pendulum energy pulser

You need the pulse at the top of the arc.

Insert a small iron bar into the arm of your pendulum near the top. Build a coil which surrounds the bar but is large enough to not interfere with the swing of the pendulum. On the bottom mass of your pendulum affix a small magnet at each side of the mass, each opposite from each other. The location of the magnet should be such that it will barely touch a Hall device at the top of the arc of the pendulum's swing. The activation of the Hall device should then generate a pulse that goes to the coil. The magnetic field in the coil will tend to center the iron bar you have installed at the top of the pendulum arm. This will give a small amount of energy to the arm which will transfer to the mass. Of course the arm of the pendulum must be stiff enough so it will not oscillate nor bend from the pulse in the coil. The pulse width going to the coil will have to be adjusted so that just enough energy is transferred to overcome friction, drag, etc. This you need to do by experiment.

Have fun!

Al

We have a fake-pendulum wall clock. It has a small coil near the top of the pendulum, which works against a small magnet mounted on the plastic shaft. The module runs from two AA's. Probably, it consists of a magnetic kicker, which is little more than a transistor and transformer. As the magnet swings by, it induces EMF in the transformer, which biases the transistor on, pulling a gulp of current and pushing the magnet. EMF reverses and the transistor remains off for a while. It takes a good minute to get swinging to full amplitude, so it's quite low power.

A two-dimensional version of this pendulum-kicker makes a chaotic pendulum. My physics prof has one in his office.

Tim

-- Deep Fryer: A very philosophical monk. Website @

Right, and it needs to be directed exactly perpendicular to the pendulum arm, because otherwise it'll change the effective value of g, which will change the period.

Cheers,

Phil Hobbs

If you put a magnet on it, you can use a reed switch and a coil to do it.

I think it will change the period almost no matter how you do it. It is like a crystal oscillator. The phase of the feedback can pull the frequency.

A reed switch could be activated by a magnet, and energize a small coil, maybe half of a ferrite pot core, or a transformer partially diassembled. That would be simple.

John

The usual approach is to put a small permanent magnet and a coil under the pendulum. When the iron pendulum bob is at nadir, it causes a small pulse in the coil (due to permanent-magnet field lines 'attracted' to the bob). So, first you detect these pulses. Then you make a timer to wait for 95% of the swing-out-swing-back time, then fires a current pulse in the coil to attract the (presumably nearby) bob downward. You want that pulse to have very short duty cycle. And you might want to skip a dozen or so swings until you trigger another pulse.

It's a self-oscilating unipolar motor, similar to how those perpetual "Top Secret" tops work.

It certainly can, and will if it's misaligned, but it doesn't have to.

If you adjust it so that it pulls down as well as outwards, you'll effectively increase g, which will shorten the period. If you make it pull up a little, it'll effectively decrease g, which will lengthen the period. Somewhere in the middle it will have no effect. If the timing is exactly at the bottom of the swing, the right direction is exactly horizontal.

Cheers,

Phil Hobbs