Electronics Kit for Control, Display and Logging a Foucault Pendulum.
Description of the Options.
Latest change
2025-11-03
In brief:
The kit is prepared for several options regading the drive timing and amplidude control of the pendulum.
In most cases changes must be made in the firmware and probably the PC-program.
Options for the Center Pass detection.
The detection of the bob passing the Center (Rest poition of the bob) is essential for all timing related functions.
Center Pass detection can be done in 4 ways:
1/ With a dedicated Center Pass detection coil located below the rest position of the bob. The magnet in the bob
will induce a voltage impulse during the fly-over. The voltage is
amplified and A/D converted by the Arduino. Detection of the pulse
serves as the starting point for all timings.
2/ The Drive coil itself can be
used for the Center Pass detection. For this special circuitry is
present, but not yet fully tested.
The advantage of this method is that there cannot be a misalignment between the Drive coil and Center Pass detection coil.
3/ With the capacitive method. This requires a central electrode to be
present below the bob. The wire, and so the bob, carry the 465 kHz
signal which will be picked up by that center electrode. The signal is
then amplified and rectified on the Receiver PCB, and routed to the
Arduino. This option has been tested to basically work in a pendulum, but no long term performance tests have been done.
4/ The hardware in the kit supports detecting that the
wire touches a Charron ring. This is a method often used in pendulums
with a simple driving system based on a delay circuit such as a 555.
The firmware and PC software have to be modifyed to use this method for DriveTiming. I
have never used this method and I can only give some suggestons for
modifying the software for this.
Contact between a grounded CharronRing and the wire is detected in the
hardware and firmware and reported in a Status bit on the GUI.
Note: When the wire contacts the Charron ring the 465 kHz signal on the
wire will change, or be shorted, which causes the PMS to go wild. At
this moment I cannot oversee the consequences of this.
If you do use a Charron ring but do not use it for drive timing, you can
insulate the wire where it touches the ring. That will have only a
slight effect on the operation of the PMS.
Options for the Amplitude Control
Generally some sort of Amplitude Control is required. Many (museum) pendulums do
not hvae an active system for this, but relay on a constant amount of
drive energy to be pumped into the system, and a constant amount of
damping (energy loss) which increases with the amplitude of the pendulum. Normally this balances at a certain amplitude.
However, the amount of energy pumped by the driving coil heavily depends
on the distance between the bob and the driving coil, so on the length
of the wire, which will be tempertaure dependent, and on the exact moment
the drive pulse is given and where the bob's magnet is above that coil at that
moment.
The "Schumacher" drive timing to suppress the intrinsic precession of
the ellipse requires the pendulum amplitude to be constant.
There are several methods to control the amplitude of the pendulum.
1/ Use of a Rim Coil. Below
the bob we have a coil wit a radius of ca.
2/3 of the desired amplitude. When the bob's magnet flies over it a
voltage is induced which is amplified and A/D converted by the Arduino.
The time it takes for the bob to go from the center to the rimcoil is a
direct measure of the amplitude. The signal amplitude of succesive
RimPasses tells us how well the rimcoil is laying in the horizontal
plane
2/ If one uses the capacitive method to detect the Center Passes one can
measure the half-height width of the signal from the center electrode.
This is a rather direct measure of the bob's velocity at the center
pass, BUT only if there is no ellipse. Generally we have an amount of
ellipse and so a correction is needed. If the PMS is operating and well calibrated this should be possible.
Detecting CenterPasses and determining the width of the centerpulse at
half-height and using that info for Amplitude Control is fully
supported. However, no corrections for ellipticity are implemented.
Using the Resonance Method to drive the pendulum:
This method is based on the circumstance that the period time of a
pendulum is amplitude dependent. The difference is small, but it might
work. See my article about the amplitude dependecy.
A.o. prof. Pippard has experimented with this method when he used the
so called "Piston Drive". I think it can be done without the expensive
and wearing-prone piston drive, but using the more reliable
electromagnetic drive.
The idea is to drive the pendulum with the frequency of the major
axis, and
not excite the slightly higher frequency of the minor axis. For this we
need a very
stable frequency source which can be adjusted in very small steps.
As I have planned to do such an experiment some provisions are made in the hardware and software.
The frequency of 465 kHzfor the PMS is generated with an AD9833 Direct Digital
Synthesizer. This device can be adjusted to generate that frequency in steps of 0.1 Hz.
To get to a frequency of 0.5 Hz for a 4-meter pendulum (e.g.) we need to
divide this by nearly 106, giving a 1:10 million resolution for the frequency steps.
The challenge is to keep this under control, where the period time might change due to a.o. temperature differences.
We also may want to use the "Schumacher" drive timing, which sets a second burden on the control system.
The current hardware and software support driving the bob with such a
frequency and controlling that frequency in very small steps.