Success Stories & Case Studies

Digitalizing Quality Control Data Problem Statement: Manual data recording in high-speed manufacturing is prone to “Human Error” and creates “Data Silos.” At ETON INDUSTRY, operators had to manually write down critical parameters (Weight, Thickness, Length) for every finished product, leading to delayed reporting and inaccurate traceability. The Challenge: The system needed to interface directly with the existing machine’s PLC to capture data exactly when the process finishes, without disrupting the operator’s workflow. It also required a robust “Target Counter” to alert operators when a specific order quantity was met. ...

Auto Transfer Laser Mark System Problem Statement: In a high-precision machining line, operators were manually transferring production data files from CNC machines to Laser Markers. This manual process, involving walking with USB drives or hand-keying serial numbers, created a bottleneck and introduced a high risk of “Human Error”—resulting in mislabeled parts and traceability issues. The Challenge: The core challenge was Cross-Platform Integration. We needed to bridge “Islands of Automation” that spoke different languages: ...

Digitalization of Batch Process in Rubber Industry Problem Statement: In traditional rubber mixing, reliance on standalone PLCs often limits the complexity of recipe management and data storage. The client needed to upgrade a Banbury Mixer 75L to handle complex step sequences based on Time and Temperature, while ensuring 100% data traceability for quality assurance (QA). The Challenge: The project required high-level interoperability. We had to centralize data from diverse legacy devices speaking different protocols (Ishida Scales, Delta Temperature Controllers, Ammeters) into a single control interface, all while maintaining millisecond-level stability for motor control. ...

Visualizing Invisible Energy Costs Problem Statement: In large-scale manufacturing, electricity bills are a major operating cost, yet they often remain a “Black Box.” Factory managers struggle with manual meter readings, lack of real-time visibility into power consumption trends, and inability to correlate energy spikes with specific production processes. The Challenge: The client needed a centralized system to collect data from multiple Digital Multimeters distributed across the factory. The system had to handle high-frequency data (Voltage, Current, Power, THD), process it for cost analysis, and present it on a responsive web dashboard accessible from anywhere. ...

Mass Production Aging Test System Problem Statement: The client required a mass production testing facility for MXR100030 (30kW) and MXR100050 (50kW) charging modules. The scale involves testing 20 units simultaneously per workstation. The Challenge: The 10-Second Deadline The most critical engineering constraint is the module’s internal Watchdog Timer. If a module does not receive a valid communication packet within 10 seconds, it automatically cuts off the DC power output. With 20 devices flooding the bus with telemetry data, network congestion could easily cause a “Heartbeat Miss,” leading to a failed test and production downtime. Our Solution: High-Throughput CAN Architecture We designed a Multi-Threaded Control Application capable of handling high-density traffic without latency. We moved away from a single daisy-chain loop to a Tree Topology using USB-CAN Hubs to distribute the bandwidth load. ...

High-Precision Sliding Tester Control System Problem Statement: The client, a leading automotive parts manufacturer, relied on a manual or semi-manual process for endurance testing (cyclic motion) of their products. The existing control software was built on legacy technology (WinForms), making it difficult to maintain, prone to human error in parameter settings, and lacked the flexibility to adapt to new hardware sensors. The Challenge: Human Variation: Manual setups led to inconsistent test results (speed, acceleration, cycle counts). Legacy Debt: The old codebase was tightly coupled with specific hardware drivers, meaning replacing a broken motor or sensor required a complete software rewrite. Reliability: The system needed to run thousands of cycles continuously without crashing or losing data. Our Solution: Clean Architecture for Industrial QC We re-engineered the entire control system using Modern C# (WPF) and adopted Clean Architecture principles. The focus shifted from simple “control” to creating a robust “platform” for quality assurance. ...

Project Overview Problem Statement: In high-precision manufacturing, Quality Control (QC) relied on manual data recording from Linear Scales. This led to human errors (typos), lack of traceability, and slow reporting. Furthermore, the client used a massive Excel template filled with complex statistical formulas (CPK, Histogram) to generate reports. The Challenge: The critical technical bottleneck was the Excel Report Generation. Standard libraries (like Interop or standard OpenXML wrappers) failed or crashed because the client’s template file was too large and complex. Opening the file to write data programmatically caused memory overflows and corrupted the complex formulas embedded in the file. ...

Gearset Matching Data Collection System Problem Statement: A Tier-1 automotive manufacturer in Pluak Daeng, Rayong, faced a traceability challenge with their Oerlikon T60 (Bevel Gear Testing) machines. The machine performs critical matching tests for Ring and Pinion gears, but the test results were stored locally in proprietary text files. The Challenge: Closed System: The machine runs on legacy software. Modifying the internal code to send data to a database was deemed too risky and would void the manufacturer’s warranty. Data Silo: Quality engineers had to manually check files, leading to data gaps and lack of real-time traceability in the central ERP/MES. Our Solution: The “Sidecar” Integration Pattern We implemented a Non-Intrusive Data Collection System. Instead of hacking the machine’s software, we treated the machine as a “Black Box” and monitored its outputs externally. ...

CSIAPipeBender: High-Precision Tube Bending Controller Problem Statement: In traditional tube bending, operators often struggle with complex geometry calculations. converting a 3D drawing (XYZ coordinates) into machine actions (Length, Rotation, Angle) requires advanced mathematics. Additionally, physical factors like Spring Back (metal elasticity) and Elongation (stretching) often lead to inaccurate bends, requiring significant trial and error. The Challenge: Complex Math: Manual calculation of bend sequences is prone to human error. Material Behavior: Different tube materials react differently; a standard fixed program cannot account for spring back variance without a smart compensation algorithm. Our Solution: PC-Based CNC Architecture We developed CSIAPipeBender, a Windows-based HMI/Controller that acts as the “Brain” for the machine, while the Mitsubishi PLC acts as the “Muscle.” ...

Real-time Crepper Motor Monitoring System Problem Statement: In a large-scale rubber processing plant, the “Crepper Machines” are the heart of production. The client, Sri Trang Agro-Industry PCL, needed to monitor the electrical current (Amps) of 3 critical motors in real-time. Previously, operators had to manually check readings from local panel meters, making it impossible to analyze historical trends or detect anomalies (like motor overload) instantly. The Challenge: Legacy Connectivity: The existing meters were standard Digital Meters (Primus TCM-94N-AB) communicating via Modbus RTU over RS-485, requiring a reliable bridge to the IT network. Data Integrity: The system needed to handle potential connection drops and ensure data continuity for accurate reporting. User Accessibility: The dashboard had to be accessible via a web browser on the local network without complex software installation on client machines. Our Solution: Hybrid IoT Gateway WP Solution architected a Web-based Monitoring System that bridges the gap between the factory floor and the control room. We utilized Node-RED as a robust middleware to poll data from the meters and Vue.js for a high-performance frontend dashboard. ...