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).
Letter from The Morgan Crucible Company regarding pump design, water pressure, and carbon seals.
| Identifier | ExFiles\Box 149a\1\ scan0119 | |
| Date | 4th August 1937 | |
| -3- EH/ETR/ To Rolls-Royce Ltd., DERBY. The Morgan Crucible Company, Limited. 4th August, 1937. Firstly dealing with designs similar to that shown on our Drawing No.100/640. It is obvious at once with this design that it is not possible to completely balance the pump so that if there is a water pressure in the involute casing there is a tendency for the sleeve, complete with impeller, to be blown off the carbon seating. This pressure should not be allowed to exceed 15 lbs. per square inch. Resisting this action is the spring which we recommend should be made to give also 10 to 15 lbs. per square inch on the carbon gland. The highest pressures which we have dealt with, with this type of design, are occasions on which there are thermostat restrictive devices in the water circulating system. When the engine is cold and the thermostat closed it is possible to get a build up of pressure as high as 25 lbs. per square inch, but this is only of a very temporary nature; we have overcome this difficulty by increasing the spring pressure up to approximately 20 lbs. per square inch. We would recommend that if you are in the design stage, you arrange that the spring pressure exerted does not exceed 15 lbs. per square inch as this has been found to give a life in service of over 100,000 miles, a mileage with which no doubt you would be satisfied. You will realise that when there is little or no pressure in the circulating system the full spring pressure is exerted on the carbon seal thus, the lower this can be kept the longer will become the life of the gland. Referring to the two questions in your postscript:- 1). We do rely on the carbon providing a seal by being a press fit in the housing. The interferences would be .001" minimum and .003" maximum, thus allowing you a .001" tolerance on your housing, and we would use the remaining .001" as a tolerance on the outside diameter of our carbon seal. We would assure you there is no danger of leakage at this point. | ||