A little investigation shows that the subject is one that is a profession in itself; consequently, this subject is taken up, not with the idea of writing a treatise on ball bearings, but with the idea of noting certain general principles upon which the theory and practice of ball bearings depend. DEFINITION-PURPOSE A ball bearing is a bearing device consisting of hardened balls and ball paths. (See Sketch 1 for names of parts.) This design reduces friction of motion as compared with ordinary surface bearings, because: 1. Reduction of surface of contact. paths. Surface contact is reduced, as the ball has either two or three points of contact with the races instead of large surfaces in contact. The reduction of friction, as claimed for accurate spheres rolling in accurate paths, may be illustrated by comparing the friction of running a motor-car on a smooth asphalt pavement (smooth path) with same on mud road (no defined path). The friction coefficient of a well-made annular ball bearing is from 0.001 to 0.0002 of the load referred to the shaft diameter and is independent of speed and load. The ball bearing, therefore, has the following advantages over plain bearings, viz.: 1. Reduction in friction. HISTORY Ball bearings in modern use date from introduction of the bicycle. This brought in the adjustable cup and cone (threepoint contact) type. Under the demands of greater load resistance and reliability the two-point contact type, without adjustability, was evolved for use, carrying loads of from a few pounds to many tons. SIZE OF BEARINGS The ball diameter determines the size of the bearing. The load a bearing can carry depends upon the size and number of balls and the speed of rotation. Races must be heavy enough to carry the balls at their designed load without deformation. As the load is taken up between the races by the balls it can be readily seen from Sketches 1 and 2 that the carrying capacity of the bearing must depend upon the strength of the balls and the number of the balls. This is expressed by formula: L=Knd? in which =Load capacity in pounds. n=No balls. d=Ball diameter in 1 of an inch. K=Varies with the type of bearing; also with material and speed. The radial load up to 2000 r. p. m. is unaffected by speed. Above that speed the radial load sharply increases. The strength of a ball depends upon its size (diameter) and is expressed by formula: W=47d’S? -E. d=2VWEES?. S=Maximum safe unit stress of material. d=Diameter of ball. E=Modulus of elasticity. W=Load. The strength of a bearing can be increased by increasing the number of rows of balls. TYPES OF BEARINGS Ball bearings are of many types, each of which has its champions to sing its praises. There are certain sound reasons THRUST PART RADIAL PART INNER RACE HALF IN SECTION HALF IN SECTION THRUST BEARING COMBINED RADIAL AND THRUST BEARING SKETCH 4. SKETCH 3. why there are different designs or types of ball bearings. These 1. Duties. DUTIES (a) Radial bearings. (See Sketch 2.) In radial bearings there can be a slight variance in size of balls, depending on the accuracy required. In thrust bearings, variance 0.001" or above in ball diameter seriously affects the bearing, because the whole load is taken up by the large ball, tending to fracture the races at point of contact. Careful selection of balls is consequently essential in this type of bearing METHOD OF ASSEMBLY The following divisions are made according to method of assembly, viz.: (a) The open type. RETAINER HALF IN SECTION HALF IN SECTION 1 BEARING ETCH 5. The open type bearings are of several kinds, with the principal differences in refinements of retainers, races, etc. The general divisions of this type are: (a) The box or square bearings, as shown in Sketch 7. ing. (See Sketch 6.) The closed type are of several kinds, the difference being in method of filling balls in races. The general divisions under this type are: (a) Slotted (interrupted race), and The slotted types (interrupted race) are of two kinds, viz. : (a) One race slotted. (Sketch 9.) (b) Both races slotted. (Sketch 8.) These types may be either full ball or retainer types. This type of bearing cannot take end thrust, as balls tend to ride over slot edges ; and this type cannot be disassembled without tending to wear the slots into the ball paths. The constant for the carrying capacity of bearing with one race slotted is K=5, 2000 r. p. m. in formula L=Knda. The constant for the carrying capacity of bearing with both races slotted is K=2.5, 2000 r. p. m. in formula L=Knd?. The non-slotted (uninterrupted race) types differ in design in the method of filling, and are divided into three principal types: (a) SKF self-aligning bearing filled by means of retainer. (b) Fafnir type in which races are tapered so that the balls may be entered by springing the races apart. (Sketch II.) (c) Hess-Bright type in which the race diameters are such that six balls can be entered by springing the races, as shown in Sketch io. METHODS OF BALL SPACING There are two methods of ball spacing, the non-retainer type and the retainer type. The non-retainer type or full ball type is a bearing in which centrifugal force is depended upon for spacing the balls. In this type the race is filled with balls, so that there is a space of about one-fourth (1) the ball diameter between any two balls when they are forced apart. Centrifugal force tends to keep the ball spaced when the bearing is in operation at high speed; but, due to there being no retainer to cushion balls and reduced vibration, this type is noisier than the retainer type. It is generally considered that this type of bearing is undesirable for high-speed work where great accuracy is necessary, but in actual practice this type is efficient; examples, torpedo gyro, Sperry gyro-compass, Fafnir bearing. It makes a better bearing for slow speeds with great weight, for according to the formula for carrying capacity of a bearing, it will be noted that the greater the number of balls the greater the carrying capacity of the bearing. |