Help! I'm stumped trying to figure this out.

My sister has one of those perpetual motion
thigamajigs. It's a pendulum, presumably
ferromagnetic, about 10" long. It swings above
a base which contains a battery and a coil.
Simple enough, n'est-ce pas?

Obviously, the coil current transfers a 'kick' of
energy to the pendulum on each pass. Now here's
the riddle: the pendulum's motion is inherently
symmetric, which implies that there should be a
cyclic transfer of energy back and forth between
the B-field and the pendulum. I.e. on the downswing,
it should gain kinetic energy, and then on the
upswing, should lose energy to the field (or perhaps
vice versa). Hence, no net transfer, and the
mechanism should simply halt.

But the dang thing goes on and on! There IS a
one-way energy path. Anyone care to take a crack at this?


--
Rich

you already answered your own question... the base contains a battery, read
this as 'source of energy', the 'kick' on each swing that makes up for
losses due to mechanical friction and hysterisis heating or induced current
heating of the metal parts.

"Rich Delaney" wrote in message
m...
Help! I'm stumped trying to figure this out.

My sister has one of those perpetual motion
thigamajigs. It's a pendulum, presumably
ferromagnetic, about 10" long. It swings above
a base which contains a battery and a coil.
Simple enough, n'est-ce pas?

Obviously, the coil current transfers a 'kick' of
energy to the pendulum on each pass. Now here's
the riddle: the pendulum's motion is inherently
symmetric, which implies that there should be a
cyclic transfer of energy back and forth between
the B-field and the pendulum. I.e. on the downswing,
it should gain kinetic energy, and then on the
upswing, should lose energy to the field (or perhaps
vice versa). Hence, no net transfer, and the
mechanism should simply halt.

But the dang thing goes on and on! There IS a
one-way energy path. Anyone care to take a crack at this?


--
Rich

In article ,
Rich Delaney wrote:
Help! I'm stumped trying to figure this out.

My sister has one of those perpetual motion
thigamajigs. It's a pendulum, presumably
ferromagnetic, about 10" long. It swings above
a base which contains a battery and a coil.
Simple enough, n'est-ce pas?

Obviously, the coil current transfers a 'kick' of
energy to the pendulum on each pass. Now here's
the riddle: the pendulum's motion is inherently
symmetric, which implies that there should be a
cyclic transfer of energy back and forth between
the B-field and the pendulum. I.e. on the downswing,
it should gain kinetic energy, and then on the
upswing, should lose energy to the field (or perhaps
vice versa). Hence, no net transfer, and the
mechanism should simply halt.

But the dang thing goes on and on! There IS a
one-way energy path. Anyone care to take a crack at this?


--
Rich


It's not a static magnetic field in the base. If it were, the toy
wouldn't need a battery.

It's all in the timing. There's probably one or two pickup coils in the
base that will sense the swinging magnet's position.


--
"Very well, he replied, I allow you cow's dung in place of human
excrement; bake your bread on that." -- Ezekiel 4:15

Rich Delaney wrote:
Help! I'm stumped trying to figure this out.

My sister has one of those perpetual motion
thigamajigs. It's a pendulum, presumably
ferromagnetic, about 10" long. It swings above
a base which contains a battery and a coil.
Simple enough, n'est-ce pas?

Obviously, the coil current transfers a 'kick' of
energy to the pendulum on each pass. Now here's
the riddle: the pendulum's motion is inherently
symmetric, which implies that there should be a
cyclic transfer of energy back and forth between
the B-field and the pendulum. I.e. on the downswing,
it should gain kinetic energy, and then on the
upswing, should lose energy to the field (or perhaps
vice versa). Hence, no net transfer, and the
mechanism should simply halt.

But the dang thing goes on and on! There IS a
one-way energy path. Anyone care to take a crack at this?

Are you sure there isn't also a package of electronic circuitry in the
system somewhere? Obviously the battery isn't merely running current
constantly through the solenoid. Otherwise, as you note, there'd be no
important difference between it & a pendulum swinging in the gravity
field. Also, the battery would be drained awfully quickly.

But there's a permanent magnet in the pendulum bob. When it swings it
induces current in the coil. On one swing, the current runs one way, on
the the backswing, the current runs the other way. In fact, it's a
continuously alternating current, graphed by a sine wave. A circuit
could be made to recognise when the bob is at some phase in its swing,
and it'd then tell the battery to let a burst of stored energy into the
coil, which then pumps the pendulum with just enough energy to
suppliment its losses.

-Mark Martin

I suspect all the base needs to do is apply a repulsive magnetic pulse at
regular intervals (remember how pendulums work). No need to sync it to the
swing of the pendulum - get the frequency and pulse width right and that
will probably happen all on it's own.

In article ,
(Rich Delaney) wrote:

Help! I'm stumped trying to figure this out.

My sister has one of those perpetual motion
thigamajigs. It's a pendulum, presumably
ferromagnetic, about 10" long. It swings above
a base which contains a battery and a coil.
Simple enough, n'est-ce pas?

Obviously, the coil current transfers a 'kick' of
energy to the pendulum on each pass. Now here's
the riddle: the pendulum's motion is inherently
symmetric, which implies that there should be a
cyclic transfer of energy back and forth between
the B-field and the pendulum. I.e. on the downswing,
it should gain kinetic energy, and then on the
upswing, should lose energy to the field (or perhaps
vice versa). Hence, no net transfer, and the
mechanism should simply halt.

But the dang thing goes on and on! There IS a
one-way energy path. Anyone care to take a crack at this?

There is probably a transister involved as well. There is probably a lead
from a sensor coil on the driver coil core that is connected to the base
or gate of the transistor. There are thus probably three leads off the
coil co a ground lead, a sensor tap, and the main driver lead. The
pendulum has a magnet in it. As it approaches the coil the flux change
dB/dt in the coil core generates a voltage in the sensor coil that causes
the transistor to on the driver coil. The driver coil is arranged so as
to increase the flux in the coil core. This creates a positive feedback
that sustains the driver coil current until the magnet starts to depart
the vicinity of the core. At this time dB/dt reverses polarity and
negative feedback shuts off the driver coil current, leaving the pendulum
free to contiue on with the momentum it gained when attracted to the coil
core as it approached. If there is no change in local magnetic field,
i.e. no swinging pendulum, then the device is automatically in off mode.

Regards,

Horace Heffner

Rich Delaney wrote:

Help! I'm stumped trying to figure this out.

My sister has one of those perpetual motion
thigamajigs. It's a pendulum, presumably
ferromagnetic, about 10" long. It swings above
a base which contains a battery and a coil.
Simple enough, n'est-ce pas?

Obviously, the coil current transfers a 'kick' of
energy to the pendulum on each pass. Now here's
the riddle: the pendulum's motion is inherently
symmetric, which implies that there should be a
cyclic transfer of energy back and forth between
the B-field and the pendulum. I.e. on the downswing,
it should gain kinetic energy, and then on the
upswing, should lose energy to the field (or perhaps
vice versa). Hence, no net transfer, and the
mechanism should simply halt.

But the dang thing goes on and on! There IS a
one-way energy path. Anyone care to take a crack at this?

1) Parametric amplifier.
2) A kid "pumping" on a swing.
3) Remove the battery. Damp the pendulum to hang absolutely
still. Replace the battery without disturbing the system --
nothing. Now, give the pendulum a little horizontal tap.
4) Cheap compass: the field is pulsed and synchronized.

--
Uncle Al
http://www.mazepath.com/uncleal/
(Toxic URL! Unsafe for children and most mammals)
http://www.mazepath.com/uncleal/qz.pdf

Uncle Al wrote in message ...
Rich Delaney wrote:

Help! I'm stumped trying to figure this out.

My sister has one of those perpetual motion
thigamajigs. It's a pendulum, presumably
ferromagnetic, about 10" long. It swings above
a base which contains a battery and a coil.
Simple enough, n'est-ce pas?

Obviously, the coil current transfers a 'kick' of
energy to the pendulum on each pass. Now here's
the riddle: the pendulum's motion is inherently
symmetric, which implies that there should be a
cyclic transfer of energy back and forth between
the B-field and the pendulum. I.e. on the downswing,
it should gain kinetic energy, and then on the
upswing, should lose energy to the field (or perhaps
vice versa). Hence, no net transfer, and the
mechanism should simply halt.

But the dang thing goes on and on! There IS a
one-way energy path. Anyone care to take a crack at this?


Yes the pendulum contains a magnet "N" side down (as per a compass ).
The base actually shows a little "S" magnetism. I can see a large iron
core electromagnet there. Up to a ~7º deflection the pendulum
oscillates and damps down. Larger than that and it will "pump up" to
full deflection. You can actually see the speed up when the bob
approaches the base.
I got this these for my kids. They won't let me take it apart.

Richard

Richard wrote:

I got this these for my kids. They won't let me take it apart.

In the name of all that's empirical, buy another one and tear its gutz
out man!!!



-Mark Martin

"Mark Martin" wrote in message oups.com...
Richard wrote:

I got this these for my kids. They won't let me take it apart.

In the name of all that's empirical, buy another one and tear its gutz
out man!!!



-Mark Martin

OK they're asleep now. I'm getting better at cracking stuff like this
open. Two gentle taps with my rubber hammer.... OK there are just
three basic parts to this thing.
1. 9V battery
2. 1 transistor? marked "LC845P" with the Motorola "M" logo on it.
Anyone know this type? I only get a reference to it being obsolete.
3. A large COAXIAL (inner and outer) coil. Tiny wire, measures .08mm
The coil itself is 20.87mm Dia. X 23.97mm Long. The inner coil
(different color) is about 10.65mm Dia.
Connections a
B- inner coil
Reading Xsistor L-R camfer up base up
L other end of inner coil 1 lead of outer coil
C B+
R other end of outer coil

That's it!
I think Gregory had it right with the sense coil and one drive coil.

Richard

Now I get to use that Plastix (Locktite) glue which BTW works really
well. Maybe solvent just in case.