The Final Assembly: Ready for Performance

As the mechanics had been fully tested I proceed to finish the project with the final assembly.

As a start I painted the MDF parts of the built white to match the over all colour scheme. I then 3D printed a series of "shoulder shelves" that would serve a double purpose of acting as a removable hatchway into the animatronic but also as a plot device.

Using small neodymium magnets in both the shelves and the shoulder struts I created a magnetic connection.

Below is another shelf that acted as a book holder for the animatronic. There is also a model that holds a small tea cup.

The book holder uses 100mm x 20mm x 3mm clear acrylic strips attached with zip ties to hold the book in place with pressure on the pages. This allows the book to be attached for a performance with little to no damage to the book.

To attach the battery securely I used a laser cut acrylic mesh that had perforations along tabs. Using a heat gun I then bent each tab whilst clamped in a vice.

The remaining tabs where lined up with holes on the black platform, drilled and bolted into place.
As the battery holder was hand bent I was able to provide a snug fit for the batteries and the extra cabling.

After my previous tests I knew that the weight distribution of the head was an issue for the motors. The original design was particularly top heavy and to counter that I moved the battery pack below the axle and the head itself closer to the axle. This provides a far more consistent speed or rotation for the head.

Below is the progression of assembly for the most important joint in the animatronic. The large bearing provides stability to the connection but the small mounting hub in the centre is attached tot he motor on the bod and bolted to the upper deck of the platform. This mounting hub is secured with two grub screws and Loctite Boltlock.

Because every bolt in the assembly is fitted with a Nyloc nut these grub screws are the only fitting that needs regular tightening after each performance. The stress from the 360 degree rotation does loosen the grub screws over time.

The sides needed to be easily removable and so are twist tied on.

 

Decals were applied as part of the finishing touches. A warning sign to the back panel and a model number for the motor.

With everything zip tied or Velcroed in it was time to close up the animatronic. 

Small screws were used at the four corners of each panel for easy access to the inner assembly.

The animatronic finished and ready for performance. 

The Rewire and Rebuild

As the week moved on I ordered a new bearing to stabilize the construction. As I waiting I finished designing the lens of the face of the animatronic. Below is the current revision with an RGB LED inserted into a clear 3D printed lens used to diffuse the light. 

I then redesigned the MDF pieces of the body to fit the larger bearing. The axle mounts were redesigned to not have a cutout in the top half and for the upper motor to be closer the the axle to reduce the length of the timing belt.

I also took the opportunity to start producing the finishing aesthetic pieces of the character. the below image shows the "McCloud Security" body plate. This being the manufacturer of the character.

Below we have another aesthetic piece, the sides of the character. Keeping with vintage style of the logo above I opted to involve a simple Art Deco design in black to break up the plain and solid white of the body plates. The MDF will be sealed and be painted in a mat white.

I then moved on to finalizing and rewiring the whole unit's circuitry. As I am using Arduino units I wanted to optimize their construction for better durability during movement.

Below is one of the smaller units. For all the motor controllers I have added stackable header pins and desoldered the drivers in an effort to compartmentalize the units. If there is an fault with the drivers, batteries or microcontrollers They can be swapped out for a swift recovery.

Below is the rewired base circuitry.

The base circuitry with stick-able Velcro adhering them to the base platform itself. 

The batteries ready and set.

The final construction of the unit minus the decoration.  Note the added cross bars on the body to add vital support. The large bearing was poorly manufactured and bent, therefor I have returned to the original bearing. Though with the reconstruction using nylock nuts, crass bars and properly fitted bearings the unit is far stabler than before, demonstrated in the video below. 

My next step is to tweak the controls for more precise movement and finish the decoration of the animatronic. Home stretch.

The batteries and the face

After much deliberation I purchased a set of batteries to power the system. Two NiMH 5000mAh 6v packs for the lower platform and one NiMH 5000mAh 12V pack for the upper body.

I am very happy with these batteries. Their weight, size and duration will be perfect for the final unit.

As you can see by the video the upper body is very unstable. I have isolated this problem to the poor construction of the Lazy Susan bearing connecting the two sections. I have sourced another wider bearing to combat the problem. I believe this will eradicate the instability.

I have also moved on to a final design for the "Face" of the animatronic. In an effort to make the surveillance personality shine through I am designing a futuristic camera for the "Face".

I will be using the lens section of an open source 3D model details below.(http://www.thingiverse.com/thing:1805762)

With an RGB LED and a diffused acrylic dome I will be able to create an effective eye for the animatronic.

The above digram shows the construction.

The above image is of the housing of the electronics and the adapter of the lens to the "head" of the animatronic.

Above shows the RGB LED controlled by and Arduino Mini Pro and with wireless connectivity over an Xbee module. 

I will 3D print the parts and stabalize the unit then work on the final decoration.

Body second iteration

After the success of the first iteration of the mechanical design I have now moved on to scaling the animatronic up to it's intended size.

During this second prototype I changed the type of motor from steppers to worm screw geared DC motors for the body section. These may not be as accurate but they have a far greater holding torque and consume less current. Because of this new addition I redesigned both the axle base and the axle mount show above and below.

I attempted to increase the size of the head though after stress testing the weight I knew I had to return to its original diameter of 20 cm.

In the photo above you can see a four inch "Lazy Susan" bearing sandwiched between the two halves of the body and platform. This bearing givs the two halves stability and a smooth turning action. Though due to the bearing's construction I will have to replace it with a sturdier model.

Above is the large scale platform test.

Above is the large scale platform test.


Next I will be working on the "face", correcting the instability of the body and sourcing the appropriate batteries.

Body First Iteration

To test the mechanical integrity of my new design I set about cutting the pieces, not out of cardboard but with MDF.

 Using stepper motors I was able to ascertain the structural compliance of the design and exactly what parts I would need to build a larger version.

 As you can see from the above photo The axle is made up of two bearings sandwiched between two nylock nuts with a timing pulley that holds the timing belt in place.  We have four angle brackets to keep it secure and downward facing stepper has a mounting hub attached to it to spin the body horizontally.

Above we have the fully assembled small scale version. This version was a success, the platform moved with ease and the body did not sway. I shall now move on to scaling the design up in size to test the limits of its stability.

Mobile platform

Based on my findings from my Physical Computing project, the "omnimac", I decided to use pre-fabricated mecanum wheels.

Setting up the motor drivers. In this case I am using the TB6612FNG.

Using a power supply to test the motor capabilities.

New design prototyping

As the design was decided I set about 3D modeling the different components of the animatronic. Using Fusion 360 I sketched out a rough version for each of the components so that they can be fabricated. 

The axle mount with mounting holes for the motor and a press fit hole for a bearing at the top.

This base of the body that has cut outs for the axle mounts and a central mount for a motor.

The rough design of the "cornea" of the eye that will house the lenses.

The head of the animatronic. Intentionally simplistic so that weights can be attached for testing.

To get these models in the real world I took advantage of the university's Epilog M2 40 Laser Cutter. I was able to cut everything out of cardboard and construct a model.