EtherCAT Motion Controller for Delta Robot Programming Guide

EtherCAT is a real-time industrial communication protocol where the master sends cyclic data packets to slaves at regular intervals, with each packet containing a specific node address that determines which slave receives and processes the data; the protocol uses a distributed clock system where each slave introduces its own timing offset, allowing the master to compensate for delays and ensure synchronized data exchange, while supporting flexible topologies (series, ring, tree, star) and utilizing standard Ethernet cables for cost-effective implementation.

EtherCAT is a real-time industrial communication protocol where the master sends telegram messages containing data for all slaves in sequence, with slaves writing data back into the same telegram. This exchange occurs at fixed intervals (e.g., every 200 microseconds), providing deterministic timing. Sync Manager channels 0-3 manage Process Data Objects (PDOs) for real-time control data like position and velocity, while channels 0-1 handle Service Data Objects (SDOs) for configuration. The object dictionary organizes parameters hierarchically using main indexes and subindexes, with specific memory address ranges for different data types. Controlling an axis requires four elements: control word, target position, state feedback, and current position feedback. PDO lists in memory area 1C12 define which data objects are transmitted through each sync manager channel.

EtherCAT is an Ethernet-based field bus protocol designed for real-time industrial automation, achieving cycle times under 100 microseconds with low jitter; it uses a daisy-chained topology where data travels through slave nodes that read and write process data on-the-fly, utilizing distributed clocks synchronized to achieve sub-100 nanosecond timing accuracy for precise control applications.

EtherCAT (Ethernet Control Automation Technology) is an industrial communication protocol enabling real-time control of automation devices through Ethernet networks. Unlike standard internet with peer-to-peer architecture causing data delays, EtherCAT implements a master-slave architecture with token-passing mechanism ensuring deterministic, real-time communication. The daisy-chain topology reduces wiring complexity and costs compared to traditional point-to-point connections. Hardware setup involves the NX12 controller with IN/OUT EtherCAT ports connecting to R88D1N servo drivers in a daisy-chain fashion. Each device requires dual power supplies: 24V DC for control circuits and 220V AC for motor operation. The controller also includes an additional port for standard Ethernet/LAN connections for programming and configuration.

In EtherCAT networks, the master sends data using a single frame that passes through the entire node network. The master is the only node permitted to send frames, and this frame is then forwarded downstream by each node in the network. This approach ensures real-time operation and avoids delays by eliminating the need for individual request-response cycles between nodes.

Trapezoidal velocity profiles consist of three phases: acceleration at constant rate, constant velocity, and deceleration at constant rate. They were commonly used in early motion control systems but result in very high jerk values at transition points. S-curve velocity profiles provide smoother motion by gradually changing acceleration rather than having abrupt transitions, eliminating the infinite jerk of trapezoidal profiles. S-curve profiles result in more elegant, efficient, and less stressful motion that reduces wear on mechanical components and improves product quality.

Closed-loop motion control continuously compares desired positions (setpoints) with actual positions using feedback sensors. Error—the difference between target and actual position—triggers corrective action through proportional, integral, and derivative (PID) algorithms. Proportional control generates corrections proportional to error magnitude, but cannot eliminate steady-state error alone. Integral action accumulates persistent errors to drive the system to exact targets. Derivative action predicts future error based on velocity trends. Proper tuning balances responsiveness against stability, preventing overshoot and oscillation while achieving precise positioning even in dynamic environments with external disturbances.

Trapezoidal motion ramping is a control technique that regulates motor position and speed by generating smooth acceleration and deceleration profiles, preventing excessive torque demands and reducing overshoot in closed-loop motor systems; this is achieved through a state machine that transitions through acceleration, constant velocity, and deceleration phases, with the ramp generator outputting requested positions that the position PI regulator tracks to achieve precise motion control.

Profiled PID controllers with trapezoidal profiles provide smooth robot motion by constraining acceleration and velocity. Unlike standard PID which could cause jerky movements, trapezoidal profiles enforce accelerate-cruise-decelerate behavior. Configurable constraints include max velocity (m/s), max acceleration (m/s²), and max angular velocity (rad/s). Separate controllers handle X, Y, and yaw axes independently. Continuous rotation mode allows wrapping from -180 to +180 degrees for seamless turning behavior.

This comprehensive section covers the complete process of configuring a rotary axis in Linux CNC and tuning the PID controller. Key topics include: PID error at zero position indicating motor stationarity, encoder scale configuration (213 pulses per degree for joint 5, 199.8 counts per degree for joint 4), PWM signal scaling (0-100 where 50 is neutral), angular axis parameters (60 degrees per second speed), limit switch configuration (0.25V minimum, 9.9V maximum), and the complete home sequence involving limit switch triggering, index pulse detection, encoder zeroing, and final position movement. The section covers PID output interpretation and direction verification, demonstrating how to verify PID direction by moving axis to known position and observing output. The complete PID tuning process is demonstrated: starting with low values (P=1, I=0, D=0), gradually increasing P values (1, 10, 20, 30, 40, 50, 100, 150), observing axis behavior, reducing oscillation by adjusting P and I values, using mechanical damping (touching shaft) to reduce inertia, and adding D value for derivative feedback. The section emphasizes that PID parameters may need adjustment when load changes and covers documenting final parameters (P=50, I=400, D=0.01).

This video tutorial demonstrates the fundamental programming workflow for IEC 61131-3 PLC programming using Yaskawa's MotionWorks IEC software, covering the complete cycle of creating a new project, adding function blocks (such as counters), configuring variables with proper data types and properties, compiling the program (Make), downloading code to the PLC controller, and debugging the application; it also explains key concepts including the difference between Warm Start and Cold Start, the Retain property for preserving variable values across power cycles, and the Toggle Boolean feature for interactive testing.

IEC 61131-3 is an international standard for PLC programming established by the International Electrotechnical Commission (IEC), providing a unified methodology that enables code reuse and technique sharing across different hardware platforms. The standard encompasses five programming languages: Instruction List (IL), Structured Text (ST), Function Block Diagram (FBD), Ladder Diagram (LD), and Sequential Function Chart (SFC), each suited for different programming objectives. It defines twenty data types including BOOL, INT, REAL, TIME, BYTE, and WORD, with strict data typing preventing automatic conversions. The software model organizes code hierarchically through Configurations containing Resources, which execute Tasks running Program Organization Units (POUs)—comprising Programs, Functions, and Function Blocks—where only Programs execute directly in Tasks while Functions and Function Blocks run within Programs.

PLCs use IEC 61131-3 standard with Structured Text and Function Block Diagram languages. Programs consist of Program Organizational Units (POUs) with Functional Blocks containing variables and executable code. The CAP1131 software generates C code from these definitions, compiled via GCC and uploaded over UDP/RS232. Safety-critical components are marked yellow and cannot be modified. Reverse engineering TriCore binaries required extensive scripting due to separate data/address registers, dynamic OOP table indices, and erased type information. The researchers exploited debugging functionality to extract firmware images from running PLCs.

The IEC 61131-3 standard defines five programming languages: two textual (Instruction List and Structured Text) and three graphical (Ladder Diagram, Function Block Diagram, and Sequential Function Chart). Instruction List is low-level, providing closer access to system operations. Structured Text is high-level and more human-readable. Ladder Diagram is the most widely used graphical language, considered universal for PLC programming. Function Block Diagram uses algebraic notation similar to digital logic circuits. Sequential Function Chart represents sequential control using steps and transitions, similar to state machines.

The IEC 61131-3 standard, published in 2013, defines five PLC programming languages: two textual (Structured Text and Instruction List) and three graphical (Ladder Diagram, Function Block Diagram, and Sequential Function Chart), each serving different control application needs from simple relay-like logic to complex mathematical algorithms.