STUDY OF SURFACE FILMS ON COMMUTATORS OF ELECTRICAL MACHINES

R. F. Bekishev, B. I. Kostylev, A. I. Skorospeshkin

(Presented by the scientific seminar of the Departments of Electrical Machines and General Electrical Engineering)

The surface film that forms on the commutators of electrical machines has a very substantial influence on commutation. The reliability of operation of the sliding brush contact and of the commutator electrical machine as a whole is determined to a large extent by the condition and properties of the surface film on the commutator. Therefore, the study of the actual nature of commutator films and the investigation of their phase composition and structure are highly relevant. Moreover, no thorough studies on this question are available [1].

At present, the physical properties and structural features of thin films formed in the process of surface oxidation of various metals, including copper, have been studied fairly fully [2]. Such investigations are usually carried out using modern physicochemical methods, such as electron-microscopic study of thin films, electron-diffraction investigation, and X-ray structural analysis.

In studying commutator films, the general principles characteristic of ordinary oxide films on metals remain valid when conducting electron-microscopic and X-ray structural investigations.

We studied surface films formed both on the commutators of electrical machines and on slip rings by brushes of various grades in three main directions:

1) Investigation of the microstructure of commutator films.
2) Determination of the nature, orientation, and phase composition of the films.
3) Measurement of the thickness of surface films.

The UEMB-100 electron microscope used to study the microstructure of commutator films has a range of electron-optical magnification of 250–150,000 and a resolving distance of 15 Å. Consequently, it is possible to distinguish details on the surface film of the order of several atoms in size and to determine the structure and morphology of the films with sufficient accuracy.

Determination of the nature and orientation of the constituents, as well as of the phase composition of the surface films, was carried out by the method of electron-diffraction investigation. The capabilities of the UEMB-100 electron microscope make it possible to conduct a parallel investigation of surface films, namely, to obtain microscopic images and to carry out microdiffraction.

Quantitative phase analysis of commutator films, that is,

determination of the number of phases and their content was carried out by us by the method of X-ray structural analysis using the URS-50 I/M apparatus.

Both electron-microscopic and electron-diffraction studies, as well as X-ray structural studies of the surface films, were carried out by transmission examination on specially prepared specimens. In this case it is necessary to separate the surface films from the substrate without destroying them.

Measurement of the thickness of the surface films is also possible after their separation from the copper of the commutator. As the studies showed, the most promising method for measuring the thickness of commutator films is the interference method, which has a sensitivity of \(10\ \mathring{\mathrm A}\) for an area equal to \(10^{-4}\ \mathrm{mm}^{2}\). In the present study the film thickness was measured with a Linnik MII-4 microinterferometer.

The main difficulty in preparing specimens for carrying out the above-mentioned studies lies in the possibility of separating thin surface films from the copper of the commutator. When separating surface films, the following conditions must be fulfilled:

1. The commutator films must be separated without destruction.

2. In the process of separating the film from the substrate, the structure, surface morphology, and phase composition of the commutator film must remain unchanged.

3. The thickness of the commutator film before separation from the substrate and after separation must not change.

At present, electrochemical and chemical methods have come into wide use for separating oxide films of metals without destroying them. Analysis and experimental verification of numerous procedures [3] developed for separating films from a copper substrate showed that some procedures are applicable for separating the surface film formed on commutators and slip rings. In the present work, electrochemical and chemical methods were used for separating the polish film from the copper of the commutator.

The electrochemical method is based on anodic dissolution of copper beneath the layer of polish film. The compositions of the electrolytes and the conditions of anodic dissolution of copper adopted for separating the surface film from commutators and slip rings in the present study are presented in Table 1.

Table 1

The process of separating the surface film from commutator bars and slip rings by a chemical method was carried out in a 4–5% solution of bromine in methyl alcohol for 3–5 min.

Before immersion in the solution (when using any of the indicated methods), the specimen was prepared. A commutator bar or part of a slip ring was coated with an inert film (zapon varnish, paraffin) so that only the areas with the patina were in contact with the solution. To intensify the dissolution of the copper beneath the surface film, the patina was cut into squares measuring 2–4 mm. After immersion in the solution, the commutator film swells, but is still rather firmly bound to the specimen. Therefore the actual separation was carried out in a washing liquid, for example in methyl alcohol, making use of the film’s poor wettability. In some cases, for example when separating very thin commutator films, it is impossible to detach them in the washing liquid. Therefore, after removal from the solution, washing, and air drying, a layer of gelatin dissolved in warm distilled water was applied to the areas of the commutator film to be separated. After 5–10 min the gelatin layer hardens and easily pulls the surface film away from the commutator copper. During subsequent washing in distilled water, the gelatin layer dissolves readily. The use of gelatin in separating commutator films greatly facilitates the process.

The use of the above methods for separating the film from the substrate is possible only if the film is undamaged. In this connection, special attention was paid to preparing the commutator and to the process of building up the surface film.

The surface film was built up on a special installation, whose rotating parts were carefully balanced; the runout amplitude of each of the commutators did not exceed 2.5 μm. The treatment of the commutators and slip rings, mounted on a shaft with pressed-on bearings, was carried out with a diamond tool. The surface finish after diamond machining reached class 8–9. The commutators and slip rings were made of M1 copper. After careful bedding-in of the brushes, the commutators operated for 120 h without current and for 300 h at a current density of 10 A/cm².

Electrographitized brushes of grades EG-4E, EG-14, EG-63, and EG-74 were used for the investigation.

After the surface film had been built up, the commutator was divided into individual bars, the patina being separated from them by one of the methods considered above.

Further preparation of the specimens was determined by the chosen method of investigation.

Electron-microscopic and electron-diffraction study of the surface film was carried out in a transmitted electron beam. As the investigations showed, the structure and phase composition of the commutator film did not change when different methods were used to separate the film from the substrate.

In order to obtain more complete information from the electron-microscopic photographs, images were studied both in bright and in dark field. Bright-field microphotographs give an image of all crystals of the surface film through which the undeflected electron beam passes. Contrast in the image is produced by different absorption of electrons in film areas having different density. Interpretation of bright-field photographs makes it possible to reveal the distribution pattern of the individual components of the commutator film and the total number of crystals.

Dark-field micrographs make it possible to obtain an image only from that part of the surface of the collector film under study which has been preliminarily selected by means of electron diffraction. Interpretation of micrographs obtained in a dark field makes it possible to determine the shape and dimensions of individual crystals when the magnification of the image is known.

Figure 1 shows a bright-field micrograph of the surface film formed by EG-74 brushes. The photograph was taken at an electron-optical magnification of 87,000. This image reflects the general structure of the film surface sufficiently completely. The broad dark bands (in the positive) are thickened regions of the film consisting of amorphized graphite mixed with finely crystalline copper oxide, Cu₂O. The dark particles of rounded shape are larger Cu₂O crystals.

The micro-electron diffraction pattern from a thin region of the surface film is shown in Fig. 2. Calculation of the electron diffraction pattern shows that the diffraction rings correspond to the lines of copper oxide, Cu₂O.

Fig. 1

Fig. 2

A dark-field micrograph in the light of a bright ring in the electron diffraction pattern is shown in Fig. 3. The light regions (in the positive) are particles of copper oxide, Cu₂O. The electron-optical magnification is 105,000. As can be seen from the micrograph, the copper oxide crystals have an irregular shape. The sizes of the crystallites vary from 150 Å to 2000 Å.

The study of surface films formed by EG-63 grade brushes was also carried out using various methods of separating the film from the collector lamella. Figure 4 shows a micrograph of a film separated by an electrochemical method in a solu-

Fig. 3

Fig. 4

copper sulfate. Electron-optical magnification—102,000. Analysis of the microphotograph shows that the structure of the film is characterized by considerable nonuniformity in thickness. The thickened bands of the film are elongated along the sliding track. Dark graphite flakes are visible on the surface of the film.

A microelectron diffraction pattern from an individual area of the film is characterized by the presence of continuous rings from a fine-crystalline copper oxide deposit, Cu$_2$O, and spot diffraction from graphite crystals (Fig. 5).

Fig. 5

Analysis of the electron-microscopic and electron-diffraction studies carried out showed that the surface films formed on commutators and slip rings by brushes of various grades include mainly copper oxide, Cu$_2$O, in the fine-crystalline state, as well as amorphous and crystalline graphite. The sizes of the copper oxide particles vary over wide limits (from 200 Å to 2000 Å). The graphite crystals have sizes of 100–4000 Å. In some electron diffraction patterns, lines with spacings of 1.7 Å and 1.5 Å are encountered; these may be assigned to the crystal lattice of copper oxide CuO. However, these lines occur rarely and are very weakly expressed; therefore it may be considered that copper oxide CuO, if it is present in commutator films, occurs in very small quantities (not more than 10%).

In the overwhelming majority of cases, the electron diffraction patterns lack lines with spacings of 2.45 Å and 2.12 Å, which may be explained by the orientation of Cu$_2$O crystallites perpendicular to the (111) directions to the copper surface of the commutator.

The lines of amorphous graphite are practically not distinguishable in the electron diffraction patterns in those cases where there are distinct lines of copper oxide. This phenomenon is typical of a mixture of two phases, one of which is more ordered [4]. However, cases are possible in which the lines of copper oxide Cu$_2$O may be present together with the lines of finely divided graphite.

Quantitative phase analysis of the films was carried out by the X-ray structural method.

The task of X-ray phase analysis consists in determining the interference pattern of the intensities of interference lines when X-rays interact with the substance under study.

Analysis of the commutator films formed by brushes of various grades was carried out on a URS-50IM ionization X-ray unit intended for measuring the absolute intensities of X-ray reflections.

X-ray phase analysis begins with qualitative determination of the phases constituting the surface film under study. Determination of the components included in the composition of the commutator films was carried out by comparing calculated interplanar spacings

and of the measured relative intensities of the interference maxima with the theoretical data [5] for the assumed elements or compounds.

Intensity curves of X-ray reflections

Fig. 6

Figure 6 presents a recording of the intensity curves of reflections obtained on a URS-50 IM X-ray unit from surface films formed by EG-4E and EG-14 brushes. The ordinate axis gives the absolute values of the intensities of the X-ray reflections; the abscissa axis gives the double angle of reflection.

Analysis of the X-ray diffraction patterns and of the results of their calculation shows that, for the surface films investigated, the interference maxima of the reflections correspond in most cases to the interplanar spacings of cuprous oxide and graphite. Interference maxima are observed from all the principal reflecting planes \((hkl)\). In the X-ray diffraction patterns, lines corresponding to the interplanar spacings for copper oxide CuO occur in small quantities; however, the intensity of these lines is very weak.

Thus, the surface films formed on commutators by brushes of different grades consist, as a rule, of cuprous oxide and graphite, present in different amounts depending on the brush grade.

Considering the surface film as a two-phase mixture, the ratio of the amounts of cuprous oxide and graphite in the film can be determined by the method of homologous pairs. The intensity of the interference lines of each of the phases is proportional to the amount of this phase in the surface film and to the reflecting power of the crystal for reflection from the given plane \((hkl)\).

The quantitative phase analysis of the commutator films investigated, carried out by the indicated method, made it possible to determine with sufficient accuracy (10–15%) the content of cuprous oxide and graphite in the surface films (Table 2).

Table 2

As can be seen from Table 2, the highest content of cuprous oxide is found in commutator films formed by hard electrographitized brushes (EG-74, EG-63). The amount of cuprous oxide in surface films formed by the soft electrographitized brushes EG-49 is considerably smaller; however, in this case as well the graphite content does not reach 50%.

The large amount of cuprous oxide in all the surface films studied requires that, when studying phenomena in sliding brush contact, the semiconducting properties of commutator films be taken into account. In this case, however, it is necessary to know the thickness of the surface film.

Electron-microscopic and electron-diffraction studies of commutator films showed that their thickness may vary over wide limits across the width of the sliding track.

By the gravimetric method it is possible to measure only the average value of the thickness of surface films over an area of not less than 10 mm². The accuracy of the method does not exceed 25–30%. The use of this method for determining the thickness of commutator films is practically impossible when the content of the individual constituents of the film is unknown. Therefore, measurement of film thickness must be preceded by complex investigations for the quantitative determination of the content of the components in commutator films.

The thickness of the surface films was measured by us by the interference method on a Linnick MII-4 microinterferometer, which makes it possible to determine film thicknesses within the range from 300 Å to 15000 Å. In the field of view of the microinterferometer, the interference fringes and the investigated region of the surface film are observed simultaneously. The surface film is placed on a glass substrate and coated with a thin layer of aluminum to increase reflectivity. At the boundary of the layer (film–glass), the optical path difference between the interfering rays changes, as a result of which the interference fringes are displaced. During measurement, the magnitude of the curvature of an interference fringe is expressed in fractions of the interval between interference fringes. The magnitude of the interval between fringes, as well as the displacement of the interference fringes, was determined with the aid of an MOV-1-15 screw ocular micrometer.

The interference pattern obtained from the commutator film formed by EG-74 brushes is presented in Fig. 7.

As calculations of the interference patterns showed, the thickness of the commutator films in different regions varies over wide limits. The results of measurements of the thickness of commutator films formed by brushes of different grades are presented in Table 3. The measurements were carried out at three points of the commutator film differing in thickness.

Figure 7

Table 3

Analysis of Table 3 shows that, for all the films studied, their thickness undergoes significant changes across the width of the sliding track. The films formed by EG-4E brushes are considerably thinner than the films produced by EG-74 brushes, although the running-in conditions and the method of preparing the specimens were identical.

Thus, hard electrographitized brushes create collector films of greater thickness than soft brushes.

In studying films formed by brushes of different polarities, the interference method revealed no substantial difference in the thickness of the tarnish under cathodic and anodic brushes. In both cases, the film thickness at various points of the section under study underwent significant changes.

In this connection it should be noted that the different level of the contact voltage drop under anodically and cathodically polarized brushes cannot be explained by the influence of the thickness of the surface films. This is all the more so because the resistance of the body of the surface film is insignificant.

In our view, the differences caused by the polarity of the brush are determined by the thickness of the blocking layer formed between the surface film and the copper of the commutator or contact ring.

Conclusions

1. Electron-microscopic, electron-diffraction, and X-ray structural studies make it possible, with sufficient accuracy, to determine the structure, phase composition, and construction of the surface films formed on commutators and contact rings.

2. Commutator films consist, as a rule, of finely crystalline cuprous oxide and graphite, occurring both in the amorphous state and in the form of oriented crystals.

3. For quantitative determination of the phase content in surface-

... film it is advisable to use X-ray phase analysis. Studies carried out by the method of homologous pairs showed that electrographitized brushes form commutator films with the following copper-oxide content: EG-49—55%, EG-14—62%, EG-63—67%, EG-74—70%.

1. The most promising method for measuring the thickness of surface films is the interference method; using it, it was found that the thickness of commutator films formed by brushes of various grades varies within the following limits: EG-49—270–1000 Å, EG-14—360–2000 Å, EG-63—480–3500 Å, EG-74—600–5000 Å.

The differences due to the polarity of the brush are determined not by the thickness of the surface film, but by the thickness of the rubbing layer formed between the film and the copper of the commutator.

REFERENCES

1. E. B. Granovskii, V. S. Titov. Electron-microscopic study of the surface film of commutators of electrical machines, Elektrotekhnika, No. 1, 1968.

2. V. S. Benar. Oxidation of Metals. Vol. 1, Metallurgiya Publishing House, 1968.

3. P. Jacquet. Electrolytic and Chemical Polishing. Metallurgizdat, 1959.

4. Z. G. Pinsker. Electron Diffraction. Publishing House of the Academy of Sciences of the USSR, 1950.

5. L. I. Mirkin. Handbook of X-ray Structural Analysis of Polycrystals. Fizmatgiz, 1961.