WO1983003918A1

WO1983003918A1 - Magnetic head having highly saturable gap liner

Info

Publication number: WO1983003918A1
Application number: PCT/US1983/000260
Authority: WO
Inventors: Richard J. Mcclure; John F. Bagby; Frederick J. Jeffers
Original assignee: Eastman Kodak Company
Priority date: 1982-05-04
Filing date: 1983-02-25
Publication date: 1983-11-10
Also published as: EP0108074A4; EP0108074A1

Abstract

The general concept of the invention is to coat, at least, the trailing pole face of a pair of ferrite pole faces (26), which define a transducer gap, with high-saturation material (30) such as Sendust. By so structuring a head with a gap liner, not only is the critical part of its gap defined by material having a high saturation magnetization -- and thus such head will no be easily subject to critical pole tip saturation of its trailing pole face -- but also (1) the long-wearing ferrite will desirably constitute the principal head part that slidingly coacts with a recording medium, (2) no high reluctance glue line will exist between a core and its tips, and (3) the signal flux will enter the gap region uniformly over the surface of the gap, thereby providing enhanced high frequency efficiency.

Description

MAGNETIC HEAD HAVING HIGHLY SATURABLE GAP LINER

This invention relates in general to magnetic heads and to their methods of manufacture. More particularly, though, the invention is directed to magnetic heads which are comprised of gapped ferrite cores.

The invention as well as the prior art will be discussed in relation to the figures wherein:

Fig. 1 is a diagram useful in describing a prior art problem solved by means of the invention, Fig. 2 is a side elevation view of a prior art head over which the invention provides improvement,

Fig. 3 is a not-to-scale diagram useful in describing the invention, Figs. 4-6 illustrate various procedures employed in producing a magnetic head incorporating the invention, and

Fig. 7 is a diagram useful in describing the invention. U.S. Patent No. 4,302,790 issued to

James U. Le ke and assigned to the assignee hereof, discloses a new and improved technique for effecting magnetic recording. Basically, the concept disclosed in U.S. Patent No. 4,302,790 is to effect recording in a magnetic medium by means of a magnetic record head having an extremely short transducer gap, e.g. one that is on the order of 0.4 microns or less. This teaching is in direct contrast with the widely practiced prior art technique of using record heads having relatively long gap lengths which are typically about 1.25 or more microns. Such prior practice is predicated on the notion that a relatively long record gap will effect recording deep into the recording medium by virtue of a large bulbous fringe field (and thus be productive of a strong playback signal) .

Further, because the trailing edge of the record gap

OMPI is, in fact, the particular head structure which last coacts with the record zone(s) of the medium, it is widely accepted that the definition of the record zone(s) will be pronounced so long as the record gap trailing edge is sharply defined (i.e. the record zone(s) will be easily discernible during playback) . By contrast, the teaching of U.S. Patent No. 4,302,790 has been found to provide not only as good, but better, playback of recorded signals, despite the fact that the depth of recording is only about as great as the length of the record head gap. One explanation for this playback improvement is that, with the recording practice of U.S. Patent No. 4,302,790, there is little or no overlapping of (arcuate) record zones which algebraically add to knockdown the playback signal, the record zones being (resolution-wise) extremely narrow and sharp because of the extremely short transducer gap employed to effect recording.

Whether with respect to recording practices as taught in U.S. Patent No. 4,302,790 or with respect to the prior art practice of recording with a long record gap, there is a trend toward the use of ferrite material to form magnetic recording heads. Ferrites exhibit many properties which are highly desirable in magnetic heads: ferrites are hard and wear well; they have high resistivity and thus tend to inhibit eddy current losses at high recording frequencies; and they have relatively high permeability. Although the saturation magnetization of ferrite material (typically, ferrite saturation occurs at about 4000 gauss) is considerably less than that of other head materials such as ternary alloys of aluminum, iron and silicon (which are often referred to as Sendust; and typically saturate at about 10,000 gauss), ferrite material nonetheless has found wide usage in connection with magnetic recording heads, especially - -

headε having conventional, relatively long, transducer gaps. Ferrite use, however, in connection with recording practices as taught in U.S. Patent No. 4,302,790 and/or when employed with so-called high energy recording media having coercivities of around 850 oersteds is severely limited as will be discussed below. Before discussing such problem, though, it must be pointed out that the parameters which are identified are merely representa- tive ones, being characteristic of head material, recording media, and dimensions which the assignee hereof has employed in the course of its developments.

Fig. 1 is a widely accepted showing of the field strength distribution in the vicinity of the gap of a recording head. Such field strength distribution has been discussed by C. D. Mee, The Physics of Magnetic Recording, Vol. II, 1964, page 40, and by C. B. Pear, Magnetic Recording in Science and Industry, 1967, page 32, the field H identified in Fig. 1 being the field within the depicted record head gap g, and the field H being that at various distances from such head gap.

A high energy recording medium which the assignee hereof has employed has a typical coercivity of 850 oersteds, a saturation magnetization of 1600 gauss, a permeability of about two, a nominal 90 percent of saturation remanence that occurs for an applied field H of about 1375 oersteds, and a nominal 10 percent of saturation remanence that occurs for an applied field of about 500 oersteds.

As noted above, the depth y of recording in a medium, when recording as taught in U.S. Patent No. 4,302,790, approximates the lengthwise dimension of the record gap g. Thus, the depth y within the recording medium at which the aforesaid 1375 oersteds of applied field appears equals the length of the record gap g.

It is well known (0. Karlquist, Transactions of the Royal Institute of Technology, Vol. 86, page 3, 1954) that the ratio of the field Ho within a record gap to the field H at various distances from the gap is related by the expression

*o _β __* (1)

^ 2 tan^"1 g

which, for y_ 1.0.5, may be approximated by S

^o 2.95 y_ +0.44 . (2)

H g

Given, as noted above, that y=g, and that H = 1375 oersteds, the requisite field H (without bias) within the record gap to effect 90% saturation remanence is ^Ho ^{= (2*95 + 0,44) 1375 = 4661 oersteds}- (3) It will be appreciated that the particular determinant concerning the resolution of a record zone within a record medium is the field gradient near the trailing edge (g¹, Fig. 1) of the record gap in question ... and such field gradient is directly dependent on the permeability of the core material that defines the trailing edge of the gap. Were such edge g' to saturate during recording (a problem known as pole tip saturation), whereby the permeability of the edge-defining core material became equal to one, the effective gap would not only enlarge, but its dimensions would become fuzzy, i.e. the recording field gradient would become slight. The relationship between core permeability and field gradient has been discussed both by T. Suzuki et al, IEEE Transactions on Magnetics, MAG-8, September 1972, page 536, and by

H. Bertram et al, IEEE Transactions on Magnetics,

OMPI

^ MAG-12, November 1976, page 702; and both conclude (1) that the permeability vε field gradient relationship is flat for permeabilities greater than 50, and decreases for permeabilities less than about 50, and 5 (2) the permeability of gap-defining core material approaches 50 when the field H within the gap is about half the saturation flux density 0 of such material. That is, when

°s ⁽⁴

10 ~T^~ " ^H ₀ *

Since, in order to effect recording in the afore¬ mentioned recording medium pursuant to the teaching of

U.S. Patent No. 4, *302,*790, a field Ho = 4661

, c oersteds would be necessary, it is quite clear that a gapped core of ferrite material -- with its inherent saturation magnetization of only 4000 gauss (see above) -- cannot effect such recording. That is,

(2 x 4661) gauss >> 4000 gauss. (5)

20 Corollarly, it is just as clear that a gapped core of material such as Sendust -- with its inherent saturation magnetization of 10,000 gauss -- could, most definitely, effect the described recording. That is,

25 (2 x 4661) gauss < 10,000 gauss. (6) Recognizing that Sendust (and similar such materials) , with its saturation magnetization of about 10,000 gauss could be employed to effect recording pursuant to U.S. Patent No. 4,302,790, the prior art

30 (as exemplified by U.S. Patent Application 184,553 in the name of D. F. Cullum et al and assigned to the assignee hereof) has taught the application of Sendust pole pieces to an otherwise all-ferrite core. See Fig. 2 which shows a magnetic head comprised of a «₅ ferrite core 10 having a non-magnetic transducer gap 12 defined by Sendust pole pieces 14 bonded at 16 to the ferrite core 10. As will be appreciated, such a magnetic head constitutes a compromise at best: relatively high permeability and eddy-current limiting is provided by the ferrite core 10; and high satura¬ tion magnetization is provided by the Sendust pole pieces 14 at relatively little sacrifice in overall permeability.

Some disadvantages, however, of the recording head depicted in Fig. 2 are as follows:

(a) Since Sendust is not as hard as, and doesn't wear as well as, ferrite material, a head as depicted in Fig. 2 will wear out more quickly than one in which the tape contacting surface of the head is ferrite material. Thus, the pole tips of the Fig. 2 head will have to be replaced when worn down, or the head will have to be discarded.

(b) Since the Sendust pole tips 14 are glued to the ferrite core 10, the glue line causes the reluctance of the magnetic circuit comprising the core 10 and tips 14 to be greater than would otherwise be desired.

(c) Because of eddy currents induced in the Sendust pole tips, the recording flux must enter the gap region at the bottom of the gap at 17 in Fig. 2. With eddy currents in the gap region, signal flux is confined to a thin layer on either side of the gap. This results in considerable impedance to the flow of flux at high frequencies; and, therefore, the record efficiency of the head of Fig. 2 decreases dra¬ matically as the signal frequency to be recorded is increased.

Rather than apply a Sendust topping (i.e. tips) to a ferrite core to prevent pole tip satura¬ tion, the general concept of the invention is to coat at least the trailing pole face of a pair of ferrite pole faces, which define a transducer gap, with high-saturation material such as Sendust. By so structuring a head with a gap liner, not only is its gap defined by material having a high saturation magnetization — and thus such head will not be easily subject to pole tip saturation — but also (1) the long-wearing ferrite will desirably constitute the principal head part that slidingly coacts with a recording medium, (2) the high reluctance glue line will be obviated, and (3) the signal flux will enter the gap region uniformly over the surface of the gap, thereby providing enhanced high frequency efficiency. (An additional advantage provided by the invention, which is apart from any of the above considerations, is addressed below under the title Ferrite Integrity.) In accordance with the invention, the Sendust coating on the ferrite pole face is produced, prefer¬ ably, by sputtering. The sputtering of Sendust onto ferrite has been demonstrated in the literature. See, for example, IEE Conference Publication [33], 1967, pp. 213-216, "Sputtering of Ternary Magnetic Films and Their Use in the Manufacture of Magnetic Recording Heads," by V. W. Vodicka, and IEEE Transactions on Magnetics, Vol. MAG-13, No. 4, July 1977, p. 1029, "Preparation by Sputtering of Thick Sendust Film Suited for Recording Head Core," by H. Shibaya et al, the former pertaining to the use of a Sendust medium- contacting topping on a ferrite core, and the latter pertaining to the fabrication of a thin film head structure per se. While no claim is made herein as to the details of the sputtering technique employed in the practice of the invention (indeed any suitable deposition technique will do), a pre-sputtering pro¬ cedure involving the removal of Beilby layers from the ferrite pole faces provides improved performance in magnetic recording heads according to the invention. This will be discussed below. Reference should now be had to Fig. 3 which is similar to Fig. 1 but shows the existence of (sputtered) coatings 18 of Sendust on both ferrite pole parts 20. Preferably, the coatings are between about 0.25 and 2.0 microns thick, the showing of Fig. 3 being clearly not to scale. Notwithstanding this intentional mis-showing (for ease of comparison with Fig. 1), however, it will be readily apparent that all of the above discussed advantages associated with ferrite material obtain, without the disadvantage of pole tip (18') saturation. That is, assuming the existence of a flux density of 2 x 4661 ^β 9322 gauss at the Sendust-defiήing pole tips 18' (thereby to effect recording in the high energy medium in question), such pole tips, which require 10,000 gauss for saturation, will remain effective and unsaturated.

In making a ferrite head embodying the invention (see Figs. 4, 5), the prior art practices of forming a rectangularly shaped ferrite half-bar, chamfering it to form a half-window 25, and lapping the chamfered bar to provide a precisely defined pole face part 26 are employed. Rather than deposit high- saturation material directly on the lapped down pole face part 26, however, the invention (preferably) calls for the removal of the Beilby layer that exists as a result of the lapping operation. As is known, cold-working, e.g. lapping, ferrite material has the effect of causing a stress induced permeability change at the surface of the material being worked, such sur- face change constituting a relatively high reluctance (Beilby) layer. Were Sendust (or the like) to be deposited directly atop the Beilby layer, a high reluctance gap would exist between the Sendust and ferrite, and such gap would cause (in most applica- tions of the invention) undeεired recording of ghost signals. The Beilby layer is removed, for example, by ion milling away several microns of the ferrite pole face part prior to the sputtering procedure.

Having prepared the ferrite as indicated above, the chamfered half-bar is then provided with a deposition 30 of Sendust (or the like) . Such a coating may be produced by a sputtering set-up as depicted in Fig. 6, wherein a target of Sendust is asεaulted by the ions of a plasma discharge, thereby causing Sendust to sputter onto the milled pole face part 26. Care must be taken that the Sendust target is devoid of an oxide coating since such may cause a "ghost-producing" gap just like the above noted "Beilby-produced" gap. Since Sendust may have lower permeability than the indicated ferrite material, it may be desirable to prevent a coating of Sendust from appearing at the back gap region 34 of the half-bar. This precaution aεεureε that, when two half-bars are brought together for purposes of forming a structure that is diceable into gapped cores, there won't be any Sendust in the back gaps of such diced cores which would render their overall reluctance higher than would otherwise be. Attendantly, a poεitionable mask 40 is provided as part of the sputtering set-up and, when properly positioned, the mask 40 will limit the deposition of Sendust to the pole tip region of the half-bar being coated.

Having coated the half-bar pole face part 26 with Sendust, an extremely thin (about 0.25 microns) coating of a non-magnetic gap spacer material 36, e.g. SiO, is then deposited on the Sendust, this procedure being commonplace in the prior art. Thereafter, as shown in Fig. 5, a second, similarly produced, Sendust-coated chamfered half-bar 37 (with or without a non-magnetic gap spacer coated thereon) is brought into face-to-face relationship with the first half-bar (39). Then, with the two half-bars so aligned, they are placed within a clamp (schematically shown at 40) and bonded together, for example, by epoxy or glass, with the non-magnetic gap spacer(s) 36 therebetween. (It will be noted that despite the presence of coatings of Sendust on the ferrite half-bars, glass bonding can indeed take place.)

After bonding, the unitary aεεembly of half- barε iε contoured; and then diced into individual tranεducer coreε, each having a Sendust liner in its transducer gap.

Notwithεtanding the indicated efforts to preclude a ghost-producing gap, a vestigal gap almost always remains; and it is to this characteristic that the above-noted definition of the Sendust layer thick- ness as being between 0.25 and 2.0 microns is addressed:

Using the same parameters noted above, con¬ sider the minimal effect "10% of saturation remanence" that occurs while using a gap length of, εay, 0.37 microns. See Fig. 7. Per equation 2, the distance y from the gap at which the "10% of saturation remanence" contour occurε iε about 1.13 microns, i.e.

H ⁼ ?95 ⁽ΪT - °-⁴⁴> ⁼ 79T <^50ϋ ' °-⁴^)⁼¹-13 micronε (7)

From Fig. 7, this would suggeεt that --in thiε example— any highly saturable layer greater than about one micron thick would have essentially no dis- torting influence on the recording field. For a variety of reasons, however, it is desirable to make the high-saturation layer as thin as possible. Among these reasons are the following:

(a) Interfacial stress between the deposited layer and the ferrite pole face will be less for a thin layer than for a thick one. Therefore,

^0RE ^

OMPI εtresε-produced magnetostriction in the thinly- coated ferrite --which reduces permeability-- will be minimal.

(b) Washout of the layer because of tape wear will be minimal, thereby minimizing spacing loss in the event the head in question is also used for playback purposes.

(c) Aside from taking less time to deposit, a thin layer —from the standpoint of psuedo- epitaxial growth— will produce a straighter gap edge, thereby providing sharper definition for the record zones produced by the head in question.

(d) A thin layer locates the vestigal gap in close proximity to the main record gap. Assuming the head in question is, again, to be used for both record and playack, both the main and vestigal gaps will see, depending both on the spacing between such gaps and the wavelength of the recorded signal, essentially the same signal flux. Widely spaced main and veεtigal gaps (thick layer) will see the same flux-sense for long wave¬ length recorded signalε, whereas closely spaced main and vestigal gaps (thin layer) will see the same flux-sense for short wavelength recorded signals. Thus, by using an extremely thin pole face layer, any "bumps" that may occur in the reproduce curve by virtue of the coaction of the two gaps in queεtion will be located well-out in the higher frequency region of such curve. Given the conflicting requirements for layer thickness, it would seem that a layer from somewhere between about 0.25 and 2.0 microns would be a reason¬ able compromise for layer thickness.

Ferrite Integrity U.S. Patent 4,017,899, issued to J. F. Bagby and asεigned to the assignee hereof, discusses at conεiderable length the vulnerability of ferrite heads to crumbling in the vicinity of their core gaps. This is due to the relative brittleneεε of ferrite materials. Whereaε U.S. Patent '899 concerned, in particular, the εide crumbling of gapped ferrite cores, the crumbling of ferrite material into a gap is an even more critical problem. By virtue of the invention, it has been found that the application of a coating of Sendust to the pole faces of ferrite cores has the additional benefit of eliminating gap erosion, albeit that the primary purpose of the Sendust is to prevent core pole tip saturation.

The invention has been described in detail with particular reference to preferred embodiments thereof, but it will be understood that variations and modifications can be effected within the spirit and scope of the invention. While throughout the above specification, reference was continually made to a material known as Sendust, it will be appreciated that whenever pole face parts, comprised of a first material having a saturation magnetization of a first given amount, are coated with a magnetic material, e.g. permalloy, of a εecond greater saturation mag¬ netization, the concern for pole tip saturation will be lessened in a head formed of such pole parts.

WIPO

Claims

WHAT IS CLAIMED IS:

1. A magnetic head of the type having a transducer gap defined by the faces of first and second pole parts formed of magnetic material having a given εaturable level of magnetization, said transducer gap having an effective magnetic gap length of lesε than about 0.4 microns, characterized in that gap liner magnetic material, having a) a thickness of between about 0.25 and 2.0 microns and b) a saturable level of magnetization greater than said first saturable level of magnetization, is intimately bonded, within said gap, to at least one pole part face, whereby the saturability of said magnetic head in the vicinity of said gap is principally determined by the saturability of said gap liner material.

2. The^" magnetic head of Claim 1 wherein said magnetic material of said pole parts is comprised of ferrite material, and wherein said gap liner material is comprised of Sendust.

3. The magnetic head of Claim 2 wherein said gap liner Sendust material iε compriεed of reεpective films thereof deposited on the faces of said pole parts.

4. The magnetic head of Claim 3 wherein said first and second pole partε are glasε bonded together in the vicinity of εaid tranεducer gap.

5. The magnetic head of Claim 3 wherein both said films of Senduεt material on εaid pole part faceε are reεpectively between 0.25 and 2.0 micronε thick.

6. The method of making a magnetic head characterized by the steps of:

(a) lapping flat a pair of ferrite pole parts to provide a pair of planar faces thereto,

(b) removing the Beilby layers which reεpectively exist on the planar faces, (c) bonding respective layerε of magnetic material having a saturability level greater than said ferrite to said planar faces after said Beilby layers are removed, εaid layer being between 0.25 and 2.0 microns thick and

(d) aligning said ferrite pole parts with their reεpective layerε of magnetic material in face-to-face relation with a non-magnetic εpacer therebetween.

7. The method of Claim 11 wherein said bonding of layers of magnetic material to εaid planar faces is effected by depoεiting reεpective filmε of Sendust on said planar faces.

8. The method of Claim 12 wherein the deposition of Sendust is effected by sputtering

Sendust from a target thereof onto said planar faces.

9. The method of Claim 13 including the additional step of removing oxide from εaid Senduεt target prior to εaid εputtering.

10. The method of Claim 11 wherein εaid

Beilby layerε are removed by ion milling.