SELECTION OF A MAGNETIC ALLOY FOR MAGNETIC MEASURING SCALES OF KINEMATOMETERS

A. A. Asmus, Ya. A. Belikhmaer

(Presented by the scientific seminar of the Departments of Machine Tools and Metal Cutting and of Mechanical-Engineering Technology)

At the present time, magnetoelectric kinematometers are beginning to be used for monitoring the kinematic accuracy of gear-cutting machines, as well as instruments for single-profile inspection of gears by means of pulse circuits [1, 2, 3].

Magnetoelectric kinematometers operate according to the following scheme.

Sensor disks with a magnetic coating are mounted on the input and output shafts of the kinematic chain being checked. Magnetic marks are applied to the surfaces of the disks. The number of magnetic marks is proportional to the transmission ratio of the kinematic chain.

The signals picked up by magnetic heads from the disks are fed to an amplifier and a phasemeter. If there are no errors in the kinematic chain, the phase shift of the two signals being read remains constant. Any nonuniformity of rotation appears as a phase deviation, which is registered by the phasemeter and recorded by a self-recording instrument.

The reliability of the circuit operation depends to a considerable extent on the quality of the magnetic alloy on the disks. At present the following alloys are used as the magnetic layer:

1. CuNiFe alloy (coercive force 360 Oe, residual induction 5600 G).

The alloy is used in the Czechoslovak kinematometer IMO—S [2]. The alloy is not produced by an electroplating method.

1. Co—Ni alloy (200 ÷ 300 Oe, 5000 ÷ 6000 G) is used in the sensors of kinematometers made by the Chelyabinsk Measuring Instruments Plant.

On the basis of studies carried out by the All-Union Scientific Research Institute of Sound Recording [4], the use of a Co—Ni—P alloy (500 ÷ 800 Oe, 4000 ÷ 5000 G) is recommended for coating drums used in various magnetic storage devices.

At the Tomsk Polytechnic Institute, studies were carried out with the aim of establishing the possibility of using the Co—Ni—P alloy as a signal carrier for the sensors of magnetoelectric kinematometers.

Deposition of the alloy was carried out from the electrolyte:

\[ \begin{aligned} \mathrm{NiCl_2\cdot 6H_2O} &\ldots\ldots 120\ \mathrm{g/l}\ (120\ \mathrm{kg/m^3}),\\ \mathrm{CoCl_2\cdot 6H_2O} &\ldots\ldots 120\ \mathrm{g/l}\ (120\ \mathrm{kg/m^3}),\\ \mathrm{NH_4Cl} &\ldots\ldots 100\ \mathrm{g/l}\ (100\ \mathrm{kg/m^3}),\\ \mathrm{KH_2PO_2} &\ldots\ldots 9\ \mathrm{g/l}\ (9\ \mathrm{kg/m^3}). \end{aligned} \]

Acidity of the electrolyte: \(\mathrm{pH}—3—4\).
Current density: \(10—12\ \mathrm{A/dm^2}\) \((1000—1200\ \mathrm{A/m^2})\).
Electrolyte temperature: \(50—60^\circ\mathrm{C}\).

The characteristics of the alloys Co—Ni—P and Co—Ni were compared; the latter was deposited from the electrolyte:

\[ \begin{aligned} \mathrm{NiSO_4\cdot 7H_2O} &\ldots\ldots 135\ \mathrm{g/l}\ (135\ \mathrm{kg/m^3}),\\ \mathrm{CoSO_4\cdot 7H_2O} &\ldots\ldots 120\ \mathrm{g/l}\ (120\ \mathrm{kg/m^3}),\\ \mathrm{H_3BO_4} &\ldots\ldots 25\ \mathrm{g/l}\ (24\ \mathrm{kg/m^3}),\\ \mathrm{KCl} &\ldots\ldots 15\ \mathrm{g/l}\ (15\ \mathrm{kg/m^3}). \end{aligned} \]

Acidity of the electrolyte: \(\mathrm{pH}—4 \div 5\).
Current density: \(1.5\ \mathrm{A/dm^2}\) \((150\ \mathrm{A/m^2})\).
Electrolyte temperature: \(50 \div 60^\circ\mathrm{C}\).

Influence of Alloy Composition and Coating Thickness

According to the data of B. Ya. Kaznachei and V. M. Zhogina [5], for the Co—Ni alloy there is no dependence between the coating thickness and its magnetic properties. For the Co—Ni—P alloy, this dependence was not studied in [4]. We investigated the dependence of the magnitude of the signal picked up on the coating thickness for the Co—Ni and Co—Ni—P alloys. The dependence was studied in the range of readout speeds \(38 \div 380\ \mathrm{mm/sec}\). It was taken into account that the magnitude of the signal picked up by the magnetic head depends not only on the properties of the magnetic alloy and the coating thickness, but also on a number of other factors, which remained constant when coatings of different thicknesses were compared.

Recording and readout of signals were performed with a magnetic head from a “Melodiya” tape recorder. Contact recording was used, with no gap between the disk and the magnetic head. The signals were recorded from a direct-current source with a voltage of \(4.5\ \mathrm{V}\); the current during recording was \(\approx 8 \div 9\ \mathrm{mA}\). The distance between the magnetic marks along the disk surface was \(\approx 2.6\ \mathrm{mm}\).

The thickness of the magnetic coating was measured on a UIM-21 universal microscope using the IZO-1 attachment.

The amplitude of the output signal was measured with a V3-4 vacuum-tube voltmeter.

From the graph shown in Fig. 1 it follows that the thickness of the magnetic coating has a significant influence on the magnitude of the output signal. As the coating thickness increases, the magnitude of the output signal increases.

In studying the Co—Ni—P alloy, the coating thickness varied within the range from 5 to \(39\ \mu\).

In all cases, both in carrying out the principal investigations and in repeated check tests, a quite definite dependence was found: as the coating thickness increases, the magnitude of the picked-up signal increases.

For all coating thicknesses, a linear dependence was observed between the readout speed and the magnitude of the picked-up signal.

Figure 1 presents the dependences for coating thicknesses of 5, 9, and \(15\ \mu\) for the Co—Ni—P alloy.

Similar studies were carried out for the Co—Ni alloy with a coating thickness from 3 to 16 μm. The results of the studies are shown in Fig. 1. The nature of the effect of coating thickness on the magnitude of the output signal for the Co—Ni alloy is analogous to that for the Co—Ni—P alloy.

Fig. 1. Effect of coating thickness and alloy composition on the magnitude of the signal taken off.

The magnitude of the output signal also depends on the composition of the alloy. In Fig. 1 this effect is visible for different coating thicknesses. In all cases, at the same coating thickness, the magnitude of the output signal for the Co—Ni—P alloy is greater than for the Co—Ni alloy.

Effect of recording density

In all the above-mentioned studies, signal recording was carried out with low density (the distance between signals was 2.6 mm), in order to exclude interaction between signals.

To increase the accuracy of magnetoelectric kinematometers and instruments for single-profile inspection of gears, it is necessary to use recording with maximum density. Recording density is characterized by the number of magnetic marks per 1 mm of length.

The effect of recording density on the magnitude of the output signal was studied according to the following procedure.

On each disk with a specified coating thickness, sinusoidal signals from the ZG-10 audio generator were recorded. Recording and readout of the signals were carried out at a constant linear speed of 38 mm/sec. To obtain different recording densities, the frequency was varied.

The dependence of the magnitude of the output signal on recording density for various coating thicknesses is given in Fig. 2.

Fig. 2. Effect of recording density on the magnitude of the detected signal for various coating thicknesses. Readout speed: 38 mm/sec.

Vertical axis: \(U\), mV.
Horizontal axis: recording density.

Curve labels shown in the figure:

• Co–Ni–P; \(t = 39\,\mu\text{m}\)
• Co–Ni–P; \(t = 35\,\mu\text{m}\)
• Co–Ni–P; \(t = 15\,\mu\text{m}\)
• Co–Ni–P; \(t = 9\,\mu\text{m}\)
• Co–Ni–P; \(t = 5\,\mu\text{m}\)
• Co–Ni; \(t = 16\,\mu\text{m}\)
• Co–Ni; \(t = 9\,\mu\text{m}\)
• Co–Ni; \(t = 3\,\mu\text{m}\)

As the recording density is increased, the amplitude of the output signal decreases. This decrease becomes sharper as the coating thickness is increased.

At a recording density of more than four points per 1 mm, the greatest value of the output signal occurred at a coating thickness \(t = 5\ \mu\) for the Co—Ni—P alloy and at \(t = 9\ \mu\) for the Co—Ni alloy. Moreover, the output signal at \(t = 5\ \mu\) (Co—Ni—P alloy) is greater than at \(t = 9\ \mu\) (Co—Ni alloy).

The decrease in the output signal with increasing coating thickness at a recording density of more than four points per mm can be explained by the fact that, as the coating thickness increases, the mutual influence of the magnetic fields of two adjacent magnetic marks increases.

It was found that the amplitude of the output signal depends on the recording method. For the magnetic head of the “Melodiya” tape recorder, recording signals from a pocket battery (voltage 4.5 V, current 8–9 mA) does not make it possible to obtain the maximum signal during readout.

When signals were recorded from the ZG-10 audio-frequency generator, the signal during readout was larger than when recording from a pocket battery.

Conclusions

1. For coating the disks of magnetoelectric kinematometers and other similar instruments, it is advisable to use the Co—Ni—P alloy instead of the Co—Ni alloy currently used.

2. The coating thickness should be selected depending on the recording density.

At recording densities up to four points per mm, it is advisable to apply a coating 20–30 \(\mu\) thick. At recording densities of more than four points per mm, the optimal coating thickness is 5 \(\mu\) for the Co—Ni—P alloy and 10 \(\mu\) for the Co—Ni alloy.

REFERENCES

1. Ya. B. Gozman. Method for checking the kinematic accuracy of gear-milling machines. Invention, class 49d, 15₀₆, No. 125894, dated 29/IV-1957.

2. K. Shtepanek, Ya. Bauer. Magnetic measuring scales. Article in the collection of works of Czechoslovak institutes. Investigation of Metal-Cutting Machine Tools, pp. 66–96, Mashgiz, 1962.

3. B. Ya. Verkhoturov, M. I. Kandalov. Progressive Technology of Machining and New Measuring Means. Chelyabinsk, 1964.

4. B. Ya. Kaznachei, V. M. Zhogina. Electrodeposition of a high-coercivity nickel—cobalt alloy. Proceedings of the Institute of Sound Recording. Gosizdat, “Iskusstvo,” 1957, issue 1, pp. 91–93.

5. B. Ya. Kaznachei, V. M. Zhogina. Electrical production of a nickel—cobalt alloy with specified magnetic characteristics. Proceedings of the Institute of Sound Recording. Gosizdat, “Iskusstvo,” issue 1, 1957, pp. 79–90.