Zhuojin Pipe Fitting Co., LtdARM-Based Optimal Control System for Pulsed MIG Welding Current Regulation
Literature Overview
This paper by Ding Gaojian, Zhu Jinhong, Shi Hongxin, and Dai Leyi from Henan University of Science and Technology, published in Welding (2011, No. 4, pp. 49–52), presents a digital control system for pulsed MIG welding based on an ARM development board. The system implements timing control, digital PWM generation, interrupt protection, and constant-current closed-loop control using Bang-Bang (on-off) control theory instead of conventional PID regulation.
System Architecture and Control Methodology
ARM-Based Digital Control Platform
The system architecture centers on an ARM microcontroller development board, which provides the computational platform for real-time control of the pulsed MIG welding power source. The main functional modules include:
| Functional Module | Description | Implementation Method |
|---|---|---|
| Timing Control | Generates pulse-on and pulse-off periods | Timer interrupt with microsecond resolution |
| Digital PWM | Produces switching signals for IGBT/MOSFET | Software-based PWM with configurable duty cycle |
| Interrupt Protection | Responds to fault conditions (overcurrent, overvoltage) | Hardware interrupt with priority levels |
| Current Closed-Loop Control | Maintains constant welding current | Bang-Bang (switching) control algorithm |
Bang-Bang Control vs. PID Control
The most notable technical contribution is the use of Bang-Bang control instead of the traditional PID algorithm for current regulation. Bang-Bang control operates on a simple binary logic: if the measured current is below the setpoint, the output is at maximum; if above, the output is at minimum. This approach offers several advantages in the context of pulsed MIG welding:
- Faster response time — The binary switching eliminates the proportional-integral-derivative calculation delay, enabling near-instantaneous correction of current deviations.
- Simpler implementation — The algorithm requires minimal computational resources, which is advantageous for embedded ARM processors with limited processing power.
- Robustness to disturbances — The high-gain nature of Bang-Bang control makes the system inherently resistant to load disturbances such as arc voltage fluctuations and wire feed variations.
Pulse Parameter Control
In pulsed MIG welding, the welding process is characterized by two distinct current levels: the pulse current (I_p) during the on-time (T_on) and the background current (I_b) during the off-time (T_off). The ratio of pulse current to background current, the pulse frequency, and the duty cycle collectively determine the droplet detachment behavior and weld pool dynamics. The ARM-based system allows independent and precise control of each of these parameters, enabling the operator to optimize the welding process for specific materials and geometries.
Engineering Practice Integration
Application Scenarios
This type of digital control system is particularly valuable in the following engineering contexts:
- Repair welding of dissimilar metals — Precise pulse parameter control allows the operator to adjust heat input in real-time when welding across material boundaries.
- Thin-gauge welding — The ability to fine-tune pulse parameters enables high-quality welds on thin aluminum or stainless steel without burn-through.
- Robot welding integration — The digital control interface facilitates communication with welding robots and CNC systems, enabling adaptive welding strategies.
Performance Characteristics
The experimental results confirm that the system operates reliably with flexible control of all major functions. The current closed-loop control achieves fast and accurate regulation of welding current, and the algorithm satisfies the real-time control requirements of the welding process. The response time of the Bang-Bang controller is typically in the range of 1–5 milliseconds, which is adequate for the dynamic behavior of the MIG welding arc.
Key Questions and Reflections
While the Bang-Bang control approach offers simplicity and fast response, it has inherent limitations. The binary nature of the control signal can lead to current oscillation around the setpoint, which may cause micro-splatter and arc instability in certain welding conditions. For applications requiring extremely smooth current waveforms, such as welding high-reactivity metals like titanium or welding in high-vacuum environments, a more sophisticated control algorithm may be necessary. Additionally, the paper does not address the long-term reliability of the ARM-based system under industrial conditions, including electromagnetic interference (EMI) from the welding arc and thermal stress on the electronic components. For industrial deployment, the system would require additional hardware protection such as isolated power supplies, EMI shielding, and thermal management. Nevertheless, the paper represents a significant step toward affordable, programmable welding power sources that can be customized for specific welding applications.