Narrative:

While under control of approach, we were vectored onto the localizer for the localizer runway 27 approach. Once cleared for the approach, all aircraft indications were normal until shortly just prior to 'reebo' intersection. Approximately 1 mi east of 'reebo' intersection (which has an intermediate altitude of 2000 ft) our aircraft's automation malfunctioned, and with autoplt engaged it began a spontaneous descent. We disconnected the autoplt to regain altitude, but saw that our aircraft had inadvertently descended to approximately 1600 ft MSL, 400 ft lower than the published altitude to reebo. We returned the aircraft to the profile for the localizer approach manually and the rest of our descent and landing was uneventful. At no time did our controllers mention the momentary deviation, and at no time did we present a threat to traffic or terrain. WX was VFR and we corrected immediately. This experience has caused me to be much more vigilant of the airbus's automation, particularly during non precision approachs. I will continue to aggressively monitor its automation and more immediately intervene when encountering deviations in the future. Callback conversation with reporter revealed the following information: the pilot reported that he has heard of several A319 uncommanded dscnts. This company's non precision approach procedure is to set a flight path angle (fpa) prior to the descent point, and at .3 mi prior, activate the descent then set the missed approach altitude in the FCU altitude window. In no case should the aircraft begin a descent out of the FCU altitude set unless either fpa or open descent are selected. When asked if the crew has selected an RNAV back up in the mcdu approach page, the reporter stated no, only the localizer runway 27. The RNAV would have allowed 1800 ft at the OM and would have automatically begun a descent.

Google
 

Original NASA ASRS Text

Title: AN A319 ON A NON PRECISION APCH BEGAN AN UNCOMMANDED DSCNT PRIOR TO THE INITIAL APCH FIX DSCNT POINT.

Narrative: WHILE UNDER CTL OF APCH, WE WERE VECTORED ONTO THE LOC FOR THE LOC RWY 27 APCH. ONCE CLRED FOR THE APCH, ALL ACFT INDICATIONS WERE NORMAL UNTIL SHORTLY JUST PRIOR TO 'REEBO' INTXN. APPROX 1 MI E OF 'REEBO' INTXN (WHICH HAS AN INTERMEDIATE ALT OF 2000 FT) OUR ACFT'S AUTOMATION MALFUNCTIONED, AND WITH AUTOPLT ENGAGED IT BEGAN A SPONTANEOUS DSCNT. WE DISCONNECTED THE AUTOPLT TO REGAIN ALT, BUT SAW THAT OUR ACFT HAD INADVERTENTLY DSNDED TO APPROX 1600 FT MSL, 400 FT LOWER THAN THE PUBLISHED ALT TO REEBO. WE RETURNED THE ACFT TO THE PROFILE FOR THE LOC APCH MANUALLY AND THE REST OF OUR DSCNT AND LNDG WAS UNEVENTFUL. AT NO TIME DID OUR CTLRS MENTION THE MOMENTARY DEV, AND AT NO TIME DID WE PRESENT A THREAT TO TFC OR TERRAIN. WX WAS VFR AND WE CORRECTED IMMEDIATELY. THIS EXPERIENCE HAS CAUSED ME TO BE MUCH MORE VIGILANT OF THE AIRBUS'S AUTOMATION, PARTICULARLY DURING NON PRECISION APCHS. I WILL CONTINUE TO AGGRESSIVELY MONITOR ITS AUTOMATION AND MORE IMMEDIATELY INTERVENE WHEN ENCOUNTERING DEVS IN THE FUTURE. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE PLT RPTED THAT HE HAS HEARD OF SEVERAL A319 UNCOMMANDED DSCNTS. THIS COMPANY'S NON PRECISION APCH PROC IS TO SET A FLT PATH ANGLE (FPA) PRIOR TO THE DSCNT POINT, AND AT .3 MI PRIOR, ACTIVATE THE DSCNT THEN SET THE MISSED APCH ALT IN THE FCU ALT WINDOW. IN NO CASE SHOULD THE ACFT BEGIN A DSCNT OUT OF THE FCU ALT SET UNLESS EITHER FPA OR OPEN DSCNT ARE SELECTED. WHEN ASKED IF THE CREW HAS SELECTED AN RNAV BACK UP IN THE MCDU APCH PAGE, THE RPTR STATED NO, ONLY THE LOC RWY 27. THE RNAV WOULD HAVE ALLOWED 1800 FT AT THE OM AND WOULD HAVE AUTOMATICALLY BEGUN A DSCNT.

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.