November 21, 2025
By Alba Garcia-Sarabia

Get to know mechanical engineering Capstone Design team Mental Mechanisms and learn about their project, "Automated End-Effector for Variable-Spacing Sound Intensity Measurements," before they showcase it at the Fall 2025 Capstone Design Expo on December 1. The team is advised by Professor Tequila Harris. 
 

Can you tell us about your team and project?

Our team is comprised of six interdisciplinary students, including:

• Ramy Aldabbagh – Mechanical engineering undergrad from Suwanee, Georgia, who is responsible for programming, specifically the UR robot interface and motion handlers
• Ansh Arora – Mechanical engineering undergrad from Alpharetta, Georgia, focused on FEA, machine design, and video editing/graphic design
• Timothy McMurray – Mechanical engineering undergrad from Wolcott, Connecticut, who is well-versed in machining, PCB design, CAD, and electrical systems
• Ansh Mishra – Mechanical engineering undergrad from Mumbai, India, focused on acoustic analysis, modal analysis, and FEA
• Jason Moore – Electrical engineering undergrad from Montgomery, Alabama, who worked on microcontrollers, buck converter, and decoupler design
• Akil Quadir – Mechanical engineering undergrad from Stockbridge, Georgia, who primarily worked on CAD and programming aspects, including object detection and image processing

Our project, sponsored by Rheem Manufacturing Company, is focused on building a robotic end-effector for automated variable-spacing sound intensity measurements. Put simply, we’re building a robotic tool that is attached to the end of a UR5 robotic arm that automatically changes the spacing between two microphones used to measure how sound moves around different objects, like air conditioning units or water heaters. This process is done to identify noise sources and improve acoustic performance in products.

Traditionally, Rheem’s engineers have to manually adjust spacers between the microphones for each test, which can take almost eight hours. Our robotic system can do this in under 30 minutes, while also making the results more consistent and accurate.

What makes our system unique is that the end-effector doesn’t just move the microphones; it also maps the surface of the object being tested. Using image processing, it creates a 3D mesh of the surface and generates an S-pattern path made up of key nodes. The robot then follows these nodes to precisely scan the entire object, ensuring complete and repeatable sound measurements.

To build this system, our team combined: electrical design (motor drivers and microcontrollers that power and control the motion), mechanical design (motor mounts, linear rails, and structural supports that hold and move the microphones), and software development (computer vision, object recognition, and motion control for the UR5 robotic arm, all programmed in Python).

We’re using automation, robotics, and smart sensing to make sound testing faster, smarter, and far more reliable for industrial applications.
 

What was your team’s design process, how do you go about separating tasks, and what are some of the biggest challenges you have faced?

We divided tasks based on each team member’s strongest skill set, which made our design and prototyping process highly efficient. Members experienced in coding and object detection focused on software development, while those skilled in fabrication and CAD handled mechanical design. Team members with strengths in video editing and graphic design worked on the informational video and capstone poster.

We also applied design tools learned in prior courses such as ME 2110, including the house of quality, function trees, and specification sheets, to gather critical information from our sponsor, Rheem, and clearly define the system’s functions in noun-verb pairs. This structured approach improved communication during our weekly sponsor meetings.

The biggest challenge arose when the UR5 robotic arm became unavailable, which initially slowed progress. However, the team adapted by collaborating with third-party labs and switching gears on what can be done now to continually progress and validate our end effector design.
 

Have there been any highlights while working on your project?

One of the biggest highlights of our project has been developing an efficient object detection system that works without large dataset classifiers, avoiding the need for AWS servers and extensive Ubuntu-based training. Instead, we implemented a foreground identification method that saved significant time and required minimal memory while maintaining strong performance. Another major highlight was integrating the software with the UR5 robotic arm and camera entirely through Python, which streamlined control, improved operator clarity, and reduced overall code complexity.

Beyond the technical achievements, the most rewarding part has been how multidisciplinary the project is. It combines robotics, electronics, software, mechatronics, acoustic engineering, and precision machining into one cohesive system. Designing custom mechanical components, fabricating parts, building an embedded control system, and seeing it all interface seamlessly with the UR5 robot has been an exciting and hands-on learning experience that truly brought together our team’s diverse engineering strengths.
 

Can you tell us about your experience working with a sponsor?

Working with Rheem Manufacturing Company was one of the most valuable parts of this project. Our sponsor contact, Shantanu Chavan, gave us the freedom to innovate while guiding us with industry insight. His mentorship helped us align our academic work with practical applications in acoustic testing and automation. Regular meetings ensured we stayed on track and met deliverables while learning how R&D engineering operates in a corporate environment.
 

What team attributes have contributed the most to your success?

Our team’s interdisciplinary mindset, open communication, and mutual respect were key to our success. Everyone brought unique expertise, from FEA and CAD modeling to circuitry design and coding, and we trusted each other’s strengths. We also shared a strong sense of accountability; every member consistently delivered, which made collaboration seamless.
 

If you could go back in time, what advice or message would you give your team on day one?

We’d tell ourselves to embrace iteration early- test fast, fail fast, and refine. Early prototyping, even at low fidelity, saves enormous amounts of time later. We’d also remind ourselves that effective communication is just as critical as good design; documenting decisions and maintaining open dialogue with sponsors and advisors makes all the difference.

The Fall 2025 Georgia Tech Capstone Design Expo will be held on Monday, December 1, at McCamish Pavilion on the Georgia Tech campus. At the Expo, over 600 seniors from various disciplines of engineering and industrial design will showcase their innovative projects designed and built during their Senior/Capstone Design course. Register to attend here!

This semester's Expo includes industry sponsors whose donations support Transforming Tomorrow: The Campaign for Georgia Tech.

Companies, entrepreneurs, and organizations interested in sponsoring projects in the Capstone Design class can contact Amit Jariwala. For more information on what projects are a good fit for the course, please review this brief YouTube video.