VISUAL ELECTRICITY                                     William J. Beaty

While working on an Electricity/Electronics exhibit for the Museum of
Science in Boston I discovered a number of serious problems in attempting
to explain simple electricity.  One problem was the obscure way that
exhibit devices typically present electrical effects: by using meters. 
Sophisticated skills are required to interpret meter readings and to
imagine the invisible events they imply.  

Exhibit designers who are familiar with electronics sometimes fall into
the trap believing that the general public also has their skills.  And so
designers include all sorts of digital and needle-meters as part of a
display.  But an electronics specialist sees a meter needle in quite a
different way than does the unskilled public.  Meter readings usually
serve more to obscure and complex-ify the exhibit than to reveal and
enlighten.  Think about it: if an exhibit makes you feel stupid, will you
end up LIKING science?  No.  The opposite occurs.

A second problem: nearly all of the electricity explanations I found in
children's science textbooks were wrong, so I couldn't use textbooks as a
guide for explaining electricity at a simple level.  Go here
for more about this.

And third, as an electrical engineer I had a gut-level feel for the math
behind electronics, yet my entire non-math picture of electricity was
mostly based on the incorrect explanations found in K-6 textbooks.  As a
designer, I'd been living in a world of electronics math, never realizing
that my verbal and visual explanations for electrical phenomena were
totally incompatible with the mathematical description.  My verbal and
visual explanations were WRONG.  But as long as I stuck with engineering,
used design equations and CAD software, I was fine, but if I tried to use
any non-math concepts to tell others what I knew, I would be spreading

Visible Electricity

The device depicted here was my best shot at a solution.  It's an
interactive realtime simulation of the behavior of charges within a wire.  
Rather than explaining anything, instead it just makes electric charges
visible so museum visitors can play with them.  The motion of its pattern
of lights follows the actual motion of charges in the wires, and when the
device is connected into a real functional electrical circuit, it provides
a window on the true nature of "electricity."  If all the individual
conductors in a complex circuit are replaced by a number of these devices,
the operation of a complicated electrical device can be directly observed
and intuitively understood.

"Visual Electricity" is simply an amp-meter having a "chaselight" circuit
as a readout device.  A Chaselight is identical to a movie marquee
lightbulb border, with a pattern of on and off lights which advance along
and simulate motion.  The input terminals of my device are physically
placed near the ends of a row of LEDs, and electric current passing
between the terminals is measured and used to drive some logic circuitry.  
The current being measured does not drive the LEDs directly of course.  
The logic circuitry lights every fourth LED, and by advancing this pattern
of LEDs, a row of glowing "electrons" can be made to flow along.  The
direction and speed of chaselight action is proportional to the direction
and amperage of electric current inside the wire.  To the observer this
device behaves as a wire which contains large, visible, movable electrons.



The input to the device is a standard current-meter circuit: a low-value
sampling resistor connected to an opamp.  The voltage from this first
stage is adjusted in amplitude and zero-level by the next stage.  The
signal is then applied to a precision rectifier circuit which outputs a
fullwave signal and a polarity signal.  The polarity signal is
level-shifted and is used to control the fwd/rev direction of a shift
register.  The fullwave signal is applied to a Voltage-to-Frequency (V/F)
converter whose output is used to drive the clock of the shift register. 
The four outputs of the shift register are buffered and applied to four
interleaved strings of LEDs.


When a small positive current is applied via the input terminals, the
opamps drive the V/F converter at a pulse frequency proportional to the
amperage, the shift register begins advancing, and the pattern of lights
starts moving slowly along the row of LEDs.  If the current is doubled,
the frequency doubles and the pattern "flows" twice as fast.  If the
direction of current is reversed, the polarity signal changes state, the
shift register starts decrementing, and the LED pattern flows in the
opposite direction.  If a very slow (1 Hz) sine wave is applied to the
input, the LED pattern will swerve back and forth, just like the charges
actually do within the wire. 

The above device is a wire with visible electrons.  It also is an electric
current microscope, since it greatly amplifies the speed of electron
motion which occurs in wires.  Contrary to popular belief, electric
current is a very slow flow of electric charge.  At normal densities of
electric current, electrons move at speeds on the order of inches per
hour, like the minute hand on a clock.  Even if individual electrons could
be seen by human eyes, their flowing motion would be invisible because
it's just too slow.  Visual Electricity depicts this motion several
thousand times faster than it actually is. 

The whole circuit could be made cheap and simple through use of a single
chip microcomputer having a A/D input, such as PIC16C7x.  Right now the
price is very high.  If the sales volume went up, I could change over to
SMT packages and automatic production; maybe get the production price down
below $10 each, so each student could have one, as opposed to a single
expensive device for one classroom as things stand now.

The next advancement is to make the LED pattern stop jumping.  Electrons
in wires don't jump (they behave more like water molecules; moving along
smoothly but also jittering around with thermal/quantum vibrations.)  
Also, one common misconception involves the (wrong) idea that electrons
jump from atom to atom.  So I need to get rid of the jumping.  Add a PWM
cross-fader algorithm, so the light pattern moves smoothly from LED to LED
without jumping.

A further advancement is to add a "visible voltage" function.  This could
be done through use of red/green LEDs, by measuring the voltage on the
input terminals with respect to a third "ground" terminal connected to
conductive rubber legs, then driving the R/G LEDs with a pulse-width
modulated signal.  If Visual Electricity as a "wire" is then connected to
a positive voltage source, the display turns red.  If negative, it turns
green.  If disconnected, it could either maintain an "electrostatic
charge" like a real wire and remain at its last color, or it could turn
yellow (50-50 red/green).  The voltage and current on a single wire could
then be displayed simultaneously with "electrons" which change color, flow
along, or both.   A square wave voltage signal would be a red/green 
flashing.  A sine wave signal would change smoothly: red/yellow/green and 
back again.

- Bill Beaty, 6/28/95

Brief history:

1987  Brainstorm!   While on a trip to Exploratorium in California, I 
      realize that their displays with current meters could be replaced 
      with schematics with "chaselight" LEDs, with current probes to move
      the LEDs proportional to the actual current in the wires.

1988  I design an etch PCBs as "chaselight" current meters, and build a
      temporary exhibit: a motor-generator pair with green LED chaselights
      to indicate the electron flow as one or the other crank is turned.
      I want to build a microprocessor version but don't know assembler
      for the MC68705.  I also want to build a version with red/green
      LEDs controlled by conductor voltage (with some sort of reference,
      perhaps use metal magnets on the bottom and stick them to a metal
      board mounted on the wall.)

1989  Chaselight meters are built into two exhibits in the VandeGraaff
      Hall (this just after I moved to Seattle.) 

1990  In Dinamation Inc., I demonstrate the device to the company pres.
      I imagine building a hall-effect current sensor or clamp-on probe, 
      so the LEDs will flow in the correct direction regardless of which 
      way the row is oriented.

1992  I etch a long thin two-PCB version using bar-graph LED modules and
      rechargable batteries.  TEKNOS Inc. sells a set of ten "chaselight
      current meters" to LLNL (or was it LANL?), delivered Jan/93.

1993  Arbor Scientific produces a less expensive version in a plastic tube 
      using discrete LEDs and 9V battery, advertised in their '94 catalog.

1994  I learn assembler for MC68705 and microchip PIC, but *still* haven't
      got off my butt to build a one-chip version or a red/green version.

1995  I start a webpage; upload my schematics and PCBs so anyone can
      build the device.

?1996? Arbor Scientific starts a website store, so finally teachers on 
      internet can buy my invention.

2004  I talk with D. Durlach, who uses the electric current chaselight 
      idea in an exhibit for New York Hall of Science.

2005  Completely independantly, Kamata and Hara of Tokyo Gakugei University 
      invent a much improved color-changing version with voltage/current 
      sensing, based on a PIC microprocessor.  See:
Created and maintained by Bill Beaty. Mail me at: .