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).
Analysis of an 'F' Type oil cooler failure on Chassis 3.BU.26.
| Identifier | ExFiles\Box 131\5\ scan0079 | |
| Date | 17th May 1938 | |
| 1114 or 261? To Hs.{Lord Ernest Hives - Chair} from Hm{Capt. W. Hallam - Head Repairs}/Std. By: c. RHC.{R. H. Coverley - Production Engineer} c. Br.{T. E. Bellringer - Repair Manager} c. Hd.{Mr Hayward/Mr Huddy} 'F' TYPE OIL COOLER FAILURE - CHASSIS 3.BU.26 - COOK. We have received the cooler (No.F.263) which failed on this car and have examined same to determine cause of failure. Cooler was pressure tested and found to leak from a fine hole (see attached photographs) in the solder at the junction of the cooler top plate with the matrix tubes. It was then noticed that the cooler had one aluminium joint washer on one union (the one near the failure) as called for by scheme and LOP., but that two washers were fitted on the other union - see attached sketch and photographs. Further, these washers were not loose, but were stuck to the unions by the Heldite jointing compound which had obviously been used when assembling the cooler in its casing. Since the seating faces on the cooler top plate are held in line and flat within .004", and each cooler is inspected for this as it comes in from the makers, the fitting of the extra washer puts the faces out of line by its thickness (.048"), and when the cooler is tightened down inside its box, on to the machined seating faces inside same, a total of .048" distortion must take place in the top plate - as shown on attached sketch. This distortion will amount to about .010 at point where failure occurred and means that either the matrix block must distort that amount in shear or else the top plate must lift locally from the tubes. We have stripped off a portion of the top plate where failure occurred and it can be seen that although the plate was once soldered to the tubes, it has locally parted from them and the solder faces are oil blackened, the one having penetrated to the edges of the tubes and plate. As a side issue, the removal of this piece of plate has disclosed the presence of an air bubble in the solder but this has not been in contact with oil. We have endeavoured to find traces of a similar effect on the other side of the cooler but no such break has occurred there - this may be partly due to the fact that the depth and strength of soldering on that side were better than on the failed side, and partly due to the fact that the baffle fitted to the sound side has a strengthening effect. | ||