Really good long-form article from IEEE Spectrum about semiconductors aging and how to improve measurements.
You know when it’s time to get a new car. Your odometer is far into six digits, perhaps the engine is burning lots of oil, or the transmission is growling. Fixing all that might well cost quite a bit more than your ancient vehicle is worth.
But what about your microprocessor? Unlike automobiles, microprocessors don’t have convenient little gauges that reflect how much wear and tear they’ve endured. And wear they do—though you’ll probably never notice it. The degradation of their transistors over time leads slowly but surely to decreased switching speeds, and it can even result in outright circuit failures.
You generally don’t perceive this deterioration because semiconductor companies always play it very safe—they set the clock-speed rating of their microprocessors so conservatively that almost every one of their products will continue to operate flawlessly throughout its intended lifetime. That strategy works. But it’s kind of like never taking your Ferrari out of the slow lane because you’re concerned that its engine might throw a rod 10 years down the road.
We and other researchers are trying to improve that situation. One critical aspect of the work we did at the University of Minnesota was to develop better ways to study the different physical mechanisms of transistor aging. Today semiconductor engineers measure those aging effects primarily by examining transistors one at a time, using microscopic electrodes to probe a silicon wafer. The necessary equipment can cost tens of millions of dollars, and probing transistors individually is arduous when you’re trying to gather many observations. Sometimes you can’t do those measurements well, no matter how much time you spend.
With better ways to measure transistor aging, chipmakers could let their microprocessors run faster—appreciably faster—than they do now. In the future it might even be possible to use these techniques to build circuits into microprocessors that continuously measure the subtle effects of aging and adjust clock frequency or operating voltages so that the transistors, old or new, could always run at peak speeds.
Stop breadboarding and soldering – start making immediately! Adafruit’s Circuit Playground is jam-packed with LEDs, sensors, buttons, alligator clip pads and more. Build projects with Circuit Playground in a few minutes with the drag-and-drop MakeCode programming site, learn computer science using the CS Discoveries class on code.org, jump into CircuitPython to learn Python and hardware together, TinyGO, or even use the Arduino IDE. Circuit Playground Express is the newest and best Circuit Playground board, with support for CircuitPython, MakeCode, and Arduino. It has a powerful processor, 10 NeoPixels, mini speaker, InfraRed receive and transmit, two buttons, a switch, 14 alligator clip pads, and lots of sensors: capacitive touch, IR proximity, temperature, light, motion and sound. A whole wide world of electronics and coding is waiting for you, and it fits in the palm of your hand.
Get the only spam-free daily newsletter about wearables, running a "maker business", electronic tips and more! Subscribe at AdafruitDaily.com !
Transistors (bipolar or MOSFET) are made by diffusing dopants into a silicon crystal substrate. Careful control of temperature and time determine the dimensions of this doped volume, which in turn establishes the transistor’s electrical characteristics. It seems logical to me that elevated temperatures in operation may cause a continuation of the diffusion process for some devices. Thus, I have always followed the Engineer’s Rule of Thumb, “If you can’t hold your thumb on it, it’s too hot.” I suspect that this is a correlate of another Engineer’s Rule of Thumb that device life doubles for every 10 degrees C that you lower its temperature. I have always assumed that semiconductor life is partially a function of the temperature x time product. Therefore, I endeavor to design what I build, or modify things I repair, to run cool. A lifetime of adhering to this practice seems to have given me the reputation of reliable designs and repairs. Be cool!
Much ado about nothing, because at the end of the day what does this buy system designers? Not much in my opinion.
The point of the original paper leans toward maximizing the process limit of a device library. Good – for the manufacturer of course. Perhaps instead of using a higher voltage process in order to get that clock frequency or transconductance, maybe a lower voltage (or mature – read cheaper) process can be utilized thereby saving die area and cost. But this doesn’t mean anything for a system designer.
Guardbands exist for a reason and that is to ensure that the company hits maximum yield from the wafer. The guardbands come from the characterization and reliability data taken over temperature, voltages, loads and whatever other operating conditions required of the device – usually dictated by the application. The MIN, TYP, and MAX specifications you see in a datasheet are always within this guardband.
The hope, of course, is "how much more can I push it?" Easy, you don’t when you want to adhere to minimal failures. This would NOT change even with the manufacturer stating new limits.