New Rollie Gmt Model, Modified Eta 2836-2

[RWG Technical]

The NEW GMT Rollie with the Correct hand stack…

This is the new version of the GMT model, it differs from the older version in a few area’s…

1. the hand stack is a-la Rolex, meaning the Hour hand is closest to the dial and the GMT hand is in between the hour and minute one

2. the movement is a 2836-2 (not even close to being designed as a GMT movement), which is poorly modified to try and sort-of-mimic-a-GMT movement, it fails on the functional side.

3. the Hour hand is Adjustable NOT the GMT one, this is an important point, as you will see later on.

Now, if your planning on taking one of these watches apart, be warned, on each and everyone I have worked on, the GMT hand gear comes apart (the tube pulls out of the gear) and requires re-assembly…as you will see in the photo’s.

Here is what happens when you take the GMT hand off, the post that is part of the GMT gear is SUPPOSED to stay attached to the GEAR, not come off with the hand… Have a look and see here…

This is the movement, and next to it, is the GMT gear, and the GMT hand with the post still attached to it…

So I take the post out of the GMT hand, carefully ‘cause if you loose it, your screwed big time…or if you bend the hand…

Here is what we have to work with, small parts that have to be perfectly re-assembled (or they will jam up the gears)

Reassembly is by using my jewel press to make sure the parts are exactly set at 90 degrees when they are put back together…

GMT gear is re-installed on the movement and the rest of the parts are installed…as you can see, quality and parts usage leaves something to be desired…this is why I don’t like these bastardized movements, no parts, poor quality of the modifications, and hard to get working after overhaul….

In pieces after service…

Overall view.

IMPORTANT POINT

This is not a GMT movement, it may look like a GMT, have the correct hand stack, but it doesn’t work like a GMT model…

Why???

As with all our reps that take a movement designed to do one thing, and add new parts to make it work like something else (Daytona Running seconds at 6 for example), there are compromises. The extra gears that make this ETA 2836-2 into a GMT model means that although it looks like a GMT and the hour hand is independently adjustable, it wont’ work correctly if you independently change the hour hand. Mechanically the GMT gear is actually taking the place of the hour gear, with an added step down gear (to slow the GMT hand to half the hour hand speed, i.e. one GMT turn for each 2 turns of the hour hand).

Here’s what I am trying to explain, but first lets back up a bit to assembling a Date movement, before we get into this model…it will be easier to follow.

When you put the dial and hands on a Date movement (which the 2836-2 is), you need to do one important thing, that is to pull the crown to the 3rd position (the time setting one) and advance the crown (and therefore the time) until one thing happens: the date switches over…

As soon as the date switches, it’s Midnight, and you don’t move the crown anymore. At midnight the hands (hour and minute) should be pointing at 12. So you install the hour hand pointing at 12.

The next hand to install would be the GMT one. The GMT hand is independent and indicates…well GMT time….where it’s installed on the dial – does not matter except for one thing. That is that the GMT hand has to line up with a Marker on the dial. Which marker is irrelevant, as long as you line it up on one of the hour markers. In this example, I lined up and installed the GMT hand with the 9 marker. Since the GMT hand is independent of the Hour and Minute ones, and the GMT hand does one revolution every 24 hours, it’s installation criteria is only to line it up with one of the hour markers.

After the GMT hand is on, I install the minute one, again lining it up with the 12 position like the hour one, after all it’s MIDNIGHT on the movement. Remember that as long as I don’t move the crown in the 3rd time setting position, it’s still midnight.

So far so good…well that is until you start messing around with the hour hand…independently of the others (minute and GMT). You see on this watch, if you move the crown to the middle position (date change AND Hour hand position), you can turn the hour hand without affecting any other part of the movement… This is important to note, as I said earlier that only the 3rd position of the crown actually moved the date change time…so if that is true, then moving the hour hand – with the crown in the 2nd position – would…not affect the date change time.

On any movement with a date, the only way the date change gear moves is by:

1. the watch running

2. you pull the crown to time setting position and advance the time.

In the time setting position, all the gears on the top of the movement turn, cannon pin, hour wheel, GMT wheel, minute wheel, transfer gears, date change gears etc…. this only takes place in the time setting positon.

So back to our modified movement. It stands to reason, that if the only time the date change gears move, is when your in the time setting position, then if you decide to move the Hour hand by itself in the second position of the crown, the date change gears are NOT going to move.

As you see this dial here, it has just turned to 12pm midnight…the date switched over, I put the hands on and lined the hour and minute with the 12…the GMT is on 9 as that is a good position to install the hand and line it up with the dial markers….

If I push the crown in to the second position, and turn the HOUR hand alone, what time is it as far as the movement is concerned??? If you said anything but midnight, you didn’t follow along… As far as the movement and date change gears are concerned, nothing has changed, Midnight it is… As the hour hand moves along alone, the movement is not moving at all, it’s frozen at midnight… If you move the hour hand back by 3 hours to indicate 9PM on the dial face, the date will change over at – you guessed it – 9PM. I hope this is clear and easy to understand.

If you want to use this watch as a GMT watch, then you can certainly set the Hour hand independently to wherever, as long as you don’t care when the Date changes over…every hour off of “Midnight” you move the hour hand, is an hour away from “Midnight” that the date is going to switch over…it’s designed this way…I don’t build them, only try and fix them…

This is probably the reason ETA GMT movements have the GMT hand move independently in the second crown position…that way the date change time stays at Midnight.

Not sure on the Genuine Rolex and how it works, but I suspect they have designed it to work the right way as a true GMT model, and retain the midnight date change with the Hour hand independently adjustable, or maybe not, I don’t really know…

As with all these reps, they may LOOK like the real deal, hand stack and all, but for $200, and a Basterdized 2836-2 QUAZI GMT movement put together with case clamps and some new gears, it’s not going to work like the real watch costing $5000 or more…it can’t… expecting it to do otherwise is unrealistic.

Thanks for reading,

RG

[By-Tor]

Another excellent writeup Z.

This answers all the typical questions. I have the old version of this movement (GMT hand under the hour hand). While it's not a real GMT it has worked flawlessly for the last 16 months. I've never excepted functional perfection of a $4000 watch for $180.

[iglutikut]

Thanks for the write up, not sure if this will interest you or not but I have had this bookmarked for sometime now.

http://fejack.net/misc/gmt24/gmt24.htm

[r11co]

As with all these reps, they may LOOK like the real deal, hand stack and all, but for $200, and a Basterdized 2836-2 QUAZI GMT movement put together with case clamps and some new gears, it’s not going to work like the real watch costing $5000 or more…it can’t… expecting it to do otherwise is unrealistic.

Well well! My phraseology for these crud modified movements is catching on I see

I remember when this latest 'GMT' 2836 movement with 'adjustable hour hand' broke cover in the Omega SMP, and I dissed it big time saying I'd choose the Asian version of the watch as it was otherwise identical and at least came with a proper functioning GMT movement as opposed to a bodged ETA. I was almost burned at the stake at the time for daring to suggest an Asian was a better choice over the 'Swiss'.

I hope that, seeing those mangled case clamps holding the hour gears on, those people will come back here and apologise. Feel sorry for the peeps with this rubbish in their watches though. The 'Swiss' supporters should be apologising to them too....

[Guest avitt]

Here's a suggestion that may make the watch more usable as a GMT. When assembling that hand stack, align the GMT hand with the rest of the hands at 12:00. This way, the date will always turn over with the GMT hand.

With my Explorer II, I use the GMT hand to indicate "Home Time" when traveling. I set the independently adjustabe hour hand to the time where I am traveling. While it's true that the date will not change with the hour hand, it will change with the GMT hand, so I always know date back home.

When I return home, I just reset the hour hand to local time, and everything is synchronized again.

This is actually a pretty good solution, as long as one is traveling within a reasonable time zone range (say, +/- 3-4 hours). When you start taking trips to locations where the time is 12 hours different, and the date is changing at noon, you might want a better option (Digital always works).

[kingkitesurf]

Thanks Zig. Great review. Nice pics.

I recently operated mine and I agree with Avitt. I did exactly the same. Lined all hands (inc GMT) up at 12 when datechange took place. GMT is always home time, I think that goes for the gen as well.

Furthermore in my Explorer2 the cannon gears are held together with a much better piece than the two bend over metal pieces. With mine (Andrew Expl2) it is a tiny fork overrunning the whole gear set. Further I took the hands of twice and the GMT hand came of without the tube. Maybe there is a difference between 2836-2 GMT emulating movements. The movement has been working fine so far. I took the dial of to paint the hour markers black.

[RWG Technical]

Here's a suggestion that may make the watch more usable as a GMT. When assembling that hand stack, align the GMT hand with the rest of the hands at 12:00. This way, the date will always turn over with the GMT hand.

With my Explorer II, I use the GMT hand to indicate "Home Time" when traveling. I set the independently adjustabe hour hand to the time where I am traveling. While it's true that the date will not change with the hour hand, it will change with the GMT hand, so I always know date back home.

When I return home, I just reset the hour hand to local time, and everything is synchronized again.

This is actually a pretty good solution, as long as one is traveling within a reasonable time zone range (say, +/- 3-4 hours). When you start taking trips to locations where the time is 12 hours different, and the date is changing at noon, you might want a better option (Digital always works).

Good idea to use the GMT as "Home" instead of GMT time, I would never have thought of using it to indicate anything but GMT time...

It's always a compramise with these modified models, looks fine, but workings leaves something to be desired.

The only GMT model I have is in my Genuine Glycine Airman 7, and I dont' wear it very often, as it's soooo friggen big....

RG

[RWG Technical]

Well well! My phraseology for these crud modified movements is catching on I see

I remember when this latest 'GMT' 2836 movement with 'adjustable hour hand' broke cover in the Omega SMP, and I dissed it big time saying I'd choose the Asian version of the watch as it was otherwise identical and at least came with a proper functioning GMT movement as opposed to a bodged ETA. I was almost burned at the stake at the time for daring to suggest an Asian was a better choice over the 'Swiss'.

I hope that, seeing those mangled case clamps holding the hour gears on, those people will come back here and apologise. Feel sorry for the peeps with this rubbish in their watches though. The 'Swiss' supporters should be apologising to them too....

I have been using the term basterdized for these models since they first came out...they are not my favorite due to the mods...

Not sure how long they will keep working, I really have a problem with the mods, and the way the gears are held in place...very poorly made...

RG

[RWG Technical]

Furthermore in my Explorer2 the cannon gears are held together with a much better piece than the two bend over metal pieces. With mine (Andrew Expl2) it is a tiny fork overrunning the whole gear set. Further I took the hands of twice and the GMT hand came of without the tube. Maybe there is a difference between 2836-2 GMT emulating movements. The movement has been working fine so far. I took the dial of to paint the hour markers black.

I have never seen any other models than the one I have shown here...seems as if yours is a much better made one without some of the flaws and defects this one has.

RG

[ubiquitous]

Great write up, Rob! It's always interesting to see the inner workings of these modified movements, and also to see how the 'maufacturers' did things. Your posts enable us to see such things, and to me, bastardized or not, it's all fascinating!

I do like Avitt's idea of using GMT as home time... And it also provides a good indication of when 'midnight' is on the movement. So... If the hour hand ever gets out of synch, you know where to set it to put it back to baseline....

Good stuff, as always!

Best,

R

[kingkitesurf]

Rob,

I will try to take a picture of my Expl2 modded 2836-2 GMT movement so you can see the different items. I am really curious what your first impression is on that one...

[TK471]

This is an extremely enlightening write up. I've got a 63 with the 2836-2 movement as the GMT. I just sent it off to be fixed. It came back working, and worked perfectly for about a week, then started to fall behind. I think what he said is happening (and forgive the very layman's explanation here) is that there is a "clutch" that keeps the GMT hand moving, and it appeared to be slipping.

The GMT can be independently set, and it will keep good time for a couple of hours, then it just sort of slows down.

[t]

if you like REAL gmt movement

look my decorated eta 2893-2 movement in my shop

[Guest avitt]

This is an extremely enlightening write up. I've got a 63 with the 2836-2 movement as the GMT. I just sent it off to be fixed. It came back working, and worked perfectly for about a week, then started to fall behind. I think what he said is happening (and forgive the very layman's explanation here) is that there is a "clutch" that keeps the GMT hand moving, and it appeared to be slipping.

The GMT can be independently set, and it will keep good time for a couple of hours, then it just sort of slows down.

That's pretty interesting...The first "real" rep that I bought was one of Paul's Seamaster GMT's. It exhibited the same bahavious (GMT hand was losing time). I brought to my local watchmaker for an overhaul. When I got it back, he told me that he had to actually fabricate a part, which he called a "clutch". I didn't really know what he was talking about, but all hand have been keeping perfect time for almost 2 years now.

[RWG Technical]

This is an extremely enlightening write up. I've got a 63 with the 2836-2 movement as the GMT. I just sent it off to be fixed. It came back working, and worked perfectly for about a week, then started to fall behind. I think what he said is happening (and forgive the very layman's explanation here) is that there is a "clutch" that keeps the GMT hand moving, and it appeared to be slipping.

The GMT can be independently set, and it will keep good time for a couple of hours, then it just sort of slows down.

I am guessing that your model has the GMT hand closest to the dial face. I am trying to remember how it's done on that model (the GMT gear to movement connection), typically the GMT gear (or on the one I reviewed the Hour one) has notches in it (which act sort of like a clutch). It's doubtful that these notches would be slipping (you can feel how much tension is required to overcome the "clutch" as you change the GMT time when you set the hand), I would have to guess that your problem would be more related to the hand slipping on the post, than the GMT gear slipping...only a guess but seems more likely.

There is little room on the 063 between the hands and dial, GMT hand getting out of sync with the rest is fairly common and whenever I have noted it, it was the hand hanging up on the dial face...

RG

[TK471]

I am guessing that your model has the GMT hand closest to the dial face. I am trying to remember how it's done on that model (the GMT gear to movement connection), typically the GMT gear (or on the one I reviewed the Hour one) has notches in it (which act sort of like a clutch). It's doubtful that these notches would be slipping (you can feel how much tension is required to overcome the "clutch" as you change the GMT time when you set the hand), I would have to guess that your problem would be more related to the hand slipping on the post, than the GMT gear slipping...only a guess but seems more likely.

There is little room on the 063 between the hands and dial, GMT hand getting out of sync with the rest is fairly common and whenever I have noted it, it was the hand hanging up on the dial face...

RG

I can only respond with a bit more speculation. Stating the obvious, the guy said it was clear that the movement was modified so that the day function would control the GMT.

What's interesting is that I can take the main time hands and move it around the dial, and the GMT tracks just as it should, keeping a three hour lag, for example. I can do this for several days' worth. All the way around, the main time works around the dials twice, the GMT once, with the date changing over at midnight. I don't know why it would work when wound manually, but not work on its own. But, that's exactly what's happening.

I'm not too worried about it though, Zig. This is a watch that I've contacted you about before. I've got a crown on the way from Palp, so I'd planned on sending it to you for relume and maintenance once that 183 you've got is finished, and once I've got the crown in. There are enough people on the board with the 63 that they might like to see what's really causing the problem.

[kingkitesurf]

Rob,

As promised here are the pics of my TWG Expl2 with 2836-2 GMT mod.

[dabom]

Good idea to use the GMT as "Home" instead of GMT time, I would never have thought of using it to indicate anything but GMT time...

It's always a compramise with these modified models, looks fine, but workings leaves something to be desired.

The only GMT model I have is in my Genuine Glycine Airman 7, and I dont' wear it very often, as it's soooo friggen big....

RG

Hi Rob,

thanks for the excellent break-down of this movement (no pun intended). I also set the GMT hand to home time, that's why when I wrote to you I said that the date now changes at 1800 - since you set the GMT hand at 9 O'Clock marker (or 18 on the GMT hand, which = 6pm).

When I travel, I set the hour hand to the timezone of the holiday location, so at all times I can see what the time is at home by looking at the GMT hand. Then when I go home I can reset the hour hand back to home time without ever needing to look at another time source. I don't particularly care what the time it is in London, so the GMT hand is never set to Greenwich, it's always set to home time.

So whilst I am at home - the GMT hand and hour hand are "synchronized". If the hour hand is pointing at 6 O'clock, the GMT hand is pointing at 6 O'clock or 1800 hours. Since I always have the hour hand synchronized to the GMT hand at home, the date will therefore change over at 6pm (since you set the GMT hand pointing at the 9 marker, or 18 on the GMT bezel).

[dabom]

Not sure on the Genuine Rolex and how it works, but I suspect they have designed it to work the right way as a true GMT model, and retain the midnight date change with the Hour hand independently adjustable, or maybe not, I don’t really know…

As with all these reps, they may LOOK like the real deal, hand stack and all, but for $200, and a Basterdized 2836-2 QUAZI GMT movement put together with case clamps and some new gears, it’s not going to work like the real watch costing $5000 or more…it can’t… expecting it to do otherwise is unrealistic.

Thanks for reading,

RG

This is how a real Rolex works:

1) Crown pulled to first click: moves hour hand forward or backward 1 hour. Date can move forward AND backward 1 day once the hour hand passes "midnight" (two revolutions of the dial). The date is synchronized to the position of the HOUR hand.

2) crown pulled to second click: all hands move together (along with the GMT hand).

This is how the rep works:

1) Crown pulled to first click: moves hour hand forward 1 hour, or the date forward 1 hour.

2) crown pulled to second click: all hands move together (along with GMT hand). Date is linked to the GMT hand position (date will always change at the same GMT time).

The crucial difference is that the real Rolex can change date forward and backward, the hour hand can move forward and backward one hour, and the date is linked to the hour hand; whereas for the rep the date is linked to the GMT hand and the hour hand can only move forward.

Like you mentioned in your article, it doesn't matter where the hour hand is positioned on the rep, as it has no bearing on when the date changes. The date and hour hand are completely independent of each other. The date is in fact synchronized to the GMT hand, as regardless of how you set the hands - the date will ALWAYS change at the same GMT time. So if you put the GMT hand pointing at the 9 O'clock marker (= 18 GMT marker), the date will ALWAYS change when the GMT hand is pointing at the 9 O'clock marker regardless of where the other hands are.

You are right when you say it is bastardized - I think the rep is a classic case of form before function. The rep is only designed to "look" like the real thing, with none of what truly makes the genuine watch an awesome watch to wear.

Would you want to drive a car that looks like a Lamborghini, but has the engine of a Lada? Personally I'm through with watches that have this movement!

[Guest avitt]

This is how a real Rolex works:

1) Crown pulled to first click: moves hour hand forward or backward 1 hour. Date can move forward AND backward 1 day once the hour hand passes "midnight" (two revolutions of the dial). The date is synchronized to the position of the HOUR hand.

2) crown pulled to second click: all hands move together (along with the GMT hand).

This is how the rep works:

1) Crown pulled to first click: moves hour hand forward 1 hour, or the date forward 1 hour.

2) crown pulled to second click: all hands move together (along with GMT hand). Date is linked to the GMT hand position (date will always change at the same GMT time).

The crucial difference is that the real Rolex can change date forward and backward, the hour hand can move forward and backward one hour, and the date is linked to the hour hand; whereas for the rep the date is linked to the GMT hand and the hour hand can only move forward.

Like you mentioned in your article, it doesn't matter where the hour hand is positioned on the rep, as it has no bearing on when the date changes. The date and hour hand are completely independent of each other. The date is in fact synchronized to the GMT hand, as regardless of how you set the hands - the date will ALWAYS change at the same GMT time. So if you put the GMT hand pointing at the 9 O'clock marker (= 18 GMT marker), the date will ALWAYS change when the GMT hand is pointing at the 9 O'clock marker regardless of where the other hands are.

You are right when you say it is bastardized - I think the rep is a classic case of form before function. The rep is only designed to "look" like the real thing, with none of what truly makes the genuine watch an awesome watch to wear.

Would you want to drive a car that looks like a Lamborghini, but has the engine of a Lada? Personally I'm through with watches that have this movement!

I disagree. This is a very nicely functioning travel watch... See my comments above regarding how to correctly assemble and use the watch to display home/away time.

GMT is a misnomer when used to describe the Explorer II....This is becuase the watch does not have a rotating 24 hr. bezel. However, it could be assembled to display GMT time. (But this would make it much less practical for use when traveling).

I took a close look at the date fonts on my Andrew derived Exp. II, and compared them with the pictures in this thread. Mine looks like kingkitesurf's, so I'm feeling pretty good....

[dabom]

I disagree. This is a very nicely functioning travel watch... See my comments above regarding how to correctly assemble and use the watch to display home/away time.

I'm talking about the makeshift module. They are just unjewelled gears stacked ontop of each other. Over time they will just wear out.

[TJGladeRaider]

I'm talking about the makeshift module. They are just unjewelled gears stacked ontop of each other. Over time they will just wear out.

Personally, I'll never buy another GMT that doesn't have a GMT movement. I have had nothing but trouble out of these bastardized 2836's.

The 2893s are wonderful!

Bill

[kingkitesurf]

I have a TWG correct hand stacked Explorer2 working fine and a PT GMT2 with 'wrong handstack' working super... I have them 6 months approx. Both are adjusted 2836-2. TJ can you explain what the trouble was....

[RWG Technical]

Rob,

As promised here are the pics of my TWG Expl2 with 2836-2 GMT mod.

Great info and pics, I have never seen this model in person.

Thank you.

RG

[TK471]

I don't think there's any question that the 2836 movement is not as good as a 2893 from the standpoint of GMT function, and isn't that what we're talking about here? On the other hand, the 2836 can be regulated to be extremely accurate (see my post about replacing the movements in this forum), so the base movement of good quality and produces solid results. But I think you can expect some problems at some point with the GMT function.

That said, I think it's possible to get the GMT hand working properly, provided someone (like The Zigmeister) who knows what they are doing gets their hands on it. The 2836's only inherent inferiority to the 2893 is that one was designed as a GMT, the other not. Granted, in this thread, that's a MAJOR inferiority.