[JustBlueFish]

You may have seen a thread I posted a little while ago (Sarb033 Accuracy Disappointment). In that thread I mentioned that I received a new Sarb033 and I was worried about some accuracy issues with the new watch, and it's odd behavior I had been seeing. I have since spent some time closely monitoring this watch and I thought I would share what I've seen so far. Perhaps some members might find it interesting.

What Was My Monitoring Methodology?

1.) I hand wound the watch, 40-45 rotations of the crown, until feeling a slight change the action that I perceive as the over-wind protection or clutch kicking in.
2.) I set the clock's time as exact to the website Time.is as possible.
3.) I placed the watch in my storage case, dial up, in a 72 degree environment. The watch was not moved from this position.
4.) Every 4 hours (except when sleeping) the time on the watch was compared to the current time on the Time.is website and noted.

Note: A couple data points were estimated based on the the previous and following data points, for the purposes of sleeping and not being able to check the watch.

Why bother doing this?

When I purchase a new mechanical/automatic watch, I like to test it's accuracy in a simple static environment. Since a watch that has been worn can't be returned, this test gives me enough piece of mind to "unbox" the watch, resize the bracelet, and begin wearing it.

I am aware that the accuracy of the watch will be different when it's worn on the wrist, but for my piece of mind I want to find out if the watch performs reasonably well in a static and controlled environment first. Normally I am not this thorough, I just wind up the watch some and check it a couple times over the course of 24 hours. However, I noticed some wildly innacurate results with this Sarb033 initially, started to panic a little, and decided to test the watch more completely.

What were my findings?

During my first little test of the watch I was finding that the Sarb033 was wildly inaccurate. It was losing double digit seconds over the course of hours, not days. After that informal test I got much more thorough.

I found the following to be true about this Sarb033 while doing this test:

• It takes about 45 full rotations of the crown to get a complete charge on the 6R15c Movement.
• A complete charge lasted exactly 56 hours 28 minutes.
• The watch was extremely accurate over the course of the first 48 hours.
• The watch ranged between as much as +3 seconds faster than Time.is to -1 second slower than Time.is over the first 48 hours.
• Over the last 8 hours, at the end of the power reserve, the watch lost a lot of time.
• In those 8 hours the watch fell as far behind the Time.is official time as -26 seconds.

I now believe that during my initial testing of the watch, I was getting such wildly inaccurate results because I simply didn't charge up the watch enough. That said, I was always under the impression that a watch would typically speed up as the power reserve runs down. For whatever reason this 6R15c doesn't behave that way. I did see a youtube review where one user pointed out that his Sarb033 also lost a lot of time at the end of the power reserve.

I now wonder, is this typical of the 6R15c, but overlooked by the vast majority of users because they don't monitor their watch while it's power reserve is at the last 15%?

Or is this truly an anomaly with this particular example watch that I own?

I'm not sure...

The Data and Graph

Text data that is represented on the graph below. 

Duration (hours) 	Error (seconds)
----------------	----------------
0			+0	
4			+1
8			+2
12			+2
16			+3
20			+3
24			+3
28			+3
32			+3
36			+2
40			+1
44			+0
48			-1
52			-10
56			-26

 

[Mr. James Duffy]

This is consistent with most of my Seiko watches except for my Baby Tuna with its 4R36 movement and my Cocktail time with its 6R15, both stubbornly running +25s per day.

 

[obomomomo]

Very similar to what I've found, all my 6R watches does like to stay fully or near fully wound and a low state of power reserve will cause it to slow quite a lot in the final few hours.

 

[jupiter6]

To be honest I'm underwhelmed by the performance of the 6R15 in general. It veers from very accurate to unacceptably inaccurate from day to day.

I'm interested in your test results. Thanks for that.

 

[JustBlueFish]

Thanks for the feedback. Appreciate it. Now I need to decide, do I bother to do the test a couple more times to get a feel for the watches consistency? Or perhaps a test that involves measuring accuracy while wearing the watch at least 12-16 hours a day for a few days?


Sent from my iPad using Tapatalk

 

[LightDot]

Great data you have here! Thank you for sharing. I wish more watch owners would publish such findings.

I would suggest measuring the watch like you did in all 6 positions (so 5 of them remaining) and than during use. Differences between the 6 positions will tell a lot about the accuracy of your particular watch, and, as you probably know, changes in behaviour can also warn that the mechanism needs service once it comes to that.

Since your measurements give an average over a time period, you can augment them by using a timegrapher. Can't remember if this came up already, but there is a great free and open source timegrapher software available, which, combined with a decent microphone, can turn your computer into a capable timing device: https://www.watchuseek.com/f6/open-source-timing-software-2542874.html

 

[obomomomo]

Thanks for the feedback. Appreciate it. Now I need to decide, do I bother to do the test a couple more times to get a feel for the watches consistency? Or perhaps a test that involves measuring accuracy while wearing the watch at least 12-16 hours a day for a few days?

Sent from my iPad using Tapatalk

I don't think further tests will reveal anything very much different from what you've already noted. End of the day it's the result from wearing it that counts. I've found that accuracy can vary depending on the daily activities and watch wearing habits. Not sure why but short sharp movements like shaking or repeatedly flicking the wrist or 'skimming stone' throwing type motions to make the rotor spin will cause it to speed up momentarily and you can even get it to gain a few seconds this way. The effect is greater when the watch is low on power reserve and less when fully or nearly fully wound. The 6R is not alone here though, it happens with my Swiss watches too.

 

[wasf]

My SRP775 became more accurate, when finally, I stopped caring about accuracy

 

[JustBlueFish]

My SRP775 became more accurate, when finally, I stopped caring about accuracy

I don't think I could stop caring about the accuracy of my timepieces, when their main purpose in the world is to keep time as accurately as possible. At that point, I feel like I might as well just wear a bracelet or some other jewelry instead of a watch. But I don't fault you for your point of view, but for me that probably wouldn't work.

 

[jupiter6]

My SRP775 became more accurate, when finally, I stopped caring about accuracy

I never cared that much, but 2 or 3 minutes slow in the space of a week is enough to make me miss the train, which makes the watch useless.

 

[Makeitfromscratch]

Actually, I get it. When I first got into watches, it was all about accuracy and how to make it better, how to acquire the tools and technique and skills to make those adjustments or repairs. Then, as time went on I came to realize that in some cases it's just o.k. to let the watch be. It's cleaned, serviced, and well within its parameters, that's good enough. Sure, I can throw $2,000.00 worth of labor into a $400.00 watch, but to what end?

By the way, your watch is running really well it seems. I trust that you decided to keep it? Enjoy.

 

[JustBlueFish]

By the way, your watch is running really well it seems. I trust that you decided to keep it? Enjoy.

The plastic's been removed and the bracelet has been resized.

Sent from my iPad using Tapatalk

 

[Makeitfromscratch]

Congrats, it's a keeper.

 

[Makeitfromscratch]

Nice spread on the data, by the way. Enjoyed seeing it.

 

[JustBlueFish]

I just found the iPhone app Hairspring. Interesting little tool to measure mechanical watches precision. I just ran my Sarb033 through a test for 5 minutes. Assuming the app works and is an accurate testing tool, I got a result of the Sarb033 running +5.3 s/day.

I'll test it again tomorrow after it's had some time to burn off some power reserves and see what I get.

For those interested, according to the app my Hamilton Khaki Field was running -6.3 s/day and my Seiko SNK809 was -3.0 s/day.

Yep, the $58 watch was the most accurate.


Sent from my iPad using Tapatalk

 

[Drubbing]

I never cared that much, but 2 or 3 minutes slow in the space of a week is enough to make me miss the train, which makes the watch useless.

I think this is less of a watch failing, than a time management thing. There are always quartz watches if you run that close to making it onto public transport

 

[Drubbing]

I never cared that much, but 2 or 3 minutes slow in the space of a week is enough to make me miss the train, which makes the watch useless.

I don't think I could stop caring about the accuracy of my timepieces, when their main purpose in the world is to keep time as accurately as possible. At that point, I feel like I might as well just wear a bracelet or some other jewelry instead of a watch. But I don't fault you for your point of view, but for me that probably wouldn't work.

There's quartz if this is people's main reason for buying a watch. All styles and price ranges. I think the SARB is a nice watch that does a decent job. I haven't checked the time on mine for a couple of weeks, as I have a Seiko 5 too, and they both read the same time, so I figure they're both running well. But having done so, it's 45 seconds out.

AS per my reply to the above poster, if that, or a couple of minutes inaccuracy, is causing people to miss trains, the problem isn't the watch...

 

[obomomomo]

I just found the iPhone app Hairspring. Interesting little tool to measure mechanical watches precision. I just ran my Sarb033 through a test for 5 minutes. Assuming the app works and is an accurate testing tool, I got a result of the Sarb033 running +5.3 s/day.

+5 spd is an excellent result. Like many others here, I like my watches running a little fast than slow. Positional variance, 6Rs tend to run a little fast resting horizontally either dial up or dial down overnight, and a bit slower when placed vertical, say crown up. Useful for fine tuning while you sleep.

 

[yankeexpress]

To be honest I'm underwhelmed by the performance of the 6R15 in general. It veers from very accurate to unacceptably inaccurate from day to day.

I'm interested in your test results. Thanks for that.

Agree. I avoid the 6r15 movement.

 

[jupiter6]

I think this is less of a watch failing, than a time management thing. There are always quartz watches if you run that close to making it onto public transport

There's what you think, and then there's what you know. Your revelation on quartz watches is very helpful. That aside, I'm honoured that you were compelled to quote me not once, but twice.