Your experience with the Rotax engine is helpful to know, and the google search link is a better compilation on the issue than I've pulled together in one place. Thanks.mrburnette wrote: ↑Mon Jan 15, 2018 12:26 amFaraday shielding is perfect if the shield is perfect, but "perfect" does not occur in deployment. I have done some work with an experimental aircraft with a Rotax engine. Faraday shielding and aggressive bypassing has been shown to work adequately.
Google search for best practices...
For the aircraft, I selected the AVR Mega at 5.0 Volts.
It might be that the non AEC-Q100 rated MCUs and others would perform fine in the auto engine compartment environment with the shielding, but it would be seen as a "weak point" of the design in terms of being a commercial product. If I was doing this for some of my other, non-commercial, work (such as putting instruments on unmanned aircraft for climate research), then I'd certainly look into that option.
Digi-Key still has a supply of a QFN-package version of the AEC-Q100 ATMega328P but they're no longer stocking the TQFP package I'd been using, and the aftermarket parts manufacturer I'm working with doesn't want to get locked into using a microcontroller series that has been discontinued. I modified my board to accept the QFN, and I'm trying to convince them that it would be years (if ever) before they sold enough of the ECUs to use up the existing supply of QFN versions, but they're not buying that argument.