One of the most important lessons of these past twenty-five years is that to ensure faultless running, the critical dimensions of the track and wheelsets must be clearly defined and the settings must be checked by the modeller and proven before being released for use. This is really quite simple, since no ruler measurements need be involved. Simple gauges are all that is required to ensure accuracy in the components. In order to manufacture such gauges, the standard dimensions of the wheels and the rail need to be known and the sooner the SIG settles for a particular set of standard dimensions, the sooner gauges are likely to be produced.
It is this factor of dimensioning and tolerances that is at the heart of the matter. As mentioned in the Journal, it has been possible for individual modellers to use exactly-scaled reductions of prototype dimensioning successfully, but only because they have total control over the adjustment of wheels and track on their individual layouts and work in isolation.
However, what is required for modellers in general is "exact scale" that allows faultless exchange running on the layouts of other modellers who use the same nominal dimensioning and this is an altogether different matter. In this situation, control over manufacturing tolerances is critical - all modellers must have identical track and wheels. Today, in 1996, despite "man on the moon" achievements, no wheelset manufacturer can guarantee perfect control over all dimensions. It is therefore necessary to provide a "safety net' in the standard dimensions which accommodates these manufacturing inadequacies.
It was these restrictions that caused the original Protofour dimensions to incorporate a slightly wider-than-scale flangeway and a slight compensating reduction in wheel back-to-back setting. The Proto:87 standards, set out in conjunction with 4mm scale Protofour operations, have always incorporated these adjustments. However, if suitable wheels and track are not produced, then one is left with an unreliable set of components that will almost certainly give unsatisfactory operation, no matter how carefully they are assembled.
The experiment described in [Gourley, 95] should reinforce the view that the wheels and track are, indeed, "two halves of a carefully engineered system". Their dimensioning is fundamental to the success of correctly-scaled railway modelling and those who do not take this factor into account are almost certain to encounter unreliability in the running of their stock.
To summarise the issues facing the SIG: the dimensions need to be decided, and there needs to be sufficient tolerance so that components can be manufactured economically.
