UDC 62-272

INFLUENCE OF THE WAVINESS OF S-SHAPED SPRINGS ON THE ACCURACY CHARACTERISTICS OF CLOCKS

V. I. KALYATINA, Yu. Ya. KOVYLIN

(Presented by the Scientific-Methodological Seminar of the Department of Applied Mechanics)

In the manufacture of S-shaped mainsprings for wristwatches, waviness (corrugation) is observed in some of them.

In the literature known to us, this fact has been reflected twice: in T. A. Gevondyan’s monograph Spring Motors (1956) and in the technical report Investigation of the Reliability of Spring Motors for Special Purposes, carried out at the Scientific Research Institute of the Watch Industry in 1967 under the direction of T. A. Gevondyan. It was noted that the causes of the occurrence of waviness had not been clarified, its influence on clock parameters had not been established, and no inspection instrument for detecting it was available.

In connection with this, we carried out investigations, as a result of which it was established that an insufficient value of the diameter of the ring for the first winding is one of the possible technological causes of the occurrence of waviness, while increasing the diameter by 20–25% compared with that prescribed by the technology substantially reduces the number of defective springs.

The present article is devoted to an experimental study of the influence of waviness on the accuracy characteristics of clocks. The object of the study was mass-produced “Polet 2409” watches of the First Moscow Watch Factory (balance oscillation period 0.4 s, balance lift angle 49°) with an S-shaped spring \(0.105 \times 1.38 \times 325\) mm made of cobalt alloy K40TYu, manufactured according to the existing technology. One of the springs tested had no defects, while the other had waviness and kinks at the second turn.

The experiments were carried out at the Scientific Research Institute of the Watch Industry on an “Isochronograph-Ampligraph” measuring complex manufactured by GREINER electronic S. A., Zongenthal/Suisse.

The complex is intended for simultaneous measurement of the instantaneous daily rate and the amplitude of balance oscillations of lever watches of any caliber. The measurement results were recorded by an \(x\)-\(y\) recorder in linear form. The chart paper moved at a speed of 5 mm/hour, which made it possible to analyze the operation of the watch mechanism with the balance spring under study from full winding of the motor to complete stop. Dynamograms of the springs under study were recorded on the “Tochnograph” instrument.

Before the start of the experiment the watches were brought into a state corresponding to absolute fulfillment of all requirements of the existing assembly technology. The watches were tested in the position with the dial upward.

Simultaneous information in the form of three records (Figs. 1, 2), characterizing the change in the spring torque over the entire possible operating time of the clock from one full winding, as well as the amplitude and the instantaneous daily rate, made it possible to form an idea of the influence of spring defects on the output accuracy parameters of the clock.

Figures 1a, b show the amplitude of balance oscillations and the instantaneous daily rate of a clock with a spring without defects.

Fig. 1. Characteristics of a spring without defects: a — amplitude of balance oscillations; b — instantaneous daily rate; c — dynamogram

The initial mean value of the amplitude of balance oscillations was 305°, the rate \(+22\) sec/day.

Throughout the entire test period no increase in the scatter of the rate and amplitude was observed.

After one day the amplitude was at the level of 280°, i.e., its decrease during one day of clock operation amounted to 25° (8%). The instantaneous daily rate was sufficiently stable (\(+22\) sec/day). Periodic deviations 1, 2, 3, 4, 5, 6 of the amplitude are explained by a violation of the processing technology of the escape wheel, and the spikes in the range 1-1 by interference in the supply circuit. Diagrams 1a, b satisfy all the requirements imposed on clocks of good quality. The dynamogram of the spring under investigation without a defect is presented in Fig. 1c. Dynamograms 2a, b characterize the change in the amplitude of balance oscillations and the instantaneous daily rate of the clock when an S-shaped spring with waviness (corrugation) and bends is installed in the movement. The dynamogram of the spring with defects is presented in Fig. 2c. In Figs. 1a and 2a the portions preceding stopping are not reproduced, since their analysis does not rep—

is of interest. It is not difficult to see that diagrams 2a, b differ sharply from diagrams 1a, b in the following respects. The amplitude of oscillation of the balance has a high degree of instability, characterized by a fairly large scatter lying in the range from \(315^\circ\) to \(215^\circ\). This also determines the instability of the oscillation period, which is vividly illustrated by diagram 2b. Thus, segment 1 has a rate scatter of 32 sec/day at an instantaneous daily rate of \(+22\) sec/day. At the same time,

Fig. 2. Characteristics of a spring with waviness and bends: a—amplitude of oscillation of the balance; b—instantaneous daily rate; c—dynamogram

in segment 2 the rate scatter is 4 sec/day, but the running of the clock continues for only 30 minutes. Segment 3 is characterized by a further sharp increase in the scatter of the clock rate due to the interaction of the wavy sections of the spring. In segment 3 the clock operated in this way for 7 hours, which constitutes the main operating time during the day. Segments 4, 6, 7 are characterized by noticeable instability due to smaller defects in the spring.

Thus, during the entire operating time the clock functioned with a sufficient degree of accuracy for only 3 hours 30 minutes out of 48 hours.

As follows from the foregoing, the main defects of the spring that affect the output accuracy parameters of the clock are located in the most “dangerous” section, which characterizes the operating time of the clock before the next winding.

The experiment carried out makes it possible to draw two main conclusions:

— poor-quality manufacture of the spring entails instability of the amplitude of oscillation of the balance, which, through the non-isochronism of the clock, causes

leads to a change in the accuracy of the rate. The isochronous characteristic of a watch with a defective mainspring is unstable throughout the entire time of operation of the watch over a day;

— from the form of the dynamogram (see Figs. 1c, 2c), one can judge the presence or absence of defects in the spring. Thus, among the recording forms “Technograph—Isochronograph—Amplitgraph” a high degree of correlation is observed, which provides grounds for recommending the use of the “Technograph” for operational monitoring of the quality indicators of mainsprings.

Great assistance in organizing the experiments and discussing the results obtained was provided to us by V. A. Lysyi, a staff member of the Scientific Research Institute of the Watch Industry.