By Daryl Bender - Ottawa Canada
(a newer but still unfinished page . . . I've now completed the coil "bobbins" and I am researching coil winding)
For those who are unfamiliar with Foucault pendulums I include a bit of an explanatory diagram but in simplest terms they are simply an oscillating mass suspended by a wire. They usually have a separate drive only so as to counteract the inevitable frictional forces which would otherwise cause them to stop swinging. Because of inertia, a pendulum, once started, tries to maintain its plane of oscillation. Classically Foucault pendulums are made very long so as to minimize any extraneous influences which might affect their planer oscillation. When one sees a Foucault pendulum however the plane that the bob is swinging in appears to sweep around like the slow hand of a clock. What is happening, and what it is intended to demonstrate, is that the rotation is not that of the pendulum bob itself but that of the massive Earth turning on its axis underneath it! Hence, it is a unique device which can demonstrate the tiny but influential Coriolis force. It is Coriolis forces which cause the more familiar pattern of weather spiraling around a pressure system such as can be seen at the top of this page. It also raises fundamental questions about matter, inertia, relativity (see "Questions" by Nickel below) and even some mysteries which are still not understood.
I have always had a long standing fascination with Foucault pendulums. I think this was set off when attending Carleton University. I would pass the Foucault pendulum designed by Karl-F. Hafner (see pg 5) in the lobby of the Herzberg Physics building (below) several times a week.
Here next to the stairwell was a 12 inch diameter 231 lb brass sphere suspended from the roof, 5 floors up (55ft), by a 0.078" wire making slow (4 sec) oscillations each way. I would always look down and note its position while going up the stairs to my physics lab and, 3 hours later when leaving, never cease to marvel at how it demonstrated the cumulative effect of Earth's rotation beneath its swaying mass. So simple a device yet so marvelously profound!
Years later I was searching the web for information on Foucault pendulum drives for other horological (clock) pursuits when I bumped into the most remarkable page entitled "A short discussion by Professor B. Nickel, Physics Department University of Guelph". Here professor Nickel not only described one built at the university but provided a wonderful background on the history of the Foucault pendulum, explained the effects and raised some most fascinating further questions. What also astounded me was this working pendulum was only 0.7m long! It immediately rekindled my old interest in them. His page described how their design was based on an American Journal of Physics article by Prof. H. Richard Crane. After contacting the OJPS (sellers of the journal) I was able to obtain a copy of the magazine containing the article. While the article contains a set of plans for a 0.7m pendulum Crane also describes the improvements for which he was able to get accurate operation on pendulums as short as 5.90"!! Here at last was a unit which was of reasonable proportions for the home.
This discovery occurred at about the time I was doing my Blick Propeller Clock.. Hence, at the same time I did the board for the clock, I also made a PCB for the pendulum drive. The schematic capture ultimately resulted in the layout seen below. The PCBs were manufactured by APCircuits of Calgary Alberta while the parts were obtained locally and at Digi-Key.
The picture above shows the following; A transformer is used to set up power on the board. The main drive board has a 555 timer. The timer is triggered by a small external sense coil below the pendulum bob. Having been triggered by a sense pulse the timer then provides a pulse to the larger drive coil also below the pendulum bob. These coils are connected to the DB-9 connector. Separate external pots and switches (connected to the DB-25) are used control the duration and amplitude of the pulses described above. The large white battery-powered isolated-output timer is used to switch ON & OFF a third coil which is part of a clock function. This third coil lays flat on the deck below the bob and when energized will prevent the pendulum bob from "advancing" due to Coriolis forces. It works like this. The pendulum takes approximately 36 hours to "rotate" around the deck at the latitude of Ottawa. This makes a clock scale of hours inconvenient however that also means it also does a 1/2 "rotation" in 18 hours, so, by preventing the bob from "advancing" for approximately 6 hours each night a clock scale of hours can be provided. I set this feature up as optional. My aim was to have the PCB and all the controls contained in a separate small wooden box which would sit just below the finished unit.
It was at this point I started my version of the design. The fundamentals are basically a reverse engineering of the design in the Crane article but the form was my own interpretation. The overall pendulum design can be seen in the following pictures.
The first picture shows the overall pendulum design. The second picture shows the lower coil bowl area and the pendulum bob. There will be an acrylic air shield around the lower bob to guard against gusts and busy fingers. For the wooden frame of the pendulum I'm inclined to choose walnut. I have some walnut in the basement and in my opinion walnut is a very good compliment to all the brass parts that are used in the design. The design of the frame is quite simple with the only challenge being the round section containing the coils at the bottom. My plan is to turn the wood for the outer bowl rather than use cardboard. It will be a challenge to glue up the multisided blank for turning. The middle picture is similar but from below and you can see the adjustable magnet setting at the bottom. The next picture shows the "catch" assembly. I have currently built the version as called for in Crane's design namely a Starrett pin vise for holding the bob-suspending 0.032" music wire at the top. I'm now second guessing myself with an alternate idea used by Marcel Betrisey. The picture reflects that design. It uses a clamp and wire die arrangement to hold and guide the wire. The die acts a bit like a Charron Ring. I like the idea that the wire will bend on a curve rather that sharply as in a pin vise. This should reduce wire stress and increase wire life. I've now downsized the wire to 0.012" which should further improve performance. In either case there are some blue shock isolating grommets (in case of wire breaks). These came from Small Parts. The last picture shows the lower coil (electromagnetic drive coil). The copper coloured ring on the top is the Charron ring to help minimize ellipsity. The whitish colour is to represent acrylic plastic,
The next major challenge was the pendulum bob itself. I went and bought a 4" dia. cylindrical piece of brass about a foot long. A little later I bought some 1.5" thick aluminum plate to make a spherical turning attachment for my lathe. A four inch diameter brass sphere is not an easy thing to make. As the full scale of this undertaking began to hit me I could see devoting a large percentage of my spare time in the summer on this alone. When I came across Liberty Brass I gave them a call. They said they could do it no problem and even had the basic 4" brass sphere in stock. After considering the alternative for a day or two I sent a fax stating what I wanted etc and within a week it was in my hands (as it was a cold January there was no end of snide comments on the contents of my package when it arrived). Per my request to Liberty Brass I was going to do the final reaming of the 15/32" axial hole to 1/2" so as to ensure the brass tube that was to be fitted to it was snug. I reamed the hole as follows. Using a forstner bit I drilled a 2" hole in a piece of 2" x 4". This served as the soft yet firm ball support. I then chucked a 15/32 drill into my mill/drill machine and, with machine off, used it to manually align the hole to the axis of the chuck. Taking care not to move the ball in the slightest I then chucked up my 1/2" reamer and held the ball with my hands while my wife advanced the quill. 30 seconds later it was all done and the heavy 0.065" wall 1/2" brass tubing that goes through the bob now fits very snugly.
I had a lot of difficulty finding a source of magnet wire for the drive coil and the sense coil. In the end I bought these at the "store of last resort for impatient fools", eBay. I obtained the specified cylindrical magnets that are used in the base of the pendulum bob and below the pendulum bob at Edmund Scientific. The music wire to suspend the bob was obtained locally at a hobby shop as was the balsa and brass wire for the indicator. The balance of the brass for the other major parts and the sheet for the scales was obtained at the local Metal Supermarket.
Machining the Brass-work
Above is the lathe roughing out some 2 3/4" diameter brass for a sense coil support for my pendulum. As you can see a 2 3/4" piece is a trivial exercise for this big lathe yet it also has the finesse to drill a 0.035" dia hole with a tiny fragile #65 drill
Turning the brass "top nut" for the Foucault support. Remember etch-a-sketch? After rough shaping my home-brew spherical turning attachment has now been set-up to turn the 3/4" ball end on the "top nut". Note the "etch-a-sketch" curve has been smoothed out. Voila! Hey that was fun! Let's try it again on a tiny 3/16" radius ball.
Polishing "the ball" on a 1/2" mandrel. The brass bits are coming together here. Seen here are the 4" bob (the "ball") in the upper left, the top support and pin vise clamp assembly upper right, The bob support upper center, the washer like catch plate (for wire breaks) below that, The bob height adjustment and jamb nut in the center. Along the bottom is the lower magnet height adjustment rod (with the lower and upper magnets beside it). At the very right is the large assembly for centering and clamping the lower coil assembly. It also supports the sense coil. More of this can be seen in the pictures below. It was the piece made from the 3" stock shown above.
Making the Coils
Here the acrylic blank for the lower coil "table" is being cut on my bandsaw.
his is followed by a little lathe work to true it up and size it. Above are the side and top views of the lower coil assembly roughed out but yet to be glued. The scraggly look is just the acrylic's protective plastic. It tends to come free at the edges that have just been machined.
To protect the coils until completion I made a box for them (I have a similar box for the brass parts). The next two pictures show the protective wrap has been removed and the parts are now glued together ready for winding. In order to wind them they will have to be held in the winder so I made a winding spool for each coil.
Not Started ... but I have the walnut now
Constructing the Drive
Apart from the board shown near the beginning I only have the plan at present
Putting it Together
Not Started :-(
Here's an idea for a micro controlled Azimuth Tracker. This would allow automated data collection of the plane of swing for a Foucault pendulum thus enabling a detailed performance evaluation.
To be continued...
Links to other Foucault Pendulums