Narrative:

During arrival into gpi; WX was reported 3200 ft overcast 6 mi light rain and mist. Copilot was flying the airplane on autoplt and captain was PNF. We were given the option of a VOR 30 circle to land or the ILS/localizer runway 2 approach. We requested the ILS. The approach was set up in the FMS with angil as initial waypoint. At some point; controller cleared us direct sak which bypassed angil. There were some assumptions that upon nearing sak; the controller would provide vectors to the approach; but this was never clarified with the controller; and this was not what happened. At or near sak; the controller cleared us for the approach and to contact tower. PNF was acknowledging the controller and at the same time; the autoplt turned us inbound on to the approach and tracked the localizer rather than outbound for the procedure turn. At this point; the approach essentially became a localizer only and a descent was programmed to stop at 5600 ft MSL at glaci; the final approach fix approximately 7 mi distant. The airplane appeared not to take command for the steep descent and it was apparent that we would not be at the specified altitude of 5600 ft when we arrived at the final approach fix. We declared a missed approach somewhere outside glaci. We believe the altitude was about 7000 ft by the time we actually initiated the missed approach. At about the same time; the autoplt disengaged without command. The copilot went to go around power. When the autoplt disengaged; the airplane severely pitched up and we began an aggressive climb in excess of 3000 FPM with a rapidly decreasing airspeed. It took both of us holding the control yoke forward in order to maintain airspeed while the pitch trim was manually reset. During this period of time; we flew through the missed approach hold altitude of 8500 ft and didn't get the airplane level until approximately 11000 ft. We reported the altitude discrepancy to the controller and that there had been an autoplt malfunction. We requested vectors for the ILS. The controller cleared us to descend to 10000 ft and this time; provided vectors back towards the approach intercept. Initially the autoplt would not reengage. The copilot was hand-flying the airplane while the PNF performed the autoplt reset procedure. We reengaged the autoplt and took time to assess the situation before intercepting the localizer and GS. When we intercepted the localizer and GS; the autoplt initiated a normal approach down the GS; however; we disengaged the autoplt at glaci and flew to a normal landing. Post analysis showed that; there were 2 sak fixes in the FMS flight plan; one before the procedure turn and one after. When we were cleared to sak; the PNF selected the second sak fix when the controller cleared us direct; which eliminated the pt phase of the approach. We don't know exactly what caused the autoplt to select such a high pitch load; but it was probably caused by excessive use of the vertical speed control while attempting a steep descent. Once the missed approach was executed and the airplane pitched up; the critical point was that we stopped the airspeed from degrading and got the pitch/climb under control. Unfortunately; this wasn't done within the altitude specified; partially because we were already near that altitude and partially because we were more focused on stabilizing the airplane. Assessing improvement for the future; it is important to evaluate the approach better and when waypoints are listed multiple times in the FMS to assess the effect of selecting one over the other. Never assume what the controller is doing when he clears us off of a published waypoint. When we turned inbound; we should have immediately aborted the approach rather than continue as a non-precision approach; advised center; and started over. Although there was some delay; we correctly abandoned the approach and when the trim was out of pitch; the priority was to fly the airplane.

Google
 

Original NASA ASRS Text

Title: CE550 FLT CREW ATTEMPTS LOC RWY 2 APPROACH INTO GPI AFTER FMC PROGRAMMING ERROR OMITS PROCEDURE TURN. ENSUING MISSED APPROACH RESULTS IN EXCEEDING PUBLISHED ALTITUDE BY 2500 FEET.

Narrative: DURING ARR INTO GPI; WX WAS RPTED 3200 FT OVCST 6 MI LIGHT RAIN AND MIST. COPLT WAS FLYING THE AIRPLANE ON AUTOPLT AND CAPT WAS PNF. WE WERE GIVEN THE OPTION OF A VOR 30 CIRCLE TO LAND OR THE ILS/LOC RWY 2 APCH. WE REQUESTED THE ILS. THE APCH WAS SET UP IN THE FMS WITH ANGIL AS INITIAL WAYPOINT. AT SOME POINT; CTLR CLRED US DIRECT SAK WHICH BYPASSED ANGIL. THERE WERE SOME ASSUMPTIONS THAT UPON NEARING SAK; THE CTLR WOULD PROVIDE VECTORS TO THE APCH; BUT THIS WAS NEVER CLARIFIED WITH THE CTLR; AND THIS WAS NOT WHAT HAPPENED. AT OR NEAR SAK; THE CTLR CLRED US FOR THE APCH AND TO CONTACT TWR. PNF WAS ACKNOWLEDGING THE CTLR AND AT THE SAME TIME; THE AUTOPLT TURNED US INBOUND ON TO THE APCH AND TRACKED THE LOC RATHER THAN OUTBOUND FOR THE PROC TURN. AT THIS POINT; THE APCH ESSENTIALLY BECAME A LOC ONLY AND A DSCNT WAS PROGRAMMED TO STOP AT 5600 FT MSL AT GLACI; THE FINAL APCH FIX APPROX 7 MI DISTANT. THE AIRPLANE APPEARED NOT TO TAKE COMMAND FOR THE STEEP DSCNT AND IT WAS APPARENT THAT WE WOULD NOT BE AT THE SPECIFIED ALT OF 5600 FT WHEN WE ARRIVED AT THE FINAL APCH FIX. WE DECLARED A MISSED APCH SOMEWHERE OUTSIDE GLACI. WE BELIEVE THE ALT WAS ABOUT 7000 FT BY THE TIME WE ACTUALLY INITIATED THE MISSED APCH. AT ABOUT THE SAME TIME; THE AUTOPLT DISENGAGED WITHOUT COMMAND. THE COPLT WENT TO GAR PWR. WHEN THE AUTOPLT DISENGAGED; THE AIRPLANE SEVERELY PITCHED UP AND WE BEGAN AN AGGRESSIVE CLB IN EXCESS OF 3000 FPM WITH A RAPIDLY DECREASING AIRSPD. IT TOOK BOTH OF US HOLDING THE CTL YOKE FORWARD IN ORDER TO MAINTAIN AIRSPD WHILE THE PITCH TRIM WAS MANUALLY RESET. DURING THIS PERIOD OF TIME; WE FLEW THROUGH THE MISSED APCH HOLD ALT OF 8500 FT AND DIDN'T GET THE AIRPLANE LEVEL UNTIL APPROX 11000 FT. WE RPTED THE ALT DISCREPANCY TO THE CTLR AND THAT THERE HAD BEEN AN AUTOPLT MALFUNCTION. WE REQUESTED VECTORS FOR THE ILS. THE CTLR CLRED US TO DSND TO 10000 FT AND THIS TIME; PROVIDED VECTORS BACK TOWARDS THE APCH INTERCEPT. INITIALLY THE AUTOPLT WOULD NOT REENGAGE. THE COPLT WAS HAND-FLYING THE AIRPLANE WHILE THE PNF PERFORMED THE AUTOPLT RESET PROC. WE REENGAGED THE AUTOPLT AND TOOK TIME TO ASSESS THE SITUATION BEFORE INTERCEPTING THE LOC AND GS. WHEN WE INTERCEPTED THE LOC AND GS; THE AUTOPLT INITIATED A NORMAL APCH DOWN THE GS; HOWEVER; WE DISENGAGED THE AUTOPLT AT GLACI AND FLEW TO A NORMAL LNDG. POST ANALYSIS SHOWED THAT; THERE WERE 2 SAK FIXES IN THE FMS FLT PLAN; ONE BEFORE THE PROC TURN AND ONE AFTER. WHEN WE WERE CLRED TO SAK; THE PNF SELECTED THE SECOND SAK FIX WHEN THE CTLR CLRED US DIRECT; WHICH ELIMINATED THE PT PHASE OF THE APCH. WE DON'T KNOW EXACTLY WHAT CAUSED THE AUTOPLT TO SELECT SUCH A HIGH PITCH LOAD; BUT IT WAS PROBABLY CAUSED BY EXCESSIVE USE OF THE VERT SPD CTL WHILE ATTEMPTING A STEEP DSCNT. ONCE THE MISSED APCH WAS EXECUTED AND THE AIRPLANE PITCHED UP; THE CRITICAL POINT WAS THAT WE STOPPED THE AIRSPD FROM DEGRADING AND GOT THE PITCH/CLB UNDER CTL. UNFORTUNATELY; THIS WASN'T DONE WITHIN THE ALT SPECIFIED; PARTIALLY BECAUSE WE WERE ALREADY NEAR THAT ALT AND PARTIALLY BECAUSE WE WERE MORE FOCUSED ON STABILIZING THE AIRPLANE. ASSESSING IMPROVEMENT FOR THE FUTURE; IT IS IMPORTANT TO EVALUATE THE APCH BETTER AND WHEN WAYPOINTS ARE LISTED MULTIPLE TIMES IN THE FMS TO ASSESS THE EFFECT OF SELECTING ONE OVER THE OTHER. NEVER ASSUME WHAT THE CTLR IS DOING WHEN HE CLRS US OFF OF A PUBLISHED WAYPOINT. WHEN WE TURNED INBOUND; WE SHOULD HAVE IMMEDIATELY ABORTED THE APCH RATHER THAN CONTINUE AS A NON-PRECISION APCH; ADVISED CTR; AND STARTED OVER. ALTHOUGH THERE WAS SOME DELAY; WE CORRECTLY ABANDONED THE APCH AND WHEN THE TRIM WAS OUT OF PITCH; THE PRIORITY WAS TO FLY THE AIRPLANE.

Data retrieved from NASA's ASRS site as of January 2009 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.