How many stories can a picture tell?
A year ago, AirAsia flight QZ8501 fell from the sky following an engine stall. All 156 passengers and six flight crew were killed.
In December, Indonesia’s National Transportation Safety Committee issued a report fingering, in part, cracked solder joints on the plane’s rudder control warning system as responsible for a warning system failure that might have misled the pilot.
“The degraded performance and ambiguous commands might have decreased the SIC’s (co-pilot) situational awareness, and he did not react appropriately in this complex emergency, resulting in the aircraft becoming upset,” the report read. (See the full story, including photos, here.) Tragically, the same malfunction occurred nine times alone in the month of the crash.
Publication of the report was juice for industry engineers on the IPC TechNet listserv for several days, where discussion ranged from observations of cracked joints and poor wetting to a lack of redundancy on mission-critical parts to oxidation. The possibility that lead-free solder could have been used – the craft was an AirBus 320 built in 2008 – added a bit of intrigue as well.
Looking over the report’s photos, there is a clear crack on the right side of the solder joint. It is known that plating can come off whenever rework takes place or if a board is heated too long. If the module was reworked several times, as the report suggests, it means it was reheated and solder reflowed, and naturally some flux would be added each time.
All of which could be an indicator that an intermetallic compound was growing, and the joint was losing its strength as a result, suggests Jim Raby.
The report suggests the module shown was the one involved in the crash. What is not known from the information published is the degree to which the cracking or any concurrent solder embrittlement problems due to intermetallic compound growth were involved in the module’s failure.
Raby, of course, is a soldering guru, with nearly 60 years of manufacturing high-reliability electronics under his belt. He is credited for developing the NASA and Navy (the famous China Lake) soldering schools, and was instrumental in developing the IPC soldering certification curriculum, used by the vast majority of the industry today. Although he retired in 2014, Raby serves as a living reminder that experience is a great teacher.
In a phone conversation, he recalled his introduction (in 1958) to the plated through-hole, compliments of an IPC meeting attended by many technology leaders of the day, including Ford Aeronautics and Philco. “I saw a bare copper board, since they weren’t plated with gold at first. We would dissolve the copper, and it would continue to grow IMCs. We would grow IMCs to eat up the copper in the PTH, and the termination would become brittle and crack very quickly.”
As the crash report makes clear, the intermittent failures tied to the rudder control module made up only one piece of the puzzle. Multiple mistakes, including pilot error, led to the tragic outcome.
Even with the investigation completed, there’s much that might still be learned here. Over time, OEMs have cut back on engineering staff and put the onus on contractors to get everything right. For Class 3 systems, that’s an increasingly uncomfortable situation.
To save on costs, many airlines use third-party specialists to perform maintenance. Shops have been set up in Mexico (Delta), China (United), and El Salvador (US Airways and Southwest), and reports on the subject indicate most of the mechanics in these new “inspection hubs” lack the necessary FAA certifications or even the language skills to communicate problems to the proper authorities. Granted this accident involved an offshore carrier, but every major player in the industry faces similar overhead circumstances.
In all likelihood, those performing critical QA checks and maintenance on these key systems have little to no knowledge of solderability or plating or dissolution rates.
With all due respect to the labs that performed the post-crash testing and analysis, it be would interesting to get our hands on the affected module or others from the same lot. Imagine the possibility of subjecting them to thermal cycle testing and ageing to determine whether similar outcomes could be created.
Readers, if you have access to or contacts at the AirBus suppliers who made the modules, please consider contacting me. Perhaps from this tragedy some wisdom can be gained.
PCD&F/CIRCUITS ASSEMBLY; firstname.lastname@example.org editor in chief of