Page images
PDF
EPUB
[graphic]

000

FIG. 1.-PINIONS RUPTURED BY BEING FORCED ON TAPERED PLUG. THEY PRESENT RUPTURES WHICH WOULD HAVE BEEN ENTIRELY UNSUSPECTED BEFORE PHOTOELASTIC ANALYSIS.

[graphic][graphic]

FIG. 2.-V-SHAPE RUPTURE. THE PINION TOOTH HAS BEEN SUBJECTED TO A HIGH RADIAL PRESSURE AND NORMAL TORQUE.

FIG. 3.-HIGHLY STRESSED PINION. AN EXAMPLE OF A SERVICE FRACTURE RESULTING FROM VERY HIGH RADIAL MOUNTING STRESSES.

perature, due to the desuperheating. In addition, there will be no tendency for the high-pressure safety valves to blow off, thus relieving them from unnecessary wear.

The 350-pound steam system will be fully protected by the safety valves mounted on the discharge of the reheater. These valves are sufficient to discharge the full capacity of the high-pressure boiler should the pressure tend to rise in the 350-pound system.

After such a shut-down, the turbine may be restarted by opening its throttle and gradually closing the by-pass, or the boiler can be shut down and taken off the line at the convenience of the operating force. In either case, the routine will be approximately the same as in normal starting or shutting down.-"General Electric Review," Nov., 1924.

CAUSES OF FAILURE OF METAL SURFACES SHOWN BY POLARIZED LIGHT.

By PAUL HEYMANS,

ASSISTANT PROFESSOR OF THEORETICAL PHYSICS, MASSACHUSETTS
INSTITUTE OF TECHNOLOGY.

This article will illustrate how phenomena of double refraction, sometimes looked upon as exclusively of the domain of physics, particularly of optics, are at present used with rather surprising results for the solution of problems encountered by the mechanical and structural engineer, and by the metallurgist.

We shall not dwell upon technical discussion or description of the method. We have done this in several other papers and articles published in this* and other journals. Let us rather consider in more detail one of the recent investigations which we have carried out, partly in our own laboratory at the Institute, partly in the Research Laboratory of the General Electric Company of Schenectady, N. Y., with A. L. Kimball,† J. L. Williamson,‡ G. R. Brophy and T. H. Frost.§

The investigation of the causes of failures of railway motor gear_pinions of apparently only endemical nature had been brought up by the Railway Motor Engineering Department of the General Electric Company.

Metallurgically, the steel appeared in excellent condition and capable of long and satisfactory life.

Structurally, the pinions seemed to be properly designed and assurance was given that they were not carrying an excessive load.

In the study of the design of the pinion, the ordinary formulae for calculation of gear teeth strength could not be considered satisfactory, but a direct stress analysis by the photoelastic method was made, which alone could offer full reliability for the statement regarding the structural strength of the pinions under consideration.

During this direct analysis of the state of stress at the different points of the pinion, attention was soon called to the changes in the conditions of stress resulting from variations in the mounting stress of the pinion on the shaft. These pinions present a slightly tapered bore, slightly smaller than the shaft of same taper. For mounting, the pinion should be brought to the temperature of boiling water, slid on the shaft, and thus allowed to shrink on the shaft in cooling. This operation, when properly executed, secures perfect adherence on the shaft during service and easy removal of the pinion

* "Tech Engineering News," June, 1922.

Research Laboratory, General Electric Co., Schenectady, N. Y.

Railway Motor Engineering Dept., General Electric Co., Schenectady, N. Y.
Department of Physics, M. I. T.

when replacement becomes necessary. However, when defectively mounted, the pinion is apt to get loose on its shaft while in service. The mounting of the pinions for motors in railway service is usually made by repair shop mechanics generally unaware of the necessity of careful mounting. It has been found that the mounting had been done by putting the pinion into a forge oven, sliding it on the shaft and driving it by means of a heavy sledge-hammer as far as possible on the tapered shaft. The mechanic then felt satisfied that he would never again encounter the trouble of a pinion getting loose from the shaft. However, what is the effect of the stresses set up by such an abnormal mounting going to be on the service strength of the pinion?

When examining under polarized light the celluloid models of the pinions after they are shrunk on their shaft, it was shown that the maximum stress occurred at the points where the radial lines passing through the middle of the teeth cross the bore-the dangerous section occurring along this radius. Mechanical tests on steel pinions were made subsequent to these photoelastic tests, forcing tapered plug into the tapered bore. Fig. 1 is a photograph of the ruptured pinions. They present ruptures which would have been entirely unsuspected before photoelastic analysis but are a remarkable verification of it.

The photoelastic analysis also revealed that the sections of dangerous stresses are different for different values of the radial pressure and the applied torque load, and that as a result of highly localized stresses, a pinion under a high radial pressure and normal torque load would fracture as shown in Fig. 2, with a V-shape. The V would become deeper and sharper as the radial pressure increased until, as a limiting case, with maximum radial pressure and no torque, an approximately straight radial crack is obtained as shown in Fig. 1.

Service fractures show, as expected after such an investigation, all degrees of angularity, from the flat break with low mounting stresses, to the very sharp angle of highly stressed pinions. As an example, Fig. 3 shows the result of a mounting stress so high as to nearly burst the pinion which with light torque was soon fatigued in operation.

The investigation, of which we have here brought up a few salient points, shows how the causes and responsibilities of failure may be traced. It is also one of the many possible illustrations of physics at the service of engineering.

We might mention that several important investigations are in progress at present in our laboratory of photoelasticity, among which we are permitted to mention the study of the stresses in rigid airships and the study of the dynamic stresses in rapidly rotating gear pinions.-"The Tech Engineering News," Jan., 1925.

COPPER WHICH BENDS ONLY ONE WAY.

Copper bars that can be bent double with one finger, but which require strength to straighten again, are expected to lead to a greater understanding of the properties of metals. The bars, which are really single crystals of pure copper, were produced in the Research Laboratory of the General Electric Company, at Schenectady, N. Y., and have been subjected to many kinds of examinations, with the revelation of numerous unexpected facts. Knowledge about the properties of metals has been limited in the past to observations of masses of small crystals. The usual piece of metal is a conglomeration of small, closely packed crystals, with the crystalline structure usually apparent at a glance. Zinc, for instance, is known as a brittle metal; a rod of it can be bent but slightly without snapping. Yet investigations of small, single zinc crystals show that any one crystal of the metal can be drawn out to six times its length in one direction; in another direction

it is extremely brittle. The properties of zinc thus depend upon how the crystal is examined-whether "with the grain" or against it. The usual piece of zinc is really a collection of small crystals pointing in all directions, so that the properties are the combined qualities of the small crystals in the different axial directions. The same holds true for other metals and other substances.

A single crystal of copper seven-eighths of an inch in diameter and six inches long, as well as numerous smaller crystals of the same metal, have been produced by Dr. Wheeler P. Davey, of the Research Laboratory. These crystals, obtained by a modification of the method devised by Dr. P. W. Bridgman, of Harvard University, are much larger than any previously recorded.

Very gradual heating and cooling of pure copper in an electric furnace is the secret of the success in producing them. The necessary amount of pure copper, in the form of a bar, was placed in a closed, cylindrical carbon crucible, and slowly passed through the electric furnace. If molten metal is cooled quickly, the resultant mass is composed of very small crystals; if the melt is cooled slowly, the crystals are larger. Doctor Davey cooled the melt so slowly that only one crystal was produced, and that included the entire melt. The atoms had plenty of time in which to arrange themselves as they desired-to build up a single crystal rather than a multitude of small

ones.

Several interesting results have been obtained with the large crystals. A piece about the size of a lead pencil, if given a jerking motion, bends as easily as does a stick of soft wax; it cannot be bent back, however, any more easily than a similar piece of ordinary copper. When the copper is a single crystal, all of the atoms are arranged in columns, equally spaced. When the bar is bent, the spacing is changed; the atoms on the inside curve are pressed together, and those on the outside are spread apart. Strains are set up and the crystal structure is altered. The bar becomes an ordinary piece of copper, of smaller crystals, facing in all directions.

If the surface of the large crystal is nicked or dented, the structure in the neighborhood is changed in the same way. It is similarly affected by filing or polishing. When one of the bars is polished it is necessary to take off a mil or less at a time. Even then the structure of the new surface is altered. The condition is remedied by etching away the surface with the usual acid bath.

An etched bar of the copper appears to be rough. There seem to be alternate dark and light lines. The appearance of the lines is due to the fact that the acid etches more easily in some directions than in others. The directions in which it etches with the greatest difficulty are parallel to the axis of the crystal.

Externally the large, single copper crystals differ little from the usual metal. X-ray analysis, however, furnishes conclusive evidence that such a crystal has been produced. Doctor Davey, by means of special apparatus, was able to prove that he had one crystal. In the usual examination a small tube of finely powdered crystalline material is placed in the path of a narrow beam of X-rays of a specified wave-length. The substance turns the X-rays in different directions, according to the arrangement of the atoms in the minute crystals. A series of lines is produced on a photographic film, and these lines are used in calculations which reveal how the atoms are arranged and how far apart they are. Copper crystallizes in the face-centered cubic system, i.e., the atoms are arranged at the corners of an imaginary cube, with another atom in the center of each face. In studying the single crystal Doctor Davey revised the method of examination so that the large crystal was used, rather than crystalline powder. The specimen was swung slowly back and forth through an angle of thirty degrees, with the edge in the path of the X-rays. The rotation of the single crystal

produced the same effect as using a stationary powdered sample, and a pattern was received on a stationary film. At the same time a moving film was used, mounted on the turntable with the crystal. If the specimen had not been a single crystal, no lines would have been obtained on the movable film, since the X-rays would have affected the entire film uniformly. The lines were obtained, however, and calculations based on a comparison of the two negatives showed that the axis of the crystal was parallel with the direction of cooling the ingot.

It is difficult to foretell the results which will follow a study of large metal crystals. It has been thought for several years that such specimens would have unexpected properties, and now the prophecy is substantiated."Journal of the Franklin Institute," Jan., 1925.

X-RAY EXAMINATION OF METALS AT THE WATERTOWN ARSENAL, WATERTOWN, MASS.

By T. C. DICKSON,* WATERTOWN, MASS.

Sound metal in many articles of ordnance is essential to the safety of personnel, particularly in those articles designed with a factor of safety very little in excess of unity. The service utility of many articles of ordnance depends upon their lightness, and a definite knowledge of the character, location, and size of defects in many components materially assists the ordnance engineer in meeting service requirements.

With the hope of preventing the acceptance into the military service of unsound metal components and of assisting the ordnance designer to meet service requirements in regard to lightness, a 280,000-volt X-ray equipment was put into operation in the laboratories at the Watertown Arsenal in September, 1922.

To insure protection to persons, the room in which the equipment was installed was lined with lead 4 inch thick and all joints and securing screws were similarly covered. A lead-lined periscope permits the operator to view with safety the Coolidge tube when making an exposure.

The first few months of operation were devoted to learning the technique, the practical limits of the equipment, and time of exposure required for different thicknesses of metal, and to determining the character, location, and sizes of defects revealed by the films.

When about 200,000 volts are impressed on the Coolidge tube an exposure of about one minute is required for 1-inch thickness of steel, about five minutes for 2-inch, and about thirty minutes for 3-inch. Efforts to obtain clear films up to 250,000 volts impressed on the Coolidge tube with a thickness of steel much in excess of 3 inches have not been successful.

The experience so far had with the equipment has demonstrated that it has great practical value in showing the methods by which sound steel castings can be produced, and as an instrument of inspection.

It is the practice at the Watertown Arsenal to make one casting from each pattern, which is X-rayed. As the films, when properly interpreted, reveal the character, size, and location of all defects, it has generally been found possible by changes in method of molding, location of risers, etc., to obtain castings free from defects or to locate them where they do no harm. If the method of molding is changed, one or more castings made in accordance with the change are X-rayed to verify the accomplishment of the object desired, and a certain percentage of those made are also filmed for verification.

*Colonel, Ordnance Dept., U. S. A.

Contributed by the National Defense Division for presentation at the Annual Meeting, New York, December 1 to 4, 1924, of "The American Society of Mechanical Engineers." All papers are subject to revision.

« PreviousContinue »