The Incident of the Perpetual Drying Cycle

An early-model, electronically controlled spin dryer never seems to finish drying and engineers worry they might have to scrap the circuit and start over

By Mike Metzger

About 20 years ago and fresh out of college I got a job with a small electronics firm where I was designing a control circuit for a laundromat spin dryer.

The circuit used about 30 old-fashioned combinational logic gates to lock the lid when the dryer was spinning, start and stop by pushbuttons, light a run light (we hadn’t been taught microcontrollers), it included a timer to time the drying cycle. The client wanted selectable time intervals for different applications so I chose a programmable timer with timing intervals selectable by dip switches.

It worked fine on my breadboard, but on the production PC boards the timer was horribly inaccurate. On short time cycles (< 1 minute) it was okay. But the longer the chosen cycle, the less accurate it became, always longer than what was programmed. On the longest interval (supposed to be about 10 minutes) it NEVER seemed to finish.

I checked and re-checked my wiring, added bypass capacitors on all the power pins, put the scope on my signals to make sure they were clean, and mostly began to panic since changes to the circuit at this point would be expensive in cost and schedule.

Finally at my wits end I asked a fellow worker to look at my circuit. He was not a design engineer, but was a self-proclaimed “blood and guts repairman” who knew how to use his troubleshooting tools. After turning the intensity on the scope trace way up, starting the timer and looking at the board for a few minutes he immediately noticed a flicker on the trigger lamp on my scope.

“Hey, you’ve got a trigger here,” he told me. I pointed out the probe wasn’t even hooked to anything, but still wondered why it would have just triggered like that. I quickly traced it to my heat-controlled soldering iron that cycled about every minute… and then got the “a ha” moment.

The timer reset pin had been left unterminated due to a late design change. The small voltage spike from the soldering iron was enough to trigger reset of the timer. Due to the temperature cycle of the iron, this phenomenon only showed up on the longer cycle times, with the shorter cycles able to complete before the iron went through its temperature swing.

Contributing Writer Mike Metzger graduated from the University of Toledo with a BSEE in 1988. Since then he’s worked in electronics design, industrial equipment sales, consulting engineering, even TV production. He now works as a control systems engineer for an oil refinery.