370 



THE CIVIL ENGIXEER AND ARCIIITECrs JOURNAL. 



[December, 



possible the deterioratina: effect of these descriptions of strains 

 upon tlie axle, the following conditions are important: — 



That the bearings or journals of the axle fit as closely to tlie 

 brasses as is consistent with freedom, the allowance of flange 

 gauge of wheel being quite sufficient for the carriage to move 

 freely round curves and meet any irregularity in the gauge of the 

 rails. That the wagons or carriages be as equally loaded as possi- 

 ble, and the draw-chains be exactly in the centre; and as side 

 chains are dangerous, they should be completely removed, pro- 

 vision being made for a duplicate centre draw-chain should a 

 failure take i>lace. As the damage to the loading of wagons is in 

 proportion to the oscillation, they should all be screwed together 

 by means of screw-couplings, having spring buffers upon both ends 

 of every wagon. It is well known that the injury to the wagon, 

 to the load which it conveys, to the axle which carries it, and to 

 the road over which it runs, is very much aggravated if the wagons 

 are allowed to oscillate from side to side, and become like so many 

 battering-rams, injuring themselves and all substances in contact 

 with them. A ti-ain of wagons or carriages should be jointed 

 together similar to the vertebrae of an animal, by which means any 

 sudden lateral action would be neutralised by the support derived 

 from the neighbouring vehicle. The road to be kept as accurate 

 as possible to gauge and line. The tliird class of strains to which 

 axles are liable are the shocks produced by starting and stopping 

 a train, and which are in proportion to the momentum of the 

 wheel and axle at the time of collision when stopping, and to the 

 velocity of the impelling force and the inertia of the wheel and 

 axle when starting; these strains are felt principally on the neck 

 of the journal. Fourth strain, the torsion or twisting produced 

 owing to wheels travelling over curves of the line; the difference 

 in length of surface of the inner and outer rail compels the one 

 wheel to grind or slide upon the rail, while the other is free to roll. 

 This strain is proportionate to the load on the wheel, determining 

 the amount of friction upon the rails, and the length of axle between 

 the wheels; a slight amount of torsion is also produced from any 

 variation in the diameter of the wheels on the same axle, by any 

 inequality of load upon each journal, the quality of the brasses, or 

 the amount of lubrication ]iroportionately, and the strain of the 

 break-block on one side, because when any of these occur sepa- 

 rately or jointly, one-half of the extra strain on one journal is 

 transmitted through the axle to the other, and twisting or weaken- 

 ing the axle is necessarily produced. To lessen the amount of the 

 above strain, it is obvious that the wheels should be kept in the 

 best possible state of repair, so far as equal diameters and true 

 circubir surfaces are concerned, the wagons or carriages should be 

 loaded e(|ually on each side, the journals carefully lubricated, and 

 all break-blocks to bear the same pressure on both wheels of the 

 same axle. Fifth strain, the constant vibration of the whole axle. 

 This is more particularly the case, and is accelerated when the 

 axle is fixed in a rigid, unyielding wheel. Aly experience has 

 proved that the axles fixed in cast-iron wheels are very much more 

 liable to deterioration than those in wrought-iron wheels, and the 

 jar or vibration tending to deteriorate the quality of the iron, by 

 altering its texture from fibrous to crystalline, is clearly visible in 

 its effects in several fractures which I have seen. It would appear 

 that the cast-iron wheel acted more like a hammer on the axle, 

 and as in the cold-swaging process a gradual breaking-up of the 

 fibre at the back of the wheel goes on, which is shown by an annu- 

 lar ring, varying from g-incli to j-inch in breadth, the strength is 

 com]ili'tely destroyed of this outer portion, and a sudden shock of 

 the wheel ui)on some point of the road completes the fracture. 



Among other causes which contribute to the deterioration of 

 axles may be mentioned — the practice of throwing cold water on 

 the axle to cool it, when it has become nearly red iiot for want of 

 proper lubrication in the journal. With regard to the strain to 

 ivhicli tlie portion of the axle between the wheels is subjei't, there 

 can bo no d(uibt if the form of the axle is so proportioned that any 

 blow transmitted through the wheel is received equally along the 

 \vh(di' body of the axle, and the sectional strength at each point is 

 fairly balanced to resist the effect of the blow, the axle will then 

 be best suited to prevent deterioration at any particular place. 

 With the view of determining the weakest point of a common 

 wagon axle under different circumstances, 1 made a few experi- 

 ments, as follows: — 



In the first experiment the power was applied to the flange of 

 the wheel, and the resistance (as in the case of a railway axle 

 wlien ruiuiing) at the centre of the opposite wheel; the result was 

 that the axle began to bend from a straight line 12,1 inches from 

 the bo^s of that wheel to which the power was applied, and there 

 is no doubt that if the power had been continued the fracture 

 would have taken place within the 12j inches. 



As a proof of this, in the second experiment, an axle of the 

 precisely same dimensions and form, on being bent alternately 

 backwards and forwards (the power being always applied on the 

 same wheel at opposite points) was broken at the twelfth time of 

 bending, within G inches of the back of the wheel. 



In the third experiment the power and resistance were exactly 

 in a parallel line to the centre of the axle, and the result, as 

 might be expected, was a curve of a nearly uniform radius; prov- 

 ing that although the foi-m of this axle was adapted to receive the 

 blows of both wheels at precisely the same instant, and to the 

 same extent (an impossible circumstance in practice), it was not 

 suited to receive alternate strains or shocks, to which all axles are 

 subject in ordinary use. The sizes of the axles in Xhe above 

 three experiments were precisely alike. 



In the fourth experiment another axle of the same dimensions 

 was taken, and reduced at the centre in a lathe to the following 

 dimensions: — The axle was divided into eight equal spaces from 

 the back of the wheel to the centre of the axle. Immediately at 

 the back of the wheel the axle was 4 inches diameter, and the de- 

 flection was 9'i inches; at the first space the diameter was 3| 

 inches, and the deflection 8j| inches; at the second space the dia- 

 meter Sfif inches, and deflection 7 inches; at the third space 

 the diameter Sy's inches, .ind deflection 5| inches; at the fourth 

 space the diameter 2}^ inches, and deflection 4-\ inches. Up 

 to this point the axle maintained a straight form from the back of 

 the wheel; and from this point to the centre of the axle, as shown 

 by the deflections, it assumed a fair curve, proving that the axle 

 was weaker towards the centre than it ought to have been, and 

 that the first 12 or 14 inches from the wheel having maintained 

 the straight form was stronger in proportion. 



In the fifth experiment the axle was reduced to 2i inches in the 

 centre, and with power applied similar, as in the last case, the 

 weakness at the centre was more perceptible. 



In the sixth experiment the axle was made of another form, 

 weaker immediately at the back of the wheel and at the centre. 

 We had here two bends or curves, with a straight jjortion between 

 them. 



In the seventh there was an improvement upon the sixth, but it 

 did not realise a perfect balance of strength at the different points. 



In the eighth experiment, this was fairly accomplished, the pro- 

 portion being as follows: — ^From the back of the wheel to the 

 centre of the axle, the sizes were 4y^in. diameter, 3^ in. dia- 

 meter, 3 in. diameter, 2| in. diameter, 2^-^ in. diameter, 2| in. 

 diameter, 2f^ in. diameter, 2i-^ in. diameter, 2'j| in. diameter; 

 the half-length of the axle being divided as before, into eight 

 equal spaces. 



It must be evident that this can only be an appi'oximate result, 

 but we found that these proportions enabled us to attain the 

 nearest approach to a regular curve in bending the axle; and it is 

 worthy of notice, that when the dimensions of the axle at the 

 journal and in the boss of the wheel are determined, a calculation 

 to ascertain the exact proportion between the wlieels seems to 

 confirm the above statement of dimension in the eighth experi- 

 ment. The greatest strain to which this portion of the axle is 

 subject being received at the bottom flange of the wheel, and 

 transmitted through its radius, the amount of strain which any 

 portion of the axle has to resist is inversely as its angular distance 

 from the point of impact is to the radius of wheel. Assuming the 

 blow on the flange of the wheel to exert a breaking force equal to 

 102,229 lb., and the diameter of the axle to be 4-71 inches to resist 

 this blow, then, dividing the axle into four equal spaces to the 

 centre, the proportionate breaking force at each point would be 

 as follows: — At the first, 91-,381 lb., relative diameter, 4,59 inches; 

 at second, 80,697 lb., relative diameter, 4-35 inches; at thii-d, 

 67,798 lb., relative diameter, 4-11 inches; at fourth, 58-829 lb., 

 relative diameter, 3'92 inches. ^Vith regard to engine axles, these 

 proporti(uis will apply where no circumstances exist of emj)loying 

 the centre of the axle for transmission of power. The crank axles 

 of locomotive engines cannot be treated by any of the rules ap- 

 ])licable to straight axles; and our experience would seem to prove 

 that, even with the greatest care in manufacturing, these axles are 

 subject to a rapid deterioration, owing to the vibration and jar 

 which operates with increased severity, on account of their pecu- 

 liar form. So certain and regular is the fracture, at the corner 

 of the crank from this cause, that we can almost predict in some 

 classes of engines the numlier of miles that can be run before 

 signs of fracture are visible: a certain amount of injury can be 

 prevented by iiutting ciuinterbalance weights opposite to each 

 crank, which lessens the vibration very considerably. It is right 

 to observe in this place, that to some extent the injury to all axles 

 may be increased, if the wheels in which they are fixed are not 



