ear, which will be the ultimate interpreter of the reproduced sound, does not respond to variations in phase angle of a signal. Thus the inclusion of 0 as a phase angle is unimportant in this system and will not cause distortion as far as the ear is concerned.

In another form of my invention I use a plurality of pick-up coils and a plurality of phase shifting networks so arranged that the phase shifts of the networks are correlated with the angular positions of the pick-up coils. For example, three pick-up coils arranged spatially at 60 intervals and three phase shifters designed for 60 phase angle differences provide an output independent of the twist of the wire.

Still another variation of my invention uses in place of the plurality of magnetic structures, a single magnetic structure built and wound in a manner similar to that employed in the construction of polyphase motor stators.

My invention has been described, heretofore, with respect to a single frequency of magnetization. However, it is well known that ordinary sound to be reproduced will be composed of a great number of frequencies. From the wellknown principle of superposition, it will be found that a complicated sound may be considered to be made up of a summation of individual frequencies. Since the entire system here described is a linear system, the system will behave, in regard to each frequency component, just as it would behave with regard to that frequency independently of the presence of other frequencies.

In consequence of this, the output signal of the system will contain a great number of frequencies each being shifted by the phase angle 0 and in addition, each being shifted by the phase angle minus a. In accordance with my invention, the phase angle a is different for each frequency and consequently each frequency component in the output will be shifted in phase from the corresponding frequency component in the input by a specified amount, which will be different from the phase shift for each other frequency component. This, however causes no difficulty since the ear is unresponsive to changes in the phase angle.

In order to accomplish the phase shift of between the two signals E1 and E2, the phase shifters H2 and H3 are designed in accordance with my invention as described in my co-pending patent application, Serial No. 613,457, filed August 30, 1945. I

Inaccordance withsthis invention, I obtain a' phase shift curveeforphase shifter H3 in accordance with curvershown' in Figure 4. Likewise, with respect-to. phase shifter H2, I obtain a .phase shift in accordance. with the curve of Figure 4.

It is to be observed that these two curves indicate a changingwphase shift with respect to frequency over the-whole frequency band, but that the essential characteristics of these two curves are that thephaseangle differs between the two systems substantially 90 over the audio frequency band. This is shown as curve 24 which is the difference in phase between the two systems.

For theoretically perfect operation, this phase difference should be exactly 90. However, if the phase angle Variesby slight differences from 90, there will be only aslight variation in the resultant amplitude, which will result in only Slight variations in the output amplitude as angle varies. Inconsequence of this, the difference in phase shift between phase shifter H2 and phase shifter ll3l= need not be exactly 90 but may be between a range of approximately degrees of 90;

Thus, the network whose curves are shown in Figure 4, accomplishes this satisfactorily. This network which gives the curves shown in Figure 4 is shown in Figure'5i This shows a complete reproducing system, with. pick-up coils, phaseshifting networks forcombining the two signals, amplifier, and reproducing speaker.

In Figure 5, ML energizes the two pick-up coils I04 and W5: One. ofthese pick-up coils is fed into the top-channel consisting of two bridge circuits 33 and 34, and1a differential amplifier comprising vacuum: tube 39; the output of which feeds into a differential" amplifier composed of tubes 3| and 32;

The other pick-upcoil feeds into the lower channel which is composed of phase shifting bridges 36 and 31 and" a differential amplifier composed of tube 4|, the-output of which feeds into the differential amplifier composed of tubes 3| and 32;

These phase shiftin channels are built in accordance with my invention described in copending application Serial No. 613,457, filed August 30, 1945. In the lower channel, for example, bridge 36 has a predetermined phase shift characteristic with respect to frequency. Bridge 31 has a similar, but different phase shift characteristic with respect to frequency; that is, its phase shift characteristic has a point of inflection at a different-frequency than does bridge Since the two of these bridge networks are eifectively connectedin tandem in this circuit, their phase characteristics are additive. Consequently, the phase characteristic of the lower channel consisting of bridge 35 and bridge 31 together, is the sum of the individual phase characteristic of the two networks.

Such a curve is given, for example, by curve 20 in Figure 4. This curve is almost linear with respect to th logarithm of the frequency. The upper channel is soproportioned that it has the same type of curve forits phaseshift curve, but it has this curve shifted in frequency to a higher frequency.

Accordingly, the .curve of theupper channel in Figure 5 is shown hmFigure 4 as curve 23. This is accomplished in theediagram by usingsmaller RC products-in the upper; channel thanare used,

RC product in bridge 33has a lower value than the RC product in bridge 36 and the RC' product" in bridge 34 has a lower value than the RC productin bridge 31.

cycles is 5.3 approximately. This, then, is the ratio of the frequencies of the two curves. Consequently R1C1=5.3R2C2.

This ratio, of course, does not apply to capacitors such as 38, which are merely blocking capacitors and do not function in providing the phase shift in these networks. only to those capacitors in the bridge circuits themselves. The output circuit of Figure 5 is a conventional push-pull connection with a transformer and a loud-speaker connected into the secondary of the transformer.

In the description and specification which I have given for my invention above, I have described it with respect to a specific network. However, I do not wish it to be limited to a specific network; but instead, I prefer to have my invention described by the following claims.

I claim:

1. In a system for reproducing sound from a magnetized wire which is subject to arbitrary twists, two pick-up coils at substantially the same location longitudinally on the wire and arranged to be responsive to mutually perpendicular components of magnetic flux from said wire, the amplitude of the induced signals in said coils varying in accordance with said twists, two phaseshifting networks, one of said phase-shifting networks being fed from one of said pick-up coils and the other of. said phase-shifting. networks being fed from the other of said pick-up coils, said phase-shifting networks having phase characteristics which differ from each other by substantially over a band of frequencies, and means for combining the output signals of said phase-shifting networks to produce a resultant signal which varies in amplitude in accordance with the magnetic recording on said wire and independent of any twist thereof.

2. In a system for the reproduction of sound from signals transversely recorded on a magnetized wire, two pick-up devices, each including a magnetic circuit and a coil linking said magnetic circuit, said pickup systems being oriented to be responsive to magnetization at the same longitudinal location on said wire butto be responsive to mutually perpendicular components of magnetic flux set up by the magnetization of said wire, two phase-shifting networks, each having input and output terminals, the input terminals of one of said phase-shifting networks being connected to one of said pick-up devices, and the input terminals of the other of said phase-shifting networks being connected to the other of said pick-up devices, the output terminals of said phase-shifting networks being connected so as to provide a signal which is the sum of the output signals of said phase-shifting networks, and said phase-shifting networks havin phase-shift characteristics which differ from each other by substantially 90.

3. In a systemfor. reproducing-sound from a For all of these RC products. the same ratio exists.

This ratio applies magnetized wire of circular cross-section which is subject to arbitrary twists, pick-up coil means arranged to be responsive to magnetic flux from said wire at substantially the same location longitudinally on the wire, means including circuit connections to said pick-up coil means for producing from said flux two signals, one of the signals being shifted in phase substantially 90 with respect to the other, and means including circuit connections to said last circuit connection means for vectorially adding the resultant signals to produce a signal which is varied in accordance with the magnetization of said Wire and independent of the orientation angle of the magne ism of the wire.

4. The method of reproducing sound from a magnetized wire which is subject to arbitrary twists, comprising generating two components from said flux, producing therefrom two signals, one of the signals being shifted substantially 90 with respect to the other, and adding the two signals vectorially.

5. In a system for reproducing sound from a magnetized wire of circular cross-section which is subject to arbitrary twists, pick-up coil means arranged to be responsive to magnetic flux from said wire at the same longitudinal location of said wire, means including circuit connections to said pick-up coil means for producing from said flux two signals, one of the signals being shifted in phase substantially 90 with respect to the other over a predetermined frequency range, and means including circuit connections to said last circuit connection means for vectorially adding the resultant signals to produce a signal which is varied in accordance with the magnetization of said wire and independent of the orientation angle of the magnetism of the wire.

6. The method of reproducing sound from a magnetized wire which is subject to arbitrary twists, comprising generating two components from said flux, producing therefrom two signals, one of the signals being shifted substantially 90 with respect to the other over a predetermined frequency range, and adding the two signals vectorially.

'7. In a system for reproducing sound from a magnetized wire which is subject to arbitrary twists, two pick-up coils at right angles to each other at substantially the same location longitudinally on the wire and arranged to be responsive to component magnetic flux from said wire, means including circuit connections to said pick-up coils for producing from said flux two signals, one of the signals being shifted in phase substantially 90 with respect to the other, and means including circuit connections to said last circuit connection means for vectorially adding the resultant signals to produce a signal which is varied in accordance with the magnetization of said wire and independent of the orientation angle of the magnetism of the wire.

8. In a system for reproducing sound from a magnetized wire of circular cross-section which is subject to arbitrary twists, a pair of magnetic structures at substantially the same location longitudinally on the wire and one of said structures being arranged to pick up flux from said wire in the horizontal direction, and the other of said structures being arranged to pick up flux from said wire in the vertical direction, a coil wound in each of said structures and energized by the flux therein, a circuit connected to each of said coils, the signals in each of said circuits being substantially 90 in phase displacement 10 from each other, and a common circuit connected to the output of each of said first-mentioned circuits.

9. In a system for reproducing sound from a magnetized wire which is subject to arbitrary twists, a pair of magnetic structures at substantially the same location longitudinally on the wire and one of said structures being arranged to pick up flux from said wire in the horizontal direction, and the other of said structures being arranged to pick up flux from said wire in the vertical direction, a coil wound in each of said structures and energized by the flux therein, and means for generating a signal equal to the square root of the sum of the squares of the signals induced in each of said coils.

10. In a system for reproducing sound from a magnetized wire which is subject to arbitrary twists, two pick-up coils at substantially the same location longitudinally on the wire and arranged to be responsive to predetermined components of magnetic iiux from said wire, the amplitude of the induced signals in said coils varying in accordance with said twists, two phase-shifting networks, one of said phase-shifting networks being fed from one of said pick-up coils and the other of said phase-shifting networks being fed from the other of said pick-up coils, said phaseshiftin networks having phase characteristics which differ from each other by said predetermined angle over a band of frequencies, and means for combining the output signals of said phase-shiiting networks to produce a resultant signal which varies in amplitude in accordance with the magnetic recording on said wire and independent of any twist thereof.

11. In a system for the reproduction of sound from signals transversely recorded on a magnetized wire of circular cross-section, two pick-up devices at substantially the same location longitudinally on the wire and each including a magnetic circuit and a coil linking said magnetic circuit, said pick-up systems being oriented to be responsive to magnetization at the same longitudinal location on said wire but to be responsive to predetermined components of magnetic flux set up by the magnetization of said wire, two phase-shifting networks, each having input and output terminals, the input terminals of one of said phase-shifting networks being connected to one of said pick-up devices, and the input terminals of the other of said phase-shifting networks being connected to the other of said pick-up devices, the output terminals of said phase-shifting networks being connected so as to provide a signal which is the sum of the output signals of said phase-shifting networks, and said phaseshifting networks having phase-shift characteristics which difier from each other by said predetermined angle.

12. In a system for reproducing intelligence from a magnetized wire which is subject to arbitrary twists, a pair of magnetic structures at substantially the same location longitudinally on the wire and arranged to carry mutually perpendicular components of flux from said wire, a coil wound in each of said structures and energized by the flux therein, and means for generating a signal equal to the square root of the sum of the squares of the signal induced in each of said coils.

13. The method of reproducing intelligence from a magnetized wire which is subject to arbitrary twists, comprising generating a plurality of components from said flux, producing therefrom a plurality of Y signals, said signals being shifted "in-phase with respect to one another by electrical angles which are substantially the same as the ;mecham'cal angles of said componentsof said -flux, and addin said signals vectorially.

W 14. A reproducer for magnetic recordings where the recorded signal is recorded magnetically transversely of the recording medium comprising apair of signal channels having input and output terminals for transmitting signals of the same frequency to the same extent; means for modifying the phase of signals passing through one of said channels relative to the phase of signals passing through the other of said channels to apredetermined extent; means connected to said output terminals for combining the output signals of said channels; magnetic signal pick-up devices for individually applying signals of the "same frequency to the inputs of said channels; said pick-up devices being located in the same plane, transverse to the direction of motion of the recording medium, and being angularly related with respect to each other by an angle substantially less than one hundred and eighty degrees; and circuit connections for producing the same output signal regardless of the instantaneous rotary orientation of the magnetic signal in the wire, said circuit connections comprising means for maintaining constant the relative phase angle between the signals at the output terminals of said channels regardless of the frequency of said signals.

15. A reproducer for magnetic recordings where the recorded signal is recorded magnetically transversely of the recording medium comprising a pair of signal channels having input and output terminals for transmitting signals of the same frequency to the same extent; means for modifying the phase of signals passing through one of said channels relative to the phase of signals passing through the other of said channels to a predetermined extent; means connected to said output terminals for combining the output signals of said channels; magnetic signal pick-up devices for individually applying signals of'the same frequency to the inputs ofsaid channels; said pick-up devices being located in the same plane, transverse to the direction of motion of the recording medium, and being angularly related with respect to each other by anangle substantially less than one hundred and eighty degrees; and circuit connections for producing the same output signal regardless of the instantaneous rotary orientation of the -magnetic signal in the Wire, said circuit connections comprising means for maintaining a substantially ninety degree phase angle between the signals at the output terminals of said channels regardless of the frequency of said signals.

16. A reproducer for magnetic recordings where the recorded-signal is recorded magnetically transversely of the recording medium comprising a pair of signal channels having input and output terminals for transmitting signals of the same frequency to the same extent; means for modifying the phase of signals passing through one of said channels-relative to the phase of signals passing through the other of said channels to a predetermined extent; means con- 12 nected to said output terminals for combining the output signals of said channels; magnetic signal pick-up devices for individually applying signals of the same frequency to the inputs of said channels; said pick-up devices being located in the same plane, transverse to the dire'ctionof motion of the recording medium, and being effectively ninety degrees with respect to each other; and circuit connections for producing the same output signal regardless of the instantaneous rotary orientation of the magnetic signal in the wire, said circuit connections comprising means for maintaining constant the relative phase angle between the signals at the output terminals of said channels regardless of the frequency of said signals.

17. In a reproducer for magnetic recordings on a wire, a pair of pick-up magnetic structures oriented at right angles with respect to one another and at substantially the same location longitudinally on said wire, circuit arrangement for shifting the signal outputs of said structures substantially with respect to each other, said circuit arrangement including a first phase shifter connected to the output of one of said structures for shifting the signals through an angle a., and a second phase shifter connected to the output of the other of said structures for shifting the signals through an angle a90, and a common output circuit for said phase shifters in which the resultant signals are added vectorially.

18. In a reproducer for magnetic recording on a wire, a pair of magnetic pick-up structures, one of said magnetic structures being arranged to pick up one component of the signal on the wire equal to the signal which would be induced in one of the structures if said one of the structures were oriented to be parallel to the magnetic axis of the wire multiplied by the cosine of the angle of orientation of the structure to the magnetic axis of the wire, and the other structure being arranged to pick up the other component making up the signal on the Wire equal to the signal which would be induced in oneof the structures if said one of the structures were oriented to be parallel to the magnetic axis of the wire multiplied by the sine of the angle of orientation of the structure to the magnetic axis of the wire, individual circuit arrangements connected to the output of each of said structures, said circuit arrangement shifting the phase of the output signal substantially 9O electrical degrees with respect to each other, and a common output circuit for said individual circuit arrangement for vectorially adding the signals in each of said circuit arrangements.

HOWARD E'. TOMPKINS.

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

UNITED STATES PATENTS Number Name Date 1,949,409 Cohen Mar. 6, 1934 2,251,300 Star Aug. 5, 1941 2,272,821 Roys Feb. 10, 1942 2,334,510 Roberts Nov. 16, 1943

Images