First-Hand:The First Quartz Wrist Watch: Difference between revisions
From ETHW
mNo edit summary |
mNo edit summary |
||
Line 21: | Line 21: | ||
== In November, 1965 Roger Wellinger, director CEH and responsible for the yearly strategy and plan, declared the "montre-bracelet à quartz" to become the primary strategic goal for the year 1966.<sup>7)</sup> This was a direct consequence of Armin Frei having designed, built and operated successfully a miniturized quartz oscillator prototype showing feasibility of a quartz wrist watch in the 10 kHz range in 4Q65. In 1Q66 Frei disposed already of a miniaturized quartz oscillator prototype with an 8192 Hz quartz resonator (picture), a novel fully integrated driver circuit running at less than four microamps current consumption (black epoxy covered IC with red dot) and a frequency adjustment set up (upper black epoxy covered IC without a dot), all these components survived till and including the industrial phase with minor improvements only. At that time the term "montre-bracelet à quartz" appeared the first time officially in CEH's documents, and it had to be defended against the many internal and external opponents of this new direction. Never mind the rest of the quartz wrist watch was still not worked out yet or decided upon, Wellinger requested that a general systems invention disclosures on the "Montre-bracelet électronique à quartz" had to be worked out.<sup>8)</sup> == | == In November, 1965 Roger Wellinger, director CEH and responsible for the yearly strategy and plan, declared the "montre-bracelet à quartz" to become the primary strategic goal for the year 1966.<sup>7)</sup> This was a direct consequence of Armin Frei having designed, built and operated successfully a miniturized quartz oscillator prototype showing feasibility of a quartz wrist watch in the 10 kHz range in 4Q65. In 1Q66 Frei disposed already of a miniaturized quartz oscillator prototype with an 8192 Hz quartz resonator (picture), a novel fully integrated driver circuit running at less than four microamps current consumption (black epoxy covered IC with red dot) and a frequency adjustment set up (upper black epoxy covered IC without a dot), all these components survived till and including the industrial phase with minor improvements only. At that time the term "montre-bracelet à quartz" appeared the first time officially in CEH's documents, and it had to be defended against the many internal and external opponents of this new direction. Never mind the rest of the quartz wrist watch was still not worked out yet or decided upon, Wellinger requested that a general systems invention disclosures on the "Montre-bracelet électronique à quartz" had to be worked out.<sup>8)</sup> == | ||
== [[Image:Prototyp 27 mm.jpg|center|625x500px]]<br>With the change of the strategy Roger Wellinger terminated abruptly Max Forrer's low frequencies Beta project with the metallic tuning fork and electromagnetic and piezoelectric actuators. – The situation at Seiko, the Japanese competitor, had been much different. They could rely on the experience with quartz clocks and electronic time keeping which they collected since 1956 (Olympics). They were developing their own quartz technology since 1958 and gradually developed quartz resonators for clocks, pocket watches and then for men's wrist watches successively.<br><br><br> | == [[Image:Prototyp 27 mm.jpg|center|625x500px]]<br>With the change of the strategy Roger Wellinger terminated abruptly Max Forrer's low frequencies Beta project with the metallic tuning fork and electromagnetic and piezoelectric actuators. – The situation at Seiko, the Japanese competitor, had been much different. They could rely on the experience with quartz clocks and electronic time keeping which they collected since 1956 (Olympics). They were developing their own quartz technology since 1958 and gradually developed quartz resonators for clocks, pocket watches and then for men's wrist watches successively.<br><br><br><u>4. The first Quartz Wrist Watch: Beta 1</u><br><br><u>4.1 World first:</u> == | ||
== The first quartz wrist watch was Beta 1 and it had been built at the Centre Electronique Horloger. The first unit of a series of five was assembled and tested at the CEH in July 1967. Since Seiko does not communicate any details about their first quartz wrist watch, we can firmly conclude that Beta 1 was the world's first quartz wrist watch world wide. The new watch was packed into a standard square [[Image:Image006.jpg|thumb|right]]case (picture), this was necessary because the quartz case itself was straight with a length of 27 mm. However, the overall dimensions satisfied perfectly the requirements set by the watch industry to qualify for a men's wrist watch. The watch with the identification number CEH-1020 was tested at the Observatory in Neuchâtel as of August 13, 1967,<sup>4) </sup> and reached a classification of 0.189, which means an improvement of about one order of magnitude as compared to classical chronometers.<sup>5)</sup> The classification was also much better than the one which was reached by tuning fork watches during the same period. Beta 1 was equipped with a stepping motor activating the seconds hand step by step. The alternative and later model Beta 2 was equipped with the same quartz oscillator like Beta 1, but the second hand was actuated by a 256 Hz vibrating motor and a ratchet wheel. <br><br><br><u>5. Components of Beta 1</u><br><br><u>5.1 Beta 1:</u> == | |||
== Beta 1 became the code name of the first quartz wrist watch. The miniaturized quartz oscillator was developed 1965/66 by Armin Frei. The same setup was used for Beta 2, a current saving alternative quartz wrist watch and later on with minor improvements in the industrial version Beta 21. The first working prototype of Beta 1, CEH 1020 had been assembled by Jean Hermann and Fraçois Niklès in July 1967.<sup>3)</sup> The main elements of the watch and their primary contributors, including parts and as well as concepts are described below:<br><br><u>5.2 Quartzresonator:</u> == | |||
== While the size of commercial quartz standards of those days was as big as radio tubes, we had to strive for physical dimensions to be small enough to allow the device to be mounted inside a men's wrist watch case. [[Image:Image008.jpg|thumb|right|220px]]To keep the electronics simple, the frequency had to be 2 to the power of n (n being a na- tural number) in Hertz in order to produce pulses with a period of one second at the end of the divider chain. Re- quirements, which are very much contradictory, because if the dimensions are reduced the frequency goes up and vice versa. Further we learned from experiments that quartz resonators with the shape of a tuning fork and fabricated with the technology of those days exhibit a much inferior factor of quality Q as compared to straight quartz bars. The solution to all these requirements was an x-y cut quartz bar with a length of 24 mm and with an eigenfrequency of 2<sup>13</sup> = 8192 Hz (picture). The small dimensions of the quartz in its metal case as well as the extremely stringent requirements of mechanical precision, stability and life time required special attention with regards to most of the physical parameters: Leakage rates of the case and its feeds through had to be inferior to 5 10<sup>-12</sup> Torr ltr/s, organic and anorganic deposition on the surface had to be less than 20 Angstrom thick, metallurgy and soldering of wires onto the quartz surfaces had to be free from any unwanted inclusion, high precision soldering within a fraction of one millimeter was required to reach high quality factors of the resonator and many others. The quartz on the picture was developed and tested by Armin Frei in 1965, Oscilloquartz in Neuchâtel provided for the raw material and Richard Challandes was responsible for the assembly. X-Y cut bar quartz, similar to the one on the picture but with increased frequency, had been produced in Switzerland for watches until 1977. As of 1979 the Swiss watch industry produced their 32 kHz quartz tuning fork resonators in Grenchen, Solothurn licensing Jürgen Staudte's patent on etching quartz tuning forks (USA Patent Jürgen Staudte, 1972). <br><br><u>5.3 Driver circuit:</u> == | |||
== A number a different circuits for driving quartz oscillators were available at the time (Clapp oscillator, etc.), none of them fulfilled the necessary requirements for our quartz wrist watch: Say no coupling capacitances, low total resistors [[Image:Image010.jpg|thumb|right|200px]]value on the chip, tolerance to the integrated circuits fabrication process and its deviations, rigid operational stability and low power consumption. This for bipolar IC's, as well as for low battery voltages. – The newly developed, symmetric cross coupled driver circuit as shown on the picture incorporates a mini- mum of four resistors with pair wise equal values R<sub>c</sub> and R<sub>e</sub> as well as two transistors Tr<sub>1</sub> and Tr<sub>2</sub> and fullfils the above requirements extremely well. The emitter resistors serve as current sources and the collector resistors provide for the negative impedances to drive the quartz. The circuit exhibits a negative impedance of -2R<sub>c</sub> approximately measured be- tween the contacts 1 and 2. The circuit was very tolerant to various applications and conditions, and easy to fabricate. It took our specialists of the semiconductor pilot line, Raymond Guye and his colleges, less than two months to ship the first fully integrated properly working chips. The circuit was developed and tested by Armin Frei in 1965 and 1966.<br><br><u>5.4 Frequency adjustment:</u> == | |||
== The first step in the process of adjusting the frequency to the desired value was carefully grinding off surplus material and weight at the ends of the quartz bar until a frequency was reached which, after [[Image:Image012.jpg|thumb|right|280px]]evacuation of the case, resulted in exactly 2<sup>13</sup> Hz . A very difficult and tedious job indeed. A fourteen stage divider chain would bring this frequency down to exactly one half of one Hz required to drive the stepping motor. – What about aging and other disturbing effects afterwards and during wear? To take care of this a fine tuning mechanism was needed. A stepwise variable capacitor was hooked up in series with the quartz to change the oscillating frequency of the quartz assembly by 0.2 sec/day upward or downward (picture). At that time Fritz Leuenberger of the semiconductor department started his research on MOS transistors, an excellent chance to integrate on a single chip a series of discrete MOS capacitors, high value and small volume, exactly what we wanted. The design and layout was made in 1966 by Armin Frei, the semiconductor department delivered the MOS capacitor chip and the watch maker technician Claude Challandes designed the miniature switch.<br><br><u>5.5 Temperature compensation:</u> == | |||
== The irregularities in time keeping of quartz wrist watches are due to the temperature sensitivity of the various physical parameters of the quartz crystal itself and not of the electronics attached to it. The deviation [[Image:Image014.jpg|thumb|right|220px]]in time is measured in se- conds per day as a function of temperature. The resulting plot, usually displayed between 4° C and 36° C is a com- plicated function of the cutting angels relative to the axes of the quartz crystal itself. At the time it was well known, that the x-y cut quartz crystals exhibit parabolic curves according to the curve a) in the picture. We were very much aware that the manufacturers of the current me- chanical watches were keen to keep the temperature deviation as small as possible, so we engaged strongly in the disciplines of temperature compensation. First inves- tigations using temperature sensitive resistors and capa- citors were not very successful. Jean Hermann proposed in 1967, shortly before the very first quartz wrist watch was operating, a scheme using the parabolic behavior twice and a switch to connect a compensating capacitor at 12°C ac- cording to curve b). This scheme was easy to implement, produced favorable results with the observatory tests and was effectiv during daily usage. Yet it required extensiv interventions by the laboratory director on behalf of the department head in order not to drop the brilliant idea. The resulting Thermo Compensation Module (TCM) was developed by Jean Hermann and was implemented using MOS technology by Fritz Leuenberger and his group in 1967. <br><br><u>5.6 Frequency divider:</u> == | |||
== Since the very beginning of the quartz wrist watch project Armin Frei decided that the oscillating frequency of the quartz had to be to 2<sup>13</sup> = 8192 Hz. Consequently for Beta 1 using a stepping motor to drive the second hand, see below, a [[Image:Image016.jpg|thumb|right|180px]]total of 14 binary flip-flop stages were required to drive the motor of the watch with pulses of half a Hertz repetition frequency. The flip-flops which were finally incorporated in the Beta 1 prototypes (picture), were designed by Jean Fellrath and implemented in integrated form by Raymond Guye and his group in bipolar technology. The circuits were optimized for low power consumption of approximately one microamp per stage and for high operational stability. While integrated counter stages at microwatt level existed since before, the flip-flop shown exhibits a new type of triggering circuits, one of them on each side, and it also uses direct coupling between collectors and opposite base similar to the oscillator circuits, see above. - In bipolar technology the divider chain with 14 stages represented a big load for the battery, limiting its lifetime to less than one year, much to the concern of certain members of the governing board. The immediate solution to that would have been a reduction of the number of divider stages, say 5, reduce the frequency to 256 Hz rather than half a Hz and use a vibrating motor rather then a stepping motor. This was alternatively proposed with the Beta 2 and the Beta 21 projects 1967 and 1968 respectively. - However the real solution to the power consumption problem were ultimately the Complementary Metal Oxide Silicon circuits (CMOS), which had been invented in USA (Frank Wanlass, 1963) a couple of years before. <br><br><u>5.7 Stepping motor:</u> == | |||
== All that was left was to convert the pulses appearing at the end of the divider chain into step by step advancements of the second hand. Jean Hermann and François Niklès proposed in 1967 a simple solution of such a stepping [[Image:Image018.jpg|thumb|right|150px]]motor for Beta 1 as seen in the picture. The setup incorporated an anchor wheel and an anchor which were responsible for the go and stop of the second hand. Much different as in mechanical watches, where the anchor and the anchor wheel act as escapement, here the anchor drives the anchor wheel by wiggling forth and back. The anchor was activated by means of a bobbin coil, which was attached to the anchor and which was magnetized by bipolar electric pulses. The duration and the amplitude of the pulses had to be well controlled to warrant proper operation and to save battery power. The duration of the pulses were derived from the pulse pattern appearing along the divider chain using boolean logic. The concept and the basics had been worked out as early as 1966. == | |||
== <br>'''<u>Beta 2</u>'''<br><br><u>Autonomy:</u> The office of the governing assembly had little sympathy for the new direction with the quartz wrist watch. They were hoping for an electronic watch, exhibiting at least one advantage compared with existing electronic watches, remember the strategy and here was a watch with a battery life time of less than one year! One representative of the office, not very familiar with electronics and its progress made it a must: Life time had to be equal or longer than one year.<br><br><u>Wrong strategy:</u> This in turn was the starting point of a fatal new strategic direction: A current saving alternative project, called Beta 2, was proposed and initiated in November 1966 by Max Forrer. Beta 2 was using the same oscillator like Beta 1, but had only five flip-flops instead of fourteen as Beta 1. With five stages, a frequency of 256 Hz was reached, just right to drive one of Henri Oguey's vibrating motors. Battery life time was above one year. The first Beta 2 calibre was assembled in August 1967, one month later than Beta 1. <br><br><br>'''<u>Beta 21</u>'''<br><br><u>Industrial Version:</u> After the splendid celebration of the outstanding results reached with Beta 1 and Beta 2 following the observatory tests, on February 15, 1968 investigations on how to establish a technology transfer from the prototypes towards a product started immediately. First it was decided to favor Beta 2, not Beta 1. This decision was commented by Henri Oguey and Henri Schneider simply by: "Au vue de l'expérience aquise sur les prototypes, seul le system Bêta 2 entre en ligne de compter pour assurer une durée de vie de la pile supérieur à un an." The industrial calibre was named Beta 21, consisted of the 8192 Hz oscillator, a five stage binary divider chain and the vibrating motor at 256 Hz. On April 10, 1970, during the "Foire Suisse de Bâle", four months after Seiko's Astron SQ35, some 20 Swiss firms launched their products all incorporating the Beta 21 movement. <br><br><u>Strategic error:</u> Beta 21 turned out to be a flop. Some 6000 units were sold, then the line with the vibrating motor was discontinued. To solve an intermediate power problem (autonomy) by pushing the vibrating motor version turned out to be a severe strategic error with a number of consequences. Indeed, the power problem was not a systems problem but a semiconductor problem. Using Complementary Metal Oxide Semiconductors (CMOS) circuit technology instead of bipolars (BIP) would have saved exceeding amounts of energy by orders of magnitude. CMOS was invented in 1963 and was applied later on throughout the watch industry. The CEH started early on MOS, however produced stable CMOS with silicon gate technology at a voltage level of 1.35 V only after 1972. Instead of pushing research after the splendid success at the observatory CEH engaged excessively in technology transfer and manufacturing, losing lead time and research competence. – Not enough of all that, with their Astron 35SC the Japanese proved feasibility of an industrial watch with a stepping motor and battery lifetime of over one year.<br> <br><br>'''<u>Chronology of the first quartz wrist watch</u>'''<br><br>May 7, 1965: Armin Frei and Rolf Lochinger started the quartz wrist watch project. Frei proposed to concentrate on quartz oscillators suitable for wrist watches and Lochinger proposed to investigate into divider circuits and drivers. Although the strategy did not plan for any such high frequency project Frei and Lochinger started their investigations immediately. <br><br>November 26, 1965: Roger Wellinger, director CEH declared the "montre-bracelet à quartz" to be a strategic goal. This was a consequence of the availability of a miniaturized quartz oscillator with low energy consumption and with a size smaller than 27 mm by Frei. This is the first time that the "montre bracelet à quartz" was mentioned in a strategic document of the CEH. The task was formulated as: "Construction, mise au point et evaluation du prototyp du calibre".<sup>7)</sup> <br><br>July, 1967: The first fully operational and complete quartz wrist watch, probably worldwide, had been assembled by Jean Hermann, François Niklès et al. The watch Beta 1 consisted of a 8192 Hz quartz oscillator, a fourteen stage divider stage and an electro dynamic stepping motor. Testing of the temperature compensation module during five days demonstrated proper operation. With this the CEH disposes of a reliable documentation of the priority of the first quartz wrist watch worldwide.<sup>3)</sup> <br><br>August 13, 1967: About one month later, the first quartz wrist watch Beta 1 with connotation CEH-1020 was delivered to the Observatory of Neuchâtel and was immediately submitted for tests in the category of "Chronomètres-bracelet". The resulting number of classification was 0.189, a value at least one order of magnitude better than the other mechanical competitors in the same category. Later on, a total of ten other quartz wrist watches from CEH (Beta 1 and Beta 2) were delivered for tests. On November 11, 1967 a total of four quartz wrist watches from Seiko, Japan were submitted. The average number of classification of the four best watches of either CEH or Seiko were 0.166 and 0.873 respectively.<sup>5)</sup> With this the CEH watches were winners in time and quality - at least in 1967. The differences in the number of classification within the group of the CEH watches result from the thermo compensation and its adjustments and not from the difference of systems.<br><br>August 1967: The first calibers Beta 2 were completed. They were submitted for tests at the observatory on September 17, 1967. Beta 2 consists of the 8192 Hz oscillator, a five stage divider and a vibrating motor. In 1968 it had been decided to start an industrial version based on the Beta 2 principle for reasons of battery lifetime.<br> <br>December 26, 1969: Introduction of Seiko's Astron SQ35 on the market, the worldwide first commercial quartz wrist watch. Seiko's watch was nice, had a battery lifetime of more than a year, even though it had a stepping motor, and its electronics were based on a new hybrid technology incorporating some 200 parts. <br><br>April 10, 1970: Introduction of the Beta 21 caliber, the industrial version of Beta 2, marketed by some 20 Swiss watch manufacturers during the "Foire Suisse de Bâle". <br> == | == <br>'''<u>Beta 2</u>'''<br><br><u>Autonomy:</u> The office of the governing assembly had little sympathy for the new direction with the quartz wrist watch. They were hoping for an electronic watch, exhibiting at least one advantage compared with existing electronic watches, remember the strategy and here was a watch with a battery life time of less than one year! One representative of the office, not very familiar with electronics and its progress made it a must: Life time had to be equal or longer than one year.<br><br><u>Wrong strategy:</u> This in turn was the starting point of a fatal new strategic direction: A current saving alternative project, called Beta 2, was proposed and initiated in November 1966 by Max Forrer. Beta 2 was using the same oscillator like Beta 1, but had only five flip-flops instead of fourteen as Beta 1. With five stages, a frequency of 256 Hz was reached, just right to drive one of Henri Oguey's vibrating motors. Battery life time was above one year. The first Beta 2 calibre was assembled in August 1967, one month later than Beta 1. <br><br><br>'''<u>Beta 21</u>'''<br><br><u>Industrial Version:</u> After the splendid celebration of the outstanding results reached with Beta 1 and Beta 2 following the observatory tests, on February 15, 1968 investigations on how to establish a technology transfer from the prototypes towards a product started immediately. First it was decided to favor Beta 2, not Beta 1. This decision was commented by Henri Oguey and Henri Schneider simply by: "Au vue de l'expérience aquise sur les prototypes, seul le system Bêta 2 entre en ligne de compter pour assurer une durée de vie de la pile supérieur à un an." The industrial calibre was named Beta 21, consisted of the 8192 Hz oscillator, a five stage binary divider chain and the vibrating motor at 256 Hz. On April 10, 1970, during the "Foire Suisse de Bâle", four months after Seiko's Astron SQ35, some 20 Swiss firms launched their products all incorporating the Beta 21 movement. <br><br><u>Strategic error:</u> Beta 21 turned out to be a flop. Some 6000 units were sold, then the line with the vibrating motor was discontinued. To solve an intermediate power problem (autonomy) by pushing the vibrating motor version turned out to be a severe strategic error with a number of consequences. Indeed, the power problem was not a systems problem but a semiconductor problem. Using Complementary Metal Oxide Semiconductors (CMOS) circuit technology instead of bipolars (BIP) would have saved exceeding amounts of energy by orders of magnitude. CMOS was invented in 1963 and was applied later on throughout the watch industry. The CEH started early on MOS, however produced stable CMOS with silicon gate technology at a voltage level of 1.35 V only after 1972. Instead of pushing research after the splendid success at the observatory CEH engaged excessively in technology transfer and manufacturing, losing lead time and research competence. – Not enough of all that, with their Astron 35SC the Japanese proved feasibility of an industrial watch with a stepping motor and battery lifetime of over one year.<br> <br><br>'''<u>Chronology of the first quartz wrist watch</u>'''<br><br>May 7, 1965: Armin Frei and Rolf Lochinger started the quartz wrist watch project. Frei proposed to concentrate on quartz oscillators suitable for wrist watches and Lochinger proposed to investigate into divider circuits and drivers. Although the strategy did not plan for any such high frequency project Frei and Lochinger started their investigations immediately. <br><br>November 26, 1965: Roger Wellinger, director CEH declared the "montre-bracelet à quartz" to be a strategic goal. This was a consequence of the availability of a miniaturized quartz oscillator with low energy consumption and with a size smaller than 27 mm by Frei. This is the first time that the "montre bracelet à quartz" was mentioned in a strategic document of the CEH. The task was formulated as: "Construction, mise au point et evaluation du prototyp du calibre".<sup>7)</sup> <br><br>July, 1967: The first fully operational and complete quartz wrist watch, probably worldwide, had been assembled by Jean Hermann, François Niklès et al. The watch Beta 1 consisted of a 8192 Hz quartz oscillator, a fourteen stage divider stage and an electro dynamic stepping motor. Testing of the temperature compensation module during five days demonstrated proper operation. With this the CEH disposes of a reliable documentation of the priority of the first quartz wrist watch worldwide.<sup>3)</sup> <br><br>August 13, 1967: About one month later, the first quartz wrist watch Beta 1 with connotation CEH-1020 was delivered to the Observatory of Neuchâtel and was immediately submitted for tests in the category of "Chronomètres-bracelet". The resulting number of classification was 0.189, a value at least one order of magnitude better than the other mechanical competitors in the same category. Later on, a total of ten other quartz wrist watches from CEH (Beta 1 and Beta 2) were delivered for tests. On November 11, 1967 a total of four quartz wrist watches from Seiko, Japan were submitted. The average number of classification of the four best watches of either CEH or Seiko were 0.166 and 0.873 respectively.<sup>5)</sup> With this the CEH watches were winners in time and quality - at least in 1967. The differences in the number of classification within the group of the CEH watches result from the thermo compensation and its adjustments and not from the difference of systems.<br><br>August 1967: The first calibers Beta 2 were completed. They were submitted for tests at the observatory on September 17, 1967. Beta 2 consists of the 8192 Hz oscillator, a five stage divider and a vibrating motor. In 1968 it had been decided to start an industrial version based on the Beta 2 principle for reasons of battery lifetime.<br> <br>December 26, 1969: Introduction of Seiko's Astron SQ35 on the market, the worldwide first commercial quartz wrist watch. Seiko's watch was nice, had a battery lifetime of more than a year, even though it had a stepping motor, and its electronics were based on a new hybrid technology incorporating some 200 parts. <br><br>April 10, 1970: Introduction of the Beta 21 caliber, the industrial version of Beta 2, marketed by some 20 Swiss watch manufacturers during the "Foire Suisse de Bâle". <br> == |