One of my jobs as an engineer was working in the engineering department of an electronic contract manufacturer. Our department designed test equipment for the manufacturing lines, but we also assisted customers with their products issues.
Do you have a memorable experience solving an engineering problem at work or in your spare time? Tell us your Tale
The product being built on the line was a PCB assembly for a coffee maker. One day, the boss came to me and said the customer is getting some complaints about the coffee maker display and controls locking up. He assured the customer we could fix their problem (even though no one knew what the issue was).
The first task was to reproduce the problem. With no clue as to how this problem happened, I started by just letting the coffee makers run. After a few days, nothing—so I tried other things like banging it and shaking it…nothing. I then pushed the buttons in every combination and cadence I could come up with – still no luck. Next ,I tried varying the line voltage slowly from the specified minimum to the specified maximum. The coffee maker still worked fine. I was running out of ideas. Finally, I tried one last thing: I plugged it into a controller that turns the line voltage on and off at varying rates. After a while it locked up.
This behavior hinted at the micro’s reset circuit so I dug into that. It wasn’t the typical design used in those days, a simple RC circuit. The resistor tied to Vcc and the capacitor tied to digital ground, then the other ends of the resistor and capacitor were tied together, and that tied to the micro’s reset.
After a little more testing I concluded that it was due to a fast interruption in the line voltage or a brownout. Either would cause the reset capacitor to discharge partway and then to reset back to Vcc. The micro wasn’t happy with this as it lost power but didn’t get a valid reset.
The boss was happy I found the root cause but now said “fix it” and the fix had to be something easy to tack on to the existing PCB. I spent a few days looking at things like changing the resistor or capacitor value or adding a 555 timer, comparator, or op-amp, but these either didn’t work or were too difficult or expensive to add to the PCB.
That’s when I remembered an obscure device I had read about, a programmable unijunction transistor, or PUT. This has some properties like an SCR. The PUT has 3 pins: anode, cathode, and gate. So, the circuit I came up with was this:
The schematic of the coffee maker fix, introducing the PUT to successfully manage failures due to interruptions in the line voltage or brownouts.
When Vcc is good, the anode to cathode is not conducting, but when the Vcc drops the capacitor stays up for a while, but the PUT triggers if the gate drops below the anode by 0.7v or more. This trigger turns on the anode to cathode path and the capacitor is quickly discharged. This fixed the problem and the PUT and resistor were easily tacked onto the existing PCB.
Epilogue: After the fix I went to the boss and asked if we could apply for a patent for this reset circuit but he turned me down. So as consolation to myself I submitted it to EDN’s Design Ideas column, which they printed. Later in my career I happen to notice that this Design Idea was referenced as prior art in patents from Texas Instruments, Dallas Semiconductor, Shlumberger Technology, and Ericson Inc. Looks like a good patent opportunity we missed. But then again, what’s better; a patent or an EDN Design Idea being published?
Damian Bonicatto is a consulting engineer with decades of experience in embedded hardware, firmware, and system design. He holds over 30 patents.
Phoenix Bonicatto is a freelance writer.
Related Content
- Audible ohmmeter cuts down on false readings
- Analog of a thyristor with a controlled switching threshold (comments)
- Exploring software-defined radio (without the annoying RF) – Part 1
- A beginner’s guide to power of IQ data and beauty of negative frequencies – Part 1
The post PUT a reset in its place appeared first on EDN.