Steam turbine locos were an interesting, but failed concept. They only work efficiently when operating at maximum power output. At lower output, they can be thermally very inefficient in terms of fuel / waterconsumption. And for what proportion of a typical journey was maximum output needed ?
That's more of a problem with gas turbines, where the compressor sucks about twice as much power as you get as maximum shaft output, and continues to suck about the same amount even when the shaft output is zero, so "idle" is still about two-thirds of full throttle. Steam turbines don't have that problem, they just have their own version of the usual engine characteristic of being most efficient in a particular speed range, so you design the turbine and its gearing to get the best overlap between the turbine's best operating range and the locomotive's expected most significant operating conditions. (So you might even have the maximum output being developed when the turbine has passed through its most efficient range and come out the other side, if you expected the loco to mostly be running at some lower output and only needing its full capability occasionally - as was also the case with reciprocating engines.)
Express passenger services in particular tend to involve alternation between full pelt and idle, with little time spent operating in any other region. This was one of the problems the Western gas turbines ran into - they did fine at full pelt, but nobody seemed to have realised quite how much of the time they would need to be at idle, so because gas turbines "can't idle" (as described above), a tremendous waste of fuel was the result. (Later it turned out to be a problem for diesels too, but of a different nature, being a matter of the thermal cycling involved.) They had anticipated being able to run on bunker oil to counteract the inefficiency with the cheapness of the fuel, but the problems with the incombustible solid content rogering everything downstream of the burner put the kybosh on that.
Steam turbines don't have a problem idling any more than reciprocating engines do, so by choosing to convert an express passenger locomotive the LMS were experimenting with an instance where the most significant operating conditions for the loco and the most efficient operating range for the turbine could be well matched, and it would rarely be operating outside that optimum range. And the indications from 6202 were that the steam turbine
did have the potential to reduce both fuel consumption and locomotive maintenance if it was used appropriately (and if all fast workings went over to turbines, the absence of hammer blow would make a big difference to track maintenance too).
But it's a particularly infuriating instance of the recurrent theme of the railways having some new idea which looked like a good one and was just beginning to demonstrate that when The Bloody War Got In The Way and crapped all over everything. There was nothing wrong with the idea, it seemed to do what was expected of it and didn't have any counteracting disadvantages apart from the usual difficulties with running just one of something that was completely different from everything else, which would not have applied if there had been a whole fleet of turbines. It just never got a fair crack of the whip because Something Happened to make those disadvantages the only thing anyone cared about.
Water tube boilers were tried on steam locos, but there were problems keeping them water-tight when subject to the typical levels of vibrations and stresses encountered by locos moving on rails. The problems were much easier when using water tube boilers for ships.
Yes, but the difficulties were made much worse because the experiments were along the lines of "unbolt the normal boiler off a loco and bolt a water-tube one on instead", even if they did do it by building a prototype from scratch. A normal locomotive-type boiler is so strong and rigid that in a traction engine it can actually
be the chassis, and in a locomotive it is highly forgiving of what you put it on and had allowed the crude, inherently wobbly and basically crap tradition of chassis design to be perpetuated with nobody really having any motivation to come up with anything better. A water-tube boiler is itself a wobbly thing that relies on a decently rigid support to prevent it wobbling, so when instead you put it on four sides of a shoebox with the bottom cut out and then twist that repeatedly from side to side with asymmetrical piston thrust forces while requiring maximum output from the boiler, it's not too surprising that it starts to get a bit weepy.
It's a more promising idea in connection with a turbine, which as well as giving you a steady and symmetrical thrust, gets rid of a great deal of the constraints on chassis design arising from considerations of where do you put all this machinery and still have it at least sort of vaguely possible to get at it, etc. You can concentrate on providing a rigid platform to bolt all the other bits to without having to worry so much about not being able to put a member where you want it because something else needs that particular bit of space and suchlike. There should be a decent amount of scope for saving weight too, as an additional advantage.
The "it's got springs hasn't it, what the heck is a second-order differential equation anyway?" approach to suspension "design" didn't help matters either. That persisted basically until the problems with increasing freight speeds causing wagons to hunt themselves off the rails forced BR into taking a scientific look at what was really going on down there, and their results have been fundamental to suspension design ever since. But it was still before computers got into everything, and the mathematics involved is all well-established standard stuff, so it could have been done long before if anyone had bothered; after all, the problem itself had arisen long before in different guises.
Someone or other (
possibly CJ Allen) has written an account of Crewe mechanical engineering department trying to tackle a problem with pony truck oscillation that made some class of locomotive unusable at speed. They did a lot of trial-and-error experimenting with different spring rates and fiddling with the pivot and what have you, but without any apparent consideration of the mathematics of the dynamics of the system, so they got a lot of results which were close to doing more to confuse them than to help them predict what might work better, which is much what you'd expect from that approach to that kind of problem. When they did eventually hit upon an effective modification it was more or less through going "we've tried everything sensible, so let's try something silly" and being amazed to find it did the trick.
Then they had the problem that the works department said "stuff that, we're not doing that, we're going to keep doing it like this because we always have done therefore your idea is wrong", which took them longer to deal with than the mechanical problem had. Which is kind of the same sort of attitude that the idea of trying to deal with the mechanical problem non-mathematically came from, just at a different level. The mathematical theory required to understand the pony truck oscillation was already very old stuff, so the knowledge to enable them to tackle it scientifically existed, they just didn't use it.
A later example was the Warship bogie design which went unstable when pushed to higher speeds than the Germans had been doing with it. They tackled that a bit more competently, but they still had to deal with silly stuff deriving from the Warships being a small class that had become a magnet for silly squabbling before they even built one.
In the context of the "this is what happens instead of diesels, back then" situation I'm trying to think in terms of, I reckon a water-tube turbine locomotive, with a decently rigid chassis and decently effective suspension, could have been a goer. I've been pretty much assuming the chassis and suspension improvements anyway for the other stuff I came up with to work reliably, and the basic requirements are the same however you take it: to recognise that it is a new kind of design and design it thoroughly, with parts made for their purpose rather than pinched or kludged up from something you've got already; and to have people working on and with it who place value on having it work, rather than taking every instance of development not going completely smoothly as a reason to tell everyone it won't.
