I have just finished a rectification on a Rolex fitted with a 3135 movement which has a free sprung balance.

On previous watches I've worked on, the adjustment of beat error was either done by rotation of the collet on the balance staff (Omega 161 etc.) or adjusting the length of the hairspring (Seiko 7S26 ec.).

On looking at the 3135 it was obvious that the length of the hairspring was set at the factory, so, in effect, the beat rate was fixed.

Correct or not?

The rate, or how fast the oscillator spins, is controlled by two weights that must be adjusted out to slow rate, or screwed in to increase rate. I simplistically see this as the "ballerina principle".

Now on to my questions.

Just for interest (I had no intention of "servicing" the movement) I put the watch on my timegrapher at full wind and checked readings at Full Wind. I then waited 24 hours and then did the same checks and took snapshots of the readings as time passed.

I noticed that the watch is running at a consistent gain (although very minor) where I would have expected one or more of the positions to lose time which would allow for "user regulating".

I've attached screenshots for six positions at full wind and the same at 24 hours later and would welcome comments from @Archer, @ChrisN, @Canuck, @François Pépin et al.

What do you interpret from these readings?

(Purely for the sake of my continuing education).

At Full Wind:



And the readings after 24 hours.

 

I realise that this isn't a very scientific assessment, the movement hasn't been serviced for a very long time but I was interested to see if the results indicated any problems.

By the way, the power reserve from full wind to dead was a respectable 47.6 hours.

 

The stud carrier on this model is held in place by a "bib" on the balance cock. This bib is secured by a single screw. Loosening the screw a bit allows the stud carrier to be turned to bring the watch to beat. Tighten the screw. The screen shots tell me that you have no concerns about adjusting the watch for beat error. Otherwise, I suspect the watch is performing well within limits.

 

Sorry I rarely visit this area of the forum, so missed this thread.

I think the main subject you are asking about is isochronism, rather than beat error...beat rate and beat error are different things. As Doug noted, the adjustable stud carrier has to be unclamped on the 3135 before you can make a beat error adjustment.

So for the rates and how they change over time, if you imagine regulating pins and how they act upon the balance spring al full wind, you should note that depending on how far apart the pins are, they may not be touching it as the watch winds down some. This is why a free spring balance has some advantages with regards to isochronism over a watch that uses regulating pins.

Overall this watch appears to be running well in terms of timing. Delta of 5, average rate of +3.7 at full wind is well within tolerances. Rolex doesn't set out a target rate for their watches like Omega does, so as long as the average rate is within the often quoted COSC parameters of -4 to +6, it passes. Omega specifies the range for average rate at -1 to +6, with the target rate at +3.

Also note that the particular amplitude the watch is at for most of the second set of vertical readings is in the area where positional variation due to poise errors are minimized, so that is certainly helping that second set.

So if we look at the timing tolerances for a typical Omega COSC watch, they express them in a way that let's you get a good feel for how much the variation is allowed to increase due to isochronism. At full wind (0 hours) the delta (positional variation) is allowed to be as much as 12 seconds over 5 positions (crown right is left out). At 24 hours after full wind, the delta is allowed to be 15 seconds over 5 positions, so the increase in variation is only 3 seconds across 5 positions. So this is an increase of 25% in delta over the first 24 hours, so not a huge increase. Most well adjusted watches (using regulating pins or not) will only have a small increase in variation due to isochronism in the first 24 hours. It's after that when the timing starts to drift more with a watch of a "typical" reserve is involved (approx. 48 hours).

And yes the method of regulation is the timing screws on the balance. These are Microstella screws and as you noted in the other thread they require a special tool to adjust the timing. These can be purchased on the open market very easily - plenty of genuine and aftermarket tools out there. However regulating this kind of watch takes a little more care than with one where you can turn a fine adjustment screw, or just push the regulator around. The screws must be turned exactly the same amount, or you will introduce a poise error into the balance, and you will have more positional variation as a result. Of course since you are working on the balance, care has to be taken with the pivots and of course the balance spring itself, so no inadvertent damage occurs.

Adjustable mass balances come in many forms, so the screws on many modern Omega watches are square head instead of star shaped as Rolex uses. Patek uses the Gryromax system, and even the very inexpensive Tissot Powermatic 80 that has the synthetic escapement uses a system similar to the Gyromax, although the execution is very crude in comparison.

Hope this helps.

Cheers, Al

 

The position indicator in the picture is pretty neat.
Watch is performing well.
I'm not sure what you mean by user regulating (put it a specific position when not wearing the watch?).
Do companies like Rolex and Omega adjust for a slight gain to compensate for effects of aging between service?