While we have all been hearing about the chip shortage, here is a closer look at the analog chip shortage and its potential impact.
We have all heard of the automotive chip shortage; it’s the reason some used cars are still worth double what they were a couple of years ago. Generally, when the chip shortage is discussed, even in industries outside of automotive, the discussion is around microcontroller units (MCUs). Many industrial and automotive applications rely on MCUs based on manufacturing process nodes above 40 nm, and these MCUs are built on proprietary designs that can only be manufactured at approved or contracted fabs.
These are often MCUs that are matured and even aging, but keep fueling certain industries that don’t necessarily have the resources or desire to build with newer technologies. Clearly, the chip shortage has put pressure on some of these industries to move to MCUs with more modern process nodes boasting a better supply outlook.
More advanced process nodes are often available, as the majority of the semiconductor industry has moved on to embrace these nodes and the highest volume applications. These applications include system-on-chips (SoCs), memory, and storage. Also, more advanced MCUs could also support greater centralization of systems, such as automobiles, that often require a multitude of MCUs, sensors, and analog processing chips.
Analog chip shortage
However, this move toward more advanced processing nodes isn’t necessarily viable for all chip components. There is an often-overlooked aspect of the chip shortage, the analog chips shortage, which are often built on much larger/older process nodes: 90 nm to 300 nm. Unlike MCUs, there has been even less pressure for many analog chips—such as sensors, analog signal processing, amplifiers, audio systems, and AM/FM radio technology—to move to more advanced nodes.
In fact, for many analog chips and sensors, the migration to smaller nodes will likely require complete redesigns and generation of a whole new set of IPs. This could be far beyond the cost accessibility of these technologies. Nevertheless, it appears that the demand for automotive and other industrial circuits built with these analog chips is only increasing.
Considering the ongoing electrification and sensor proliferation within automobiles and industrial electronics, availability issues of analog chips and sensors will likely only get worse. There is another aspect to this: as more MCUs move toward more advanced process nodes, there may actually be less availability and support for these legacy nodes that analog chips and sensors are built on.
Outcome of analog chip shortage
The outcome of this analog chip shortage may be manifold. One outcome is that there are additional delays in production of automotive and industrial electronics relying on these legacy analog chips and sensors. Another is that some applications that are agile enough may move toward more centralized and integrated solutions, such as more capable MCUs that have a broader suite of analog sensors and processing capability onboard.
It’s already a trend to avoid much the same issue of having to track down and source a variety of analog sensors and chips. Though the upfront cost may favor older and less expensive analog sensors and chips, using more modern MCUs equipped with a wider range of analog sensors and processing could help some applications avoid the need to redesign or struggles with replacing unavailable components.
This shift toward more modern MCUs with integrated analog sensors and features also means that there may be less need for traditional analog circuit board design talent or services as these efforts would shift to MCU implementation. Conversely, the lack of availability of certain analog sensors and chips may require additional analog circuit design services to do redesigns on legacy boards and IPs that may have been used for quite some time.
There may also be greater demand for new analog sensors and chips that operate on more available process nodes, though this may only be a short-term phenomenon as the general trend is toward heightened integration and eliminating the cost and complexity of external discrete components in many applications. It’s especially true for the growing Internet of Things (IoT) and smart sensor industries incorporating a wealth of new sensor types and use cases as well as enhanced integration and connectivity for these sensor platforms.
This article was originally published on Planet Analog.
Jean-Jacques (JJ) DeLisle, an electrical engineering graduate from Rochester Institute of Technology, moved to technical editing and writing work for an RF publication after spending six years in the industry, working as an IC layout and automated test design engineer. He writes about analog and RF for Planet Analog.