After unboxing the PCBs, we began the assembly process, which was quite straightforward. We first applied solder paste to each component pad using a solder-paste dispensing syringe. For this build, we used Sn/Pb-63/37 solder paste, which melts at around 200°C.
Once the paste was applied, we gathered all the SMD components (MOSFETs, resistors, and other parts) and started placing them on the board using tweezers. For MOSFET ICs, we paid special attention to the orientation, aligning each one with the dot marked on the silkscreen to ensure they were placed correctly.
With all components positioned, we carefully lifted the PCB and placed it onto our Miniware reflow hot plate. The hot plate heats the board from below, and as the temperature gradually rises to about 200°C, the solder paste melts and securely holds every component in place.
After the reflow process, we moved on to the through-hole components. We added the two JST connectors for the motor on the bottom side of the board, along with a CON6 male header pin that allows the board to be plugged into a breadboard. Flipping the PCB over, we soldered all the JST and header pins using a standard soldering iron, completing the assembly.
2
MOTOR and POWER SOURCE
Before moving on to testing the driver, we’re using a DC gear motor rated for 3V to 12V as our test motor. For the power supply, we’re using DFROBOT’s Rainbowlink 4, which serves as a USB converter for multi-channel serial communication and power output. By connecting it to a PD adapter, we’re able to get a stable 5V to power our entire setup.
3
BREADBOARD VERSION LEVEL 1
We now move on to testing our homebrew motor driver, starting with a simple breadboard setup.
The driver is based on an H-bridge configuration, where two MOSFETs work together to control the direction of current flow. The motor’s rotation and speed are managed through two gate inputs. For Motor A, we use pins A1 and A2—setting A1 HIGH and A2 LOW spins the motor in one direction, and reversing the logic (A1 LOW and A2 HIGH) flips the rotation.
For the breadboard test, we placed two buttons on the left and two on the right side of the motor driver. Button 1 is connected to A1 and VCC, while Button 2 goes to A2 and GND. On the right side, Button 3 connects A1 to GND, and Button 4 connects A2 to VCC. Pressing Button 1 and Button 2 together completes the circuit and drives the motor in one direction. Pressing Button 3 and Button 4 instead reverses the connections, causing the motor to spin the opposite way.
4
DUAL MOTOR LEVEL 2
Next, we added the second motor to the Motor Driver Patch and placed another set of four push buttons for the B1 and B2 pins. The wiring is the same as before: two buttons on the left and two on the right. Button 5 connects to B1 and VCC, and Button 6 connects to B2 and GND. On the opposite side, Button 7 goes to B1 and GND, while Button 8 connects to B2 and VCC.
Just like the first motor, pressing Button 5 and Button 6 makes the motor spin in one direction. Pressing Button 7 and Button 8 reverses the current flow, causing the motor to rotate the other way. With this setup, we were able to confirm that both motor channels are working exactly as expected.
5
MICROCONTROLLER LEVEL 3
Now comes the main test of this whole project—pairing the Motor Driver Patch with a microcontroller. For this, I decided to use a Seeed XIAO ESP32-C6 and connected A1 and A2 to GPIO0 and GPIO1. But the moment I uploaded the test sketch and powered everything up… nothing happened.
The reason turned out to be something I completely overlooked at first: the MCU’s GPIO pins weren’t providing a high enough signal. The XIAO outputs less than 3.3V on its pins, while the MOSFETs on our driver need at least a solid 5V to properly switch their gates. So essentially, the MOSFETs never toggled, which meant the motor wouldn’t move at all.
To fix this, the humble BC547 transistor came to the rescue. I added two of them in a simple transistor-switch configuration, one transistor to pull A1 up to VCC and the other to pull A2 down to GND. With this setup acting as a level shifter, the MCU could finally control the MOSFET gates properly, and the motor driver responded exactly the way it should.
Now that this issue has shown up, it’s clear that the next revision of the Motor Driver Patch needs to include these transistors built directly into the board. That way, any MCU—3.3V or 5V—can drive the H-bridge without extra wiring or modifications.
The motor driver performed well overall, though we did encounter one limitation: the microcontroller couldn’t drive the input pins directly since it outputs only 3.3V, while the driver requires at least a 5V signal.
To solve this, we used transistors to shift the 3.3V signals up to 5V and pull the pins properly to HIGH or LOW, allowing reliable control of the MOSFETs. In the next revision of this project, we plan to integrate this transistor stage directly into the H-bridge design so the driver can be easily controlled by any microcontroller.
For now, the project is complete, and all related files and details are provided on this article page. If you run into any issues, feel free to reach out.
Arnov Sharma
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.