Altium Business Card Challenge

In January 2026, a challenge was posted on LinkedIn→: design a business card with a twist — make it a PCB! I'd already drawn out my logo for my portfolio, but I thought it'd be cool to make it light up. Just having LEDs turn on seemed boring, though; I wanted them to flash somehow. But integrating an MCU and programming it for a circuit design challenge seemed overkill. Besides, I wanted to learn something new!

So instead I designed a 5-stage transistor loop oscillator on a breadboard with RC circuits in between for timing, sketched out a schematic for 7 stages (because who doesn't want more lights?), went through part selection, and laid out the PCB to the left. 

A significant challenge at the beginning of this process was the startup of the loop. Often, when I'd hold the button to power the circuit, all LEDs would fade up into a half-on state, which I later learned was a result of the loop being too symmetrical; without a push at one end, the stable state operation wasn't oscillating at all. I could force start oscillation by shorting the collector of one transistor to GND or the base of one to PWR (or several other potentially questionable practices), but that wouldn't fly on a PCB. But after I incorporated startup asymmetry created by an extra capacitor and a diode on the first stage, oscillation started from the get-go. 

Another issue I had was power — I wanted the business card to be portable so I could demo it at job fairs, so naturally that meant battery power. But I couldn't hardly strap a pack of AAs to the back. Using a coin cell seemed like a logical move to keep things relatively flat but still mobile, but that had its own problems with voltage sag and higher internal resistance. I did what I could to minimize current draw by using green LEDs (since the human eye sees green best, they can be dimmer and still appear bright) and fairly high series resistors (2.2k). Designing so the card is only lit when a button is held helps to extend battery life, but also makes draw from the battery more of a pulse rather than extended; during short demos, the drop in speed of oscillation is essentially unnoticeable since during pulses battery sag isn't nearly as bad.

The battery holder is the only component on the back of the board and is hand-solderable so manufacturing would be cheaper. Traces from the battery are thicker and go solely to the button, so the circuit can only ever be powered when the button is held. The power plane is shaped to give visual dimension to my scarab decal, but also is contiguous, doesn't narrow, and is fully functional to power the loop.