From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Remedies for sliding joint issues and an analysis of cracked exhaust manifolds across various models.
| Identifier | ExFiles\Box 47\3\ Scan285 | |
| Date | 1st November 1931 | |
| He/Lid.{A. J. Lidsey}1/AMA.11.31 contd. -5- In accordance with Hd.{Mr Hayward/Mr Huddy}199, this is a definite cure and all cars turned out from 161-GY onwards are immune. There are 700 P.II cars made prior to this, however, from which we must expect complaints. The remaining complaints are varied and scarce and warrant no special treatment. Copper and asbestos steel lined joint washers are now being used on P.II cars and will help complaint (c). ---------- Exhaust Manifolds Cracked or Broken. 42 of the total exhaust complaints during the September quarter are due to cracked or broken exhaust manifolds. (1) 20 HP. cars - total 19. (2) 20/25 HP. cars. - total 16. The manifolds on the above cars are the same type (prior to the present standard F.O.W. type manifold). The fracture is always at the same place, on the centre port. (3) P.I. - total 5. The rear manifold is usually the one that fails and it is generally found that on cars which have failed, the butterfly valve has seized in a partly closed position (the controls having been strained). (4) P.II - total 2. Manifolds fail at the cylinder port fixing stud bosses in every case. Modifications for the above. (1) and (2) 20 HP. and 20/25 HP. Cars. We have no proved modification, two alternative designs are available but have not been proved out. Owing to there being no works cars available for testing these out, it appears advisable to fit them to cars on which the standard type have failed in preference to fitting the standard job which we know is liable to fail again. There are no customers complaints from the latest F.O.W. feathere type of manifold in spite of the fact that they fail in France on 10,000 mile test. We have an alternative design for this type which will | ||