Rolex Caliber 3135 – Still worthy of the crown after all these years?
By A. Watchmaker

After reviewing Rolex’s new ladies movement a few years ago, Caliber 2235 review, I have finally acquiesced to Hans’ request and followed it up with this much delayed review of Rolex’s top of the line men’s movement, the caliber 3135. To Hans a huge thank you for this article, because without his help and kindness, and especially his patience, this review would never have happened, at least not from me. This review is based on his own Rolex Sea Dweller model 16600N series that he bought brand new in 1993, and I photographed it during servicing. Fortunately, it wasn’t in bad condition even although it was about fifteen years before its first service, because as we all know Hans owns a few watches, so it hadn’t been worn continuously for much of that time.

Prior to the introduction of this movement Rolex used their caliber 3035 – which was in turn a replacement for their caliber 1575. Although caliber 3035 was a strong, reliable and very accurate movement – the main plates and basic design even being used as the basis for their now defunct quartz movement, caliber 5035 – it was only in use for a little over a decade.

Caliber 3035 was a radical departure from their previous movements in that the balance speed was increased from 19,800BPH to 28,800BPH, the balance screws were done away with, except for the four timing screws, which were now placed on the inside of the balance rim in order to keep its diameter as large as possible, thereby increasing its moment of inertia without increasing its mass, and to reduce aerodynamic drag. As mentioned in my earlier review of the 2235, Rolex followed ETA’s lead in using the new micro-gear toothing in the movement (this was their first movement to do so) and dial trains 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 barrel makes approximately one turn every five hours, and runs down after about ten turns, thereby giving the movement a generous power reserve of approximately fifty hours.

See my review of the ETA 2892-A2 for more info on these type of teeth Review ETA 2892. Curtis Thomson also has an explanation on these special type of teeth as well. Note that some companies, JLC in particular, have also modified the basic ETA style micro-gear profile very slightly and given it their own name. Rather than give an inadequate description of what these teeth look like – see the photos below.

Originally introduced around 1988 Rolex’s 3135 has been their workhorse movement since then.  Of course things haven’t remained static these past 20 years and it has seen numerous improvements and upgrades in order to both refine and improve its performance and reliability. Like most Swiss watch companies Rolex don’t make a big deal of these upgrades, so most of the time even their own watchmakers are not aware of them, unless of course they are specifically instructed to upgrade the parts and movements when they come in for repair. Some of the improvements, like their new Parachrom hairspring, they choose to anchor to new marketing campaigns, in which case potential buyers now want to make sure that their new purchase has all the latest and greatest upgrades. But others, like subtle refinements to pivot and teeth shapes, improvements in the alloys of the metals used etc, often go unheralded.


Basic technical info for the 3135

Diameter = 28.5mm
Casing diameter = 28.1mm
Height = 6mm – note that the non-date Cal. 3130 is the same height
Jewels = 31
Power Reserve = Approximately 50 hours


A closer look at the movement

(pic#1) Overview of the completely disassembled movement. Cleaned and ready for assembly. Movement screws on the top right hand side and dial screws on the top left next to the mainspring.

Overview

(pic#2) A close up of the barrel, mainspring and barrel arbor. Like the ETA 2892, and most modern movements today, Rolex also use the micro-gear toothing throughout the movement, including the barrel and both the movement and dial trains.

Barrel


(pic#3) A comparison of the shorter ETA 2892 mainspring compared to the much longer one of the 3135. This longer mainspring enables Rolex to squeeze a 50 hour power reserve out of the latter. Although the 2892 mainspring is both lower in height and shorter, it is marginally thicker – i.e. slightly stronger.

2892 - 3135


(pic#4) A size comparison between the rhodium plated 2892 barrel and the plain uncoated brass barrel of the 3135.

2892 -3135 size


(pic#5) The mainspring fitted into the barrel, ready to be lubricated and then closed up with the cover.

Barrel


(pic#6) The 3 glucydur train wheels, plus the escape wheel and pallet fork. Note that although the teeth and pinion leaves of the wheels are different sizes in order to facilitate the power reduction from the barrel to the escapement, the diameter for the various wheels, excluding the escape wheel, are almost identical.

train wheels


(pic#7) The movement side of the rhodium plated stamped brass main plate. IWC nickel plate theirs. Other manufacturers, like Lange & Söhne, who use nickel silver instead of brass, don’t plate their main plates at all. Note the beryllium bronze bushing in the center for the canon pinion, and the two brass nuts near the top. The latter two facilitate quick minor adjustment of the end shake of the balance wheel.

main plate


(pic#8) The canon pinion placed in position.

canon


(pic#9) And fitted with its own separate bridge. The jewel in the center of this bridge supports the second wheel and not the canon pinion, as the latter is supported by beryllium bronze bushings both at the top and bottom. It’s interesting to note that like the JLC 889, the canon pinion is driven directly by the barrel. This arrangement provides maximum torque in order to drive the dial and calendar train wheels, but is not part of the drive train.

canon fitted


(pic#10) A clear view of the layout of the drive train wheels without the bridge in place. Fitted adjacent to the stem and crown are the crown wheel, stop lever and crown wheel bridge. This system is unusual in that the not only is the crown wheel under the bridge (a practice usually implemented in very thin watches), but it also has its own separate bridge.

train wheels


(pic#11) And with the bridge in place.

train wheels


(pic#12) The complete movement minus the balance wheel. The winding and wig-wag wheels on the right hand side of the brass ratchet winding wheel facilitate hand winding. The wig-wag swings easily out of the way when the automatic winding system is engaged.

barrel bridge


(pic#13) And with the balance wheel installed. Its diameter is approximately 10mm, about the same size as that of the 7750.

balance bridge


(pic#14) The dial side of the movement with stem and winding system installed, but prior to the installation of the barrel and balance wheel.

dial


(pic#15) The upper half of the canon pinion, which is friction fitted to the lower half, has been installed together with the two set wheels. Setting of the hands is accomplished via the latter two wheels and the sliding pinion.

dial setwheels


(pic#16) A clear view of the layout of the dial wheels prior to the installation of the calendar plate. The latter supports the calendar disc as well as keeping all of these wheels in place. On the left hand side you can see the brass date driving wheel with its steel cam riveted to it. The spring, lever and jewel resting up against it provide the tension and torque necessary to effect the instantaneous date change. The shape of this cam has been carefully calculated (together with the thickness of the spring) in order to limit both the force applied and the amount of travel, so that the date always jumps exactly one day. The small pin near the 12 ‘O clock position on the cam drives the date jumper which fits on top of it.

dial wheels


(pic#17) And here with both the calendar plate and date jumper in place. At the 10’O clock position one can see the finger on the date jumper. The latter strikes the date indicator and moves it ahead one day every 24 hours. The three domed shaped jewels equidistant in the center of the calendar plate cut out, support the calendar disc. Doing it this way reduces friction to a minimum and helps ensure an instantaneous date change at midnight.

dial


(pic#18) For comparisons sake here’s the dial side of a caliber 3130 – i.e. no date. Here one can clearly see the jewel inside the top part of the canon pinion tube. As far as I know, Rolex is the only one to use a jewel to support the upper part of the second wheel pinion like this. The advantage of this is that it’s supported more accurately and with less friction too. Caliber 3035 originally used a Teflon washer, but Rolex discovered that a jewel was much better in the long term because the Teflon eventually deformed and lost its smoothness, and this extra friction greatly reduced the amplitude of the balance wheel.

3130


(pic#19) With the calendar disc installed, the dial side of the movement is now complete and ready to be fitted with the dial and hands. The date finger is between the 18 and the 19 a few minutes after the date has jumped. Near the 3 and 4 of the date indicator disc is the quick set date corrector wheel.

calendar


(pic#20) A close up of the small brass post that the set wheel sits on. The second set wheel sits on a metal post as can be clearly seen in pic 15.

3135 post


(pic#21) A side view of this post shows how severely it can be damaged if the movement is not serviced regularly. This is most likely to happen if moisture gets into the watch, which was the case in this particular instance, due to the fact that the heavier moisture displaces the lighter lubricants.

worn post


(pic#22) To put things in perspective for you, I photographed the 3035 and 3135 set wheels side by side. The latter is on the left, and the much larger and more robust 3035 on the right.

3135 3130


(pic#23) And a side view highlighting the differences in thickness.

3135 3035 thickness


(pic#24) Overview of the automatic winding system prior to assembly. The large wheels and simple design result in the most efficient automatic winding system that I’m aware of. The red reversing wheels are Teflon coated in order to reduce friction and don’t require any lubrication. Only the posts of these wheels require a small amount of lubrication because they are steel, as are the inner wheels that pivot on them. The small wheel below the brass ratchet driving wheel is the winding pinion. This fits onto the rotor axel and is driven by it. The clip at the bottom holds the rotor in place and has a small hole on the left of it for sharp tweezers (preferably brass), so that it can be easily unclipped in order to remove the rotor from the automatic winding unit.

auto overview


(pic#25) The underside view of the fully assembled automatic winding unit and rotor.

axel bridge


(pic#26) And installed on the movement sans the rotor. Note the crescent shaped bridge by the balance wheel. In case of a hard perpendicular blow to the watch, this bridge prevents the edge of the rotor from hitting the rim of the balance wheel.

auto


(pic#27) The complete movement, carefully cleaned and lubricated, ready to be installed in the watch case. This Swiss mechanical marvel of precision micro engineering is once again ready to offer years of accurate and trouble free time keeping.

complete


A few observations:

As far as I am concerned it’s very sad that even with the huge slew of new movements that have been introduced during the 3135’s past two decades of continuous use and refinement, there still aren’t too many challengers to the Rolex triple crown of accuracy, reliability and durability – i.e. its toughness – for example, its ability to withstand the abuse of everyday life that most active people, both men and women, would hurl at it, and still keep on ticking. And not just keep on ticking, but to continue doing so accurately too. The only ones that come to mind are the ETA 2892-2A, the ETA 7750 (including their numerous variations and incarnations) and possibly Omega’s new caliber 8500. But unlike the first two movements, the latter one, notwithstanding the fact that it is their 3rd generation of Dr. George Daniel’s co-axial escapement, doesn’t have the advantage of over 20 years worth of use, abuse and refinement, so it’s still an open question as to its long term reliability. So in my opinion, that leaves only two serious contenders to Rolex’s 3135 throne. Of course I would be remiss not to mention Rolex’s own caliber 2235 for its ability to match the three men’s movements. Which is an astounding achievement given the fact that this is a lady’s movement and a lot smaller in diameter and volume than the aforementioned ones.

Initially I was also going to include the Zenith Caliber 400 in this small selection above, but ultimately decided against it because although it’s virtually on a par with the ETA 7750 in most respects, unfortunately unlike the latter, it is only available in various chronograph options.

I’ll be the first to admit that none of these movements will win any prizes for their aesthetics, or their level of decoration. Plus there are other movements that I have mentioned before that can match these three in terms of accuracy, but I haven’t included them here because they are too fragile to be classified as tough movements, or haven’t been around long enough to prove their long term reliability. A perfect example is the JLC 889. An excellent, accurate and reliable movement when serviced by competent watchmakers, but not known for its ability to withstand abuse Then of course we have a few Seiko and Citizen movements that have an outstanding reputation for their toughness and reliability, but most of them fall short in the accuracy department, and one is fortunate if they are accurate to better than +- 15s per day. As a note of interest that standard was more than acceptable for certified COSC chronometers in the 1960s, but most people today expect better, having been spoilt by the standards of their thermo-compensated, atomic clock adjusted quartz watches. And so the COSC have raised their standards accordingly.


And the winner is…

Of these three movements which one do I like the best? If accuracy is your only criteria, then it doesn’t matter which one you choose as there really is virtually no difference between them in that regard. Sure some individual movements might time out marginally better than others, but overall the differences in time keeping between them is insignificant. All three are capable of exemplary accuracy in all six positions, and do so with a minimum of variation and loss to the balance amplitude. More importantly, they should provide excellent accuracy and reliability under real world conditions too.

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.

I wouldn’t have any qualms about someone who preferred either of the other two movements though. At 8mm high the 7750 is the thickest and ugliest of the three. It also has the noisiest rotor of any automatic watch that I have worked on, but one cannot question its accuracy, reliability and toughness.

The 3135 is the youngest, most sophisticated and best looking of these three and it has many admirable strengths. A longer power reserve and instantaneous date change to name a few. As for the weaknesses of the 3135? In my honest opinion there are only two glaring weaknesses. The first is that the oscillating weight pivots on a steel post that is riveted to it and held in place by two jewels. The small circumference of said post, coupled with Rolex’s simple and efficient reversing wheels and gearing ratios, greatly improves the winding efficiency of the automatic unit. This is probably the most efficient automatic winding system available today. But unfortunately its tiny diameter doesn’t give enough support to the weight to stop it from hitting against the movement plates every time the watch is subjected to even light perpendicular blows, let alone strong ones. I think that an upgrade to an oscillating weight pivoting on ball bearing races, like they’ve done in their new chronograph movement caliber 4130, is long overdue in order to eliminate this problem. It would be even better if they used lubrication free zirconium oxide ball bearings like JLC, PP and others that are doing so today, not only for their strength, but also for their efficiency over steel ones.

The other weakness is something that may or may not manifest itself as readily, depending on the circumstances and how often the watch is serviced. This potential problem is easily understood by any watchmaker who has serviced a lot of these movements. The problem is that the 2 setting wheels under the dial, and the two small and thin posts that they pivot on, can be easily damaged if the lubrication runs dry. If the grease on the canon pinion dries up due to age or moisture in the watch case, the teeth on these small thin wheels will break off. The more severe problem is if the lubrication on the posts runs dry, then the first post will be worn away in no time at all, as shown in the photo above. This is less of a problem on the second post as it is a steel pin that is not riveted into the main plate. So it can be easily replaced if it is worn or damaged. Unfortunately the first post is part of the main plate, and is made of brass just like it. Therefore if this post gets damaged like that, the only way to repair it is by replacing the entire main plate. An expensive proposition at best. It’s worth noting that this was not a problem on Rolex’s older caliber 3035 because the diameter of the post was quite a bit thicker, as was the set wheel itself. Please note that this shouldn’t be a problem for those who take care of their watches and have them serviced at regular intervals – every four to five years as recommended by most factories today. I am 100% in agreement with this recommended service interval.

Whether Rolex is interested in making any more radical upgrades to this movement is anybody’s guess, but my feeling is that by now they probably have a replacement waiting in the wings for it. With the economy being in the shambles that it’s in right now, the timing might not be right just yet. Only time will tell. But hey, if anybody gets an email from Bruno Meir about their new movement be a sport and tell your friends on Time Zone about it.


Final thoughts:

A short note about Rolex’s latest innovations – their Parachrom hairspring and Paraflex shock absorber.

I’m not sure why Rolex went this route instead of going with silicone hairsprings like the Swatch Group, JLC etc. This project is a joint venture between Rolex and them, and with the money that they have invested in it, they obviously have the same right as their partners to use this technology. The only reason that I can proffer for them not going with the silicone hairspring, is that they might be obliged to give up using their Breguet overcoil in it.

As an aside, most people aren’t aware of the many joint ventures that the major Swiss watch companies undertake. From a financial point of view it makes a lot of sense if one compares Intel’s R&D budget (or even a much smaller company like Apple) to Rolex’s, one will understand that it makes a lot of sense to pool limited R&D resources with the Swatch Group and Richemont etc.

Having said that though I commend them most highly for finally coming up with a 100% non-magnetic hairspring material. It’s also worth noting that in most of the balance wheels fitted with the Parachrom hairspring that I have seen, they have elected to use the brass Greiner collet instead of their usual metal one. As I mentioned in my review of the 2235 I have a preference for it because it is also 100% non-magnetic.

As far as their new Paraflex shock absorber is concerned, there’s not much I can say about it without putting it though its paces. And no, I have no intention of dropping a $5,000 watch ten feet onto a concrete floor in order to test its efficacy. Especially one that doesn’t belong to me. I know that Rolex has done these tests, but I haven’t personally witnessed them, or examined the watches afterwards. Only time will tell if both their new hairspring and shock absorber are as good as what they claim to be, but I can vouch for the non-magnetic capabilities of the former. Of course if Rolex send me a sample to test, I’ll be more than happy to do so for free!


Dissecting the Helium Escape Valve:

In the first photo you can see the disassembled helium escape valve placed in order of how it fits together. On the left is the piston with its “O” ring gasket. This moves out (i.e. to the left) if the pressure inside the case exceeds the tension of the spring holding it in place. Next is the body of the valve which screws into the watch case from the outside via the slots on its right. The flat portion on the left where the threads end accommodates a flat gasket. The spring rests inside the body of the valve, and this ensures that it is water resistant even without any pressure being exerted on the piston from the outside. Last is a flat metal washer. This is firmly friction fitted to the right hand side of the piston which is tapered in order to accommodate it. The washer also provides a base for the spring to rest on, thereby pulling the piston firmly to the right with enough pressure to squeeze the gasket thus ensuring the water resistance of the valve.

Helium 1

The second photo shows the piston and its “O” ring gasket on the left. On the right is the body of the valve showing the recess that the piston and its gasket fit into. The flat part of the piston and the valve should be flush with the watch case when installed correctly.

Helium 2


This is a simple and reliable over pressure relief valve that should perform exactly as designed. The whole unit should be replaced, together with all the other gaskets, every time the watch is serviced, in order to ensure the water resistance of the case. My only minor criticism is that the gaskets are regular nitrile rubber and not the superior longer lasting Viton gaskets that Sinn is now using in their dive watches. Viton is a synthetic rubber developed by Dupont. Although slightly less flexible, it was developed to withstand high oxygen environments and other hostile chemicals and gases that will deteriorate natural rubber gaskets in short order.