Powering the rest of the electronics should be relatively simple, as they all simply plug into the power distribution WAGOs; however, it actually turned out to be one of the hardest parts of putting the electronics assembly together. 

I started with Pi’s power module, which would convert the power of the 6S battery to 5V, 5A, and plug in via USB-C. I used some adhesive strips to stick the module onto the underside of the assembly, near the WAGOs. 

Then, I had to figure out a way to fit all of the ESCs into the remaining space. After fiddling around with various configurations for a while, I found that the best way to fit them all would be to arrange them in a sort of semi-circular pattern around the edges of the tube, taking advantage of every last bit of empty space left. I also made sure to label each with a number 1 through 5, making sure to plug each signal wire into the corresponding port on the Pixhawk. Ultimately, it looked a bit like this:

Now, theoretically, all that needs to be done is the final connections between the motors and the ESCs, but before I do that I first needed to verify that everything was connected, functioning, and tuned. 

So I used some crimps that I had on hand to create a simple connection between the motors and the ESCs, fixed the Y-splitter cable for the batteries, and prayed as I plugged the batteries in for the first time.

Although there was a slight spark, nothing exploded, and I could hear the Raspberry Pi’s fan whir up as all the systems came to life. 

After everything went online, I used my computer to connect to the Raspberry Pi over the internet, and opened up BlueOS’s online control panel. In reality, the whole process of figuring out how to work the controlling of the drone took a really long time, almost two nights worth of running into problem after problem, I’ll condense the details here.

First, I tested each of the 5 motors manually, using sliders to gradually ramp up the power. Depending on whether the direction of the motor was what I wanted, I would click the “reverse” option to reverse the direction of the motor; for example, since the propellers on my two main thrusters were mirrored, I had to reverse the direction of one of the motors so that they would both provide thrust in the same direction.

Then, I had to go to the parameters section to change a couple of key settings, primarily the ROV’s frame setting and joystick gain. I set the frame to the BlueROV1 setting, which has a similar motor configuration to my own drone, and turned the maximum joystick gain down a little bit, meaning that the joystick wouldn’t be able to turn the motors to maximum throttle (which would be too overkill for underwater). Furthermore, turning down the maximum throttle would make testing much safer for me and quieter for my neighbors. 

Finally, I would make my way to the actual ground control station, called Cockpit. Here, I would be able to hook up my controller to my computer, map any controls, and finally control the motors. Because I set the frame to the preset, the joystick controls to control motors were already mostly set up, with the left joystick controlling forward-backward thrust, and the right controlling pitch and yaw. I also bound the disarm button, which disables the drone, to a couple of the extra buttons on the controller. 

In the next post, I’ll go into a bit more detail about the issues I ran into while tuning and how I solved them.