Good article:

http://home.watchprosite.com/show-f...t/fi-17/pi-3662881/ti-587501/t-horological-meandering-free-sprung-vs-regulated/

....For a free-sprung system, the main advantages are reported to be stability of rate, once properly timed (no regulator to knock about, as Steve said), and better inherent [author's emphasis] positional accuracy. This is because the hairspring, particularly when paired with a Breguet overcoil, is free to "breathe" concentrically as it contracts and expands with balance oscillation. This concentric breathing theoretically means than the watch's rate changes less when put into different positions.

The main disadvantage to the free hairspring is the difficulty in regulation. Rate is changed only by fiddling with the balance, and any non-symmetric changes will throw the balance out of poise (like a wobbly bicycle wheel). The main disadvantage to the Breguet overcoil is its height: it adds something like 1mm to the balance assembly.....

A regulated balance can be made quite indifferent to position, so the possibility of lower positional variation may not out-weigh the difficulty in regulation.

 

So...what's the deal? I would think that for a 20k dress watch with no complications, no expense would be spared in terms of perfecting the movement.

I would argue that you have not gotten to the point of no expense spared, even on a three hand movement, at that point. Manufacturers at that price point are still adding more and more hand finishing to the movement etc. vs trying to get that last bit of accuracy. The point being accuracy doesn't sell high end watches, high levels of fit and finish do. Also don't discount doing something the "hard way" better which may or may not have anything to do with this situation. Look how long many of the high end brands stayed with low beat movements, certainly there were other issues at work, but they took pride in making the low beat movements more accurate than the high beat movements of other manufacturers.

 

Hi,

All 3 are superb watches , for mine I would prefer the regulated because it offers a more simple easy and elegant solution. But seriously would be happy to have any one of them.

Regards,

 

this has always been my line of thought as well. seeing all those screws sticking out of the balance wheel just looks less elegant, and less advanced, than a regulator with an intuitive lever to move.

 

Accuracy isn't the point of mechanical watches; they're good enough, certainly, but there is absolutely no point in trying to go further in a world that contains thermocompensated quartz, GPS-sync and pretty soon chip-scale atomic clocks. So, what is the point? Art. Once you understand that a mechanical watch is a kinetic sculpture that happens to be useful, these questions make a lot more sense.

 

Art is certainly part of the reason for adding features like a free sprung balance that have at best a theoretical impact on performance. I can look at some movements and just smile and enjoy the design and fine machining. But features like that also allow manufacturers to craft a distinction between watch movements that are otherwise all but identical.

 

....So...what's the deal? I would think that for a 20k dress watch with no complications, no expense would be spared in terms of perfecting the movement....

I suppose that this would suggest that they should all contain silicon escapements and hairsprings, too, as well as Breguet overcoils. To your question, since one can obtain COSC specs with a traditional regulation, maybe it's a matter of weighing the marginal benefit of a fsb versus its difficulty to adjust and service cost to owner. As mentioned, some of the 'perfecting' is in the aesthetics and not the performance.

 

I don't know all the details, but some of the most accurate watches in the world have been made with smooth balances, so to the degree FSB helps, it must not make a night and day difference. It's worth remembering that Grand Seiko beat essentially all mechanical watch movements in the world, including one-off custom made prototypes, in 1968 with a smooth balance. That doesn't mean it's better, it just tells me that it can't be the whole story.

I'm also curious about how silicon hairsprings may affect the need for free sprung balances and Breguet overcoils in the future since you can have variable width.

In the early days, you would actually have both a regulator and timing screws\collets for fine adjustment.

 

I'm also curious about how silicon hairsprings may affect the need for free sprung balances and Breguet overcoils in the future since you can have variable width.

Indeed, being able to adjust the width of the hairspring along its length should in principle allow you to more closely approximate a linear spring, and thereby improve isochronism. I should assign this as a problem in a mathematical modeling and numerical optimization class...

 

Thanks for the share OP - I missed that article first time round

 

It's a good read but maybe what we ought to be taking away from it is not that we should be disappointed that many legitimately high end watches have smooth balances, but that smooth balances are apparently so good that they can be used in legitimately high end watches.

I certainly don't see the forums filling up with people who are disappointed in their Grand Seikos or Zenith El Primeros in the accuracy dept., or even people with more affordable movements like 2824 chronometers. Yet we do see a number of people disappointed with their free sprung, Breguet overcoil Portuguese Automatics (in terms of accuracy). There's just so much going on, I don't think we can get hung up on a single design decision.

 

I've seen on both of my Sea-Gulls, that they use smooth balance wheels, but they are carved in one spot (looks like a deep scratch on a balance wheel), apparently they removed a bit of material from the balance wheel to improve it's motion and stability.

 

Yes, this is done using a laser to poise the balance.

 

Good thread.

 

Silicon springs don't do particularly well when you go 3-D. I have seen a number of designs and the best are rather clumsy.

The problem seems to be that to "step-up" with a silicon spring requires a riser block which changes the poise of the hairspring. Since, the silicon spring can be made flat AND relatively isochronic,there really isn't a need to make them overcoils.

An isochronic ideal spring will maintain a constant frequency no matter the input energy (amplitude), ie the mainspring weakening due to unwinding will not affect the rate.

 

I don't really know of anything that I would call a disadvantage with the freesprung system. Rolex uses it to good effect. One of - if not the most successful of all time - competition wristwatches was a freesprung 2824. The Tissot bested everything in the Chromometrie Trials except a couple of tourbillons(it beat the rest of them). Even if you reqire an annular balance - this doen't bar you from having a freesprung watch.

 

This thread started out for the masses but now my head hurts.

 

Let's keep it civil.

Whatever the mathematics of the thing, actual watch manufacturers are attempting to modify the shape of silicon hairsprings to achieve better positional accuracy:

"What makes Syloxi unique is its shape and geometry. Though it is flat, the hairspring breathes constantly in all positions, compensating for gravitational errors, due to the varying pitch and thickness along the length of the hairspring.

Additionally, the ends of the hairspring are pinned in a patented manner, which further improves its concentricity. The inner coil of the hairspring is flexible, allowing it to be fixed without glue. The other end of the hairspring is rigid and crescent shaped, allowing it to be fixed to the balance bridge at two points on either side of the balance staff. This patented hairspring geometry ensures it remains perpendicular to the balance staff while being self-centring."

Watches by SJX: Baselworld 2014: Explaining the Syloxi Hairspring Inside the Datejust Pearlmaster 34, the First Rolex Equipped with a Silicon Balance Spring

It doesn't say anything about isochronism.

I still believe the greatest threat to silicon hairsprings is probably temperature, judging from the documents we have left over from like 2001 lab testing at Ulysse Nardin, where their prototype was noticeably more affected than the metallic hairspring by temperature. Silicon hairsprings are coated for thermal stability, but I still want to see the numbers--like an anachron versus si-14 hairspring in an 8500 at different temperatures.

 

Let's keep it civil. . . . .

Let's keep it accurate. : )

 

I will continue to work on solving world hunger, global warming and providing universal justice but in the interim I'm going to mount a piezo crystal behind the roller jewel, swap the brass collet for my new polymer unit with integral microprocessor and accelerometer, replace the blue hairspring with my new special sauce variable rate composite elastomer and use a sapphire lensed light emitting diode for a stud.

Can anyone else say it? E X H I B I T I O N - B A C K

 

V1 of my new special sauce variable rate composite elastomer was generally configured as a series of planar concentric coils connecting the traditional stud holder position on the balance cock to the staff vis a vie the new polymer collet located just above the balance arms.

The outer layer of the special sauce is a super flexible resin coating applied to a thin layer of conventional antimagnetic hairspring material.

The outer layer serves as a raceway for special sauce's conductors and protects its strain gauges and reversible polarity electromagnets which are mounted in a series of nodes outside the antimagnetic hairspring material layer.

A series of polymer shackles are bonded to the outer layer of super flexible resin.

These shackles straddle the antimagnetic hairspring material layer and loosely bind and align a thinner steel hairspring material layer to the middle antimagnetic layer.

This inner and third layer of steel hairspring material is slightly shorter than the outside and middle layers and is unrestrained, without terminal attachment, allowing it to float within the shackles in the absence of activation of one or more of the electromagnets.

 

Sorry to revive an old thread, but I have a couple of questions:

1) Would a free-sprung escapement set-up without an overcoil have improved positional variance over a regulated balance? I'm just thinking (based on the basics I've learnt so far), that one end of the hairspring is attached to the balance wheel, while the other is either attached to a regulator index or, after an overcoil, to somewhere along the balance cock. But if an overcoil is absent, wouldn't the other end of the hairspring just be attached to a fixed point, which would function as if it was attached to a regulator index?

2) For free-sprung balances, how does the swan neck regulator adjust the rate? To what does it connect to? The only answer I've found thus far on the internet is "Lange has retained the swan's-neck regulator on its free-sprung watches not a a rate adjuster, but rather to adjust the beat (impulse from the pallet fork to the balance should be symmetric in both directions of the fork swing) by centering the fork at rest between two guide posts", but I don't really understand how that works.

Thanks for taking the time to respond to these queries!

 

. . . . 1) Would a free-sprung escapement set-up without an overcoil have improved positional variance over a regulated balance? . . . . .

By "regulated balance" are you referring to a balance regulated by curb pins straddling the hairspring? [Note: If a balance isn't regulated the rate of the movement is ungoverned.]

. . . . I'm just thinking (based on the basics I've learnt so far), that one end of the hairspring is attached to the balance wheel, while the other is either attached to a regulator index . . . .

This phase reads fairly nonsensical from a watchmaking perspective.

. . . . or, after an overcoil, to somewhere along the balance cock. But if an overcoil is absent, wouldn't the other end of the hairspring just be attached to a fixed point, which would function as if it was attached to a regulator index? . . . .

Curb pins - when properly adjusted - don't always contact the hairspring as the balance oscillates. [Don't confuse beat adjustment with rate.]

. . . . 2) For free-sprung balances, how does the swan neck regulator adjust the rate? To what does it connect to? The only answer I've found thus far on the internet is "Lange has retained the swan's-neck regulator on its free-sprung watches not a a rate adjuster, but rather to adjust the beat (impulse from the pallet fork to the balance should be symmetric in both directions of the fork swing) by centering the fork at rest between two guide posts", but I don't really understand how that works. Thanks for taking the time to respond to these queries!

SEE ABOVE. The stud pin holder isn't fixed in this case and it's fine adjustment is controllable via this micro postionable device.

--------------------

You may be on the verge of a meaningful line of inquiry here but you'll have to state it for yourself and it appears that in order for your to do that you will:

1. Need to get the lexicon down

2. Review some technical explosions

3. Study the basic interactions between these conponents

 

Thank you both for your replies. I do understand a bit better now and will read up more!