prr s2 - Trains Magazine

The biggest 'teething problem' turned out to be not so much the turbine as the boiler and its front-end arrangement.  A little watch-building first:  Steam turbines need low exit pressure for best efficiency.  Conventional wisdom holds that turbines on locomotives aren't 'preferred' as they are on ships because the latter have efficient condensing at the required high mass flow.  There is also a requirement for very high volume (and good flow smoothing) in the exhaust tract from the turbine to clear the required large volume of low-pressure steam.

Meanwhile, while turbines produce very good torque at low speeds (a point the PRR motive-power department was at some pains to emphasize) there is higher steam flow through the device at low rotational speeds (in an axial-flow turbine, the potential reaction energy can't be extracted as rotation, with the effect that some of the steam just 'flows through' -- as an amusing side-note, this apparently got Parsons, the inventor of the axial-flow turbine, in trouble at school because he'd invented an engine that "obviously" had a 40% free-area leak straight through... ;-}).  As it turns out, there is also relatively higher leakage at the 'tips' of the blades, both due to required clearance there and to vortex generation.

The S2 handled the additional mass-flow requirement by having four stacks and nozzles (you can see this in overhead pictures of the locomotive.  The problem, as it turned out, was that the combination of leakage and power flow through the turbine at low speed ALSO produced sharply higher drafting on the fire, with comparatively rapid onset.  I also suspect that any propensity for the engine to slip would result (as with the poppet valve gear on the T1s) in sustained very high mass flow and, in the S2 design, monstrously amplified draft flow.

Remember that the S2 boiler was of conventional design and construction (riveted and staybolted) running at what was by that time well-known to be an 'edge-of-the-envelope' pressure for that kind of structure.  Perhaps unsurprisingly, by 1946 we begin to see evidence of staybolt failures -- some of them VERY severe and extensive.  Doing a bit of 'reading between the lines' in the surviving correspondence at the Hagley, it's pretty clear that this was the nail in the S2's coffin.

It should be remembered that this was NOT the PRR's developed design for steam-turbine power.  That was the V1 locomotive, greenlighted for production in 1944.  This used two turbines, each geared to 4 individual axles with 48" wheels, and a "modified Q2" boiler.  I have a copy of the tractive-effort chart developed for this locomotive, and it is interesting to observe just how competent the design was -- even before it was given a self-excited variable-torque transmission, the Bowes drive, which imho would have assisted dramatically in reducing the problem with low-rotor-speed "slip".

Unfortunately the V1 shared with the Q2 a colossal appetite for water -- something that F-units, for example, did not suffer from...

RME 

Welded construction had been approved by ASME (in 1943) but none of the postwar PRR engines had it (this might have helped with the waist-sheet problems on the Q2s, but I digress) --