Mechanical Design, Machine Design

[Hero shot]

1.1 Assignments of the Week

Mechanical Design (part 1 of 2) Machine Design (part 1 of 2)

Here's our production process and final demonstration:

1.2 Design and Planning:

We brainstormed a lot of ideas, including an automatic T-shirt printing robot, screen printing, and more. Eventually, we decided on a 3-axis CNC machine because our lab currently needs a small CNC for educational purposes. It needs to be compact, simple, and portable, making it suitable for teaching middle and high school students. So, we decided to build this 3-axis CNC.

Goal Setting:

The CNC machine is mainly intended for engraving on circuit boards and wood, with a 150x150 mm engraving platform.

Division of Responsibilities:
  • Siyuan Fan is responsible for mechanical design.
  • Ziyi Zhu handles electronic control and software debugging.
  • I am in charge of assembly and all records.
  • 材料准备表单

    1.3 Materials Preparation

    Mechanical Materials
    Materials numbers
    4
    2
    4
    1
    2
    4
    4
    16
    1
    1
    n
    1
    1

    Electronic Materials
    Materials numbers
    4
    1
    1
    1
    1
    1

    1.4 Mechanical Design

    Frame Structure - Primarily constructed with aluminum profiles and 3D-printed parts.

    Since we aim to make a compact machine, a moving Y-axis setup would increase the distance required, so we adapted a gantry-style movement instead.

    The base is framed with aluminum profiles on all four sides, enclosed by 3D-printed parts that allow space for the lead screws, aluminum profile platform, bearings, etc. To maximize the internal workspace, we positioned the motor externally. The spindle section uses pre-made injection-molded parts, connected to the gantry by custom 3D-printed components.

    For further details, please refer to Siyuan Fan's documentation.

    Specifications

    1.5 Electronic Control

    We use the Makerbase MKS LS ESP32 PRO control board.

    Specification Description
    Microcontroller Equipped with an ESP32 chip (32-bit, dual-core) that supports both Wi-Fi and Bluetooth connectivity.
    Input Voltage Supports a DC power input ranging from 12V to 24V, suitable for various power requirements.
    Output Interfaces
    • X, Y, Z Axis Control: Three-axis control interface for stepper motors.
    • Laser/Spindle Control Interface: PWM output support for adjusting laser intensity or spindle speed.
    • Limit Switch Interface: Limit switch interface for each axis, enabling homing operations.
    Communication Interfaces
    • Wi-Fi and Bluetooth: Allows users to wirelessly upload G-code.
    • USB Interface: Enables direct connection to a computer.
    • Driver Interface: Compatible with standard stepper drivers (e.g., A4988, DRV8825, TMC2208), with four driver slots available.
    Driver Interface
    • Compatible with standard stepper drivers (e.g., A4988, DRV8825, TMC2208), with four driver slots available.
    Firmware Pre-installed with open-source GRBL_ESP32 firmware, compatible with popular laser engraving and CNC software.
    SD Card Interface Includes an SD card slot for offline operation with G-code files.

    Advantages

  • Wireless connection via Wi-Fi and Bluetooth.
  • Processor supports more complex and high-precision engraving tasks.
  • Disadvantages

  • Poor heat dissipation; suitable only for low-power devices.
  • Might be challenging for users unfamiliar with the ESP32.
  • Connect stepper drivers and motors, configure parameters, and set up software. For detailed steps, refer to Ziyi Zhu's link .

    Conduct tests and debugging: Test each stepper motor's smooth movement, axial alignment, and rotational accuracy. Perform individual tests.

    Testing a single motor:

    Testing two motor:

    Everything seems fine...

    1.6 Assembly - My Part

    This week, I was mainly responsible for the assembly.

    The project was quite complex, so we planned ahead and purchased the materials in advance. However, we still encountered several issues, such as receiving the wrong lead screws and delayed arrival of the couplings.

    I've already listed the material list above, and we started installing from the bottom.

    The installation of the bottom frame went smoothly until we reached the lead screws. We had originally bought three, but the seller only sent two, and one of them was too short. With time running out, we decided to repurpose a lead screw from an idle CNC machine in the lab, but we found that the lead screws had different pitches. In the end, we had no choice but to use the incorrect short lead screw for the lower motor.

    It was a bit too short, so we tried using a bearing to secure it in the middle and shortened the lead screw a little to fit the coupling. Then we fixed the bearing with hot glue - it seemed to work! However, after testing, the lead screw wobbled a lot.

    After thinking for a while, We remembered that the lab had some double-pass hexagonal copper pillars. So, I found four M3*30 copper pillars, which helped alleviate the issue of the lead screw being too short.

    With the main problem resolved, we continued installing the rest of the components, and the assembly was finally completed!

    For more assembly process, please watch the video at the beginning

    Hero Shot:

    1.7 Milling Test

    We used Candle for path generation, as we were trying out some more user-friendly, easy-to-use software for generating paths, making it more convenient for students.

    Here's a tutorial video on using Candle:

    The tool alignment on this machine is currently manual. After completing the tool alignment in Candle, we upload the path and can see the interface below, which synchronously displays the milling progress. It's very intuitive!

    It works very well, it's really exciting! (Here we used a piece of wood for testing, and the instructor said to use PCB testing later, but the first version has been dismantled... We will show the new machine test later... Please look forward to it)

    But there are still some areas for improvement

  • 1.The cutting marks are rough.
  • 2.The noise was loud, and the vibration was intense.
  • 3.There was no automatic Z-axis calibration.
  • 4.The lack of an enclosure made the cutting process dangerous due to exposed parts.

  • Four months later — 【QIAO-JIANG】 was born.

    We underwent a major redesign and started over.

    We tested it with a sound synthesizer PCB.The cutting process is very stable

    The PCB turned out beautifully! The precision of the cutting path is exceptional, and the overall finish is incredibly clean and sharp. After vacuuming, the surface is spotless, with every trace and detail perfectly visible. It's truly an outstanding piece of work!

    This equipment is now sold commercially (18 units sold), and detailed structural information cannot be disclosed at this time. Our team is developing the second generation, and once it's complete, we will open-source the first generation! Stay tuned...

    1.8 Design files

    3D-W10 cnc