Taylor - Machining Productivity Improvement - Part V

HOW TO MAKE AND RECORD EXPERIMENTS ART OF EXPERIMENTING

DEFINED AS DETERMINING EFFECT OF VARYING ONE ELEMENT WHILE ALL OTHERS ARE HELD CONSTANT

210 In paragraph 55 (Part 1) we have broadly defined the art of experimenting on this subject as an attempt to hold uniform and constant all of the elements which affect the final results under investigation except the one variable which is being studied, while this one is systematically changed and its effect upon the problem carefully noted.

211 It is the necessity of holding these variables constant which makes all these experiments so difficult, causes the apparatus and forgings tested to be so large and expensive, and consumes perhaps four-fifths of the time of the experimenter. Time and again in our work it has required days and sometimes weeks to prepare for an experiment which, after we have succeeded in obtaining uniformity in all the elements, has been made in a few days or hours.

212 A description of how to maintain uniform conditions is then virtually a description of the art of experimenting on cutting metals. The writer will, therefore, begin by explaining in detail the precautions which must be taken to secure this uniformity.

THE METAL WHICH IS TO BE CUT

213 In determining the effect which any one of the more important elements has upon the cutting speed,—such, for instance, as the effect of the feed or thickness of shaving upon the speed,-—it is necessary to make many standard cuts of 20 minutes’ duration in succession for each thickness of shaving, in order to find the exact speed at which the tool will be ruined at the end of 20 minutes, as described in paragraph 137. For this reason the information required for each thickness of shaving may cause hundreds of pounds of metal to be cut up into chips; and since in determining the law many different thicknesses of shaving must be experimented with‘, and the metal cut must be of uniform hardness, in order to obtain accurate results not only a. large experimental forging is called for, but the metal throughout the forging must be as nearly as possible uniform.

214 We have finally adopted as the standard size of forging best suited for experiments apiece about 10 feet long by 24 inches in diameter and weighing about 15,000 pounds. For this purpose we prefer  forging made from a large ingot cast from a carefully melted heat of open hearth steel, and forged with a large reduction in the diameter so as to obtain as nearly as practicable uniform working of the metal throughout from the outside to the central portions of the forging. This forging in all cases requires thorough annealing,and in some cases an oil hardening previous to the annealing. Before using it, experimental standard tensile test bars should be cut from each end of the forging and broken so as to prove the uniformity of the metal, and, if necessary, the metal should be re-treated by tempering and annealing until it has been demonstrated through breaking a sufficient number of test specimens to be uniform.

215 Masses of cast iron upon which tools are to be tested should in all cases have their central portion cored out, so that the metal will be cooled from the center outward as well as from the outside inward, thus producing an annular ring of cast iron from three to four inches thick; which, if sufficient precautions are taken to insure regular cooling,will be comparatively uniform throughout. We have used annular masses of cast iron, made in this manner, of 24-inch diameter with 12- inch cores, 12 feet long; of 15-inch diameter, 10-inch cores, 8 feet long; of 13-inch diameter, 7 feet long; of 13-inch diameter, 9-inch cores, 10 feet long. On the whole, we have obtained the most uniform results from a casting of 24~inch diameter, 16-inch core, and 10 feet long, walls of casting having a maximum thickness of 4 inches poured on end. We have found the central portion of the annular rings which were 6 inches in thickness to be very materially softer than the outer or inner portions of the ring. It has proved to be much more difficult to obtain sufficiently large and uniform masses of cast iron than of steel.

216 A large diameter is called for in both forgings and castings, not only for the sake of obtaining a large quantity of uniform metal, but also to avoid as far as possible the effect of chatter upon the tool, caused by the deflection of the work; and the writer wishes to emphasize the necessity for large masses of metal if the experimenter hopes for trustworthy results.