My First Experience of a Rolex Movement- 3035 Full Service

I am now in the final stretch of the WOSTEP course and preparation for our final exams in July has begun. The final exams involve a written element and a practical element. The written element is theory paper that will test our knowledge and understanding of horology however it mainly focuses on the technical aspects that enable a timepiece to function. The practical element culminates in servicing 3 watches within a two day period, a quartz, an automatic and a chronograph. Of course these watches will contain faults that need to corrected because the exam is designed to test the watchmaking skills that have been learnt through the duration of the course.

In order to prepare us for the practical element of the final exams we have now begun to practice and perfect our servicing skills. Throughout the course we have had the chance to service a few movements namely the ETA trio of 6497/8, 955.112 and 2824 however most of the focus has been on the 6497/8. It’s origins as a pocket watch calibre make it ideal for learning for a number of reasons, it’s sheer size being the main advantage as most components are double the size of those found in the majority of watch movements.

The large size of the 6497/8 presents a whole host of advantages when trying to demonstrate how a watch movement works, how to inspect it and how to adjust it all, to a group of students that have mostly had no experience of any of these things. The WOSTEP course is clearly a well thought out training course and for this reason the 6497/8 features heavily throughout the course until the final stages. The reasoning being that the basic skills of a watchmaker are developed on this calibre before being transferred to the normal size calibre’s that are common throughout the industry. I have to say that this method works really well, I couldn’t begin to imagine starting the course with a 2824 or similar, trying to learn everything would have been a real struggle!

We had only serviced the 2824 and 955.112 a handful of times of each so practice as part of course on these normal sized calibre’s has been pretty limited. A number of us, including myself, have undertaken project watches of our own so we have gained some experience of working on smaller and more diverse calibre’s but our skills are clearly far from perfected. We should be able to service these calibre’s in a number of hours but mainly due to our lack of experience it has often taken days and in the case of some vintage watches, weeks!

Fellow student, Nikolai, hard at work…


The path leading up to the final exams gives us the opportunity to service a diverse range of quartz and mechanicals movements, ranging from a basic quartz to an automatic chronograph with triple date and moon phase. The experience will hopefully further develop our servicing skills and dramatically increase our speed. The prospect of getting the chance to work on all these different calibre’s is pretty exciting and it’s the part of the course I’ve been looking forward to the most. My day to day role when I return to my employer will be doing exactly what we are practicing now so it’s good preparation for that too. Although I hope my career path takes a slightly different direction in the future.

To start us off on this path of practice we spent a couple of weeks servicing the 2892, 2824 and 255.411 respectively. I’ve decided not to cover these though because I’ve posted about both the 2824 and 2892 before when working on my own projects and there isn’t much more I can add. Likewise the 255.411 is similar to the 955.112 which I have also covered before and there isn’t really many parts to a quartz watch. All ETA movements are pretty similar in construction too and for that reason it’s quite simple to move from working on one calibre to another.

The third week of practice presented us with an all together different challenge and one, that to me, was also much more exciting than the rather mundane ETA stuff we had spent our time on thus far. The calibre placed in front of us was the Rolex 3035, the predecessor to the current calibre 3135, it was a development of revered 15xx series and the first high-beat (28,800 BPH) Rolex movement. The 3035 boasts a free-sprung balance with Breguet overcoil and Rolex’s own microstella regulating system. It also contains an instantaneous jumping date display.

Rolex Calibre 3035…


Caliber 3035 was a quite radical departure from the previous 15xx series movements in that the balance speed was increased from 19,800 BPH to 28,800 BPH. A high-beat movement is generally perceived to provide more accurate timekeeping due to its ability to maintain a stable rate even when there are external influences, such as a shock. I believe that each Rolex calibre produced since the 3035 has stuck to the same high-beat of 28,800 BPH which is common in today’s watch industry.

The balance screws from the 15xx were ditched with the exception of the four timing screws which were now placed on the inside of the balance rim in. There positioning enabled the diameter of the wheel to be as large as possible, creating an increased moment of inertia without a gain in its mass and also a reduction in aerodynamic drag. The four timing screws, referred to as the Rolex microstella system, are used to both poise the balance and regulate the movements timekeeping. The screws have a specially shaped head which require the correct key to be used in order to adjust them.

Rolex also followed ETA’s lead by incorporating micro-gear toothing in this movement for the first time, in order to reduce both the free play between the gearing and to simultaneously reduce the friction of the glucydur teeth rolling on the steel pinions. They also made the switch to a fast rotating barrel in order to improve both the torque and stability of the drive train. The 3035 barrel makes approximately one turn every five hours, and runs down after about ten turns, giving the movement a power reserve of approximately fifty hours.

Rolex began using a free-sprung balance in combination with a Breguet overcoil a long time ago and its something that even to this day ETA still choose not incorporate into their movements. ETA continue to use a regulated balance with a normal hairspring which is adjusted using their ETAchron system. There are advantages and disadvantages to both free-sprung and regulated balances which I will now explain.

A free-sprung balance has a fixed-length hairspring, the rate is adjusted by changing the inertia of the balance itself by moving weights (screws) either out from the center (increasing inertia and slowing rate) or in towards the center (decreasing inertia and increasing rate).

Rolex free-sprung balance with Breguet overcoil…


The main advantages of this system are a more stable rate, once its properly timed, and better positional accuracy. This is because the hairspring, particularly when a Breguet overcoil is used, is free to breathe concentrically as it contracts and expands with each oscillation of the balance. This concentric breathing means that the watch’s rate changes less when put into different positions. There is also no regulator to accidentally knock out of position.

The main disadvantage to the free-sprung balance is the difficulty involved during regulation. The rate can only be changed by adjusting the weights on the balance and any non-symmetric changes will throw the balance out of poise. The main disadvantage to the Breguet overcoil is its height, it adds at least 1mm to the balance assembly inevitably making the movement itself thicker.

Regulated balances generally have a fixed-inertia balance and a variable-length hairspring. The hairspring’s effective length is changed by the regulator, as you move the index on the balance cock, the curb pins slide along the hairspring, any movement in toward the centre shortens the spring and increases rate, and vice versa.

Omega (ETA) regulated balance…


The main advantages are the ease of regulation (just the regulator to move) and the fact that you rarely have to adjust the balance itself so it stays poised. The main disadvantage is the influence of the regulator pins on the breathing of the hairspring. The breathing of the hairspring is asymmetric because of this, there are positional rate errors caused by the effects of gravity. In order to achieve high performance this problem has to be addressed.

The conclusion from these advantages and disadvantages of both systems is that a free-sprung balance complete with Breguet overcoil is the more accurate time keeper once correctly regulated. The Regulated balance is easier to adjust but suffers from a greater amount of positional error due in part to the interference of the regulating pins on the hairspring and the lack of an overcoil. Of course the regulated system is much cheaper to manufacture and means it will always be the main stay in lower end movements and to be honest in its modern guises performs pretty well.

Anyway I think I’ve given you enough background on the 3035 and some of the features it utilises so I’ll move on to its step by step disassembly. Along with the basic technical guide, which explains a few of the key requirements for lubrication and adjustment, we were given a few tools specifically designed for this calibre. The first was a movement holder that incorporates a side for the movement when it’s dial down and when for when its dial up. It also provides better support and accessibility compared to the basic Bergeon movement holders we normally use. The second tool was another holder, this time to aid with the reassembly of the automatic module. The final tool, which looks a bit like a fork, is used to remove the calendar wheel nut. It’s important to check components along the way when disassembling as any faults need to corrected before cleaning.

Ready to begin disassembly…


After checking the watch winds, changes the date and sets the time, disassembly can begin. My movement wasn’t running even though all the above were ok so I knew I had a fault somewhere in the gear train or escapement to find along the way. On close inspection I spotted the problem but I’ll come to that later.

The first step is to remove the three screws that keep the automatic module in place and remove the module from the movement.


The automatic module is placed to one side for now and the balance is removed to prevent any risk of it being damaged during disassembly. It’s important to check the end shake of balance at this point although with Rolex it can easily be adjusted later thanks to a clever system. It uses a screw and nut system that enables the balance to be moved up and down at the turn of a screw, the benefit being that the end shake can be adjusted at any point during a service.


Balance in more detail, notice it has been poised by hand a number of times, probably the result of a few balance staff changes. These movements have been used for training over the last few years so do show quite a few battle scars!


The movement is then turned over so the calendar disc can be removed, the retainer is twisted out of the way with a screwdriver and the disc removed.


The calendar disc can then be removed and safely stored.



The movement is then turned back over to continue the disassembly, the pallet bridge and pallets are next to be removed. It’s important to check the depth of lock and the end shake on the pallets. I couldn’t check it at this point due to a problem with the escape wheel…



The end shake of train wheels is checked next and then the train bridge can be removed along with all of the train wheels. Note the escape wheel sits within jewelled end stones, the pivots are conical like on a balance staff and help to further reduce friction. The capped jewel also prevents the oil from being contaminated.




The watch wasn’t running before disassembly and the escape wheel was dancing around like a spinning top! I suspected a broken pivot and I was right. You could still see the pivot in the jewel hole when under a microscope.


The movement is then turned over once again in order to remove the calendar and motion work. The calendar wheel is the first thing to come off, note the special tool required to remove the nut that holds it in place.





The three screws that secure the ring that seats the calendar ring are removed, followed by the ring itself. Note the jewels that the calendar ring rotates against.



The hour wheel and cannon pinion are removed next, followed by the date corrector and date corrector wheel.



Followed by the setting lever jumper and the yoke.



At this point you could remove the rest of the keyless and the stem but its easier to handle the movement with them in so I leave them in for now and turn the movement over. The minute pinion bridge and the minute pinion itself can then be removed.



The ratchet wheel and click spring are removed next.


Followed by the winding bridge, the wig wag pinion and the intermediate crown wheel.



The last of winding work is removed next, the crown wheel, crown wheel core and crown wheel friction spring.


The barrel bridge can now be removed followed by the barrel itself.



The movement can now be turned over again and the remainder of winding work removed. The winding stem, sliding pinion, winding pinion, setting lever and spring. The balance stop can also be removed at the same time.


The balance can now be positioned back on the movement ready for cleaning, the shock settings and escape wheel end stones are also removed at this point.


The barrel can be disassembled.



At this point we can return to the automatic module and also disassembled it for cleaning. Note the special holder and the coated reverser wheels that do not require any lubrication.




Finally everything is disassembled and ready for cleaning. Note that the calendar disc and any springs are not put through the cleaning machine as they would get damaged. The mainspring is also not put through the cleaner.


Unfortunately I didn’t have the time to take any pictures of the assembly but its basically the reverse of the above with lubrication applied to the usual areas. The hairspring can sometimes be pushed slightly off centre during cleaning so its important to correct this when assembling the movement. It’s an easy correction to make when there is a Breguet overcoil, it’s simply a case of pushing against the overcoil slightly in the desired direction. If the hairspring is not centred, the movement will suffer from a large amount of positional error in the vertical positions due to the spring not breathing concentrically.

Ready for assembly…


Now lets get to the timekeeping, due to the combination of free-sprung balance and Breguet overall this movement is able to keep fantastic time. The average timekeeping of my movement was +2 seconds a day and its delta (the difference in timekeeping between the highest and lowest figure) was just 5 seconds so well within chronometer standards. This performance was much better than most of the ETA movements I’d serviced and pretty impressive for a thirty year old design!

I have to say that the quality and finishing of every component is way ahead of anything ETA produce, the whole movement also seems more robust and the systems employed by Rolex clearly improve timekeeping. Overall I have to say that compared to a modern ETA 2892, the 3035 is clearly the better movement, the fact that it was replaced by the 3135 more than twenty years ago demonstrates that Rolex have been producing excellent movements for a long time. Obviously Rolex watches command a higher price point than most brands that contain ETA movements but there are brands out there using the 2892 in the Rolex price bracket. You would definitely be getting a better engineered and manufactured movement with the Rolex.

I really enjoyed working on this movement, not only is well engineered and designed, its also easy to service and produces great time keeping with little adjustment. Well, providing that the previous watchmaker has adjusted the microstella system correctly! This can be tricky to adjust especially without much practice but I’m sure in time it would become easier and quick to do. The instantaneous date change is also a welcome addition, it just clicks straight over whereas the ETA system is semi-instantaneous and takes a while to change. The 3035 is a high end movement that has proved to be reliable over a long period and for that you have to give Rolex huge credit.

I will be tackling a well known chronograph calibre next, it will be the first time I’ve touched a chronograph movement and it seems slightly daunting at this point. The extra components and mechanisms add another dimension to a service. At the same time though I’m really looking forward to it and as has been proved in the past, most things aren’t as difficult as they first seem with practice.

Posted in The British School of Watchmaking and tagged , , , .