There is nothing wrong with the watch. It is well within production tolerances for an inexpensive movement. The machine is useful but the time the watch tells on your wrist is the reason you have it. A good mechanical will be better than a minute a week. A good Chronometer, a minute a month. Horology is about accuracy but there is a reasonable limit to production tolerances. It may well be possible to tune the watch to run slightly better given experience and patience but it is more likely that you'll just ruin the water resistance and let in dust.

 

I have no experience with Seikos, but I would have thought the positional variation on a new modern watch should be smaller than what you quote. Maybe you could drag your timegrapher to the store and pick the best one they have?

 

After measuring the performance of quite a few similar watches, I tend to think both the Beat Error and the positional variation are on the high side. I am doing the next best thing to your tongue-in-cheek suggestion. I've ordered a second one of the same watch, and will keep the one that performs best!

CHAPTER 2. Yesterday afternoon I received my second Seiko 5 Sport SRP165K1 with the new 4R36A automatic movement that hacks and winds manually. By the way, these latter features,being novel on a lower priced Seiko, are the points of special interest to me,that caused me to purchase this watch. The second sample of this watch is giving much better timekeeping performance, but the Beat Error values are still high although quite uniform with position. Specifically, the rate varies between + 1 sec./day and +10 sec./day over the six usual positions. The most important on-wrist positions for me(recall I'm a lefty and wear a watch on my right wrist) give +6 sec./day crown up, +8 sec./day 6 o'clock up and +10 sec./day dial up. These values are quite acceptable to me, and will probably actually be better on the wrist and after a month or so for settling time. The Beat Error vales over the six usual positions vary between 1.1 and 1.5msec. While the Beat Error values are high, the variation is not too bad. In any event, as the English gentleman noted, the on-wrist timekeeping is the important thing, and I'm confident that will be pretty good. Needless to say, the first example of this watch is on its way back to the seller for a refund!

 

generally, beat error should not exceed 0.8ms in any single position. your rate delta is also high, but really, for a seiko 5 you cannot expect very high levels of regulation from the factory for the price.

 

With adjustment, it should be possible to bring the Beat Error on this watch within the range of 0 to 0.4ms. Although I have the equipment to measure it, I'm not sure how difficult this adjustment might be. Since there is no plus or minus to Beat Error, this could be pretty tricky to get right. What really bothers me more is the positional error. I can deal with regulation, but I can't adjust for the positional error.

 

Let me begin by saying Seikos are not your ordinary watch. They consistently have low amplitudes, even when new. Keep in mind these movts come out of a machine and as far as I know, are not actually assembled by human hands. (Our Professeur had a tour of the factory) While I would say your beat error is a bit high, your positional error is par for the course with this type of watch. Before the Seiko enthusiasts jump on me, let me say I have nothing against Seiko, they are an inexpensive watch, but you do get what you pay for. I have many Seikos, (over 20) and they range from over forty years old to just a few years old, auto and manual, hi-beat and chrono. I have had them all on the Witschi, and they are all surprisingly similar, save the hi-beat. Since most of them do not have the manual wind option in addition to the auto, I was curious to see what they did at full power, so we wound them internally til the bridle slipped, and the amplitudes were still low. These watches are designed a certain way, and it would not surprise me to find out that your watch had been regulated to have the errors cancel each other out, or basically timed while on a simulator. While it is possible to regulate the watch for a better result, keep in mind this would involve a fairly complex series of adjustments, from the jewel shake to the hairspring (centered, flat, perfectly in the centre of the regulating pins, and said pins then adjusted for minimum positional error) . That's assuming the watch is not magnetized or the balance out of poise. What makes the difference between a Seiko 5 and a Swiss mechanical is the quality of the materials. The metal in the movt is very cheap and the screws are brittle, and tend to break, as do pivots and arbors. While you do see a lot of old Seiko 5's around, still running, that's about all they are doing, running. Most of mine lose or gain minutes every other day, and that's after a proper service and regulation, but hey, they are 40 years old and had never been properly serviced before, so it's a miracle they run at all.

I would also like to add a postscript to the effect that, yes, Seiko makes the higher end lines like the Spring Drive, and these watches are very very good. The majority of the mechanical Seikos out there are just good workhorses that commonly live long past their designed replacement date.

 

Ed.P., no, it actually wasn't tongue in cheek. I remember a post, I don't think it was watches, maybe lenses, where after several returns the (brick and mortar) store manager got tired of it and allowed the customer to just pick the specific item from what they had in stock. To the store it was all the same, since no other customer would have cared about performance at that level. But if it's mail order that won't work, at least not as efficiently. Outta Time, so you are saying Seikos are significantly worse than say a Sea-Gull ST16 (as far as positional variation is concerned)?

 

They can be. I have a few Seiko 5's that perform much better than the OP's, but I service them myself. And I had a Seagull that was decent, but didn't last more than 5 years without structural failure. It is the material that is critical. Russian movts have also suffered from this. This is not to say all Russian watches are bad, or even all Seikos are bad. They are what they are, affordable decent quality watches. I have a Cardinal with a Russian movt, and it runs extremely well and seems to be made with good quality metals. If you were to take a Segull or Seiko or any mechanical movt, it can be tweaked to run quite well, but final fine adjusting and tolerances for these watches at the factory are not the same as say, a Swiss maker's similar product. I have come across Omegas that were pretty bad, design wise, and suffer from horrible errors, regardless of tweaking, and I don't mean vintage ones either. But these are the exception. A good example is the Rolex 3100 series movts: Excellent materials, several design flaws. The rotor will eventually hit the balance while being shocked. There is an extra bridge of metal to protect the balance because of this. I say why not fix the rotor post problem? It has actually gotten worse, as the new posts are smaller.
But I digress.
There is an inherent danger in pointing at particular brands and trying to assess their quality level as a whole, it really must be done carefully. Seikos are designed well, and are durable, and have many excellent case designs. Their quartzes, in particular, are very good. Have you ever seen an ETA 805 series? Complete garbage. You will find these in $300 Swiss quartz watches. My Seiko Panda has a vertical clutch column wheel movt, and from that standpoint, is a good design. It suffers from poor materials, however, and Vintage Seiko parts aren't easy to come by. A good watchmaker can adjust one to run quite acceptably, and this can be said for most mechanical watches when they are new.
So I wouldn't say they were worse, in fact, I have a soft spot for Seikos and on my own personal scale, I like them better than Seagulls. But I'm the guy that has to crawl inside the movt and really assess what it is and what it's doing, so my point of view is somewhat different. For many ppl, Seiko mechanicals have the warm and fuzzy factor, they seem to last forever, and everybody knows them and probably has owned one or known someone in the family who did. And while you might think it pedantic for someone to rage about a broken screw because of quality, it speaks volumes about cost cutting measures in the watch. I recently had a customer's Longines Railroad watch, and what struck me first as I serviced it, were the screws: They were flawless and of such high quality, they were the standard on which the watch was built, with it's perlage and decoration even in places you wouldn't see it. The only other time I was so impressed with screws was in a Patek.
So, Seiko good, Seagull bad, or vice versa? I'd say they were comparable, but Seiko has invested a lot more as a company, and been around a long time. It's got brand recognition, and they have had a lot of years to experiment with their movts. Again, if you have a watch that seems like it's not performing, take it back under warranty by all means, or if not, get to know your local watchmaker and have it tweaked.