International projectionist (Jan-Dec 1939)

current in the armature coils during commutation, from the resistance of a glazed commutator surface to the passage of load current, from the unbalancing effects of selective action, and from the friction of the brushes may at times be enough to seriously affect the efficiency or the satisfactory per FIGURE 1 Exaggerated drawing illustrating nature of contact between brush and commutator formance of a motor or generator. However, with proper brush selection and operation, these losses can be kept to a small value. Quietness of operation is dependent, primarily, on the maintenance of uninterrupted contact between brushes and commutator. Even where freedom from noise is not an essential characteristic of performance, the firm contact on which quiet operation depends has an important influence on satisfactory performace in other respects. Brush life should not be made a primary objective in brush application. If the other elements of satisfactory brush operation are attained, reasonable brush life may well be expected; but long brush life obtained at the sacrifice of performance in some other respect may prove far from economical. • Nature of Brush Contact Before undertaking a discussion of the numerous mutually dependent factors on which the attainment of satisfactory brush operation depends, it would be well to consider the nature of the contact between a brush and the commutator. It should be realized that the commutator and the brush, however highly polished they may be, do not necessarily present to each other perfectly smooth surfaces of identical curvature. Due to the play of the brush in its holder, and the yielding of supporting members under stress, the brush face may have a slightly longer radius of curvature than the commutator. In such case the actual area of solid contact, for a particular brush at a given instant is probably confined to a relatively short span of the brush face in the direction of rotation. Furthermore, there are indications that within this area actual contact between brush and commutator is limited to a number of points carrying current at an extremely high current density. Movement of the commutator under the brush face, mechanical abrasion and the destructive effects of the high local current densities cause constant shifting of these points of contact so that solid contact on any single point of either commutator or brush face is probably of extremely short duration, except for firmly imbedded points which protrude to a substantial degree beyond the surrounding surface. There are three types of paths through which the current flows between commutator and brush: the points of solid contact, mentioned previously; adjacent areas in which free particles of carbon, graphite, copper or other electrically conducting dust provide a conducting path; and open gap, across which some current may pass in the form of an arc. Fig. 1 illustrates, in exaggerated form, this conception of brush to commutator contact. The assumption that actual contact between brush and commuator is limited, for a given instant, to only a small portion of the brush thickness does not justify the further assumption that commutation must be effected while the mica gap between two commutator segments is traveling that short distance. Not only do the actual points of contact shift rapidly, but the same is true of the area within which those points lie. It is quite possible for contact to occur at points scattered over essentially the full area of the brush face during a single commutation interval. Therefore, the fact that a brush shows uniform polish over the entire face does not preclude the possibility of contact at any given instant having been limited to a relatively small portion of the total face area. When there are several brushes of hke polarity, connected in multiple, one seems fully justified in considering the cycle of commutation to occupy the full time required for the mica gap to traverse the brush face from leading to FIGURE 2 Commutator surface film greatly magnified trailing edge. There may be instances in which the movement of the area of contact during a single commutation interval fails to cover the full brush thickness and completion of commutation is forced to take place within a very short arc of rotation. However, with normal performance and good brush and commutator surfaces, it seems probable that such cases are the exception rather than the rule. There appears to be no necessity, as has at times been suggested, of abandoning the classical concept that the normal period of short-circuit for a coil undergoing commutation extends from the leading edge to the trailing edge of the brush. Inasmuch as firm contact between brushes and commutator is essential to satisfactory brush performance, protruding mica cannot be tolerated. Fortunately, the practice of undercutting the mica below the surface of the copper segments has become general. If this operation is carefully performed, and no fins of mica are left at the sides of the slots, brush application is made substantially simpler. In the following discussion it is assumed that the mica has been properly under-cut and no reference is made to complications which would certainly result were protruding mica present. It is further assumed that the commutator has been finished to a smooth surface of essentially true cylindrical form, free from high or low spots of appreciable eccentricity or area. • Commutator Surface Film One of the most important factors affecting brush performance is the surface film on the commutator. The exact nature of the surface film varies greatly th operating and atmospheric conditions, and there is evidence that the presence or absence of copper oxides in this commutator film may greatly influence performance. Graphite is also frequently found to be a constituent of the surface film. In some cases this may be rubbed into the irregularities of the commutator surface by the wiping action of the brush. In other cases, there are indications that the graphite is transferred from brush to copper by electrolytic action. The brush face and commutator surface, in highly magnified form, may be considered as resembling the sectional drawing in Fig. 2. No matter how care FREE PARTICLES GRAPHITIC FILM OXIDE FILM fully prepared, or how highly they may be polished, a powerful microscope will disclose minor irregularities in brush and commutator surfaces. A raw copper surface on the commutator does not lend itself to good brush performance. Some oxidation of 8 INTERN ATIONAL PROJECTIONIST