Counterfeit Electronic Components Identification: A Case Study
Counterfeit electronic components are finding their way into today's defense electronics. The problem gets even more complex when procuring DMS (diminishing manufacturing source) parts. This paper will provide a brief introduction to counterfeit prevention and detection standards, particularly as they relate to the aerospace and defense sector. An analysis of industry information on the types and nature of counterfeit components will be discussed to illustrate those most likely to be counterfeited, followed by a specific case at a major defense contractor.
The case involved two circuit card assemblies failing at test, whereby their root cause for failure was identified as "unable to write specific addresses at system speeds." The error was traced to a 4MB SRAM received from an approved supplier. Fifteen other suspect parts were compared with one authentic part directly purchased from a supplier approved by the part manufacturer. Defects or anomalies were identified but not enough to unequivocally reject these parts as counterfeit as the defects could have also happened in the pre-tinning process, which is a program-specific requirement if the parts were stored for more than three years. Through the subsequent analysis, subtle differences between the authentic and suspect parts were identified and isolated. The methodologies and process chosen to identify counterfeit parts will be reviewed and an assessment of the results will be presented along with the defects found in relation to the defect types reported in relevant test standards.
The Defense Federal Acquisition Regulations DFARS 252.246-7007 Contractor Counterfeit Electronic Part Detection and Avoidance System defines a counterfeit part as:
An unlawful or unauthorized reproduction, substitution, or alteration that has been knowingly mismarked, misidentified, or otherwise misrepresented to be an authentic, unmodified electronic part from the original manufacturer, or a source with the express written authority of the original manufacturer or current design activity, including an authorized aftermarket manufacturer. Unlawful or unauthorized substitution includes used electronic parts represented as new, or the false identification of grade, serial number, lot number, date code, or performance characteristics.
Highlights for the DFARS Case 2012-D055 final rule include:
• Applying requirements to the acquisition of electronic parts and assemblies containing electronic parts, including commercial items (COTS)
• Defining “Counterfeit” and “Suspect counterfeit”, is limited to electronics, including embedded software and firmware
• The costs of counterfeit electronic parts or suspect counterfeit electronic parts and the cost of rework or corrective action that may be required to remedy the use of inclusion of such parts are unallowable (unless electronic parts were provided as GFE and timely notice of discovery was provided by contractor)
Based on the highlights for the ruling and the impact that counterfeit parts could have on the performance of fielded systems, it should be obvious in terms of the importance of understanding, identifying and addressing suspect counterfeit parts in the aerospace and defense industry. Although the current definition and ruling applies to electronics, the expectation is the definition will eventually broaden to include nonelectronics (i.e. optics, mechanics, MEMs, and materials). Therefore, a robust process to ensure parts that are received and used in systems to support the aerospace and defense industry is paramount to not only the business and industry, but to the users of the products that rely on these systems, especially the warfighter.
Counterfeit Parts Business is a Multibillion Dollar Industry
The discussion of recognizing that counterfeit parts have been introduced into the supply chain is not new, with various companies, and technical journals publishing as early in 2002.
In a 2006 article published by Pecht and Tiku and noted in the UK Electronics Alliance (UKEA) position paper, “UKEA Position on Counterfeit Electronic Components”:
Alliance for Grey Market and Counterfeit Abatement (AGMA), based in the USA, estimates that, in 2006, up to 10% of technology products sold worldwide are counterfeit, which amounts to $100 billion of sales revenues. However, this does not take into account consequential losses. In 2007, the US Patent and Trademark Office estimated that total ‘counterfeiting and piracy (activity) drains about $250 billion out of the US economy each year and 75,000 jobs’.
A primary driver of counterfeit parts has been part scarcity, or diminishing manufacturing source and material supply (DMSMS). Realizing that as the consumer market began to grow exponentially in the 1980s and 1990s, the supply base for manufacturing parts rated for military and high-reliability applications was having a difficult time keeping up with demand, and part availability was becoming more difficult. These market forces drove the opportunity to introduce counterfeit parts into the supply chain through ‘gray market electronics brokers’. According to a 2001 article on fake parts, “One U.S. independent distributor, which asked to remain anonymous, said it paid a broker in China $70,000 for 1206 case-size ceramic capacitors about three months ago. The 90-cent parts, which under less-constrained market conditions would have cost 20 cents, slipped through two quality inspections before arriving on the OEM's production floor".
Bad Parts are not Always Counterfeit
It is important to recognize that, just because there are anomalies identified on electronic parts, it does not signify that the parts are counterfeit. It does, however, require the incoming inspection organization to assume the responsibility to make initial determination as to whether there is enough evidence to suggest the parts from a lot or shipment should be evaluated for additional anomalies. Three important points to consider when creating a system to screen for counterfeit parts are:
• They are not easy to identify, even with sophisticated analytical methods
• They are in the supply chain, even with authorized distributors
• They are more of an issue with obsolete parts
Background on Case Study
During functional test of control module boards used in a multiple sub-array of a testable antenna, two boards failed. The root cause for the failures was identified as "unable to write specific addresses at system speeds." When diagnosing the issue, it was narrowed down to an SRAM that was supplied by an electronics part broker. The parts in question were procured from the broker, an approved diminishing material supply (DMS) supplier, due to unavailability from a franchised distributor of the original components manufacturer (OCM). When reviewed by the internal Failure Review Board, it was determined that a comparison of SRAM parts supplied by the broker should be compared with parts from the distributor to determine if there were any observable differences in the parts.
Figure 1: Comparison of two SRAM parts with different lot numbers.
Analysis Approaches and Techniques
A total of seven different methods which ranged from nondestructive to destructive were used to make a determination about the SRAM parts being suspect counterfeit. Any individual analysis does not make a clear case on its own merits. However, to make a legal case for suspect counterfeit, enough due diligence is necessary.
The following outlines the seven analyses used to make the case:
1. Visual inspection by optical microscopy
4. Scanning acoustic microscopy
6. Electrical test
7. Discussions with OCM
Visual Inspection by Optical Microscopy
Once the failure occurs on a component or subsystem, typically there is an optical inspection to determine if there was any physical damage to the part either before or during testing. Damage can occur from a variety of sources including handling, testing conditions and setup, foreign object damage or debris (FOD), fixturing, etc. Figure 1 shows a comparison of an SRAM received by an authorized distributor and the broker in question. It was noted that the lot number of the broker part was not in the OCM database.
Figure 2: Lead and mold inspection. Different mold interface and pin width.
In and of itself, this does not constitute a smoking gun, but it does inspire one to continue the investigation. Upon further visual inspection, it appeared the workmanship, or quality of the part around the leads suggested a difference in mold processing (Figure 2). Because visual inspection is subjective and directed by any given customer requirements, incoming inspection (5-10X at AQL) easily can miss the inconsistencies. This is especially true when suspect counterfeit parts are mixed in the same delivery packaging and 100% inspection is not performed.
Finally, there was a measurement of pin width between the two different leads. The leads from the distributor parts were on the order of 14.5 mils wide, whereas the lead width from the broker parts was 12 mils. The difference led to the next step in the investigation, namely X-ray.
To read this entire article, which appeared in the July 2017 issue of SMT Magazine, click here.