I was recently asked to continue to make posts showing some of the watch rebuilds that I've been working on as of late. With that request in mind, I did my best to remember to take photos of my most recent project. This one is a Seiko from 1980 - model # 7009-8580.
I got this watch on eBay. It isn't particularly valuable by any means, but I thought it looked pretty good and had a movement in it that I hadn't yet worked with. Per the eBay seller, it was 'Running well" & "Keeping good time". I had my doubts with this being a 42 year old watch, but the price was right so I figured I'd pick it up.
Before I started I put it on my timegrapher to see how the watch was actually running. Here are the results (pre-service):
The seller's definition of 'Keeping good time' and mine are vastly different it seems. Nevertheless, I was encouraged with these poor beginning readings because it really needed a service and I wasn't just going to be working on it for the sake of practice. The watch needed some attention. The amplitude reading of 220 degrees as shown above is not even close to what the watch was actually running at. With the beat error so badly off at 9.3ms, the amplitude reading won't be accurate. To get a more accurate reading of amplitude (degrees in which the balance swings in each direction), the beat error must be sorted first.
I got the watch fully disassembled down to each individual component. It was very filthy. The lubricants had gummed up and all of the gaskets were compromised.
Here is the watch fully disassembled:
The mainspring had a slight deformation in it towards the bridal end (right side of the photo). Thankfully I was able to reshape the spring without putting any kinks in it, and was able to re-use it for assembly.
After cleaning (no photos), I began assembly. While fitting the rear side of the movement back together, I found that the mainspring barrel had too much side shake when the bridge was installed on top of it. Over time, the barrel arbor had worn out the bearing race on the top barrel bridge, causing the round hole to become an oval. Having too much slop in the barrel will cause the watch to run inefficiently, amplitude will drop, and the power reserve will be diminished. I used my recently acquired staking set to correct this error:
I used a domed anvil on the underside and a domed stake on the top side of the bridge, and with the staking set it began to move material inward. I moved enough material to shrink the hole size down to where it was too small (intentionally) for the barrel arbor to fit in. Once the hole size was brought down, I used a smoothing broach to open the hole back up to the specific size needed with just the right amount of side shake (very small amount). Here is a photo of the work being done on the staking set. The smoothing broach is a two handed job so no photos could be taken:
After I got the watch movement assembled I put a full wind in it, let it run for about 2 hours, then began to regulate it. If you reference the timegrapher photo above, showing the results pre-service, it was running slow by 95 seconds per day, with a beat error of 9.5ms (HUGE!!). Those results were in the 'dial up' position. Here are the results post service and regulation:
Dial Up (movement is still in holder, and is oriented with the back end facing up - which is why it is turned upside down to get the dial up position):
Dial Down:
I'm very happy with those results. This is well above the stated accuracy of this movement, but I probably put more work into fixing the mainspring, adjusting side shake in the barrel, checking end shake in the wheel train, and proper use of modern synthetic lubricants than the factory did when they built it.
I went on to assemble the rest of the watch and I noticed that the day wheel was not centered when the dial was installed. This bugged the hell out of me so I took the day wheel apart, adjusted the gear on the back side of the wheel that indexes it into position when it moves, and then riveted it back together:
The watch is now fully assembled again. I decided not to touch the case. It was in decent shape for a 42 year old watch, so I kept the light scratches on it that were there, and only cleaned it up. I did however take some fine grinding paste and put some elbow grease towards the crystal. It has nearly all marks gone and really makes the watch look good.
It probably took me as much time to write this out as it did to work on the actual watch!! Only kidding, but I do hope that those of you who've expressed interest in seeing this watch repair work being done enjoy this write up.
I got this watch on eBay. It isn't particularly valuable by any means, but I thought it looked pretty good and had a movement in it that I hadn't yet worked with. Per the eBay seller, it was 'Running well" & "Keeping good time". I had my doubts with this being a 42 year old watch, but the price was right so I figured I'd pick it up.
Before I started I put it on my timegrapher to see how the watch was actually running. Here are the results (pre-service):
The seller's definition of 'Keeping good time' and mine are vastly different it seems. Nevertheless, I was encouraged with these poor beginning readings because it really needed a service and I wasn't just going to be working on it for the sake of practice. The watch needed some attention. The amplitude reading of 220 degrees as shown above is not even close to what the watch was actually running at. With the beat error so badly off at 9.3ms, the amplitude reading won't be accurate. To get a more accurate reading of amplitude (degrees in which the balance swings in each direction), the beat error must be sorted first.
I got the watch fully disassembled down to each individual component. It was very filthy. The lubricants had gummed up and all of the gaskets were compromised.
Here is the watch fully disassembled:
The mainspring had a slight deformation in it towards the bridal end (right side of the photo). Thankfully I was able to reshape the spring without putting any kinks in it, and was able to re-use it for assembly.
After cleaning (no photos), I began assembly. While fitting the rear side of the movement back together, I found that the mainspring barrel had too much side shake when the bridge was installed on top of it. Over time, the barrel arbor had worn out the bearing race on the top barrel bridge, causing the round hole to become an oval. Having too much slop in the barrel will cause the watch to run inefficiently, amplitude will drop, and the power reserve will be diminished. I used my recently acquired staking set to correct this error:
I used a domed anvil on the underside and a domed stake on the top side of the bridge, and with the staking set it began to move material inward. I moved enough material to shrink the hole size down to where it was too small (intentionally) for the barrel arbor to fit in. Once the hole size was brought down, I used a smoothing broach to open the hole back up to the specific size needed with just the right amount of side shake (very small amount). Here is a photo of the work being done on the staking set. The smoothing broach is a two handed job so no photos could be taken:
After I got the watch movement assembled I put a full wind in it, let it run for about 2 hours, then began to regulate it. If you reference the timegrapher photo above, showing the results pre-service, it was running slow by 95 seconds per day, with a beat error of 9.5ms (HUGE!!). Those results were in the 'dial up' position. Here are the results post service and regulation:
Dial Up (movement is still in holder, and is oriented with the back end facing up - which is why it is turned upside down to get the dial up position):
Dial Down:
I'm very happy with those results. This is well above the stated accuracy of this movement, but I probably put more work into fixing the mainspring, adjusting side shake in the barrel, checking end shake in the wheel train, and proper use of modern synthetic lubricants than the factory did when they built it.
I went on to assemble the rest of the watch and I noticed that the day wheel was not centered when the dial was installed. This bugged the hell out of me so I took the day wheel apart, adjusted the gear on the back side of the wheel that indexes it into position when it moves, and then riveted it back together:
The watch is now fully assembled again. I decided not to touch the case. It was in decent shape for a 42 year old watch, so I kept the light scratches on it that were there, and only cleaned it up. I did however take some fine grinding paste and put some elbow grease towards the crystal. It has nearly all marks gone and really makes the watch look good.
It probably took me as much time to write this out as it did to work on the actual watch!! Only kidding, but I do hope that those of you who've expressed interest in seeing this watch repair work being done enjoy this write up.
Last edited:
