Alternative to a ‘fly’ for braking a clock striking train

[lfoggyParticipant @lfoggy]

A few years ago I built an 18th century style English bracket clock based on plans published by John Tyler. The clock strikes the hour through conventional rack striking. Although the clock has been running well, I was always aware that the striking train was quite noisy when operating. This is a feature of all clocks of this design as well as long case clocks. The whirring noise comes from the fly which spins rapidly and serves to brake the striking train when it is unlocked. Most of the noise seems to come from the engagement of the warning wheel with the pinion on the fly arbour. How could this be reduced? First thing I tried was making the warning wheel out of nylon. This reduced the noise considerably, but it became evident that this would not be a durable solution. I thought of dispensing with the gear/pinion arrangement completely and driving the fly arbour with a belt, maybe using an O ring. This would probably have also been quieter but again vulnerable to failure. I do want my clock to have the same durability as an original 18th century example. At around this time, I made a desktop gadget consisting of a thick copper tube and magnet. The magnet dropped down the tube is dramatically retarded by the Lenz effect. This led me to wonder if eddy current braking could be used to control the striking train. A quick test on the clock depthing tool with an aluminium disc mounted on an arbour and some small neodymium magnets was encouraging.

With no magnets the disc spins freely.

With the magnets close to, but not touching, the disc, there is obvious retardation.

I therefore removed the fly from the clock completely and replaced the warning wheel with a 1mm thick 45mm o.d. aluminium disc. Four small neodymium magnets were mounted in a pure iron yoke and arranged with a 0.25mm air gap between the magnets and disc. This braked the striking train very effectively. Too effectively in fact, as the striking frequency was reduced to a ponderous one strike every three seconds. To reduce the braking effect, I reduced the thickness of the disc to 0.5mm and added some slots to interrupt eddy current formation.

I now have a clock that strikes at one second intervals with no audible additional sounds at all – success. Added benefit is that the system has much less inertia than the fly arrangement meaning reduced strain on the parts when the striking train is suddenly arrested by the gathering pallet hitting the pin on the rack. Durability is ensured as this arrangement has fewer moving parts than the traditional design.

The magnets will eventually lose their magnetism, but this happens over centuries -easy to replace as well. The only other problem I can think of is iron dust sticking to the magnets, maybe from the pinions wearing over time. This design may even be more reliable than the traditional fly. Only time will tell.

Few things in horology or model engineering are new, and I can’t imagine that this is a new idea, but I’ve not seen reference to it anywhere hence this post which I hope is of interest to other clock constructors. I will be incorporating this into future clock projects.

[Author]

Posts