I originally designed the control board in Fritzing. I went through a few designs before settling on the single board concept that I ended up sticking with.
Designing and Planning
I wanted to leverage the simplicity of an Arduino, but have the strength of a soldered PCB, so I decided to build the “Arduino on a bradboard” circuit on a solderable PCB. As much as I wanted to, I didn’t have time to actually design and etch a PCB, so I picked up two solderable breadboards on eBay. I picked up a pack of four ATMega 328P microcontrollers, with the IC sockets, crystals, and capacitors required to get a basic circuit up and running.
The circuit is very simple actually.
- The ATMega gets is power in accordance with this great video
- The rightmost pin on each MOSFET gets connected to ground
- The center pin on each MOSFET gets connected to the grounds of the LED tape (one MOSFET per color)
- The leftmost pin on each MOSFET gets connected with a 10k ohm resistor to its PWM pin on the ATMega
- The ground of the 12 volt supply is connected to the ground rail on the board
- The +12 Volt of the 12 volt supply is connected to the +12 Volt terminal on the LED tape
The two LEDs are not actually imperative to the circuit, but one LED indicates that the board is powered up (it is tied directly to the +5v rail and ground), and the other is tied to Arduino pin 13, which will blink when the chip is flashed with the “Blink” sketch. This allows us to verify that the chip is functioning properly. It is really useless once you get the MOSFETS going, but can be useful as you are putting everything together.
While working on this project, I was in the shop Friday after Del Val let out for about 8 hours, and Saturday for about 9 hours. The first time I installed the board was a Friday evening. I connected the outer circuit. The outer circuit consisted of the outer two channels of three strips of LED tape. Getting this lit up and fading was a beautiful moment. We finally felt that everything we had worked toward was coming together, and we finally had something to show for what we had been working on.
I spent the rest of the night working on cleaning up wiring, getting ready to connect the inner loop to the board, and working on the second board.
Blowing the Board
The next morning, we were all particularly looking forward to getting the inner loop connected and lit up. Excitement was high all around. I connected the ground leads to the second set of MOSFETS, connected the +12 Volt lead to the battery, turned on the board, and *POOF*. A small puff of smoke rose from the 328, signaling that it was fried. I tried to figure out what had happened and was immediately burned when I touched one of the MOSFETs. I unplugged the board power supply and felt the 328 – it was hot. Excitement turned to concern and then disappointment. Now I was a week behind where I was just a few minutes before – not good. Not realizing that there was a short in one of the MOSFETS I put another 328 into the socket to see what would happen. It fried too. At this point I knew that my board was shot, and was trying to figure out what to do about it. For a while, I started to finish building the second control board I had started the day before.
After some time of working on the second board, I realized that we hadn’t tested the inner loop like we had tested the outer loop. I went over to the jukebox and touched the individual wires to the battery. They sparked. This signified to me that there was a short somewhere along one of our nearly 1000 solder joints in the inner loop. Unfortunately, not only one of the colors sparked – they all did. Now I had a broken board, and multiple solder bridges creating a shorts to ground in my lighting strips. Ugh.
The Next Week
The next Thursday that I came home, I finished rebuilding the original board, and successfully tested it with my testing rig. The next day (Friday), I carefully connected it to the outer loop, and turned it on as I held my breath. It lit up. It felt good to at least see some progress in the right direction, but the daunting task of finding all of the shorts in the inner circuit was still ahead.
I had Dante start looking for solder bridges on that inner circuit while I started to get the wire ready for another set of lights for the record player compartment. He found two, and I found a third when he went to take his dinner break. After testing each of the colors, we determined that we had repaired all of the shorts. As Borat would say, “Great Success!”
It was time to connect the inner circuit to the second set of MOSFETS. This was probably the most stressful moment of the entire build process. I connected the battery to the connector, and the board to +5 Volts. Another landmark – it lit up.
This was huge for us. All of our designing, soldering, wiring, and taping so far had led to this. It was beautiful. Dante had made excellent progress on the record player assembly (as seen in the picture), the speaker grille was nearly done, and progress was being made on the button panel, paint job, button panel, and extra trim pieces. I think it was at this point that I actually felt like we were going to deliver this on time. I spent the rest of the night working in the auditorium on installing my fog machines in the stage.