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).
Paper on the use of an electric telemeter to measure and analyze valve-spring surge.
| Identifier | ExFiles\Box 56\2\ Scan086 | |
| Date | 15th January 1929 guessed | |
| 2 ELECTRIC TELEMETER AND VALVE-SPRING SURGE pressure on the other. Each stack forms one arm of a Wheatstone's bridge, and the current through the bridge instrument—milliammeter or galvanometer—varies as the pressure on the stacks is varied. This bridging instrument may be the galvanometer element of an oscillograph, in which case oscillograms can be made and preserved for record. In addition to the telemeter itself and to the bridge, the apparatus consists of an oscillograph, a photographic drum and an oscillating-mirror mechanism. The last can be used with the oscillograph for visual inspection of the dynamic operation. This equipment is seen in Fig. 2. The telemeter is used with the valve-spring testing-machine developed by our company and shown in Fig. 3, which consists of a shaft and a flywheel driven by an induction motor through a variable-speed transmission. The outboard bearing of the shaft is mounted in a cast-iron head upon which is mounted a block containing the valve-train parts assembled in their correct relation. The cam used to actuate the train is mounted on the extreme end of the shaft. To adapt the telemeter to the work, a C-shaped spring was interposed between the valve-train block and the upper end of the valve-spring, as seen in detail in Fig. 1. OPERATION OF THE EQUIPMENT In operating the equipment, the valve-train testing-machine is started and adjusted to the speed at which observations are desired. The bridge is balanced and the telemeter gage properly adjusted beforehand. When the deflection, as shown by a milliammeter incorporated in the bridge, is adjusted to a convenient amount by either a range-changing plug or by varying the bridge voltage, a double-throw knife-switch is thrown to substitute the oscillograph element for the bridge milliammeter. An oscillogram can then be made, or the oscillating mirror can be driven at synchronous speed to give a visible standing wave. To record the whole range of operation, the motor current can be cut off with the machine running at high speed, and an oscillogram made as the speed of the machine falls off. This will show the entire range, including the surge points. The drum for exposing the photographic film takes a film 36 in. long and 6 in. wide, and is driven by a sewing-machine belt from a speed-reduction unit. An interesting feature of the drum is an arrangement for axial movement, obtained by a worm and gear. When the drum switch is closed, a solenoid shutter opens; at the same time a magnetic clutch holds the worm-gear stationary so that the worm advances axially as it rotates. As a result, the wave motion has a slight lead across the film. By this means it is possible to obtain on a 36-in. film a record 12 ft. long, covering an elapsed time of 15 to 20 sec. The electrical circuits of the equipment, including the telemeter, the Wheatstone's bridge, the light control for the oscillograph, the solenoid shutter for the drum, and the magnetic clutch which governs the axial motion of the photographic drum, are all shown in the diagram of Fig. 4. OBTAINING QUANTITATIVE READINGS By means of a comparator having a micrometer microscope, the telemeter gages employed can be so calibrated that each milliampere through the bridge corresponds to a known deflection of the gages. The deflection of the gage is the deflection of the C-spring, which in turn is proportional to the deflection of the valve-spring. Hence the calibration provides a means of relating the amplitudes of the complex valve-lift curve to the actual dynamic stresses present at any instant. One of the most important considerations of the method deals with the frequency of the C-spring. This must be high enough to prevent the spring from being brought into resonance at any time. The C-spring finally adopted deflects 0.001 in. under a load of 50 lb. This stiffness, with small mass, gives the C-spring a natural frequency of about 1000 cycles per second. This is too high to be observed on any of our records, and if present it can do no more harm than to introduce a high-order harmonic in the form of the spring-frequency wave. Means are provided also for damping this C-spring, when necessary. The basic idea of the C-spring is that, as the valve-spring surges and the wave motion travels up and down the spring, the pressure at the top of the spring varies during every cycle of the wave motion. At one time this pressure is more than the normal static value, while at another time the pressure is less. This difference in pressure varies the deflection of the C-spring correspondingly, and its effect is thus transmitted to the recording apparatus by means of the telemeter gage. Since the normal full deflection of the C-spring will be only 0.001 in. when the difference between valve-open and valve-closed loads is 50 lb., the deflection-load curve is a straight line. Another important item in a discussion of this method of testing valve-springs deals with the oscillograph galvanometer-elements. It must be admitted that the oscillograms do not and never can give a strictly true picture of the wave shapes encountered, because the galvanometer elements have a certain slight mass and therefore a period of vibration in themselves. However, any errors on this account will be negligible provided the elements are well damped and have frequencies high enough to produce no objectionable harmonics in the wave-form. In the case at hand the elements have frequencies of about 1200 cycles per second, which is sufficiently high for our purposes. In making observations with the oscillograph, one element is used to obtain the complex valve-lift curve, while a second is connected in series with suitable resistance units across the secondary of a bell-ringing transformer to obtain a 60-cycle alternating-current timing wave. THE CAUSE OF SURGE Telemeter oscillograms, as shown in Figs. 6 to 20, are of material aid in the determination of the true cause of valve-spring surge. In using the oscillograms for this purpose it is important to investigate the operation at speeds which indicate both resonant and non-resonant conditions. The points of resonance are the points with which we are most concerned, because they indicate the presence of harmonic forces which cause the spring to vibrate. Since spring frequencies ordinarily lie between 10,000 and 30,000 vibrations per minute, it is obvious that the harmonic forces exciting the spring must, to cause resonance at various speeds, have frequencies far in excess of the camshaft speed; yet these harmonic forces are definitely linked with the cam, because the cam is the agent by which they are set up. Referring to the telemeter oscillograms again, it can be seen that at each resonant speed there is an integral number of vibrations of the spring per camshaft revolution. It can be further observed that the number of complete vibrations or waves per revolution times the resonant speed in revolutions per minute equals the spring frequency in vibrations per minute. This establishes the fact that for resonance, where there are n free waves per revolution, there must be acting an harmonic force which has a frequency of n times the fundamental or camshaft frequency. At times the spring can be observed to be vibrating in halves at double its fundamental frequency, when the harmonic frequency of n times the resonant speed equals twice the spring frequency. The spring then seems to vibrate in much the same way as a vibrating string, in that it is capable of vibrating in either its fundamental frequency or its first overtone. Undoubtedly, if there were present strong harmonic forces of higher order than those commonly encountered, the spring would vibrate in thirds as its second overtone comes into resonance with the harmonic forces. Actually, the spring can be observed vibrating in either its fundamental or its first overtone, or in its fundamental with the overtone superimposed. In other words, a resonant condition will result when ns{Norman Scott} = NF, where n is the order of the harmonic force; s is the camshaft or fundamental speed, in revolutions per minute; N is the order of the spring vibration; and F is the fundamental frequency of the spring, in vibrations per minute. RESONANT AND NON-RESONANT HARMONIC FORCES The presence of harmonic forces can be accounted for by investigating the valve-lift curve throughout its complete cycle from one valve-open position to the next, considering the first valve-open position as 1 deg. and the second as 360 deg. Such a curve, which is continuous and symmetrical about the 180-deg. point, can be represented by a constant term and a simple trigonometric series. The solution of such a curve involves the evaluation of the coefficients of the various terms of the series, which coefficients indicate the amplitudes and signs of the various harmonics. Expressed mathematically, the series can be equated as follows: f (y) = A₀ + A₁ cos wt + A₂ cos 2wt . . . + Aₙ cos nwt (1) [Text from Figures] FIG. 2—RECORDING APPARATUS OF ELECTRIC TELEMETER FIG. 3—VALVE-SPRING TESTING-MACHINE FIG. 4—DIAGRAM OF ELECTRIC TELEMETER Text within diagram: Solenoid for Shutter 2¹/₂-6-Volt Lamp Carbon-Stacks 0-to-3-Amp Ammeter Rheostat Range-Changing Plugs Oscillograph Light Switch Milliammeter Drum Worm Magnetic Clutch for Drum Worm Drive Drum-Key Drum Switch FIG. 5—LIFT CURVES OF TWO CAMS Text within graph: Lift, in. Cam Angle, deg. Cam No. 1 Cam No. 2 Axis values (Y): 0, 0.040, 0.080, 0.120, 0.160, 0.200, 0.240, 0.280, 0.320 Axis values (X): 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120 | ||