Copy that

A copy machine causes trouble for two engineers chasing phantom oscillations.

John E Rogers, Test Engineer -- EDN, November 11, 2011

After exhausting every bypass, decoupling and damping scheme we had in our collective engineering toolbox, we reluctantly called in the “big gun,” the department’s principal engineer, for his advice. My colleague and I were confident that the senior engineer would have no better luck than we did in discovering the root of the mysterious, random oscillations, but we thought we would give it a try anyway. We explained all the design objectives, set the attenuated input signal level to the nominal frequency of 20 kHz, and monitored the amplifier output with a 456B scope we had in the lab. We all stared at a clean-looking sine wave for several minutes with no oscillations occurring. My co-worker and I explained that we could sit for hours before noticing the troublesome anomaly and that it lasted only a few seconds before vanishing again.

This Tale is a runner-up in EDN’s Tales from the Cube: Tell Us Your Tale Contest, sponsored by Tektronix. Read the other finalists’ entries here.

After quietly looking at the scope trace for a little while longer, our senior engineer said he would make a copy of the schematic and run some analysis. We gave him the drawing we had been working from and he made his way to the copy machine. While he was gone, we looked at the scope and wondered when the next random burst of oscillations would occur. A few seconds later, there it was, a two second burst of oscillation followed by a normal signal and then another two second burst. We couldn’t believe it! No sooner had the department engineering guru left the room than the circuit was mocking us again!

The principal engineer returned and told us that he had made two copies of our schematic, one for himself and a spare for us because the original had some obvious wear and tear (coffee stains, pencil notes, etc). My fellow engineer and I looked at each other in one of those eureka moments—“Two copies!”


The copy machine was located on the other side of the lab’s back wall where we had our bench setup. It seems that the electrostatic corona wire drive circuit of the copier was the source of our phantom oscillations. The amplifier front end was pushed into oscillation whenever RF from the copier was emitted, which lasted for about two seconds during the copy cycle. Over the past week, we had been chasing a ghost that would seem to appear randomly, only to find out it was coincident with the use of the copy machine! Our embarrassment was probably justified—we should have looked for external causes—however we ensured that adequate RF shielding was incorporated in the equipment that was going to manufacturing. Manufacturing had several copy machines on the floor.