Civil Engineering August 2021 | Vol 29 No 7

48 August 2021 Civil Engineering at the front of the tractor. This kept the tractor free to follow the bogie exactly without being jerked from its wheelway, while also allowing the driving wheels of the tractors to negotiate bumps and dips in the wheelway of up to 150 mm in relation to the rails. The roadrail tractor is shown in operation in Figures 1 and 5. The trial runs at Canada Junction showed that the tractor could maintain 29 km/h on the straights and 19 km/h on the balloon curve; that it could start and stop with full load on a 1:17 grade; and that the driving wheels of the tractor aligned themselves with accuracy, every time, in exactly the same spoor. Some observers were concerned that the tractor could not reverse; others feared that the extra width required by the driving wheels would lead to more expensive cut- tings and embankments, making roadrail less suitable for mountainous terrain. SAR&H General Manager William Hoy deemed the trial to “prove successful, the tractor riding steadily and hauling a substantial load over a very rough track”. Dutton was happy that the tests adequately proved the roadrail concept: The trial at Canada Junction was obviously merely intended to demonstrate, at the smallest expense, the fundamental principle of the system; namely, that the high tractive effort of driving wheels fitted with rubber-tyres running on a roadway could be successfully applied as the motive power on Light Railways. The Canada tractor could not reverse more than a few feet, which feature its critics did not fail to make the most of, and they expressed the opinion that a rail bogie at each end of the tractor would prevent it from taking sharp curves, and would not admit of the necessary flexibility for change of grade or for use on uneven ground. The trial runs at Canada Junction were made in 1917, a year before the end of World War I. The war left South Africa in dire economic straits which brought railway development to a standstill. Under these conditions, the official development of roadrail had to be put on hold. It took a few years of slow economic recovery before South Africa could afford to embark on a substantial new rail-building programme with Parliament’s approval of Act 30 of 1922 which authorised the construction of 18 new lines throughout South Africa, totalling 1 372 km. MODEL EXPERIMENTATION During the economic lull, Dutton invested considerable private effort to convince the sceptics of the benefits of roadrail. At a scale of 1:12, he experimented with a model railroad in his garden, shown in Figure 6, to systematically address the shortcomings and questions raised by the Canada Junction experiment of 1917: Q Q The model tractor of 15 lb (8 lb on the driving wheels) developed a drawbar pull of 0.65 times the load on the driving axle – practically the same as the 0.67 found at Canada Junction; see Figure 7. Q Q The main complaint that the tractor with a single bogie could not reverse was addressed by showing that bogies at both sides (which would allow reversing) did not pose any problems; see Figure 8. Q Q The model tractor could negotiate extreme grades going suddenly from 1:3 to 1:7 in opposite directions to Figure 6 Frank Dutton’s model roadrail in his garden (Credit: THL LS_18_014) Figure 7 Dutton’s testing rig for roadrail drawbar pull (Credit: THL LS_18_007)