I have learned a lot from the correspondence on this topic, but I must confess I have not found any cause to modify my views that twin o.h.c. were commercially unacceptable in 1930, or that Jaguar were the first firm to make them so.
Your correspondents have suggested some half-dozen twin cam engines made up to 1939 which achieved a high standard of refinement, namely Alfa Romeo, Sunbeam, Lagonda, Salmson, Bugatti and Duesenberg. The firms making the first three failed, thus supporting my case; the last two, I venture to suggest, were in no sense commercial productions, being aimed at the ultimate prestige market in which price was of very little significance. To paraphrase a famous remark, if you wanted to know the price of a Duesenberg or Bugatti, then you could certainly not afford one. I do not know how Duesenberg made out for money (and in the end they too failed), but I have long suspected that Bugatti was as individual about this as about everything else and subsisted comfortably on other people’s, to their very great pleasure.
We are left with the Salmson, which at first sight appears to he the exception to the rule. I accept Martin White’s view that this twin-cam unit – not, as I now learn, first built for racing – which remained in production until 1954 was a delightfully civilised engine, but from the technical papers he has kindly loaned me it is clear why this was so: it was detuned to such an extent that it gave far less power than most pushrod units. The figure quoted for the 2.3 Salmson in 1939 was 68 b.h.p., which – unless (like the Sunbeams) it was tested on the town gas supply – is hardly comparable with the 70 b.hp. of the most “cooking” of all 2.3 Talbot engines in 1930.
The question remains of how such an inefficient and expensive engine remained in production for so long; I believe the answer is that the Salmson car, like the Bristol after it, existed only by the good grace of the parent aero-engine division, using its foundry, capital, and no doubt in slack times even its machinery and labour, with no overheads at all and no need to stand on its own feet. Lastly I would say that I agree with every word in Brian Morgan’s letter. To his list of snags in pre-war twin o.h.c. engines I would add the difficulties in disposing of the large quantities of oil required at the very roof of the engine; they tended either to escape down the outside, which was messy, or down the inside past the inlet valves, which was disastrous; the Lagonda semi-o.h.c. design was very good here. I also rather wonder whether the prophecy in his final paragraph is as well founded as it would have been years ago; from my reading of the technical press the long run of the inclined valve hemispherical head may be coming to an end, displaced by Roesch-shape combustion chambers in the piston crowns and lots of vertical valves in a fiat head surface.
One thing I would dearly like to know from Mr. Morgan: what exactly was the b.h.p. of the Model-J Duesenberg engine?
Callington. Anthony Blight.
[Except perhaps for comments about Model-J Duesenberg power output, this lengthy correspondence is now closed – ED.]. *