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 valve-spring surge using an electric telemeter, with spring characteristic data and oscillograms.
| Identifier | ExFiles\Box 56\2\ Scan089 | |
| Date | 15th January 1929 guessed | |
| 8 ELECTRIC TELEMETER AND VALVE-SPRING SURGE TABLE 2—SPRING CHARACTERISTICS | | Spring No. 1 | 2 | 3 | 4 | | :--- | :--- | :--- | :--- | :--- | | Pitch Diameter, in. | 0.998 | 1.000 | 1.008 | 0.998 | | Free Length, in. | 2 23/32 | 2 3/4 | 3 | 2 25/32 | | Total Number of Coils | 9 1/2 | 9 1/2 | 12 1/4 | 9 | | Pitch | Uniform | Variable | Uniform | Variable | | Load, valve open, lb. | 74.0 | 77.0 | 72.0 | 104.0 | | Load, valve closed, lb. | 46.5 | 47.0 | 50.5 | 68.0 | | Stress, valve open, lb. per sq. in. | 57,800 | 60,000 | 56,800 | 73,700 | | Stress, valve closed, lb. per sq. in. | 36,300 | 36,700 | 39,800 | 48,200 | | Static Stress-Range, lb. per sq. in. | 21,500 | 23,300 | 17,000 | 25,500 | | Free Frequency, vibrations per min. | 16,500 | 16,500 | 12,000 | 21,500 | Note.—The ends of all springs are closed and ground square. Gage of wire for all springs, Washburn & Moen No. 9 (0.148 in.). The resonant points for all the speeds and harmonics can be seen in Fig. 6. It should be noted that the highest speed at which there is no overtone in the wave motion is 1030 r.p.m., at which speed the exciting force seems to be the sixteenth harmonic alone. This indicates that harmonics above the thirtieth are either absent or are very feeble. It should also be noted that nothing on the oscillogram indicates a resonant condition due to the nineteenth harmonic, which was found in the harmonic analysis to be negligible. In a general way the magnitude of the surge at the resonant points checks well with the calculated values for the amplitudes of the harmonics. This can be verified by comparing Fig. 6 and the detail views with the harmonic values for cam No. 1 in Table 1. A further point of is interest in connection with Fig. 6. At the peaks of the valve-lift curve at resonant speeds, if the free waves have a crest, or rise above the lift curve, the corresponding calculated harmonic is positive; and if the free waves have a trough, or fall below the lift curve, the corresponding harmonic is negative. SLIGHT CHANGE WITH DIFFERENT CAM The oscillogram in Fig. 7 shows the operation of spring No. 1 and cam No. 2 through a range of speed from 1750 to 750 r.p.m. Except at certain resonant points, the oscillogram is very much the same as that of Fig. 6. One point of difference between the two is that the twelfth harmonic has a very feeble effect in Figs. 6 and 10, but causes very appreciable surge in Figs. 7 and 12. Also, no twenty-fifth harmonic is noted in Fig. 6, but in Fig. 7 it causes resonance at a speed of 1300 r.p.m. the spring vibrating in its first overtone. The sixteenth harmonic which causes appreciable surge in Fig. 6, is absent in Fig. 7; and the nineteenth, which is absent in Fig. 6, has an effect in Fig. 7. All these differences are borne out fairly well by the differences in the harmonic values of Table 1. The oscillogram shown in Fig. 13 was made to record the operation of the same spring and cam at still higher speeds, from 2150 to 1150 r.p.m. The most interesting point on this oscillogram occurs at 1835 r.p.m., at which point there is a slight resonance-peak due to the combined effects of the ninth and eighteenth harmonics. It is to be noted that this peak is bordered by violent surges due to the eighth harmonic on one side and the very bad tenth on the other. Table 1 shows that the eighth and tenth are strong harmonics, while the ninth is very feeble. Another interesting fact to be learned from Fig. 13 is that there is a decided tendency for the higher harmonics to cause the spring to vibrate at double frequency at high speeds. VARIABLE PITCH IMPROVES THE SPRING The oscillograms shown in Figs. 14 and 15 represent the operation of spring No. 2, which is identical with spring No. 1 except that it is wound with a variable pitch. The speed range is from 1750 to 750 r.p.m., and cams No. 1 and 2 were used, as indicated in the caption. A fraction of an active coil is closed in the valve-open position so that spring No. 2 has a slightly higher frequency than spring No. 1. As the valve opens, the effective coils of spring No. 2 decrease until valve-open position is reached, at which point about 2 1/2 active coils are closed up. This variable-pitch action causes the frequency of the spring to vary continuously during the valve-lift part of the cycle, therefore the spring can never attain a completely resonant condition. However, the spring does vibrate rather badly at times on the base-circle part of the curve, which can be accounted for by the fact that the spring is rather low in frequency and that the lower harmonics tend to excite the spring when the valve is closed and the frequency of the spring does not change. Figs. 14 and 15 demonstrate that, simply by varying the pitch of the spring as described, bad resonant points can be avoided. However, there are at times slight maximum and minimum points in the surge amplitude, as determined by the amount of interference of the present in the wave motion at any time. SPRINGS DESIGNED TO COMPRESS SOLID Figs. 16 and 17 are oscillograms made with spring No. 3, which was so designed that, when the valve is wide open, the spring lacks but 1/64 in. of being compressed solid. Speeds and cams were the same as for Figs. 14 and 15. The results show that a certain damping is obtained by having the spring compress nearly solid once during each revolution of the cam. The top of the lift curve is either almost non-oscillatory or the oscillation is of very high frequency. The spring has a very low frequency when operating on the base-circle part of the curve, however, and it surges badly then under the influence of the lower harmonics. When observed visually and audibly, these two set-ups appear to be even worse than the oscillograms show. Above a certain speed they seem to surge badly all the time, and they are very noisy because of coil-clash. It is doubtful if the full effects of the coil-clash are transmitted through the apparatus to the photographic film. A HIGH-FREQUENCY VARIABLE-PITCH SPRING Figs. 18 and 19 show the operation of a well-designed high-frequency variable-pitch spring operating through a speed range from 1750 to 750 r.p.m. with the same two cams that were used for Figs. 16 and 17. The freedom of these two oscillograms from resonant points should be noted. Comparison of these oscillograms with those of Figs. 14 and 15 shows the marked superiority of this spring over a low-frequency variable-pitch spring. REFERENCE OSCILLOGRAM FROM HARMONIC CAM The oscillogram shown in Fig. 20 was taken to show the operation of spring No. 1 at speeds between 1750 and 750 r.p.m. when actuated by a harmonic cam. The FIG. 18—OSCILLOGRAM OF SPRING NO. 4 AND CAM NO. 1, FROM 1750 TO 750 R.P.M. FIG. 19—OSCILLOGRAM OF SPRING NO. 4 AND CAM NO. 2, FROM 1750 TO 750 R.P.M. FIG. 20—OSCILLOGRAM OF SPRING NO. 1 AND HARMONIC CAM, FROM 1750 TO 750 R.P.M. 9 | ||