Super-sized Animatronic Ball Ornaments

• Supersized Animatronic Christmas Ball Ornament - Click HERE - from Tory Street Lights

More below...

What is it? These are lip-syncing ornaments. The idea for how to put them together is valid for any similar item you want to make.

Where did this idea come from? Brian "Lightman" Lincoln did this:

Singing Christmas Tree

Terry Sutton (Tory Street Lights aka T.D.Sutton) wanted to supersize the idea for his front porch.

Bill of Materials

Big ball (25 inches) PVA/paper Cheesecloth

Planting container

Servos of various sizes 4 servos for the eyes. 2 servos for the eyebrows 1 servo for the mouth. 1 motor and gears for the ornament to turn away from crowd when not performing

brass rod foam board

Renard board Optocouplers Drivers

1000 watt slide projector with template

Chris Maloney's Audacity

Many cans of paint.

Some further comments about the "Supersized Christmas Balls" to clarify some points about the project may be in order.

Setting up the eye mechanism required an inordinate amount of time. Placing the eyeball mounting rod with the universal joint exactly in the geometric center of the ball required a center mark to be made on the outside of the ball with a black marker pen while the ball half was spun up in my metal lathe to locate the eye center. A 1/8 inch brass rod, with a washer soldered on the end, roughly shaped to a convex curve to match the internal surface of the ball, was then "5 minute Epoxied" into the ball. The rod was located over the ball's center mark, while being held at the outside end by the centering plate. The brass rod was then cut to length once the Epoxy had cured.

This was centering was most important, as the eyeball had to be concentric with the eyelids to allow them to open and close with the eyeballs in any position, but not touch one-another. The clearances involved in this project were very fine and created problems when I glued the photograph of the eye in position. Many hours of adjustment was required to get everything operating and in the correct place.

The pictures of real eyes, obtained from the internet, were adjusted to the required size and then printed as photographs in a local photo lab. They were cut to shape and then had three radial cuts made in to the black iris. To make them to conform to the eye's spherical form they were soaked in Methylated spirits, then pressed between waste disk sections removed from the rear of the eyeball when it was opened up for mounting and control.

The eye mechanism in the photographs is a spare set, and as such has micro servos to operate the eyelids, the display ornaments have digital servos in that position that are three or four times faster. These were necessary because human eyelids "Blink" very quickly, to replicate this, super fast servos were required.

Each "Ball" is supported on a 3/8 inch threaded rod running through ball races mounted under the two marine ply plates that are fiber-glassed inside the ball. The weight is taken on nuts threaded on the rod with two nuts locked together at the bottom, and another pair under, and on top of, the top bearing. At the bottom, inside the ball, an arm is lock-nutted to the rod, which is connected to an arm on a jumbo servo to allow slow controlled rotation of the ball, over about 60 degrees, so that they can "look" at each other.

The top end of the rod has a purpose made threaded female adaptor screwed on and lock-nutted to it that converts from 3/8 inch to 6.0 mm. The 6.0 mm socket end has a 6.00 mm threaded "Rawl Anchor" hook that is made for concrete "Rawl Anchor" attachments. These were used because I did not trust my welding should I welded the two rods together.

The hook is simply hooked over a bolt running through two plates that are bolted through another eyebolt in the same plates, this eyebolt is attached to the 180 degree rotating mechanism in a similar way to the ball's hook. This setup is visible in the picture under the rotating mechanism. Once the ball is hooked onto its cross bolt, the bolts are tightened up very tight, because all the rotating forces are transferred through the fixture. The hook makes for "easy" mounting and demounting of the Ball.

The rotating mechanism may require some further explanation, it is based on a 24 volt gearbox out of a soft drink dispenser, or similar. It has another three times reduction to it's output shaft using some discarded heavy duty gears from my work at the time. The 3/8 inch threaded output shaft is once more supported on ball races top and bottom, lock nutted as required. The large output gear has bolts on it to operate the Micro switch end-stops. Electrically the mechanism has 12 volts DC supplied permanently, and is controlled by a signal from one of the display Renard controller channels. When the channel is "Off" the ball is "Parked", when the channel is "On" the ball rotates 180 degrees and faces the street, then returns to parked again when the channel is switched "Off".

The Rotating mechanism is mounted on a discarded satellite dish mounting arm, locked up good and tight at right angles. The arm is attached to a wooden plate that is "F" cramped to a verandah rafter when on display. This plate has hooks to slip under the iron roof for added support. The result being the balls are kept out of the weather for protection, and out of sight when not "Performing".

Another item that may require further explanation is the Renard controller. This is housed in a re-purposed Electric Drill case, with the electronics separated inside plastic sandwich boxes for safety. The controller is a stripboard 32 channel Renard controller, using regular PIC16F688 chips, loaded with Chris Maloney's Ren Servo software, running at 11500 baud. There are 32 Channels because it was initially intended to have four "Ornaments". The controller is capable of running from the ESP8266 Pixelstick in the lower lefthand corner of the picture, but I prefer it to be cabled from the end of a 448 Channel controller chain.

The Opto-coupler and buffer box also has 32 channels available, with the buffer/driver chips hidden under the linking cables on the Right hand side of the picture. It was fortuitous that the buffer chips have a tri-state gating pin. I was able to take advantage of this to disable the data to the Ornaments when they were not "Performing". This was necessary because the Renard chain was outputting pulse width data outside the range of the servos when not "performing", the effect was that all the servos went to random positions way outside their expected positions. The gating signal was gained by using the projector control signal from a display Renard controller channel, applied through an opto coupler. The gating to "On" occurs 100 milliseconds after the start of the sequence, and "Off" occurs 60 milliseconds before the end of the sequence,