USC STEAM 2024: Adjustable Saddle Generator

Scope of Our Mission

The project aimed to develop a parametric support system for ventilation ducting, utilizing Autodesk Inventor to create a model that could be easily modified based on user-defined parameters. The goal was to design a flexible, modular system that allowed for quick adjustments to dimensions, the number of supports, and other key elements. The team applied advanced CAD techniques, including multibody modeling, parametric design, and iLogic programming, to automate and streamline the design process.

Our Journey

The team was composed of Aaryan Kumar, Shriya Narasimhan, and Komel Nulwala, each contributing their expertise. Aaryan was primarily responsible for developing equations and defining the key parameters necessary for parametric modeling, while Shriya focused on project management, communication, and troubleshooting. Komel worked extensively on 3D modeling using parametric design tools, including tackling the creation of the angled saddle and integrating various modeling techniques to meet the project’s requirements.

The project began with initial research into Autodesk Inventor’s capabilities, particularly in creating plug-ins and multibody parts that could be used to make a horizontal saddle. Aaryan defined the key parameters needed to be adjustable in the design, such as height, width, and the number of supports. The team also organized meetings with their EAFab mentor to align on the project scope, after which they began working on initial models. Early attempts revealed challenges with ensuring the parametric design was both accurate and flexible. Komel faced difficulties with using various tools in Inventor to achieve the desired angle in the supports, which led to multiple iterations and adjustments. During this time, Aaryan worked on defining user-driven dimensions and incorporating them into the model, using tools like the rectangular pattern feature to control the number of supports. However, the team faced complications when trying to automate the generation of multiple supports with varying heights.

Over the following weeks, the team continued to iterate on the design, seeking guidance from our EAFab mentor, who provided feedback on various aspects of the project. His suggestions, such as using revolved features for the angled saddle and splitting the solid body for easier patterning of supports, were incorporated into the model. However, the angled saddle model remained a challenge due to its geometric complexity. The team explored multiple solutions, including iParts and iLogic, but encountered difficulties in automating the design process. They also experimented with alternative tools like the Patch and Sculpt tools to address issues with the angled saddle’s geometry. Despite some progress, the model continued to present setbacks, especially with the angled supports and the need to create a flexible system that could adjust to the varying angles of the ducts.

By early January 2025, the team focused on finalizing the third and final model, which was the rectangular saddle, and perfecting the customizable number of supports.

At the same time, they continued to refine the angled saddle, though the challenges persisted. Despite revisiting the loft tool and considering other approaches, the team was unable to fully resolve the issues with the angled saddle. The angled design required much more complex math to define the supports’ dimensions and placement than the timeframe allowed, leading to significant delays. The last element that the team worked on was the iLogic forms that allowed the user to control the parameters of the model. The team grouped parameters to make the form intuitive and created technical drawings that depicted what each parameter would change.

Final Solution

The team utilized Autodesk Inventor’s multi-body parts, parametric design tools, and iLogic, along with trigonometric calculations to define relationships between parameters like the height of supports and saddle angle. Despite challenges with automation and geometric complexity, their efforts in learning new techniques enhanced their CAD modeling skills. Our EAFab mentor's feedback was crucial in troubleshooting and refining their approach. The project made progress in developing a flexible support system for ventilation ducting, providing valuable insights into advanced modeling and automation. While the angled saddle remains a challenge, the team is optimistic that the final model will meet the specifications and provide a robust solution for varying duct configurations.

USC STEAM Students who worked on the project: Komel Nulwala, Aaryan Kumar, and Shriya Narasimhan.