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).
Necessary primary effects and permissible errors for producing silent gearing.
Identifier | ExFiles\Box 136\5\ scan0313 | |
Date | 3rd September 1937 | |
By.{R.W. Bailey - Chief Engineer} To. WB{Mr Brazier/Mr Bell}/Mr.Ashby. c. Hs.{Lord Ernest Hives - Chair} c. Hm.{Capt. W. Hallam - Head Repairs} c. BY/CS. c. BY/TWD.{T. W. D???Aeith} FTH 1152 RHC{R. H. Coverley - Production Engineer}/McW.1/FM.3.9.37. RE: SILENT GEARING In our efforts to produce silent gearing it may be noted the primary effects which were necessary to achieve this end, and obtain a standard which resulted in producing gears to run quietly under load and testing conditions. It is stated in memo' BY.6/G.31.8.37. under item (a) that the point of major importance is accuracy with smooth finish, but reference to limits in any detail of gear measurement is not made. Experience in this direction, obtained by tests upon Phantom III Constant mesh gears, has suggested that figures relative to this can be laid down, and maintained with machines of the Maag generating type. Whilst we agree that minimum error is the object, it is considered that permissible limits of error and their effects should be recorded. (a) It is of major importance, that as a comparative reading, the spacing error between successive profiles any teeth of a gear of the Constant mesh type, shall not exceed .0002, with the stipulation that this error shall accumulate and diminish in even progression, and not exist in successive teeth. Defects in this direction result in the uneven vibration, amplitude and period dependant upon the degree and spacing of such errors. (b) The eccentricity of any pair of gears when running in close mesh, should not produce changes in centre distance exceeding .002 providing that at working centres and a position of maximum eccentricity, the backlash shall be .0005, represented by a .001 centre distance change with a 14 1/2º tooth in close mesh. It may appear that this error is rather high, but it has been conclusively proved that with even higher degrees of eccentricity, providing that spacing errors are a minimum, gears will run quietly. The effect of large eccentricity upon the effective running results in the reproduction of an uneven low frequency vibration, which (cont'd) | ||