My brother seems to have a knack for giving me watch projects with unforeseen new challenges. This Seiko from the 1970s had become the latest example. Seeing it on an online marketplace, he was somehow drawn to its striped dial and very square features, and decided to buy it. Upon arrival, however, it turned out to be in a rather poor state so he asked if I could give it some attention.
With a square case reminiscent of a TV set, Seiko’s design inspiration clearly came from the popular Rado Manhattan and NCC ranges from the same period in the 1970s.
From the outside the watch was fairly dirty, with scratches on the case and crystal and gold plating that had worn through. The plated bracelet in particular was doing a good job of shedding its coating of gold. My brother casually asked if I’d be able to re-plate everything – I had to lower his expectations as I don’t have the equipment, space or skills to do my own re-plating just yet!
Turning to one of the things I could actually assist with – the mechanical repairs, I noticed a knocking noise when the watch was shaken. Perhaps an issue with the automatic winding rotor? The movement did appear to be running after a few shakes, giving some hope that the situation inside wasn’t seriously bad.
Opening it up
The first challenge was to get it open. The square case felt pretty solid, and it didn’t have a traditional case back, which would have been screwed down or pressed on. Instead, hidden by the bracelet was a pair of wide slots. My Seiko case opener just so happened to have a blade with the same width… Pushing it in forced some retaining springs to withdraw, which freed the inner case.
The old crystal was then removed – this was just sitting on top of the dial on a rubber seal.
Now, how to remove the movement from the inner case? Without access to the rear, presumably the only thing holding it in was the stem and crown. So how was I meant to access the stem release button?
The secret lay in the small cutout near the top corner of the dial – a pointy tool inserted here would push against a hidden lever that releases the stem, allowing the movement and dial to come out as one assembly. But this neat trick wasn’t required as the dial feet were broken and the dial simply lifted away!
I was fairly confident that the loose dial was the cause of the knocking noise heard earlier. In a way it was great as it meant the automatic rotor was likely to be fine, but it was also bad as I currently don’t have a good way of fixing broken dial feet. Hmm, this issue could prove to be a challenge.
With the dial out I could now see the movement in the inner case, as well as the small lever to release the stem. This isn’t a view a repairer would normally see as the movement and dial would normally come out together.
After I took out the stem, it was then a simple matter of flipping the movement from the case. This revealed the rear of the Seiko calibre 7006A – a movement containing 19 jewels and beating at a rate of 21,600 beats per hour (6 beats per second). The serial number of the watch dated it to around 1973.
The timegrapher showed performance to be… ahem… terrible. In Dial Up position it was losing 2 minutes per day, and in Dial Down, 4 minutes per day!
Now for the fun part – dismantling the movement to prepare it for cleaning. Here’s the movement with the rotor removed. Grey gunk can be seen on the ratchet wheel, most likely excess grease from the “magic lever” automatic works.
The dial-side parts were next, followed by the balance, wheel train and escapement.
A few observations could already be made. The barrel arbour hole was filthy and worn on one side – fairly typical for a Seiko given little maintenance over the years.
The balance shock protection jewels were covered in dust and completely dry – they hadn’t seen oil in years.
The mainspring barrel had a serious build-up of dried grease to scrape away.
Finally it was interesting to discover ‘959’ inscribed onto the mainplate by a previous watchmaker – why was it there?
Here are all the parts dismantled.
Getting it back together
After a run through the cleaning machine it was ready to go back together. But due to the time delay between doing the cleaning and starting the reassembly (around a week), in my mind I thought I’d inspected all the parts under the microscope, when I actually hadn’t… This would cause issues later.
In the meantime, I began by getting the mainspring back into the barrel, and continued reassembling and oiling the main parts on the mainplate.
When done, the movement was placed on the timegrapher to see if there was an improvement.
The traces were looking more normal now, and all parameters had improved to a reasonable level. But there was still room for improvement, particularly with the amplitude which was a little low, but also with the wavy trace. The 30 second period of the wave suggested an issue with the fourth wheel (seconds wheel) – perhaps there was still dirt on one of the pivots, or one of the bushes was worn.
It was only then I realised I hadn’t actually inspected any of the parts after cleaning. The late evenings and frequent stop-starts had impacted my memory of what I’d done…
I begrudgingly removed the fourth wheel to clean it again and inspect it. It appeared to be fine (apart from a blob of oily dust), but reinstalling it made a mess of the oiling I’d previously done on the other parts. It seemed the best thing I could do was start over – this would remove any doubt over the cleanliness and condition of not just the fourth wheel but all the parts in the movement. I also decided to order a spares movement so I could get hold of a better mainspring and have parts available to swap out if needed. Waiting for the spares meant the project went on hold for a few weeks – just enough time to get over my frustration of having to repeat work (which I really dislike!)
The parts arrived a few weeks later, and having forgotten my earlier frustrations I treated the project as a new one, cleaning all the parts and now remembering to inspect them! Interestingly nothing serious was revealed – not even the balance cap jewels had any dimples, but I did clean away some leftover muck on the wheel pinions and pivots. I tried having a close look at the fourth wheel bushes but found it difficult to confirm if they were worn, however there may have been a little more sideshake than I would have expected.
With it all back together again and re-oiled, the timegrapher revealed further improvement over the earlier effort. The amplitude had increased significantly as a result of the replacement mainspring and it was more consistent between the different positions. And after adjusting the timing, performance also showed an improvement. The following photo shows it at its best in Dial-Down position, with a high amplitude and an almost flat timing rate neither gaining nor losing time (although it appears to be on the cusp of a downturn).
Dial-Up had similar performance, but the vertical positions were a little more flaky. Crown-Right was the worst with the rate fluctuating between +14 and +24 secs/day and still a slight wave in the trace, which I suspect could be due to worn bushes. I think we’ll have to live with the slight wave for now, and I might come back to it in the future if or when I’m more skilled.
I proceeded with installing the calendar works. Getting to the date jumper, it wouldn’t go in on account of one of the screws underneath protruding up too far. Turns out I’d fitted the wrong screw in this spot…
The culprit was the centre wheel bridge screw, which I’d inadvertently swapped with the balance cock screw. Thankfully I was able to swap them around with only minor dismantling and not too much of a headache.
The rest of the parts went in without issue, and this meant work on the movement was finally done.
I could now turn to tidying up the dial and case. The front of the dial cleaned up okay although a number of scratches and scuffs couldn’t be fixed. Most of it wouldn’t be visible along the dial’s edges but a large circular scuff around the centre hole could be seen under certain light conditions. The hands had previously been pushed down too far and allowed to run; this had also damaged the ‘O’ in the Seiko logo. I had to leave all this as it was.
Now for the next challenge – repairing the dial feet. After much deliberation over how to go about it, I ended up buying a dial feet milling tool from the UK after hearing some positive reviews. I have to say I was pleasantly surprised by the build quality on this Chinese-made unit – it’s weighty, solidly built and the milling shaft rotates well. The milling bit is effective at making a clean recess for the new dial feet to sit in. After first practicing on a scrap dial, using the tool on the Seiko’s dial wasn’t too difficult.
To attach the dial feet most people online seemed to recommend J-B Weld epoxy (I have no affiliation) so I gave this a go. I discovered there were a few obstacles to achieving a good attachment, the first being the date disc. As this sits beneath the dial with minimal clearance, I tried filing the new feet to a shape that allowed the disc to operate freely. This left only a small contact area for the adhesive, and even with the highly regarded J-B Weld cured over 24 hours, one of the feet managed to break free when I installed the dial.
Next I tried using slightly more epoxy around the feet, but another 24 hours later one foot still broke free while installing the assembly into the case. Perservering, I completely smothered the feet in the epoxy and filed away the smallest amount possible for the date disc to still operate. This appears to have worked…
The situation was further complicated by the dial spacer sitting directly alongside the dial feet. Small sections of the ring needed to be filed away so it could sit flat on the dial.
I eventually got it all back together, and after cleaning up the hands the assembly was ready to get cased up. There was a tiny speck of missing lume on the hour hand that could have been fixed (this meant replacing all the lume on both hands, for consistency), but considering all the other flaws on the dial it didn’t seem too much of a distraction so I left it as is.
Case and crystal
Cleaning the case in the ultrasonic cleaner freed it of gunk, but it still needed something to make it look presentable.
Considering the poor condition of the gold plating, I thought a possible interim solution was to simply give the case a light hand polish with metal polish, and ended up using Mothers Mag & Aluminium polish (no affiliation). This was done as gently as possible, rubbing in the direction of brushwork where present, and keeping in mind that using metal polish on gold plating is not recommended as it can remove the plating!
The final result was surprising, as the brassed areas had shined up nicely after losing their tarnish, which helped to hide them amongst the remaining gold areas. The case was looking fairly respectable!
Everything was put back together, which included a new genuine crystal from Seiko. It must have been sitting in stock for quite some time…
So here it is all back together.
The watch was certainly designed to catch the eye – almost every design detail is intended to reflect light in a different way. From the alternating striped brushwork on the dial, the stepped indices, creased hands, and the bevels on the case and crystal, these elements all combine to create a riot of reflections when the watch is angled to the light.
It’s interesting to see the other ways Seiko tried to differentiate their square-cased model from those of its Swiss counterparts. The yellow minute track gives the dial more complexity, aligning with the lume on only the horizontal indices. Then there’s the touch of red on the seconds hand, which together with the yellow track and the copious amounts of gold plating make the watch feel dressy but playful at the same time. I find the style hard to pin down – is it a casual dress watch, or maybe a dressy sports watch, or something else entirely? In spite of its boxy shape, it resists being boxed into any particular category…
Unfortunately the bracelet would have to wait for it’s own moment of glory one day, so in the meantime I suggested fitting a gingerbread suede strap to complement the light dial colours and slightly shabby gold case.
Well, this has been another interesting project with a range of new challenges for me to attempt. It was a genuine surprise to see how well the final watch presented itself in spite of its flaws, although there was no hiding how worn-out it really was, inside and out.
Gold-plated watches in poor condition can be difficult to make nice again due to their need for re-plating. This is something I’d like to attempt in the future if I ever have the appropriate equipment and space – I’d be particularly interested in restoring the bracelet and reuniting it with the watch.
I also need to work on some process improvements, like identifying with more detail what stage each of my watch projects are at (e.g. needing to be cleaned, cleaned, cleaned and inspected) as some projects can sit on hold for quite some time and I end up forgetting exactly what I’ve done. Working late in the evening while tired also hasn’t helped. I might need to stop trying to squeeze so much into my day…
Finally I have to wonder if broken dial feet are a common source of trouble with these models as the design appears to be slightly flawed. Inside the case, it would seem the movement is only secured by the two dial feet and the stem, placing considerable stress on these parts. The hidden stem release lever compounds the problem as its location may not be obvious to many repairers, leading them to pull up on the dial to get a look underneath. It would be interesting to find out if the design was ever altered for later square-cased Seikos, with the movement secured in other ways to reduce the stress on the dial feet.
Please note I have no affiliations with any of the brands mentioned in this post.