Lucid Automation and Security

**GMP INTERNSHIP PROJECT: PID CONTROL SYSTEM** During my internship Lucid Automation and Security I developed a **PID-controlled water level system** with the following features: ### **System Overview** - Two vertically stacked tanks with a solenoid valve between them - A pump moves water from the **lower tank** to the **upper tank** - A load cell weighs the upper tank. - Two operation modes: - **Manual Mode** – Operator controls the pump manually - **PID Mode** – Automatically adjusts pump speed to maintain a setpoint ### **PID Control Functionality** - Operator selects a **setpoint (SP)** (0-100% of the upper tank) - The system measures the **process variable (PV)** (real tank level) using a **load cell** - The **PID controller** calculates the error (SP - PV) and adjusts the **pump speed (CV: 0-100%)** - Operator can fine-tune the controller with appropriate gain values ### **Technologies Used** ? **PLC Programming** – Allen Bradley **Micro820** (Connected Components Workbench) ? **HMI Design** – FactoryTalk View ? **C++ program for **Arduino Uno** ? **Motor Control** – H-Bridge motor driver L298N for **direction & PWM speed control** Improvements since last post: - The 0-10V signal from the load cell is now filtered and averaged for more stable readings. See my previous post for how I got the Arduino and PLC communicating with a more technical electrical overview.

This past quarter I’ve been working on a PID water level control project at Lucid Automation and Security with lots of help from Bill Mueller. One of the biggest challenges was establishing communication between an Arduino Uno R3 (Arduino IDE, C++), a Micro820 Allen-Bradley PLC (CCW, Ladder), and a PanelView Plus HMI (FactoryTalk View). Here's what I've accomplished so far. Arduino - PLC communication Analog communication: (motor speed) * The load cell generates a 0-60mV signal which is converted to 0-10V by a conditioner.? * This 0-10V is fed into the PLC AI pin and scaled by half to 0-5V. * The 0-5V AO from the PLC is fed into the arduino which outputs a 0-5V PWM signal. * This 0-5 PWM signal is converted to 0-24V PWM by the L298N motor driver.? * Overview: 0-60mV – 0-10V – 0-5V – 0-5V PWM – 0-24V PWM… Digital communication:?(motor direction) * Digital Out 0 from the micro820 (connected to the DUCK button) generates a 24V digital signal which activates a steady state relay. The “normally open” pin grounds the arduino’s digital pin (11). * The relay is “sinking” the digital pin which is using an internal pull up resistor, meaning its normally HIGH. - If (pin 11 is 1) {the arduino will set DO 5 to HIGH and DO 4 to LOW (motor forward) - If (pin 11 is 0) { the arduino will set DO 5 to LOW and DO 4 to HIGH (motor reverse)}? The purpose of this relay is to isolate the 5V arduino digital pins from the 24V from the micro820 PLC Solenoid control? * Digital Out 1 from the micro820 (connected to the DOOR button) generates a 24V digital signal and activates a steady state relay. * This relay is in the sourcing configuration as its normally open pin provides 24V to a solenoid.? * The solid-state relay protects the PLC from harmful back EMF generated by the solenoid coil when it turns off, preventing voltage spikes and potential damage to the digital pins. Note: The arduino can likely be eliminated as this Micro820 is capable of producing a PWM output for speed control. And I could use two digital outputs with relays for motor direction… but it was a good challenge. Let me know if you would implement the arduino - PLC communication differently or have any suggestions for the HMI layout.

We're #hiring a new Automation Engineer Intern in Davis, California. Apply today or share this post with your network.

In GMP life sciences, upgrades are tricky with all the legacy validation to consider. Swapping out a reporting solution, the core of GMP quality systems, at end-of-life is f’ing rough... With FT VantagePoint reaching end-of-life, we’ve spent years evaluating alternative solutions. Are AI dashboards a possibility? Nope. They’re not even in the conversation with quality teams. Don’t get me wrong, they’re extremely cool and useful. But right now, they’re mostly home-brew solutions. We use AI internally to uncover non-obvious connections between data and operations, but always with an engineer in the loop to evaluate and implement. A lot of AI’s suggestions don’t work in the real world, but there are gems to be found. For migrating existing batch reports, there are solutions: SSRS, RT Reports, Dream Reports, Power BI, and PI Vision. However, none provide a direct migration path from VantagePoint. If you’ve stuck with VantagePoint long enough, you know its power comes from using a familiar tool (Excel) that directly accesses multiple SQL databases and PI direct. Any process engineer can create custom reports, and any quality person can see exactly how the data flows into them. That builds a lot of trust in the product. One that must be re-established when moving to a new platform. We’ve become skilled at migrating reports to new platforms by managing expectations and simplifying the reports to leverage each platform’s strengths. Change is rough, but necessary. There are solid paths to upgrade existing batch reports. Just don’t wait until the last minute, chasing flashy AI dashboards, to start the process. #BatchReports #VantagePoint #LucidAutomation

GMP compliance is time-consuming, don’t make it harder with a bolt-on solution. A turnkey system designed for easy documentation, validation, and standard practices is just faster. Faster to qualify, easier to maintain, and helps avoid prolonged audits that can bog your internal quality teams down. Our expertise in designing systems from the ground up, using commercial off-the-shelf tools, integrated change control, and a risk-based testing approach, results in more predictable execution. Projects more predictable in costs, schedules, and the reliability of the turnover package to stand up during audits. For existing systems, we can retrofit the same validation plan and tools to close potential gaps. Developing a trace matrix makes the entire process smoother and more efficient. Lucid Automation is here for all your Life Science GMP Automation, Networking, and Data Integrity needs. Specializing in automated control systems for biotech and pharmaceutical manufacturing, everyone benefits from having a trusted partner like us. #Automation #LifeScience #LucidAutomation

We're #hiring a new Automation Engineer Intern in Davis, California. Apply today or share this post with your network.

Happy Holidays! It’s been a fantastic 2024 working together, and we’re proud of the exceptional projects we’ve accomplished as your automation partner. Lucid Automation will be running with a skeleton crew from December 16th to January 3rd as our engineers focus on off-site training and preparations for exciting new projects in 2025. Rest assured, we’re still available during this time, though responses may be a little slower. If you need immediate assistance, please contact your favorite Lucid engineer and cc either Marie Fiala or Bill Mueller for the fastest response. From all of us at Lucid Automation, we wish you a joyful holiday season and a prosperous 2025!

This month, we proudly celebrate five years of engineering excellence that Natalie has dedicated to delivering exceptional solutions for her clients. Congratulations on this significant milestone! #EmployeeRecognition #LucidAutomation

Nobody reads them, but our case studies show one thing: when Process and QA engineers start with a blank slate, they rarely know what they need. But give them a simple, clear report, and suddenly everyone has ideas for the “perfect report” once it’s in use. Taking that first step takes courage and a commitment to improvement. Is that you? Life sciences companies are buried in GMP process data with minimal context. We see it all the time: stacks of raw data, recorded, stored, but ultimately, not useful. What good is a Pi Historian full of numbers or a Vision system that can’t show the trends relating process value to a triggered alarm? At Lucid Automation, we make data actually work for you. We bring context and clarity to your existing GMP data, turning it into insights your team can act on. Process Engineers can finally see meaningful trends, automation engineers can troubleshoot faster, and quality teams can make confident, data-backed decisions. Release by exception anyone? Using tools like Pi Historian, Vision, FT Historian and Dream Reports, we simplify data interpretation. Our methods are industry-standard and proven – no guesswork. This isn’t about fancy dashboards; it’s about seeing exactly what’s happening in real time and knowing how to respond. The payoff? Less time digging for information and more time making decisions. Productivity goes up, bottlenecks get solved faster, and compliance becomes less of a headache. If you’re ready to make your GMP data work as hard as you do, let’s talk. At the end of the day, clear, contextualized data isn’t just a nice-to-have – it’s a game-changer. #GMP #GMPData #ManufacturingIntelligence #LucidAutomation