All things being equal, how can a unidirectional winding be more efficient than bidirectional?

[Chascomm]

Also, this thread is interesting in that respect ...
IWC Schaffhausen | Fine Timepieces From Switzerland | Forum | Bidirectional vs. Unidirectional winding

Some useful points raised in that article. For example AS's methodology for determining which was more efficient. If you've ever examined one of AS's bi-direction systems, the amount of dead-angle is plainly evident compared to an ETA bi-directional. By eliminating one direction, and the switching gear, they got a more efficient winder. I would feel more confident of this experiment had it been carried out by ETA using reverser wheels as the behaviour of the various components would be more consistent between uni- and bi-directional versions.


Something which was not mentioned in this discussion is the way some winding systems (Pellaton, Magic Lever and everything along that developmental continuum) have variable winding efficiency depending on a combination of direction and position of the rotor. This can be 'tuned' (simply by the adjusting the position of the rotor at the time of assembly) to deliver maximum efficiency in a particular arc e.g. when the crown is pointing generally downwards (i.e. the wearer is walking). Another instance of the difference between theoretical efficiency and 'real world' efficiency.

[bigvatch]

I would say the same since bi-directional winding is common among affordable Seikos.

Maybe mentioned already, but the most common Miyota auto , the 8215 is a uni-directional winding auto, and the cost is a bit cheaper thant he cheapest seiko auto.

[austinnh]

I suspect that it mostly has to do with the dead-angle. When the wearer is walking, for example, the rotor swings back and forth. With a reverser (typical bi-directional winding system), each time it reverses direction, some energy is lost as the rotor swings through it's "dead-angle."

A uni-directional winding system, on the other hand, has negligible dead angle (I expect it's not zero, but very tiny). But what about all the energy that is lost while the rotor swings in the non-winding direction? It's not lost. As the rotor swings freely in the non-winding direction, it keeps that energy as kinetic and gravitational potential energy (there should be very little friction in this direction), so once it reaches the end of it's swing, it will simply swing back in the other (winding) direction and all that energy will then be transferred to the mainspring.

The Seiko magic lever system has essentially no dead angle around some parts of the circle, but effectively huge dead angle around other parts. Seems to me this would be better in a situation where the angle of the watch relative to gravity is controlled (for example walking, always crown down), but worse where when the angle is more random and the high dead-angle parts of the circle will come into play (just about anything else). In any case I believe Seiko went to a traditional reverser mechanism in some of its GS automatics, suggesting to me what I had already suspected: the magic lever is simpler, but less efficient.

Comparing any two complete watches is not really useful because in any well-designed watch, other things can be compromised to bring the winding efficiency up, for example, using less torque in the gear train and/or a smaller diameter balance and/or a lower beat rate. This also means that even though most modern automatics have little trouble staying wound, winding efficiency is still important because more of it means higher torque/diameter/beat rate. Of course depending on who you ask, torque, diameter, and beat rate are not important anyway...

This summarizes what I THINK to be a correct-ish explanation.

[CitizenM]

It'd be interesting to see if they'd still want the reverser on the current 9S calibres if they lowered the BPH to achieve their power reserve, like most manufacturers do.

I tend to think people don't give enough credit to the accuracy implications of efficient winding mechanisms. The conventional sentiment is that as long as the watch is fully wound when you take it off at the end of the day, it didn't matter what automatic winding mechanism was used. But due to fact that accuracy is affected by the tightness of the mainspring, the mechanism that winds the watch from 50% to 95% in just a few hours will tend to be more accurate, at least on days it is worn, than another mechanism that winds the watch the same amount in 6 hours.

[aladin_sane]

I read a couple of post trying to explain that the energy of the freewheeling rotor spinning in the non winding direction would somehow be captured when the direction of the rotor reversed. This is patently false.

[austinnh]

I read a couple of post trying to explain that the energy of the freewheeling rotor spinning in the non winding direction would somehow be captured when the direction of the rotor reversed. This is patently false.

Maybe you should enlighten us. Seems to me that when the rotor is in a non-horizontal plane (most of the time), the kinetic energy of the rotor is turned into gravitational potential energy as the rotor swings up. Once all it's kinetic energy has been turned into gravitational energy, it stops and begins to swing in the other direction, like a pendulum. That gravitational potential energy will be turned back into kinetic energy. Oh but wait, it won't because it will all be captured by the winding system, now that the rotor is traveling in the winding direction.

Yes, it's true that if the rotor has enough energy to make full circles, or if the watch is in a perfectly horizontal plane so gravity is not affecting it, energy will be lost due to friction until the rotor slows down enough to reverse direction due to gravity rather than making more full circles, or until the watch is rotated away from the horizontal plane. This, however is not typical, and if you're doing an activity that causes the rotor to frequently spin in full circles, you're probably having no trouble keeping your watch wound.

You're probably a pretty smart guy, but you're not the only one here, so when you're going to call someone's carefully justified understanding "patently false," some justification of your own is called for. If I haven't explained this clearly, I'm happy to explain it in more detail. If there's something wrong with my reasoning, then please be more specific.

[CitizenM]

Maybe you should enlighten us. Seems to me that when the rotor is in a non-horizontal plane (most of the time), the kinetic energy of the rotor is turned into gravitational potential energy as the rotor swings up. Once all it's kinetic energy has been turned into gravitational energy, it stops and begins to swing in the other direction, like a pendulum. That gravitational potential energy will be turned back into kinetic energy. Oh but wait, it won't because it will all be captured by the winding system, now that the rotor is traveling in the winding direction.

Yes, it's true that if the rotor has enough energy to make full circles, or if the watch is in a perfectly horizontal plane so gravity is not affecting it, energy will be lost due to friction until the rotor slows down enough to reverse direction due to gravity rather than making more full circles, or until the watch is rotated away from the horizontal plane. This, however is not typical, and if you're doing an activity that causes the rotor to frequently spin in full circles, you're probably having no trouble keeping your watch wound.

You're probably a pretty smart guy, but you're not the only one here, so when you're going to call someone's carefully justified understanding "patently false," some justification of your own is called for. If I haven't explained this clearly, I'm happy to explain it in more detail. If there's something wrong with my reasoning, then please be more specific.

I take it for granted that at least once in the ownership of a unidirectional winding watch the rotor will swing up in a non-winding direction, but due to gravity, come down in the winding direction, returning some part of the energy that it cost to move it up there.

[drunken monkey]

I imagine that no matter how much the rotor freewheels in the non-winding direction, it will tend to always fall back in the winding direction (otherwise it'd just tip over and continue).
The question is then where in the swing does gravity overcome the intertia of the rotor spinning and how much time does it spend freewheeling compared to winding.

[CitizenM]

I think the question can't be that simple, given that it's almost certainly true that all watches worn will at least once in their lifetime go in the other direction, energy that'd have been captured (overcoming the dead angle, presumably) in a bi-directional winder. So that efficiency is almost a total loss, excepting those times the rotor swings up in the dead direction and comes back down in the live one. But, if we overcome that loss with superior efficiency in the one direction (as opposed to inferior efficiency in both), then we might be able to get a net positive return on that modification. That seems to be the conclusion.

[drunken monkey]

Well, you can't really compare the non-winding direction of a uni-directional winder to the same relative movement in a bi-directional winder as there are different resistance forces at work
i.e none in the uni-direction, winding resistance in the bi-direction.

However, in the case of the bi-directional winder, you do also get an extra little bit of wind when the rotor comes back down.

On the other hand, as I tried to illustrate in my previous point, the free-wheeling rotor movement will tend to free-wheel until gravity is greater than the inertia. I imagine that often, this leaves the rotor positionally higher up in its swing so the resultant winding rotation is greater. I can also then see this resulting in some non-winding rotation as the rotor completes its winding movement and begins to return in the other direction.
i.e it rocks.

incidentally, I just had a look at my various watches and observations show that

ETA 2824 - bi-drectional winding - rotor has quite a high amplitude with a gentle shake
TAG Heuer1887/SII T78 - bi-directional magic lever - rotor tends to move at a lower amplitude with same sort of shake
Omega 1012 and 1020 - uni-direction - rotor behaves similarly to the 2824 in it's amplitude and ease of movement.

What I find interesting is the comparison of the 1887 to the 2824.
As I said before, I recall something about the magic lever being a more efficient winding system than the usual reversing wheel. Perhaps this is what is being translated into the lower amplitude
i.e resistance to rotor movement=winding resistance.

Either that or it needs a service already...

Of course, we have no idea what they mean when they say something like "more efficient in the real world" as they could be looking a very specific set of figures gained from very specific data entries. I mean, they could be comparing like for number of rotor rotations for each movement/system and equating it to power reserve or they could be looking at how much input it takes to get comparative power reserves into both systems or they could be looking at how little inertia there is in the winding system and taking that to equating to "ease of winding". All could be taken to be to reinforce the "more efficient" argument.

Alas, without being a engineer with a phd in tribology, I'm not sure how the layman could know for sure what is what.

Sidenote - aren't Rolex reversing wheels teflon coated to reduce friction in the winding system to increase winding efficiency?
Seems to support my initia thoughts on it to do with resistance forces added by having a reversing wheel.