Patek Philippe

MGM
30

• Show profile
• Show other posts

Patek single-impulse escapement (animation)

This is a single direct impulse, variable torque escapement with additional locking/unlocking for the part of oscillation when the escapement doesn't deliver the impulse (4 stable positions per oscillation).

Patent publication number: EP1522001

(3 MB file)

Different layers are in different colors to make it easier to see which parts interact and how.

dark blue - impulse wheel, impulse "jewel", locking part

blue - 2 locking/unlocking parts, unlocking "jewel"

grey - escapement gear train

black/yellow - rotor/stator

The blue part of the torque indicator presents the range of torque when escapement is unlocked.
The gray part is minimum torque to zero. The red part is negative torque during the unlocking.

Variable torque

The torque consists of two components:

1) "Constant" torque from the mainspring barrel

2) Sinusoidally varying torque from the rotor

You can see in the animation how the torque changes as the rotor turns.

Rotor - Stator

Rotor - Permanent magnet diametrically magnetized (indicated by the arrow)

Stator - Ferromagnetic ring with two recesses. When rotor is set in rotation, the presence of these two recesses creates magnetic couple/torque having an effect on the rotor.

When the rotor's axis of magnetization is on the same axis as the recesses the rotor is in position of unstable balance. The position of stable balance is perpendicular to it. When the rotor is in position of stable balance slight angular shift will tend to bring the rotor back to its position, but when it's in position of unstable balance slight angular shift will tend to move it further from that position (to the position of stable balance).

The last picture shows what a pure mechanical analog would look like. Of course there are many ways of making a similar solution. The rotor and stator basically act like a spring - but with the simplest possible construction.

The impulse

Single impulse - there is only one impulse per oscillation (instead of one per each semi-oscillation as in lever escapement). This means less disturbance for the oscillator - the balance wheel.

Direct tangential impulse - the impulse is delivered to the balance directly by the escape wheel (and not through a lever) - there is less sliding friction (than the radial impulse of the lever escapement). This means more efficient energy transfer and no need for lubrication.

In this aspect it is similar to detent chronometer escapement or even more similar to Robin with its additional locking.

The angle during which the impulse wheel and the balance are engaged seems to be around 43 deg but that's not the whole story. When the escapement unlocks, the torque increases and when the impulse wheel and the balance come in contact the torque is still not at the maximum (<90%). Shortly after that, it reaches its maximum (the tangential component of the force at the max also) and then drops to around 40 % of the max at the point where they disengage. This means that the maximum power is transmitted over a limited angle of the oscillation. Maximum torque is achieved when the balance is at its maximum velocity (kinetic energy) - that's the dead point of the balance wheel (neutral position of the hairspring). This way the isochronism of the oscillator is preserved to the maximum.

Maximum torque from the escapement is (at least) around 65% greater than torque from the mainspring (fully wound).

Torque from the escapement during the impulse:

The unlocking

You can see from the animation that all the parts are very small in diameter and that center of mass=center of rotation.

The lever escapement has a safety feature that prevents unlocking at the wrong time (the draw). It holds the lever against the banking pin (or the locking notch in the new Pulsomax escapement). When the balance unlocks the escapement it also moves the escape wheel slightly backwards and that's the additional energy loss.

With this design this was not necessary. There is no recoil. Energy loss minimized.
You can see in the animation the small amount of negative torque during the unlocking.

During the unlocking the locking/unlocking part slides over the escape wheel tooth (or another locking part) while the escape wheel doesn't move. During that motion the gears can not be engaged and that explains the missing teeth from the pinions and the big wheel.

Short version:

efficient energy transfer, minimized disturbances of the natural oscillation of the balance

no lubrication of critical surfaces

compact size, low inertia

shock-proof

relatively simple construction - only uni-directionally rotating parts

easier lubrication of pivots (non-critical)

Very clever and outside the box solutions - but it's still just a patent and obviously very high tech design so I guess we shouldn't expect to see it in action any time soon.

Mishko

Site Info.