DE1951741A1 - Device and method for detecting errors in flexible magnetic storage media - Google Patents

Description

PATB MTANWAIiT

DIPL. ING. ß. WOODS

r-9 AUOHJiUIiG

PHIUPPINB-WEISEK-STBASSB U

I.

Augsburg, October 9, 1969

International Business Machines Corporation, Armonk, N.Y. 10504, United States of America

Apparatus and method for locating defects in flexible magnetic storage media

The invention relates to a device and a method for Plagues of defects in flexible magnetic storage media '.

In previous magnetic tape test methods, the procedure is such that either a strong one that saturates the tape

009825/1760

Direct current signal or one that also saturates the band Alternating current signal or a high frequency alternating voltage signal which is so weak that it can only be used up to penetrates the tape to a shallow depth. Each of these recording methods is for plague suitable for special types of tape defects. However, none of these methods can provide all types of P belt errors can be successfully detected.

In the case of the test method using a DC-saturated tape, the tape connected to the read head takes Test circuit on the pulses which are picked up by the reading head from the saturated tape. The read head generates these pulses when there is a flaw in the tape while the DC saturated tape is moving past causes a positive change in the read head under the read head. fe The weakness of this test procedure is that it for shallow holes in the tape or bumps on the tape,

-4 which are on the order of 2.5 * 10 mm, only has very low sensitivity.

In the case of the test method using an alternating current-saturated A test circuit connected to the read head provides the envelope of the saturated tape

0Ö9Ö25 / 1780

Tape picked up AC voltage signal. The amplitude of the envelope of this AC voltage signal corresponds to the thickness of the oxide layer on the magnetic tape.

Large flaws with a diameter of for example

"2
2.5 * 10 mm mean that the tape is much thicker or much thinner than normal at these points. The envelope of the AC voltage signal changes when a hole or a bump moves past the read head. This test method is inadequate in that it does not allow the plotting of shallow holes or bumps or deep, small-diameter holes to be determined.

In the tape test method using alternating voltage signals with a low penetration depth, the one with the read head connected test circuit determines the amplitude of each half-wave of the AC voltage signal and checks whether this amplitude is below a certain threshold value, This test circuit is composed of several parts, because it contains a circuit which takes a sample from the peak part of the recorded AC voltage signal and also a threshold value circuit which determines when the selected sample part of the AC voltage signal exceeds a certain threshold value. Between the test circuit and the Frequency of the signal read out by the read head must

009825/1760

BATH ORIGINAL

synchronous operation can always be guaranteed. The frequency of the signal recorded on the tape and the tape speed must accordingly be precisely controlled. If the test circuit, the recording frequency and the If the belt speed is set correctly, this method can produce very shallow imperfections in the Belt surface can be determined. The weakness of this method, which works with alternating voltage signals of low penetration depth, consists in the fact that there are flaws small diameter, i.e. on the order of 2.5 * 10 mm regardless of how deep they are formed in the band, cannot be determined. The many parts of the circuits and their correct setting make that Processes are expensive and also difficult to work with.

Current full testing procedures require that the tape be tested by a combination of Circuits is passed through according to the methods mentioned above, so that all types of faults are detected can. In a typical combination, direct current saturation will be used to determine any deep holes in the tape and furthermore a test with alternating voltage signals of low penetration depth to determine the »ridges flat defects applied. With this combination

00982sVi760

BATH

however, no small diameter imperfections can be found. In addition, there is a large outlay on equipment required because two different tests have to be carried out and because that is less with AC voltage signals Penetration depth working method requires a particularly large amount of equipment.

The aim of the invention is to solve the problem of serving all fault locations in flexible magnetic storage devices including those with a depth or depth

-li a diameter of the order of 2.5 x 10 mm.

The aim of the invention is to use a single test method to determine all small and flat To be able to detect defects in a flexible, magnetic storage medium. Finally, through the invention includes all imperfections, i.e. both flat and small in size, in a flexible one magnetic storage medium can be determined with the help of a test procedure, in which on the storage medium a direct current gradient is recorded.

The invention makes it possible, with the help

009825/1760

simpler, cheaper test devices in which little or nothing needs to be adjusted to be able to measure the surface roughness of flexible magnetic storage media.

In order to achieve the above object, the invention includes a device for plaguing Defects in flexible magnetic storage media which, according to the invention, are identified by a device for recording of a constant magnetic field in the medium, the field strength values of which are each a puncture of the depth of the constant magnetic field are in the medium and change as a gradient over the entire depth of the medium, whereas if there is a flaw in the medium, part of the magnetization gradient is lost, and which according to the invention further by a detector device for determining differences in the magnetic flux between adjacent surface parallels Levels of the medium are characterized, these flux differences being due to the differences in the magnetization gradient these adjacent levels of the medium are caused.

Reading out the errors in the storage medium

^003825/1760 BADOR ₁ G ₁ NAL

caused flux fluctuations is further improved by the fact that the readout head with an integrating Circuit is connected. The integrating circuit generates an output signal, its amplitude fluctuations exactly the surface fluctuations of the storage medium correspond. The integrating effect of the integrator is similar to the differentiating effect of the reading head from that it reproduces the flux fluctuations caused by the defects in the surface of the storage medium.

The advantages of the test device or the test method according to the invention are very diverse compared to previous test methods or test devices:

1. The test of the invention detects all defects, i.e. large, small, flat or deep bugs, and therefore there is no need to run a multitude of tests as in the past was necessary.

2. The test device according to the invention is characterized by simpler structure and operation compared to previous test devices. Because of this, the manufacturing and Maintenance costs of the device according to the invention are much lower.

- 7 -00 9 825/17 60

8AD ORIGINAL

19517Λ1 p

3. The signal of the test device according to the invention generated by the integrator connected to the read head is either an exact copy of the surface roughness or a measurement of the surface roughness depending on how wide the read head used is.

In summary, it can be said that the test device according to the invention provides more precise information about the condition the surface of the flexible magnetic storage medium in question, costs less and is easier to use is than previous devices.

The previously described and further features and advantages of the invention emerge more clearly from the following detailed description of a preferred embodiment shown in the accompanying drawings is shown. These show in detail:

Pig. I a strip of magnetic tape,

on which with the help of broad gauge magnetic heads is recorded or which is scanned with these,

Fig. 2 is a schematic circuit diagram of the in Pig.l

shown tape flaw detector,

- 8th - "
009825/1760

Fig. 3 shows a hysteresis loop for the

Magnetic tape,

Fig. 4 is a diagram of the field strength of the

Magnetic tape as a function of the distance from the magnetic head,

Fig. 5 in a schematic representation

Circuit diagram of a voltage comparator shown as a block diagram in FIG. 2,

Fig. 6 shows some typical frequency sensitive

speed curves of the circuits used in the tape defect detector,

Fig. 7 Schemes showing the interaction of

Explain read / write heads and magnetic tape of the device according to the invention, assuming that the device according to the invention has a hole in the magnetic Determines the oxide layer of the tape,

Fig. 8 Schemes showing the interaction of

Explain read / write heads and magnetic tape of the device according to the invention, wherein assume that the device is on the

- 9 009825 / 17SO

Surface of the tape detects adhering particles or bumps on the tape surface.

Referring to Figure 1, a preferred embodiment is shown in FIG Invention shown. A magnetic tape 10, which has a magnetic oxide layer applied to a base is moved past three magnetic heads. The magnetic tape 10 is in a known manner with the aid of negative pressure or kept in tension by spool drive motors. This tension creates a touch between the tape and the magnetic heads. If necessary, between the magnetic tape and the erase head or the read head an air bearing can be attached; however, the non-saturating print head must contact the tape directly. These three Magnetic heads are used for erasing or saturation, for unsaturated recording and for reading.

The magnetic head 16 is the erase head. It is fed from a current source 18. The current source 18 supplies the magnetic head 16 with such a high current that the oxide layer of the magnetic tape 10 is saturated. This deletion or ₀ saturation function is applied either by a DC power source or from an AC power source. According to the invention, the erasure can be done either with direct current or with alternating current; however, the best result is achieved by direct current erasing or by direct current saturation of the magnetic tape. This is described in detail below.

The magnetic head 20 records a non-saturating signal on the tape 10. It is made from a gradient current source 22 and creates a flux gradient in the magnetic tape. Plus gradient means that the flux parallel to the surface of the tape 10 decreases with increasing distance from the magnetic head. This flow gradient has result in a magnetization gradient which is recorded in the oxide layer 12 of the tape 10.

The variable power source is a direct current source. The person skilled in the art knows that a low-frequency increase in current source is equivalent to a direct current source if, in practical applications, the frequency is relative to Tape speed is so slow that DC recording is being made on the tape. The best or most effective Tape test method results when using a DC power source, which the tape 10 in opposite Direction of direct current saturation magnetization of the tape by the magnetic head 16 magnetized.

A magnetic head 2 * is used to detect errors. the tape 10 from changes in flow. The flux changes are amplified, integrated, and then in a tape flaw detector 26 compared with a reference value. When the change in flow exceeds a certain threshold,

- 11 009825/1 7SO

BATH ORIGINAL

the tape failure detector shows the presence of a

Error. The tape is then stopped and under examined under a microscope to see if the error can be corrected.

The illustrated in Fig. 1 magnetic read head 24 is a wide track head, which each over the edges of the Magnetic tape protrudes. The use of a single broad gauge reader head adds to the simplicity and low cost Cost of the test device according to the invention. If necessary, however, a single narrow read head can also be used, which only covers part of the band. In this case, however, a plurality of tape defect detectors are required and the cost of the test device according to the invention increases unduly.

Details of the tape defect detector 26 are shown in FIG shown. A winding 28 on the magnetic head 24 is grounded via a center tap. Each terminal end of the magnet winding 28 is connected to the base of an NPN transistor tied together. The npn transistors 30 and 32 are powered by a positive voltage source at the collector and by a current source 34 controlled at the emitter. The power source 34 acts on the transistor 30 through a resistor 36 and on the

- 12 00 9 825/1760

Transistor 32 through a resistor 38. The transistors 16 and 32 act as amplifiers for the magnetic head 2 * 1, which the latter amplifies the current from the winding 28 and forwards it to an integrator 39. The waveforms of the output signals each at the emitters of the transistors 30 and 32 are identical. However, these curves point with respect to earth an opposite polarity.

The outputs of transistors 30 and 32 become forwarded to the integrator 39, which consists of a differential amplifier 40 with input resistors 42 and 44 and with Feedback capacitors 46 and 48 are made. The integrator acts in a known manner as an integrating operational amplifier. This amplifier is intended to reduce the noise and thus to improve the signal-to-noise ratio a differential circuit has been chosen.

The two integrated signals emitted by the integrator are sent to a further differential amplifier 50, which reinforces this. The output of the amplifier 50 feeds a differential voltage comparator 52.

The differential circuit is selected for the tape defect detector shown in FIG. 2 because it is in operation

- 13 -

009825/1760

is more advantageous than a one-sided, unbalanced circuit. The differential circuit has a better one Signal-to-noise ratio, because the noise that is usually present is compensated for. The expert is however, it is clear that the device can also be operated in such a way that only one end of the reading coil 28 is one-sided Feeds integrator, amplifier and voltage comparator.

In Fig. 5 is a schematic circuit diagram of the differential voltage comparator 52 shown. With respect to earth positive pulses appear at input 54 and are via a Capacitor 57 passed to the base of an npn transistor 56. Negative pulses with respect to earth appear at input 58 and become the base of an npn transistor via a capacitor 6l 60 headed. The base of transistor 60 is normally at a threshold potential + TH. This ^ Potential or this voltage is applied from a voltage source + TH via a resistor 62 to the base.

The transistor 60 is normally biased so that it is "open" and conducts the current from a voltage source + V * which is connected to the collector of the transistor via a _; Resistor 64 is connected to a current source 66 which is connected through a diode 68 to the emitter. Of the

- 14 -

Q0-8'26 / V7tO

Transistor 56 is in the same way by that with his Collector connected voltage source V. biased and in addition, its emitter is connected to the current source 66 through a diode 70. The transistor 56 is normally non-conductive because its base is held at ground potential by a resistor 72 connected to ground. There the base of transistor 56 is connected to ground and there the The emitter of transistor 60 is at a somewhat positive voltage, is the base-emitter junction of transistor 56 reverse biased and the transistor 56 is therefore non-conductive.

Another npn transistor 74 in that shown in FIG Comparator circuit is normally non-conductive because transistor 60 is conductive and thus the Base of transistor 74 at a value below ground potential holds on. When transistor 60 becomes conductive, however, the voltage at the collector of the transistor increases on and the voltage at the base of transistor 74 in turn rises above ground potential and the transistor becomes conductive. When transistor x 74 becomes conductive, it falls the voltage at its collector connection from + V, to ground potential and the entire voltage V, is thus at a resistor 77. This voltage drop from V to

- 15 -009825/1760 BAD ORIGINAL

Earth serves as the output signal. Resistors 75 and 76 each serve only to set a bias voltage and thus to control the opening and blocking of the transistor 7 ¹ U. A capacitor 78 conducts high-frequency transient processes to earth. These transients occur while transistors 60 and 74 are switching over.

The differential voltage comparator shown in Fig. 5 provides an output signal at the collector of the transistor "Jk" whenever the differential voltage between the positive pulses at input 5 ^ and the negative pulses at input 58 exceeds the threshold voltage + TH. If this occurs, the Base-emitter junction of transistor 56 is forward biased, while the base-emitter junction of transistor 60 is biased in the reverse direction, so transistor 60 blocks and transistor 7M becomes conductive. The voltage drop at the collector of transistor 7k shows the presence of pulses, the differential sum of which is greater than the threshold value The threshold value can of course be set to any desired value and thus the size of the pulses which cause an output signal and thus indicate a fault in the magnetic tape can be controlled.

- 16 009825/1780

Another important feature of the tape error detector illustrated in Figure 2 is frequency sensitivity curves shown in FIG. 6. In Fig. 6A, the frequency sensitivity curve of the magnetic head 24 is shown. The magnetic head acts as a differentiator and has a frequency sensitivity of 6 dB / octave.

In Fig. 6b, the gain is shown as a function of the frequency of the integrator amplifier. Because of the integrating effect, the gain of the integrator amplifier decreases by 6 dB / octave. With the appropriate setting the input resistors and the feedback capacitors in the integrator it is possible to control the frequency range so that this gain drop is linear.

In Fig. 6C, the overall gain of the magnetic head and Integrator amplifier shown. The combined action of the head and the integrator amplifier results in the System has a flat frequency sensitivity curve in the range of approximately 500 Hz to 200 kHz. This area of flat frequency sensitivity is dependent on the selected values of the input resistors 42 and 44 and the feedback capacitors 46 and 48 adjustable. The bandwidth from 500 Hz to 200 kHz is for tape speeds in the

- 17 009825/1700

A magnitude of 2.5 m / s or 5 m / s is required.

The advantage of a flat frequency sensitivity of the tape flaw detector is that the amplified signal applied to the voltage comparator changes in direct proportion and, averaged over the width of the read head, represents an image of the surface roughness and the surface flaws. The differential signals from the integrator 39 shown in Figure 2 can be visually displayed so that they directly indicate the surface roughness of the tape. When using a wide magnetic read head, the signals from the integrator 39 represent an average Bandrauhigkeit along the width of the magnetic head _o when using a very narrow magnetic head can. actual surface changes moved by under the read head are displayed.

The manner in which tape defects were plagued The invention can best be explained with simultaneous reference to FIGS. 1, 3 and 4. The tape hits when moving from left to right, first hit the erase head l6. The erase head 16 is powered by a direct current source fed with a high DC current and saturates the

- 18 C09825 / 1780

Magnetic oxide layer 12 with a magnetic field directed directly to the left. The tape then moves into a position below the magnetic head 20. The magnetic head 20 is also fed by a direct current source and generates a magnetic flux or an H field in the tape, which is opposite to the magnetic head 20 magnetic field generated by the erase head 16 is directed in the opposite direction. In FIG. M, the field strength H is shown as a function of the distance from the magnetic head 20. The field strength H is shown on the ordinate and the distance from the magnetic head is shown on the abscissa. The curve shown in Fig. ¹ I is the familiar exponential curve and shows the removal of the field strength H with increasing distance from the magnetic head.

While the strength of the Η field below the magnetic head 20 is represented by the curve in FIG. ¹ J, one obtains the magnetization of the tape 10 from the hysteresis loop shown in Fig. 3 of the tape. In Figs. 1, 3 and 4, reference characters V, W, X, Y and Z respectively indicate magnetization at a certain distance from the magnetic head. For example, for a distance Z from the magnetic head, the field strength in the tape below the magnetic head is equal to H ". Before the tape is moved under the magnetic head 20, its magnetization is at a point 90 on the hysteresis loop according to FIG

- 19 -00982571760

Representation in Fig. 3. The H ₂ field of the magnetic head causes the magnetization of the tape to be shifted towards a point 92. When the tape leaves the magnetic head 20, the H field of the head no longer acts on the tape and the magnetization B_ of the tape thus follows a curve 93 up to a point 9 ^ at which the H field is equal to zero and the end magnetization is B-,

. The vector B ₁₇ in FIG. 3 corresponds to that in FIG

the magnetic head 20 for a depth Z from the magnetic head shown vector. In the same way, the other vectors in the depths V, W, X and Y calculated from the magnetic head are determined by 'that in each case the strength of the H-field at these depths in Fig, ¹ I curve shown is determined, and that the field strength values so obtained are then entered into the position shown in Fig. 3 hysteresis loop. The respective vector B or the respective magnetization at these depths then results in each case from the vertical distances according to the illustration in FIG.

As shown in Fig. 1, in the magnetic oxide layer 12 and recorded a magnetization gradient over the entire thickness of this layer. In the vicinity of the the upper edge of the oxide layer is the magnetization against

- 20 00 9825/1760

above the saturation magnetization recorded by the erase head vice versa. Further into the magnetic oxide layer, the magnetization decreases to zero. Even further into the magnetic oxide layer inside the magnetization remains in the direction of the saturation magnetization, but is opposite to this of smaller size.

Those skilled in the art know that the value of the gradient depends on the strength of the flowing in the winding of the magnetic head 20 Current and by the length of the writing air gap of the magnetic head 20 can be controlled. The magnetic head 20 can still record a magnetization gradient in the tape even when the magnetic tape passes through the magnetic head is saturated or not completely saturated with an alternating magnetization. However, the amplitude of the magnetization gradient is caused by an erasure or magnetization of the tape in the opposite direction to the gradient magnetization generated by the magnetic head 20 is increased.

As shown in FIG. 1, the tape is moved from the magnetic head 20 to the read head 24. Since the tape goes through the Magnetic head 20 has been magnetized by means of a constant field, the read head 24 only emits an output signal when if defects in the tape or surface roughness of the tape cause a change in flux in the magnetic head 24.

- 21 -

00 9 825/1760

The manner in which the invention improves the tracking of tape defects will be described with reference to FIG Figs. 7 and 8 explain. In Fig. 7 the interaction of read and write heads with a magnetic tape is shown, wherein the magnetic tape has a hole or a space.

In previous methods, holes in the tape are usually determined by first opening the tape by means of a Saturated constant field and that this saturated tape is then scanned by a read head. At a Such previous methods, the tape is fed past a saturation write head. If there is a hole in the tape there is only a small amount of magnetic oxide available in the vicinity of the hole for saturation. If this If the saturated tape is then passed over a reading head, a drop in flow through the reading head can therefore be detected will. The reduction in the flow read out in a certain position of the tape is made up of all of them Magnetization fields over the entire depth of the magnetic oxide layer this place together. If there is no hole in the magnetic oxide layer, it is the composite one Magnetization relatively large. On the other hand, the composite magnetization points in the vicinity of a hole or a vacancy

- 22 009825/1760

the same direction as the saturation magnetization, but it is smaller than the saturation magnetization.As is well known, the change in magnetization when the tape moves past a read head causes a change in flux in the read head, which is picked up as a pulse by a winding on the read head.

7A, 7B and 7C show the interaction of the magnetic heads with the magnetic tape in the test method according to the invention. The magnetic tape is saturated by the writing head 100. The magnetic tape then moves to a position above the non-saturating write head 110. It was already mentioned above that the non-saturating write head 110 creates a flux gradient through the magnetic oxide layer of the tape and accordingly a magnetization gradient in the magnetic oxide layer of the tape. If there is a hole in the tape above the non-saturating write head, part of the magnetization gradient disappears at that point on the tape, simply because there is no magnetic oxide at that point. The effect of this missing part of the gradient can be seen when reading out the gradient recording of the tape, cf. the illustration in FIG. 7C.

- 23 -

009825/1760

The read head 112 sees; facing a flow component, which is composed of the total magnetization gradient at a certain point on the strip. If over If there is neither a hole nor a blank space in the tape for the reading head, the reading head 112 reads the indicated by arrows 116 and 116 respectively. According to the As shown in Fig. 7C, arrows 114 and 116 have a

^m small size, since in the example shown the magnetic tape contains magnetization components of variable strength and opposite polarity. Combined from these components, there is in each case a small vector directed to the right, which is represented in FIG. 7C by the arrows 11h and 11 respectively. If, on the other hand, a hole or a void passes the read head 112, all magnetization components are absent in the gradient, except for a larger component directed to the left. The read head

|| So this one sees itself as a larger one, directed to the left Component, which is represented by an arrow 118, opposite. Accordingly, the flow in the read head changes at the movement of the tape past read head 112 from a small one to the right to a large one to the left directional vector. This substantial change in flux creates a pulse in the winding of the read head 112. The change in flow detected by the reading head in the method according to the invention is thus somewhat greater than one according to one of the previous methods from a read head

0 0 9 * 8 2.57 1760

detected change in flow.

Compared to previous constant field test methods, the invention brings a very substantial and important improvement in the detection of flat defects or flat bumps as shown in FIG. 8. A flat bump and the gradient constant field band test method of the present invention are illustrated in FIG.

With previous tape test devices, which were based on the direct current saturation principle work, a particle adhering to the surface of the belt causes only one very small change in flow in the read head. As shown in Fig. 8a, the foreign particle causes it to move past of the magnetic tape on the saturation write head 120, a bulge or tension of the magnetic tape. As the stylus however, when the tape saturates, the slight movement of the magnetic oxide layer away from the write head does not diminish the Ability of this write head to completely saturate the magnetic oxide layer, which has been deformed like a tent by the foreign particle. When the saturated tape is then passed over a read head, the foreign particle causes it again a noticeable deformation of the band. The effect in terms of on the read head is that the saturated magnetic tape moves away from this by a small distance. There the tape is saturated, is caused by the movement of the tape

- 25 009825/1760

it

the composite flux picked up by this head is not diminished by the read head. For this reason, the read head find no stool on the belt.

In the device according to the invention, however, the tape moves from a saturating write head 120 to a non-saturating write head 130. The write head 130 records a magnetization gradient in the tape which is opposite to the saturation magnetization recorded by the write head 120. Since the foreign particle pushes the magnetic oxide layer in its vicinity away from the non-saturating write head, the magnetization gradient in the tape opposite the flaw differs noticeably from the magnetization gradient immediately in front of and immediately behind the flaw. This is shown in FIG. 8C, in which the magnetic tape with recorded magnetization gradients is guided over a read head 132. The tape error in turn causes the magnetic tape to bulge when it passes the read head. If necessary, however, an air bearing can be attached to the reading head 132 so that the tape does not touch it and thus does not bulge. Important _; is only the composite flow which in every position

- 26 -

00 98 25/1760

is removed from the reading head 132 as the tape moves past. For parts of the band that do not have a hump or no foreign particle adheres, the composite flow generated in the reading head 132 is indicated by arrows 134 and 136 shown. As shown, these are short arrows pointing to the right. In the vicinity of the fault or however, part of the rightward magnetization of the foreign particle is missing because the tape defect has prevented the write head 130 from recording it. Accordingly the composite flux picked up by read head 130 is very strong and directed to the left, indicated by an arrow is indicated. When the tape is moved over the read head 132, a large change in flow occurs because of the composite river 136 a composite river 138 and this, in turn, a composite river 134 follows. This very large change in flux from 136 to 138 and from there to 134 is generated in the winding of the reading head a strong impulse. This fact is in clear contrast to the previous DC tape test devices which in the case discussed here no impulses would have been generated.

As previously explained, the device according to the invention and the test method according to the invention allow the determination very flat or very small flaws down to the

- 27 -009825/1760

-4

Of the order of 2.5 x 10 mm in diameter or depth in Magnetic tapes. The limitation to this value results only from the current state of magnetic recording technology and does not constitute a limitation of the main features of this In other words, in this test method

-4
only flaws of up to 2.5 x 10 mm can be detected, simply because

1. the current state of the art in the contact between tape and magnetic heads

-H gives distance on the order of 2.5 x 10 mm, and because

2. the frequency sensitivity of the heads, that is particularly limited by their structure through the air gap length and thereby is situated in a region which is to receive enough energy on moving past very small flaws on the magnetic head too narrow.

Therefore, if in the future magnetic heads are developed with a smaller air gap length and the contact between Magnetic head and tape is improved, so one will be able to use the test method according to the invention to detect errors

-4 digits to be able to determine which ones are smaller than 2.5 · 10 mm are in diameter or in depth.

- 28 009825/1760

ia

While the invention is only with respect to several embodiments has been shown and described in detail, but it will be understood by those skilled in the art that various Changes in form and details can be made without getting away from the thought and from removed from the scope of the invention. Although the invention is particularly described in connection with magnetic tapes has been, it is readily apparent to a person skilled in the art that the device according to the invention or the method according to the invention in the same way for testing any movable magnetic storage medium, such as magnetic Thin films, magnetic disks and magnetic drums, can be used.

- 29 009826/1780

Claims (1)

Claims ' Ij device for detecting errors in flexible magnetic storage media characterized by means (20, 22) for recording a magnetic Constant field in the medium (12), the field strength values of which are a function of the depth of the magnetic constant field in are in the medium and change as a gradient over the entire depth of the medium, whereas in the presence a flaw in the medium a part of the magnetization gradient is lost, further by a detector device (24, 26, 28) for determining differences in the magnetic flux between adjacent ones surface-parallel planes of the medium, these flux differences due to the differences in the magnetization gradients of these adjacent planes of the Medium. 2. Apparatus according to claim 1, characterized in that the detector device has a transmitter (24, 28), which from the magnetic flux differences between these respective adjacent levels of the medium (12) produces an electrical signal and has a frequency sensitivity curve of predetermined slope that further the detector device has a circuit (30, 32, 39) which the electrical generated by the transmitter - 30 009825/1760 Signal amplified and which one with respect to the frequency sensitivity of the transmitter behaves inversely Has frequency sensitivity, so that the overall frequency sensitivity of the transmitter and the amplifier together one result in a flat curve. 3 * device according to claim 1, characterized by a Saturation circuit (16, 18) which saturates the medium (12) with a strong magnetic constant field before the Recording device (20, 22) records a magnetic constant field in the medium, which is opposite to the latter is opposite to the magnetic field recorded in the medium by the saturation circuit. H. Apparatus according to Claim 1, characterized by a saturation circuit which saturates the medium (12) with the aid of a strong alternating magnetic field before the recording device (20, 22) records a direct magnetic field in the medium. 5 device for detecting errors in magnetic tapes, which has a write head, a read head and a tape drive, which latter the tape first on the Write head and then moved past the read head, marked - 31 - 0 0 9825/1760 BADORiGINAi. characterized by a source * (22) for exciting the write head (20) with DC voltage signals which are so strong that the write head generates a magnetic field in the tape, the field strength of which is a function of the depth of this magnetic field in the tape, the Magnetization of the tape normally changes as a gradient along the thickness of the tape (12, ΛΗ) , but is partially lost in the presence of an error P, and further characterized by a detector device (26), which decreases the magnetic flux changes from the read head (24, 28), the latter in this due to differences in the magnetization gradients in each case on top of one another following arranged surface-parallel planes of the tape are caused. 6. Apparatus according to claim 5, characterized in that fc that the detector device has an integrator (39), which integrates the signals from the reading head (24) and whose integration function behaves in reverse as the Differentiation function of the read head, so that the output signal of the integrator reflects the changes in flux in the read head is directly proportional. 7. Apparatus according to claim 5, characterized by a - 32 - 009825/1760 Erasing device (l6, 18) for direct current erasing of the tape (12, 14) before it is under the write head (20) is moved past, the magnetic field generated by this erasing device being the reverse of the magnetic field Field of the write head (20) is directed. 8. Apparatus according to claim 5 »characterized by an erasing device for erasing the tape with alternating current, before this is moved under the writing head (20). 9. A method for detecting errors in flexible magnetic storage media, characterized by the following Process steps: a) Writing in the medium with a constant magnetic field, the strength of which is a function of the depth this magnetic field is in the medium, b) storing the strength of the DC magnetic field, wherein the magnetization along as a magnetisation gradient changes in the medium, each corresponding to the thickness of the medium, except at the points> to which a flaw in the medium only storing a portion of the - 33 - 009825/1780 Gradients allowed, c) Reading out differences in the magnetic field in each case from a plane parallel to the surface of the Medium to an adjacent level, whereby these changes are caused by Pehler, which only allow a part of the magnetization gradient to be saved in the defective plane. 10. The method according to claim 9 »characterized in that that process step c) is divided into the following sub-steps: oil) differentiating the through the magnetization gradient generated in each plane of the medium magnetic field, so that changes in the gradient of generate a signal in each case from a plane to an adjacent plane, and / 7) Integrate this signal so that the integrated Signal changes directly with the magnetic level of the magnetization gradient of the medium. 11. The method according to claim 9 »characterized in that - 34 - 009825/17 60 that the medium is saturated with an equal field before something is written into the medium , so that the direction of magnetization of the medium before writing is opposite to the direction of the field applied to the medium during writing. 12. The method according to claim 9 »characterized by erasing the medium by means of an alternating voltage signal which saturates the medium before writing. - 35 009825/1780 ORIGINAL BATHROOM Blank page