Twin Tweaks ...
IN THE BRAIN OF THE BEAST.
GRAVITY SWITCH MATERIALS.
demographic seem to deliberately interfere with
hacking by making the casing difficult to remove.
Screws are numerous and hidden. The casing
might be difficult to pull off and fit back on. And
sometimes once the casing finally does submit, the
circuit board is an inaccessible rat’s nest of wires
that leave no open places to solder in your own
Given the fact that the Roboni-i appears to be
first and foremost an electronics entertainment
system, we were afraid its hackability might be a
limited afterthought. Thankfully, we were
completely mistaken. The main body panel of the
Roboni-i is held on with six screws that are easy to
access. After removing the bot’s head, the panel is
easy to slide out from under the wheels. The panel
isn’t obnoxiously held on with wires or sensors
inextricably bound to it with glue or stubborn
connectors. Upon taking a look at the circuit
board, we were excited to begin hacking. We
were hoping to find some open pins or solder
pads, and we were not disappointed.
The Roboni-i does indeed have a multitude of
open solder pads; three clusters, in fact. One
cluster is labeled JTAG, another JC, and another
AC LED. We thought that at least one of these
pads would provide the perfect way to incorporate
our Skinner switch, but just to make sure we
PROTOTYPING THE SWITCH ON A BREADBOARD.
72 SERVO 03.2010
pulled out our trusty multimeter for a quick
diagnosis. The JC and JTAG clusters gave us some
small voltages, and the AC LED read about 3.7V.
We were hoping a look into the programming
interface might give us a few more clues about
how to implement our switch, but we were
confident that we would find a way to include our
Before soldering up our gravity switch, we
wanted to prototype it on a breadboard as a quick
proof of concept. We were confident that our
simple circuit design was sound, but we wanted to
make sure that the gravity switch was consistent
after our troubles with the straw. All we needed
for the circuit was our cylinder, ball bearing, LED,
100K Ω pull-down resistor, and some extra wire.
Anticipating a connection to the free solder pads,
we used a male PWM cable for our connecting
wire. When we first flipped over our cylinder, we
were disappointed to see that the LED failed to
After inspecting the switch, we determined
that one wire was catching the ball bearing before
it could hit the second wire. After a little finesse
with some needle-nose pliers, the switch was
working like a charm. Each time we flipped it over,
the LED lit up with a bright red and we were
ready to solder it all up. Ideally, we would have
liked to use some heat shrink to make everything
look nice, but we were fresh out and electrical
tape was a fine (albeit not as aesthetically
pleasing) substitute. It looked like the Roboni-i had
performed its last unpunished flip. All that was left
to do was to hook it into the robot and program a
Like a Cat in a Puzzle Box
Our multimeter investigation was inconclusive,
so we hoped that looking into the programming
for the Roboni-i would give us some clues about
how to implement our new switch.
The Roboni-i comes with a comprehensive
software package known as the Command Center