Back in 2003/04, at the age of 20 y/o the professor and my tutor at Uni was hugely into Patek Philippe watches. I had a bit of interest and because Patek make hand made watches they after show images of their production shops and this brought out my hidden engineering interests. Knowing I couldn`t buy a Patek, I purchased some old pocket watches off ebay to take apart (Smiths for those interested). I later found out it must have been total luck, but I managed to fix one and suddenly thought I was a clock maker. I decided to buy a Unimat 3 lathe off ebay that I somehow won it at a fantastic price. I made a few posts on NAWCC firum and a chap over in Yorkshire invited me around to see his Unimat 3 and clock builds and repairs. In his home we discussed the current ME article showing John Wilding who had just finished writing his plans for Dr Woodward’s Gearless Clock and was posting them in Model Engineer mag. Being a fool, I ordered all the materials and purchased the copy of ME. I asked my dad for some help and one Saturday morning we drilled 3 mounting holes for the brass back plate and failed when it came to soft soldering 3 washers on the back!!! I realised very quickly that I didn`t have the skills to make a clock and the materials and lathe were sold.
So now in 2011, with a couple of years experience behind me, I`m going to give this another go.
I recently ordered all the brass plate and received this a couple of weeks ago and I`m just waiting on the remaining materials, mainly imperial bar stock to arrive from College Engineering Supplies. I`ve also taken delivery of a length of nylon cord and some 0.006” spring steel. Last week while on holiday, I appreciatively took delivery of some 1/8” ID bearings from Clive off Madmodder!
This is the clock built by John Wilding:
Dr Woodward originally wrote about the design in his book – My Own Right Time but sadly I can`t find it in the library. I have of course purchased John Wildings lovely write up of the plans and build process. He mainly cuts the clock using a Unimat, I`ll be following his methods unless I can utilize my small mill to better effect which he doesn`t use.
I should state right away that this build log has be written already by the notorious GadgetBuilder who details fantastically his build of this clock. He has made some excellent modifications of Wiliding’s design, especially the automatic maintaining works. Unfortunately I`m not confident, intelligent or daring enough to deviate from the plans other than a couple of fasteners where I`ll be using metric instead of the specific BA series.
The clock uses a really unique method to run. A series of colliding pawls, hooks and rods give the pendulum an impulse, with the energy provided by the large weight.
Until I have pics to explain in more detail, the best way I can explain is:
There are two main aspects, the running train and the daisy motion. I`m going to start back-to-front and talk about the daisy motion first. Basically the “clock” mechanism rotates a main arbor once each hour – the minute hand is connected to this. The daisy motion is used to convert that 360 degree hour spin, into a 1/12 of a turn for the hour hand. Here is a youtube video made by gadgetbuilder (since I didn`t make the video I`ve given it as a link rather than making it display on the website): http://www.youtube.com/v/fdb7_VCaiOE&rel=1
This is a really neat and novel idea allowing you to generate your hour movement from the minute movement – remember, without gears!! The smart thing is, a taper pin is removed and the entire daisy motion mechanics can be removed! This is why I mentioned this first, that just leaves us with the rest of the motion works.
Probably best to look at gadgetbuilders explanation on his website but here goes:
This clock only receives an impulse to the pendulum once every 60 seconds! This means everything involved with the count wheel must be very low on friction. Look at the pendulum and you will find two brackets. The upper one catches a tooth on the count wheel (the top wheel that looks like an escape wheel, NOT the pin wheel) and the pendulum is of such a length that in 60 seconds the count wheel goes around once. I`m wondering if a minute hand could be attached to the countwheel??
You probably can`t see in the photo, but one tooth on the count wheel is slightly deeper than the rest. When the pawl on the pendulum drops into this tooth, it pulls a lever down at the rear of the wheel. This lever has two section a bit like this “ > “ so as the top lever is pulled down anticlockwise, the bottom lever also moves down. This bottom section pushes down on the bottom “impulse pawl” of the pendulum (the pawl connected to the bottom bracket of the pendulum rod). This actually pulls the count wheel (the pin wheel) anti-clockwise for a second but at this point the weight which is wrapped around the count wheel to pull it clockwise, pulls the wheel back clock wise again. Now however, the levers are not bearing down on the impulse pawl (the count wheel pawl is back in a normal small tooth) and it is free to lift up out of the pin it is holding (the pin wheel also moves through a gate and moves forward by one pin, i.e one minute!!!!) but now the pendulum is at a higher position again – i.e the pendulum has been given an impulse! The pin wheel moves on by one pin every minute and therefore with 60 pins, the pin wheel does one revolution every hour – the minute hand! Add the daisy motion to this and you have your minute and hours!
I`m not going to rush this but to be honest, with no gears the parts don`t look overly complex so I expect to make good progress. I expect like I have experienced with my Webster IC engine, most of the time will be in the troubleshooting, getting it to run!