US2198080A - Phonograph recorder - Google Patents

Description

April 23, 1940.

.1. R. FOUCH PHONOGRAPH RECORDER Filed Dec. 15, 1938 2 S heetS -Shet 1 James RFouch IN V EN TOR.

BY A TTORNEYS.

April23, 1940. JR PO 2,198,080

PHONQGRAPH RECORDER Filed Dec. 15,- 1938 2 Sheets-Sheet 2 Fig. 0".

I I I I I I I l I l I I I I l I I I I I I I I 3 l I I I I 4 l D I I I I k I I I I I I I I 1 IIIIIII Fi IIII A TTORNEYS.

Patented Apr. 23, 1940 UNITED STATES PATENT OFFICE-3 PHONOGRAPH RECORDER A James R. Fouch, Inglewood, Calif. Application December 15, 1933, Serial N6. 245,845

11 Claims.

the principal end to which the present inven- 10 tion is directed.

Some additional objects of my present invention are: (1) to avoid the use of rubber and rubber substitutes as damping media; (2) to provide a recording head in which the resonant peak is variable as to frequency whereby it is made possible to compensate to better advantage for the inherent falling off of sound intensity at the higher frequencies, due to the compliance of certain record blank materials, and at the same time to overemphasize those frequencies which include the major part of the needle scratch noise, so that said noise can be more adequately suppressed; (3) to provide a recording head of structurally simple design which is easy to assemble and easy to adjust and; (4) to providea recording head which is not susceptible of being adversely affected by climatic conditions or changeshence one which'can be assembled and adjusted'in a temperate zone and shipped to the tropics or to the poles with assurance that it will function properly.

In order to facilitate a full comprehension of the present invention it is desirable to explain at the'outset the problem or problems which I have 7 tire useful audible frequency hand must actually be recorded' in true proportion throughout; and, as a matter of fact, to do so would not be com mercially feasible, and would result in reproductions having a high noise'level as well as inadequate intensity at the higher frequencies.

- To record the lower audible bass frequencies in truevolumetric proportion to the high frequencies would require a record groove of such inordinate width as to reduce the playing time 55 of record of ordinary size' to such an extent as to render it impracticable commercially.

is equally impracticable to maintain proportionality betweenthe lower bass frequencies and the highest audible frequencies vby decreasing the recording amplitude over the whole frequency rangebecause the-record undulations at the highest frequencies would be .too microscopic for any stylus to follow and, furthermore, indistinguishable from the grain of the recordmaterial.

As a practical matter the onlyway in which-it at thesame time maintain the ,intermediateand' lowest audible bass frequencies at a sufficientgam plitude. Theresult is that it has not heretofore been found commercially practicable to record all the,,audible frequencies, or a totally satisfactory part thereof, on a commercial phonographrecord, at, theirproportionate volume intensities.

I do not wish to be interpreted as implyingthat the results heretofore achieved in phonographic recording and reproduction have not been exceedingly good. The fact is that the phonographic recording and reproducing art had been developed to a very high degree long prior to the present invention; and my contribution is simply a; further but marked improvement in an art already approaching the pinnacle of perfection. V

Another of the problems which have to be dealt with in modern phonographic sound recording and reproductionis the elimination or suppression 'of background noise-commonly called needle scratch. Thevolume level of needle scratch is a constant. That is to say, the scratch level remains the same regardless of the amplitude of the recorded sound. If, then, the amplitude of the recorded sound undulations in the record groove is relatively large the objectionable background noise will be masked and thus rendered inaudible, or, at least, unnoticeable It is not feasible to thus blot out the background noise by simply amplifying the record undulations throughout the entire band of recorded frequen-' cies be'cause'thatwould increase too much the amplitude of travel of "the recording stylus at the lower bass frequencies-which means that the record grooves would overlap, unless the groove pitch were increased and the playing time thus proportionately decreased.

But, fortunately, the background noise level always occurs principally at frequencies far above the bass frequency band; and for that reason it is possible to suppress the background noise by greatly over-amplifying the recorded undulations in that frequency band which embraces the greater part of the background noise-the overamplified frequencies, including the background noise, being selectively attenuated during reproduction through the agency of suitable band filters. This serves to restore the over-amplified frequency band to the proper level while eifecting a corresponding diminution of the noise level. A record so made is, of course, distorted, so to speak (purposely so) and if played on a reproducing machine which is not designed to suppress the over-amplified frequency band it would give a more or less distorted reproduction. But when played on a suitable reproducer, it will give substantially noise free results together with a close approach to absolute fidelity throughout the entire recorded frequency band.

Closely related in its solution to the problem of suppressing background noise is the matter of compensating for what is called record blank compliance. The term compliance refers to the fact that all record blank materials are more or less yieldable to the reproducing stylus and that certain materials such as aluminum and aluminum alloys and cellulose acetate and nitrate out of which record blanks are often made are more yieldable than are the molded materials. This yieldability or compliance is significant at the higher frequencies because at the higher frequencies the record undulations are of extremely small amplitude; and if the walls of the groove are particularly compliant they will yield to the stylus considerably at the higher frequencies, instead of causing the stylus to oscillate with full amplitude.

The loss of amplitude attributable to record compliance can, obviously, be offset by exaggerating to an appropriate degree the amplitude of movement of the recording stylus at those frequencies at which record compliance is an important factor.

It is perhaps not within the realm of possibility in phonograph recording and reproduction to completely and perfectly compensate for the distortion arising from record compliance and at the same time eliminate completely all needle scratch noise; but this can be done to a degree which is completely satisfactory to the ear, and I have been able to, accomplish that result with.

the herein described recorder by adjusting its natural period-that is, by increasing or decreasing its natural frequency through the agency of adjusting means which are providedi'or that phrpose.

It has been common practice heretofore to mount the armatures of electromagnetic phonograph recording heads on pivotal supports of resilient material, such as rubber or one of the several known viscous synthetic compositions having more or less the characteristics of rubbersuch material functioning as a damping agent and being operative to suppress the natural frequency peaks of the mechanical vibrating system, whichv comprises the armature and recording stylus. But, as I shall endeavour to show hereinafter, the use of such pivots inhibits the recording of very high audio frequencies because at those frequencies the rubber or other similar damping material yields laterally and permits the armature and stylus assembly to oscillate as a whole about the cutting end of the stylus-instead of oscillating about the mounting pivot; and this means that the upper high frequency vibrations are not transmitted in any substantial degree to the cutting end of the stylus and, therefore, are not recorded to any substantial extent.

In other recording heads of the prior art the armatures are mounted on non-resilient pivots, and rubber or other viscous damping material is employed at some other point or points for the same purpose. This latter is an improvement over the resilient pivot practice, but still has its drawback in that the yieldability of the damping means varies with temperature and age and thus causes the characteristics of the recording head to change-a condition which the recording technician frequently is neither prepared to discover nor to cope with. The desirability of doing away with rubber and similar damping media in phonograph recording heads is something which is thcroughly well recognized by those who are skilled in the recording art. Rubber and similar damping materials were also employed in recorders of the prior art for restricting the amplitude of movement of the armature and recording stylus, so as o e en Overttin at ei is quencies.

The order; h a f the present invent on is a product of extendedresearch a d egeperimentation and is characterized in-that it is ,capable of recording t e ao d nar .zfiii th em tire band of useful audible frequencies from the lowest audible bass to the highest frequency readily discernible by the ordinary human ear; s ap l f being .s ad u te hat it w l .ove emphasize that band of frequencies which embraces the normal background noise (needle scratch) so as to enable the background noise to be suppressed during reproduction; a-nd is in: herently operative -to confine the oscillatory movements of the armature and stylus withinthe limits determined by the pitch of the record groove, but without at the same time cutting off the lowest audible frequencies which it is desir: able to record. 1

My invention can best be described by refer..- ence to the accompanying drawings which depict the preferred embodiment andwherein:

Fig. 1 is a front elevational view, partially in cross-section, on a greatly enlarged scale, of ,an assembled recording .head--the protectivexcover being removed;

Fig. 2 is a view looking at the bottom ofF-ig. 1-; Fig. 3 is a crossesectional. viewtaken along the line 3-3 of Fig. 1; 1

Fig. 4 is a view along the line 4?- l.of 1, portraying the under side of the armature and the armature-suspension plate;

Fig. 5 is a cross-sectional view taken alongethe justing the tension of the armature-suspensio plate. i

In the original drawings Figs. 1 to 7 ;ir eli sive e drawn to asca e o t v1.:1. 2m hich it w l 75 be self-evident that the actual structure is quite small. All parts have been drawn as nearly as possible to scale; but the dimensions of some parts are quite critical and for thatreason I shall specify certain important dimensions in order that my disclosure may be absolutely complete. It is not my intention to restrict myself to any particular dimensions'or to any particularmaterials, but it so happens, in this case, that the excellent results realized have been accomplished,

to a considerable extent, through the cut and try method, and that equally good results cannot, according to my experience, generally be achieved through haphazard selection of dimen'sions or materials.

In the structure illustrated, l is a rectangular mounting plate which is attached in any suitable manner to the carriage of a recording machine. In use it stands upright as shown in Fig. 1.

Secured to the mounting plate by means of screws 2 are a pair of softiron or steel polepieces 3 and 4, respectively.

The two poles of a permanent horse-shoe type magnet 5 rest, respectively, on pole- pieces 3 and 4 and are secured thereto solely by magnetic adherence.

Pole- pieces 3 and 4 have inwardly projecting opposed pole-tips 3a and 4a and downwardly ex-, tending arms 3b and 41). At 6 is a magnetic gap between the pole-tips, in which is centered the free end la. of an armature l-which armature formsthe magnetic core of an actuating coil 8 Said coil usually consists of a large number of turns of fine insulated copper wire. 7

To the lower ends of arms 3band 4b is attached, by means of screws 9, an assemblage comprising a relatively rigid rectangular bronze backing plate I0, a pair of thin rectangular spacers ll and H, a thin rectangular armature suspension plate I2 (see Figs. 4, 5 and 6) to which the armature T is securely attached, two sets of brass leaf springs 13 and M (see Fig. '7), each comprising a pair of spring fingers l3a, I31) and.

Ma, Mb, a rigid, rectangular bridge member I5, and four adjusting screws 16, each of which, individually, bears against the under side of :one of the four spring fingers. Interposed between the ends of the spring fingers and the adjacent surfaces of the armature-suspension plate [2 are Armature 1 and its suspension plate l2 are illustrated in detail in Figs. 4, 5 and 6, and it will be observed that the armature comprises a narrow upper end portion la, a vertical cylindrical body portion 1b and a horizontally extending hub portion '10. The armature has a vertical stylusreceiving recess lb and a tapped hole 'le for the accommodation of a stylus clamping screw It. The armature is attached to the suspension plate l2 by means of a rivet l9 and a U-shaped strap 20, the ends of which project through the suspension plate and are headed over.

Formed on the under side of backing plate H1 and in line, with the vertical axis of the armature are a pair of downwardly projecting teats 2| and 22 which serveas fulcrum points about which the armature rocks oscil ato fly .when a 'c tuated. 'The four spring fingers-are pressed up- 3 wardly by adjusting screws IS with suflicient force to hold the armature-suspension plate securely against the teats 2| and 22, but the spring fingers have sufficient flexibility to permit them to yield slightly as the armature oscillates about its fulcrum. v A recording stylusZii is inserted in armature recess 1d and firmly clamped in place by means of clamping screw l8.

The assemblage comprising parts ill to [5 in: elusive is soldered together at its two extremities so that it can be removed as a whole from the pole-pieceatogether with the armature, without disturbing the adjustment thereof; and the soldering operation performs an additional useful function in that the applied heat causes the armature-suspension plate 2 to expand more (due to its smaller bulk) than the backing plate l0 and bridge member l5 and, as a consequence, the armature-suspension plate is put under tension when the assemblage cools off-and that is a highly desirable condition. Since the abovementioned assemblage, when soldered, forms a' self-contained unit which can be removed without disturbing its adjustment, it is a simple mat- 0W ter to remove andreplace the coil 8. g

The four spring fingers and their'adjusting screws it perform two distinct useful func tions. Firstly, they facilitate centering the armatur in the magnetic gap; and, secondly, they permit the natural period of the vibratory system' to be raised and lowered. Tightening up on the adjusting screws increases the tension on the armature-suspension plate'and thusincreases the natural frequency of the system. slacking off onthe adjusting screws has the opposite effect. It so happens that record background noise (needle scratch) occurs most pronouncedly in a band of frequencies which peaks in the neighbor'- hood of 3000 to 4000 cycles per second, varying somewhat with different record materialsjand it is very desirable that the response curve of the recording head be peaked in the vicinity of the background noise peak so that the recorded sound will greatly exceed in amplitude the noise level-the distortion thus produced being easily compensatable during reproduction, screws l6 and the spring fingers permit the recording head to be adjusted so as to produce a response peak at the desired point in the frequency band. v

As previously stated, certain record blank materials-particularly aluminum and aluminum alloys and cellulose acetate and nitrate-are characterized by relatively high compliance; which means that in order to obtain proportionate reproduction intensity at the higher frequencies, those frequencies must be overemphasized in the recording or else inordinately amplified during reproduction. -By peaking the response curve of the recording head somewhat higher than the scratch frequency peak it is possible to obtain both a desirable over-emphasis of the very high frequencies and a satisfactory amplification of the frequency band in which the major part of the needle scratch occurs. Generally speaking, when using a recorder in accordance with the present invention on record blanks having high compliance, better over-all results may be realized by adjusting it to a somewhat higher natural frequency than the needle scratch frequency peak. But in records of low compliance the ultimate over-allresult may be somewhat better. if the recorder isadjusted fora somewhat lower natural frequency peak.

Adjusting In operation coil 8 is ordinarily connected to the output terminals of a power amplifier from which it is energized by sound-modulated current. In the absence of current flowing through coil 8 the armature is magnetically neutral, or substantially so, because it is centered mid-way between the pole-tips. When the coil is energized by an alternating modulated current, a correspondingly modulated alternating magnetic flux is set up in the armature and the upper end la thereof is attracted first toward one pole-tip and then the other-depending upon the direction of flow of current through the actuating coil. Suspension-plate i2 is made of a special magnetic alloy which will be discussed more at length hereinafter, and it serves contributorily to complete the magnetic path for the armature flux.

The armature oscillates vibratorily in a single plane about its pivotal center and, of course, the cutting end of the recording stylus which engages the record blank oscillates correspondingly.

There are, inter alia, three things which must be accomplished in order to realize an approximation to the theoretically maximum attainable quality in commercial phonograph recording and reproduction, and I shall allude to these briefly and then proceed to point out the features of my recording head whereby I have accomplished my objectives. One of the three things to which I refer is the suppression of needle scratch and, incidentally, compensation for record compliance, and I have already made it quite clear that I accomplish these by adjusting the resonance period of my recorder either to coincide with the scratch frequency peak and thus over-emphasize the recorded undulations in the scratch frequency band, or I adjust the resonance period to a peak frequency somewhat higher than the scratch frequency peak and thus effect a compromise between suppression of needle scratch and compensation for record compliance.

Two factors remain to be considered and these are: (1) the recording, reproduceably, of all frequencies within the useful audible range and (2) the proper suppression or attenuation of the lowest frequencies, so that the amplitude of movement of the cutting end of the stylus will never become so large as to cut through the walls which separate adjacent portions of the record groove.

To be able to respond satisfactorily to very high frequencies the armature and stylus of a recorder must ordinarily have very little inertia and the armature magnetic path must be magnetically efficient. That is to say, the armature path must be characterized by low reluctance and low eddy current losses under the conditions to be encountered.

The armature l is of exceedingly small mass and low inertia and is, for that reason, inherently capable of responding readily to high frequency energization without requiring an excessive driving force to overcome inertia; but if the magnetic actuating force at high frequencies is ,unduly dissipated or otherwise reduced eXces sively there will not be sufficient high frequency response notwithstanding the low inertia of the mechanical oscillatory system. The high frequency energizing force must in some way be either conserved or amplified, if superior recordings are to be had, and I conserve this force by providing a very efficient magnetic bath.

Numerous tests were made to determine the most suitable magnetic materials available for ,the various parts of the magnetic circuit and it was found, strangely enough, that for the armature ordinary soft iron gives the best results. But for the armature-suspension plate [2 through which the magnetic circuit is completed I use a magnetic alloy which is most commonly known to the public as Permalloy, having the characteristic of high permeability for very small magnetizing forces, accompanied by a very low magnetic saturation point. This material is eminently satisfactory for the instant purpose because it provides a path of low reluctance for the high frequency magnetizing force-the magnitude of which force is small by comparison and never rises sufiiciently to exceed the magnetic saturation of the paths through the suspension plate. In that way I obtain a very high magnetic efficiency at high frequencies and at the same time I accomplish another very important result which I shall nowproceed to explain.

It is a simple enough matter to make the armature magnetic path eflicient at high frequency, or any frequency for that matter, and I would not need to resort to the use of any special magnetic alloy having peculiar characteristics to accomplish that object alone. I have said that the special magnetic alloy which I employ is characterized not only by low reluctance at low magnetization, but also by a very low saturation point. So long as the magnetizing force dealt with never rises to a sufficient magnitude to saturate the special magnetic alloy path, said path is very efiicientits permeability is high and reluctance low; and that is the condition which prevails at all frequencies here dealt with except the bass frequencies. As the frequency diminishes the magnetic lines of force traversing the armature increase in numberan inherent condition-and if the reluctance of the armature magnetic path remained constant the oscillatory movement of the armature and stylus would be excessive in the bass frequency range. The stylus would out too Wide and break through the walls separating adjacent portions of the record groove. And this cannot be avoided by the use of any means in the nature of a mechanical stop operating to limit abruptly the travel of the armature or stylus because any such provision would produce very objectionable noise effects.

The usual practice heretofore has been to use damping means in the form of rubber or some viscous material which would operate to resist the oscillatory motion of the armature and stylus with increasing vigor as the amplitude of movement tended to increasethus functioning as a .yieldable stop to prevent over-cutting. Such damping means work reasonably well but, to the best of my knowledge, no recording hea d provided with such damping means has ever been devised which would give the excellent results I have been able to achieve at bass frequencies and at the same time record satisfactorily the entire range of useful audible frequencies.

It should not be difficult to apprehend that there are two possible waysand two only,of limiting the amplitude of movement of an armature, without utilizing abrupt mechanical stops-- which are out of the question in any recorder. One way is to interpose yieldable resistance to the armature movement-which resistance increases concurrently with the applied forcewhich is the familiar damping means above men.- tioned. The alternative is to limit in some apiii) propriate manner the-actuating forcewhich can be made effective upon the armature. And the lattert is the method which-the present invention follows. The armature suspension plate [2, being made of special magnetic alloy having a low saturation point, is so proportioned, as to cross section, that it becomes magnetically saturated and thus definitely limits the magnetic actuating force which can be applied to the armature. In this way the magnetic force, which normally reaches its maximum at the lowest frequency, is definitely restricted so that the oscillatory movement of the armature cannot exceed at low frequency, or, for that matter, any frequency, the permissible amplitude.

The armature-suspension plate 12 is shown in detail in Figs. 4, 5 and 6 and, as will be seen, is rectangular in form, being cut out so as to form twohorizontally extending strips in and 12b,

and a cross piece I20 interconnecting I2a and there then remains only the long highly reluctant space path over which the excess of magnetic lines of force can complete their circuits. Hence, the excess magnetization has extremely little influence by way of increasing the force applied to the armature. i

As a matter of fact, I have not actually determined whether the armature-suspension plate l2 becomes saturated only when the magnetic lines of force are added thereto by the actuating coil 8 or, alternatively, said plate is normally.

saturated by the permanent magnetic field. The same results would, hypothetically, seem to follow whether the suspension plate is fully saturated normally or under-saturated normally, provided the suspension plate is so proportioned as to appropriately limit the=change in magnetization therethrough. Having achieved the results I was after, it was unnecessary for me to determine whether or not the suspension plate is normally saturated by the permanent magnet, but I will say that there is respectable opinion to the efiect that such is the case.

The armature-suspension plate is tensioned lengthwise, as previously explained, and firmly secured at both ends. The armature is so securely attached thereto by rivet l9 and strap 20 that no appreciable displacement of the armature can occur laterally in any direction; and the long hub portion 1c of the armature, extending all the way across the suspension plate practically precludes any rotational movement of the armature otherwise than in the plane in which it is intended to oscillate. It follows then that in ordinary use the only possible appreciable movement of the armature is the intended rotational movementabout its fulcrum points and. that being so, substantially all the movement imparted to the armature will be transmitted to the cutting end of the stylus in the proper plane;

" and this is especially important at high freots of rubber orother resilientmateriaL -The rubber is capable of yielding laterally to such an extent that the armature oscillates about the cutting end of the stylus and to a diminished extent or not at all about its own pivotal center; and that accounts for the seemingly innate lack of ability of recorders so constructed to respond to very high audible frequencies.

In the instant case a knowledge of the underlying principles of the invention is not enough, I have found, to enable a man skilled in the art to design and construct, without exper'imentation, a thoroughly successful recorderunless by mere good fortune he should hit upon optimum proportions at the outset.

The recorder here illustrated is a product of protracted trial and elimination as well as a product of the. application of my underlying inventive ideas; and since I am unable .to lay down any preciseformulaby means of which an engineer could, with assurance and without ex- Suspension plate I2 is made of a special magnetic alloy called Alleghany Electric Metal,v

manufactured by Alleghany Steel Corporation. This alloy is the kind more widely known as Permalloyf? The suspension plate is .012 inch thick and the dimensions 1-! and .J are /8thinch each.

Armature I is made of ordinary Norway iron; and its diameter .K is .110 inch; dimension L.

is inch; M is inch and N is .O l inch. Coil 8 comprises 375 turns of-No. 35 single silk, enameled copper wire. 1

Dimensions not given are not critical; nor are all the dimensions given necessarily critical. All

figures are drawn to the same scale and dimensions not stated above may be determined closelyenough with a pair of proportional dividers.

What is claimed is: I I 1. In a phonograph recorder, a magnetic field structure including a magnet and a pair of spaced pole-tips of oppositemagnetic polarity, an armature and a flexible armature-suspension plate,

said plate being made of a magnetic alloy characterized by high permeability at low magnetization and a relatively low saturation point, said plate constituting a magnetic path between the poles of said magnet, said armature being se cured to the mid-portionof said armature-suspension-plate and projecting transversely therefrom in a direction normal to the plane of said plate, said armature being disposed in the gap between said pole-tips at a point removed from saidplate, fulcrum structure for said armature,

and a coil for magnetically energizing said arma ture. j

2. 'In' a phonograph recorder, a magnet, struca ture defining a pair of spaced opposed pole-tipsof opposite magnetic polarity and associated individually with the poles of said magnet, an armature, an actuating coil surrounding said armature, a flexible armature-suspension plate, said armature being secured to the mid-portion of said plate and projecting transversely therefrom in a direction normal to the plane of said plate, said armature being disposed between said poletips at a point removed from said plate, fulcrum means for said armature, said fulcrum means being operative to oppose lengthwise movement of said armature in one direction, and yieldable means pressing laterally against one side of said plate at either side of said fulcrum means.

3. In a phonograph recorder, a magnet, structure defining a pair of spaced opposed pole-tips of opposite magnetic polarity and associated individually with the poles of said magnet, an armature, an actuating coil surrounding said armature, a flexible armature-suspension plate, said armature being secured to the mid-portion of said plate and projecting transversely therefrom I in a direction normal to the plane of said plate,

said armature being disposed between said poletips at a point removed from said plate, fulcrum means for said armature, yieldable means pressing laterally against one side of said plate at either side of said fulcrum means, and adjusting screws for varying the pressure of said yieldable means against said plate.

4. In a phonograph recorder, a magnet, a pair of pole-pieces, one for each pole of said magnet individually, said pole-pieces each including a pole-tip, said pole-tips beingin spaced opposed relation, there being a magnetic gap therebetween, a I flexible armature-suspension plate bridged between and secured to said pole-pieces, said plate being made of a magnetic alloy which is characterized by high permeability at low magnetization and a low saturation point, an armature secured to the mid-portion of said plate and projecting transversely therefrom in a direction normal to the plane of said plate, said armature being disposed in said magnetic gap at a point removed from said plate, an actuating coil surrounding said armature and disposed between said pole-tips and said plate, springs bearing upwardly against the under side of said plate at either side of the pivotal center of said armature, and adjusting screws for varying the pressure exerted by said springs against said plate.

5. In a phonograph recorder, a magnet, a pair of pole-pieces, one for each pole of said magnet individually, said pole-pieces each including a pole-tip, said pole-tips being in spaced opposed relation, there being a magnetic gap therebetween, a backing plate bridged between said pole pieces and having one or moredownwardly projecting fulcrum points in vertical alignment with the center of said magnetic gap, a thin armaturesuspension plate bridged between said pole-pieces and situated below said backing plate and parallel thereto, said armature-suspension plate being spaced from said backing plate but in contact with said fulcrum? point or points, an armature secured to the mid-portion of said armature-suspension plate and projecting transversely therefrom in a direction normal to the plane of said plate, a portion of said armature remote from said armature-suspension plate being disposed in said magnetic gap, an actuating coil surrounding said armature and disposed between said poletips and said armature-suspension plate, a. plurality of spring fingers situated below and hearing' against the under side of said armature-suspension plate, there being at least one such spring finger at either side of said armature, a bridge member situated below said spring fingers, and a plurality of adjusting screws carried by said bridge member and bearing against said spring fingers individually.

6. A phonograph recorder in accordance with claim 5 wherein the armature-suspension plate is cut out to form two parallel horizontally extendingstrips, there being four of said spring fingerstwo at either side of said armature -two of said spring'fingers bearing against one of said strips and the other two spring fingers bearing against the other of said strips, the pressure of said spring fingers against said strips being individually adjustable by said screws.

7. In a phonograph recorder, a magnet, a pair of pole-pieces, one for each pole of said magnet, each of said pole-pieces comprising a pole-tip, said pole-tips being situated in opposing relation with a magnetic air-gap therebetween, each of said pole-pieces also comprising a downwardly extending arm, an actuating coil disposed between said arms and having an axial opening substantially in alignment with said magnetic air-gap, and a self-contained assemblage bridged betwen said arms and including a backing plate, an armature-suspension plate situated below and spaced from said backing'plate, spring fingers below and bearing against the lower surface of said armature-suspension plate, a bridge member situated below said springfingers, adjusting screws for said spring fingers carried by said bridge member, and an armature mounted on and secured to the mid-portion of said armature-sus pension plate, said armature projecting upwardly" through said coil and having its upper end disposed in said magnetic air-gap, said assemblage being soldered together at its two ends so thatit can be detached from said arms without losing its adjustment.

8. A phonograph recorder in accordance with claim '7 wherein the armature has a hub portion extending laterally across said armature-suspension plate.

9. A phonograph recorder in accordance with claim 7 including a mounting plate to which said pole-pieces are attached and by'which they are held in spaced relation.

10. In a phonograph recorder, magnetic field structure including a pair of magnet poles of opposite magnetic polarity, a pair of spaced pole-tips connected respectively to said poles, a flexible armature-suspension plate bridged between said poles, said plate being made of a magnetic alloy,

characterized by high permeability at low magnetization and a relatively low saturation point, an armature secured to the mid-portion of said plate and projecting transversely therefrom in a JAMES R. FOUCH.

Images