Improving Performance and Efficiency

On the subject of improving locomotive performance and efficiency, Wardale offers some erudite observations on page 144 of his book, drawing attention to the fact that the four categories listed in the table of modification options, all impinge on locomotive performance and efficiency in the broadest sense of the terms:

Before considering the modifications in detail the underlying philosophy upon which such an integrated rebuilding scheme was based must be explained. Improving the performance and efficiency of steam locomotives was essentially a question of minimizing avoidable losses, called ‘loss control’ in modern parlance. This applied both in the thermal and mechanical sense. Taking the latter, maintenance and less-­than-perfect reliability were both due to losses – wear, the consequences of which consumed much of the maintenance effort, is by definition a loss of material, and any kind of failure is a loss of the ability of the component concerned to function correctly, which might be the loss of the clamping force of a bolt which stretches or fractures, or the loss of thermal conductivity in a heat transfer surface which becomes scaled, etc. In the broad sense mechanical deterioration meant a loss of the ability of a locomotive to perform as designed and a locomotive which was loss-free in this sense would have been perfectly reliable. Although such mechanical losses could not all be reduced to zero in the real world – for example traction required unlubricated creep between wheel tyres and rail heads which had to lead to wear – there was nothing in the laws of nature which said they could not be made much less than typically found in steam locomotives (for example the wear rates of steam locomotive cylinder liners, pistons, rings, etc.. could be some thirty times as much as for the corresponding diesel engine components).

Considering thermal losses, the performance limit of any heat engine is dictated by the laws of thermodynamics and the related heat loss is unavoidable, its magnitude at any given work output depending on the upper and lower boundaries of the thermodynamic cycle on which the machine operates. The discrepancy between this theoretically attainable efficiency and that actually achieved represents losses which are not imposed by the laws of thermodynamics and which are therefore potentially avoidable (or at least largely so) or, put another way, it is a measure of the extent to which the performance of the actual machine falls short of that which can ideally be achieved. In the case of steam locomotives avoidable thermal losses were much too high, and at worst could be simply appalling.

Taking the 25NC class, the Rankine cycle efficiency for the thermodynamic limits between which the engine operated, i.e. the potential efficiency allowed by the laws of thermodynamics, was 17%, yet it can be deduced from dynamometer car tests that at 90 km/h and maximum power the thermal efficiency at which useful work was produced at the drawbar was only 3.3%, or one fifth of this theoretical maximum. The carpet of coal which lined railway tracks wherever coal burning steam locomotives worked hard was a silent testimony to this unfortunate fact. Therefore a scheme to improve the performance of existing locomotives had to attack avoidable losses – especially where they were highest.

It was clear that there was great scope for reducing the thermal losses from the 25NC class: moreover cost data showed that in normal service between Kimberley and De Aar the fuel cost of these locomotives was about three times their maintenance cost, hence the priority given to attacking thermal losses. As input minus losses equals output such loss control would achieve both improved thermal efficiency and power. This philosophy was exactly what had guided Chapelon and Porta but not, unfortunately, many other locomotive engineers, and much of the rebuilding work could be classified as simply correcting the mistakes in the original design, such as poor internal streamlining. In the thermodynamic sense it was bound to succeed if properly applied, and was the reason why apparently astounding improvements in performance could be made without altering the overall size of locomotives nor many of their important parameters, such as boiler pressure, grate area, evaporative heating surface area, or cylinder dimensions.