To reduce friction in the moving parts of machinery, the parts contacting are made smooth, and their area or surface is kept small, as in roller-bearings and ball-bearings. (See BEARINGS). Provision is also made for some form of lubrication, the common plan being to bore small holes leading to the bearing or other point of friction and supply such holes with oil occasionally. In large bearings grooves are often cut diagonally in the surface of shafts to carry the oil to the entire surface. The principle of lubrication is the placing of a film of slippery substance between rubbing sur faces not so slippery. Oil is the most common lubricant and thin oil is used on light-running machinery and heavy, thick oil or grease on heavy machinery where the weight might tend to drive out or use up the light oil quickly. Water is sometimes employed, being cheaper.
i Graphite is an efficient lubricant, employed in cases where it is difficult to retain oil, as on a bicycle chain. In kinetic friction between surfaces that are smooth and well lubricated, the results depend far more upon the nature of the lubricant than they do upon the nature of the rubbing surfaces; and they are also greatly dependent upon the method by which the lubricant is applied. When a journal is lubricated by means of an oily pad placed underneath it, the friction may be more than six times as great as when the same journal is run in a bath of oil. Speed also has a great influence upon the amount of friction developed.
Early books on mechanics and engineering contained assumed laws of friction which were quite erroneous and are now disregarded.
There is no method of determining in advance the actual friction in a mechanism, because all the conditions are never certainly known, and some of them are subject to change daily or hourly. In a cotton-spinning frame, for in stance, the cotton bands that drive the spindles tighten up through shrinkage on a damp day, causing undue pressure on the bearings and often doubling the friction of the machine. The most nearly frictionless machine ever con structed is doubtless the Emery testing ma chine, which is employed for testing great bars of steel, as by pulling them apart. Large sizes of this machine will exert a pull of 500000 pounds and, of course, this involves tremendous strain on the parts of the machine, and when the piece tested is pulled apart and broken and the strain is thus suddenly released, the reac tion of the parts strained is equal to the break ing strain and would rack the machine terribly were not almost frictionless devices introduced. These are the "weighing-heads,° constructed on the principle of hydraulic weighing machines. The blow of reaction is received by a plunger bearing against a body of water, and the shock is absorbed by the water without injury to the machine. It would appear that nature makes frictionless machinery, for the earth and planets revolve in a medium in space so tenuous that no evidence of friction is observable. Were friction present, it would show itself in retarda tion of rotation as well as as orbital revolution. For the internal friction of liquids see VIS