the smooth-surfaced gumballs I used for inspiration — ping
pong balls are created with an abrasive surface so they can
get traction on the ping pong paddle and table. Unfortunately,
this extra friction caused the balls to stick together in the
jug, jamming the stirring rods. The only way to un-jam the
system was to fill the jug to less than 10% of its capacity
(Figure 5) which really defeated the purpose of the jug
itself. This was about the time we noticed a second
problem with the design (insert deep sigh here).
It seems that rotating the ping pong balls inside the plastic
jug caused them to build up a static electric charge! This
would cause the ping pong balls to cling tenaciously to the
sides of the jug making the stacking/jamming problem worse
and, of course, stopping them from dropping down through
the indexer to the print pedestal (insert second deep sigh here).
DOES SIZE REALLY MATTER?
As we were pondering what to do about the sticking,
stacking, and static problems, our bulk-order of blank white
ping pong balls arrived. We were excited by this good news
and dumped a bunch of these new balls right into the jug
in the hopes they would be smoother and might alleviate
the jamming. No such luck. These new balls jammed the
exact same way. We weren't getting off that easy! We then
dumped out the jug and put just a handful of these new balls
back in for testing. We activated the printer and the first ball
dropped right onto the pedestal for printing as expected
(Figure 6). When the printing was done, the "indexer"
(Figure 7) started to cycle, then jammed without releasing
the second ball. We removed the jammed ball and tried
again. Same thing, the indexer was stuck and would not
release the second ball. A lot of head scratching later, we
discovered the root cause of the indexer jams.
It turns out that there are two "types" of ping pong balls.
The first type is "professional" 40 mm balls and the second
type are "hobby" ping pong balls that are 38 mm. We had
started buying the test balls from local sporting goods stores
and they were of the 40 mm variety. Subsequently, when
Rick had machined the parts, he had used 40 mm as his
reference size. When the indexer would try to cycle, it would
bump into the bottom edge of the next ball in line to be
printed and stall without completing the cycle. Rick took the
printer back to his shop and reworked the indexer push-bars to accommodate the 38 mm balls. Once this was
done, the bulk purchased balls worked fine but we still
had the existing multiple water jug issues to contend with.
CHANGE TO CHAIN DRIVE
When Rick had the Ping Pong Printer at his shop to
re-engineer it for the 38 mm balls, he also took time to
look at the plastic jug and try to come up with a new way
to serve ping pong balls to the printer. After a few different
designs were tossed around, he settled on an ingenious
chain drive belt-scoop feed system which he then fabricated
from aluminum, polycarbonate plastic, and steel (Figure 8).
This new system would ferry the balls up to the top of the
Ping Pong Printer, then let them roll down a ramp and drop
into the indexer. This new design solved both the static
problem as well as the stacking/sticking problem, but at the
cost of a bit more software complexity. I had to figure out
how to sense when to run the motor and — more importantly
— when to stop it. To deal with this, I added an IR beam-break detector to determine when a ball was rolling down
the ramp. A bit of coding had the software turning on the
belt drive motor, waiting for a ball to occlude the sensor,
then switching the motor off. With these modifications, the
new and improved Ping Pong Printer was all ready to go.
Since then, we've used the Ping Pong Printer not just
to make "ammo" for the Ponginator, but to create souvenirs
for many events such as movie premieres (Figure 9), South
By South West Austin, and multiple Dorkbot events. Around
this time, the video of the Ping Pong Printer in action (with
its original water jug) was picked up by a number of tech-blogs
and ended up with over 30,000 views (and counting)!
NOTHING EVER GOES AS PLANNED
The point I'm trying to make with all this detail on what
went wrong and how it was overcome is that we have yet
to encounter a problem that could not be bested with the
proper amount of perseverance. As the above examples
show, given the right motivation and the right people, you
can overcome just about any engineering challenge. Just
expect that Murphy is alive and well and living in your
■ FIGURE 6. A ping pong ball on the
print pedestal after printing.
■ FIGURE 7. The indexer stage that
controls the fall of balls into the printer.
■ FIGURE 8. The new chain-drive
ping pong ball hopper.