Even if the investigation from the AAIB is still in progress and the preliminary report is expected to be released in about one month, new theories are emerging about the BA038 crash landing’s root cause. Since a simultaneous engine failure is extremely unlikely, a software glitch affecting the engine control system is among the possible causes being investigated.
An interesting article dating back to October 2006 (http://www.iasa-intl.com/folders/belfast/BA777_Unthrustworthy.htm), focuses on errors introduced with a FADEC (Full Authority Digital Engine Control) software update that affected the B.777 equipped with GE90 engines. The article reports about two thrust rollback recorded on the 777-300ERs that suffered the failure during take off (and 5 occurred in flight). Subsequent troubleshooting found that the rollbacks were caused by a glitch in the software of the FADEC and that the reductions “only likely to occur at reduced powers”. The article explains that the flawed software was installed after a FADEC software update.
So, the FADEC has already caused worries to the B.777 operators using GE engines. The British Airways aircraft was equipped with RR Trent engines, even if the software used to control them is probably the same (or mostly similar). Even discarding the possibility that the current software may still cause Loss Of Thrust Control or LOTC for the same flaw (the AD was issued in 2006 and by now the software should have been patched), the above mentioned article provides also details dealing with an Airworthiness Directory applied to the GE90 engines that confirms the risks of corruption of the FADEC signals because of clogging of the sensors feeding the engine control system. The GE90 engines incorporate now a design modification aimed to prevent signal corruption but what about Trent engines?
For pictures and a more in-depth analysis, I suggest visiting this link that was provided from a visitor: http://www.iasa.com.au/folders/Safety_Issues/others/The_BA038_LikelyCause.html