August_EDFA_Digital

edfas.org ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 20 NO. 3 14 Third, temperature response of ceramic capacitors can vary widely. The ceramic dielectric within the part determines whether the capacitance value will be stable over a wide temperature range. While the C0G dielectric formulationmaintains a stable valuewithin1percent from - 55° to 125°C, the Y5V formulation loses over 80 percent of its 25°C value when the temperature reaches 100°C. A counterfeiter can realize a profit be selling a Y5V dielec- tric as aC0Gpart. A fewof the reported suspect counterfeit ceramic capacitors were noted as having failing capaci- tance value over temperature, which suggests those parts were the wrong dielectric. Fourth, capacitor characteristics such as voltage and dielectric are not easily detected during receiving inspection. While value and dissipation factor are com- monly measured, voltage is not an easy characteristic to verify. Determining whether a 10V capacitor has been illegally substituted for a 50V part might not be possible without submitting a sample toanovervoltage test tonote whether any parts fail. Checking for the wrong dielectric might require testing the parts at elevated temperature. Neither test is commonly performed at receipt. Finally, system failures due to counterfeit ceramic capacitors may be particularly hard to detect. Because many ceramic capacitors serve to filter out noise or unwanted voltage spikes, the failure mode for a counter- feit capacitor is likely to be an intermittent glitch or reset of an IC. These failure modes are notoriously diƒicult to troubleshoot andmay bemore likely to result in changing the integrated circuit rather than identifying the faulty capacitor. COMMON TEMPERATURE RANGES FOR COUNTERFEITING Counterfeit electronic parts are of particular concern when the application is mission or safety critical. In general, parts intended for higher temperature range applications tend towards higher criticality. There are exceptions, but most would agree that electronic parts tested and used inmilitary or automotive applications are more likely to be critical than parts for commercial and to some extent industrial applications. In addition, parts in these applications are usually more expensive than their commercial counterparts, always an important factor in the counterfeiter’s businessmodel. Finally, itmakes sense for a counterfeiter to falsify a lower temperature range (i.e., commercial) marking in order to reap the higher cost of the high temperature range part. So can we identify any trends in counterfeiting for these temperature ranges? ERAI’s reporting data on integrated circuits since 2012 was sampled for temperature range, and those parts classified as commercial (0° to 70°C), industrial ( - 40° to 85°C), automotive ( - 40° to 125°C), and military ( - 55° to 125°C). Many part temperature listings did not fit these ranges perfectly, so they were categorized into the closest applicable range. The results are as follows: Data in Table 1 was compared with IC usage estimates taken fromStatista and IC Insights, which estimatemilitary tem- perature range usage of only 1 percent, and commercial temperature range applications using over 70 percent of ICs. Basedon this information, it appears that even though they comprise only a small portion of suspect counterfeit integrated circuits, a military temperature range part is much more likely to be counterfeit than a commercial or automotive part—perhaps as much as 15 times likelier. Industrial temperature range parts are also more likely to be counterfeited. COMMONLY COUNTERFEITED MANUFACTURERS The data can also be analyzed to determine which original component manufacturers’ (OCM) parts are most commonly counterfeited. It must be noted that the manufacturers arenot aparty to counterfeiting. Therefore, the presence of an electronic part manufacturer on this list should in no way be construed negatively for the manufacturer. The data presented in Fig. 5 for 2012 through 2017 is stunning, in that one manufacturer’s parts are reported much more frequently than the rest. The chart captures RECENT TRENDS IN COUNTERFEIT ELECTRONIC PARTS (continued from page 12) Table 1 Reported integrated circuits by temperature range, 2012 – 2017 Temperature range type Range Percent of reported ICs, 2012‒2017 Commercial 0° to 70°C 41% Industrial - 40° to 85°C 45% Automotive - 40° to 125°C 3.7% Military - 55° to 125°C 11%

RkJQdWJsaXNoZXIy MjA4MTAy