Analysis of Spindle Position Errors in a Retrofitted Three-Axis Training Unit CNC Machine using Statistical Methods

The use of Arduino is becoming more common due to its ease of programming and low cost. However, the accuracy of Arduino-based control systems when used to retrofit CNC machines has not been widely studied, especially in educational settings where many machines are no longer functional. In this study, Arduino was used to control spindle position for retrofitting a Training Unit (TU) CNC three-axis machine, which is an important tool in mechanical engineering education. The proposed system consists of an Arduino Mega connected to three TB6600 motor drivers and NEMA 23 stepper motors, providing a low-cost retrofit solution. The Arduino was programmed using the Arduino IDE. After assembling the system, experiments were carried out to evaluate the accuracy of the spindle position compared to the commanded settings. The results showed that the maximum spindle position error was measured with a 95% confidence level. Errors in the x and y axes showed an increasing trend along with the spindle position in each respective direction, while errors in the z axis remained relatively stable. Quantitatively, the maximum error increased by 24.5 m per mm on the x-axis and 20.7 m per mm on the y-axis, while the z-axis error stayed nearly constant at around 0.23 mm.