Following on from Peter the Pumpkin - pt2


Progressing the Eyes

I've progressed over the week with the eyes Mk5 e.g. paper mache. However it did take several nights of adding additional layers before I considered them to be strong enough to break out of the silicone egg cup moulds....and then I added some more paper mache for good measure. I also coated the outside in thinned down wood filler, allowed to dry, sanded, repeated, primed, sanded, primed and then added a white gloss finish. I was aiming for a mirror smooth finish. Knowing that all the geometry in the pumpkin are different I realised I would end up adjusting angles and rotations to get the final effect I wanted. Gluing the aluminium frame bars directly into the paper mache eyes would give no extra adjustment. Therefore I glued in some softwood strips and have mounted the frame cross bars to that using a couple of 1/4" wood screws.


I printed off some iris designs are used PVA glue to fix them to the eyeballs. I'm wondering if I should have done this at the end when I can adjust them to the required positions. When assembling I also added a small disk of polypropylene from a milk carton lid into the servo joint. The rod and coupling occasionally fouled and this was to help prevent that and the subsequent straining of the servo motor. Below is a picture galley:


{gallery} Progressing Eyes

Wood filler on Eyes

More Priming Coats

Wooden Internal Mounting

Some white gloss finish
Back of eyes - servo frame fixed to wooden internal frame
Final Eyes - adjusted and ready for a test :-)

I also spotted this today. It is the top from a clothes washing detergent bottle and could have been used directly for the eyes (isn't that always the way).

Detergent Bottle - possible future eyes?

[post edited later in the day to add this small test video - because I'm really pleased with it ]


if (thisMode==eyesA){

int posLeft=90;
int posRight=90;






Movement Sensor

I ordered a new PIR movement sensor but went for another slightly different, but popular, style. This requires a 5v supply and the output is a digital high when movement is detected. The unit can also repeat detections whereas the first PIR unit had a built in 8s timer. I just need to find a way to mount it on the animatronic's stand as I want to be able to point this sensor to the direction of 'trick or treat' arrivers.

PIR Movement Detector


I've never used an Arduino for driving WS2812/Neopixels and have used my own PIC microcontroller with Configurable Logic Cells (CLC) like in this project: PIC Microchip LED Effects: Firefly Part 1

For this ShareTheScare project I made the lights for the eyes (now sidelit) and loaded up the Adafruit Neopixel library. This worked a treat and was very easy to use - well worth a look for anyone wanting to make something for the coming festive season or other national/cultural events. At work there is often something happening for Diwali and I've noticed a lack of lights on the stalls. Hopefully I can help out this year with what I've found the humble Uno can do to make those stalls shine .


Head Movement

I still plan to utilise a modified version of the code I used on My First Stepper Motor  with Arduino pin changes as necessary. The head will require fixing to the stepper motor pedestal. One thing I haven't checked yet is the ability of the stepper to actually move this load. I made this stepper motor turn table whilst applying for the Molex 2.4GHz / 5GHz Antenna Kit roadtest. I have used a set of toothed pulleys to slow the speed down (and hence increase torque) and also improve accuracy. The bearing consists of some steel bearings (as popular with fidget spinners) and a wooden housing. The end of the shaft has been sharpened and rests on a steel plate recessed underneath to take the axial force. I also need a pair of microswitches to prevent any software errors making Peter rip his own head off (which might be scary but would be a one off event).



The Arduino Uno software is progressing well and I've written test code for all the main components (lighting, stepper motor, servos and PIR) and are currently finding the best way to integrate them.


To Do By 31st October - no pressure!

  • Mount eyes into the void of Peter's head - might use lumps of polystyrene to get to the approximate locations
  • Finish software
  • Mount the lights on the servo frames so they side illuminate the white eyes - will be using hot melt glue for that
  • Adjust the servos so the eyes move freely, the servo rods do not jam and the servos are not overloaded
  • Fix Peter's head to the turntable
  • Mount the PIR looking the required direction


To Do by 3rd November

  • Make a Fritzing diagram in my final blog
  • Add code to my Github page (and link in blog)
  • Take some videos of Peter in action


Final Thought of Today

A 12v airhorn would definitely add to the scare....or perhaps a bit too much so.