Ferrite magnetic heads for sound recording and features of their application. Encyclopedia of radio electronics and electrical engineering

Encyclopedia of radio electronics and electrical engineering / Audio equipment

 Comments on the article

In the first part of the article, the designs of ferrite magnetic heads, commercially produced in the CIS, are considered: their parameters are indicated, and application features are noted. The following parts describe the manufacturing technology of the heads, provide additional methods for measuring the parameters of the heads and recommendations for adjusting tape recorders with such heads. This information will be useful to radio amateurs and specialists involved in the repair and design of magnetic sound recording equipment.

The era of cassette recorder dominance is probably coming to an end. However, taking into account economic considerations and the presence of a huge number of phonograms on compact cassettes among the population, it can be assumed that in our country it will last and cassette recorders will serve their owners for at least another 15-20 years.

Publications devoted to magnetic heads (MG) for sound recording have already appeared on the pages of "Radio" [1, 2]. And yet, information, in particular, about ferrite MGs, unfortunately, is clearly not enough. Over the past ten years, we can recall only a few materials on ferrite heads that appeared in [3,4,5]. Moreover, some materials [1,2, XNUMX] contain inaccuracies that lead to serious problems in their use.

The author tried to give more complete information about the ferrite MGs currently produced and to talk about the features of their use in cassette recorders.

Just as the general name "metal" MGs means heads made of different materials (permalloy, sendust, amorphous alloys), and the name "ferrite" (or "glass-ferrite") MGs means heads made of different materials with different manufacturing technologies, which is essential affects their parameters and performance properties. For domestic MGs, information about materials and manufacturing technology is contained in a two-digit number - the modification number - after the dot in the MG symbol. Certain technologies and materials correspond to specific areas of modification numbers; this was standardized back in the 70s and, with rare exceptions, is now in effect (Table 1). Foreign companies mark heads according to a variety of (often closed) internal company standards, so it is practically impossible to extract the necessary information from the designation of foreign MG.

The most obvious advantage of ferrite MGs - their durability - is determined by the material of the working surface. There are ferrites of polycrystalline and single-crystal structures. Polycrystalline ferrites used for the manufacture of MGs are obtained either by hot pressing technology - hot-pressed ferrites (HPC), or by isostatic pressing (IPF) or "Oxostat" technology. With isostatic pressing, the compression of the press powder occurs evenly from all sides, while with hot pressing, it is only in one direction. As a result, the porosity of the GPF grade 10000 MT-1 does not exceed 0,5%, and the porosity of grade 10000 MT-2 (IPF) does not exceed 0,1%. The M1500NMZ ferrite ([ 1]) has a porosity of up to 5% or more. The porosity of the material determines not only the wear of the MG itself. but, more importantly, the wear of the working layer of the magnetic tape (ML). The working surface of the erasing heads (for such tape recorders as "Or-bita-205") made of ordinary ferrite with a porosity of up to 20% is, in fact, a "grater *, mercilessly peeling off the working layer of the ML (remember the slides of powder on the tape drive mechanism). Only in MG type 6S24.710, BFA is used, which ensures low wear of the ML (in [1] it is inaccurately indicated that the material was obtained by hot pressing).

Single-crystal ferrites (MCF) are obtained using the technology of growing artificial rubies and sapphires using the Verneuil, Czochralski or Bridgman methods. The first two methods are more productive, but the crystals are of lower quality, so the Bridgman method is more often used [6, 7]. Growing a crystal (the so-called "boule") weighing 8 kg, together with cooling, takes about 20 days. A single crystal is an anisotropic material and requires orientation along crystallographic axes in the manufacture of MG.

The nature of wear of the working surface made of HPF or IPF and a single crystal is very different. The working surface of the head is affected by the abrasiveness of the ML, adhesion (sticking) to the ML, thermal and electrostatic effects of frictional origin (especially in high-speed rewriting devices), as well as impacts from microinclusions in the working layer of the ML (typical for domestic and worn foreign ML). If permalloy heads, as the softest ones, fail due to a change in the shape of the working surface (“saw through”), sendastoo ones - from loss of linearity of the edges, covering the gap under the action of adhesion (Fig. 1), then heads from HPF (to a greater extent) or from IPF (to a lesser extent) wear out due to erosion, chipping of polycrystal grains. The grain size in HPF is 15..,30 µm, in IPF - 10...15 µm. Erosion occurs from the impact of electrostatic forces, temperature microstresses and impacts of microinclusions on the weakest areas - grain boundaries. A "pothole" 10..30 µm wide is formed at the working gap. Single chipping quickly turns into massive ones, and the head fails. With a gap depth of 60...80 µm, the restoration of such heads is difficult. In addition, the edges of the "potholes" scratch the working layer of the tape, leading to an increase in the noise level.

In contrast to heads based on HPF and IPF, the wear of heads made of MCF is predominantly abrasive in nature, erosion (i.e., tearing out of material particles) is practically not observed. First, the softer glass that fills the gap wears out, the resulting dimple exposes the edges of the gap, then the “blockage” of the edges, leading to a gradual expansion of the effective width of the gap. It is important that the ICF-based head retains the mirror surface of magnetic tapes even when the head itself is heavily worn.

By the way, the consequences of moderate wear of heads from the ICF are easily eliminated without removing it from the tape recorder by running a polishing tape (electrocorundum with a grain of 10 microns), cut to a width of 3,81 or 6,3 mm. Such a tape is produced by many factories (in St. Petersburg - LOMO, the Magneton factory). Run time - 1...2 min. During polishing, a layer with a thickness of only 2...4 microns is removed, which completely restores the parameters of the MG (when polishing, the frequency response is monitored every 30 s until it is completely restored). Thanks to this, MG from MKF can be produced with a gap depth of only 40...60 µm. After running the polishing tape, it makes sense to drive the tape recorder for several hours on a low-value tape with increased abrasiveness (Sound Breeze or TASMA MK 60-7) to finish the surface.

It is well known that when a tape recorder is operated for 2 hours a day, Permalloy heads fail after 1,5 ... 2 years, Saint-Dust ones - after 2 ... 2,5 years. For comparison: MGs from the IFF serve 2...4 years, and, moreover, are easily restored. In high-speed rewriting devices, the service life decreases in proportion to the increase in speed and daily operating time, except for MG from the GPF or BPF, which fail faster (especially the recording heads). An unexpected feature: MKF heads with IEC II (CrOg) tape usually last longer than with IEC I (y-Fe6O10) tape. On fig. Figure 2 shows the nature of the destruction of the gap of the head sample ZD3 from the GPF 2 MT-24.712 after 10000 hours of operation, and in fig. 1 - head clearance 1000V3 from MKF after 6 hours of operation. Nearby (below) one can see an inter-channel screen made of HPF, corroded by erosion.

The electromagnetic parameters of the heads are given in Table. 2. For heads ZD24.012 (PO EVT, Penza) and 6A24.510 and 6V24.510 (Yerevan), passport data are given, for the rest - real, measured on a large number of heads. The measurement conditions are given in accordance with [8]. The shunt coefficient Ksh characterizes the losses in the magnetic head and is calculated by the formula

where E is the electromotive force (EMF) of a real head, mV; End - EMF of the head without loss, mV.

In general,

End \u2d 0p f F103 h W XNUMX.

where f is the measurement frequency, Hz;

Ф0 is the effective value of the short circuit magnetic flux per 1 m of the track width according to [9], Wb/m;

h - track width, m;

W is the number of turns.

Substituting the values, we obtain for cassette tape recorders at f=315 Hz, Ф0 = 250 nWb/m, h = 0,6 mm, W = 1000 turns

End = 2,97 10-4 V; and for reel-to-reel tape recorders with h = 0,94 mm

End = 4,6 5-10-4B.

The amplitude-frequency response of the reproduction (AFC) of the lossless head, Dpnd dB, is calculated by the formula

Dpnd = 20lg(fmeas Jf) + Nmeas

where fmeas - the nominal frequency of measurement of the frequency response, Hz (upper frequency);

f - reference frequency equal to 315 Hz;

Nmeas is the relative recording level at the nominal measurement frequency according to [9]. db.

In table. 2 does not provide data on the erasing heads (HS). This is due to the fact that the parameters of HS for reel-to-reel tape recorders are given in [1], and domestic HS for cassette recorders are of no interest, since they are made of pressed ferrite and mercilessly peel off the tape. Also, these heads do not work with IEC IV ("Metal") tape. High-quality demagnetization of such tapes is the topic of a separate article.

There is a special class of erasing heads used in cheap foreign devices - heads with a permanent magnet. A core made of high-coercivity ferrite is magnetized according to a special law, obtaining a sign-alternating falling magnetic field. The number of poles is from three to ten or more. The erasing quality is not high: increased noise and non-linear distortions. We use such heads in tape recorders "Electronics-402C", "Electronics 331C" and their modifications (manufactured in Zelenograd and Voronezh).

As for the heads for recording and playback, the Magneton plant (St. Petersburg) produced them with a magnetic circuit both from the GPF or IPF, assigning them the index "P", and from the IFF with the index "M". Since the mid-80s, according to the test results, heads have been produced only from the IFF. PO EVT (Penza) produced heads from GPF 10000 MT-1 (ferrite produced by the Magneton plant). The Yerevan plant produces heads from HPF of its own manufacture. Ferrite heads coming to our market from abroad, almost all, even those considered high-end (Hitachi, Sony, JVC), are made of HPF or, at best, of IPF.

(click to enlarge)

Ferrite heads (Table 2) are manufactured according to two design schemes (Fig. 4,5): with "P"-shaped and with linear contactors. The first design has a larger volume of remagnetizable material, which leads to increased nonlinearity in the reproduction of signals with a low recording level ("ferrite sound"), but it allows you to place a winding with a large number of turns. It is used in heads for reel-to-reel tape recorders.

The second design (Fig. 5) provides good linearity during reproduction, but the number of turns is limited by the size of the window for the winding and the external dimensions of the MG.

At one time it was believed that with such a constructive scheme it was impossible to obtain an acceptable EMF value of MG. However, the detailed calculation of the magnetic circuit carried out by the author according to the refined method revealed the area of ​​design parameters in which MG according to such a scheme are competitive. This made it possible for the first time to create ferrite MGs for cassette tape recorders, which are distinguished by the absence of a "ferrite sound" during playback.

The manufacturing technology of a two-channel cassette head in general terms is as follows: - so-called limiters of normalized thickness are sprayed onto the semi-blocks (Fig. 6), depending on the required width of the working gap.

Next, the semi-blocks are soldered with glass. In the gap formed by the limiters, the glass flows capillary. Then blocks of 1,55 mm in size (the width of two channels) are cut out of the soldered workpiece, a groove for the interchannel screen is cut on each block (Fig. 7), the interchannel screen is glued in and the jumper is ground off (Fig. 8, 9).

Having finished gluing the elements that form the working surface, the workpiece is ground along the radius (Fig. 10), while maintaining a gap depth of 40 ... 60 microns. After sorting, the gapped pole pieces are ready for assembly.

The advantage of such a labor-intensive technology is that the parallelism and alignment of the gaps of the stereo head unit is provided automatically.

A simpler method is "element-by-element" assembly: the channel heads, the screen and other elements are made separately, and then either glued together or soldered with glass in a "stack". But such simplicity, as they say, "goes sideways": it is almost impossible to maintain alignment and parallelism of the gaps. Using this technology, heads were produced in the Penza PO EVT, in particular ZD24.012.

The main areas of application of heads from the MKF:

• devices for high-speed dubbing, operating at speeds above the nominal. Bias currents, depending on the speed, have a frequency of 200 kHz to 2 MHz;
• high-quality household tape recorders, designed for a long service life and consistently high quality of work;
• medium-class tape recorders (complexity groups 1-2), which, due to the use of such heads, not only benefit in durability, but also improve sound quality [3].

Of course, extremes are also possible: the installation of the ZD24.751 head in a tape recorder of a very low class (in the Melodiya-106 radio tape recorder) instead of the MG type BRG ZD24.M (Hungary) completely transformed the sound (as they say, "do not know!").

It should also be borne in mind that the heads given in Table. 2, do not write on ML M3KIV ("Metal").

When calculating the costs, it can be assumed that one head from the MKF is equivalent in terms of durability to three from the sendust (the term is limited by the complete wear of the tape recorder). If you buy at the factory, then the cost of one ZD24.750 head ranges from 20 to 24 rubles, depending on where you buy it - in the sales department of the Magneton factory or in the factory store. In the market, dealers are added to this.

When tuning tape recorders with a ferrite head, features associated with the properties of the material used appear: for example, the bias current is 2 ... 2,5 times less than that of metal heads, and a high quality factor leads to a sharp influence of resonance phenomena on the tuning process. The parameters of the ferrites used for the manufacture of MG are given in Table. 3. For comparison, the parameters of some magnetic alloys are given (for other materials, see also [10, 11]).

Before installing the MG, it is desirable to determine its inductance Lmg, its own capacitance Cmg and quality factor Qmg. Previously, the manufacturer in the passport for MG gave individual values ​​​​of Lmg, EMF, as well as recording and bias currents. Now in the passport only unreasonably extended limits of their values ​​are given, which, given the significant cost of the heads, only causes bewilderment. If the values ​​​​of currents can be taken on average from the table. 2, then the inductance needs to be determined more precisely. We can recommend the following method for measuring Lmg, Smg. The measurement scheme is shown in fig. eleven.

The inductance of the magnetic head Lmg forms an oscillatory circuit with a total capacitance Cmg + Spar + Cdop, where Cmg is the head's own capacitance Spar - mounting capacity; Sdop - additional capacity. For measurement, it is desirable to have 4 - 5 Cdop values ​​​​from 5 to 80 pF, known with an accuracy of no worse than 5%, this directly affects the measurement accuracy. The tolerance on R1 and R2, input capacitance and input resistance of the millivoltmeter are not critical. Connection to the MG is best done using sockets from a suitable small-sized connector (for example, from RG35-ZM, etc.). The wire connecting to the MG and the conclusions R1, Sdop must have a minimum length to reduce Spar.

The required error in setting the frequency of the generator is 1 ... 2%, the output voltage in the range of 20 ... 200 kHz is at least 3 V. The required sensitivity of the millivoltmeter is 3 mV.

By connecting in turn capacitors Cdop of different ratings, starting from small values, the resonant frequency of the circuit is counted according to the minimum millivoltmeter reading when the generator frequency changes. Transforming the well-known formula, we obtain

CΣ=(2,53/Lmg)x104/f2res. where СΣ - total capacitance, pF;

Lmg - inductance, H (for ferrite heads in this frequency range, the value is almost constant); fpez - resonant frequency, kHz.

It follows from here that there is a linear relationship between CΣ and 1/f2pez, which can be used to determine Cm. This is done as follows [12]:

• for each used value C, the value 104/f2pez is calculated (see the example in Table 4);
• a graph is constructed (Fig. 12), where the values ​​of Cdop are plotted along the abscissa axis, and 104 / f2pez are plotted along the ordinate axis.

A straight line is drawn along the obtained points until it intersects with the abscissa axis. The intersection point and gives the value (Smg + Spar). When the length of the connections between R1, Сdop and MG is less than 2 cm, the capacitance Сpar can be taken equal to 2pF. In the above example (Smg + Cpar) = 13 pF. From here we find

Lmg \u2,53d 04 / (Smg + Spar) x 2 / f2,53res. = 13/0,485x0,0944 = XNUMX H;

Smg \u13d 2-11 \uXNUMXd XNUMXpF.

The measured values ​​of Cmg for different specimens of heads of the type ZD24.750 - ZD24.752 lie within 7 ... 20 pF. This capacitance differs for different channels and varies depending on the connection of the common wire to one or the other output of the MG

For metal heads, this method of determining the intrinsic capacitance and inductance is unsuitable because of their low quality factor and, as a result, the strong frequency dependence of the inductance.

Accurate measurement of Qmg in amateur conditions is difficult. In the general case, the quality factor of the circuit Q is determined from the resonance curve (see [12]):

Q=fres/(fmax - fmin)

where f is the resonance frequency, kHz; fmin and fmax - frequencies at which the voltage on the circuit drops to the level of 0,707Umax, kHz.

The measurement accuracy depends on the degree of circuit shunting by the input impedances of the measuring instruments, the reading accuracy 0,707Umax and the frequencies fres fmin and fmax. For measurements with an error of up to 5% at Q = 20...40, it is necessary that the shunting resistance be at least 10 MΩ, and the values ​​fpez, fmin, fmax 0,707Umax should be measured with an error of no more than 0,2%. According to the diagram in Fig. 11, the shunt resistance is approximately equal to R1, which gives a decrease in Q by 50 ... 70%.

The use of field-effect transistors with a large input impedance makes it necessary to take measures to protect against static electricity (the static potential of the voltage on the operator's hands relative to the ground can reach 20 kV!).

In practical work, you can focus on the measurement data given in Table. 5.

The measurements were carried out both in the range of upper sound frequencies and in the frequency range of the bias current. The measurement error is about 5%. Low-loss capacitors were used in the measurements, and Cmg and Cpar were taken with a large tolerance of 15 and 6 pF, respectively. This assumption and the error in the calculations gave a spread in the values ​​of the inductance Lmg, which were calculated using the formula given earlier. The resonant resistance of the parallel oscillatory circuit Rres and the active loss resistance Rs were calculated by the formulas [12]:

where Rres - resonant resistance, MΩ; Lmg - head inductance, H; CΣ - total capacitance, pF; Rs - active loss resistance, Ohm. For those who want to understand in more detail, we recommend [13].

An analysis of the data obtained shows the following: the quality factor decreases with a wider MG gap and with an increase in СΣ, remaining very high (tens of units) in the region of upper sound frequencies. At bias current frequencies, the quality factor of ferrite heads is also quite large (on a metal MG it is less than unity, it cannot be measured). At the same time, Rpez is such that if the frequency fpez coincides with the frequency of the bias current in the recording mode, it becomes impossible to set the nominal bias currents in the usual scheme of their supply (it turns out to be "brute force"). Rs of ferrite MGs is much less than that of metal MGs, for example, type ZD24.211 ("Mayak"), especially at medium and higher frequencies (200 Ohm versus 3 ... 5 kOhm!). This explains the significantly lower level of thermal noise in ferrite heads.

Before turning to specific issues of optimizing the parameters and adjusting tape recorders with ferrite heads, it is necessary to recall some terms and provisions adopted in the technique of magnetic sound recording. The reference frequency, adopted as 315 Hz (earlier, before 01.07.88/400/8, the nominal frequency - 14 Hz), allows you to compare the measurement results [XNUMX]. At this frequency, the EMF of the heads is measured during playback, the frequency response is also measured in relation to this frequency. For this, a signal-gram is used, recorded in accordance with the recommendations of the International Electrotechnical Commission (IEC). The frequency response of the short-circuit magnetic flux of this signalogram N, dB, is calculated by the formula [XNUMX]:

where f - frequency, Hz;

τ1, τ2 - time constants, s. The relative short circuit magnetic flux recording level is calculated as the difference between N(f) and N(315 Hz), where 315 Hz is the reference frequency. The numerical values ​​of the relative recording level are given in [9]. These values ​​are used to calculate the lossless 0Rid of the head. In table. Figure 6 shows the calculated values ​​of the relative recording level (reference frequency 315 Hz, τ2 = 3180 µs, τ1 = 70 and 120 µs).

The frequency correction of the playback channel, i.e., the path of the playback head-amplifier (HC), must ensure that the requirements for uneven frequency response in a given frequency range are met. Thus, the standardization of the dependence of the frequency response N(f), proposed by Heegard in the fifties, leads to the standardization of the frequency response of the playback channel. The choice of pre-distortion distribution between the recording and playback channels is made, as stated in [15], "based on the frequency response of the residual magnetic flux of the recorded phonogram, which can be obtained with existing tapes and a reasonable amount of pre-distortion in the recording amplifier." On the one hand, this allows you to exchange records, but on the other hand, it hinders the development and use of new, "non-standard" magnetic tapes. We will not consider the reasons for choosing specific values ​​of τ1 and τ2 here.

In table. 6 shows the values ​​of Drid frequency response of the lossless head, and in fig. 13 shows its view together with the frequency response of heads of types ZD24.752 (τ1 = 120 μs), ZD24.751 and ZD24.750 (τ1 = 70 μs).

The high cleanliness of the working surface of the heads makes it possible to obtain low contact losses. By the way, due to the "slipperiness" of the MG surface, they practically do not get dirty and do not require frequent cleaning. The high magnetic properties of single-crystal ferrite provide negligible losses for Foucault currents and remagnetization of the material. Nevertheless, the course of real wave characteristics is distinguished by some "flattening" of the top and a more gentle decay in the high-frequency region. This could be explained by the wedge shape of the gap, as shown in [16], but measurements of the gap width did not reveal this (within the measurement accuracy). The most likely explanation for this is a change in the magnetic permeability of the material in the gap zone due to glass diffusion into the core (which can be represented by the parallel operation of several gaps of different widths). The frequency response in the low-frequency region lies approximately 1 dB above Drid and in Fig. 13 is not detailed.

The block diagram of the playback channel is shown in fig. fourteen.

The playback amplifier has a frequency response that is inverse to the frequency response of an ideal head Drid (see Fig. 13), and the frequency response correction at the upper audio frequencies is usually carried out due to the resonance of the series circuit formed by the inductance Lmg and the total capacitance, consisting of Smg, mounting capacitance Spar. input capacitance of the amplifier Svh and additional capacitance Cdop. The voltage on the total capacitance, i.e. at the SW input, for such a circuit at the resonance frequency increases by a factor of Q, where Q is the quality factor of the circuit. The rise in the frequency response at the resonance frequency from the signal level without taking into account the resonance is 20lgQ, dB. Due to the shunting action of Rin and Rsh, the quality factor decreases. The influence of Rin without taking into account losses in the total capacitance can be estimated with sufficient accuracy by the formula

Qsh=Q Rin/(Rres+Rin)

where Q is the initial quality factor of the MG (see Table 5);

Rin - input resistance SW, kOhm;

Rres - resonant resistance (see Table 5), kOhm;

Qsh - quality factor of the shunted circuit.

So, at Q = 15, Rres = 150 kOhm, Rin = = 100 kOhm, we get Qsh = 6, i.e., the frequency response rises by 15,6 dB. At Rin \u1000d 13 kOhm, Qsh \u22,3d 6 (frequency response rise by 10 dB). Due to the high quality of the MG surface from the ICF, the actual required rise in the frequency response is only 2 to 3 dB, which corresponds to Qsh = XNUMX...XNUMX. You can calculate the approximate value of Rsh necessary to obtain the desired quality factor using the formula

1/Rsh=(Q-Qsh)/(QshRres)-1/Rin,

where Rsh - shunt resistance, kOhm;

Qsh - required quality factor of the shunted circuit;

Rres - resistance according to the table. 5, kOhm;

Rin - input impedance of the amplifier, kOhm;

So, for Qsh = 3 (amplification of the frequency response is not more than 10 dB) at Q = 15, Rres = 150 kOhm, Rin = 100 kOhm, Rsh = 60 kOhm; for Rin = = 1000 kOhm - Rsh = 39 kOhm.

Two tasks can be distinguished, in the solution of which it is advisable to use ferrite MGs:

• replacement of worn-out MG on the existing apparatus;
• development of a playback channel in order to maximize the capabilities of ferrite MGs, namely, to achieve a low noise level (due to low Rs, see Table 5), good linearity in playback, a wide bandwidth of reproducible frequencies and good impulse responses.

The last task is reduced to the development of a "compatible" reproduction amplifier having its own noise no worse than -65...-70 dB and low intermodulation distortion. The SW must work stably with high-Q MGs (most of the existing SWs are excited in the absence of Rsh). In addition, there is a requirement to be able to operate at twice the speed for rewriting. This requires changing the value of the total capacitance from 1300 ... 630 pF at Lmg = 100 mH (upper frequency 14 ... 20 kHz) to 320 ... 160 pF (upper frequency at double speed 28 ... 40 kHz). If Cm = 11 pF, Spar = 20...40 pF, then with a large input capacitance of the SW, obtaining good quality at double speed becomes impossible. Changing the total capacitance is achieved in two ways:

• switching capacitor C add;
• electronic control Svh. A detailed description of the options for electronic control of the SV and an analysis of the effect of dynamic capacitance are given in [17], but the SW option proposed there has a noise level of -58 dB, which is clearly insufficient. Much better parameters have an amplifier based on field-effect transistors with a pn junction, described in [5]. Perhaps the parallel connection of N of the same type of low-noise field-effect transistors, in which the EMF of self-noise decreases by √N times, allows you to develop a "compatible" HC for a cassette recorder with a noise level below -70 dB (the heads from the ICF allow this). But the question of using double speed remains open - the dynamic input capacitance is large.

Let us determine the upper frequency fepx for different types of produced ferrite MGs, based on the requirement for the necessary unevenness of the frequency response of the playback channel. Typical frequency response of playback channels for three types of MG without correction at high frequencies is shown in Fig. 15.

These frequency responses were obtained from the data for the MG ZD24.750 - ZD24.752 (see Fig. 13). By superimposing the resonant curves of the input circuit on these curves at different frequency response decays, we can make sure that an acceptable unevenness of the total frequency response is obtained if we take the frequency at which the frequency response decay without correction does not exceed -10 dB for fvepx. For 3D24.752fvepx = 14...16 kHz, for ZD24.751 fbepx = 16...18 kHz, and for ZD24.750 fbepx = 18...20 kHz. Figure 15 shows the resonance curves from the level of -10 dB at a frequency of 20 kHz with Qsh equal to 10, 3 and 2, as well as the view of the total frequency response of the playback channel. As can be seen, the optimal correction of the high frequencies for the MG ZD24.750 occurs at a value of Qsh between 2 and XNUMX.

Therefore, when installing a ferrite MG in a tape recorder, if the SW has a high-frequency correction adjustment (except for the formation of standard time constants τ1 and τ2), and/or a positive feedback circuit to increase the quality factor of the input circuit [17], it is necessary to bring their adjustments to a minimum. After that, in parallel with the MG, it is necessary to connect a small-sized tuning resistor with a nominal value of 80 ... 100 kOhm as Rsh, setting its maximum value and turning off the shunt resistor available in the SW.

When installing the MG, in addition to the usually checked inclination (azimuth), centering and "nod" of the head, it is necessary to check the depth of entry of the MG into the cassette. Due to excessive pressing of the tape to the working surface, in addition to increased wear of the MG, frictional "whistles" also occur, especially if the working surface is contaminated with traces of glue from the adhesive tape used to glue the leader cords.

It is more convenient to check using a cassette, in the top cover of which there is a cutout in the place where the head enters the cassette. The zone of contact of the working surface with the magnetic tape should lie within 3,5...4,5 mm symmetrically with respect to the gap.

If, when the device is turned on, the SW is excited, it is necessary to reduce the value of Rsh until the excitation disappears.

The upper frequency is taken either equal to fBepx of a particular type of MG, or is lowered if the tape recorder does not provide the necessary stability of the azimuth of the magnetic tape movement or the SW has a limit on the upper frequency. The input circuit is tuned to this frequency by selecting Cdop. Due to the high wear resistance of heads made of MKF (wear of 3 microns per 1000 h), adjustment during operation is not required. The resonance frequency is determined by the maximum output signal of the SW when the magnetic field of the signal is applied to the gap of the MG using a coil on the shaper according to [9]. The frame of such a frame has dimensions of 8x75x3 mm, the number of turns is 20 ± 5 with a PEV 0,2 wire. The signal from the generator is fed through a 100 ohm limiting resistor. This method does not require unwanted soldering on the tape recorder boards. The magnetic field can also be applied to the gap using a flexible conductor glued to the working surface of the MG in the area of ​​the gap (it is convenient to glue it with alcohol-soluble glue such as BF-6).

It is most convenient to tune to fvepx and the frequency response of the playback channel using a signalogram of measuring tapes of the ZLIT1.4.4-120 type [9], consisting of packs of frequency messages. The burst repetition rate is 18 Hz, the duration of one frequency burst is at least 3 ms, the pause between bursts is 1 ms, the maximum frequency is 14 kHz. The resonance frequency is determined using an oscilloscope by the maximum amplitude of the corresponding frequency message. If fvepx is more than 14 kHz, or there is no such measuring tape, then it can be formed using a personal computer. A number of necessary messages are recorded in memory, which are recorded on a cassette using a well-tuned tape recorder with a sufficient frequency range. The duration of the parcels and the repetition frequency are the same as for ZLIT.CH.4-120. The number of frequency bursts is up to 10. With a sampling frequency of 44 kHz, a maximum frequency of up to 20 kHz can be obtained, with a sampling frequency of 54 kHz - up to 24 ... 25 kHz. A tape of the ZLIM.UNCHK.4 type manufactured by Magnolia JSC (about $ 8 ... 10) is also suitable, on which there are all the necessary signals (for checking the frequency response, detonation, nominal level, balance, etc.).

After setting the input circuit to ftop, set the nominal level at the line output and the corresponding indicator readings in playback mode. This requires a measuring tape with a signalogram of the reference frequency of the nominal level. The linearity of the frequency response is adjusted with a tuned resistor Rsh, which is then replaced with a constant one. When using a homemade measuring tape to adjust the frequency response, you must ensure that the recording level is at -20 dB. To do this, when recording on a reference tape recorder, the input voltage is reduced by 10 times relative to the nominal. With sufficient experience, it is permissible to adjust the frequency response without a measuring tape according to the scheme in Fig. 16, setting the treble boost equal to the typical frequency response rolloff (see Fig. 15). It is quite satisfactorily possible to adjust the frequency response by setting Rsh with a resistance calculated from the data in Table. 5 for Qsh = 2 with known RBX. Tuning "by ear" using musical backing tracks, as a rule, gives negative results due to the masking of the highest frequency signals by the mid-frequency ones and the difference in the quality and spectral balance of the recordings. At the same time, RBX can be easily measured, for example, by the compensation method.

Estimation of the non-linearity of the playback channel is usually required in the development of CFs or when comparing MGs from different materials. If such a need arises, it is recommended to evaluate the nonlinearity using the Twin-Ton-Test difference tone method [18]. In this case, two test signals of the same amplitude with a frequency ratio of 1:1,06 are applied to the input. If the amplitude of their intermodulation products is 4,7% of the amplitude of the test signals, then this corresponds to a coefficient K3 = 3% for one of the test signals.

To obtain a good sound, as it has long been substantiated abroad and finally recognized in our country [19], it is necessary to achieve an intermodulation distortion coefficient Ki less than 0,003%. In practice, a qualitative assessment of Ki is carried out by applying a magnetic field of test signals to the MG gap, as described earlier. In this case, it is convenient to choose the signal frequency from fvepx to fvepx / 2 with a difference between them of 0,5 ... 1 kHz. The amplitude of the signals is increased from zero to the nominal level at the linear output of the SW. If, when listening acoustically to such a combination, better on high-quality headphones, a difference tone begins to be heard, this means that Ki becomes more than 0,003% [18; 19]. For a more accurate assessment of Ki, a spectrum analyzer is needed.

As already noted, due to the minimum volume of remagnetizable material, normalization of the coercive force Hc, and good high-frequency properties of the material, MCF cassette heads have a fairly low non-linearity during playback: lower frequency for sendust heads and comparable with the best permalloy heads. However, when recording on an IEC type IV ML, phenomena associated with saturation of the edges of the working gap are observed. The results of studying this effect are given in [20], where it is shown that an increase in the field in the gap HG (in oersteds, Oe) above the value corresponding to half the value of the saturation induction Bsat (in gauss, G) leads to saturation of the edges of the working gap. As a result, the recording area expands, losses increase and non-linear distortions increase. There is also an empirical formula for determining the required field in the gap HG (Oe) with a gap width g (μm) required for recording with a limiting signal level with a wavelength λ (μm) on a carrier with a coercive force Hc (E):

HG \u1,7d (0,33 / g0,8 + 0,78VgXNUMX) x Hc.

It was also shown in [20] that this value is close to the optimal bias field strength for recording with high frequency bias.

The coercive force Hc of various types of ML lies within the limits [18]:

• 24...28 kA/m (300...350 Oe) for type I ML (Fe2O3);
• 35...40 kA/m (440...500 Oe) for type II ML (CrO2 and its substitutes);
• 80...120 kA/m (1000...1500 Oe) for ML type IV (Metal).

Hence the required field in the gap HG (E):

• for type I ML at fup = 14 kHz (λ = 4,76 cm/s (104/14000 Hz = 3,4 µm) d = 1,8 µm, HG = 940...1100 Oe;
• for type II ML at ftop, = 16 kHz (λ = 3,0 µm), g = 1,5 µm, HG = 1400... 1620 Oe;
• for type IV ML at ftop = 20 kHz (λ = 2,38 µm), g = 1,0 µm, HG = 3600...5400 Oe.

To work with type I ML, a material with Vsat > 2900 Gs (0,29 T) is required:

• with type II ML - with Vnas = 3250 Gs (0,33 T);
• with ML type IV - with Vnas = 7200 ... 10800 Gs (0,72 ... 1,08 T).

Comparing the obtained values ​​for Vnas with the data of Table. 3, we can conclude that not only ferrite MGs, but also metal MGs do not guarantee recording without excessive distortions on all available Type IV MLs. the requirement for Vnas material is up to Vnas > 160 T.

There is a design of a ferrite MG, in which, to protect the edges of the gap from saturation, a layer of a metal alloy with Vmax > 1,4 T and a thickness of 2 ... 10 μm is applied to the inner walls of the gap. These are the so-called "MIG" heads ("Metal-ln-Gap" - metal in the gap) [21; 22]. Such heads are quite widely used in video technology, but for the purposes of sound recording, our (and foreign) industry practically does not produce them, probably due to the limited distribution of type IV tapes (increased cost, and most importantly, the lack of devices that realize their advantages).

For a commercially available MG type ZD24.750 with g = 1 μm, when recording a signal with fbepx = 20 kHz on a type II ML, a core material is required in the gap zone with Vmax > 0,36 T, which is done with a sufficient margin (according to Table 3 at MCF Vmax = 0,43 ... 0,5 T). Therefore, the statement that "ferrite heads ... give the highest level of non-linear distortion (in the recording mode)" [2], as applied to heads from the ICF, seems incorrect. Direct measurements show the opposite.

And finally, about setting up the recording amplifier when installing ferrite MGs. When setting up the recording channel, first of all, it is necessary to make sure that the bias frequency fsubm is less than the resonant frequency fpez of the circuit formed by the MG inductance Lmg and the total capacitance CΣ, consisting of the MG's own capacitance, the output capacitances of the generator and amplifier (filter tube) and the mounting capacitance. It is desirable that fsubm < 0.8 fpez or, according to Table. 5, fsubm < 84...96 kHz. If the capacitance Cmg has been measured, as discussed earlier, then a more accurate limit on the value of fsubm can be obtained. With fsubm = fpez, the circuit LmgCΣ works as a filter resonator, while any temperature change in the values ​​of Lmg and CΣ leads to a change in the bias current, and its value is greatly overestimated. If fsubm>fpez, then the bias current is shunted by CΣ and, if it is regulated not by resistors, but by trimmer capacitors, the load on the generator may increase sharply.

Due to low bias losses for ferrite MGs, the optimal current turns out to be 2–3 times less than that of metal heads (ceteris paribus). The write current is less, but not significantly. This leads to the fact that there are not enough regular adjustments to set (reduce) the bias current, you either have to introduce an additional resistance of 50 ... 200 kΩ into the current circuit break, or, if the erasure level allows, reduce the generator supply voltage (which is worse). If the bias current is supplied through the separating capacitance, then it should not be reduced (it is better to put a series resistor) so as not to fall into the series resonance of this capacitance and the inductance of the head.

Particular attention should be paid to this when installing the recording MG ZA24.751 and ZA44.171 on high-speed dubbing devices. If the frequency fpodm is more than 200 kHz for ZA24.751 and above 500 kHz for ZA44.171, adjustment of the bias current may not be possible due to resonance phenomena. When setting the bias current for the MG type ZA44.171, due to the penetration of the bias from the adjacent channel, sometimes there are not enough adjustments that reduce the bias current (at a frequency of 500 kHz, the penetration level of this MG is -30 dB). Penetration can be combated by shunting the channel where this phenomenon affects with a 10 kΩ resistor.

Before setting the optimal bias current, it is advisable to choose the main type of ML with which it is supposed to work.

The choice is usually made on the basis of "price-quality" ratio. As a rule, each user has a proven, "familiar" type of ML, but when installing a new durable MG, other types can be used, guided by the data [23, 24, 25]. From experience, good results, especially in relation to frequency characteristics, distortion and "transparency" of sound, are shown by tapes produced by the not very well-known Korean company Sunkuong Magnetics Corp. (trademark SKC).

As already noted, earlier in the individual passport for the MG, the values ​​\u723b\u4592bof the recording and bias currents obtained for typical MLs - R23DG (IEC I) and S24A (IEC II) were given. Based on these data, by recalculation [6,3, 1], it was possible to determine the currents for the selected type of ML. Now this data is not available. Setting the optimal current Ipodm begins with determining the regulation zone and, if necessary, setting additional resistance. To do this, by decreasing Isubm, find the point at which the signal with a frequency of 3 kHz is recorded at the maximum level. Then, by increasing this current, the level is reduced by 315...1 dB. The optimal current is set either by the minimum noise of the selected type of ML, or by the minimum non-linear distortion when recording a tone with a frequency of 120 Hz. These values ​​are usually close. The final setting depends on the capabilities of the tape recorder. If the SW (at τ54 = 57 μs) has noises worse than -XNUMX...-XNUMX dB (alas, there are many such SWs), then tuning to a minimum of ML noise is difficult.

Adjustment to the minimum distortion can be done without a selective voltmeter, using the method described in [18]. Harmonic distortion is determined by the deviation of the transfer characteristic when the reference frequency signal is recorded from a straight line (on a logarithmic scale in dB). A deviation of 0,5 dB corresponds to 3% harmonic distortion (Fig. 17). This method is described in [18] for reel-to-reel tape recorders; for cassette tape recorders, it is required to check the accuracy of the results obtained. Generally speaking, an experienced tuner notices distortion of 3% or more in the distortion of the sinusoid shape.

After setting the bias currents, it is necessary to check the linearity of the frequency response over the entire frequency range. It may be necessary to reduce the correction of high frequencies in the US. Setting the "zero" of the recording indicators to the nominal level is carried out, as usual, after calibrating the SW on a measuring tape or based on the overload capacity of the ML (and US) by a compromise between noise and distortion.

This article is devoted only to serial heads, so the influence of such design parameters as gap width, back gap, etc., on the recording quality was not considered here.

In conclusion, one word of warning: due to the good high frequency and dielectric properties of the material, ferrite heads are susceptible to high frequency interference from radios, motor commutator sparking, and pulse controlled motors. This requires careful decoupling of their power circuits, including the common wire. Sometimes, to reduce interference, it is required to rotate the collector motors around the axis (which is usually provided for in the design of tape recorders), and when this does not help, it is necessary to install a copper high-frequency screen under the landing pad of the MG. If the design allows, it is useful to shield the cassette receiver.

Literature

1. Polov Yu. Magnetic heads of reel-to-reel tape recorders. - Radio, I989, No. 12, p. 34
2. Dry N. Magnetic heads for cassette recorders, - Radio, 1995, No. 5, p. 15 - 17.
3. Meleshkin N. Replacing the magnetic head. - Radio, 1988. No. 10, p. 36.
4. Kolotilo D. Recovery of magnetic heads. - Radio. 1988, no. 11, p. 38.
5. Fedichkin S. Field-effect transistor in the input stage of a low-noise ultrasonic frequency converter. - Radio, 1988. No. 10. p. thirty.
6. Smith G. Precious stones. - M.: Mir, 1984. p. 186-195.
7. Lodiz R., Parkor R. Growth of single crystals - M.: Mir. 1974.
8. Heads are magnetic for sound recording. General specifications. GOST 19775-81. - M.: Goskomstandart.
9. Tapes magnetic measuring laboratory and technological for household and automobile tape recorders. General specifications. OST4.306.002-86. - M.: VNII.
10. Tereshchuk R., Tereshchuk K., Sedov S. Semiconductor receiving-amplifying devices. Handbook of the radio amateur. - Kyiv: Naukova Dumka, 1982, p. 28 - 30,33 - 37
11. Kryukova V., Lukyanova N., Pavlov E. Status and development prospects of magnetic heads from metal alloys. Surveys on electronic technology. Series 6, "Materials". Issue. 4 (961). - M.: Central Research Institute "Electronics", 1983.
12. Kroneger O. Collection of formulas for a radio amateur. - M.: Energy, 1964, p. 44 - 53.
13. Aseev B. Fundamentals of radio engineering. - M.: Svyazizdat, 1947, p. 71 - 74.
14. Household tape recorders. General specifications. GOST 24863-87. - M. Goskomizdat, p. 6.
15. Korolkov V., Lishin L. Electric circuits of tape recorders. - M.: Energy, 1967, p. 42, 43.
16. Parfentiev A., Pusset L. Physical basis of magnetic sound recording. - M.: State publishing house of technical and theoretical literature, 1957, p. 177-179.
17. ---
18. Vasilevsky Yu. Magnetic recording media. - M. Art, 1989, p. 200-215,231.
19. Kostin V. Psychoacoustic criteria for sound quality and choice of UMZCH parameters. - Radio, 1987, No. 12, p. 40-43.
20. Jeffers F. High Density Magnetic Heads. - TIIER, 1986, v. 74, No. 11, p. 78-97.
21. Jeffers FJ et al. A "MIG" - type head for magnetic recording. - IEEE Transactions on Magnetic, 1982. v MAG-18, no. 6, p. 1146-1148.
22. Analysis of Metal-in-Gap Heads. - IEEE Transactions on Magnetic, 1984, v. MAC-20, no. 5, p. 872, 873.
23. Karnaukhov E. Audio cassettes. - Radio, 1995. No. 8, p. 51,52.
24. Sukhov N. 66 compact cassettes in the CIS market. - Radio, 1993, No. 10, p. 10-15.
25. Cassettes for magnetic sound recording. Radio, 1991, No. 4, p. 82, 83.

Author: V.Sachkovsky, St. Petersburg

See other articles Section Audio equipment.

Read and write useful comments on this article.

<< Back

Latest news of science and technology, new electronics:

Machine for thinning flowers in gardens 02.05.2024

In modern agriculture, technological progress is developing aimed at increasing the efficiency of plant care processes. The innovative Florix flower thinning machine was presented in Italy, designed to optimize the harvesting stage. This tool is equipped with mobile arms, allowing it to be easily adapted to the needs of the garden. The operator can adjust the speed of the thin wires by controlling them from the tractor cab using a joystick. This approach significantly increases the efficiency of the flower thinning process, providing the possibility of individual adjustment to the specific conditions of the garden, as well as the variety and type of fruit grown in it. After testing the Florix machine for two years on various types of fruit, the results were very encouraging. Farmers such as Filiberto Montanari, who has used a Florix machine for several years, have reported a significant reduction in the time and labor required to thin flowers. ... >>

Advanced Infrared Microscope 02.05.2024

Microscopes play an important role in scientific research, allowing scientists to delve into structures and processes invisible to the eye. However, various microscopy methods have their limitations, and among them was the limitation of resolution when using the infrared range. But the latest achievements of Japanese researchers from the University of Tokyo open up new prospects for studying the microworld. Scientists from the University of Tokyo have unveiled a new microscope that will revolutionize the capabilities of infrared microscopy. This advanced instrument allows you to see the internal structures of living bacteria with amazing clarity on the nanometer scale. Typically, mid-infrared microscopes are limited by low resolution, but the latest development from Japanese researchers overcomes these limitations. According to scientists, the developed microscope allows creating images with a resolution of up to 120 nanometers, which is 30 times higher than the resolution of traditional microscopes. ... >>

Air trap for insects 01.05.2024

Agriculture is one of the key sectors of the economy, and pest control is an integral part of this process. A team of scientists from the Indian Council of Agricultural Research-Central Potato Research Institute (ICAR-CPRI), Shimla, has come up with an innovative solution to this problem - a wind-powered insect air trap. This device addresses the shortcomings of traditional pest control methods by providing real-time insect population data. The trap is powered entirely by wind energy, making it an environmentally friendly solution that requires no power. Its unique design allows monitoring of both harmful and beneficial insects, providing a complete overview of the population in any agricultural area. “By assessing target pests at the right time, we can take necessary measures to control both pests and diseases,” says Kapil ... >>

Random news from the Archive Ratoc RP-MP1 pico projector 17.12.2015 Once huge, consuming a lot of electricity and glowing with heat, projectors - first film, and then the first electronic ones, shrink from year to year. So, the new RP-MP1 pico projector, released by Ratoc System, has dimensions of only 100 x 100 x 22 millimeters and weighs only 240 grams. At the same time, it supports broadcasting technology from Miracast wireless devices in 802.11b / g / n mode, for which it is equipped with the appropriate hardware stuffing running Android 4.2. Of course, the dimensions leave their mark on the technical characteristics of the novelty: although the projector uses a backlight based on a powerful modern diode, the maximum image brightness does not exceed 80 lumens. That is, the scope of the RP-MP1 is a darkened room or night gatherings in nature in the company. The projection system is micromirror (DLP), the chip itself has a native resolution of 854 x 480, but the projector recognizes an input signal up to 1080p, scaling it accordingly. The distance to the screen can be from 20 centimeters to 2,8 meters, the diagonal of the created "screen" in this case varies between 7-100 inches. In addition to Miracast, the projector has HDMI (MHL compatible) and USB inputs. The device is also equipped with a built-in 4000 mAh lithium-ion battery, which guarantees an hour and a half of continuous operation. The Ratoc RP-MP1 is available in two colors: black and with a golden cap. The cost of new items is approximately $290.

Other interesting news:

▪ High voltage DC/DC converter with integrated frequency switch

▪ JVC DLA-Z4 1K projector

▪ Panel computer from EVOC

▪ Electric coupe-crossover Skoda Enyaq Coupe iV

▪ FMS6407 - video driver filter

News feed of science and technology, new electronics

Interesting materials of the Free Technical Library:

▪ section of the website Electrotechnical materials. Article selection

▪ article Germany above all. Popular expression

▪ article What is the average color of all light sources in the universe? Detailed answer

▪ article Indoor fountain. Children's Science Lab

▪ article Installation of laser effects. Encyclopedia of radio electronics and electrical engineering

▪ article Transformerless power supplies. Encyclopedia of radio electronics and electrical engineering

Leave your comment on this article:

All languages ​​of this page

Arabic	Hebrew	Polish
Bengali	Hindi	Portuguese
Chinese	Italian	Spanish
English	Japanese	Turkish
French	Korean	Ukrainian
German	Malay	Vietnamese

Home page | Library | Articles | Website map | Site Reviews

www.diagram.com.ua
2000-2024