TML 2020 #02: Exploring Eye Mechanism

by fingerplayers

This journal is an entry in The Maker’s Lab journal series. Click here to read the previous entries.

Relearning & Refining

For the month of July, I focused on refining the eye mechanism and finding materials that could help with the rotation of the eye. One of the things I had to relearn was the process of ideation. When designing for corporate projects, having ample time to ideate and sketch is a privilege due to the usual time constraints of a project. Most concepts had to be quickly churned out in short amounts of time. Thus, many designers dive straight into execution upon receiving the brief. The Maker’s Lab is a process-oriented programme and it was a timely reminder for me to relearn this aspect of design thinking.

Prior to experimenting with the joints of the eye mechanism, a quick sketch was done to find out how the position of the servo motors and joints could be placed.

Figure 1: Initial Sketch of eye mechanism

Factors to Consider

1. Weight – The weight of the materials used would largely impact the puppeteering experience. If the mechanisms used are already heavy, more strength and stamina would be required from the puppeteer.

2. Material Availability – Can the material used be easily replaced when damaged? Does it require maintenance?

3. Cost of Materials

These are three factors that influence the design/making decisions during this process.


For my eye mechanism, I decided to use a ping pong ball for the eyeball as it is already a good working size and the soft plastic allows me to cut and drill into it easily. This allows me to prototype efficiently.

Joints Between Eyeball & Rod

I tried a few ways to build the joint between the rod and the eyeball starting with a spring and a rod (Figure 1). I thought the spring could act as a holder between the rod and the ping pong ball but it was unstable.

Figure 2: Pegs with a round tip (purchased from Art Friend) and ping pong ball
Figure 3: Bent wire holder attached to the eyeball with hot glue
Video 1: Eyeball with Spring (not sitting very well in the rod tip)

The glue used to secure the “spring” to the ping pong ball added friction to the eyeball and caused more problems to the rotation of the eyeball. The “spring” also had problems securing to the ping pong ball due to its small surface area and eventually fell off. It is also not a feasible idea as it was time consuming to bend the wires to fit the round tip of the pegs (Video 1).

While taking a break from making, I found a joint used in Daiso’s Petite Blocks (Figure 4) that could be used for the eyeball.

Figure 4: Petite block ball joint (purchased from Daiso)
Figure 5: Petite block ball joint attached to eyeball with hot glue

I used the joint on the Petite Blocks to form the ball joint between the ping pong ball and the rod. Although it worked well, a question that came up was the availability of the petite blocks in the future – thus, this solution was only temporary.

Figure 6: Eyeball attached to the wooden peg using the petite block ball joint

As suggested by Daniel, I also tried using a ball bearing to act as the ball joint (Figure 7).

Figure 7: Ball bearing purchased from Shinko Bearings & Beltings Pte Ltd
Figure 8: Ball bearing fitted into the ping pong ball

Although the ball joint was smooth and fits well into the ping pong ball, a problem that came with it was its maintenance and the weight of the ball bearing. It was heavy and frequent oiling was needed to ensure the smoothness of the mechanism. The cost of one ball bearing ($12+) was also hefty in comparison to the petite block ($2).

Figure 9: Up-down joint – the metal rod attached to the eyeball rotates freely up and down inside of the hollow aluminium rod

After a brainstorming session with Daniel and looking around the workshop, I found this way of creating the joint to be most feasible (Figure 9). It is lightweight and the materials can be found locally. However, the hollow aluminium rod on the inside of the ping pong ball needs to be longer to reduce the noise generated when the eyeball is rotating. This is to reduce the left and right sliding of the eyeball as it is rotating up and down. The thick rod attached to the ping pong ball also causes the eyeball to bend out of shape.  It will have to be reduced in size. Also, because the hollow aluminium rod is quite thick, a larger hole in the wooden dowel had to be drilled. This weakens the dowel.

From this, I further refined the eyeball and came up with the following (Figure 10).

Figure 10: Up-down joint using a thinner hollow aluminium rod (purchased from Art Friend)

The rods in the eyeball have been reduced in size so as to keep the eye mechanism as light as possible.

Building the Base Structure

Figure 11: Servo motors fitted into a wooden base structure

Figure 11 shows the base plate of the eye mechanism. The end result of this also differs very much from the sketch that I had made earlier. In the sketch, I envisioned the base plate to be at the bottom of the eyeball. While making, I feel that the eyeball would work a lot better if the plate was positioned in the middle of the eyeball as I will have more space to house the motors and wires.

Figure 12: Eyeball attached to the base structure

Figure 12 shows how the ping pong ball would sit in the structure. I found that building the pivot point closer to the back of the eyeball allows more control over the left-right, up-down rotation of the eyeball as compared to having the pivot point built inside of the eyeball (see Figure 6).

In the last photo, I found that I had made a mistake in building the horizontal plank too closely to the ping pong ball. I had to cut some holes to allow the eyeball to turn better (refer to Figure 12). You can see from the video above that the eyeball does not turn to the right as easily as to its left.

Different servo motors also have different sizes. I was not able to build the right eyeball in time for this reflection as the servo motor I was using was different in size and more alterations need to be made to the size of the base plate.

Positioning of Servo Motors

From the previous making of the eyeball, I found that the position of the servo motors played an important role in ensuring smooth rotation. This would be one area that I will be working on in the coming days.

Currently, the servo motor controlling the left-right movement of the eyeball is placed at the back of the plate and the servo motor controlling the up-down movement is placed in front of it. With my current set-up, the up-down servo motor blocks the path of the left-right servo motor, hence I had to bend the wires. As the servo motors pulled the eyeball, the wire will slightly flex and unflex due to the bends which causes the movement of the eyeball to be quite inaccurate. I will be redesigning this part to allow the wires to run directly to the motor without bending.

Stores Explored

  • Art Friend – Basswood Sheet (4105), 3/32″ Aluminium Hollow Rod and 3/8″ Wooden Dowels
  • Shinko Bearings & Beltings Pte Ltd – Ball Bearings
  • MakerSupplies Singapore (on Lazada) – SG90 servo motor

I found that the servo motors made by different suppliers to differ in terms of the smoothness of the rotation. The servo motors that I had bought from a local supplier produced louder sounds and jerks a lot more in comparison to the motors that came with Arduino Starter Kit.


Getting exposed to puppet masters and makers makes me question, “What is my influence?”

I was first introduced to the term, “animatronics”, through the Jim Henson’s Creature Shop Challenge. In 2014, the Jim Henson’s Creature Shop organised a reality competition show where contestants were given design briefs to be completed within the time limit. Each episode featured a new design challenge and contestants had to quickly conceptualise creatures using their vast knowledge about different materials and techniques in making. I found episode 3 of the series (titled, “Assembly Inspired”) to be most exciting as the contestants had to search through a junkyard to find used junk and conceptualise a creature based on the found objects.

In 2017, I watched a production Afternoon of a Foehn by Compagnie Non Nova that completely changed my understanding of object puppetry. Through the control of fans placed in a circle, pieces of plastic bags resembling people begin to dance in the performance space. There is a presence of the live body but technology now becomes the puppeteer of the object, while the performer plays the role of the indirect puppeteer by controlling the fan speed. Live music is also played based on the plastic bags’ movements. Not all the performances are equal, as there are aspects of the performance that cannot be controlled.

I recently found a blog post about Richard Teschner and how he adapted Javanese Rod Puppets in Western puppet theatre. His puppets were contemporary but at the same time I could see the influence of the Javanese Rod Puppets.

I feel that there is so much to explore and learn in making puppets. Through refining the simple joints of this eye mechanism, I learnt about how different materials react with each other and how materials used in engineering could possibly be used as a joint. I think I am starting to understand why puppet masters are willing to spend their whole lives refining their craft.

This article is a monthly reflection by Sim Xin Feng, the maker of our inaugural The Maker’s Lab as part of an ongoing 9-month experimental laboratory. The Maker’s Lab is curated and managed by Daniel Sim, a core team member of TFP. The ideas and reflections within the article are drawn from Xin Feng’s observations and discoveries as a maker, designer and researcher. Instead of being taken as conclusive, we hope that they serve to be a starting point for thought-provoking conversations and perhaps even debates. We would love to hear from you and can be reached at

If you wish to follow Xin Feng on her journey with The Maker’s Lab, click on the ‘Subscribe’ link below now!