An ST18 and an ETA 2892-2, a review

40219 Views 18 Replies 14 Participants Last post by Scottish Steve, Oct 21, 2012

lysanderxiii

Discussion Starter · Apr 27, 2009

17

Foxint requested I review and compare an ST18 with an ETA 2892, since he was supplying the two movements, how could I refuse?

The ST18 was a 23 jewel version, operating at the normal high beat speed of 28,800 bph. Initial timing showed the movement to be fairly well timed. The beat error was 0.4 milliseconds and the average rates were +8.2 s/d in the dial up position and +15.2 s/d in the crown down position. In a wristwatch this would result in approximately +10 to +12 s/d when worn.

The ETA 2892 was an older version, dating to approximately the late eighties to the mid nineties, being a -2 variant, not the current A2 version produced today. This was a 21 jewel movement. Being used, no comparison of initial timing could be made.

Some general comparison shots (The ETA 2892 is the first or left hand image.)

Front

Here they both are again without the Autowind bridge installed so we can see the gear train:

From this point on we will stick with the ST18's disassembly, showing where the parts go.

Here the train cock has been removed, exposing the third, fourth (in the center) and escape wheels:

Here the gears are removed, you can see the balance stop lever, the gold/copper colored part up near the stem:

Now the entire barrel bridge, barrel and off-set center wheel has been removed. The barrel bridge has all the parts for manual winding installed on the underside. It's a fairly busy assembly, we'll get into that in a later post when we discuss the operation of the movement in detail.

Balance removed:

Pallet cock and pallet removed:

Here is the mainspring out of the barrel, seems well formed:

And the balance, This also seems well made:

Here is the underside of the autowind bridge.

And here are all the parts of the automatic winding mechanism

And the rotor:

The 2892-2 disassembled the exact same way, so no shots of it coming apart.

This is just the introduction, in a future post on this thread we will get comparison shots of all the major, and some of the minor, parts of both movements and how they stack up against one another.

Then when we have compared the parts we will discuss the ETA 2892 design in general, and the ST18 specifically.

Be seeing you...

See less See more

Reactions: 4

1 - 19 of 19 Posts

gigfy

· #2 · Apr 27, 2009

Thanks! I'm waiting with bated breath. For the comparison AND the pictures! (pics are blocked at my work)

Cheers,
gigfy

igorschutz

· #3 · Apr 27, 2009

WOW! Nice post!!!

I can't wait for the upcoming posts!

Many thanks for the effort,

Igor

jason_recliner

· #4 · Apr 28, 2009

gigfy said:
(pics are blocked at my work)

you understand my pain!!!!!

Pawl_Buster

· #5 · Apr 28, 2009

Your 'tear down' posts are the best :-!

:thanks

I also wait in anticipation of your detailed descriptions

lysanderxiii

Discussion Starter · #6 · Apr 28, 2009

24

Comparison shots (ST18 first or on the left).

The pallet forks and jewels. Bothe forks are well made, with little polish on them, which is atypical of ETA. The ST18 shows the typical heavy dose of shellac on the pallet jewels, and its unsightly bubbling, this does not affect the operation of the movement.

Underside of the balance, showing double roller and the notch for the safety pin. You may note some slight differences. The ST18 uses the traditional fixed pins, while the the ETA uses their ETACHRON design.

Upper view, the only notable difference is the slot in the eccenteric goes all the way through the head in ST18.

The top of the barrel.

The ST18 has a serial number, apparently, to make it look as if it is a chronometer.

Shot of the jewels. We can see traces of oiling on the ST18.

Cutout for the balance, showing the ETA markings

Post for the cannon pinion

Underside of the barrel bridge, the ST18 has a jewel for the barrel arbor.

Top of the barrel bridge, showing the nice decoration on the ST18.

You can see the ST18's fit and finish is comparable to the ETA 2892-2. A little later we will get into the other parts.

See less See more

lysanderxiii

Discussion Starter · #7 · Apr 28, 2009 (Edited)

42

More comparison shots:

Here is the offset center (or second) wheel. For those who may not know, the first wheel in the gear train is the large gear on the drum of the barrel. This pivot takes the largest load of any of the wheels, that is the reason for the large diameter axle.

Bottom view:

The third wheels:

Bottom view:

The escape wheel, with the focus on the pinion gear:

The focus on the escape wheel:

The fourth wheel and the post for the second hand. The are two noticable differences, 1) the ETA post is shorter, as the movement was made with the shortest hand stack, the ST18 is the small as the "normal" hand height, and 2) the number of spokes in the gear. This last difference may seems silly, but it is important, as ETA uses the number of spokes in the 4th wheel to denote the movement it works on, 6 spokes for 21,600 bph movements, 8 spokes for the 28,800 bph movements, and 5 spokes for the 36,000 bph movements. The only difference between the fouth wheels for the different speed movements would be the number of teeth the the big gear.

Here is the bottom of the winding bridge, note the post as one gear axle on the ST as opposted to the jewel on the ETA.

But, the ST still keeps a jewel on the post bearing:

Top view of the barrel bridge:

The rotor mount area:

The main gear train cock, with its three jewels (underside):

Top:

Another major difference. The ST uses the more ecomonical method of putting pins in the mainplate, as opposed to the ETA method of trepanning a post onn the bottom of the cock. Seagull uses this on all their ETA copies for all the bridges and cocks.

Also, here we see that ST had to adjust the end-shake of the balance. They did so with the simple method of driving a burr into the bottom of the balance cock. It may not be an elegant solution, but it does work.

The bottom view of the winding bridge:

The single reverser gear:

The underside of the pallet cock:

The rotor bearing:

Here is the underside of the barrel bridge. The single "V" shaped spring is the click spring, and the single weakest part of the 2892 design. As yoou can see, the gap between the plate and the bridge is slightly greater than the width of the spring. The spring is bent upward (relative to the viewer) to engage the gear, it can slip downward and disengage the gear, this will release the tension on the mainspring. (cont below)

Moreover, only after the movement is assembled can tension be placed on the barrel, and the first application of tension will cause this spring to slip off. Worse yet, the barrel bridge, unlike all earlier ETA designs cannot be removed independently of the other bridges and cocks, so if this spring slips off, almost the entire movement must be disassembled to fix it. The solution seems to be put enough bend in the spring so that its natural tendency is to slip off in a upward (relative to the view), this way the ratchet gear keeps the click in place.

.
.
.

See less See more

lysanderxiii

Discussion Starter · #8 · Apr 28, 2009 (Edited)

Now, that we have seen all the pictures what does it all mean? And, how does the ST18 compare to the ETA 2892-2? How good a movement is it?

The ST18 compares very well with this ETA 2892-2, the finish and workmanship of this ETA is similar. The Seagull has some minor alterations to reduce the cost of manufacture. Now, we must remember that this is not a current 2892A2, one of the top-of-the-line automatic movements produced by ETA, but apparently a one made some twenty years ago. This 2892-2 was likely the second level grade, the Standard (now it would be called Elabore grade), judging from its balance, which seems to be gilt-nickle, not glucydur.

Both movements, after re-assembly, were timed and regulated. This was done fairly swiftly, as the movements were fairly good in positional variation. The ST18 ended up +8.4 s/d, with a positional variation of about 5 s/d, the ETA about -0.4 with a positional variation of about 4 s/d. They were both very stable.

The ETA 2892-2 design (and the 2892A2, as nothing major has changed) is not in my opinion superior to the 2824-2. Other than being a slimmer more compact movement, it has nothing I can see that would give it an engineering edge over the 2824, and several things that I think are weaker.

The click of the 2892 is a very small thin spring that can slip off the intermediate crown wheel, it is not the use of a spring as a click that I dislike, the ETA 2850/2870 series also uses a spring as a click, but the 2850/2870 this part is a short stiff spring that cannot flex and slip off it gear.

The 2892 has about 10% more parts, most small gears used to link parts that where separated by the need for thinness.

The 2892 does not disassemble and re-assemble as easily as does the 2824.

The fact that ETA has placed the 2892A2 in the Mecaline Specialites line (the high-end line), and these are more carefully made and better decorated than the Mecaline, is all that keeps the 2892A2 ahead of the 2824-2. Personally, I prefer the design of the 2824.

Similarly, Seagull may have placed the ST18 in a higher prestige tier than the 2824 copy, the ST2130, as this movement does appear slightly (if very slightly) better finished than the ST2130 I looked at a few months ago.

See less See more

Reactions: 2

ezcheese

· #9 · Apr 28, 2009

Great post! I enjoyed reading about these movements and seeing those super cool pics! :-!

Pawl_Buster

· #10 · Apr 29, 2009

lysanderxiii said:
Now, that we have seen all the pictures what does it all mean? And, how does the ST18 compare to the ETA 2892-2? How good a movement is it?

The ST18 compares very well with this ETA 2892-2, the finish and workmanship of this ETA is similar. The Seagull has some minor alterations to reduce the cost of manufacture. Now, we must remember that this is not a current 2892A2, one of the top-of-the-line automatic movements produced by ETA, but apparently a one made some twenty years ago. This 2892-2 was likely the second level grade, the Standard (now it would be called Elabore grade), judging from its balance, which seems to be gilt-nickle, not glucydur.

Both movements, after re-assembly, were timed and regulated. This was done fairly swiftly, as the movements were fairly good in positional variation. The ST18 ended up +8.4 s/d, with a positional variation of about 5 s/d, the ETA about -0.4 with a positional variation of about 4 s/d. They were both very stable.

The ETA 2892-2 design (and the 2892A2, as nothing major has changed) is not in my opinion superior to the 2824-2. Other than being a slimmer more compact movement, it has nothing I can see that would give it an engineering edge over the 2824, and several things that I think are weaker.

The click of the 2892 is a very small thin spring that can slip off the intermediate crown wheel, it is not the use of a spring as a click that I dislike, the ETA 2850/2870 series also uses a spring as a click, but the 2850/2870 this part is a short stiff spring that cannot flex and slip off it gear.

The 2892 has about 10% more parts, most small gears used to link parts that where separated by the need for thinness.

The 2892 does not disassemble and re-assemble as easily as does the 2824.

The fact that ETA has placed the 2892A2 in the Mecaline Specialites line (the high-end line), and these are more carefully made and better decorated than the Mecaline, is all that keeps the 2892A2 ahead of the 2824-2. Personally, I prefer the design of the 2824.

Similarly, Seagull may have placed the ST18 in a higher prestige tier than the 2824 copy, the ST2130, as this movement does appear slightly (if very slightly) better finished than the ST2130 I looked at a few months ago.

Well done and thank you for another excellent series.
Hopefully gigfy will put these valuable posts into the reference section :-!

Foxglove

· #11 · Apr 29, 2009

Amazing post. Thank you very much!

Chascomm

· #12 · Apr 29, 2009

It's interesting to see traces of typical Chinese manufacturing practice applied to a 'clone' movement; specifically the jewelled barrel arbor and the arrangement of the plate pins. It indicates that somebody was thinking while they were copying.

gigfy

· #13 · Apr 29, 2009

Alpha-Getty said:
Well done and thank you for another excellent series.
Hopefully gigfy will put these valuable posts into the reference section :-!

Done.

Excellent post lysanderxiii. I wish I was a watchmaker so that I could fully digest these technical posts. I'm picking up more and more thanks to you. You do a excellent job of explaining your points so everyone has the opportunity to learn something new. Thank you. :-!

Cheers,
gigfy

Martin_B

· #14 · Apr 29, 2009

Thank you for this comprehensive post! :-!

Regards,

Martin

foxint

· #15 · May 5, 2009

Mate

Well done. Thanks for that. I have been head down and tail up. So I have nto been around.

Thanks again. You have done a super job as always.

Dan
Orange

BullDawg

· #16 · May 5, 2009

lysanderxiii,

Great post and super photos! I like how you added the blue tent to easily recognise the two movements.

Cheers,

Bulldawg.

lildaddym

· #17 · May 2, 2010

Any idea about the stem? Any specifics?

440saw

· #18 · Jun 6, 2012

DITTO!!

gigfy said:
Done.

Excellent post lysanderxiii. I wish I was a watchmaker so that I could fully digest these technical posts. I'm picking up more and more thanks to you. You do a excellent job of explaining your points so everyone has the opportunity to learn something new. Thank you. :-!

Cheers,
gigfy

Scottish Steve

· #19 · Oct 21, 2012

Thank you very much indeed for all the work involved in this wonderful post. Looking through these pictures has given me a much better idea of how complex even a 'basic' mechanical calibre is. Very impressive. This post has also done more to convince me of the competence of Sea Gull than any other I've seen.
It should be a sticky on f2!

1 - 19 of 19 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.