It is interesting to look at GE’s position on the A1A-A1A wheel arrangement as it was at the end of the 1950s, and as embodied in this article in “Diesel Railway Traction”.
This was aimed more at the export market and models, but also referred to domestic experience. One might summarize what GE said as where the B-B wheel arrangement was precluded on the basis of its relatively high axle loading, and so six axles were required, C-C was definitely a better choice than A1A-A1A. Thus it aligns with the notion that A1A-A1A running gear is a dead end.
In the pre-Universal era, GE had built export locomotives of the six-axle, four-motor kind. The first series of “shovel-nose” units for Argentina, with the Cooper Bessemer FVL-12T engine, had the A1A-A1A wheel arrangement. Essentially the same design was built for Chile, and in Australia Queensland Railways had a licence-built, A1A+A1A version of the 70-tonner.
But none of the initial range of Universal export models was offered with an A1A-A1A option. Presumably GE would have built such had any customer insisted and could not be dissuaded from its errant path. From the above article, it is evident that GE had done the basic work-up for what would have been the “U9A1A” and “U12A1A” models, which were then compared with the U9C and U12C respectively.
As it turned out, there was no call for the A1A-A1A wheel arrangement on export Universals through the 1960s and into the 1970s. Then from the mid-1970s, GE built the UM15A1A for Sudan (1974 and 1981), the UM10A1A for Hedjaz Jordan (1974), and the U18A1A for Indonesia (1976 and 1982). I understand that some of the Indonesian U18A1A fleet have been rebuilt as U18C, but this was following track and roadbed upgrades that allowed wider use of the latter type. It might be noted that PKA Indonesia had been acquiring the U18C type from 1975 onwards, so it had not made an “either or” choice between C-C and A1A-A1A, but rather had bought on a “horses for courses” basis. Since the 1970s U15 and U18 models were direct descendants of the original U12, the respective U15A1A and U18A1A variants can be considered to be direct descendants of the notional U12A1A that GE used in its article as an example of what not to buy. Of course, by the mid-1970s, GE’s 1958 article was probably forgotten by or unknown to those involved. Perusing old trade journal articles – and looking for “lines” of historical continuity where none may exist - is much more the prerogative of armchair observers than those at the coal-face making “here-and-now” decisions.
The EMD situation was quite different. During the 1950s and 1960s it had built many A1A-A1A versions of its export locomotives, including the B, G8 and G12, GL8 and G18, and G22. After the early 1970s though, its output of A1A-A1A locomotives tailed off. The irony of its significant A1A-A1A output is that apparently, during the design phase of the G8 and G12, there was high-level internal resistance to the idea of an A1A-A1A version. Rather it was thought (wishfully, I’d say) that a B-B diesel locomotive with a 37 500 lb axle loading would be suitable for use over track where steam locomotives had been restricted to 25 000 lb. Then, perhaps as a repeat of that aversion, the GL8 was issued as a B-B design, with the subsequent A1A-A1A variants (two such) looking like afterthoughts.
The two sides of the A1A-A1A issue are well illustrated by two examples. New Zealand Railways (NZR), from 1955 through 1967, installed a large fleet of A1A-A1A G12 locomotives, for use on its mountainous (2% grades) North Island main lines. During that period, it did look at using the C-C GR12, but its calculations showed that whilst the GR12 could certainly haul much larger trailing loads over the ruling grades, the point-to-point times would be increased to the extent that they were inherently unsatisfactory, and line capacity was not increased. So there would be no return on the extra capital employed by using the GR12 rather than the G12, and no justification for buying it. I suppose one might say that the ton-miles per hour factor was pretty much the same for the GR12 and G12. Successor to the G12 in the North Island was the GE U26C. This was a C-C, with 2600 hp on 215 000 lb, and so 83 lb/hp. The G12 had 1310 hp on 130 000 lb adhesive weight, 99 lb/hp. For NZR, the A1A-A1A version of the G12 made sense.
QR, with some steep grades but overall a flatter profile, also acquired its first G12s in 1955, this being Clyde-GM’s entry point to the QR business. Here one supposes that Clyde-GM wanted to offer a (slightly modified) standard model, and the G12 was seen as the best fit. QR had already started main line dieselization with 200 000 lb C-C locomotives of around 1100 to 1200 hp from GE and English Electric (EE). Apparently it expected the 1310 hp G12 to fit into the established operating patterns, and haul the same trains as the GE and EE locomotives. Unsurprisingly, the A1A-A1A G12, in this case with 134 000 lb adhesive weight, was found to be rather slippery, and QR asked Clyde-GM for a C-C variant, which appeared as the GR12 (also confusingly advertised by Clyde-GM as the G12C) in 1957, ahead of EMD’s announcement of the same model in 1958. For QR, the A1A-A1A version of the G12 was a dead-end.
So, whether the A1A-A1A wheel arrangement is a wise and appropriate choice or a dead end is very much dependent upon the situation, and it can fulfil both roles very well. That it has been inappropriately applied in the past has given it a somewhat clouded reputation, as evidenced by the simple existence of this thread. On balance though, the ES44C4, deployed for those operations where it is an ideal fit and not expected to be a “do everything” locomotive, looks to fall into the “wise and appropriate choice” category. I imagine that it is most appropriate for larger roads with the scale and operating discipline to be able to segregate locomotive pools than for smaller roads where “one size fits all” is the inevitable operating regime, whether de facto or de jure.
You do not have the required permissions to view the files attached to this post.