He deleted the transistor. Replaced it with a different model from the vast TI library. He added a 1-picofarad capacitor across the feedback path—a trick he’d seen on a forum at 3 AM during an all-nighter.
He thought about his father. A man who fixed tractors with baling wire and a wrench. His father didn't believe in "simulation." He believed in grease under the fingernails and the smell of ozone from a live wire.
He hit send.
He hit the "Run" button.
He opened his email. He attached the file and the required 10-page analysis. He typed the subject line: "Project 3 Submission – Leo Chen." multisim student
It was enough to learn that a diode drops 0.7 volts. It was enough to understand that a Zener works in reverse. It was enough to fight a timestep error for four hours and win.
Leo zoomed in on the circuit. The problem was a feedback loop around the transistor. In the real world, it would work. But in the sterile, mathematical womb of Multisim, the virtual electrons were panicking. They were simulating infinite acceleration, dividing by zero in a digital panic attack. He deleted the transistor
But Leo loved the picture. He loved the blue glow of the oscilloscope probes in the software. He loved that he could change a resistor value from 100 ohms to 1k ohm with a single click and watch the waveform dance. It was clean. It was predictable. It was safe.