Automatic versus manual winding

Both kinds of watch store energy in the same place: a coiled ribbon of spring steel inside a barrel, described in full in our guide to how a mechanical watch works. The only question here is who does the coiling. In a manual watch it is you, through the crown. In an automatic it is a weight that swings as your arm moves, plus a small assembly that turns swinging into winding.

The practical consequences are smaller than the marketing suggests. An automatic is not more accurate, not lower maintenance, and not more advanced in any interesting sense. It keeps the mainspring in a healthier part of its torque range without you thinking about it, at the cost of thickness, parts count and a partly obscured movement.

The rotor, and how a swing becomes a wind

The rotor is a semicircular weight on a central bearing, free to spin through a full circle. It is heavy at the rim and unbalanced by design, so gravity holds it while your arm moves the watch around it. Relative to the movement, it turns.

Two details decide how well it works. The first is the rotor's turning moment: how much mass it carries and, more importantly, how far that mass sits from the pivot, since its moment of inertia rises with the square of that distance. This is why rotors carry a heavy-metal rim: tungsten alloy in most modern movements, gold or platinum in some expensive ones, not because the metal is precious but because it is dense. The second is bearing friction. A rotor on a jeweled bushing costs less; a rotor on a ball bearing costs more and starts turning with less provocation. A worn rotor bearing announces itself as a faint gritty rustle when you rock the watch by your ear.

The rotor's output is useless as it stands, because it swings one way and then the other. Something has to rectify that motion, in the electrical sense: convert alternating input into one-way output.

Unidirectional and bidirectional winding

The first mass-produced answer, and the one Rolex used in its 1931 Perpetual, was to ignore half the motion. A pawl or click lets the rotor wind one way and freewheel the other. This is unidirectional winding: simple, reliable, and it discards roughly half the energy available.

Bidirectional winding captures both directions, and nearly every modern automatic does it. The three common methods are worth telling apart, because they behave differently in the hand.

None is a quality signal on its own.

The slipping bridle: why you cannot overwind an automatic

An automatic mainspring is not attached to the barrel wall. Its outer end carries a curved strip of spring steel, the slipping bridle, pressed against the barrel's inner wall by its own springiness. Friction holds it while the spring winds. At full tension the pull exceeds that friction, the bridle slides round the wall, and a little tension is released. A specialist braking grease makes the behavior consistent.

So: you cannot overwind an automatic, by wearing it, hand-winding it or leaving it on a winder. A hand-wound watch has no bridle. Its spring is hooked solidly to the barrel wall and stops dead at full wind, and forcing past that point is how people break clicks, ratchet wheel teeth and, occasionally, springs.

Winding efficiency and the desk problem

Manufacturers quantify this as a winding efficiency figure: how much rotor rotation stores an hour of running. The numbers are rarely published, and where they are they come from a machine swinging the watch through a defined arc, which is not your day.

The consequence is what matters. A sedentary office worker, typing with forearms resting on a desk, may produce far fewer rotor revolutions than the test assumes. The result is rarely a watch that stops. It is a watch hovering in the middle of its power reserve, never reaching full wind, running permanently in the weaker part of the torque curve where amplitude is down and the rate is least stable. Owners in this position often conclude their watch is inaccurate when it is simply hungry.

The fix is trivial: if the movement can be hand-wound, give it 15 to 20 crown turns each morning for a week, then judge. Some automatics cannot be hand-wound at all, the long-running Seiko 7S26 being the best-known example, because the keyless works has no winding path from the crown. There, the only remedies are more active wear or a burst of wrist motion.

Winding a manual watch properly

From fully stopped, most hand-wound wristwatches take somewhere in the region of 20 to 40 crown turns to reach full wind. The exact number depends on the barrel and the winding gear ratio, so treat any figure you read as a guide rather than a rule.

You do not count. You feel. Wind at a steady pace, in whichever direction engages (many crowns wind one way only and slip harmlessly in the other), and stop the moment the resistance changes from smooth and even to firm and unmistakably solid. That firmness is the spring reaching the end of its travel. There is no benefit whatsoever in one more turn.

Hold the watch off the wrist while you do it. Winding on the wrist levers the stem sideways against the case tube, a real cause of stem and tube wear over the years. The same applies to setting and adjusting your watch.

Micro-rotors, peripheral rotors, and what they cost

A conventional rotor sits on top of the movement, adds height, and hides half of it from anyone looking through the caseback. Two layouts avoid that, and both charge for it.

A micro-rotor is a small weight recessed into a cut-out, sitting level with the bridges rather than above them. Universal Genève and Buren both brought micro-rotor movements to market in the second half of the 1950s and both claimed priority, so the question of who was first is not settled. The gain is thinness and a clear view. The cost is physics: a smaller, lighter weight closer to its pivot has much less inertia, so winding efficiency falls. Designers compensate with dense metals and gearing, but a micro-rotor is generally a poorer winder than a full rotor of the same era.

A peripheral rotor is a ring running around the outer edge of the movement on rollers or a bearing race, leaving the whole movement visible. Patek Philippe built one in the 1960s and abandoned it; the idea was revived commercially in the late 2000s by Carl F. Bucherer and has since appeared from Vacheron Constantin, Breguet, Piaget and Bulgari. It is thin and it shows everything. It is also expensive, more sensitive to shock and bearing wear, and historically weak at the winding itself, though modern versions are much improved.

Power reserve and service implications

An automatic module changes the mainspring maths. Because the bridle must be able to slip, an automatic barrel cannot use a stop-work to keep the spring away from its weakest coils, and full wind is a soft ceiling rather than a hard one. Barrel space is also constrained by the winding module above it. Hand-wound versions of a given caliber therefore often achieve a slightly longer or more consistent reserve than their automatic siblings.

At service the module is extra work: reversers or pawls to clean and lubricate with specific oils, a rotor bearing to inspect, braking grease to renew. That last point matters. Aged, sticky braking grease stops the bridle slipping cleanly, which shows up as a stiff, low-amplitude watch or a mysteriously short reserve. Both are covered in our guide to servicing and care.

A routine if you own both

1. Wear the automatic on active days, and give it a short hand-wind on the morning of a desk day if it accepts one.
2. Wind the manual at the same time daily, off the wrist, to firm resistance.
3. Let anything you are not wearing simply stop. There is no harm in a stopped watch.
4. Use a winder only for a calendar watch you dislike resetting, at the minimum turns per day the movement needs. Our guide to winders and storage sets out the numbers.
5. A watch that runs down overnight after a full day of wear has a fault, not a lifestyle problem.

When a manual wind is genuinely the better choice

• Thinness. Remove the rotor and its bridge and you remove real height. Most of the thinnest watches ever made are hand-wound.
• A clear view. If the movement is the point, a rotor hides half of it.
• Occasional wear. A watch worn once a fortnight is never properly wound by wearing, so a rotor buys you nothing.
• Simplicity at service. Fewer parts, fewer specialist lubricants, fewer things to wear out.
• The ritual. Not an engineering argument, and it does not need to be.

An automatic remains the better tool for daily wear, for calendar watches you would rather not keep resetting, and for anyone who would simply forget.