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).
Findings from aerodynamic research, focusing on wind tunnel testing and calculated aerodynamic forms.
| Identifier | ExFiles\Box 161\1\ scan0193 | |
| Date | 22th September 1938 | |
| To Rm.{William Robotham - Chief Engineer}. from Da{Bernard Day - Chassis Design}/Ev.{Ivan Evernden - coachwork} c.c. Ev.{Ivan Evernden - coachwork} c.c. Da{Bernard Day - Chassis Design} c.c. Da{Bernard Day - Chassis Design}/Db.{Donald Bastow - Suspensions} c.c. Rm{William Robotham - Chief Engineer}/IMW.{Ivan M. Waller - Head of Chateauroux} 1380 GBH Da{Bernard Day - Chassis Design}/Ev.{Ivan Evernden - coachwork}8/N.22.9.38. Re: Aerodynamic Researches. We would like to put on record what we have learnt from our investigation of the problem of aerodynamic resistance, as far as we have gone. (1) Wind Tunnel. The failure of our wind tunnel tests on the Paulin model to give us a reasonably accurate actual K.{Mr Kilner} for the car, basing our results on previous tunnel and road data for the Streamline Bentley B.23.AE. is due to the widely different air-flow over the two models. The former is laminar, and the latter turbulent. This would not matter if we were using full size models. The Reynold's number taken as a function of scale and velocity assumes a similar flow for the two models. As the scale velocity factors for the two cars are respectively 20.5 and 16, the scale effect will be great and not the same in each case. The method of running tests on the model at 55 ft. sec. and 70 ft: to test for scale effect does not shew up the defect referred to because the Reynold's number is only increased by 27.3% whilst the increase between the model and the car is at least 1500% Messrs.Vickers and M.Andreau have said the same, in fact the latter has given up using a wind tunnel. At the moment we appear to be in the position where we can use a tunnel only to evaluate small changes in the model which will not change the entire nature of the flow. (2) Calculated Aerodynamic Form. M.Andreau and others in Europe appear to use a method for creating a form of low drag around certain fixed limitations, and then to employ a method of calculating the 'K'. We have asked M.Andreau if he would kindly tell us more about this, since he used this procedure to produce the Thunderbolt for Capt.Eyston, with apparent success. | ||