(National Maritime Museum 59-209)
The first practical water-tube boiler was the Belleville, invented in France in the 1850s and first adopted by the French Navy in the 1880s. Its water-tubes formed a series of flattened spirals built up of straight tubes with cast-iron junction boxes connecting them. They rose from a feed box in front of the boiler to a cylindrical steam drum at the top. Most ships had economizers, which preheated the feed water and controlled steam output when it had to be changed suddenly, for example to increase speed. Pressure inside the boiler was typically 350psi, reduced to 250 for the engine (the greater pressure inside the boiler was later considered a serious defect, though it came to be commonly accepted). Observation of Bellevilles on board the French mail steamer Laos prompted the Royal Navy to try it on board the torpedo gunboat Sharpshooter and then to adopt it for numerous large cruisers, such as the Powerful class. The Belleville used large water-tubes, and it was attractive because it appeared to be sturdy, and because it was already in successful service. It offered more fire grate area (for overall size) than any other boiler then known, and its small elements did not require a large opening in a ship’s armour deck. The Admiralty did not appreciate that Bellevilles, introduced at the same time as much higher steam pressure (300psi or more instead of 160), were a considerable technological leap. There were serious breakdowns in service; HMS Hermes had to come home after only a year in commission. Europa showed extravagant fuel consumption on passage from Portsmouth to Sydney: of eighty-eight days she had to spend thirty coaling (partly due to leaky condensers and leaky steam joints). The big cruiser Terrible burned 200 tons a day on a 1902 voyage to China at an average of 11.8kts, but two years later she burned only half as much at an average of 12.6kts.
Early problems with the Bellevilles were critical because it was adopted so quickly for so many important ships. By 1900 there were calls for a Committee of Enquiry, one engineer calling the Belleville ‘the worst boiler in existence’. In September 1900 the Admiralty formed a Boiler Committee under Admiral Sir Compton Domville.44 All but one member (Chief Inspector of Machinery J A Smith) were associated with either the merchant fleet or with Lloyd’s. The first interim report was issued in 1901 and the final one in 1904. In 1904 Domville was flying his flag in the Belleville-boilered battleship Bulwark; he considered her boilers entirely satisfactory. Many of the problems attributed to the Bellevilles turned out to be due to other changes, including machinery packed too tightly together because with higher pressure it could be made more compact.
The interim report recommended fitting both cylindrical and water-tube boilers and abandonment of the Belleville as it seemed to have no particular advantages over other types. The committee listed thirty-six other water-tube boilers, of which it favoured four, already being fitted on a large scale in foreign navies: the Babcock & Wilcox, the French Niclausse, the German Dürr, and the Yarrow large-tube boiler. Of these the first two had already been tested satisfactorily in the Royal Navy, and were being adopted on a limited basis – two sloops (Espiegle and Odin) and a second-class cruiser (Challenger) were receiving Babcocks, and one sloop (Fantome) and a first-class cruiser (Devonshire) were receiving Niclausse boilers. The Babcock & Wilcox was already being tested on board the torpedo gunboat Sheldrake, but the type now contemplated was different. At the committee’s suggestion, the cruisers Medea and Medusa were reboilered with, respectively, Yarrow and Dürr boilers, as it was difficult to draw conclusions fully applicable to larger ships from torpedo gunboat trials. In addition, in 1897 and in 1899 the small cruisers Barham and Bellona were both reboilered with Thornycroft water-tube (small-tube) boilers (not as part of the Boiler Committee program). Similarly, in 1900 and in 1901 Blanche and Blonde were reboilered with Normand small-tube water-tube boilers.
For ships powered by piston steam engines output is expressed in indicated horsepower (IHP). IHP was measured using an indicator card which directly measured the work done by the engine. Although an engine produced a given IHP, by no means was all of that power transmitted to the shaft. The engine drove various auxiliaries and also had to overcome friction within it. Modern turbine engines are rated in shaft horsepower (SHP), which is the power actually available to drive the propeller. For ships described in this book it was typically 80 to 85 per cent of IHP. Yet another measure of power was effective horsepower (EHP), the power actually required to drive a ship at a given speed, which was estimated on the basis of model tests. The model basin also estimated the efficiency of a ship’s propellers, to estimate the IHP required to achieve the EHP it calculated. The overall ratio between EHP and IHP was about 50 per cent, but errors in estimating efficiency helped make it difficult to predict exactly how fast a given ship would be, even in smooth water. Blake and Blenheim seem to have been particular cases of failed estimates.
Tonnage and Cost
Until the early 1870s, ships at the design stage were described more by their burthen tonnage rather than by their displacement, as has been standard ever since. Burthen tonnage – sometimes called builder’s (old) measurement – was easy to calculate because it was based entirely on a ship’s dimensions. It was given as (L-2/3B) × B × B/2 divided by 94; one indication that it was being used in design reports was that tonnage was given as tons plus some number divided by 94. This tonnage seems to have been used initially because designers could not precisely predict the displacement of ships, perhaps partly because weights (particularly those of machinery) were difficult to estimate (but by about 1860 displacements were generally calculated, and accounts of earlier steam warships often include their displacements as well as their burthen tonnage). Burthen seems to have been considered a good way to compare hull cost. It was probably not too bad as long as ships had similar hull forms. That fell apart for ships with unusually fine hulls, which might be long but not heavy, as in the comparison between Volage, Raleigh, and Bacchante in 1871. The following year burthen was abolished as the official measure of ship size (displacement replaced it). That estimates could still be quite faulty became obvious in the unfortunate Orlando class more than a decade later.
For much the same reason, until the mid-1870s design reports gave nominal horsepower (NHP) rather than the indicated horsepower (IHP) which determined a ship’s performance. NHP was a measure of the size of the engine, not its output. It could be specified precisely, but actual engine performance seems to have been a very different matter, again until some time in the late 1860s or early 1870s. NHP was 7 × area of piston × equivalent piston speed divided