IN THE FIRST example, the SeeSaw Gravity Escpement provides impulse at every swing of the pendulum, that is, first in one direction and then the other (first impulse mode). This is by far the most common mode of impulsing used by escapements.
Alternative escapement modes
A second mode of impulsing involves impulsing the oscillator (pendulum or spring balance), in one direction only. The most well know example of this type is the marine chronometer escapement and the lesser-known pendulum version. A third mode of impulsing involves impulsing the pendulum after it has made a predetermined number of oscillations. The most well known example is the escapement used in the Synchronome clock. A fourth mode of impulsing involves impulsing on demand so that, as the pendulum's amplitude drops below a certain value, it is automatically impulsed. The most well know example involves the ingenuous mechanism know as Hipp's toggle. It is beyond the remit of this article to discuss the merits and uses of these various forms of impulsing. They are mentioned simply as precedence for adapting the SGE to work in additional impulse modes.
The SGE may be adapted to work in the second and third of these impulse modes and these will be described next. Adapting the SGE to work in the fourth mode: 'impulse on demand', is a challenge for the future!
Second impulse mode
Figure 16 is a front view of a SGE in which the pendulum is impulsed in one direction only. The construction is virtually identical to that shown is Figures 3a – 8c. However, the LH flat cam is replaced by a concave cam, and the LH weight support stop is no longer needed. The radius of cam is struck from the pivotal centre of the LH prop assembly. Hence, as the prop pivots, it supports the weight support, but the weight support does not move up or down. Furthermore, the length of the cam, and the prop assembly stops, constrain the roller to move within the radiused surface of the cam.
Oscillation is begun by manually displacing and releasing the pendulum, in the normal way. The LH Prop and its coacting components continue to lock and unlock the escape wheel as described previously. However, because of the radiused cam, the LH weight support does not move up and down, and the LH weight does not impulse the pendulum. The RH prop, and its coacting components, lock and unlock the escape wheel, and impulse the pendulum, as described previously. Hence, the pendulum is impulsed in one direction only.
Third impulse mode
Figure 17a is a front view of a SGE, where the impulse occurs once for every complete cycle of the escape wheel. The construction is similar to Figures 3a – 8c and Figure 16, but with the following differences:
The RH flat cam is replaced by concave cam, and the RH prop stop is no longer needed. The radius of the cam is struck from the pivotal centre of the RH prop assembly. Hence, when roller acts within the radius of the cam, it supports the RH weight support but does not cause it to move up or down. However, the length of the RH cam is shorter than the LH cam and, as the RH prop pivots fully clockwise, its roller exits the edge of the cam causing the RH weight support to lower, and be constrained by its stop. The RH reset lever is replaced by bent reset lever. The bend gives clearance to the resetting rollers of the escape wheel assembly, prior to their resetting action.
The pivotal shaft of the RH prop assembly extends though the front plate and on it is fixed an impulse control lever (Figure 17b). The distal end of this lever terminates in a ball bearing roller. The roller acts on an impulse control cam, which is mounted on the shaft of the escape wheel assembly. The control cam has major and minor peripheral regions. In this example, which provides a pendular impulse once per revolution of the escape wheel, the minor diameter encompasses teeth 1 and 6, relative to the escape wheels. For two impulses per revolution of the escape wheel, the minor diameter would be extended to encompass teeth 3 to 6. For less frequent impulsing than once every revolution of the escape wheel, the control cam, with an appropriate form, would be driven via the escape wheel shaft, but at a slower speed.
When the RH prop pivots clockwise, as the RH weight is lifted clear of weight support, and roller of the impulse control lever coacts with the major diameter of the control cam, the prop roller is constrained to remain within the confines of the radius region of the RH weight support cam. Hence, the weight remains at the same height, and no impulse is imparted to the pendulum by the RH weight. However, when the roller of the impulse control lever coacts with minor diameter of the control cam, the roller of the RH prop exits the edge of the cam, causing the RH weight support to lower. Hence, impulse is imparted to the pendulum by the RH weight (as described in modes 1 and 2). As the escape wheel assembly continues to index, the control cam controls when impulse occurs. In this example, it is once every complete turn of the escape wheel, or every sixth swing of the pendulum.
Discussion
The concept model of the SGE, in which the pendulum is impulsed in both directions, has been tested. Frankly, because of its departure from convention, I was surprised it worked at all! But, work it does. Furthermore, it is robust, and can handle a wide range of periodic times and pendulum amplitudes, ranging from a fraction of a degree to several degrees.
What of its limitations: Like its nearest relative the DTLGE, and particularly if weight driven, the SGE needs a higher gear ratio between the driving weight/s and the escapement, than conventional escapements. However, such clocks were made made successfully in the mid 19th century ('Big Ben' being the prime example). These days, high precision gears, and the use of ball bearings, make light work of high gear ratios.
Like the DTLGE, the tick of the SGE is louder than conventional escapements. It is thought that the strategic use of resilient materials, and/or damping elements, will reduce the noise of the tick to an acceptable level, which is particularly important for domestic or 'boardroom' clocks.
As I mentioned earlier, no attempt was made to optimize the concept model or seek its longevity at this stage. I have already alluded to some possible improvements, particularly regarding the materials used. These relate mainly to the thermal stability and wear of the SGE mechanism itself. In this respect, I do not know if the SGE is potentially better or worse than other escapements. At least in the SGE it is easy to see how components affect the stability of impulse, and to design and choose materials accordingly.
The question is: will the SGE be a good timekeeper, and will it be reliable? In theory, it should be – but that's hardly an answer! The only way to test the escapement is to build it into a precision clock. The most important component of such a clock is the pendulum. I see no reason to break with tradition and not use a one second pendulum - but what sort of pendulum? Should it be a simple, compound, or Schuler-type; and what of temperature and barometric compensation? Then there is the question of which mode of impulse is best?
But, traditionally, clock makers have never confined themselves to purely functional aspects. Reifler precision clocks of the early 20th century were beautifully elegant with their etched glass cases, nickel plated columns, and maple back boards. These days', mechanical clocks are not built primarily for timekeeping accuracy; quartz clocks and atomic clocks are far superior in that respect. My own interest in designing and making contemporary mechanical clocks lies in the way the technology and aesthetics may be combined to enhance visual interest. So, if a clock has a particularly interesting mechanism or pendulum, then these become important aesthetic features. But that is not as easy as it sounds because a precision clock needs to indicate the time clearly, and making a feature of the mechanism must not distract from that.
These are challenges for the future - which will not be solved 'overnight'.
Conclusion
The feasibility of the SGE has been demonstrated using a concept model. In addition, alternative modes of impulse have been considered and appear possible. However, further work is needed to examine the performance of the SGE in a complete clock. Only then will the utility, or otherwise, of the SGE be established.
The SeeSaw Gravity Escapement is the
subject of a recent patent application.
