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).
Investigation report on the cause of brush noise on a 72 coil dynamo.
| Identifier | ExFiles\Box 166\1\ img126 | |
| Date | 6th April 1939 | |
| 60934- 6142 To By.{R.W. Bailey - Chief Engineer} from Rm{William Robotham - Chief Engineer}/RC.{R. Childs} c. to By/RD. c. to Mx.{John H Maddocks - Chief Proving Officer} Rm{William Robotham - Chief Engineer}/RC.{R. Childs}6/JH.6.4.39. Brush Noise on 72 Coil Dynamo. As previously reported, in connection with brush noise on the 72 coil dynamo, the facts indicated that the noise had only recently arisen, and that when it occurred, it was under certain conditions of heat and commutator surface. This led us to suspect the commutator and the presence of varnish on the surface. During experiments carried out on 27.G.VI. we managed to obtain the car at a moment when the brushes were noisy. The conditions of heat and polished brush surface were confirmed and the brushes then examined. It was found that while still hot, the brush surfaces, although highly polished had a peculiar resinous dragging action when gently wiped with a finger. This disappeared when the brushes got cold. It was concluded from this that varnish was in some way getting on to the commutator. Following this, the dynamo was removed from the car and the commutator checked, when it was found to be quite true. For the purpose of the test however it was given a new surface by diamond skimming. After refitting the machine the car ran for three days without noise. On the fourth day the car was driven continuously for a very long distance and at the end of the run the brushes were a little noisy. From this it was concluded that varnish was working up through the mica insulation between the copper segments. In the meantime some of the material used by Production in the construction of these commutators was examined. The inter-segment insulating material was found to be a very inferior looking grade of mica, coated with thin paper and not the robust mica that was expected. Samples were taken and held in a gas flame with the following results - | ||