Artificial Feather Badminton Shuttlecock

Note that details are heavily redacted due to the proprietary concepts, designs, and technical information produced during the research and development of this project.

Motivation

Badminton is one of the fastest racket sports, with an all-time record smash speed of nearly 350 MPH. Modern day badminton originated in the UK in the last 1800s but has since spread globally. Unlike other racket sports, the "ball" is not a sphere, but rather a cone with unique geometric and aerodynamic properties. The projectile is known as a "shuttlecock" or "birdie". I personally began playing in late 2021, and felt there was room for innovation in the badminton space.

There are currently two types of shuttlecocks, plastic and feather.

Feather shuttlecocks are preferred among intermediate to advanced level players given their rapid stabilization, non-ballistic flight path, and sharp sound/feel. Feather shuttlecocks are used for warm up or beginner level play due to their increased durability and cost effectiveness.

But why can't there be something durable and cost effective that still has a non-ballistic flight path, feels/sounds sharp, and rapidly stabilizing dynamics?

The objective of this project is to contribute to the advancement of the sport of badminton by developing a sustainable synthetic feather shuttlecock that unites the advantages of both genuine feather and plastic birdies. Though there are a few products on the market aiming to achieve a similar goal, the badminton community remains divided on their performance. 

How Hard Could This Be?

Below is a video of the very first attempt at solving the problem

Turns out it's a lot harder that I thought... Little did I know this was going to be the beginning of a multi-year R&D effort and one of the most challenging technical problems I've faced as an engineer.

Design Process

In my initial attempt, I tried to shortcut the engineering design process.

However, especially when working on very challenging R&D problems, it is critical to follow a much stricter design process.

Each revision resulted in new learnings, prompting additional research, requirements, and solutions to be explored. Throughout each step of the process, I continously documented all findings and thought processes using a standardized template to help accelerate the process and ensure forward progress.

Anatomy of a Shuttlecock

The anatomy of a shuttlecock is shown below, which will be a helpful reference for understanding the subsequent sections. The left image is a screenshot from my CAD model, and the image on the right is of a feather shuttlecock.

Requirements

The requirements for this project can be broken down into the five aforementioned categories. Note that these requirements are defined qualitatively instead of quantitatively, since many of the requirements are challenging to validate numerically. Additionally, given that this is organized by requirement, and most design decisions effect several requirements, the design progression is presented temporally non-sequentially.

Requirement 1: Rapidly Stabilizing

A shuttlecock experiences two phases of flight: unstable and stable. Stable flight is defined as the portion of the flight where the shuttlecock's direction of travel is coincident with it's heading, and is the flight regime for the majority of the arc. Unstable flight is the portion directly after impact with the racket when the shuttlecock is changing direction. This is shown in the image below, with the red arrow pointing in the direction of travel.

The shuttlecock rotates about it's transverse axis to reorient itself in the direction of flight, but due to rotational inertia, overshoots the angle. It then corrects back in the other direction and repeats, with each oscillation resulting in less and less overshoot. The shuttlecock therefore acts as an underdamped oscillator. 

From the free body diagram above, it can be shown that the drag concentrated at the aerodynamic center of pressure is the corrective force helping achieve stable flight. Ideally, the settling time is minimized such that the shuttlecock enters stable flight very shortly after impact. This settling time is directly affected by the distance between the center of pressure and the center of mass, drag force, and transverse moment of inertia. The image below shows the angle between the heading and direction of travel for a feather shuttlecock with an initial linear velocity of 18.6 m/s^2 [Cohen, C., Texier, B. D., Quéré, D., & Clanet, C. (2015). The physics of Badminton. New Journal of Physics, 17(6). https://doi.org/10.1088/1367-2630/17/6/063001]. Figures for all these metrics are referenced in several papers and serve as targets during the design process. They are validated through a mix of simulation data and real-world testing. Below is a graph showing the angle between the velocity vector and heading as a function of time for a feather shuttlecock with an initial velocity of 18.6 m/s. 

A comparison of Rev - and a feather shuttlecock is shown below. Both are dropped from a height of about 2 meters with the skirt facing down. The feather shuttlecock begins correcting sooner, and exhibits much lower overshoot.

Between Rev - and Rev T15 (the most recent revision), several changes were made to achieve faster stabilization. This includes, but is not limited to, shifting both rings closer to the base by 1 mm, changing the feather material and adhesion method several times, using a lower density epoxy, and switching the type of cork used in the head.

Requirement 2: Non-Ballistic Flight Path

Shuttlecocks exhibit a unique flight path when viewed from the side and top. From the side, the shuttlecock begins with a linear portion in the direction of travel, then becomes parabolic with a rapid and steep dive [Cooke, A. (2002). Computer simulation of shuttlecock trajectories. Sports Engineering, 5(2), 93–105. https://doi.org/10.1046/j.1460-2687.2002.00100.x]. From the top, the shuttlecock makes a very slight right-hand turn [Woo, T., Alam, F., & Kootsookos, A. (2024). Scientific development of Badminton Shuttlecocks. Sustainability and Sports Science Journal, 2(3), 149. https://doi.org/10.55860/zqqe3823]. Both of these effects are substantially more noticable with feather shuttlecocks than plastic models.

This behavior can be explained by the axial rotation of the shuttlecock inducing off-axis moments and is attributed to primary and secondary precession  [Cooke, A. (2002). Computer simulation of shuttlecock trajectories. Sports Engineering, 5(2), 93–105. https://doi.org/10.1046/j.1460-2687.2002.00100.x].

Evaluating the angular velocity of the prototypes for flight path would be challenging since the measurement would need to be taken during free flight. Adding any sort of measurement equipment to the shuttlecock would disturb its flight. Instead of making the birdie move through the air, it would be just as valid to move air across the shuttlecock. To achieve this, I designed a plenum chamber. This takes turbulent, high-pressure air produced by a set of 8 fans in the lower chamber and converts it into laminar air in the upper chamber. A shuttlecock can then "float" in the upper chamber. Assuming all models being compared are the same weight and have similar drag coeffecients, the angular velocities can be directly compared using a laser tachometer. A demonstration of the operation of this jig can be seen below.

From the video above, it is clear that the prototype was rotating too quickly. Indeed during testing, the shuttlecock had too aggressive of a dive.

Additionally, the distance that the shuttlecock travels is critical. This is dependent on many inputs, but mass is one of the critical factors. Given two objects with the same initial velocity and drag profiles, a heavier object (A) will travel farther than a lighter object (B). Assuming both shuttlecocks achieve stable flight in the same amount of time, this weight and distance relationship can be explained by evaluating energy equations.

Several changes were made between Rev - and Rev T15 to reduce the overall weight down to 5.0 g, and the feather angle was decreased by 1.5 degrees to reduce the axial angular velocity.

Requirement 3: Sharp Feel/Sound

The feeling and sound during impact on the racket is critical feedback for the player during a match. Interestingly, the shuttlecock experiences substantial deformation during the impact, as shown in the video below [ScienceLuxembourg. (n.d.). The Beauty of Slow Motion - Shuttlecock. YouTube. https://www.youtube.com/watch?v=xTYsQTHLHk8].

This deformation is affected by both the stems and the rings. To quantify this deformation and compare the stems and rings against a feather shuttlecock, I designed a jig that effectively measures "stress vs strain". Although it doesn't directly calculate these values, it plots load cell output vs stepper motor steps; the slope of this graph can be compared between the prototype and feather shuttlecocks. The jig allows for either a single stem or an entire shuttlecock to be mounted and tested. Other than the load cell, I machined all the parts on this jig using a manual mill and lathe.

The resulting data showed that the prototypes were not as stiff (analogous to Young's modulus) as the feather shuttlecock, which I identified as a likely root cause of the prototype not having as sharp a feeling and sound as feather shuttlecocks.

Between Rev - and Rev T5, some of the major changes to remidy this included changing the ring material/geometry, and changing the stem from a constant, thin cross section to a tapered thick to thin cross section.

Requirement 4: Durable

One of the biggest issues with feather shuttlecocks is the lack of durability; multiple shuttlecocks may be discarded in one session. For my design, the stems are made of a brittle but high-strength composite, whereas the ring and feathers are made of ductile materials. The durability of a material is known as toughness, which is a measure of energy absorption and is defined as the area under the stress-strain curve. Typically, this includes both the elastic and inelastic regions, but since plastic deformation is not acceptable in this application, only the area up to the yield strength is considered in this case.  

The selected material for each portion of the shuttlecock has changed from Rev - to Rev T15 to optimize the ratio between weight and durability. In addition to material selection, durability was also improved by reinforcing areas of high stress concentration. For example, the highest stress concentration for the stems is where it meets the upper ring (as shown below); this has been proven through testing and studying the resulting failures. 

 The width of the stem in this area is now greater than in the original stem design. Although increasing the thickness would have a greater effect on the area moment of inertia and durability in the critical axis, manufacturing limitations and availability of COTS stock prevented this from being viable. Furthermore, an additional high-tensile strength composite tow was laminated to the stem at this interface to reinforce it. Given that it is furthest from the neutral axis, it is able to most effectively help absorb energy from the impact and deformation of the stem. 

A substantial amount of R&D also went into the material, coating, and adhesion methods for the feather. Most notably, the material identified as the most optimal contender for its weight-to-toughness ratio is low surface energy, which means finding an adhesive to fix it to the stem, which was also low surface energy, was challenging. The feather would frequently start peeling from the stem at the top. I solved this issue by cutting a flat pattern of the feather that could then be folded over the feather, as shown below. 

This meant that the selected adhesive would only need to stick strongly to the feather, as the stem was mechanically constrained (sandwiched) by the feather. Although the adhesive doesn't necessarily bond well to the stem materal, it only needs to prevent shear between the stem and feather instead of shear AND peel, which is much easier.

Requirement 5: Cost-Effective

This entire project would only be successful if the resulting product was cost-effective compared to existing offerings. Looking at a volume of 10,000 units, I would be able to sell these shuttlecocks for the same price as nylon birdies with a 48% profit margin. This includes the cost of generic packaging, taxes, and shipping raw materials. 

Manufacturing

This project involved many manufacturing challenges, especially given the constraints of available tooling and budget.

Stems

The stems are manufactured from oversized COTS stock material. The stock is cut to size, and a ~1.5 degree taper is added. Given the small size of the parts and the very acute angle for the taper, this is the most challenging part to fabricate. Additionally, the composite used may produce carcinogenic dust when processed, so the entire operation is done in a vat of mineral oil to minimize airborne particulates. Several prototype jigs were designed and tested, as shown below. Each one was more efficient than the last, processing more stems at once in a more automated fashion.

Feathers

The feathers underwent several manufacturing iterations, including using a drag knife cutter and laser cutter. For the folded feather design, careful considering needed to be taken in regards to the order of operations. For example, if the feathers were cut, then folded, there could be misalignment of the two halves. For this reason, it made more sense to first fold the feather material over the stems, and then laser cut them. This process required several jigs for the insertion of the stems (involving the use of a custom vacuum table) and the alignment of the laser (involving modifications to the laser cutter to include registrations for precision pins on the jig). The process of preparing the feathers for laser cutting, as well as a video of the laser process for an older revision, is shown below.

Cork

The original design for the cork involved drilling holes, but later evolved into cutting slots. Given how small the slots were, I needed to design a custom blade. The original prototype blade was made by grinding material off of an Xacto HSS chisel-edge knife, but the off-center point and relatively flexible HSS resulted in the blade flexing during the cut. To fix this, I procured custom carbide blades that had the tip centered within the cross-section. Given how brittle carbide is, I also designed a jig and blade holder to fully support the blade throughout the cut. The prototype and final blade, jig, and cork are shown below.

Assembly

Given the importance of accurate and precise angles, a jig was required for the assembly as well. The jig was designed using surfaces in SolidWorks to fully support each portion of the shuttlecock during assembly, acting almost as a mold.

Current State of the Project

Although a substantial amount of progress was made between the initial revision and now, there are still several improvement I would like to make before moving into manufacturing. Namely, I aim to increase durability by continuing to iterate on the feather design, and want to improve the feel/sound by trying additional ring materials/thicknesses. Outside of these areas, all the other major problems have thus far been solved. The next phase of the project will be further automating the manufacturing processes and filing a provisional patent.

This page was updated on 9/7/2026