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).
Technical explanation and diagrams of a 'lashlock' mechanism and its application in a torque amplifier.
| Identifier | ExFiles\Box 154\3\ scan0063 | |
| Date | 12th April 1928 guessed | |
| [PAGE 6] [Handwritten Note at top of page] Lashlock as used in torque amplifier. Rat-trap spring (Puts light load on brake shoes)... [Main Text - Right Column] in opposite directions to the gears A and B. The wedge will be pushed outward until the teeth on gears A and B lightly pinch the teeth of gear C; this may be termed the “normal” position of the wedge. There are only two circumstances which can cause the wedge to change from this normal position: either both of the rollers G and H press on it simultaneously, or both simultaneously relieve their pressure. Pressure of only one roller will cause binding friction of wedge on stem; therefore, when gear C begins to drive it will cause one of the split pair to exert pressure on its respective roller, and so on one side of the wedge; thus the driving force of C will be transmitted to the stem and so to the shaft which carries the split pair. (Gear C may, of course, be the driven gear.) If C reverses the direction of its drive the other roller will press against the wedge and the shaft driven oppositely. There will be no instant when neither roller is exerting pressure on the wedge, and no instant when both rollers are pressing simultaneously. All this is true provided the teeth on all the gears are accurately cut. If they are of different size through inaccurate cutting or wear, a thick tooth will urge the rollers together and move the wedge outward, while a thin tooth will permit the spring to push the wedge inward and cause the split to pinch a little more closely. It must be remembered that the pinching is always a light force, because the spring which urges the wedge outward need be only of light force—practically negligible compared with the other forces involved—for the lashlock action is through positive frictional locking and bears no relation to the strength of the spring employed on the wedge. In the non-lashlock type of split gear the pinching is necessarily always comparatively severe because the spring must be heavy enough to take the full load of the drive in one direction. It will easily be understood, then, how by this accommodating inward and outward movement of the wedge all irregularities of the teeth are compensated for, that the action is immediate and automatic, and that the teeth never are put under strain. In case, through lack of lubrication, the wedge should tend to stick on its stem, the outward push of the rollers would overcome such friction without subjecting the gears to any considerable strain; the only result of sticking is failure of the spring to push the wedge inward, which means that the backlash would not be adequately eliminated. The diagram does not give an illustration of the construction of a set of lashlock gears. The lashlock requires but little space, and need not be set inside the gears at all. It has been positioned, for instance, on a handwheel on the end of a shaft a foot or more from any of the gears. It will be readily understood, also, that one lashlock may eliminate backlash from a train of gears, or from a combination of gears, worms, racks, [Image Captions] Fig. 4—Double involute cam lashlock. Fig. 5—Threaded-type of lashlock Fig. 6—Three forms of wedge used in the lashlock. Straight-sided, double cam, “T” or screw type (Text on component in image: B. Bradbeer. 17113. Lawrence House. No. 77) Fig. 7—Lashlock on a screw-feed. Positive drive is obtained in both directions with reverse of direction less than 0.0001 in. of carriage travel [Main Text - Left Column] mechanisms through which only small forces are transmitted, for if the strength of the spring be increased the pinching of the teeth will become excessive, producing wear and strain. Incorporating the lashlock in a split gear, however, transforms it from a weak element with a narrow range of usefulness into a strong, positively-acting element capable of being widely applied to both light and heavy duty machinery. In Fig. 2 is illustrated, diagrammatically, the lashlock principle on a split-gear element, A and B are the split members, meshing with the solid gear C. Stem D is attached rigidly to the shaft which carries the split pair, and on the stem is mounted the wedge member E.{Mr Elliott - Chief Engineer} Rollers F and G are attached respectively to gears A and B. A light spring H, is mounted so as to urge the wedge away from the shaft, thus urging apart rollers F and G, and through them imparting rotational urge [Page Number] [6] [PAGE 8] [Handwritten Note at top of page] Drive from generator, camshaft, water pump, or even small electric motor. [Handwritten Notes near image] Conventional steering gear might be replaced by say a worm + gear with lash lock. 6" brake drums rotating in opposite directions. Could be 4" dia. [Image Caption] Fig. 4—Torque amplifier on automobile. Spark and gas controls pass through amplifier and emerge at end of steering column. Drums are driven from engine connection. The steering gear proper may be left undisturbed or the ratio reduced [Main Text - Left Column] ing the band in the opposite drum; therefore, the work arm, because of the frictional engagement of the band in one of the drums is pushed around by that band in the direction of the drum’s rotation. Moreover, the force with which the band presses against the work arm stud is many times the force applied at the control arm stud. This follows from the well-known principle of wrapping friction, commonly used in calculating belt drives, snubbing problems, etc., and equally applicable to an outside band frictioning against the outside of a drum, and to an inside band frictioning against the inside of a drum. The formula is usually expressed: F₁ / F₂ = eμa Where F₁ and F₂ are the forces applied respectively to control and work ends of the wrapping friction member, e is the base of natural logarithms, μ is the coefficient of friction, and a is the angle of wrap in radians. It is to be noted that this formula does not mention the diameter of the drum or the width of the band, these factors being immaterial to the theory. In the case of the amplifier shown we can assume a coefficient of friction of 0.6, an angle of wrap of a little less than a full circle, say six radians, giving a relation of F₁ to F₂ of about 40 to 1. The action of the work arm in following the movements of the control arm can be understood from the following consideration. Observing only drum B as shown in Fig. 2, if the control arm E is moved to the right by an infinitesimal amount, the band which was originally barely in contact with the drum, is now put into frictional engagement with it. This forces the end of the work arm also to the right, the movement of the work arm in this direction tending to loosen the [Main Text - Right Column] land, so that when the work arm has followed the control arm through exactly the same angle the band again reaches the condition of being barely in contact with the drum. If the movement of the control arm is a continuous one to the right then the work arm follows continuously so that the tape is always maintained in bare frictional contact with the drum. In the foregoing it was assumed that the work arm moved with perfect freedom. If, however, the work shaft is doing external work it lags behind, thereby tightening the band until sufficient frictional contact is obtained to overcome the external resistance. The amount of such lag depends on the nature of the friction material. If a very spongy substance were used for this purpose, the angular lag would be considerable; but in the case of nearly incompressible friction materials it is so small as almost to escape detection. The difference between a band which slides freely within the drum and one which grips with a heavy friction is only a matter of 0.001 in. in radial expansion or about 0.006 in. expansion between the ends of the band. Special attention must be called to the fact that the speed of the drum has no relation to the speed of the band and connected parts except that the drum speed must always be higher than any speed at which it is desired to rotate the control shaft. In the normal operation of an amplifier the bands are intended to be in floating frictional engagement with the drums and never to grip them positively. The fact that the friction is of this floating type accounts for the fact that even with high drum speeds the control shaft may be revolved at a very slow speed while the work shaft executes the same slow movement with perfect uniformity. Such an amplifier as described above can be controlled in either direction, the torque furnished by the work shaft being always forty times that supplied by the control shaft, this "ratio of amplification" being theoretically constant for torques of any magnitude and practically so over a wide range of conditions. A practical difficulty is encountered in constructing an apparatus as above described because of the necessity of accurately adjusting the lengths of the bands so that in reversing the direction of the control shaft the band on one side shall begin to grip just when the band on the other side releases. If the bands are made too long, both will be gripping at the same time whereas if too short there will be a neutral space when neither grips. To avoid this exceedingly close tolerance and alterations in service due to heating and wear, a floating automatic adjustment (the lashlock, which will be described in a later article) has been provided at the end of the work arm to keep both bands in light frictional contact under any condition. Thus particular care is necessary to make both bands of exactly the same length. If one happens to be longer than the other it causes a slight angular displacement between the power and control arms, an error which being constant and always in the same direction automatically cancels out when the amplifier is initially set. If, after the mechanism is adjusted and calibrated, wear occurs in both bands equally, the backlash adjuster automatically takes it up without altering the adjustment. For purposes where an amplification of 40 to 1 is not sufficient, the desired degree of amplification might be obtained by using a number of amplifiers in series, the work shaft of the first being connected to the control [Page Number] [8] | ||