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).
Disappointing progress and failures of the Phantom III Oil Cooler designs.
| Identifier | ExFiles\Box 131\1\ scan0158 | |
| Date | 7th December 1935 | |
| To E.{Mr Elliott - Chief Engineer} from Hs{Lord Ernest Hives - Chair}/Rm.{William Robotham - Chief Engineer} X114 Hs{Lord Ernest Hives - Chair}/Rm.{William Robotham - Chief Engineer}18/KW.7.12.35. Phantom III Oil Coolers. We think you will agree that the progress of the Ph.III Oil Cooler has been rather disappointing. Briefly, as we see it, the position is this. We brought the Buick Oil Cooler back from America. It was perhaps not easy to construct an oil cooler exactly like the Buick owing to the different disposition of the water pump, etc. However, knowing the amount of development work which had been put into this cooler, we were anxious that it should be adhered to as closely as possible so that troubles could be avoided. We handed the cooler over as a model. The secret of the freedom from leaks of the honeycomb block in this cooler appears to be the avoidance of flat surfaces on the edge of the matrix. The first R.R. cooler failed mechanically and it was altered. It was thought that the back pressure in the cooler was too high and that this caused failure. It was therefore re-designed with less internal resistance, which dropped the efficiency approximately 35%. Furthermore, as will be seen from the attached report, the actual internal pressure on the cooler was only reduced 7.5% by this modification, as the majority of the restrictions were not due to the cooler matrix but to the rest of the system. The second design of cooler has failed on the 10,000 miles test. We thought the quality of the solder we used might have some bearing on these failures and have tried silver solder as used on Glycol radiators, but without much success. We think, however, that the reason the cooler has failed on the 10,000 miles test must be a combination of temperature and vibration. | ||