The electronics industry is out of balance, putting too many resources into a few product sectors at the expense of others.
"Smaller, faster, cheaper, better" is the electronics industry's mantra. It has given us worldwide communications literally in your pocket, the Web in your laptop, and more music than you could ever listen to in the palm of your hand.
But the creation of these wonders depends on increasingly smaller-scale, complementary metal-oxide semiconductors (CMOS), deep submicron chips operating at +5, +3.5, or even below +1.8V. (You've noticed how much longer your cell phone runs on a single charge? The lower operating voltage is a big part of the reason why.) Single-supply, low-voltage CMOS is where the R&D money has been going, all right. But it's left other market segments out in the cold. The industry mantra is, after all, a relative statement. What constitutes smaller for an MP3 player is very different than smaller for an MRI scanner.
The industrial instrumentation electronics market, for example, benefits from CMOS mainly in the digital domain; however, medical monitors, factory automation, control loops in electrically noisy workplaces need relatively higher voltage bipolar signals (±10V typically) to cope with the environment. Many electrical testers and medical devices rely on ±5V operating systems as well. They, too, have been left behind.
High-voltage bipolar and CMOS processes typically use 5-micron gate lengths and geometries—many generations behind what's running your new digital camera. The problem this presents to industrial-segment electronics designers is that it's very difficult to take advantage of the cost reductions that benefit pure digital, submicron designs. Analog signal conditioning needs to keep pace with the digital. Sure, there are more chopper amplifiers with low drift, but they don't handle the large signals required for factory automation. Adding digital circuitry to an old, large-geometry, high-voltage analog process results in chips that quickly get quite large, and therefore more costly. Where are the state-of-the-art higher-voltage, lower-cost industrial instrumentation chips? Only a few companies are responding to this need but, in fact, much of the technology developed for submicron chips can be used to produce chips that meet the requirements of industrial and medical manufacturers and adhere to the "smaller, faster, cheaper, better" mantra.
Clearly, "smaller, faster, cheaper, better" has been interpreted by many of the companies designing semiconductors as a technology challenge for digital ICs relative to the needs of portable consumer and communications devices. Where are the improvements to benefit higher-voltage applications?
With globalization in full swing and pressures on manufacturers unrelenting, industrial instrumentation applications must be able to move into the modern age. They can only get there if chip designs and manufacturing processes bring industrial components into the modern age.