There is a sense that we may just be starting to see a shift away from the obsession with giant watches. For a while now, the dinner plate on the wrist fitted with a 24mm strap seemed de rigueur but slowly it seems to have dawned, at least in some circles, that the over-large watch is just silly. There is good reason why the Rolex submariner has remained at 40mm for the past 50 odd years – it’s large enough to serve its intended purpose but not so large to get in the way. I don’t actually have a big problem with watches larger than 40mm but if it is to be wearable then the proportions start to matter much more than they do in a smaller watch. If it becomes too thick, then it won’t fit under a shirt sleeve; if it’s too long, which sometimes happens when classic designs are scaled up to appeal to modern tastes, then the case lugs can protrude over the edge of your wrist and frequently the result can be a watch which sits uncomfortably on the landscape of your wrist. My own take on this is that the sweet spot should be somewhere between 36 and 38mm.
Today’s little project is just that. At 36.5 mm across the beam, inconceivably small for a men’s divers watch by modern standards but somehow its compact size is an essential part of its charm. I bought it on a whim, as is my wont, taken in by the lovely dial and handset, the downwards swoop of the lugs
Photo credit: Ebay seller
and a lovely ETA 2472 automatic movement surrounded by a caseback o-ring of impressive girth, swimming in grease but hardened and no longer fit for purpose.
Swiss watches don’t get much of a look-in on this blog so let’s take a break from the constant stream of vintage Japanese watchery and take a look under the hood of an example of the 1960’s Swiss competition.
Once it had landed, an initial appraisal suggests plenty of promise here; in fact it looks really rather nice and, in spite of those slightly ropey Ebay pics, this is not going to be one of those duck-to-swan transformations. Of course a presentable exterior doesn’t necessarily mean a routine scrub and brush up awaits. Let’s take a closer look at the movement:
Two things strike me immediately: it’s really a very good looking movement, with the quality of finishing standing out as a cut above. I love the shape of the cut-out around the winding weight bearing, the exposed pawl wheels on the auto winding mechanism and the fact that the balance is partially hidden, the incabloc setting peeking through a beveled hole in the winding bridge. It’s pretty grubby mind, which suggests a service is timely. Turning it over, and we get a better look at an absolute peach of a dial:
What’s not to like about this? The raised, chromed markers set against a deep black gloss dial, crisp silvered printing and wonderfully aged, golden tritium lume. The hands too are perfectly judged with elegant fluted ends to the hour and minute hands, a sweet little meatball perched on the end of the seconds hand. All three hands also curve downwards towards their ends following the downward curve of the convex dial.
Removing the hands required some care because the hour hand was sat very close to the dial and I had to use chronograph hand levers to gain some sort of purchase. With that achieved and the dial removed, we can take a look at the top and bottom of the movement prior to disassembly:
Those pawl wheels on the autowinding bridge look grubby but no more so than does the rest of the movement: the grime though lends a rather pleasant patination to the movement, giving it an appearance which reflects what it is: somewhat old, and in need of a little attention.
Removing the auto winding mechanism provides a better view both of the movement proper but also if the complexities if the autowinder itself:
You can see from the image on the left that the winding framework is secured to the winding weight by a screw fixed from the bottom of the mechanism rather than from the top which was the approach used by Seiko in their automatic movements from the same period.
Breaking down the movement from this point was straightforward (it’s rarely taxing to take something to pieces – it’s the putting it back together which can be problematic). Starting with the top,
we see the barrel and crown wheel are supported by their own bridge with the train wheels accommodated by their own dedicated bridge too. A comparable movement of the time from Seiko is the 6217/6205 which used a single bridge for barrel and train wheels
although Seiko did use separate train wheel and barrel bridges in earlier designs from the decade such as in the 7606 (see here).
The calendar side also reveals quite a different design philosophy to the Seiko approach with the cannon pinion forming an integral part of a driving wheel which sits loosely on a fixed centre pipe, with the driving wheel itself driven by the pinion at the end of the intermediate wheel located on the other side of the movement.
On the Seiko designs, the cannon pinion fits tightly onto the shaft of the centre wheel protruding through from the reverse of the movement and with the centre wheel supported within a jeweled bearing. The other neat feature of this ETA movement is the instantaneous date change mechanism but this is achieved with some perceptible mechanical preamble which also appears to affect the attitude of the driving wheel and cannon pinion, with the minute hand rising perceptibly in the 9 hours prior to date change compared with its height in the preceding 15.
Overall though, this is a sophisticated, very nicely executed design, with an air of quality that is perhaps a notch above that of the mid-tier Seiko movements of the time. I’ve no idea though what the price point of the 2472 would have been at the time compared with, say, a 6205.
The Incabloc shock protection on the balance ends is another very nice design, somewhat easier to engage and disengage than the Diashock setting used by Seiko
You can also see that the spring is held in place by a separate fitting on the lower setting seen here. This is fine but reassembling the whole thing is a bit of a fiddle (as I’ll discover later on).
Back to the auto framework to remove the pawl winding wheels and driving and reduction gears
The leftmost of the two pawl winding wheels mounted to the top of the framework has a pinion on its lower side which transmits motion to the reduction gear (the smaller of the two in the image on the left above). Each of the two pawl winding wheels is composed of two toothed disks, the upper of which meshes with the bearing wheel on the winding weight (shown earlier above) and the lower of which meshes with the corresponding disk on the neighbouring pawl wheel. The upper and lower disks of each wheel are connected by a springless pawl which is mounted in such a way to ensure that the wheel without the pinion is locked if the winding weight turns counter-clockwise while the other wheel is locked when it turns clockwise. This arrangement ensures that the reduction wheel always turns in the same direction no matter which direction the winding weight is turning. It also means that when the movement is wound manually, that motion transmits only as far as the pawl winding wheels, which then do not lock and do not transmit the winding action as far as the winding weight. An ingenious mechanism but a complex one requiring the pawl wheels to be meticulously clean and correctly lubricated. Speaking of which, we’re ready now to clean the parts prior to reassembly.
First step, post clean, is to refit and oil the lower Incabloc setting:
followed by the escape wheel cap jewel and keyless works
Thus far, we’ve been lulled, perhaps, into a false sense of security, no obvious flies in the ointment. However, when I got around to checking out the barrel, prior to reassembling the top of the movement, I discovered a problem:
The wall of the barrel is cracked and distorted and the barrel lid does not then sit tightly enough to properly secure the mainspring within the barrel. This is an old movement and the barrel is obsolete and not available either from Cousins or Jules Borel. Fortunately, I managed to find a new one from another watch parts supplier in the UK and together with a fresh mainspring from Cousins, saved the day.
But then another problem: the diameter of the new mainspring constrained within its holding disk is slightly larger than that of the new barrel inner wall which means that I cannot simply push the mainspring into the barrel. Instead, I had to extract it from the disk (carefully) and then wind it into the barrel by hand.
With that hiccup overcome, the barrel can assume its correct position back in the mainplate and we can resume rebuilding the rest of the movement. First up then, the barrel and bridge, followed by the intermediate wheel, train wheels and bridge:
Make sure everything moves freely before fitting the crown wheel, ratchet wheel, pallet fork and pallet fork bridge.
All seems fine, so flip it over and reassemble the calendar side of the movement
On the home straight now, so on with the dial and hands, having established that the calendar works properly
ready for re-housing within the case, following nothing more than a clean and a fresh armoured acrylic crystal.
You may have noticed two differences compared with where we started. The first is that I’ve fitted a crystal without a cyclops. I did actually try to find a like-for-like replacement but the case requires a 29.4mm crystal and the only crystals I could find with cyclops which correctly aligned with the date window were the 29.3 and 29.5mm crystals (in spite of the fact that the position of the cyclops relative to the centre should have been the same on all). In the end I decided to abandon the cyclops in the interests of fitting a crystal capable of preventing water ingress! The other difference is the crown. I am not sure whether the crown which came with the watch was original but it appeared to have no o-ring fitted and sat loose on the case tube, providing no water resistance at all. I tried three or four different crowns before settling on the one you see above.
The other problem needing solving was that of the case back gasket. The original was hardened and unfit for purpose but was also of a non-standard size, being flat, 2mm thick and with an inner diameter of 30.5 mm. Nothing in Cousins’ catalogue came close to fitting and so I turned instead to a supplier of nitrile o-rings intended for some other application (plumbing perhaps?). I settled on a circular profile o-ring, 1.5mm thick and with an inner diameter of 31mm. With the case back screwed down the compression of the o-ring would fill the gap nicely providing a more than adequate seal.
We are ready now to re-fit the automatic framework, a quick look at those ingenious reversing pawl wheels
before re-attaching the framework to the winding weight
and then attached the whole lot to the rest of the movement.
Refit the case back and bezel and we are pretty much there.