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).
Test report detailing modifications and performance analysis of a fuel injection pump.
| Identifier | ExFiles\Box 179b\3\ img015 | |
| Date | 6th January 1933 | |
| -4- The stops were filed so that at maximum injection the relief passage was just on the point of being uncovered, the discharge was then 396 ccs/min. and spill 151 ccs/min, no improvement was noticeable in the spill, but the delivery had slightly increased. It was then found that with the pump stationary, there was leakage past the barrel from the suction chamber to the spill chamber due to excessive clearance between the barrel and the pump body. This leakage was 40 ccs/min. and was reduced to 6 ccs/min. by nickel plating the outside of the barrel. The pump was reassembled and a delivery of 402 ccs/min. and spill of 146 ccs/min. obtained. Reducing the clearance between the pump barrel and body therefore increased the delivery and slightly reduced the spill. By setting the stops on the control lever bracket to increase the overlap between the relief passage and spill ports at maximum injection, the leakage was reduced to 133 ccs/min. and the discharge increased to 414 ccs/min. By further increasing the overlap the spill was reduced to 94 ccs/min. and the discharge again increased to 432 ccs/min. After about 2 hours running the spill had fallen to 78 ccs/min. and the delivery increased to 441 ccs/min. These tests proved that much of the spill was due to leakage from the relief passage past the plunger to the helical slots in the pump barrel. In common with many other fuel injection pumps this leakage path decreases as the pressure increases. The rather great leakage past the plunger to the spill port in this pump may be due to elastic distortion of the pump barrel at the unsupported region of the spill ports. When the pump had been adjusted to deliver 441 ccs/min. change of suction head from 12 ft. to 1ft. was found to have practically no effect in delivery or spill, but at an earlier stage of the development work, reducing the suction head from 12 ft. to 1 ft. of oil increased the delivery from 402 ccs/min. to 424 ccs/min. and reduced the spill from 146 to 122 ccs/min. The increase in delivery was probably brought about by the better filling of the pump chamber. The reduction in spill would be due almost entirely to the decrease in the direct leakage from suction to spill chamber. The variation of delivery at maximum injection with pump speed was now determined for a speed range of 335 r.p.m. to 1000 r.p.m. and the resulting curve is shewn in Fig.2. | ||