Laying Rails.— Owing to the difference in quality and in the amount of traffic over the rails, it is quite difficult to form a correct esti mate of the average life of the same. Good iron rails have been known to last, in service on the main track of a railroad doing a fair business, nine or 10 years, and steel rails 15 to 20 years. There is much discussion on the sub ject of broken or even joints. The majority of track has heretofore been laid with even joints; but at the present time all rails are be ing laid with broken joints, which consists in placing the rail joint on one side of the track opposite to the centre of the rail on the other side of the track. Outside of the general line i and surface the principal defects in steel rails are pipes, gag marks and cinders, which result in flaws. (See IlioN AND STEEL, METALLoc. RAPHY or). Owing to heavy engines and cars of great weight now employed, heavy rails are demanded to be used and as the height gives vertical strength, in such heavy rails used, height should be the main consideration and should not be less than six inches. To give good wearing heads, the proportion of metal in the head should be kept down to the mini mum, as owing to the present process of rolling steel rails at a great heat from a large ingot, a thick head will not give as good results as a thin head rail. The web should be thicker at the base than under the head, which in the high rail ensures greater strength, as the tendency is to break at this point, if too weak.
Wave Motion.—As all movements are on the principle of the lever, there is of necessity an undulatory motion on the passage of each train, the amount of which being dependent on the condition of the sub-grade, ballast, ties, rail and weight of the rolling stock. The less sub stantial the superstructure the greater ballast compression there will be, and, of necessity, a rough-riding track. The foundation of all ties being loosely compacted material, any move ment or lichurningo of the tie necessarily throws unequal loading on the ballast at differ ent times, causes its compression and movement, and destruction of the tie foundation. The wider the ties and the lighter the rail and heavier the loads, the greater such movement must be. Rails take a permanent set, as regards wear motion, in one of three forms: (a) Joint low and centre high; (b) joint and centre low, quarter high; (c) entire rail level. As the vibratory motion of the rail takes place some thing has to give way. If the fastening to the tie is by push bolt or lag screw, the tie will be raised with the vibration and c(pumpp the bal last, and in 'time the fastenings will become loose.
Elevation of Curves.— The elevation of curves has been a subject of much discussion. Thirty years ago one-half inch elevation per degree of curve was considered proper; but, now with faster and heavier trains a greater elevation is required. Engineers claim that one inch per degree for a speed of 60 miles i per hour is proper for curves of three degrees or less, but for each additional degree the eleva tion should not exceed three-fourths inch.
While practice teaches the amount of elevation needed in any given case it is a safer plan to follow the ensuing formula deduced from mechanics in following the theory of elevation; E , in which E represents the eleva 32.16 R tion in feet, V' the square of the velocity in feet per second, G the gauge or the distance between the points supporting the wheel, or from the centre of one rail to the centre of the other, and R the radius of the curve in feet. This rule, though scientifically correct, cannot always be applied in actual practice, because all trains do not travel at the same rate of speed. Practical provision must be made for the aver age speed of trains and the number in each class of service. For example, if a road have two trains whose speed is 50 miles per hour, four with a speed of 40, and six or eight with a speed of 25 miles per hour, one inch per degree of curve will be found suitable. The pressure by. the fast train against the outer rail of a curve will not be sufficient to spread the rail or throw the track out of line. The elevation will be about right for the 40-mile trains, and while it may be excessive for the slow trains, yet they will do no particular damage to the track and will have a tendency to prevent the track being moved permanently outward by the fast trains. The liability of derailment of locomotives or cars is much greater on a curve track and a large percentage of such accidents is chargeable to the defects in the rolling stock as well as to the defects in the track itself. Heavily loaded cars often leave the track owing to the failure of a track to adjust itself to the curve of the track, caused, perhaps, by a defective curve roller, and the greater part of the load resting upon one side of the track.
Frogs and Switches.— The turnout, by which a car may pass from one track to an other, consists of a frog, a rail leading to the frog, a corresponding opposite rail and a de vice connecting these rails with the main track known as the switch. If the switch is made to serve two turnouts, it is called a three-throw switch; a trailing switch is one where a train on the main track passes from frog to switch; while a facing switch is one that approaches in the opposite direction. The common or stub switch consists of a pair of connected rails so arranged that one end is fixed; the other can be moved so as to be a part of either the main track or turnout. This switch has two serious defects—the want of safety, and the necessary space at the end of the moving rail, which jars the rolling stock, batters the switch rails and causes discomfort to passengers. The first cost of a point switch is more than a stub, but the split, which is more economical to maintain and safer, is cheaper in the end. In laying switches the general rule is to cut the least number of rails. When putting frogs into a track care should be taken to have them in a true line and level with the track rails which are con nected to them. The gauge rail, opposite the frog, should be put to a perfect gauge for the full length of the frog.