I refer to your article “Elusive Talbot. In 1964 the l Mech E arranged a symposium on ‘The Design of Small Engines’. In the discussion following the reading of the papers, Roesch took the opportunity, as he often did on these occasions, of relating his pioneering work at Talbot. His contribution, much of which was not strictly relevant to the subject, included a description of his modified 10/23 engine together with the section drawing which you reproduced in your article. Roesch stated that a single 35mm carburettor was fitted but made no mention of the mysterious blower/fan fitted to the nose of the crankshaft. As you imply, a centrifugal device running at crankshaft speed could not possibly produce any useful boost, even if the rotor and casing were correctly shaped which they were clearly not. One can only conclude that it had something to do with the cooling. It is faintly reminiscent of the flywheel fan on the 14/45 Talbot. Roesch gave the maximum speed attained, presumably running light, as 7650rpm. It is surprising that Blight says the engine had a two-bearing crank, when the drawing clearly shows a centre bearing.
I have never really believed the claim of 70bhp for the 1923 200-mile race Alvis, for two reasons. Firstly, because the 1924 engine, which had larger ports and a c r raised from 6.2 to 6.6;1, was said to develop “50 to 60bh” – vide “The History of Brooklands” p136. Secondly, because 70bhp at 4400rpm represents a bmep of 138Ib/sq in, an impossibly high figure for a single carburettor engine with a c r of 6.2;1, bearing in mind also that this is the bmep at peak power and not the peak bmep. For comparison, the bmep of the Roesch engine works out at 1131b/sq in. One might add that the speed of Harvey’s car compared with that of Joyce’s AC did not reflect a power advantage of 15bhp.
Kings Lynn, Norfolk.