■ FIGURE 5.
balls like a pitching machine (Figure 4).
We had hoped to make the system
auto-loading with some type of hopper
to hold the ammo. We sketched up
and experimented with numerous
different designs with the first requirement being throwing distance (we
wanted to be able to hit the middle, if
not the end of the arena!) and the
second biggest concern being reliability. We had to consider that the end
product would be suspended 15 to 20
feet in the air, so it would be difficult (if
not down right dangerous) to reach it
for diagnostics, adjustment, or repair.
In the end, we settled on a simple
design based on one of many pneumatic spud gun plans we found on the
Internet. Our final gun consisted of a
length of 4” ID PVC pipe configured
as an air reservoir, an inexpensive 1”
sprinkler valve as a trigger, and a piece
of 1.5” PVC pipe as a barrel (Figure 5).
Due to concerns with reliability
and the short time we had to build
things, we settled on a “muzzle
loaded” design using strips of a plastic
drinking cup to hold the ping pong
ball in place after loading. Because the
gun turrets would rotate, we were
able to point them down for loading,
and then point them almost straight
up in order to allow the balls to settle
into position for firing (Figure 6).
We wanted the balls to be memorable, so we found a shop that could
print our logo onto the ball (Figure 7).
We ordered four gross of logo-emblazoned ping pong balls. They
were scheduled to arrive just two days
before the Faire! In the meantime, we
did our testing with some ping pong
balls purchased at local sports shops.
IN THE ROAD
Once we had four pneumatic guns
built and tested, it was time to test the
■ FIGURE 7.
■ FIGURE 6. The author
loading the guns.
shoulder pan motion. The
original design called for the 2
x 12 cross plank to rotate to
allow the head and shoulders
to pan the guns across the
arena. This would allow us to
shoot ping pong balls both
into the arena and also up into
the stands. The pan cross
plank was designed with outrigger wheels that would carry
the weight of the guns so the
motor would only have to provide the rotational torque to
move the shoulders (Figure 8).
When we got all four guns hung on
the turrets, and applied power to the
12V pan motor using a 12V gel cell, the
motor made a soft click and did not
move the shoulders at all! We removed
the guns and the shoulders would pan
with ease. Turns out that with the guns
mounted, the assembly was just too
heavy for the motor to get moving.
We didn’t have time to redesign
the shoulders or replace the motor, so I
decided to see if a bit more “juice”
might make the motor turn. I grabbed a
second gel cell and put it in series to get
24V. Again, the motor went click but a
bit louder this time. I picked up another
gel cell, added it for 36V, and the motor
started to barely move, but after a short
slow turn, it got stuck. I tried reversing it,
but no luck. It stayed put.
Having another gel cell handy, I
added it in series as well, and hit that
motor with 48V! The shoulders
creaked and lurched into action and
the shoulders began to pan! Success!
But then, a very loud
crackling sound came
from the motor’s gearbox and the shoulders
ground to a halt. Turns
out 48V is enough to get
the motor turning, but
the gear train wasn’t up
to the task and disintegrated into metal shards!
Lucky for us, it hadn’t
frozen up so we were
able to manually position the shoulders point-
■ FIGURE 8. Shoulder
rotation cross plank.