Narrative:

FMS was programmed with route of flight with normal EFIS indications prior to takeoff. Takeoff was accomplished in pitch and heading modes. Autoplt was engaged passing 2000 ft MSL, and climb and flight path progressed normally in heading mode. Clearance was given to proceed direct to an en route fix, at which time FMS was programmed for direct routing and navigation mode was selected. Navigation solution for direct routing presented on mfd appeared invalid, as it indicated an initial turn away from present course. As the copilot and I discussed this abnormality, it became apparent the autoplt was not responding to the navigation mode command, but was still proceeding in heading mode. Correct switchology was confirmed, FMS was reprogrammed, and navigation mode again selected, but similar results were obtained. In an attempt to comply with our clearance, the heading bug was used to intercept our assigned route, but it became apparent an intercept heading was causing us to diverge even further from our FMS course. En route charts were quickly consulted for airway frequencys and courses, CDI's were crosschecked for functionality, and a course was flown which began to reintercept our routing. Suspicions turned to our FMS-displayed heading, which turned out to differ 50 degrees from the magnetic compass. During this troubleshooting process, the aircraft failed to audibly annunciate or capture the pre-selected altitude of FL270 (as indicated on our pfd's, which also turned out to be inaccurate when compared to our standby altimeter), causing a possible altitude deviation. The autoplt was disconnected and the aircraft hand flown to maintain FL270, as indicated on the standby altimeter, and course, as indicated on our CDI's. Subsequent investigation revealed a dual INS failure with associated loss of valid air data computer indications, which was apparently caused by a single integrated avionics computer failure corrupting 2 theoretically independent system. Numerous fault warnings should have accompanied such a failure, but none were displayed at any time before or during these multiple failure modes. The possible altitude deviation might have been prevented by closer attention to (all) the altimeters, but when all instrument indications are suspect and multiple system are partly or wholly inoperative, prioritization becomes very challenging! Callback conversation with reporter revealed the following information: it was determined that the map displays were accurate and manual slaving of the compass could be done. The remainder of the flight was completed using VOR navigation. The manufacturer participated in troubleshooting in the air. After arrival. It was determined that the #2 integrated avionics computer was at fault. During follow-up testing, the crew was told that the accuracy limit for the standby compass with power on is +/-29 degrees with electric power on and +/-10 degrees with power off. The compass on the aircraft was near these limits which made navigation difficult when coupled with high en route xwinds. No failures were indicated on EICAS. The 2 system should compare with each other for accuracy but there was no 'misscompare' warning given in spite of the failure.

Google
 

Original NASA ASRS Text

Title: AN INTEGRATED AVIONICS COMPUTER FAILURE IS FOUND TO BE THE CAUSE OF ERRONEOUS INDICATIONS ON BOTH PLT'S FLT DISPLAYS AND FAULTY OP OF THE AUTOPLT SYS ON A CE750. USE OF STANDBY INSTRUMENTATION WAS REQUIRED.

Narrative: FMS WAS PROGRAMMED WITH RTE OF FLT WITH NORMAL EFIS INDICATIONS PRIOR TO TKOF. TKOF WAS ACCOMPLISHED IN PITCH AND HDG MODES. AUTOPLT WAS ENGAGED PASSING 2000 FT MSL, AND CLB AND FLT PATH PROGRESSED NORMALLY IN HDG MODE. CLRNC WAS GIVEN TO PROCEED DIRECT TO AN ENRTE FIX, AT WHICH TIME FMS WAS PROGRAMMED FOR DIRECT ROUTING AND NAV MODE WAS SELECTED. NAV SOLUTION FOR DIRECT ROUTING PRESENTED ON MFD APPEARED INVALID, AS IT INDICATED AN INITIAL TURN AWAY FROM PRESENT COURSE. AS THE COPLT AND I DISCUSSED THIS ABNORMALITY, IT BECAME APPARENT THE AUTOPLT WAS NOT RESPONDING TO THE NAV MODE COMMAND, BUT WAS STILL PROCEEDING IN HDG MODE. CORRECT SWITCHOLOGY WAS CONFIRMED, FMS WAS REPROGRAMMED, AND NAV MODE AGAIN SELECTED, BUT SIMILAR RESULTS WERE OBTAINED. IN AN ATTEMPT TO COMPLY WITH OUR CLRNC, THE HDG BUG WAS USED TO INTERCEPT OUR ASSIGNED RTE, BUT IT BECAME APPARENT AN INTERCEPT HDG WAS CAUSING US TO DIVERGE EVEN FURTHER FROM OUR FMS COURSE. ENRTE CHARTS WERE QUICKLY CONSULTED FOR AIRWAY FREQS AND COURSES, CDI'S WERE XCHKED FOR FUNCTIONALITY, AND A COURSE WAS FLOWN WHICH BEGAN TO REINTERCEPT OUR ROUTING. SUSPICIONS TURNED TO OUR FMS-DISPLAYED HDG, WHICH TURNED OUT TO DIFFER 50 DEGS FROM THE MAGNETIC COMPASS. DURING THIS TROUBLESHOOTING PROCESS, THE ACFT FAILED TO AUDIBLY ANNUNCIATE OR CAPTURE THE PRE-SELECTED ALT OF FL270 (AS INDICATED ON OUR PFD'S, WHICH ALSO TURNED OUT TO BE INACCURATE WHEN COMPARED TO OUR STANDBY ALTIMETER), CAUSING A POSSIBLE ALTDEV. THE AUTOPLT WAS DISCONNECTED AND THE ACFT HAND FLOWN TO MAINTAIN FL270, AS INDICATED ON THE STANDBY ALTIMETER, AND COURSE, AS INDICATED ON OUR CDI'S. SUBSEQUENT INVESTIGATION REVEALED A DUAL INS FAILURE WITH ASSOCIATED LOSS OF VALID ADC INDICATIONS, WHICH WAS APPARENTLY CAUSED BY A SINGLE INTEGRATED AVIONICS COMPUTER FAILURE CORRUPTING 2 THEORETICALLY INDEPENDENT SYS. NUMEROUS FAULT WARNINGS SHOULD HAVE ACCOMPANIED SUCH A FAILURE, BUT NONE WERE DISPLAYED AT ANY TIME BEFORE OR DURING THESE MULTIPLE FAILURE MODES. THE POSSIBLE ALTDEV MIGHT HAVE BEEN PREVENTED BY CLOSER ATTN TO (ALL) THE ALTIMETERS, BUT WHEN ALL INST INDICATIONS ARE SUSPECT AND MULTIPLE SYS ARE PARTLY OR WHOLLY INOP, PRIORITIZATION BECOMES VERY CHALLENGING! CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: IT WAS DETERMINED THAT THE MAP DISPLAYS WERE ACCURATE AND MANUAL SLAVING OF THE COMPASS COULD BE DONE. THE REMAINDER OF THE FLT WAS COMPLETED USING VOR NAV. THE MANUFACTURER PARTICIPATED IN TROUBLESHOOTING IN THE AIR. AFTER ARR. IT WAS DETERMINED THAT THE #2 INTEGRATED AVIONICS COMPUTER WAS AT FAULT. DURING FOLLOW-UP TESTING, THE CREW WAS TOLD THAT THE ACCURACY LIMIT FOR THE STANDBY COMPASS WITH PWR ON IS +/-29 DEGS WITH ELECTRIC PWR ON AND +/-10 DEGS WITH PWR OFF. THE COMPASS ON THE ACFT WAS NEAR THESE LIMITS WHICH MADE NAV DIFFICULT WHEN COUPLED WITH HIGH ENRTE XWINDS. NO FAILURES WERE INDICATED ON EICAS. THE 2 SYS SHOULD COMPARE WITH EACH OTHER FOR ACCURACY BUT THERE WAS NO 'MISSCOMPARE' WARNING GIVEN IN SPITE OF THE FAILURE.

Data retrieved from NASA's ASRS site as of July 2007 and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.