The mystery of the 16kHz Astron

Introducing “Off Topic”

As we approach the third anniversary of “The Grand Seiko Guy” going live, one of the challenging issues that we have decided to address is that regarding the breadth of the content presented on the website.

Up until now, we have resolutely stayed laser focused on the subject of vintage Grand Seiko. However, as anyone who has an interest in these watches will know, Seiko do make it rather difficult to remain fixated on just one small part of their history!

Your author fell off the proverbial wagon almost as soon as he got behind the reins, and now has a rather alarmingly large collection of Seiko quartz watches from the 1970’s, along with a few other choice collectibles including (and more on this to come at a later date) a very interesting vintage dive watch.

We have undertaken extensive research into the vintage Grand Seiko era, occasionally stumbling onto topics outside our usual focus. We believe that some of the things that research has uncovered may well be of interest to those that visit the site, and possibly even a wider audience. Some of the results of this research has already been covered at a high level on Instagram, but that really isn’t the place to really dig deep on subjects, so – given we’re not really interested in setting up another website just for the sake a few articles a year – we thought it would make sense to simply add a new “Off Topic” category to our article section, and post them here. Also, following a poll on our Instagram page, it seems that there is considerable interest in adding some articles to the site regarding the modern era of Grand Seiko – those musings will in time also be posted under the same category.

So, without further ado, let’s kick off with the first article. And it’s one that we think may well re-write history. Or at the very least, set it back on the correct timeline…

Double or nothing

NOTE – This article has been updated following examination of newly discovered source material, which turns the original article on its head!

Everyone knows the headline to the story of the introduction of the world’s first quartz wrist watch – the Seiko Astron, launched on Christmas Day, 1969.

Seiko Astron (c) Seiko

Beyond that though, there seem to be almost as many versions of what happened next as there were Astrons manufactured in November 1969. The purpose of this article is not to address all the inaccuracies that are out there, but rather to challenge one particular claim that has always struck your author as rather odd.

And that claim is that the frequency of the crystal resonating away in the 35SQ/SQC movement was doubled from 8,192Hz to 16,384Hz either (as some sources – including Seiko themselves state) “at some later time”, or, when the 35SQC (the Quartz Astron with added date complication) was introduced.

Before even looking at any evidence, this claim should immediately raise some questions. A change in the frequency of whatever oscillator is regulating the timing of a watch movement is a big thing. The first question that needs to be asked is “why?”. Why would Seiko do this? Well, there is an obvious answer to that, and as we will see, when Seiko did introduce a watch whose frequency was double that of the Astron, they explained the benefits. The second question is that if they did do this to the 35SQ and/or 35SQC movements – why on earth did the movement numbers not change?

If you double the frequency of the “clock” running your watch, you are going to have to make some other fundamental changes to the circuitry, because if you don’t, it’s not going to be very accurate. It’s not simply a case of dropping in a faster oscillator – the circuitry that takes those oscillations and turns them into one cycle per second needs to be altered as well.

Remember. This is Seiko we are talking about here. This is a company that introduced new movement caliber numbers simply because the physical size of the battery powering the movement changed. Can anyone find us another example from Seiko history of a caliber number not being updated when the frequency of the oscillator changed?

Seiko Sales, January 1971

Barely 12 months after the introduction of the Astron, we see the cover of the January 1971 issue of Seiko Sales proudly showing Seiko’s latest quartz watch – the 36SQC 011.

Now, disregarding the Astron for a brief moment, this is an extremely interesting watch in and of itself for a number of reasons – not the least being – can you find one? We are not aware of any examples of this reference ever coming to the market in recent years, and know of no examples in any collection (save the Seiko Museum).

The reason why the introduction of this watch – historic in its own right for being the first watch to utilise a MOS chip – is so important to us for the purposes of this article, is that the inside pages of the Seiko Sales magazine compare it to the 35SQ and 35SQC Astron –

Remember that this is an article published just one year after the Astron first went on sale, and you don’t even need to be able to read Japanese to understand the importance of what is laid out in the table on page 10.

Firstly, this contemporaneous source very clearly states that both the 35SQ and 35SQC movements had a crystal oscillator running at 8,192Hz. Just this single piece of evidence should be enough to immediately disprove any claims that the 35SQ ran at 8kHz, and the later 35SQC at 16kHz.

The article in the publication actually spans 6 pages, but for now, we won’t get bogged down into the rest of it because we’re just here to prove one thing.

Earlier we posed the question “why would Seiko seek to double the frequency of the crystal oscillator”? Well, the answer to that is in the table too – as we can see, the 36SQC caliber boasted even better accuracy than the Astron – almost double in fact, being quoted as accurate to +/-0.1 seconds per day, or +/- 3 seconds per month.

If Seiko did for some reason go to all the trouble of doubling the clock frequency of the 35SQ and 35SQC movements “later”, isn’t it a rather odd thing to do and not let people know the accuracy of the watches had improved 100%? Alternatively, if they doubled the clock frequency but the accuracy didn’t improve, then what was the purpose in doubling the clock frequency?

Now that we have evidence that both 35-series movements were originally clocked at 8kHz, one has to ask where the story of the 16kHz Astron came from.

When we first published this article, we speculated that the publication we are about to detail next was the one responsible for the “false” claims that the 35SQC movement ran at 16,384Hz. Less than 24 hours later, we discovered another article in our archives that actually explains once and for all the history of the 35SQ and 35SQC movements. We’ll get to this new article later, but we still feel that the source of the claims for the 35SQC movement running at 16,384Hz come from the below

Seiko Sales, December 1971

Fast forward to the end of the year, and we have another issue of Seiko Sales, introducing yet another quartz calibre.

It didn’t feature on the cover, but this issue introduced the 38-series of quartz watches. Seiko were not hanging about back then – just two short years after the introduction of the world’s first quartz watch, and Seiko were already on their third generation.

Once again, we’re not going to share the full article on the 38-series here, because what we need to focus on is on just one of the pages –

We believe that this article is the original source for every claim out there for the 35SQC (which of course was introduced “later” than the “earlier” 35SQ) having a crystal oscillating at 16,384Hz.

And if anyone were to cast a cursory eye over the above text, one wouldn’t need to understand any Japanese to see that yes, indeed, the article states very clearly that the 35SQ ran at 8,192Hz, and the later 35SQC ran at 16,384Hz.

But you would need to understand Japanese to know that there is a typographical error in the article.

Whilst it states “35SQC … 16,384Hz” in the 6th line, if you know a little about the history of these early quartz watches, your suspicions should have been raised immediately by what is written in the line above this – “MOS-IC”. Those of you paying attention will recall that we mentioned something about MOS chips earlier on. And that the first watch movement to utilize a MOS chip was the 36SQC.

So, even without translating the Japanese, alarm bells should be ringing here. Why does that term “MOS-IC” appear immediately preceding an apparent discussion of the 35SQC?

Let’s turn to Google for the translation of the important bit.

“With calendar device, using MOS-IC, 35SQC (Crystal Oscillation 16,384Hz).”

And there we have it. The article cannot be referring to the 35SQC when talking of a watch with a crystal running at 16kHz, because the 35SQC did not have a MOS integrated circuit.

Ergo, it is a typo. Someone, all those decades ago, made a simple little typo in an article, and ever since then, the history of Seiko quartz watches has been incorrectly recorded.

Or so we thought. Until we unearthed an article in our archives from an internal Seiko publication that was published between the two issues of Seiko Sales detailed above, in August 1971.

And this article is the missing link that solves the “mystery” once and for all.

Suwa Seikosha internal company magazine, August 1971

The monthly Seiko Sales (originally called Seiko News) magazines that Seiko distributed to its retailers from the late 1950’s through to the mid 1970’s are extremely hard to get hold of these days. We have been purchasing them to add to our reference library for over 5 years now, and are still missing 42 issues. Of the 170 issues we do have, probably more than half of them we have only ever seen a single copy come up for sale.

But those commercially distributed magazines are commonplace compared to the internal magazines that were only distributed within the factories. So we count ourselves extremely fortunate to have this one, given the importance of the article contained within it.

On page 20, there is an article that features a large photograph of the stainless steel “Astron” – the 35SQC 020.

Google Translate works best when you drip-feed it text to translate, rather than trying to get it to do a whole page at a time. So after a little effort, we have managed to get what we believe is a fairly accurate translation of the article, and it is very revealing.

“Popular in the market – 100,000 Yen level 5 [?] crystal electronic wristwatch

Seiko Quartz 35SQ
Seiko Quartz 35SQC

It has been three months since the “Seiko Quartz 35 SQ” and “Seiko Quartz 35 SQC” models of crystal electronic wristwatches were released. Its high precision and elegant design on the market have captured the hearts of many consumers.

The new crystal-type electronic wristwatch is a new model of the gold-side [?] crystal-type electronic wristwatch “Seiko Quartz 35SQ ” ​​”Seiko Quartz 35SQC”, which was released by our company at the end of last year. With a stainless metal band.

The standard retail prices are 175,000 yen and 185,000 yen (with calendar).

Looking at the movement, the new model is 16,384 Hz.

Although the crystal frequency was first released, it has twice the high vibration of 8,192 Hertz and has high stable accuracy. In addition, the frequency divider circuit (circuit that divides the vibration) has a hybrid. Instead of ICs (integrated circuits), MOS (metal oxide film) type ICs are used, and the main features are that the size has been significantly reduced and the power consumption has been reduced.

Other mechanisms have not changed, and the portable accuracy is within plus or minus 3 seconds per month at room temperature, with an average daily difference of 0.1 second. Within one second, the drive system is a step motor type, the battery life is one year or more with one battery, and the manufacturer’s warranty period is two years.”

As we know, the 36SQC was the first Seiko watch to have a MOS integrated circuit. Earlier in this article, we commented on how it was very strange to read in the Seiko Sales December 1971 issue the 35SQC being described as having a MOS-IC, and speculated that it was a typo. We still believe that to be the case, since that article only mentioned the SQC, and not the no-date SQ model.

However, the bombshell revelation in this new article is not just that both the 35SQ and 35SQC were indeed – as Seiko and other sources have claimed – updated to 16,384 movements, but that those updated movements – introduced in May 1971 with the launch of the steel cased “Astrons” – were based on the new MOS-IC technology.

There is a slight question mark over the exact interpretation of the newly discovered article, and we suspect it would need a good (human) translator to clarify. The article makes reference to the references being launched “three months” ago, yet also states “that was released by our company at the end of last year”, and then adds “with a stainless metal band”.

One possible interpretation is that the “end of last year” comment is referring to the 18K gold cased 35SQC, with the steel cased watches being released three months prior to the article being published (i.e. in May 1971). Alternatively, perhaps it is the steel cased watches that were launched at the end of 1970, and this article is specifically referring to a movement upgrade for those steel watches in May 1971.

We suspect it is the former of those two options, but this specific detail does require more investigation.

As always when discovering something new, there are often more questions raised than are answered. And here are just a few more for…

Is the chip in these watches was the same one said to be used in the 36 SQC, which was supplied by a US company, or the later one that we see in the 38-series quartz watches that was produced in-house?

How were Seiko able to provide an acceptable battery life with these references running off just one battery, whilst the 36SQC required two?

If your 18K gold cased Astron was sent in for service, would it have its core movement parts swapped out for the newer MOS-IC based caliber?

With reference to the last question, what is very interesting is that it would seem the crystal and circuit block from the later steel cased references are hot-swappable with the movement of the original 18K Astron.

Those interested in more a detailed discussion regarding the construction of these watches are encouraged to read two of the articles found over on The Watch Bloke‘s website.

In the first, dating from June 2017, we can see a detailed treatise on the servicing of both 35SQ and 35SQC 18K gold Astrons, where unfortunately only one of the movements was able to be revived.

Six months later, and parts to revive the stricken 35SQC movement are provided, and those parts bring the watch back to life. We believe from talking with the collector who owned these watches that the parts used came from a steel 35SQC 020.

Parting thoughts

[left verbatim from the original article, with edited comments in parentheses]

Yes – this is of course Seiko we are discussing here, and occasionally when looking back 50 or 60 years into their history, there are things that with hindsight may seem a little odd.

But because we are used to Seiko seemingly doing odd things once in a while, we can fall into the trap of assuming that when we see something really odd, and “that something” has been reported over and over again across countless sources, we discard our natural inquisitiveness, disregard the nagging inner voice telling us “hmm, that doesn’t seem right”, and just shrug it off.

After all, Seiko gonna Seiko, right?

Sure, it does seem a little strange for them to double the clock speed of the world’s first ever quartz movement, without changing the caliber number, and without actually mentioning it at the time. But Seiko report it as being the case, as do some well respected Japanese sources, and it’s been discussed to death already on all the forums. So we just accept it as being true.

Your author doesn’t think it is. [edit – turns out he was wrong, and it is!]

Your author believes the evidence presented in the above two articles is pretty conclusive. [edit – which it was, but it was incomplete evidence]

Is it possible there is more evidence out there to uncover – evidence that disproves the above, and that actually does prove that the 35SQ/SQC movements were upgraded with 16kHz oscillators at some point in their lives? Yes – absolutely it is possible. After all, this is Seiko…

But let’s see the evidence please. Something tells me that Internet searches won’t suffice here.

[edit – Mea culpa. We provided the evidence to correct our original errors, although we rather suspect that the article we discovered in the internal Suwa Seikosha magazine wasn’t on the internet prior to today!]

“Seiko,” it seems, after all, really do “gonna Seiko.”