A bimetallic balance wheel

[Sam StonesParticipant @samstones42903]

One evening in October, when I was one third sober – or so the story goes, I decided many years ago to build author John Stevens’ “A Skeleton Clock with Lever Escapement” (see five editions of ME commencing Feb ’72). She, who must be obeyed, was unprepared to have a model beam engine on the mantelpiece.

With one eye on “Watch and Clock Making and Repairing” by W J Gazeley, my attempts to make a bi-metallic balance wheel for this clock may perhaps be of interest and/or amusement to other model engineers. Or perhaps get a smile or two from any horologists who might be `listening’.

Making this wheel, I have to say at the outset, took five attempts before I achieved an acceptable result. For my personal satisfaction, every part of this clock had to pass close examination through a 4″ binocular magnifier.

The method advocated for attaching the brass to the steel (see page 243 of Mr Gazeley’s book) was to turn a `sink’ into a rod of `good-quality’ steel. A sink in this context is turned into the end of the rod, thus forming a circular groove with a rectangular cross-section, and whose inside diameter exactly matches the diametral interface of the brass with the steel. A ring of brass is machined whose volume will more than adequately fill the groove (sink) once the brass starts to melt. During my last attempt, and allowing for the borax paste, the brass ring was a close (tight) fit onto the inner side of the groove. Presumably the brass ring would loosen up as it got hot. This inner face was to become the steel/brass interface.

By the way, and this may be obvious, the brass must be on the outside of the finished wheel for the principle to work. The normal ratio for brass to steel (as stated) should be 3:2, although mine ended up at about 2:1. If it doesn’t keep good time in Melbourne’s wide temperature fluctuations, at least it should look good.

Even though I did all the right things (or thought so), each earlier attempt revealed either poor fusion or blow holes, once it was back in the lathe. The essential issues related to perfect cleanliness of all the parts including the borax and water needed as a flux. I didn’t have a hot source of heat at home, so I went into the maintenance department of the office where I worked, and they sprayed a nice hot oxy flame onto the assembly. An acetylene flame burns very hot, so maybe the earlier attempts got the brass too hot, I’m not sure.

Turned to size, the next step was to drill and tap ten holes (10BA) through the rim of the wheel (brass and steel), and to make ten screws to fit, each with a large head for weight. These holes are arranged in a pattern so that, by winding the corresponding screw in or out, temperature compensation can be achieved. That’s the theory. I understand that to adjust their weight for the best result, the screw heads are hollowed out from the threaded end, and so that the screw heads all look the same size and fit snugly with the wheel rim.

Phew!

Mk V balance wheel was eventually ready for the bench, so the next step required piercing and filing the bottom face to form two crossings. For some unknown reason, in clocks, they are called crossings not spokes. A couple of holes are drilled in the spokes (sorry crossings) near the centre so that the wheel can be attached to the balance-wheel collet with two tiny screws. Finally, the rim is cut through close to where the crossings meet the rim so that two semicircular `bands’ are created. These form the bimetallic `arms’ which is what all the fuss is about. That’s where it stands at the moment.

Regards to you all,

Sam

 

Edited By Sam Stones on 29/06/2010 04:03:02

[Sam StonesParticipant @samstones42903]

A couple of pit-falls for beginners like me

[Sub MandrelParticipant @submandrel]

I must admit I struggle to follow that without drawings or pictures… but it’s puicqued my curiosity why not write it up fully with some illustrations and send it to Dave Clarke?

 

Neil

[Sam StonesParticipant @samstones42903]

Thanks Neil,

 

That’s certainly `on-the-cards’. All the parts are in 3D (KeyCreator), C/U photographs can be taken, and more text can be written.

 

I have however, promised my wife that I’ll finish making the clock which has stood for more than 30 years waiting for the day it will tick. There are still 23 parts to make and, having borrowed the HobbyMat lathe and mill, I’m getting close to cutting metal again.

 

If I do go ahead, can you please advise the file format for drawings (CKD, DRG, or ???)and photographs (starting-point Canon RAW)?

 

Regards,

 

Sam

[Sub MandrelParticipant @submandrel]

Thanks Sam

 

How about 1 or 2 pictures for the gallery to keep us interested?

 

Check with Dave Clarke for the preferred formats for drawings,I think they use Autocad so DWG is probably the preferred format, but I think ‘scribble on the back of an envelope’ is acceptable. In the past I’ve sent two files for drawings, a pdf and a hi-res gif – it hasn’t been possible to judge which of the two they used.

 

My experience with photos for publication (ME and elsewhere) is to do my own tweaking then save as a jpeg with a relatively modest amount of compression. Printers aren’t fussy about having everything as lossless TIFs anymore, and jpegs don’t seem to stumble if the designer uses a Mac.

 

Neil

[Author]

Posts