PENDULUM DRIVE

Food for thought:

Tim

-- Seven Transistor Labs Electrical Engineering Consultation Website:

Thanks Tim!

That certainly looks like it'd do the job as I described it.

Not much power is required as the only energy drain is the air resistance of the slowly swinging pendulum and the miniscule friction of the pivot it's suspended from.

Jeff

On Sat, 09 Nov 2013 23:24:28 -0500, Jeff Wisnia Gave us:

I would think that too much energy would be consumed keeping it moving.

Most of those wall clock jobs last a year or even two on a single double A or triple A cell. I cannot imagine this one would last very long if it has to pull on that pendulum arm once every certain point.

But it sounds like the same "motor" action used on the hard drive actuator for the arm the head hangs on... the head arm.

I think that would mean there would be a magnet up there on the (drive) end of the pendulum rod.

On Sat, 09 Nov 2013 23:43:22 -0500, Jeff Wisnia Gave us:

No... Not if it is a "motor". If there is a kick taking place, it draws power. If it were a standing field, it would not perform the kick. The long arm and the pivot it hangs on is not where the drop takes place. Lock-stepping the proper kick to keep the proper cadence has a few factors going on. If the arm has a portion extended *above* the pivot (it should). That will be the "rotor" of the motor, and the fixed coil is the stator. So that small above the pivot extension should be what passes right in front of that coil.

Otherwise, we would have PMMs which could actually perform real work.

The long arm and the pivot it hangs on is not where the drop takes place. Lock-stepping the proper kick to keep the proper cadence has a few factors going on. If the arm has a portion extended *above* the pivot (it should). That will be the "rotor" of the motor, and the fixed coil is the "stator". So that small above the pivot extension should be what passes right in front of that coil. Since it make no full rotation, but functions within a small arc, it is hard for some to picture as a motor, but it would be just that.

As to the PMMs (perpetual motion machine) remark...

"We are only ten years from a solution, however..." --several

The bearings in atoms are pretty slick, low friction devices. ;-)

Just have to keep them out of the hands of the assholes.

Yes, Tim's link is appropriate. Might take some twiddling, but it can be handled.

The magnetic "kick" to overcome the friction is only large enough to overcome losses. Even if not, the swing will just diminish. I had one of those clocks and it ran for a year before battery replacement. I think that may be the rule for many battery-operated devices. Kinda makes sense, really

There is no portion of the pendulum rod extending above its pivot point.

The steel piece (which I agree must must be magnetized) on the back of the rod is about an inch and a half below the pivot.

I prefer to think of the system as an oscillator, where the "tuned circuit frequency" is the inverse of the natural period of the free swinging pendulum.

Anyway, it works.

Jeff

some have a D cell for the pendulum and an AA for the clock movement.

The other way it is done is that the driver only gives the pendulum a kick when its amplitude has fallen below a particular threshold. Or there is a means to limit the current so that the pulse of current is just sufficient to maintain the motion at a fixed amplitude.

Badly adjusted electric observatory pendulum clocks in the days before cheap quartz crystals used to clip when they hit the enclosing box if the drive was made too enthusiastic. It didn't do them any good.

I suppose if the Q is sufficiently high, it could drive at constant frequency with a slight excess and be done. The frequency of a pendulum is second order to amplitude and drops with increasing amplitude (something mostly unfamiliar in electronics, though a material with low initial permeability might be able to simulate the second order form; otherwise, saturation causes frequency to rise suddenly).

Tim

I recall a prototype super Foucault pendulum of extremely high Q that was pumped by moving the suspension point up and down at 2f. It was designed by Cavendish Professor Sir Brian Pippard. He was very much into parametric amplifiers and other cute engineering designs.

He was the first person to experimentally determine the Fermi surface of copper.

Speaking of B. Pippard, "Physics of Vibration Vol I." is a nice read. (for geeky science types.) I guess I should try Vol II.

Hey, did you know that if you make a swing long (high) enough it becomes too big to pump by yourself? (Personal experience.)

George H.

Do you have any idea where in the travel it is lifted and how much it needs to be lifted, (in relation to what?) and percentage of the duration it needs to be lifted. (for a decent Q pendulum) (Seems like you would need to work at it to make a low Q pendulum {-;) Don't really want the math, just trying to incorporate the concept. Mikek

Maybe I can come up with a stick, string, and weight here on my boat and figure it out.

That didn't take long. I used a ruler, string, weight and a fulcrum. At the peak of each swing you start raising the pivot. Took me several tries to do it a 2f. 1F works. It very much feels like the kick you give

• it is a swing, but, I wanted to say, I hate the safety swings that are a strap that tightens up on your butt, making it impossible to jump out and also very uncomfortable? A lawyers idea!

idth=1024

Oh If you've never done this you need to take yourself to a playground righ t away. (unless you are old enough to be worried about breaking a hip.) F ind a swing you can stand on, getting it going a bit*, and then raise and l ower yourself. (Squat at the ends and stand up in the middle of the 'swing '... well I think that's the recipe.) Oh I guess you can do this with a rope, weight and bar too... but not nearl y as fun, nor memorable. (I recall a cartoon of monks swinging an incense filled container.) Well a quick web search only turned up these,

George H.

Raise it slightly each time it goes through the vertical axis.

It is the same principle as a child pumping a swing. Although if the thing had been built full scale (in a pit at the south pole) the intention was to test GR so the mount was incredibly over engineered.

Ideally pump with a sine wave at the second harmonic of the main pendulum. Pumping with a square wave is more entertaining.

Probably more maths than you really wanted but you can still look at the diagrams (and it isn't in Russian).

This is a nice one to try if you have a suitable mass, spring and string in your junk box. It is quite impressive if the spring resonance is close to twice the pendulum resonance (easily adjusted by length).

I urge you to play with this one it is most entertaining (and cheap). Good demo for an Xmas science lecture.

You can also stabilise a rigid rod a la Indian Rope trick this way.

The movement is powered by one AA battery.

If the battery life is annoyingly short there's plenty of room inside the clock to mount a D cell holder and connect it to the movement.

Jeff

You don't think that cheap "quartz" clock is going to be temperature sensitive or even inaccurate? If it doesn't synchronize to NIST, it's junk.

If it keeps time as well as the numerous other non-synchronized $12 wall clocks I have in our home and our office I'll be just fine with it. I'm not wanting GPS accuracy you know.

BTW, my last job prior to retiring from electronic engineering about a dozen years ago was building and testing cesium atomic clocks flown on board the GPS Block II satellites. Now THEY were accurate.

Jeff