Powerful locomotives, no matter how costly, do not matriculate from the builders' shops into the class that draws the sixteen-hour trains to Chicago, or pulls the Lake Shore flyers, without first having demonstrated their capacity by a series of most exacting tests.

Railroad officials must have some better proof of an engine's capacity than the mere indorsement of the men who make the mechanism. So, when locomotives come newly painted and polished from the big Baldwin shops, for instance, the owners send them into testing plants, mount them upon a delicate and compact series of registering instruments, and run them at all speeds, until their weaknesses have been detected and their strong points emphasized almost to a fractional degree of an atom. The testing machinery is, in reality, nothing more than a treadmill, in principle. In outward appearance it seems to be just so many large wheels revolving on axles, so arranged that each wheel of the engine under test meets a corresponding wheel in the tester.

Once in position, an engineer climbs into the locomotive cab, opens the throttle until it has reached its widest point; the steam shoots into her tubes and chests, the great wheels begin to revolve, gain speed and finally become a circling blur, in which the eye is unable to detect the interstices. Deriving the full power from its fuel, the forward or backward movement of the locomotive is nevertheless barely a fraction of an inch.

This is an unromantic testing beside that which Kipling described with characteristic vigor. The "tryout" which he depicted consisted of taking an engine, hitching it on to heavy freight cars and sending it out on the line, on levels and tangents, on curves and grades, until the machinery demonstrated its worthiness to take the speedy runs of its owners. Railroad men to-day are more exacting. Figuratively, their testing plants ask questions of a mass of wonderfully constructed iron and steel, and the metal answers them in their entirety.

The chief plant of this kind is located in Altoona, Pa., being a part of the extensive shop system of the Pennsylvania Railroad. With a force of sixteen men, it has been in constant operation since November 19, 1906, and, on an average, about three complete tests are made each week.

A separate building of steel and brick has been erected for housing the apparatus. The driving wheels of a locomotive under test rest upon supporting wheels with rims shaped to correspond with the head of a rail. The axles of these supporting wheels carry absorption brakes. The turning of the driving wheels causes the supporting wheels to revolve, but these are retarded to any extent desired. The work actually done by the locomotive' consists in overcoming the friction resistance of the supporting wheels and brakes, the resulting force exerted at the drawbar being measured by a traction dynamometer. The axles of the supporting wheels run in heavy pedestals secured to cast-iron bed-plates resting upon a concrete foundation. There are two bed-plates running parallel to the track, and in order that the supporting wheels may be directly beneath the locomotive drivers, these bed-plates are provided with T-slots, so that the pedestals may be moved along parallel to the track, and secured in any position to suit the particular engine under test. The only wheels of the locomotive which move during a test are the drivers. The wheels of the leading truck rest upon rails secured to I-beams and supported upon the same bed-plates that carry the pedestals. The wheels of the trailing truck rest upon supporting wheels—which remain stationary during the test—and are carried by pedestals secured to longitudinal bed-plates.

Preparation of the testing plant to receive a locomotive consists of bolting the pedestals to the bed-plates, so spacing them that there will be a pair of supporting wheels directly beneath each pair of drivers of the locomotive. A section of special rail is bolted to the inside faces of the supporting wheels. This rail is composed of a heavy I-beam, to the top of which is secured a grooved head in which the flanges of the drivers run. The top of the supporting wheels are in line with the track entering the testing plant building, so that a locomotive can be backed in and the drivers will run on their flanges until in position directly over their supporting wheels. After a locomotive has been secured in place and its drawbar attached to a dynamometer, these grooved rails upon which it ran in are removed, leaving the drivers resting upon the supporting wheels.

The axle for each pair of supporting wheels carries upon each of its overhung ends an Alden absorption brake. Each of these brakes consists of two smooth circular cast-iron disks, keyed to the supporting-wheel axle. On each side of each one of these disks is a thin copper diaphragm secured at its periphery, and also at its inner edge to a housing which does not revolve and has its bearings upon the hubs of circular revolving disks. The stationary housing is so designed that when it is filled with water under pressure the copper disks are forced against the revolving disks, creating friction. Provision is made for securing continuous and uniform lubrication of the surfaces of these revolving disks, and the water is caused to flow through the housing In order to, carry away the heat generated. Thus the water performs two functions: it supplies pressure to cause the friction, and it carries away the heat generated by the friction.

Connection between each brake and the source of water supply is made by a flexible hose. Discharge pipes for all the brakes empty into an iron trough, and each pipe is provided with a valve located adjacent to the valve in the supply pipe for the same brake. When placing a load upon the locomotive under test, these valves are adjusted until the individual brakes each absorb their share of the work. When this preliminary adjustment has been made, the power absorbed by all of the brakes may be increased or decreased by operating a large valve in the supply main.

A special system has been installed for the purpose of supplying water under uniform pressure for use in the brakes.

An adjustable drawbar is used to connect the locomotive with a dynamometer and, in addition, the dynamometer housing is provided with a means for raising and lowering the dynamometer proper to bring this drawbar truly horizontal. Two safety bars are provided between the locomotive and the dynamometer frame, to decrease the vibration transmitted to the dynamometer through the drawbar. At their ends these bars have universal joints to insure perfect freedom of adjustment, and each bar is provided with an oil dashpot near the dynamometer end.

The Pennsylvania Railroad's traction dynamometer, which measures the drawbar pull of the locomotive, is of the lever type. The weighing mechanism is supported by a frame, which slides up and down in ways formed by the housings. These housings are very massive, rigidly secured together, and anchored to a heavy foundation. The lever system is constructed upon the Emery principle, in which flexible steel fulcrum plates take the place of knife edges used in ordinary scales. As the levers are vertical instead of horizontal, their weight would not come upon the flexible fulcrum plates in the direction in which they transmit pressure. In certain cases it has therefore been necessary to supply two fulcrum plates with their axles at right angles, one for carrying the weight of the levers, and the other for transmitting the thrust.

The mechanical and mathematical detail entering into this phase of locomotive testing is so delicate and complicated that it would be, in an article of this kind, almost wholly unintelligible to the lay machinist, though of course easily understood by trained engineering minds.

Of very great interest, however, are the records obtained on a recording table, over which an endless strip of paper eighteen inches wide is mechanically drawn, and upon which a continuous story of the test and its results is told. The paper is driven by direct connection with one of the supporting wheels of the, testing mechanism, upon which the locomotive drivers rest. The speed reduction is so arranged that when the locomotive under test travels one mile on the supporting wheels, the paper moves 52.8 inches, giving a scale of 100 feet to the inch upon the diagram. In order to obtain an accurate movement of the paper, it passes between a finely corrugated brass roller and another roller covered with rubber. The winding drum to which the paper is finally delivered is arranged to slip upon its shaft, in order to accommodate its constantly increasing diameter as the test progresses.

A datum pen marks a continuous straight line upon this paper. A traction recording pen moves across the paper perpendicular to the datum line, being dependent upon the force transmitted by the drawbar from the locomotive. The maximum travel of this pen away from the datum line is eight inches. Two sets of springs are provided. With the heaviest set the eight-inch movement of the traction pen corresponds to a load of 80,000 pounds upon the drawbar, which represents the maximum capacity of the dynamometer. With the other set of springs the eight-inch motion of the traction pen corresponds to a pull of 40,000 pounds upon the drawbar, and with all the flat springs removed the eight-inch motion corresponds to a 16,000-pound load. The total motion of the drawbar to give the eight-inch movement to the recording mechanism is about 0.04 of an inch. The multiplication of the recording and weighing mechanism is therefore 200 to 1.

An integrator is provided and attached to the traction recording mechanism, so that the foot-pounds of work performed by the locomotive is automatically summed up. Five additional electrically-operated pens are provided. They normally draw continuous straight lines. One of them is electrically connected to a clock, so that each second is indicated by a jog in the straight line which the pen normally draws. Another pen is electrically connected to a roller, which is, rotated by the recording paper, causing the pen to make a jog in the line for every thousand feet which the locomotive travels. Another pen is electrically connected to the integrator, and makes a jog in its line every time the integrator measures one square inch. The remaining electrically-operated pens are used for recording such features of the test as taking indicator cards.

For handling coal used by the locomotives under test, a very complete plant has been installed. Bottom-dumping railroad coal cars are run in on a track beside the test building. They are dumped into a large hopper, and from this the coal is carried by a bucket conveyer to two elevated reinforced concrete pockets, each of which has a capacity of about fifty tons. Each coal pocket is provided with a hopper cutoff gate at a convenient height above the main floor of the test building. Coal from the bins, as needed, is discharged through the gates into wagons holding about 1,000 pounds each, which are run over weighing scales, pushed out to the locomotive, raised by hydraulic elevator to the firing platform, and then dumped.

Ashes from the locomotive are discharged at the pit level, placed in wagons, and removed.

A supply tank located in the corner of the laboratory supplies the water used in the locomotive boiler. This water first passes through a meter, the reading of which is used as a check upon the weighing tanks. A small motor-driven centrifugal pump returns to the supply tank the overflow from the injectors used on the locomotive.

So unique and complete is this big testing plant of the Pennsylvania Railroad, that rarely is there a week that passes when engines of other railroads are not tested because the owners of the locomotives lack the facilities in their shops to determine the road value and capacity of their own transportation haulers.

For the completeness of this plant and the highly-maintained state of perfection the Pennsylvania officials attribute much credit to Mr. Theodore N. Ely, Chief of Motive Power of the Lines East.