Hi,
It seems there are a number misconceptions about compound locomotives around. If high pressure steam expands, it cools which may cause condensation which increases steam consumption and in the process cools the cylinders too. In the next stroke part of the steam is needed to reheat the cylinders, which increases steam consumption again. The compound engine attempts to minimize this by expanding steam in two stages through two different (usually pairs of) cylinders, which operate at different pressure and temperature levels. This way steam consumption may be reduced by 10 to 15 %. The price for this a heavy and complicated steam engine. The beneficial effects of compounding may be eaten up by increased mechanical friction within the more complicated engine, as well as friction between steam flowing through the two sets of valves and connecting pipes. Chapelon's successes with reconstructing older French compound locomotives were mainly due to his increasing diameters of steam ducts, and thus improving steam flow.
The design of compound locomotives varied considerably. With the exception of Anatole Mallet's articulated locomotives, they had rigid frames. And with the exception of Francis Webb's compounds, high- and low pressure cylinders were invariably mechanically linked. This could be either by having all cylinders working on one axle. During the 1890s two cylinder compounds with three coupled axles were quite popular as goods engines. Four cylinder express engines could have all cylinders act on one axle. They could be either single expansion or compound engines. Many famous German Pacific compound express locomotives like the Baden class IVf or the Bavarian class S3/6 were built that way. The disadvantage was that there was considerable strain on the bearings of the single driven axle. Usually the high pressure cylinders were between the frames and the low pressure cylinders at the outside.
A more advantageous design was to have the two sets of cylinders work on different axles, which (with the notable exception of Webb's componds) were mechanically connected by connecting rods. De Glehn's compounds, which became particular popular in France were fairly unique because they had the low pressure cylinders between the frames. The reason for this was, that his compound express locomotives were derived from English style two cylinder single expansion engines, which had inside cylinders. The power of these locomotives could not longer increased by increasing cylinder diameters because of the constricted space. So he added a pair of outside cylinders, which functioned as high pressure cylinders, retaining the large diameter inside cylinders as low pressure cylinders.
Newer designs like the Baden class IV h locomotive, the best and swan song of the German Pacific compound express locomotives had two driven axles as well, but like the other German compounds the high pressure cylinders between the frames and the low pressure cylinders outside.
Quite a different approach to cope with the problem a adiabatic cooling of expanding steam was superheating. Here the steam temperature was raised to upt to 400 decrees Celsius in the superheater. At this temperature range cooling would not cause any condensation. It used heat energy still contained in the combustion fumes, which would otherwise escape through the smoke stack. The superheater also had no moving parts and required thus only little maintenance. Its disadvantage was, that it increased the boiler weight and needed heat resistant cylinder oil for lubricating the cylinders. Once the latter problem had been overcome, even better savings of steam could be attained with single expansion engines.
Of course superheating could be combined with compounding, but as both methods attempted the solve the same problem, the results were marginal, since superheating improved steam consumption of compound locomotives by a few percent only. It did, however add the disadvantages, particularly cost and weight of a compound engine with increased cost and weight of a boiler fitted with a superheater.
A much more efficient way to optimize steam locomotives was to dispense with the complicated compound engine altogether, the saved weight not only compensated the weight of the superheater, but also allowed to fit a larger boiler to the locomotive, which would produce more steam, thus allowing for a more powerful locomotive.
To my knowledge André Chapelon did investigate the design of very large six cylinder locomotive, with intermediate superheating. Since France, however, decided after the War to abandon steam traction alltogether in favour of 25 kV 50 cycle AC electric traction, these projects came to naught.
Cheers,
Konni
