attaching the skirt. I don’t know how similar the fans are,
however, I wouldn’t be surprised if there is a small variation
in the thrust between them. That small variation in thrust in
a nearly frictionless HoverBot can certainly produce a
rotation. Until I can find a way to test the thrust of the
fans, there’s no way to know how closely the fans match
50 SERVO 11.2010
each other. Since I don’t believe it’s realistic to use a
rheostat to adjust the spin of a ducted fan, asymmetry
produced by mismatched fans may be something we have
to work around. I suspect a small source of the HoverBot’s
spin is due to the counter-torque of the spinning lift fan
blade. The ducted fan blade spins fast, but has little mass.
The greater mass comes from the
spinning armature inside the fans
motor, but its radius is even closer to
the center of rotation. Combined,
the torque from the spinning ducted
fan should be small compared to the
rest of the HoverBot. However, the
HoverBot is floating on a nearly
frictionless cushion of air and even a
tiny torque can add up over time.
This potential (weaker?) source of
anomalous spin is another one we’ll
have to work around.
Until I get a better handle on
this anomalous spin, I’ll have my
HoverBot correct for it by dead
reckoning. At times, the HoverBot
will spin in the opposite direction for
a moment. Perhaps later, I can try
adding a small accelerometer or gyro
to the HoverBot. An accelerometer
lets the HoverBot detect the spin
and as the rotation builds up, the
HoverBot can produce a counter-spin
to put itself back on track. Some
time next year, I’ll have the HoverBot
position itself based on a model star
tracker. Next, will come the larger,
HoverBot 2.0. I’ll keep everyone
updated about my experiments via
my tweets, my blog, and my
You Tube Channel (@NearSys,
nearsys.blogspot.com, and
www.youtube.com/nearsys).
I hope you’ve enjoyed my
HoverBot series. Again, feel free to
assemble your own and keep me
informed about your successes.
Here’s to a nearly frictionless
future without wheels. SV
By the time this article shows up
in print, I should be selling HoverBot
kits for those who don’t want to make
their own printed circuit boards. You’ll
find information on my webpage at
NearSys.com.