I have watches that use both movements. The watches are usually within +2 seconds of atomic time, after wearing them for a couple of days. In the five years I've had them I've not had a single problem.

 

The ETA 2892-A2 is the most accurate movement ever mass produced, in my experience and by the specs. Unfortunately, they occasionally need service... but all mechanicals exhibit that problem.

 

The ETA 2892-A2 is the most accurate movement ever mass produced, in my experience and by the specs. Unfortunately, they occasionally need service... but all mechanicals exhibit that problem.

The 5 years on the 2892-a2 service is mandatory because otherwise the ball bearing winding system system is blocking. And no it is not the best nor the most accurate movement ever produced because the rolex 3035 and its derivatives was the most produced certified chronometer from 1977 till 1988 with 300000-500000 units per year and later the 3135 and its derivatives since 1988 was and is still the most produced certified chronometer with 500000-800000 units per year. The 2892-a2 is a good movement that offers a lot for the money but it is nothing stellar nor exceptional.

 

I've owned several watches that use the 2892 or a modified 2892 ebauche: 3 Doxas, an Oceaneer, Anonimo, 4 SMPs...
It's been the most accurate mvt I've owned, especially just focusing on the Non-COSC watches like the 3 Doxas, Oceaneer, Anonimo ... They've all been within 2 or 3 secs a day.

 

I couldn't help but notice you've posted this in what looks like five or six different forums. May I ask why or to what end? Usually dupe posts are merged... let alone six.

 

I couldn't help but notice you've posted this in what looks like five or six different forums. May I ask why or to what end? Usually dupe posts are merged... let alone six.

This is the main thread in the Public Forum, and it is "linked" to 4 other brand forums as the ETA 2897 (or simpler ETA 2892) is/are known to be used by those 4 watch brands, at least.

Hence, those 4 "linked" threads are meant to capture the attentions of members who do not visit this Public Forum.

This is a deliberation. Those 4 "linked" threads are all directed to this main thread instead of having duplicated contents of their own.

 

I have both movements and they are both some of the most accurate and reliable in my collection.

 

I have this one with a 2895-2 (2892 with subsecond hand at six), and is very accuracy (+4 s/day), but i think it doesn´t load charge very well, if i use the watch 2 or 3 days continously (except sleep) the power reserve is lower than 36 hrs. In all ways, great caliber and great watch i think.


Uploaded with ImageShack.us

 

I have a 2982 in my Bell & Ross and it runs about 1 to 2 seconds a day.

 

I own a 2893-2 (GMT variant). It keeps time to +2 seconds per day (slightly in the minus range on the winder, though). Have had it for about 3 years. Did have it regulated to achieve that accuracy.

Why a Rolex movement would need service every four years as stated above by adswuk is beyond me. I own a Rolex I bought around 1999. But it was from 1990, hardly ever used though. Had it serviced in 2010. It was still going strong at +1 to +2 a day. That performance never changed over all those years. There was no need to exchange any parts of the movement. This is the cal. 3000 that Walt Odets poopoo'ed.

Given that the Rolex 3135 has more flesh to the bone than the 2892, it should be more robust overall and it should be easier to get the parts to work well together because the proportions are bigger.

Here is an article by a watchmaker with more insight:
Rolex Caliber 3135 - Still worthy of the crown after all these years?

Personally, if I had the choice and money was not a matter, I'd go for the 3135. But it is clear that in general and for the money in particular the 2892 is an excellent movement.

 

My new ETA 2892.A2 based watch is about +30/+35 sec a day. I read some articles which say that this movement is one of the most accurate, but maybe I'm not lucky enough. I'm not sure if this is still within normal variation? Is calibration/regulation normally part of warranty service, or it depends on brand?

 

So far, my ETA 2892 is about +45 a day already. It seems that the watch is definitely need service.
Pity, as this is my first swiss made watch and even though it running "legendary accurate" movement, it turned out to be crap. Or I'm so much unlucky.
Anyway, not sure I'm going to invest in swiss mechanical watches again. Maybe I'm too old fashion, but still I'm kinda expect some quality control, from ETA, for such a money paid.

 

My friend tdebug, you should go for profilactic cleaning and oiling of the watch in good watchmaker. He will adjust it so you will not have problems anymore with it.
I have Victorinox Dive Master 500 with ETA 2892-A2, from the very begining it runs +1.5 sec/24hrs and it is 1,5 years old.
As a colleague of one forums says it is all a matter of adjustment , if properly adjusted you can have with Seiko 5 +/- 5 sec/24hrs

 

ETA variations are staggering to me. I own two watches powered respectively by the base 2824.2 and 2893.2 Elabore Grade. Both run to +1 to +2 per day. I have not had them long enough to comment on their reliability. I found the following articles of interest with regard to comparisons with Rolex movements. The consensus appears to put ETA chronometre grade on an even plane with Rolex 3135 in terms of accuracy and reliability. Inhouse movements from Patek Philippe, Jaeger-LeCoultre and Blancpain are compared as well. Not as satisfactorily it seems.

It would appear, particularly in these pages and other fori, that the very ubiquity of the ETA movements leads many to assume they are of inferior quality to inhouse movements of Rolex, the other big European brands and high-end Seiko movements. Comments like "bog standard ETA" permeate many WISs opinions when comparison shopping.

From what I can see, the major advantage to the new Rolex Paracrom spring is that it not made of metal. That makes it non-magnetic and a decided advantage for Rolex.

I have an unshakeable respect for ETA movements that comes from continued reading on the subject and my experience with them on my wrist.

ETA SA - Wikipedia, the free encyclopedia

"The 2892.A2 is regarded by many horologists as being in the same 'quality' range (accuracy and reliability) as the in-house produced Rolex movements"

Rolex Caliber 3135 ? Still worthy of the crown after all these years? - TimeZone

"As for me, please keep in mind that no movement is perfect and that they all have their strengths and weaknesses. Having said that though there is absolutely no question in my mind, that I prefer the ETA 2892-2A over the other two. Okay, so it's been around almost half a century and in many ways isn't as sophisticated as the Rolex - no Breguet hairspring, or Parachrom hairspring material etc - but during its long lifetime in its best available chronometer version, it has proven itself to be an exceptionally accurate, reliable and tough movement. Its two main advantages over the 3135 are that it's quite a bit thinner, only 3.6mm thick versus 6mm, and has only one major weakness - the inefficiency of its automatic winding system, as I mentioned in my earlier review of it. While good enough for most reasonably active folk, it is not efficient enough for those people, young or old, who lead a sedentary lifestyle."

An indepth look at the ETA 2892

"So how does it compare to the competition? There are some movements that match it in terms of accuracy and reliability, but in my humble opinion, none exceed it. The Rolex 3035 and 3135 match it toe to toe. But they are a lot thicker and considerably more expensive too. The PPs, while being very pretty to look at, do not match it for accuracy and are more delicate as far as reliability is concerned. Of course they're also slightly thinner, so that does put them at a disadvantage. The JLC 889/2 does match it for accuracy, but is also too delicate to give it any competition in the reliability department. The main reason for the latter is its very weak mainspring. The whole design, while being well thought out and superbly executed, relies too much on everything being just perfect. It is just thrown out of wack too easily, when even minor things go out of adjustment. I don't have too much experience on the Blancpain/Piguet movements. But from the few that have crossed my bench, they don't seem to deliver the same accuracy that the 2892 has no trouble delivering. Let me know if I've left any of your favorite movements out, and I'll gladly comment on them."

The automotive engine maxim of "Want horsepower, reliability, and low cost? Pick any two." has been turned on its head by ETA watch movements that provide accuracy, reliability and low cost all at once. Fifty years of continuous R&D has apparently accomplished that.

 

Yes, the 2892-a2 and it's many variants are well thought of for good reasons. In my experience two non-chronometer graded movements deliver 7 to 10 seconds per day accuracy on the wrist which is pretty good for a mechanical. The Accutron Gemini 26b12 has a rhodium plated and well decorated 2892-a2 movement displayed attractively through a sapphire back. I like how smoothly the crown winds and sets the movement.

 

i have a cosc 2892 and it is quite remarkable how accurate it is.

 

I have the Jazzmaster Power Reserve with Eta 2897. I wear it a lot and have owned it since 2009 and it loses 2-4 seconds per day on avg.

 

Have the 2893 modified for gmt in an Ernst Benz world timer. It is gaining about 1 sec every 2 days.

 

I have an ETA 2897-2 (Elabore grade). I first did some initial observations out of the box without wearing it on the wrist. The watch was at +2.5 spd dial-up, -2 spd crown up, and +4 spd dial-down. Each position was observed for 24hours.

But that's when the fun really started... when I finally put it on the wrist.

So far I am at 5 (FIVE) days of constant 24-hour wear on the wrist (minus showering), and the watch is at +0 seconds and counting. Will update this thread when the watch manages to finally lose or gain a second.

And you may be thinking "well, the watch is probably running slow in the morning and then fast in the evening, to even out each day".... nope. I check it at random times during the day; after waking in the morning, early afternoon, evening, and before bed. At any given time of day, it's still spot on at +0 spd.

Needless to say I am beyond impressed with this movement, even though I am unashamedly new to automatics.

 

My opinion on the 2892 depends on the level of finish and parts used. It also depends on the price of the watch it's in.

IWC uses the 2892 in its Portofino and Mark XVIII. It's basically a drop in from the ETA factory. Yes, they order the movement in its "Top" (pardon the pun) form, but it's essentially still a 2892 movement with a premium price.

That said, in Chronometer spec, it's an accurate movement. I have had a few in my day from Omega, UN and the like. They always kept accurate and consistent time.

 

Crap. I got sucked into a necrothread.

Rick "deleted" Denney

 

This post was even more irrelevant than the one above, given the dates.

Rick "sheesh" Denney

I don't know what you wrote, but the 2892 (and its clones) are as prolific as ever. So the information is still fairly valid.

 

This post was even more irrelevant than the one above, given the dates.

Rick "sheesh" Denney

 

^^

Been looking at lots and lots of the archives on the 289x movement and very few posts speak to real-world precision results (on the wrist or via timegrapher). Tons of great info on ETA in general and overall reliability of this movement, however...

By the way, my 2897-2 is currently at +0.5 seconds over a period of SIX full days on the wrist (minus showering), sleeping included. That is 0.17 s/d! Love this movement thus far!

After another week or two on the wrist, I will test it out again by resting it fully-wound in all six positions for 24hrs each (I don't own a timegrapher yet but may borrow one from a buddy at work). And perhaps I will do the same for half-wound, although that will take a long time!

In fact, I'm not even sure how to conduct an accurate observation of the watch in the half-wound state, since 24hrs after that point it would be well past its specified power reserve. Maybe just 12hrs would suffice?

Edit: Rick, on an older thread regarding the 2892, you mention a 10-day wear test. Can you give me a bit of insight into that test? In your comment you give +\-1 s/d meaning if the watch was at +10 sec after 10 days it passed your test, correct? Curious if that is similar to what I am doing now, thanks! Here is your post:

I have watches powered by chronometer-grade 2892's that are truly excellent in that regard. Only my best watches can survive my own 10-day wear test with a variance (from the mean regulated accuracy, and thus a measure of adjustment rather that regulation) inside +/- 1.0 s/d. My chronometer-grade 2892watches all do, and are on a par with my chronometer-certified Zenith movements, which are renowned for accuracy. They are better than my top-grade 2892's (which generally fit inside a +/- 2 s/d envelope), which still meet COSC expectations.

 

ETA 2897 (or simpler ETA 2892) --- Accuracy & Reliability, your experiences ...

Edit: Rick, on an older thread regarding the 2892, you mention a 10-day wear test. Can you give me a bit of insight into that test? In your comment you give +\-1 s/d meaning if the watch was at +10 sec after 10 days it passed your test, correct? Curious if that is similar to what I am doing now, thanks! Here is your post:

I use WatchTracker. It calculates a best-fit mean rate, and then evaluates the data points for variance from that mean rate. Here are some results:

Ebel 1911 Chronometer with caliber 120 (2892A2). The mean rate was -1.9 s/d, and none of the data points ventured much more than a second from that mean rate:

This Cartier Santos XL has a mean rate of +4.6, but a very tight variance, demonstrating good adjustment. It isn't claimed to be a chronometer, but it's good enough. The movement is a Cartier 049, a 2892-2 with no date:

This Concord Mariner uses a Top-Grade (I think) 2892A2. The envelope is bigger at +/- 2 s/d, consistent with my other non-chronometer 2892's from high-quality watch companies. Mean rate is +3.9:

To compare with known excellence: here is a chronometer-certified Zenith Captain Chronograph with an El Primero movement, average rate -4 s/d:

And a La Joux Perret 8286 in a Concord Saratoga GMT (average rate -4.4):

A brief test of a Zenith 682 in a chronometer-certified, serviced but 20-year-old Concord Impresario GMT (average rate +5.5):

And, finally, an extended test of a La Joux Perret-tweaked A07.211 in a chronometer-certified Concord C1 Chronograph (average rate -0.7, really the most accurate mechanical watch I own):

As you can see, a well-adjusted 2892 has nothing to be embarrassed about in terms of accuracy.

Rick "not a formal test, but real data all the same" Denney

 

Yes I have a Hamilton Jazzmaster that uses the ETA2892 I have had for 4 days and so far no noticeable difference in time it is keeping even with my clock. It runs well and barely can hear it run which is a little unnerving since the Jazzmaster doesn't not have a second hand so can't tell if its going unless I look at the back and see the movement.

 

Rick, thanks for sharing, those are some great results! I don't have the ability to use an app like that but I can probably do similar testing by recording times more frequently when wearing throughout the day, with the only unknown values being during my sleep hours (still wearing during sleep the last 7 days).

So far I am at just over 7 days of constant wear with a mean of +0.21 s/d... but that is not an avg of many measurements like yours.. just the cumulative delta from atomic time divided by 7.

Great to see your results and how well even the non-chronometer 2892s did.