Mad Teddy's web-pages

My synchronous wheel (ca. 1967)

If you came here by way of my Electrical stuff menu, you will have seen a photograph of the front cover of "Model Making For Young Physicists" by A.D.Bulman (John Murray, 1963). That front cover featured a drawing of the project about to be described here.

It's a very simple project, but it needs to be well-made if it is to perform satisfactorily, because it generates very little power - and every milliwatt is precious. (Now, why do those last four words remind me of Monty Python??) If it's made without attention to detail, it probably won't work at all.

On the other hand, if it is made carefully, it can be a source of great fascination. Simple it may be; but it's possible to do quite a lot of very interesting physics with it. (I'd like to think that science teachers may find it a source of inspiration.)

My own model started out pretty much as Bulman described it, and worked reasonably well for a while. Over time, however, the second law of thermodynamics took its toll: parts wore out or went rusty. Eventually the electromagnet was removed and used for something else, with the remaining bits tied together with a piece of wire, stored away and forgotten about.

In 2002, I found it and decided that it would be fun to get it going again. After a few hours of re-building, fiddling and adjusting, I came up with a pretty neat little gadget. Here it is:

The spindle consists of a large sewing needle with the eye cut off. Onto this is fixed a cross made from twelve small plates cut from thin sheet iron, each 25mm x 6mm (1" x 0.25"). Each of these plates has a small central hole, just big enough for the needle to pass through. The plates are stacked up as laminations so that they run alternately "north-south" and "east-west".

The iron sheet we (my Dad and I) used for the laminations was taken from a spool which at some stage had probably held a length of wire. The outside of this spool was painted pale blue with red markings (as can be seen from the picture above). This layer of paint probably helps (if only slightly) the performance of the model, as it provides a layer of insulation between the plates and thus helps to minimize eddy currents which could have a dampening effect.

I'm not quite sure how my Dad fixed the cross to the needle. He probably followed Bulman's instructions fairly closely, the cross being wired in place while being soldered, in the hope that enough solder would hold the strips in place on the needle. Or perhaps he just glued it together with epoxy glue - I'm not sure. Whatever he did, it was a neat and effective job; and the whole assembly has stood the test of time extremely well!

In this next photograph, you can see the construction of the cross quite clearly.

Ignore the stack of coins peeping out above the flywheel - they're just propping the model up, as I had it standing on end while I took the picture!

You can also see the upper support, which is attached to the base under the electromagnet by a couple of nuts, bolts and washers. This is bent up at right-angles to rise between the rotor and the electromagnet; finally, another right-angle bend completes the zigzag, and the rotor is supported by a hole drilled in the top horizontal section.

Actually, when I revamped the model, I reinforced this upper support by attaching another short piece of brass strip as a strut to produce a triangle, as shown here:

This adds considerably to the overall robustness of the model.

You can also see a smaller brass strip, bent into a serpentine curve with a small hole drilled at one end to lend support to the lower end of the rotor. The strip is attached to the base by a single bolt through a larger hole at its other end (more about this shortly).

In Bulman's original model, there was no lower support strip. His model relied on the upper support and a strip of brass with a small indentation to support the needle at its point.

(Also, his upper support was attached near the edge of the base, where our lower support was attached).

The original flywheel was made from a disc of hardboard 0.6cm (0.25") thick and 3.2cm (1.25") in diameter. The needle was passed through a small central hole, and the disc glued in place. Like the cross, this has lasted well also.

The flywheel was painted white, with a red circular dot painted near the edge above one of the cross-arms. (The reason for this will be explained shortly.) When I revamped the model in 2002, I added a 4cm diameter red plastic bottle cap with a central hole, which fits quite snugly over the hardboard disc, and placed on it a 3.2cm white circular sticker which I had coloured black with a marker pen. I then placed a smaller (8mm diameter) white circular sticker on the black one, near to the edge, and again above one of the cross-arms. Part of the reason for doing all this was that the old flywheel was a bit tatty, and I decided that the whole thing could do with a face-lift. It turns out that the extra mass also provides a stronger flywheel effect (i.e. more angular momentum), which I think is probably a good thing.

In the original model, made following Bulman's instructions fairly closely, the point of the needle rested on a strip of brass of the same type as the supports, attached to the base by a couple of small screws. A small indentation was made in this brass strip for the needle point to rest in.

The problem with this was that, over time, the spinning needle drilled itself deeper into the soft brass strip until it actually became a rather tight fit, completely defeating the strip's purpose. One could always replace it; but it occurred to me that it was possible to do a lot better.

There used to be a glazier's shop in my home city; sadly, this became just another victim of economic rationalism and globalization a few years ago. On occasion, I'd buy various little things from them for various projects. One such project (I won't go into detail) required small glass discs. It turned out that, from time to time, the glazier found it necessary to bore holes in sheets of glass to meet some customer's special needs. They put the resulting discs aside and eventually threw them out.

Ever on the lookout for useful bits and pieces, I asked if I could have some of these little discs. Sure enough, they gave me quite a lot!

One of these discs found its way into this project. Using a glass drill, I made a small conical depression in one side. I placed a circular sticker on the other side, and glued it to the base with ordinary wood/paper glue so that, with the needle vertical, its point would rest in the depression.

I mentioned earlier that I'd have something more to say about the lower support for the rotor.

Theoretically, the lower support is not necessary. (This was obviously Bulman's point of view, too.) If the needle point is sitting in the conical depression in the glass disc, and the rotor is supported above the cross, why would it need any other support?

The reason is that the rotor is started spinning by a quick twist with thumb and forefinger ("Here comes the twister" ). No matter how carefully this is done, it is very difficult to avoid a certain jerkiness in the action, so that the needle point will quite likely not come to rest in the small depression. (It may even be damaged during the process, if too much vigour is used.) So the lower support is invaluable for keeping everything more or less in place during this operation.

Once the rotor is spinning, however, the lower support becomes redundant. Therefore, it's highly desirable that the needle should not touch the sides of the hole once it is released, so as to reduce overall frictional losses.

By attaching the lower support to the base firmly, but at a single point, it's quite easy to adjust things so that this requirement is met.

Now: to the electromagnet.

An old soldering iron I'd had since the early 1970's eventually wore out to the point where it was no longer a going concern. At some stage in the early 1990's, I pensioned it off and bought a modern temperature-controlled Dick Smith solder station.

I don't throw old things out. (My wife calls me a "hoarder".) So I still had the beat-up old soldering iron lying around in 2002, when I decided to rescue the synchronous wheel.

The soldering iron had its own transformer, which I dismantled as carefully as possible. A hacksaw was needed to cut through the laminations - a somewhat hair-raising job, if one is to avoid damaging the windings. This successfully accomplished, I unwound the wire. I used half of the primary winding to make the electromagnet for this project, and eventually used the other half for the next project described in this website (the single-solenoid electric engine ). The thick, short secondary I put aside. (It may see service as the primary of a Tesla coil at some later stage.)

I don't know exactly how many turns there are on my electromagnet. Bulman suggests winding on from 300 to 600 turns.

I used a section of the laminated core of the now defunct tranformer as the core for my new eletromagnet. I prepared two ends from a piece of aluminium sheet, and placed them on the laminations. I covered that part of the laminations between the ends, and the insides of the aluminium ends themselves, with gaffer tape, and wound the wire on, finishing the work with a covering of more gaffer tape over the windings.

The piece of transformer core I used already had holes drilled in its ends; bolts had passed through these in the original transformer, to hold it together. I passed long steel bolts through these holes, added a nut to each to anchor the bolts firmly to the core, and then placed pieces of chrome-plated brass tubing liberated from an old telescopic TV antenna over the bolts for most of their length.

I had a couple of large-diameter bolts left over from an earlier project. The first couple of centimetres of these bolts, next to the heads, was unthreaded. I cut the heads off, and also cut the threaded sections off, leaving two short rods each tapered at one end. I used a file to make flat areas on each of these near the untapered ends, and drilled holes to accommodate the long bolts. Finally I used nuts to secure these rods onto the long bolts to give the structure you can see in the photographs.

I used pieces cut from small right-angle shelving brackets bolted to the aluminium end-pieces as supports for the white banana socket terminals, to which I then connected the ends of the winding. Finally, I added some more gaffer tape to give a tidy finish to the lower part of the structure, and attached the whole thing to the base with four screws, two at each end.

I've gone into considerable detail regarding the constuction of this project. Why?

Well, basically because I feel that if you only build one of the little mechanical models described in this website, this one has the most to offer from a scientific point of view. If you do decide to build your own synchronous wheel, I'd like to try to ensure (as far as possible) that it works well, so that you may have as enjoyable and educational an experience from its use as I've had from mine.

It's necessary to have a source of AC power to operate the synchronous wheel. Mine runs very nicely from my old power supply . Because of the need for AC, the bridge rectifier is not used.

If you have access to an AC power supply, you're in business. If you don't have one, you can probably obtain a suitable plug-pack with an AC output. With my power supply set to its 14V output, the model draws about half an amp (which means that it's then running at about 7 watts).

All set?

So, you switch on, give the rotor a twist, and away it goes! Right?

Not quite. You need to spin it at just the right speed, with very little room for error. However, if you do, it will indeed keep on spinning!

So what is the right speed?

This has to do with the frequency of the AC supply in your area. In Australia, where I live, and in some other countries including the UK, the frequency is 50Hz (i.e. 50 cycles per second). In the US and other countries, the frequency is 60Hz.

In each cycle (1/50 or 1/60 of a second, depending on where you live), current flows in the coil first in one direction and then the other, in a smoothly-varying sinusoidal fashion. As a result, a sinusoidally-varying magnetic field (denoted by B, and measured in teslas) is generated between the poles of the elecromagnet, first in one direction (north-south) and then in the other (south-north).

The critical point here is that a strong magnetic field occurs twice within each cycle: one-quarter of the way through the cycle, and three-quarters of the way through it, as indicated in the diagram above. The fact that the field's direction reverses every time is immaterial for our purposes; the cross laminations are made up from un-magnetized soft iron, which will happily line up with a magnetic field oriented in either direction.

Thus a magnetic field capable of lining up a set of cross laminations in any one of four rotor orientations, 90 degrees apart, occurs 100 (or 120) times each second.

Now, the cross has four arms, each of which passes each of the two poles of the electromagnet on each revolution. If the rotor is spinning so that the arms of the cross are close to the electromagnet poles each time the field strength is at a maximum, and are as far away as they can be (i.e. at a 45-degree angle to the direction of the magnetic field) when when the field strength drops to zero, it must be spinning at one-half of a turn per AC cycle. This means that it will be spinning at 25 revs/sec (or 30 revs/sec).

Hence, under these conditions, the speed of rotation, in revs/sec, is exactly half the frequency of the AC supply, in Hz (cycles per second).

Whenever the rotor is spinning, there are forces (friction with the upper bearing and the glass disc, air resistance) tending to slow it down. If it is spinning as just described, the repeated pull of the magnet on the cross laminations as they approach the poles will give the rotor just enough angular acceleration to overcome these losses, and thus take it safely through its next quarter-turn. This condition, in which the rotor is always "playing catch-up" with the AC power supply, produces synchronization and hence constant-speed rotation.

So: how do you get the rotor spinning at just the right speed, so that it will "lock in" and continue to spin at that speed? And how will you know when you've got it right?

Simply spinning it at a somewhat greater speed, and waiting for it to slow down to the correct speed and settle there, rarely if ever works. It will just slow down through that particular speed, and ultimately stop - unless you're very lucky (it's happened for me just twice).

It requires practice and a deft touch to spin it just right. It's not easy at first, as you are guessing how fast to spin it, trying to acquire the technique, learning from experience - and you may ask yourself: "How do I work this?"

Believe me - you will know when you've spun it at just the right speed. Within a fraction of a second, you will be aware that the little machine is settling into a rhythm.

You may happen to spin it so close to the required speed that it just sits there and continues to spin quietly as though it were the most natural thing in the world. You will experience a deep feeling of inner peace.

If you've spun at it just the merest fraction slower or faster than its required speed, you will become aware of a soft, slow throbbing sound ("whirr-whirr-whirr") as it alternately spins slightly faster and slightly slower, gradually - over several seconds - settling down until everything is in equilibrium. Then all you will hear will be the quiet, jiggly little noise of the spinning needle as it lightly interacts with its bearings, and the sound of your own breath as you slowly exhale in a sigh of satisfaction. Spiritual stuff!

Of course, having done it once, and got the "feel" of how to do it, the learning curve rapidly becomes easier. Same as it ever was...

Is there any way to simplify this learning experience? Indeed there is - and this is where the dot on the flywheel comes into its own.

Assume for the moment that the rotor is already spinning as described above, at half the AC frequency. Suppose also that ambient light is at a fairly low level, and that a fluorescent light is used to illuminate the model. What will you see?

No doubt you are aware that the light from a fluorescent tube is not steady, but flickers. (It can get quite annoying after a while.) How fast does it flicker?

A moment's thought will convince you that it flickers at double the mains frequency - for exactly the same reason that the magnetic field strength of the synchronous wheel's electromagnet achieves its maximum strength at double the mains frequency.

Now, since the flywheel makes one complete revolution over two complete AC cycles, and since the light from the fluorescent tube is at its brightest four times within those two cycles, it follows that the flywheel is illuminated brightly four times during those two cycles.

Thus you will see the white dot illuminated brightly four times within one revolution of the flywheel. This means that you seem to see four whitish areas, apparently stationary, spaced at 90 degree intervals around the flywheel.

You won't see four sharply-defined circular white dots, as shown in the left-hand diagram below; but rather a ring composed of varying shades of grey, with four quite bright patches and four rather dark patches within it, somewhat like the diagram at the right.

In fact, even that is not quite correct. You will actually see something like the right-hand diagram above - but rotated a bit. If the flywheel is spinning clockwise, it will appear rotated slightly anticlockwise, as in the left-hand diagram below. Conversely, if it is spinning anticlockwise, it will appear with a slight clockwise twist, as in the right-hand diagram below.

Why does this happen? As John Lennon once said, "You might well arsk". We'll deal with the phenomenon in the next page.

If you've made a synchronous wheel of your own following these notes, and had a bit of a play with it, you may be interested in further activities involving it. If so, click on this link:

Mad Teddy's synchronous wheel, page 2

Return to Electromechanical models sub-menu

My home page     Preliminaries (Copyright, Safety)