Development and integration of the NA64-DTC automation controller for the CERN "DESY Table'' motorized platform

This paper reports on the design, construction, and successful commissioning of the NA64-DTC, an ESP32-based remote automation controller that enables non-invasive, opto-isolated control of the CERN "DESY Table" motorized platform via HTTP, thereby facilitating remote operation and reducing manual intervention for various experiments.

The Gist

Imagine you are in a massive, high-tech laboratory where scientists are shooting beams of tiny particles at various targets to study the building blocks of the universe. To do this, they use a giant, heavy-duty motorized table called the "DESY Table." Think of this table like a high-precision, industrial version of a camera slider or a telescope mount. It can slide heavy detectors (which are like giant, sensitive cameras) left, right, up, and down with extreme accuracy.

However, there was a problem: to move this table, a human had to stand right next to it, press buttons on a physical control panel, and watch a small digital screen to see where it was. If the scientists wanted to move the table while the particle beam was active, they had to leave their safe control room, walk over to the noisy, dangerous beam area, fiddle with the buttons, and then walk back. It was slow, manual, and interrupted the flow of the experiment.

The Solution: The "Remote Control" for the Table
The authors of this paper built a clever little device called the NA64-DTC. You can think of this device as a "digital puppet master" or a "ghost hand" that sits next to the table's control panel.

Here is how it works, using simple analogies:

• The "Ghost Hand" (Opto-Isolation): The device doesn't actually touch the buttons or cut any wires. Instead, it uses "opto-isolators." Imagine these as invisible light-beam fingers. When the computer wants to press the "Move Right" button, the device flashes a tiny light that tricks the table's control panel into thinking a human finger just pressed the button. This allows the scientists to control the table from their computers without ever touching the original hardware or breaking any warranties.
• The "Brain" (ESP32 Chip): The heart of the device is a small, cheap, Wi-Fi-enabled computer chip (the ESP32-C3). It's like the brain of a smart home device. It connects to the lab's Wi-Fi network, just like your phone connects to the internet.
• The "Eyes" (Encoders): The table has motors that spin to move the platform. The device has "eyes" (sensors) that watch the motors spin and count the pulses, like a pedometer counting steps. This tells the computer exactly where the table is, down to a fraction of a millimeter.
• The "Remote Control" (Web Interface): Once connected, scientists can open a web page on their laptop or phone. This page looks like a simple dashboard with buttons for "Left," "Right," "Up," and "Down." They can type in a distance (e.g., "move 5 millimeters to the right") and hit enter. The "Ghost Hand" then presses the buttons automatically.

How They Built It
The team designed a small circuit board that plugs directly into the back of the table's control panel.

1. No Surgery: They didn't have to open up the table or rewire anything. They just plugged this new device into a spare port on the back of the existing panel.
2. Power: It runs on the same power cable that powers the table's control panel, so no extra batteries or power cords were needed.
3. Safety: If the big red "Emergency Stop" button on the original panel is pressed, the device instantly knows and stops moving. It also has its own software "panic button" in case the Wi-Fi connection gets weird.

The Results
The team tested this device at CERN (the famous particle physics lab) in 2026.

• It Worked: They successfully moved the heavy table remotely while the particle beams were running.
• It Was Accurate: The table moved exactly where the computer told it to, with an error margin of less than half a millimeter (about the thickness of a credit card).
• It Was Reliable: It stayed connected to the Wi-Fi and didn't crash, even during long experiments.

Why This Matters
Before this, moving the table required a human to be physically present. Now, scientists can automate the movement. They can write a script to move the table in a specific pattern, or just click a button from their office. It saves time, reduces the need for people to walk into dangerous beam areas, and makes the experiments run smoother.

The paper concludes that while they built this for one specific experiment (NA64), the design is so simple and generic that any other experiment at CERN using this same type of table could use it too. It's like turning a manual car into an automatic one, but without changing the engine.

Technical Summary

Technical Summary: Development and Integration of the NA64-DTC Automation Controller

Problem Statement
The "DESY Table" is a motorized platform widely utilized at CERN's East-Area and North-Area experimental installations for the precise positioning of detectors and targets, particularly in fixed-target configurations involving electromagnetic and hadronic calorimeters. While capable of handling loads up to 103 kg with a ±500 mm excursion range, the platform relies on a manual control panel for operation. This necessitates frequent physical intervention by operators to adjust configurations, with position data available only via local displays. This manual dependency limits the efficiency of beam activities and prevents seamless integration into automated, remote control systems required by modern experiments.

Methodology
To address these limitations, the authors developed the NA64-DTC (DESY Table Controller), a remote automation device designed to interface non-invasively with the existing DESY Table hardware. The methodology focuses on three primary development stages:

1. Architectural Design: The controller connects to the DESY Table's rear 28BSM "switches duplication" connector. This interface allows the device to emulate button presses and read encoder signals without modifying the original control electronics. The system is designed to be powered directly from the platform's 24 V DC supply (consuming <10 W) and operates wirelessly via the CERN Wi-Fi network.
2. Hardware Implementation: The core of the device is an ESP32-C3-WROOM-2 System-on-Module (SoM), chosen for its low power consumption, built-in Wi-Fi, and compact footprint. The hardware utilizes eight MOCD223M dual-channel opto-isolators to safely interface the low-voltage microcontroller with the high-voltage (24 V) control panel signals. This setup emulates directional buttons (Left/Right, Up/Down, Go) and monitors the emergency stop circuit. An optical-isolated phototransistor reads the incremental encoder outputs (A and B channels) from both motor axes to determine position. The design includes a custom 2-layer PCB with a 20-pin IDC connector to mate directly with the platform's ribbon cable, minimizing mechanical stress.
3. Firmware and Software Development: The firmware runs on FreeRTOS, implementing a multi-threaded application that handles GPIO interrupts for quadrature encoder decoding (software-based ×4 decoding) and manages motor control logic. Safety is enforced through both hardware (monitoring the physical E-Stop) and software mechanisms. The device exposes a REST API via an embedded HTTP server on port 80.
   • User Interfaces: Two interfaces were developed: a lightweight, browser-based Web UI for direct operator interaction and a custom EPICS SoftIOC for integration into the NA64 slow-control subsystem. The EPICS implementation includes Process Variables (PVs) for position readback, motion commands, soft travel limits, and emergency stop logic.

Key Contributions

• Non-Invasive Integration: The solution achieves full remote control and automation without altering the original DESY Table control panel or requiring external power sources.
• Dual-Mode Operation: The system allows for simultaneous manual and automated operation; the original panel remains functional for manual overrides while the controller is connected.
• Standardized Interface: By utilizing a generic HTTP-based REST API, the device is designed to be experiment-agnostic, allowing integration into various control systems beyond the specific EPICS implementation used for NA64.
• Safety Logic: The system incorporates robust safety features, including hardware E-Stop monitoring, software E-Stop assertion, motor stall detection, and configurable soft travel limits.

Results
The NA64-DTC was successfully commissioned during the 2026 NA64 experimental runs at the PS T9 and SPS H4 beamlines.

• Performance: Testing confirmed stable Wi-Fi communication and correct encoder-based position tracking. The system achieved a positioning accuracy of 0.2 to 0.4 mm across the full movement range, consistent with the nominal motor configuration and the 400 counts/mm scaling factor derived from the 100 PPR encoders and firmware decoding.
• Operational Success: The device was routinely used to move the ECAL detector for calibration and alignment with the beam direction. It successfully handled an undocumented hardware behavior where movement axes were inverted, which was corrected via a firmware update.
• Reliability: The system demonstrated robust behavior during extended periods of continuous use, including reliable handling of emergency stop conditions and encoder fault detection.

Significance
The paper claims that the NA64-DTC provides a practical solution for reducing manual intervention during beam activities at CERN, thereby improving operational efficiency. While initially conceived for the NA64 experiment, the authors emphasize that the generic nature of the hardware and the HTTP-based interface make the solution applicable to any CERN experiment utilizing the DESY Table platform. The work demonstrates that legacy motorized platforms can be effectively modernized for remote and automated operation without the need for invasive hardware modifications or external power infrastructure. All design files, schematics, and firmware sources have been made publicly available to facilitate adoption by other experiments.