DE1266811B - Signal transmission arrangement - Google Patents

Description

FEDERAL REPUBLIC OF GERMANY

GERMAN

PATENT OFFICE

EDITORIAL

Int.

GlIc

German class: 21 al - 37/06

Number: 1266811

File number: J 22545IX c / 21 al

Filing date: October 24, 1962

Opening day: April 25, 1968

The invention relates to a signal transmission arrangement comprising a coupling ribbon line with a width that changes over its length, a controlling and a controlled magnetic layer arrangement, which are interconnected by the coupling ribbon line, and deflection means for the temporary deflection of the magnetizations of the magnetic layer arrangements.

It is known, magnetic layer elements with uniaxial magnetic anisotropy by means of band-shaped Coupling lines to be coupled with each other and binary information from a first (controlling) To transfer magnetic layer element to a downstream second (controlled) magnetic layer element. In this way, for. B. Shift registers or logic circuits are set up (Bull. SEV, Vol. 51, October 8, 1960, pp. 1004 to 1010). In the same context it is also known the retroactive effect to reduce a controlling element to an upstream element in that the transmission drive field generating driveline with the easy axes of the magnetic layer elements a small Angle and forms an angle of 45 ° with the coupling lines. It becomes one that way preferred transmission direction defined.

From the technology of magnetic core circuits it is known that when the signal is transmitted from a first On a downstream second magnetic core, certain dimensioning regulations and adaptation conditions must be observed in order to energetically to achieve favorable signal transmission properties. This problem is solved with magnetic cores appropriate choice of the number of turns attached to the controlling and controlled magnetic cores. Because of the fundamental topological Differences between magnetic core and magnetic layer circuits, it is not possible, however, that of the Magnetic core circuits using known dimensioning rules for the coupling circuits Magnetic layer elements, and it was not previously known in what way Matching problem in magnetic layer circuits can be solved.

It has already been proposed for the directional transmission of information between the magnetic layer elements a chain of such elements through different layer thicknesses of the individual Magnetic layer elements in the direction of information transmission to an asymmetrical field distribution generate so that the stray field of an element is always a preferred one according to the direction of transmission Neighboring element to accept a signal transmission arrangement

Applicant:

International Business Machines Corporation,
Armonk, NY (V. St. A.)

Representative:

Dipl.-Ing. A. Bittighofer, patent attorney,

7030 Boeblingen, Sindelfinger Str. 49

Named as inventor:

Henri Jean Philippe Oguey, Peseux (Switzerland)

Claimed priority:

Switzerland of October 27, 1961 (12 437)

correct magnetization state influenced. According to a modification of this proposal, the Asymmetrical field distribution is generated by means of vertical conductors of different widths. It will be uniform thick magnetic layer elements are used, the two band-shaped drive lines in alternating sequence are assigned whose width is within the range of each magnetic layer element of a maximum value decreases to a minimum value. This is done by a current in one of the conductors to the elements applied magnetic field has its greatest strength in the area of the minimum width. When the driving current decreases thus always leaves the part of the magnetic layer element assigned to the area of maximum conductor width first the magnetic saturation and decreases the magnetization of the preceding element at. A favorable adaptation of the coupling means when transmitting signals from a controlling to a controlled element is not taken into account in this arrangement.

It is also known in magnetic layer memories for the purpose of scanning one in one single memory element stored bit sequence by controlled domain wall movement one in its Use wide, wedge-shaped, ribbon-shaped interrogation conductors (e.g. British patent 867 723 or International Solid-State Circuits Conference, February 1961, pp. 64 and 65). A change in the The interrogation field causes an area to migrate in the memory element for wall switching

809 540/300

3 4

critical field strength along the scanning direction. that form the controlling magnetic layer arrangement-That Problem of matching between a controlling, thin magnetic layers, the and a controlled magnetic layer element F i g. 7 a coupling device with in series for the purpose of effective signal transmission there is a switched, controlled magnetic layer arrangement also not with this arrangement. 5 forming, thin magnetic layers,

The object of the present invention is to take measures F i g. 8 a coupling device with parallel to be indicated, by the interconnected in the coupling, the controlled magnetic layer arrangement controlling and controlled magnetic layer arrangements; forming, thin magnetic layers, gen a predeterminable adaptation with regard to a F i g. 9 a coupling device for signal transmission

energetically favorable signal transmission enables transmission over two stages, will. Fig. 10 diagrammatically shows the idealized characteristic

In an arrangement of the type explained at the beginning, line for induction B and field strength H in one, this object is achieved in that for the purpose of magnetic layer element,

the improvement of the transmission properties FIG. 11 a and 11 b diagrammatically resulting from

Overall extension of the magnetic layer arrangement 15 of the ß / if characteristic curve of Fig. 10 resulting characteristic across the direction of the coupling tape line as well as lines for magnetic flux Φ and current / for different as the width of this line in the area of the magnetic geometric shapes of magnetic layer and coupling layer arrangements for the controlling and the gelungsbandleitung; 11a shows the fy / Z characteristic-controlled magnetic layer arrangements are unequal, that line for a rectangular magnetic layer element, at least one of the two magnetic layer arrangements, its longitudinal axis is parallel, and FIG. 11 b shows the layer element having a different overall extent parallel '/ ^''for a rectangular magnetic and orthogonal to the coupling tape line, the longitudinal axis of which is orthogonal to and that the coupling tape line runs in both magnetic layer arrangements, products of the magnetic layer thickness and overall shape diagrammatically the ß / H-characteristic for

expansion parallel to the coupling tape leads to a controlling magnetic layer element in the event of modulation separations are great. tion in the hard direction,

The invention makes the signal transmission Fig. 12b diagrammatically the B / H characteristic for

between the controlling and controlled magnetic layer - a controlled magnetic layer element in modulation arrangements vastly improved. As the energy ration in the easy direction with a magnetic the bias in the hard direction from the controlled magnetic layer arrangement in 30, the coupling line fed signals relative to F i g. 13a and 13b diagrammatically result from the

is low, are energetically favorable transfer β / H characteristics of Fig. 12a and 12b resulting properties in arrangements of the type mentioned 0 // characteristics, whereby through a suitable choice of of crucial importance for achieving the geometric shape of the magnetic layer elements and a reliable way of working. Only in this way is these characteristics both in ensures that the magnetic reference from the controlling magnet to the controlling as well as the controlled Layer arrangement emitted signal component largely magnetic layer element the same initial inductance have not weakened in the controlled magnetic layer arrangement,

tion comes into effect. F i g. 14 a schematic representation of a 2-stroke

Further advantageous embodiments of the invention 40 are shift registers with non-reciprocal transfers to be seen from the subclaims. Subsequent properties for binary information, the following are various exemplary embodiments of FIGS. 15 a cross section through the 2-stroke VerErfindung described on the basis of drawings. shift register according to FIG. 14, It shows F i g. 16 the structure of the conductor arrangement for

F i g. 1 shows a basic arrangement of the invention 45 generating the preamble fields in the 2-stroke vergemäß Coupling device with two magnetic shift registers according to FIG. 14, Layer elements, which by a coupling tape F i g. 17 diagrammatically shows the course over time

line of variable width coupled with each other of the driver current for the operation of the 2-stroke-Versind, Shift register,

Fig. 2 shows a cross-section of the coupling device 50 Fig. 18 diagrammatically shows the temporal progression according to FIG. 1, which in relation to the A elements of the 2-stroke

F i g. 3 the electrical equivalent circuit diagram of the Kopp shift register effective driving fields, processing device according to FIG. 1, Fig. 19 diagrammatically shows the course over time

F i g. 4 shows a first variant of the inventive in relation to the B elements of the 2-stroke Ver coupling device with a controlling magnetic shift register effective driving fields, layer arrangement consisting of three not combined. 20 the critical curve of a magnetic layer

hanging, from the coupling ribbon cable in series elements and in vector illustration the on comprised of thin magnetic layers, controlled magnetic layer element under the Ein-F i g. 5 a second variant of the coupling influence from the left and right neighboring controlling Direction with a controlled magnetic layer arrangement control fields acting on 60 magnetic layer elements tion, made up of three non-contiguous, of in forward and backward direction for the purpose the coupling ribbon line included, thin a qualitative analysis of the operation of the 2-stroke magnetic Layers consists, shift register according to FIG. 14.

F i g. 6 a further variant of the coupling element. All the embodiments described below

direction with two unrelated, in 65 of the present invention have the coupling Series connected, the controlling magnetic layer of magnetic layer elements by means of a strip conductor arrangement forming, thin magnetic layers shaped coupling line in common. Magnet- and two non-contiguous, parallel-formed layer elements are electromagnetic, made of thin

high-speed switching elements composed of magnetic layers. A thin magnetic layer is a ferromagnetic material (e.g. 20% iron and 80% nickel) applied to a base, for example vapor-deposited. For the thickness of the magnetic layer elements, the range between 100 and 30,000 Å (IA = 10 " ⁸ cm) is preferred. Magnetic layers with uniformly aligned magnetization are of particular importance; a distinction is made between isotropic and anisotropic thin magnetic layers Magnetization in the position into which it was brought by a switching process, e.g. caused by the application of an external magnetic field. In anisotropic magnetic layers, there are certain preferred positions for magnetization. In magnetic layers with uniaxial magnetic anisotropy, the magnetization is parallel or antiparallel to a certain preferred direction, which is also referred to as the “easy direction.” If, in a magnetic layer with uniaxial anisotropy, the magnetization is deflected from the easy direction by applying an external magnetic field, it reverses in the closer direction when the external magnetic field is switched off next neighboring preferred location.

In the case of magnetic layer elements, the binary information 1 or 0 is generally represented by the parallel or anti-parallel setting of the magnetization with respect to a specific direction. A magnetic layer element can be switched from the ONE to the ZERO position and vice versa with the help of externally acting magnetic fields, either by wall or rotary switching. Because of the significantly shorter switching times - they are in the order of magnitude of nanoseconds (1 ns = 10 " ⁹ s) - the magnetization reversal processes caused by rotary switching are preferred. In rotary switching (also known as rotary switching), magnetization reversal occurs through a generally coherent rotation of the magnetization in the new direction.

First of all, a basic arrangement of the coupling device according to the invention is described, in which two magnetic layer elements are coupled to one another by a coupling tape line with a width continuously increasing in the longitudinal direction; Fi g. 1 shows the top view and FIG. 2 shows a cross section of this coupling device. Two magnetic layer elements 1 and 2 are coupled to one another by means of a self-contained strip conductor loop 3, of which the part above the elements is referred to as the outgoing line and the part running below the elements is referred to as the return line. Each individual one of the two magnetic layer elements is formed by a contiguous, approximately rectangular, thin magnetic layer of thickness d. The longer side of the rectangle is marked with a, the shorter one with b . The two magnetic layer ^{elements are offset from one another by 90 c} , in such a way that the side α of element 1 and side b of element 2 run parallel to the longitudinal axis 4 of the strip conductor. The width of the strip line 3 is adapted to the corresponding dimensions of the magnetic layer elements, ie the strip line above and below the element 1 is approximately as wide as the magnetic layer side b; above and below the element 2 it is about as wide as the magnetic layer side a. The transition from the narrow to the wide part of the ribbon cable is continuous. The distance W ₁ between the outgoing and return lines for element 1 and the corresponding distance w ₂ for element 2 are in the same ratio as the width b of the strip line with respect to element 1 and the width α of the strip line with respect to element 2. The reason for this measure will be given later.

The two magnetic layer elements 1 and 2 consist of thin magnetic layers of the type mentioned, whose uniformly aligned magnetizations can be deflected or switched from an initial position in a certain direction by externally applied magnetic fields running parallel to the layer plane. For the consideration carried out below, which is valid for the general case, the presence of anisotropy of the magnetic layers does not need to be assumed in principle. It is assumed that the magnetization of the element 1 in the starting position is oriented in the direction indicated by the arrow 5 parallel to the longitudinal axis 4 of the strip conductor. When an external magnetic driving field H _tr directed perpendicular to the starting position 5 is applied, the magnetization of the element 1 is deflected in the direction of the driving field, namely in the direction indicated by the arrow 6. The greater the magnetic driving field H _tr , the greater the deflection angle. In the event that the magnitude of the driving field is greater than the saturation field strength H _{K of} the magnetic layer, a deflection angle of 90 ° results, ie the direction of magnetization is then parallel to the external driving field in the direction of arrow 6. Such a deflection or switching of the As a result of the associated change in the magnetic flux, magnetization induces a voltage in the part of the strip conductor 3 running above or below the element 1. Since the strip conductor forms a closed conductor loop, an induction current / begins to flow. This current flowing through the strip conductor 3 in turn wrikt both the magnetic layer element 1 and the magnetic layer element 2 due to the magnetic fields it causes. The effect of the induction current in relation to the magnetic layer element 1 consists in the formation of an external field H, _r counteracting Field in accordance with Lenz's law. In relation to the magnetic layer element 2, the induction current / generates a magnetic field which - as indicated schematically by the arrow 8 - is directed orthogonally to the strip conductor axis 4 upwards. Assuming that the magnetization of the element 2 in the starting position is parallel to the longitudinal axis 4 of the strip conductor, e.g. B. in the direction indicated by the arrow 7, it experiences a certain deflection in the direction shown by the arrow 8 due to the magnetic field caused by the induction current /. This deflection of the magnetization occurring in the magnetic layer element 2 results in the generation of a counter voltage in accordance with Lenz's law, which counteracts the current flowing in the strip conductor 3. During the switching process considered here, triggered by the application of the external driving field H, _r , the coupling

treatment device a state of electrical equilibrium, according to the condition that the sum the induced voltages and the ohmic voltage drop resulting from the flowing current and the line resistance is equal to zero. '

The computational treatment of the case considered here of coupling two magnetic layer elements by means of a coupling ribbon cable is carried out on the basis of the corresponding equivalent circuit diagram, which in Fig. 3 is shown. The change

tion of the magnetic flux -j- in the magnetic layer elements 1 and 2 with respect to the coupling line has its equivalent in the two voltage sources U ₁ and U _{2 of} the equivalent circuit diagram. The line resistance of the coupling ribbon line is taken into account by the resistance R in the equivalent circuit diagram. Assuming the same polarity of the voltage sources CZ ₁ and U ₂ and the indicated direction of the current /, the relation applies that the sum of the voltages U ₁ and U _{2 is} equal to the voltage drop across the resistor!?:

U ₁ + U ₂ = IR.

(D

(The indices 1 and 2 refer to the magnetic layer elements 1 and 2, respectively)

The voltages U ₁ and U ₂ result from the law of induction:

3 °

U ₁ = -

German άΦ ₂

The magnetic fluxes Gt ₁ and Φ ₂ result from the product of the magnetic layer element cross-sectional area F induction B:

35

40

Φι = F ₁ -B ₁ , (4)

Φι = F ₂ -B ₂ . (5)

The cross-sectional areas F ₁ and F ₂ are the magnetic flux exit areas with respect to the coupling line; they are given by the vertical cross-section through the magnetic layer elements 1 and 2 parallel to the longitudinal axis of the strip conductor 4. If both magnetic layer elements are of the same thickness d, the cross-sectional areas are:

F ₁ = ad, (6)

F ₂ = bd. (7)

The inductions B ₁ and B ₂ are known to be the product of permeability μ and field strength H:

B ₁ = / HH ₁ , (8)

B ₂ = / '2H ₂ . (9)

Hysteresis characteristics (^ /// characteristics) of the magnetic layer elements determine.

The field strength H ₁ is used for the calculation as the sum of the external magnetic driving field H _tr and the magnetic field generated by the current /, the latter being inversely proportional to the width of the coupling line above the magnetic layer element 1:

H ₁ = H, + i- = H _tr + ~ H ₂ . (10)

The field strength H ₂ results solely from the magnetic field generated by the current /, which is inversely proportional to the width of the coupling line above the magnetic layer element 2:

By combining the above equations accordingly, the differential equation is obtained

fh a J

dH ₂ German

dH, _r German

for the magnetic field H 2 acting on the magnetic layer element ₂ , which represents the only decisive control field for the signal transmission from the controlling magnetic layer element 1 to the controlled magnetic layer element 2. In order to achieve favorable signal transmission conditions, efforts must be made to obtain the greatest possible control field strength H ₂ .

Assuming a linearly increasing driving field

H _tr = H ₀ - , 0 <ί <τ

and at the same time summarizing the values assumed to be known

, a lly b

k = -τ- + - * - * - -, b / I ₁ a

The solution of the differential equation results in the control field strength H ₂ as a function of time r, valid for the time segment 0 <t <r, before saturation of the magnetic layer element 1 is reached:

In order to keep the mathematical effort for the calculation to be shown here small, the permeabilities / Z ₁ and / i _{2 of} the magnetic layer elements are assumed to be given; they should depend neither on the field nor on the time. However, when performing sizing tasks in practice, it is advisable to make more realistic assumptions. However, there are only quantitative, but not qualitative, differences. The permeabilities to be used in the calculation can be found, for example, from the experimentally recorded values. The variable τ characterizes the point in time at which saturation is reached. It is assumed that the saturation of the magnetic layer element 1 just occurs when the driving field H _tr its maximum value H reaches ₀ (namely, at the time f = f _s = r) Under this assumption, it must act on the magnetic sheet member 1 resulting external field H ₁ be at least equal to the saturation field strength H _K. The control field strength effective at this point in time then has its maximum value, which will henceforth be referred to as H ₂ .

At time t = τ , equation (10) is written as follows:

the driving field until the value / Z _{0 is reached} . The determining equations are:

α β

-τ H ₂ .

(17)

In order to be able to assess the influence of the control field strength H ₂ on the controlled magnetic layer element 2 with regard to its control effectiveness, it is useful to consider the quotient of the control field strength to the saturation field strength, which is achieved by combining equations (16) and (17) with the elimination of H ₀ :

JL ₌ JIlL- ¹ L *

b 11?, I.

, (22)

1 - exp (-

- = 3 -Π / ΐ ₊ ± JL (LW ^ Ίαΐ \ _ , 1 T ₁ ft LI / 3 α Μ / ί ₂ | a J ₁ ) J "

From these equations it follows that the following secondary condition is met for the coupling circuit have to be:

Η _κ /, "2 \ H _K J

LI

The analysis shows that the quotient -? R-

"2

increasing quotient - steadily increasing, whereby increasing

it should be noted that, according to equation (14), k is also a function of the geometric dimensions α and b . Accordingly, high control field strengths are to be expected if α is provided to be much larger than b . Using a Taylor expansion, equation (18) can be approximated by the following relationship:

If the minimum value of the control field strength required for proper operation is denoted by Iw ₂ I ,,,, - ,,, the following dimensioning conditions for the geometric dimensions and for the driving field amplitude are obtained, taking into account equations (24) and (23):

\ H ₂

/ Ί

2T ₁

27 ?; a

(19) H _K

Ul
lh

, ";"

127?

This approximation has an accuracy of within 1% for

^a >? ς Λ Γ IJi ~ ^τ ■ by μι Τι

(20)

However, it can also be used for all other quotients ~ - occurring in practice without making a decisive mistake. There are only slightly smaller values for the control field strength.

From equation (19) it can be seen that the control field amplitude for the case a » b can assume the following maximum amount:

After the signal transmission from left to right, i. H. from the magnetic layer element 1 the magnetic layer element 2 has been mathematically investigated, let the case of signal transmission be the opposite Direction, d. H. viewed from the magnetic layer element 2 onto the magnetic layer element 1. We are now interested in the control field acting in relation to the magnetic layer element 1, if that external driving field acts on the magnetic layer element 2. The mathematical analysis is analogous Way to perform as above. To distinguish the two cases, let the corresponding sizes marked with an asterisk (*) for signal transmission from right to left. One goes of the following two equations [cf. the above equations (10) and (16)] from:

\H

2T _t

2 I max

(21)

m = H _tr + ^ - H ₁ *, (27)

However, it is not advisable to use control field strengths close to this maximum amount, since, according to equation (17), this may require a prohibitively high driving field strength H ₀ .

Equations (17) and (19) can be used to dimension a coupling circle. · It is assumed that the quantities H _K , H ₀ , T ₁ and the ratio of permeability / i ₂ : / I _{1 are} given. The control field strength H ₂ is also to be regarded as given, because the circuit will require that a certain minimum value of the control field strength be achieved in order to ensure proper signal transmission. What is sought is the geometric dimension ratio a: b and the rise time τ for H _1, the control field effective with respect to the magnetic layer element 1 and H ₂ the resulting magnetic field acting with respect to the magnetic layer element 2. The constant k * with the value

_fc * ₌ JL ₊ IJl JL (29)

a / I ₁ b

is greater than the above constant k [cf. Equation (14)] because of the assumptions made a> b and / i ₂ > / I ₁ .

While in the case of signal transmission from left to right, the time r _s , which elapses until the

809 540/300

With respect to the controlling magnetic layer element I effective resultant magnetic field H _{1 reaches} the saturation _{field strength H K} , was set equal to the time r that is required until the driving field H, _{r reaches} its maximum value Ho, so it is shown that in the case of signal transmission from _{right to left the resulting magnetic field H 2} * effective in relation to the controlling magnetic layer element 2 reaches the saturation field strength H _K earlier than above, ie:

<f _v or t * <τ.

(30)

The effective signal transmission ratio in both directions in a coupling device according to FIG F i g. 1 can be at least approximately quantified with reference to equation (19) by the "reflux transfer ratio" //, which for the directivity of the coupling circuit is characteristic:

[I a

Ml - ^{2 y}
T ₂

127 ?;

b_ a

JL ₊ (Hi ₊ _ £! Λ

(31)

The directivity of a coupling circuit is the better, the smaller the ratio [I with respect to 1 is. Taking into account equation (22), neglecting the difference between t * and r and assuming a » b, one can approximately set for //:

, "2

a Ih ₂

H ₀ - H _K

(32) the requirements placed on the rise time τ for the driving field can be realized more easily for a> b.

To show the practical value of the above analysis, an example is given.

The following assumptions correspond to real conditions in practice based on the current state of the art: H _K = 5 Oe; u _z : - // _t = 4;

T ₁ = / I ₁ γ = 1 nsec (= 10 ~ ⁹ see). The coupling circuit assumes a good control effect Ih ₂ ["" · ,, = I Oe and a good directivity β = y

demands.
It is first checked whether (34) is fulfilled; This is

the case, because y <1.56.

The required maximum field strength of the driving field is calculated from (33) and results in H ₀ = 8.75 Oe.

In order to satisfy (35), namely - <0.58, becomes

a = 3 b chosen.

The rise time for the driving field is calculated from (23) and results in τ = 2.36 nsec.

The control effect exerted on the controlled magnetic layer element is calculated from (22) and “results in IH ₂ 1 = 1.25 Oe> 1 Oe = \ H ₂ \ _min .

For a favorable signal transmission in a coupling ribbon conductor loop, how they z. B. also in the basic arrangement of the coupling device according to the invention (See Figs. 1 and 2) is applied, it is desirable that the characteristic impedance Z (based on the unit of length) across the entire length of the line Has value. The well-known relationship applies to strip conductor loops:

25th

35

If you think of a coupling circuit a
requires a certain directional effect, ie the ratio [> specifies, the following approximation applies

inversely proportional to the width / the strip conductor d bdi

H _K ~

-J! L

(33)

g pp

ribbon. In order to, in the case of the coupling ribbon line 3, of the FIG. 1 and 2, the impedance condition Z - const. Must be met. Using (26) one can still find the ratio must

"H-Wr) '- ^l34)

^V. "2 I " 2 I min '

If the expression in brackets in (34) is less than 1, the coupling circuit must have a certain Demand directivity so that a satisfactory mode of operation is achieved.

_{If the quantities H 0} and H _{2 are} eliminated using equations (22), (24) and (33), the coupling circuit can only operate satisfactorily if the following condition is met:

«Ι: b = us: a

(35)

This requirement (35) stands for the statement, which is important from a qualitative point of view, that in order to be able to achieve a certain directional effect in the signal transmission with the coupling arrangement {[i < 1), the dimension ratio b : α must be selected to be less than 1, i.e. .Ti. a> b. In addition, these can also be met, that is, the distance between the forward and return lines is changed in the same ratio as the width of the strip line changes.

The two magnetic layer elements 1, 2 used in the basic arrangement of the coupling device according to the invention (see FIGS. 1 and 2) are of the same volume, because it was assumed that they have the same thickness d, the same length α and the same width b . In the following, some coupling devices equivalent to the basic arrangement are shown, to which the statements made in relation to the basic arrangement apply in a corresponding manner. In the F i g. 4, 5 and 6 are used controlling or controlled magnetic layer arrangements which are composed of several individual magnetic layer elements. It is immaterial which geometric shape the individual elements have. It is assumed that the individual elements of the controlling magnetic layer

arrangement and the individual elements of the controlled magnetic layer arrangement in total have the same volume; that is, provided that the layer thickness d is the same, that the "effective" length of a magnetic layer arrangement, which is composed of several magnetic layer elements comprised in series by the coupling line, is equal to the sum of the lengths of the individual elements, and that the "effective" width of a magnetic layer arrangement, which is composed of several magnetic layer elements encompassed in parallel by the coupling line, is equal to the sum of the widths of the individual elements.

In principle, one can consider the following cases:

a) Controlling magnetic layer arrangement with individual elements lying in series (FIG. 4);

b) Controlled magnetic layer arrangement with individual elements lying in series;

c) Controlling magnetic layer arrangement with parallel Single elements;

d) Controlled magnetic layer arrangement with parallel individual elements (FIG. 5);

e) Combination of cases a) or c) with b) or d); as an example, FIG. 6 the combination of Cases a) and d) shown.

In general it can be said that the presence of several individual elements in a controlling magnetic layer arrangement is important for the execution of logical operations. On the other hand, there is presence several individual elements in a controlled magnetic layer arrangement important for the implementation of branches. When performing logical operations, however, it is important to note that that the coupling device is to be optimized for the case. that the circuit to be expected Minimum number of actually controlling elements to achieve the greatest possible control effect achieved with respect to the controlled magnetic layer arrangement.

Reference is now made to FIG. 4, which shows a first variant of the coupling device according to FIG. 1 shows that the controlling magnetic layer arrangement is no longer made up of a single. coherent, thin, magnetic layer, but from three equally large, separated, approximately square, thin, magnetic layers 11, 12 and 13 with the side length b . These three non-contiguous magnetic layer elements are encompassed by the coupling ribbon line 3 in series. They are - similar to FIG. 1 - coupled to a controlled, consisting of a single, contiguous, rectangular, thin, magnetic layer magnetic layer element 2, the length α of which is 3 times its width h . It is evident that the signal transmission ratios of the coupling device of FIG. 4 are the same as in the previously considered case of the coupling device according to FIG. 1, provided that the three separately arranged, thin, magnetic layers 11, 12 and 13 are subjected to the same operating conditions. Then they can be viewed in their entirety as a controlling magnetic layer arrangement of an effective total linear extent 3i> = α parallel to the axis of the coupling ribbon line.

F i g. 5 shows a second variant of the coupling device according to FIG. I. In this case, the controlled magnetic layer arrangement is formed from three equally large, separately arranged, approximately square, thin, magnetic layers 21, 22 and 23 with side length b . These three non-contiguous magnetic light elements are encompassed by the coupling ribbon line 3 in parallel connection. They are - similar to FIG. 1 - coupled to a controlling magnetic layer element 1 consisting of a single, contiguous, rectangular, thin, magnetic layer, the length α of which is 3 times its width b . Here, too, it is evident that the signal transmission conditions are the same as in the coupling device according to FIG. 1, provided that the three separately arranged magnetic layer elements 21, 22 and 23 are subjected to the same operating conditions, at least during the time interval of the signal transmission to these layers. Then they can be viewed in their entirety as a controlled magnetic layer arrangement with an effective total linear extent 3 b = α orthogonal to the axis of the coupling ribbon line. For the purpose of connecting the three magnetic layer elements 21, 22 and 23 in parallel, the strip conductor looping around them is branched into three sections 31, 32 and 33, each section having the same width corresponding to the side length h . As a result of the parallel connection used, the effective width of the coupling ribbon line in relation to the three magnetic layer elements 21, 22 and 23, which are to be considered as a controlled magnetic layer arrangement, is the sum of the widths of the individual branches 31, 32 and 33.

The in Fig. 6 shown embodiment shows a combination of the cases discussed above, by having both the controlling and the controlled magnetic layer assembly of e.g. B. two separated arranged magnetic layer elements are formed. In total there are four magnetic layer elements, namely two controlling 14 and 15 and two controlled 16 and 17 are provided, which are shown in FIG. 6 for example are shown square and have a side length c-. The magnetic layer elements are located on a non-ferromagnetic, electrically conductive base plate 9, which z. B. a thin one Can be copper or tantalum foil; they are optionally isolated from this by a thin layer Intermediate layer, e.g. B. silicon oxide, electrically insulated, A coupling tape conductor 10 is provided over the four magnetic layer elements, which essentially consists of three ladder sections 18, 19 and 20. One ladder part 18 extends over the two Magnetic layer elements 14 and 15 and causes them to be connected in series. The two other ladder parts 19 and 20 represent a branch, the part 19 extending over the magnetic layer element 16 and the Part 20 extends over the magnetic layer element 17 and thus a parallel connection of these two magnetic layer elements causes. This circuit has the consequence that the two magnetic layer elements 14 and

15 in its entirety like a controlling magnetic layer arrangement with an effective overall linear extent 2c parallel to the longitudinal axis of the conductor part 18 and the two magnetic layer elements

16 and 17 in their entirety like a controlled magnetic layer arrangement with an effective linear one

15 16

Total extent 2c orthogonal to the longitudinal axis of the driving field is in FIG. 6 symbolically by the two

of the ladder parts 19 and 20 act. Likewise indicated in the arrows H is _Ir2.

In this embodiment, the effective width of the magnetic layer elements 16, 17 through the layer elements 16 and 17 is twice the effective driving field H _tr2 in the “hard” direction (that is, the width of the coupling ribbon line in relation to the direction orthogonal to the easy direction) is to be considered as a controlling magnetic layer arrangement. Simultaneously with the arrival of a control transmitted by means of magnetic layer elements 14 and 15 of the coupling tape conductor 10, the ends 24, 25 and 26 of the coupling tape pulse, which by deflecting the magnetization conductor 10 are electrically conductive with the base plate 9 of the controlling magnetic layer elements 14, 15 connected, so that the strip conductor 10 is obtained together with the easy direction 30, the electrically conductive base plate 9 is switched off to a closed drive field H _tr . The magnetization switches on and forms a coupling loop. In order to meet the above-mentioned tion of the elements 16, 17 from the unstable hard impedance condition Z = const., The 15 direction is back in the stable easy direction. Distance between the strip conductor 10 and the basic The direction of switching back is determined by plate 9 according to the effective width of the coupling the polarity of the control pulse. As is well known, the binary information tape line over the length of the line is continuously changed by the orientation of the line. Dementspre- magnetization in the easy direction is represented the distance of the conductor part 18 from the 20 benefits; accordingly, in the example shown here (FIG. 6), the base plate is smaller than the distance between the conductor parts 19, one binary value (e.g. "0") through a downward and 20 from the base plate. This difference in the directional and the other binary value (eg "1") the distances allows the design of a multi-stage shift register with an upward magnetization, as in FIG. 6 indicated. of the magnetic layer elements 14, 15 and 16, 17 represent - According to this embodiment runs above 25 is placed. The deflection of the magnetization of the steuder magnetic layer elements 16, 17, which are located on the earth magnetic layer elements 14, 15, is effected by the base plate 9, the conductor part 18 'of the stage of the displacement-acting driving field H, which is connected to these elements on the right-hand side , _ri . Depending on the tape leader 10 'belonging to these registers. Only binary information stored about the elements controlling it is the 30 tion belonging to the strip conductor 10, ie depending on the starting position of the magnetization conductor parts 19 and 20. Since the magnetic layer elements tion, one obtains in the coupling strip conductor 10 a 16, 17 in the subsequent stage of the displacement-induced positive or negative current pulse, registers as a controlling magnetic layer arrangement. 35 of the pulse-shaped magnetic field (control field), which the in F i g. The coupling configuration shown in FIG. 6 - as already mentioned - the direction of the return can be used advantageously in particular when switching the magnetization of these elements from the rotational switching of magnetic layers with unidirectional in the easy direction. This axial anisotropy for information transmission control field is relatively small if it is used with the process, as it is already compared in principle with the field strength of the driving spring H _lr. Since you are beaten and z. B. in the article "Thin Magnetic can set it up through synchronization, however, Films" by S. Methfessel, WE Proebster that the control field just at the time of the off and C. K i η berg in the anthology "Informa- switching the up the controlled Magnetschichttion Processing ", Verlag Oldenbourg, Munich, 1960, elements 16, 17 acting external driving field H _Ir2 p. 439 to 446, was described; is particularly effective, it only has to fulfill a control function in fig. 4, 15 and 17, for which its field strength is sufficient. By article. the switching back of the magnetization of the controlled- In this application, the magnet elements 16, 17 have layers 14 and 17 in a predetermined position, a uniaxial magnetic aniso- parallel to the easy direction depending on tropie; The slight direction 30 runs in all four 50 of the polarity of the control field is the previously stored in layers orthogonal to the longitudinal axis of the controlling elements 14, 15 around them binary strip conductor parts 18, 19 and 20 respectively elements acting on the magnetic layers 14 and 15 are adopted.

magnetic driving field H _trl can, for. B. by a current-carrying 55 drive (not shown in FIG. 6) of the transmission of binary information is generated by a strip conductor which includes both magnetic layers during the time 14 and 15 essential for the transmission and whose longitudinal axis is parallel section, namely during a certain time to the easy direction. The driving field generated by it at the beginning of the signal transmission process then acts orthogonally to the light Rieh permeability, H _{1 of} the controlling magnetic layer line, as shown in FIG. 6 is symbolically indicated by the two 60 elements considerably smaller than the permeability / i ₂ arrows H _trl . _{The driving field / J, r2} acting in relation to that of the controlled magnetic layer elements, which alone, magnetic layers 16 and 17, can be determined in a similar way by a second (in layer elements shown) current-carrying tape assumption / i ₂ > / I ₁ is thus fulfilled, conductors are generated which includes both magnetic layers 16 65 In the arrangements considered so far and 17 and the longitudinal axis of the coupling device according to the invention was parallel to the easy direction. The orthogonal to the easy direction, which he created for the sake of simpler analytical representation, made the restrictive assumption that the

replace α by a ₂ ,
replace h with b _x ,
replace d with A ₁ ,

50

55

η, replace by / I ₁ = / I ₁ -, (38)

^fl 2

Replace / I ₁ with / i ₂ = / i ₂ , (39)

D ₁ di

Replace T ₁ with T ₁ '= T ₁ - for </, = d, (40)

^a 2

Replace k with k '= - ^ - + ^ -. (41)

optionally composed of several individual magnetic layer elements, controlling and controlled Magnetic layer arrangements have the same magnetic volume, d. i.e. that the Product of thickness, effective length and effective width of all controlling magnetic layers is the same the product of the thickness, the effective length and the effective width of all controlled magnetic layers. At the Design of circuit networks, e.g. B. for computing or control units in program-controlled digital Calculating machines will, however, often have to deal with cases that do not meet this requirement is satisfied.

With reference to FIGS. 7 and 8 we take the case of branching as an example, wherein a controlling magnetic layer has several (here for example three) downstream controlled magnetic layers must be controlled, the latter, viewed in its entirety as a controlled magnetic layer arrangement, from the controlling magnetic layer deviating, namely larger effective total linear dimensions exhibit.

The series connection of three controlled magnetic layers 35, 36 and 37, each of which is the same size and shape as the controlling magnetic layer 34, is shown in FIG. 7 shown. These layers are more similar through a coupling ribbon conductor loop 27 Way, as already described several times, coupled with each other.

The parallel connection of three controlled magnetic layers 39, 40 and 41, each of which of is the same size and shape as the controlling magnetic layer 38 is shown in FIG. 8 shown. These Layers are coupled to one another via a coupling strip conductor loop 28 by being connected in parallel.

Assuming the same thickness of the magnetic layers is in the embodiments according to F i g. 7 and 8 the volume of the controlled magnetic layers by a factor of 3 larger than the volume the controlling magnetic layer.

About sizing rules for the generalized case of deviating effective linear overall dimensions to be able to give between controlling and controlled ^ magnetic layer arrangements is modify the mathematical analysis given above accordingly. The derived equations and relationships apply accordingly to the generalized case if one follows Making substitutions:

Let the volume ratio of the controlled to the controlling magnetic layer arrangement be

γ =

Then the following ratio of permeabilities results:

El

/ Ί

Taking into account the above substitutions, equation (22) changes into its modified form:

H ₁

a ₂ [H ₂

Equation (19) also changes into its modified form:

■ (45)

2T ₁ a,

By substituting the corresponding values in relation (25) one obtains for the generalized one Case the following dimensioning condition for the geometric dimensions:

b ₂ d ₂

St.

I H ₂ 1 "" ·,

This relationship (46) can be interpreted as follows. In order to control a controlled magnetic layer arrangement, the controlling magnetic layer arrangement must apply a change in magnetic flux which is greater than a certain minimum amount Φ ₂ "" · ". On the other hand, the controlling magnetic layer arrangement generates a magnetic flux change ΛΦ ₁ with respect to the coupling line. The coupling device has an efficiency for the transmission of the magnetic flux that is less than 1, since of course a certain amount of energy is always lost. Accordingly, the following must apply:

Λ Φ ₁ > I Φ ₂ 1 _min (47)

^ d ₁ AB ₁ > b ₂ d ₂ \ B ₂ \ _min , (48)

where 1 B ₁ denotes the change in induction in the controlling magnetic layer arrangement and | U ₂ | "" - "denotes the required minimum induction in the controlled magnetic layer arrangement.

The required rise time [cf. Equation (23)] results for the generalized case from following modified form:

60

= 3-

3 « ₂

b ₂ d ₂

In practice, there may be the case that two types of magnetic layer cells are of two sizes in a transmission system occur, as in the embodiment of FIG. 9 for a two-stage coupling configuration is shown, which has a total of

809 540/300

represents branching over two levels. It is assumed that an element of the first size type controls an element of the second size type in the first transfer stage and the element of the second size type η elements (e.g. η = 3) of the first size type in the second transfer stage. In Fig. 9, the elements of the first size type are represented by magnetic layers 81, 82, 83 and 84 with the dimensions: length Ci ₁ , width b _u thickness ^ ₁ . The element of the second size type is represented by the magnetic layer 85 with the dimensions: length O ₂ , width b ₂ , thickness d ₂ . The magnetic layer 81 in a controlling function and the magnetic layer 85 in a controlled function belong to the first transmission stage. These two magnetic layers are coupled to one another via a coupling strip conductor loop 85, as already described several times. The second transmission stage includes the magnetic layer 85 in a controlling function and the magnetic layers 82, 83, 84 in a controlled function. These layers are coupled to one another via a coupling strip conductor loop 87 by being connected in parallel. It is clear that the in Fig. 9 shown two transmission stages can be part of a further branched logical system.

It is now desirable to dimension the two transmission stages in an energetically favorable manner, ie to determine the geometric dimensions of the magnetic layers so that the requirements for the driving field (amplitude H ₀ and rise time r) and the control field strength with regard to the controlled layers are identical in both stages are. For the sake of simplifying the calculation, it is assumed in the following that the resistances of the coupling ribbon conductor loops 86 and 87 are the same.

Equations (44) and (45) are used and successively applied to the two transmission stages applied:

If relation (25) is modified, the following additional dimensioning condition can be derived:

Hi

. "ι

τ ι - +
■ fr '" ² + * * 2 al) a ₂ (51) *> ₂ Η _κ 2T ₁ +: τ ² 3a \ d \
α ² ? β -) 3a, · \ Η ₂ \ 3α, ά, oil 3 μ ₂ 127? τ a ₂ d ₂ * μ \ τ ²
γη} 2 TJ

The condition results from equation (50)

a ₂ b ₂ = 3Ci ₁ O ₁ . (52)

By equating coefficients in equation (51) in combination with condition (52), one obtains the following dimensioning conditions for the geometric dimensions of the magnetic layers:

^a 2 _ Λ ^b 2 _ Λ ^d 2 _ -ri

- - J, -j- - J, -p J

O ₁ O ₁ U ₁

For practical reasons it may not be desirable under certain circumstances to provide magnetic layers with different thicknesses. Then U ₁ = d ₂ , and the volume ratio r = η = 3, according to equation (50). From the following generalized for the following, the signal transmission properties of the coupling device according to the invention are considered from the point of view of the adaptation condition. To this end, we return to the basic arrangement of the coupling device shown in FIG. The adaptation condition known from electrical engineering states that the most favorable energy transfer between a generator and a receiver. is when the internal resistance of the generator is the same as the internal resistance of the receiver, assuming an ideal transmission line. In the case of the present coupling configuration (FIGS. 1 and 2), a good coupling between the magnetic layers 1, 2 and the strip conductor loop 3 can be assumed. The controlling magnetic layer element 1 together with the left part of the strip conductor loop 3 forms the signal generator, while the controlled magnetic layer element 2 together with the right part of the strip conductor loop 3 forms the signal receiver. The use of magnetic layer elements in the mode of operation described above means that the internal resistances of the generator and receiver are essentially represented by the inductances L of the elements. It is primarily the initial inductance that is decisive for this. This initial inductance is determined from the gradient at the zero point of the magnetic flux (0) current (/) characteristic of the magnetic layer elements. The matching condition is fulfilled when the initial inductance of the controlling magnetic layer element is equal to the initial inductance of the controlled magnetic layer element.

To understand the procedure, the idealized B / H characteristic shown in FIG. 10 for a magnetic layer is assumed. The linear part of the characteristic curve passing through the origin of the coordinates ends at the saturation induction ± B _s or at the saturation field strength ± H _K. This is followed by the zone of saturation, ie there the induction B = \ B _s \ remains constant for field strengths H> \ H _K \. The initial permeability u _A is given by the slope of the linear part of the ß /// characteristic in the origin of the coordinates.

Based on the B / H characteristic in Fig. 10 result depending on the position and the geometric shape of the magnetic layer elements and their associated coupling ribbon lines have different-looking Φ / J characteristics.

FIG. 11a shows the characteristic curve for an arrangement which corresponds to the magnetic layer element 1 in FIG. 1 corresponds. The saturation flux <1> _S1 is given by taking equations (4) and (6) into account

0 _S1 = α · d ■ B _s . (55)

In order to achieve this saturation flow, one needs a current

= bH _K.

In other words, I _K1 is the current that must flow through the strip conductor loop 3 so that

With respect to the magnetic layer element 1, a magnetic field of the magnitude H _{K is} generated.

FIG. 11b shows the </> ₂ // characteristic for an arrangement which corresponds to the magnetic layer element 2 in FIG. 1 corresponds. The saturation flux Φ _Β2 results, taking into account equations (5) and (7)

B ₅ .

(57)

In order to achieve this saturation flow, one needs a current

a ■ H _K.

(58)

The initial inductance L ₁ (that is the gradient at the origin of the "/ ^ // curve in Fig. 11a) is (α: fr) ² times greater than the initial inductance L ₂ (that is the gradient at the origin of the 0 ₂ // characteristic in Fig. Lib), provided that the two magnetic layers 1 and 2 are of the same thickness d . For a magnetic layer with a given initial permeability μ _Λ , a desired initial inductance can be achieved solely through topological measures.

With reference to Fig. 6 it has already been mentioned that in the driving field modulation method used for the controlling and controlled magnetic layer elements, the initial permeability 11 _A j of the controlling magnetic layer element 1 is less than the initial permeability n _{A2 of} the controlled magnetic layer element 2. The B / H characteristic is for a controlling magnetic layer element shown in Figure 12a; it results from modulation in the "hard" direction (that is the direction orthogonal to the unaxial magnetic anisotropy). The Z? /// characteristic for a controlled magnetic film element is shown in FIG. 12b; it results from modulation in the "easy" direction (that is the direction parallel to the uniaxial magnetic anisotropy), the magnetic layer element being subjected to a magnetic bias by a magnetic driving field effective in the hard direction, which is equal to or greater than the saturation field strength H _K which the magnetization tries to keep in the hard direction.

In the case of unequal initial permeabilities. ". 42> ! Αι it can be achieved by a suitable choice of the length-width ratio a: b that the initial inductances L ₁ and L ₂ are the same for the magnetic layer elements 1 and 2, that is, the Fulfill the adaptation condition L ₁ = L _2. Assuming u _A2 \ fi _Al = 9: 1 (cf. FIGS. 12a and 12b), the geometric dimension ratio would have to be

b V! ' _Al

= f9 = 3

(59)

55

can be selected (see. Fig. 1). Which are up for this The resulting Φ / 7 characteristics are shown in 13a for the controlling magnetic layer element 1 and in FIG. 13b the controlled magnetic layer element 2.

It is clear that for controlling and controlled magnetic layer elements or arrangements which are of unequal magnetic volume, the above conclusions are to be modified accordingly. In any case, the signal transmission properties can be achieved by fulfilling the adaptation condition in the present coupling devices between controlling and controlled magnetic layer elements improve in the manner shown by purely topological measures.

Reference is made to FIG. 14, which shows, as a further application example for the present invention, a 2-clock shift register in a schematic representation. On an electrically conductive base plate 50 there are - if necessary separated from the base plate by a thin insulating layer - several approximately rectangular magnetic layer elements which are assigned to cycle groups A and B alternately. In the figure, five magnetic layer elements are drawn, namely the ^ elements 51, 53, 55 and the β elements 52, 54. The longitudinal direction of the magnetic layer elements is parallel to one another, which is at an angle to an axis 42 connecting the centers of the elements is inclined by at least about 45 °. The magnetic layer elements, which have a thickness of only a few hundred or thousand angstrom units (IA = 10 ~ ⁸ cm), consist of a ferromagnetic material, e.g. B. Permalloy of the composition 80% nickel and 20% iron. They have a uniaxial magnetic anisotropy; their easy direction, symbolically represented by the double arrow 43, is inclined in all elements with respect to the axis 42 by a small angle f. This angle e can be approximately between 2 and 15 °; he is z. B. assumed with 6 °. The binary information 1 is represented by a magnetization oriented in the easy direction to the right and the binary information 0 by a magnetization oriented in the easy direction to the left.

The magnetic layer elements are variable width via coupling tape lines in accordance with the invention coupled together. In detail, the elements 51 and 52 via the ribbon line 61 are the elements 52 and 53 via the ribbon line 62, the elements 53 and 54 via the ribbon line 63, the elements 54 and 55 via the ribbon line 64, the element 55 with an optionally connected downstream another magnetic layer element via a further ribbon line 65, etc. coupled. The end faces 44 (shown in the drawing by thick lines) the ribbon cables are aligned with the one below Base plate 50 connected in an electrically conductive manner. With information flowing from left to right is provided, the arrangement of the ribbon lines is made so that the narrow part of the ribbon line on the left and the broad part on the right. The narrow strip conductor part corresponds at least approximately to that Width of a magnetic layer element, while the wide strip conductor part is approximately the length of a magnetic layer element is equivalent to. In the embodiment shown, the ratio between length is and width of the magnetic layer elements or between the wide and narrow strip conductor parts, for example accepted with 3: 1.

Further details of the structural design of the present 2-clock shift register emerge from the cross section shown in FIG. 15 along the line PQ of the arrangement of FIG.

At the top is the conductor arrangement 60 which carries a bias current i ₂ and is described in detail below. This is followed by an insulating layer 47 and therein the _{strip conductor 45 which carries a drive current J 1} and is so wide that it covers the magnetic layer elements; its longitudinal axis 42 runs parallel to the line connecting the centers of the

23 24

Elements. On the underside of the tape conductor 45, etc .; Finally, as shown in the comparison, there is a thin insulating layer 48. Underneath it is shown in FIG. 14, the center piece 75, in which the coupling strip conductor 64, with which the current Z _{2 flows} from left to right, is located above its end face 44 connected to the conductive base plate 50 to the element 55. To achieve the desired. The space between coupling 5 alternating reversal of the current direction in the strip conductor and base plate is with insulating middle pieces of the conductor arrangement 60 are the material 49, z. B. vapor-deposited silicon oxide, filled from power supply points to the individual Mittelstük. Between the base plate 50 and the ken alternately right and left. All centerpieces; Coupling strip conductor 64, embedded in the insulating _{material 49 in which the current Z 2 flows} from left to right, lies above the magnetic layer element 10, the current is via a lead 46 attached to a left upper element 55 of the coupling strip conductor 65 and the axis of the coupling strip conductor 64 coincides with the cross section PQ at a lower right connection point, the axis of the connected discharge line 56 runs away. All central coupling ribbon conductors 65, viewed perpendicularly to it, in which the current Z _{2 flows} from the right to the position of the magnetic layer element 55. Below 15 on the left, the current becomes an information lower than an assumption made on a right of the above Connection point attached supply line 56 in the direction of flow from left to right provides the coupling and discharged via a discharge line 46 attached to a left upper connection ribbon line 64 with respect to element 55. (The one input line and the coupling ribbon line designations “top” and “bottom” relate to an output line. In the present embodiment, refer to the graphic representation in FIG Axis 42 flowing current flow and output coupling lines perpendicular to one another to achieve in the middle pieces, these are with each other, whereby a good decoupling, a plurality of narrow elongated slots 66 is achieved. The distance between the narrow part provided. To return the current Z ₂ , the ribbon line 65 and the base plate 50 are used at the 25 preferably the same location as shown in FIG. 16 of the magnetic layer element 55 and the spacing of the conductor arrangement 60 ', which is attached to elements below the magnetic layer between the wide part of the ribbon line 64 and that of the magnetic layer element from the current Z ₂ the point 55 are flowing through it in the opposite direction to each other in a ratio such that both by the constants of a bias current Z ₂ Volume lines as possible the same characteristic 30 flows through the conductor arrangement 60, 60 'come the Have impedance. The transition from the smaller to the ^! - elements and the .B elements under the influence of a larger distance of a strip conductor takes place preferably perpendicular to the axis 42 directed, constant magnetically continuously. In the case of the production of the table prestressing fields H ₁₁ . These fields are shown in ribbon lines with the aid of a vapor deposition process. FIG. 14 by correspondingly marked arrows is symbolically represented continuously by at the corresponding points. The current Z ₂ is determined in such a way that the increasing thickness of the insulating intermediate layer 49 ensures that the absolute value of these biasing fields is the same as the amplitude of the distance _{ratio of the driver current Z 1.} generated magnetic fields. Accordingly, the conductive base plate 50 can, for. B. from one in each case the superimposed fields with respect to the thin copper foil or from a group of elements (e.g. with respect to the produced thin copper layer, the B elements), while they are with respect to the other is applied to a glass plate 57. Under this group of elements (e.g. with regard to the A-Eh glass plate is the strip conductor 45 'provided for the return of the element) in its effect on these elements under currents I _1. Then follow support. The dimensioning is chosen so that an insulating layer 47 'and the conductor arrangement 60' provided for the return line 45 in the latter case as a whole for the relevant current Z _2. Magnetic layer elements effective field amplitude In the event that the conductive base plate 50 for the rear is greater than the anisotropy field strength H _K for conducting the current Z ₁ , magnetic layer elements, that is, their magnetization, of course, the strip conductor 45 'and the insulation are perpendicular to the axis 42 is deflected, layer 47 'is not required, but is then located below- 50 In the present exemplary embodiment, the outside of the glass plate 57 immediately finds the conductor arrangement directing the magnetization of the ^ elements to 60', whose function, however, may also be at the top, that of the B- Items down instead. can be taken over by the conductor 50. The pulse program for the through line 45 It is now on FIG. Referring to FIG. 16, what flowing drive current Z ₁ is in FIG. 17, further details of the conductor assembly 60 are shown. 55 The current flows alternately in positive and The shape of the conductor arrangement is chosen so that negative direction. The switching time τ is that of the constant biasing time flowing through it, which it takes until the constant current amplitude current i _{2 is} reached through the ^ elements in one direction of opposite polarity. (z. B. to the right) and over the B elements by superimposing the _{direction generated by the current Z 1 in} the opposite direction (ie after 60 magnetic fields with the biasing fields H _n are left). The conductor arrangement 60 is in relation to the Λ-elements in Fig. 18 darden magnetic layer elements 51 to 55 that the set driving fields H _A and with respect to the B-EIe middle pieces 71 to 75 of the conductor arrangement just mente the in Fig 19, the driving fields H _B shown come to rest over these magnetic layer elements.

that is, the center piece 71, in which the current Z ₂ 65 at the point in time f _x the driver _{current Z 1} im flows from left to right, is switched from minus to plus via element 51, conductor 45. That the middle piece 72, in which the current Z ₂ from has the coming into effect or switching on of the driving right flows to the left, lies above the element 52 field H ₄ of 0 or a value which is smaller than

the anisotropy field strength H _K , to a maximum value H ₀ , which is greater than H _K , result. Synchronously with this, the driving field H _{B is switched off} from the maximum _{amount H 0} to the value 0 or to a value that is smaller than H _{K at the time f t} . Thus, at time I _1, the magnetization of the /! Elements is deflected from the easy direction in a direction which is perpendicular to axis 42 (ie upwards in FIG. 14), and at the same time the magnetization of the B elements is switched off the direction perpendicular to the axis 42 back in the easy direction (from below in FIG. 14). At time t _x , binary information is transferred from the ^ elements to the B elements. Because of the non-reciprocal transmission conditions of the coupling arrangement, the B elements take over the information from the / 1 elements adjacent to the left, as will be shown further below.

At time t ₂ , the drive field H _{A is} switched off and the drive field H _{B is} switched on. The magnetization of the / 4 elements switches from the direction perpendicular to the axis 42 (in FIG. 14 from above back into the easy direction; the magnetization of the B elements is simultaneously deflected from the easy direction (in FIG g. 14 downwards) At time i ₂ , binary information is transferred from the β elements to the /! elements, namely - for the reason given above - the /! elements take over the information from the left neighboring B elements.

At time i ₃ , the drive field H _{B is} switched off and the drive field H _{A is} switched on. This results in the same operating conditions for the magnetic layer elements as at time i _t . The operating conditions at time i ₄ are the same as at time i ₂ .

For the mode of operation of the present 2-clock shift register, in addition to the non-reciprocal Transmission properties that result from the geometric shape of the magnetic layer elements and coupling lines can be achieved, the following features are characteristic: The inclination of the easy Direction 43 at a certain angle to the axis 42 of the driver lines and the inclination of the coupling lines 61 to 65 at about 45 ° to the axis 42 of the driver lines (see FIG. 14).

The provision of an inclination f in the easy direction to the axis of the drive lines means that the magnetization of a magnetic layer element is deflected by the magnetic drive field H _A or H _B in a direction which deviates from the hard direction by the angle /. This angle of inclination prevents the undefined switching back when the driving field is switched off and the possible associated splitting of the magnetization of the controlled element into many small domains in opposite directions of magnetization. An angle f between 2 and 15 ° proves to be favorable. The above-mentioned splitting of the magnetization occurs if, during the switching off of the driving field, a control field component originating from a control current in the coupling line and defining a clear switching back of the magnetization of the controlled element is not present in the easy direction. A certain angle of inclination > ■ is effectively equivalent to a constant control field component H , based on the easy direction. If, therefore, at the moment of switching off the driving field acting with respect to the controlled element, no control current of sufficiently large amplitude is induced from the controlling element into the coupling line, then - because of the angle of inclination e - the magnetization switches. controlled element nevertheless back in a predetermined easy direction, namely in the next adjacent easy direction. This predetermined easy direction is the "O" position for the /! Elements 51, 53, 55; for the B elements 52, 54, however, it is the "1" position. This difference is that the drive fields H _A for the A and drive fields H _B for the B elements are of opposite polarity.

The inclination of the coupling lines at about 45 ° to the axis 42 of the driver lines is asymmetrical Transmission properties for the binary values 1 and 0 result, as mentioned in the introduction Publication »Bull. SEV «is explained.

The asymmetrical transmission properties mentioned are added in the present exemplary embodiment on the basis of the topological measures according to the invention with regard to optimal adaptation A discrimination of the magnitude of the control effect of the signals is added as a function whether it is control signals in the direction of the desired flow of information or act in the opposite direction, which is the non-reciprocal transmission properties of the present Shift register conditional. The aforementioned discrimination in the tax effect of the "forward" (i.e. in the desired information flow direction from left to right) and that to "backwards" (i.e. against the desired direction of information flow) allows control signals acting operation in a 2-stroke system in the first place.

It is a characteristic of the 2-clock shift register considered here that, as a result of the topological arrangement of the coupling lines, the binary information is regularly inversed when information is transmitted from one element to the next. A z. B. “1” in A element 51 is taken over as “0” by the B element 52 adjacent to the right and as “1” again during the next measure from the adjacent ^ element 53 on the right, and so on.

Before time J ₁ , the binary information to be transmitted is in the /! Elements; it is represented by the "magnetization vectors", ie by the direction of magnetization in these elements, which because of the driving field H _A = 0 (see FIG. 18) run parallel to the easy direction. The magnetization vectors of the B elements are deflected downwards at the same time because of the driving field H _B > H _K (see FIG. 19). At time T ₁ , the driving field H _A begins to rise and the driving field H _B begins to decrease. As a result, a deflection moment is exerted on the magnetization vectors of the ^ elements, which deflects them upwards orthogonally to the axis 42. As soon as a noticeable change in the magnetic flux becomes effective in the / i-elements with respect to the coupling lines after the onset of this deflection process, a feedback effect occurs from the B-elements, which weakens the rotation of the / 1-magnets.

809 540/300

tization vectors (this is the magnetization in the .4 elements). This feedback effect is caused by the following processes: The change in flux caused by the deflection of the A magnetization vector induces a voltage in the relevant coupling line, and a current flows in this line which generates a magnetic field with respect to the relevant β element ; this field causes a rotation of the β magnetization vector, which also causes a change in flux and consequently a back EMF which weakens the current in the coupling line. It can be said that - if the losses are neglected - a perfect coupling between two magnetic layer elements requires equal but opposite flux changes in both elements. In detail, for example, consider the .B element 52, which is simultaneously under the control of its left and right adjacent ^ elements 51 and 53 (see FIG. 14).

As a starting point for the Λ magnetization vectors, it is initially assumed, for example, that there is a "1" in element 51 and a "0" in element 53 (case 1). The deflection of the magnetization vector of the element 51 under the influence of the driving field H _A causes a flux change three times larger with respect to the coupling line 61 than the deflection of the magnetization vector of the element 53 with respect to the coupling line 62, and it is formed with respect to the B element 52 produces a resulting control field, under the influence of which the magnetization vector of element 52 switches back from the lower position (orthogonal to axis 42) in the corresponding easy direction. The switching back of the magnetization vector of the element 52 takes place in the sense that this change in flux counteracts the above changes in flux. The back EMF induced by the element 52 in the coupling line 61 is smaller by a factor of 3 than the back EMF induced by the element 52 in the coupling line 62. This means that the element 52 presents a small resistance with a view to a large effect emanating from the element 51 and that it presents a high resistance with regard to a small effect emanating from the element 53. These two effects are somewhat attenuated by another, however: a current in the coupling line 61 produces a control field three times smaller with respect to the element 52 than an equal current in the coupling line 62 does with respect to the same element. Overall, the non-reciprocal transmission properties of the in FIG. 14 by the simple length-to-width ratio, which was assumed to be 3: 1, for example, determined and not by the square of the same.

In the above assumed case 1 (i.e. "1" is in element 51 and "0" in element 53) the Transfer of information to element 52 - as can be seen from the analysis given - by the left adjacent element 51 is determined. This is due to currents in the coupling lines 61 and 62

(see Fig. 19) is the magnetization vector of the element 52 switches back from the lower position orthogonally to the axis 42 via the hard direction into the "O" position.

To assess the transfer of information in the present shift register, three further starting points must be considered, and it must be found that the information is always transferred from element 51 (and not from element 53) to element 52 in these cases as well. If there is a "0" in element 51 and a "1" in element 53 (case 2), the change in flux caused by element 51 and resulting in relation to coupling line 61 is very small. The change in flux caused by the element 53 and resulting in relation to the coupling line 62 is also negligibly small. The control field acting on the element 52 and resulting from the very small induction currents in the coupling lines 61 and 62 is therefore also very small and of no decisive influence. The magnetization vector of the element 52 deflected downward by the driving field H _B at the point in time ^ switches to the next adjacent, in this case back into the right, easy direction, ie to the "1" position, as a result of the intended angle of inclination e.

If both elements 51 and 53 contain the same binary values "1" (case 3), then that is the result of the Element 53 caused the change in flux resulting with respect to the coupling line 62 to be negligible small. In contrast, the one caused by the element 51 and resulting with respect to the coupling line 61 is Flux change has a decisive influence, and the magnetization vector of element 52 becomes switched back from the lower position orthogonally to the axis 42 via the hard direction in ^ the "O" position.

If the same binary values "0" are present in both elements 51 and 53 (case 4), the change in flux caused by element 51 and resulting in relation to coupling line 61 is very small. The change in flux resulting from the element 53 with respect to the coupling line 62 causes a control field component in the right easy direction. The effect of this control field component is supported by the predetermined angle of inclination p, so that the magnetization vector of the element 52 switches back in the right easy direction, ie in the "1" position. It is now shown on FIG. 20, where an at least qualitative analysis of the control fields occurring during the transmission of information in the present 2-clock shift register with respect to the element 52 is given in a graphic way. The control field resulting from the induction current in the coupling line 61 is called mjt H _ic) and is called "control field in the forward direction"; the control field resulting from the induction current in the coupling line 62 is denoted by H _(r) and it is called "control field in reverse direction".

As has already been shown, because of the non-reciprocal transmission properties of the control

effect H _(r) : H _{r) with respect to element 52 equals 3: 1.

resulting control field with respect to the element 52 65 The switching back of the magnetization vector has a sufficiently large component in the element 52 is determined by the polarity of the left light direction, so that during the resulting control field components in the light sounding of the driving field H _B > / i _K to the value H _B = 0 direction at the time when the component

30th

in the hard direction of the decaying driving field H _B approximately the value of the saturation field strength | H _{K of} the magnetic layer. The in F i g. The point in time recorded 20 is shortly after J ₁ (cf. FIGS. 18 and 19). that is to say that there are already significant flux changes due to deflections of the magnetization vectors of the elements 51 and 53 with respect to the coupling lines 61, 62 .

In the diagrammatic representation used in FIG. 20, the coupling line 61 is shown schematically by its axis 161 . The control field in the forward direction H _(r) is perpendicular to the axis 161. The coupling line 62 is shown schematically by its axis 162 . The control field in the reverse direction H _fr) is perpendicular to the axis 162. The lines 45 and 60 carrying the driver currents I ₁ and I ₂ are shown schematically by their common axis 160. The driving field H ₈ is perpendicular to the axis 160. The easy direction of the magnetic layer element 52 is indicated by // _, the hard direction by H ^ . As is known, the switching behavior of a magnetic layer is represented by the so-called "critical curve", the asteroids 170. The peaks of the critical curve correspond to the saturation field strength H _{K of} the magnetic layer. The inclination of the easy direction to the axis 160 is represented by the angle e .

in the above case 1 (ie "1" is in element 51 and "0" in element 53) the control field // "" originating from element 51 and the control field H _{r originating from element 53 ) are effective, which deal with superimpose the driving field H _B to the vector 163. The tip of the vector 163 is to the left of the point

2. Control field H _{rj in reverse direction, which is based on the action of a control element adjacent to the right;

3. Constant control field component H due to the inclination angle f of the easy direction to

Axis of the driver lines.

In the following two tables, all possible starting positions of the magnetic layer elements are compiled in order to understand the mode of operation of the present 2-clock shift register. Table 1 refers to the times t _t , t ₃ , r _s etc .; In particular, the elements 51 and 53 acting as controlling elements at this point in time and the B element 52 acting as the controlled element at this point in time are considered.

Binary information in the /! Element 51

Binary information in the /! Element 53

Size factor and effect of // "., On the B element 52

-H*

during the period in which the driving field from its maximum value to a value less than
subsides. Thus, during the period of time that is decisive for the switchover, there is a resulting control field component acting in the slight left direction, under the influence of which the magnetization vector of the element 52 switches back to the "0" position.

In case 2 above (ie "0" is in element 51 and "1" in element 53) // "., And // _{(r) are} negligibly small, and the resulting control field component is practically solely due to the component in the light Direction of the driving field H _B determined. Because of the angle f, this always acts in the slight right direction, so that the magnetization vector of the element 52 switches back to the "1" position.

In case 3 (ie "1" is in both elements 51 and 53), H _{(r) is} negligible, and // · "., And H _B are superimposed to form vector 164. As in case 1, the tip of vector 164 lies to the left of the point - H _K , and the magnetization vector of the element 52 switches back to the "0" position.

In case 4 (ie "0" is in both elements 51 and 53) // "., Negligible, and H _lr) and H _B overlap to form vector 165. The tip of this vector lies to the right of point - Η _Λ , and the magnetization vector of element 52 switches back to the "1" position.

As shown in FIG. 20 , the resulting control effect on a controlled magnetic layer element is composed of the following three components:

1. Control field H _{v) in the forward direction, which goes back to the action of a control element adjacent to the left;

Size factor and effect H _K \ 35 of // ,,, on the B element 52

Size factor and effect of Η _ε on the B element 52

Size factor of the control effect and end position of the B element

Binary information in the B element 52

Binary information in the B element 54

Size factor and effect of // (,.) on the / 4 element 53

Size factor and effect of // _(P) on the /! Element 53

Tabel

"0" "0" "1" "0" "1" "0" - - three
- ♦ »0« one
—► »1« - one
-""1" one
-> »1« one
-""1" one
-> »1« two
"1" one
"1" one
"0"

"1"

"1"

three

-> »0«

one

-> »1«

two "0"

Tabel

"0" "0" "1" "0" "1" "0" three
-> »1« three
-> »1« - - one
- »» 0 « -

"1"

one - »» 0 «

Size factor
and effect
from Η _ε to that
^ Element 53

Size factor
the control effect and end position of the / 4 element 53

one one one -> · »0« -> »0« -> »0« two one one "1" "1" "0"

one

two "0"

IO

Table 2 refers to times t ₂ , i ₄ , t _b etc .; the S elements 52 and 54 acting as controlling elements at this point in time and the B element 53 acting as a controlled element at this point in time are considered there. The tables apply to the other elements in the same way.

The analysis of the possibilities presented in the two tables shows - as already mentioned - that there is a clear flow of information from left to right and that the binary information from one element merges with the neighboring element on the right in inverted form. This regular However, inversion does not pose any disadvantage to the practical usefulness of the shift register represent.

Claims (20)

Patent claims: 1. Signal transmission arrangement, which has a coupling ribbon line with over its longitudinal extent varying width, a controlling and a controlled magnetic layer arrangement, the are interconnected by the coupling ribbon line, as well as deflection means for temporary Includes deflection of the magnetizations of the magnet layer arrangements, characterized in that that for the purpose of improving the transmission properties, the overall extent of the magnetic layer arrangement transversely to the direction of the coupling tape line as well as the width of this line in the area of the magnetic layer arrangements for the controlling and the controlled magnetic layer arrangements are unequal are that at least one of the two magnetic Layer arrangements have a different overall extent parallel and orthogonal to the coupling ribbon line and that in both magnetic layer arrangements the products of magnetic layer thickness and the total extent parallel to the coupling ribbon line are of different sizes. 2. Arrangement according to claim 1, characterized in that that the controlling and controlled magnetic layer arrangements have geometrical dimensions have according to the relationship (46) Wed
M- '2IHlIH '2 »2 / Ί where: a _t = total linear extension of the controlling magnetic layer arrangement parallel to the longitudinal axis of the coupling ribbon line; b ₂ = linear overall extent of the controlled magnetic layer arrangement parallel to the longitudinal axis of the coupling ribbon line; A ₁ = magnetic layer thickness of the controlling magnetic layer arrangement; d ₂ = magnetic layer thickness of the controlled magnetic layer arrangement; //! = Permeability of the controlling magnetic layer arrangement; // ₂ = permeability of the controlled magnetic layer arrangement; | H ₂ L _n = minimum value of the control field strength in relation to the controlled magnetic layer arrangement; H _K = saturation field strength of the magnetic layers. 3. Arrangement according to claim 1 or 2, characterized in that the linear dimensions of the controlling and controlled magnetic layer arrangements in the to the longitudinal axis of the Coupling ribbon line orthogonal direction and the respective width dimensions of the Coupling ribbon lines are practically the same size. 4. Arrangement according to claim 1 or 2, characterized in that the rise time τ and the amplitude H ₀ reached after this time of the driving field generated by a first deflection means and acting on the controlling magnetic layer arrangement are determined according to the following relationships: _L = 3 ^ Γ ιΛ7 7i O ₁ L V b ₂ d ₂ d ₂ 3 a ₂ \ I H ₂ 1 Ci ₁ (I ₁ £ 1 H _K in which, in addition to the definitions already given, the following are added: Jj ₁ = total linear extension of the controlling magnetic layer arrangement orthogonal to the longitudinal axis of the coupling ribbon line; a ₂ - total linear extension of the controlled magnetic layer arrangement orthogonal to the longitudinal axis of the coupling ribbon line; T _x = / I ₁ ■ - £ ■, where R is the resistance 60 the coupling ribbon line is. 5. Arrangement according to one of claims 1 to 4, characterized in that the distances zwisehen Coupling ribbon line out and back to the widths of the coupling ribbon line are in at least approximately the same ratio at least at the location of the magnetic layer arrangements. 6. Arrangement according to one of claims 1 to 5, characterized in that at least at the beginning of the signal transmission, the permeability u _{2 of} the controlled magnetic layer arrangement is more than 3 times greater than the permeability //, the controlling magnetic layer arrangement. 7. Arrangement according to one of claims 1 to 6, characterized in that the controlling and / or the controlled magnetic layer arrangement each comprise a plurality of thin magnetic layers, which have the same geometric dimensions as one another. 8. Arrangement according to claim 7, characterized in that indicates that the controlling magnetic layer arrangement has at least two thin, magnetic Comprises layers which with respect to the current flowing in the coupling ribbon line in Series connection are arranged (F i g. 4). 9. Arrangement according to claim 7, characterized in that the controlling magnetic layer arrangement comprises at least two thin, magnetic layers which with respect to the current flowing in the coupling ribbon line in Are arranged in parallel. 10. Arrangement according to one of claims 7 to 9, characterized in that the controlled Magnetic layer arrangement comprises at least two thin, magnetic layers, which with respect to the current flowing in the coupling ribbon line are arranged in series. 11. Arrangement according to one of claims 7 to 9, characterized in that the controlled magnetic layer arrangement has at least two thin, includes magnetic layers which with respect to those flowing in the coupling ribbon line Current are arranged in parallel (Fig. 5). 12. Arrangement according to claim 7, characterized in that the total linear extent α of the controlling magnetic layer arrangement parallel to the longitudinal axis of the coupling tape line is at least approximately equal to the total linear extent of the controlled magnetic layer arrangement orthogonal to the longitudinal axis of the coupling tape line, that the total linear extent b of the controlling magnetic layer arrangement is orthogonal to the longitudinal axis of the coupling tape line is at least approximately equal to the total linear extent of the controlled magnetic layer arrangement parallel to the longitudinal axis of the coupling tape line, that the controlling and controlled magnetic layers have at least approximately the same thickness (Fig. 1) and that the controlling and controlled magnetic layer arrangements have at least approximate geometric dimensions according to the relationship a
T 45 where / ^ _{1 is} the permeability of the controlling and μ _{2 is} the permeability of the controlled magnetic layers approximately at the beginning of the signal transmission. 13. The arrangement according to claim 8 or 9, characterized in that part of the a number of thin magnetic layers belonging to the controlling magnetic layer arrangement A sufficiently large control effect with respect to the controlled magnetic layer arrangement for signal transmission is able to generate. 14. Arrangement according to claim 8 or 9, daaurch characterized in that at the control all belonging to the controlling magnetic layer arrangement thin, magnetic layers a flux change in the same direction with respect to the Generate coupling ribbon line. 15. The arrangement according to claim 8 or 9, characterized in that the modulation a certain number from the total number of those belonging to the controlling magnetic layer arrangement thin, magnetic layers one in comparison to the other controlling layers generate opposing flux change with respect to the coupling tape conductors. 16. Arrangement according to claim 10 and 11, characterized in that the controlled magnetic layer arrangement has at least two thin, includes magnetic layers which with respect to those flowing in the coupling ribbon line Electricity connected in series, and at least two thin, magnetic layers includes, which are arranged in parallel. ! 7. Arrangement according to Claims 10 and 11, characterized in that the controlling magnetic layer arrangement comprises the same number of thin magnetic layers as the controlled magnetic layer arrangement and that the Magnetic layers of the controlling and the controlled magnetic layer arrangements of the same volume have (Fig. 6). 18. Arrangement according to one of claims 1 to 17, characterized in that at least two transmission stages are present (F i g. 9), wherein in a first transmission stage a first magnetic layer arrangement (81) with the linear overall dimensions a _x (parallel to the longitudinal axis of the first coupling ribbon line), b _t (orthogonal to the longitudinal axis of the first coupling ribbon line) and the magnetic layer thickness ά _χ a second magnetic layer arrangement (85) with the linear overall dimensions a ₂ (orthogonal to the longitudinal axis of the first coupling ribbon line and parallel to the longitudinal axis of the second coupling ribbon line), b ₂ ( parallel to the longitudinal axis of the first coupling ribbon line and orthogonal to the longitudinal axis of the second coupling ribbon line) and the magnetic layer thickness d _{2 is able} to control that in a second transmission stage this second magnetic layer arrangement is a third magnetic layer arrangement (82-84) consisting of several magnetic layers with the total linear dimensions HO ₁ (Orthogon al to the longitudinal axis of the second coupling ribbon line), ^ ₁ (parallel to the longitudinal axis of the second coupling ribbon line) and the magnetic layer thicknesses d _t and that the magnetic layer arrangements have geometric dimensions according to the relationships a ₂ ■ b ₂ - a ₂ U ₁ Jt b ₂ ⁿ 'ET = η 19. Arrangement according to one of claims 1 to 18, characterized in that the magnetic layers the magnetic layer arrangements have a uniaxial anisotropy of their magnetization, that first deflecting means are present of such a nature that at the time the transmission of information has an effect on the controlling magnetic layer arrangement first driving field to operate that second deflecting means of such a nature are present that they are at about the same point in time on the controlled magnetic layer arrangement acting second driving field little 809 540/300 at least approximately make it disappear, and that the arrangement of the deflection means made so is that the driving fields are at least approximately in the hard direction of the magnetic layers. 20. Arrangement according to claim 18 or 19, characterized characterized in that for the formation of a 2-clock shift register the first (controlling) Second magnetic layer arrangement (51) connected downstream via a first coupling ribbon line (61) (controlled) magnetic layer arrangement (52) via a second (62) coupling ribbon line with a downstream third magnetic layer arrangement (53) is coupled, the same operating conditions is subject, like the first magnetic layer arrangement, that the first coupling ribbon line the output coupling ribbonline of the first and the input coupling ribbonline of the second and that the second coupling ribbonline is the output coupling ribbonline of the second and the input coupling ribbon line of the third magnetic layer assembly represent that further the axes of the input and output coupling ribbon lines belonging to the same magnetic layer assembly form a right angle with each other and the width of the input coupling ribbon line at least ^ times the width of the output coupling ribbon line and that the deflecting means relate to at least one all three magnetic layer arrangements comprise strip conductor-shaped driver lines (45, 45 ') that act, their axis in a manner known per se with the axes of the coupling ribbon lines each one Angle of at least approximately 45 ° and one with the easy direction of the magnetic layers Forms an angle between 2 and 15 °. Considered publications: British Patent No. 867,723; Bull. SEV, Vol. 51, October 8, 1960, pp. 1004-1010; Solid-State Circuit Conference, February 1961, pp. 64 and 65. For this purpose 2 sheets of drawings M9 540/300 4.61 β Bundesdruckerei Berlin