Mag-neato: A magnet based installation at the Children's Museum of Atlanta

Project Summary

In this project, a teammate and I work with the Children's Museum of Atlanta to create a museum exhibit. It was important to the museum that the exhibit provides value to child visitors, whether it be educational or motor and perception development. Most of the museum visitors range in age from 3 to 8 years old (and are accompanied by adults). We chose to introduce magnetic concepts, and develop motor skills through unguided play. This project was completed in April 2019.


The possibility of a magnet installation at the Children's Museum of Atlanta (CMA) came up in an early brainstorming session where it was noted that children find magnet interactions to be like magic. Magnets are something young children may have some exposure to, but a limited understanding of. We sought to create an interactive, magnet-based installation that would combine the magical feeling of magnets with scaled down applications that demonstrate usefulness. The Georgia Standards of Excellence were used as a base for finding a set of learning goals that could then be built off. There are a handful of kindergarten to second grade standards that relate to magnets, an area of science that appeared to be sparsely covered by the museum on our initial visit. Standards in both Math: referring to measuring and comparison (MGSEK.MD.1) and Science: obtaining, evaluating, and communicating information to demonstrate the effects of magnets, examples of ‘useful magnets’ and the relationships between balanced and unbalanced forces (S1P2, S4P3). With Mag-neato, the goal was to provide open and unguided free play with passive learning about magnetic forces.

Design Prototype

The plan was to design and build a toy maglev train for the CMA. The train cars would levitate on the track and demonstrate magnetic forces. The floating cars would connect with small magnets to form the train. This was the base idea, but the train set would also have environment pieces such as train "toppers" and other magnetic demonstrations.

The magnetic mechanism behind real maglev trains are a result of precise engineering and for the purposes of the children’s museum, were found to be too difficult and unnecessary. We opted for a simpler design.

We purchased several types of magnets to test how each worked with the maglev train setup. We experimented with standard flexible magnetic sheeting, ceramic magnets, and neodymium magnets (which grow in strength, in that order). We also tried various magnet shapes including circular (big and small), long strips, and bar shapes. We aimed to maximize the gap between the train and tracks to make it readily apparent that the train is floating. The neodymium magnets achieved the best height, but were too strong to work with reasonably. We feared the strong magnets would become a danger if the adhesive or enclosure was not designed correctly or wore down over time. We also learned that in order to achieve consistent train travel down the track, a continuous magnet track was needed. In our first prototype, we built the track out of round ceramic magnets which caused the train to jump rather than glide. The cars in this prototype were small wood blocks with magnets taped to the bottom.

As mentioned, magnets placed discretely (like the picture above) do not work to make something float consistently. We instead needed magnetic strips for continuous levitation. Typical standard flexible magnetic strips, however, do not work well as they are designed with multiple magnetic poles on one side to always attract. It was also discovered in early prototyping that achieving a smoothly guiding train car was difficult to due to the minute clearance required between train car and retaining walls. If the gap is too small, the train car gets stuck. If the gap is too large, the train car flips over or escapes the track. This technical requirement made curves in the track extremely difficult and infeasible for the installation. This information ruled out our desire track shapes shown below. In fact, we were forced to rethink the maglev train altogether.

Around this same time, while experimenting with different magnets, we discovered another interaction that represented the goal of opposing magnetic forces (shown below). This interaction is similar to the wooden bead mazes found in pediatrician's offices. If the magnets are kept on a closed path, it resembles a magnetic roller coaster. We decided to pivot to this magnetic interactoin because the maglev train was proving difficult.

Safety and durability were key concerns throughout the design process. These concerns drove material choices. Aluminum was chosen for the rods that act as the guiding paths for the bead maze. Aluminum was chosen due to its high durability, which is necessary for the poles to keep their shape while kids play.

We made a small version of to test the concept and get feedback from peers and experts. This prototype included two straight paths (for the bouncing interaction) and a curved path to demonstrate a push effect that can be achieved. This prototype was received well, with only minor concerns about the smoothness of curves and the durability of magnets.

The desire for co-play, and the large groups present during the museum site visits, drove the goal of allowing multiple users on the exhibit at one time. In order to accommodate for multiple users, the interaction points, the rods and the train, needed to be dispersed throughout a large enough area to accommodate multiple children without overcrowding. The magnets also need to be reachable by small children. Due to our wide target age range, 3-8 years old, some amount of the magnets need to be accessible, even by smallest of the museum attendees. We chose to place some rods low and beyond the base of the display, allowing for easy access.

We were also able to obtain magnetic strips with one pole per side, which allowed us to include a small maglev train. 

In order to increase stability of the elements on the base of the exhibit, placement slots were laser cut for the train, allowing the tracks and sidewalls to be secured within the base, rather than on top of it. Terrain provided vertical support and a larger surface area to adhere the rods to the base. The terrain also added an aesthetic element, increasing visual complexity and adding to the verticality at the base of the exhibit.

Finally, we added the magnets, which were coated in a liquid-rubber material and painted black. The rubber material was meant to reduce the chance of magnets breaking, which happened several times during prototyping.

Installation at CMA

From the beginning, our museum partners emphasized durability and stay time as key metrics for installations. Stay time is the primary metric the museum uses to measure success. Durability is important because kids are typically rough with installations and a broken installation can be costly.

We completed one day of user testing at the CMA with our final prototype. Children, both individually and in groups as large as eight, were able to play with the magnet exhibit as intended. The favored interaction of children and adults alike was to compress the repelling magnets on the rod and then release them and watch them explode away from one another. In a following iteration, changes should be made to the pole material, currently aluminum, to reduce friction and the forced magnetization of the metal, which reduces the intensity of the explosive repulsion of the magnets.

During the installation, we realized that we could improve the educational aspects of Mag-neato by coloring the magnets. Painting all positive poles blue and all negative poles red, would help to visually illustrate the poles and the resulting forces.

Kids asked several times about other magnetic interactions like attraction. The kids would squeeze the magnets together, thinking that they were attracting (very cute). Attracting magnets were intentionally avoided due to possible safety hazards with children getting fingers pinched between the magnets. However, this left a gap in the forces that were being presented. In future iterations, weak attracting magnets could be used like wands to move the magnets along the poles; the weak nature of the magnet would help reduce the likelihood of injury due to magnet attraction. Other additions, such as magnets of different strengths, could be included in order to help exemplify the diversity of magnets.

The magnets were covered in a layer of liquid tape in order to reduce impact force and the likelihood of magnets chipping from collisions with each other. Along with replacing the current ceramic magnets with more sturdy magnets, the coating should also be changed. The current liquid tape coating leaves the magnets with a tacky feel, even after fully curing. This tacky feel can feel uncomfortable to touch and causes magnets to stick together.

A handful of small adjustments could be made to the current components in order to increase usability.  The installation was successful both in terms of durability and stay-time, but also the subjective feedback received from kids. The children we observed seemed to enjoy playing with the magnets, had fun, and might have learned something.