Who Turns on Their Stuff at -40 Degrees?

Upon leaving a restaurant tonight, I couldn't help but notice, as a baby surely notices a wet diaper, that my nostrils were starting to freeze as I traversed the few parking spaces necessary to reach my temporary moving shelter -- my car. Turning on the radio, the weatherman noted that it was currently -32 degrees Celsius (-26 deg. F). It hadn't been this cold in Ottawa for quite a few number of years. As I was driving home, I couldn't help but think about the cold weather and how it reminded me of -40 degrees. Minus 40 degrees is the only temperature that is the same in either Celsius of Fahrenheit. It's also the lower end of the temperature range that we typically design our integrated circuits to work under. On the high end, it's typically 125 deg. C., well above the boiling point of water.

It's not unusual, when designing circuits, to run into a problem whereby circuits that work at one temperature extreme fails at another. By adjusting circuit parameters to make it work at the failing extreme, one often finds that the original passing extreme is no longer passing. It's a frustrating problem because one must now go and design something much more complicated in order to cancel out the temperature effects. And it's a problem that has caused many an engineer to ask "who turns on their stuff at -40 degrees anyway?"

Not many. But on a day such as today, if someone left their iPhone in the car, and after dinner, decided to make a phone call using the same really cold phone, they still expect it to work. And if it doesn't, it'll be the corporation that suffers the bad internet review once the annoyed customer leaves the device on for long enough that the heat from the dissipated battery power makes 3G work on the iPhone once more.

What else needs to work at -40 degrees? All the electronics in my car, of course. Tonight, more than any other night, I thank those automotive electronics engineers who designed their circuits down to -40. (And a thanks to those mechies too!)