Linkedin
Youtube
Github

Ball Feeding Mechanism

Written by

in

Comments

Jaguar Land Rover Suspension Knuckle

April 1, 2025

As part of a university-industry collaboration with Jaguar Land Rover, guided by JLR’s Ravi Londhe, our team designed a rear suspension knuckle to demonstrate our ability to deliver a real-world automotive component. The knuckle was modelled in Fusion 360 with supporting SolidWorks components assembled into a simplified but representative suspension system for integration validation. Three manufacturing methods were evaluated — casting, CNC machining, and additive manufacturing — with A356 aluminium sand casting combined with post-process CNC machining selected as the optimal production route.

My personal focus was FEA and additive manufacturing research. I led the structural validation in ANSYS Workbench, performing static load and harmonic response analyses on the cast design, confirming safety factors exceeded 2.5 under real-world loading conditions. I also investigated advanced manufacturing routes including DMLS with titanium alloy and high-performance FDM materials such as PPA-CF, PEEK, and nylon carbon fibre, contributing material selection insights and manufacturing trade-off analysis to complement the team’s work on casting and CNC machining. This project was an excellent opportunity to apply engineering skills in a real-world industry scenario.

Model & assembly

Assembly
Assembly
Assembly
3D Model
3D Model
Knuckle – Drawing

The rear suspension knuckle was designed from scratch in Fusion 360, with supporting suspension components sourced from SolidWorks and assembled within Fusion to create a simplified but representative suspension system. This allowed for integration validation, confirming correct fitment and clearance across all mounting interfaces before committing to manufacture.

The knuckle was modelled with industry-standard design practices in place throughout — including correct draft angles for casting, appropriate tolerancing on all mating faces, and geometry optimised for ease of manufacture and assembly. All fastener interfaces were designed around standard metric hardware to reduce cost and improve repairability. A fully detailed engineering drawing was produced to communicate critical dimensions, tolerances, and surface finish requirements, ensuring the design could be handed directly to a manufacturer without ambiguity.

FEA Static Load

Mesh
Force / Load
Total deformation
Total deformation
Equivalent (von-mises) stress
Equivalent (von-mises) stress
Equivalent (von-mises) strain
Equivalent (von-mises) strain
Harmonic response stress
Harmonic response Deformation
Harmonic response strain

Analysed a suspension knuckle in ANSYS Workbench 2024 R2 using a uniform triangular mesh. Maintained high-quality elements with controlled sizing and smooth transitions. Verified mesh integrity to minimize distortion. A static Vertical load of 4,200 N was applied. Results showed stresses well below material yield (13% of 125 MPa), minimal deformation (<0.02mm), and safety factors of 5-15.

Design meets static requirements but would need dynamic testing for real-world validation. Harmonic response tests were also carried out. Normal stress reveals critical frequencies where stress peaks could cause fatigue cracks, especially near engine vibration ranges (400–2000 Hz). Normal strain shows deformation risks at high frequencies (e.g., 5000 Hz from tire imbalance), which could loosen joints or amplify cabin vibrations in lightweight designs.

Additive Manufacture — FDM (PPS-CF)

Sliced model (Line type)
Sliced model – Line 400/740
Printed 3D Model (PLA)
Print Orientations

A showcase prototype of the knuckle was 3D printed in PLA for physical display and form verification. Alongside this, the viability of high-performance FDM materials was explored — demonstrating how the knuckle could be sliced and printed using PPS-CF filament (168 MPa, 227°C heat resistance) with optimised parameters to maximise part integrity and minimise supports, should the material prove sufficient.

Further investigation covered DMLS using titanium alloy and advanced polymers such as PEEK and nylon carbon fibre as potential production routes. While FDM materials have advanced significantly, the anisotropic nature of layer-based deposition makes polymer printing unsuitable for a safety-critical suspension component. DMLS remains the only additive route capable of producing a structurally viable knuckle, though its cost makes it impractical at road-car production volumes. Additive manufacture was therefore limited to the prototyping stage.

Die Casting – Integrating Coolant Access into the Clamping Device
Die Casting – Integrating Coolant Access into the Clamping Device
CNC machining simulation – slot cutting for central bore
CNC machining simulation – slot cutting for retaining ring

Die Casting & CNC Machining

For production, sand casting with post-process CNC machining was selected as the primary manufacturing route, with a die-cast variant developed as a high-volume alternative. Mounting holes were omitted from the cast geometry and finished to tolerance by CNC in a subsequent operation. The sand casting mould was designed as an 8-part assembly, incorporating M16 clamping features, pouring holes, and recessed spill catches.

The die-cast variant added an internal coolant channel network to reduce cycle time and improve dimensional consistency. Draft angles of 2–3° for sand casting and 0.5–1° for die casting were applied throughout, with A356’s 6.5% linear shrinkage accounted for in the cavity geometry. Post-casting CNC milling then finished all critical faces and bores. This combined approach delivers the best balance of cost, precision, and scalability for a standard road vehicle.

Topspin & Backspin Tennis Ball Launcher

April 20, 2026

I designed and built a dual-wheel tennis ball launcher capable of producing topspin, backspin, and flat shots — the project’s primary objective. The core spin mechanism exploits the Magnus effect through independent control of two A2212 2200KV BLDC outrunner motors, with a 5 mm wheel-gap compression validated through hand calculations and physical testing at 30–40 mph. The prototype was completed within a £300 budget (~£200 real spend), with all structural components FDM-printed in PETG across approximately 30 hours of printing time.

I also examined the feasibility of vertical and horizontal angle control as an extension of the baseline concept. The electronics system — Arduino Nano, two 30A OPTO ESCs, A4988 stepper driver, 12V→5V buck converter, and 6-way fuse box — was built and tested as the foundation for a future polished system including motorised angle adjustment and wireless ESP32 Bluetooth control. The vertical tilt axis is driven by a custom-designed 4:1 3D-printed planetary gearbox paired with a NEMA 17 stepper motor, producing a theoretical torque output of 1.62 N·m.

Model & assembly

Full Assembly Isometric
Internal Mechanism Drawing
Top Wheel Sub-Assembly
Bottom Wheel Sub-Assembly
4:1 Gear Reduction Drawing
Ball Feeding Mechanism

Designed the complete machine in SolidWorks, producing individual sub-assemblies for the top wheel, bottom wheel, vertical angle stepper system, and planetary gearbox. Each sub-assembly is backed by a full engineering drawing with BOM, tolerances, and material callouts. The 4:1 planetary gearbox — sun, planet, and ring gears all designed from scratch — steps up the NEMA 17 stepper output to a theoretical 1.62 N·m of torque for the vertical tilt axis.

Electronics system

Full Wiring Diagram
Full Wiring Diagram

The electronics system is built around an Arduino Nano as central controller. Two A2212 BLDC outrunner motors are independently driven by 30A OPTO ESCs, enabling differential speed for topspin, backspin, and flat shots. A servo motor opens and closes the ball-feeding door on a 2-second cycle. An A4988 driver handles the NEMA 17 stepper for vertical angle adjustment. All power is routed from a 12V battery through a 6-way fuse box, with a 12V→5V buck converter supplying logic power. The system lays the foundation for future wireless ESP32 Bluetooth control.

3D print & final product

Final Prototype
Final Prototype
Printed Components (Wheel and 4-1 Gearbox)
Slicer: 9 Part Grid Layout (1d 6h 50m)
Slicer: Ball Feeding Mechanism + Frame Sections (6h 21m)
Slicer: Wheel (4h 20m)

All structural parts were FDM-printed in PETG across approximately 30 hours and 8 print plates using a Bambu Lab printer. Key components include the dual-wheel housings with integrated tread (4h 20m per wheel, 150 g each), the ball feeding mechanism and frame sections (6h 21m, 252 g), the base plate, and the full planetary gearbox casing. Print orientation and infill were optimised per part for structural strength at key load points. The assembled prototype successfully launches tennis balls with controllable spin at 30–40 mph.

Full Assembly Isometric
Full Assembly Side View
Full Assembly Top View
Full Assembly Bottom View

IMECHE Design – Automated device

March 30, 2024

Designed and assembled a miniature vehicle for the 2024 IMechE Design Challenge in SolidWorks and conducted structural analysis using Ansys for weight optimisation.

Performed detailed gear and timing belt calculations. Applied FDM and resin printing techniques focusing on tolerance and manufacturability. Coordinated team roles and project phases and ensured seamless mechanical-electrical integration.

Ball Feeding Mechanism

Comments

Leave a ReplyCancel reply

More posts