The global semiconductor manufacturing process is a model of efficiency and cost reduction. Accordingly, the U.S. outsources most of its chip-making activities, which is why the country produces just 12 percent of the world’s computer chip supply.
This highly efficient process started to break down at the start of the pandemic. It has yet to recover. An unfortunate side effect of the shortage has been the rise of counterfeit chips.
Off-shore manufacturing makes it difficult to ensure the integrity and quality of the produced semiconductor chips used heavily in U.S. commercial and defense electronics. It is critical to know that all of these chips are genuine and manufactured to specifications and not recycled or counterfeit.
The Pentagon estimates that about 15% of all purchased parts are counterfeit. The Electronic Reseller Association International (ERAI) suggests the number of counterfeit electronic products in circulation is increasing, and industrial businesses lose approximately $250 billion each year, notes a blog from Manufacturing.net. These costs don’t even consider the human cost of compromised chips, resulting in loss of life, income, and safety to our highly electronic-based world.
Testing For Counterfeit Chips
There are three basic ways to test for counterfeit or reused chips. The first is to use embedded hardware security primitives and sensors. Primitives can be both hardware and software-based. Hardware-based security primitives employ instance-specific and process-induced variations in electronic hardware as a source of cryptographic data.
Hardware primitives include physical unclonable functions (PUFs) and true random number generators (TRNGs) to produce device-specific electronic fingerprints and random digital signatures. These fingerprints and signatures are used to generate cryptographic keys and IDs commonly used for device authentication, cloning prevention, etc. Further, designs, such as combating die and IC recycling (CDIR) sensors, offer additional countermeasures against counterfeits.
The second fundamental way to test for counterfeit electronics is via the electrical testing of the chip, which checks for performance abnormalities in its electrical behavior and other indicators of tampering.
The third test is, in many ways, the most basic. It requires both an external and internal physical inspection of the chip. Counterfeit chips can have visible burn markings, discoloration, blacktop coating, dents, tooling marks, and numerous other defects on the exterior. A high-intensity microscope is often used to examine each chip down to the micron level to determine if it's safe or hacked.
Internally, the compromised chip is tested using a 3D x-ray microscope to examine it layer by layer. Clues of tampering are often very subtle and require a trained eye and precise equipment to identify.
An example of a detailed test is captured in a blog from the Electronic Reseller Association International (ERAI). In this blog, five compromised chips pass all tests until the decapsulation analysis, in which cap or encapsulating material was removed from the packaged integrated circuit by mechanical, thermal, or chemical means.
This analysis revealed that a sample part showed an unknown logo die marking and 956 markings not traceable to the labeled manufacturer or the specified part number. The die markings and die topography did not match those of a known good sample. (Image Source: ERAI, Inc. (www.erai.com))
New Testing Methods
Recently, researchers at the Florida Institute for Cybersecurity (FICS) Research suggested a new way to detect recused and counterfeit electronic parts, especially chips. The technology holds promise in making supply chains more secure, protects consumer safety, and runs at almost zero cost, claim the researchers.
The basis for the new test method focuses on one ubiquitous component—low-dropout regulators, or LDOs. The researchers found that measuring how the LDO regulators respond to variations in the input presented to the chip’s power supply yielded an aging signature that could be analyzed to determine if the chip is a recycled-type counterfeit or not.
This approach is made more potent with machine learning to automate the data analysis and better understand the sources contributing to aging.
John Blyler is a Design News senior editor, covering the electronics and advanced manufacturing spaces. With a BS in Engineering Physics and an MS in Electrical Engineering, he has years of hardware-software-network systems experience as an editor and engineer within the advanced manufacturing, IoT and semiconductor industries. John has co-authored books related to system engineering and electronics for IEEE, Wiley, and Elsevier.