Narrative:

Flying a visual approach, autoplt off, autothrottles on. Small turboprop failed to turn off runway and we were given a go around at about 75 ft AGL. Throttles had gone into retard mode. Toga buttons were pushed, FMA showed go around and throttles advanced. At approximately 600 ft AGL we were given a left turn to 210 degrees then a clearance to 3000 ft. The copilot was setting the heading and I reached to set 3000 ft. Just as I set 3000 ft the copilot said 'watch your airspeed (attitude/anu degrees was normal) lower the nose.' I saw the airspeed decelerating through 120 KTS (vref flaps 28 degrees and 134 KTS). I smoothly lowered the nose and reached for the throttles. They were retarded to the idle stops and retard was on the FMA. Normal power was applied and a normal uneventful recovery was achieved. I had to turn the autothrottles off to get out off retard mode. Callback conversation with reporter revealed the following information: reporter captain idented aircraft as MD83. During the landing phase, prior to the far, the autothrottles retarded correctly. When the go around was initiated, reporter says that the FMA showed go around and autothrottles advanced. Reporter captain admits that he uses a technique which may have contributed. When in a descent, landing, for example, he spins the altitude selector up to a higher altitude so as to eliminate altitude warnings. Thus, during a go around, when setting the go around altitude, it is necessary to spin the altitude knob back down to the assigned altitude. So, reporter thinks that may have contributed, but doesn't know how or why. Also, reporter followed up with discussion with his company and other pilots and nobody had heard of a similar event on the MD83. Because the autothrottles go to retard on landing when 75 ft from the ground, and that function is tied to the radar altimeter, reporter thinks a radar altimeter malfunction might have caused the idle position. Supporting that contention was the FMA showing retard. With the throttles at idle and in a climb, the airspeed dropped rapidly and the flight crew got a stick shaker prior to advancing the throttles.

Related reports:

Google
 

Original NASA ASRS Text

Title: ACFT EQUIP PROB. AUTOTHROTTLES WENT TO IDLE DURING A GAR. AIRSPD DECREASED TO STICK SHAKER.

Narrative: FLYING A VISUAL APCH, AUTOPLT OFF, AUTOTHROTTLES ON. SMALL TURBOPROP FAILED TO TURN OFF RWY AND WE WERE GIVEN A GAR AT ABOUT 75 FT AGL. THROTTLES HAD GONE INTO RETARD MODE. TOGA BUTTONS WERE PUSHED, FMA SHOWED GAR AND THROTTLES ADVANCED. AT APPROX 600 FT AGL WE WERE GIVEN A L TURN TO 210 DEGS THEN A CLRNC TO 3000 FT. THE COPLT WAS SETTING THE HEADING AND I REACHED TO SET 3000 FT. JUST AS I SET 3000 FT THE COPLT SAID 'WATCH YOUR AIRSPD (ATTITUDE/ANU DEGS WAS NORMAL) LOWER THE NOSE.' I SAW THE AIRSPD DECELERATING THROUGH 120 KTS (VREF FLAPS 28 DEGS AND 134 KTS). I SMOOTHLY LOWERED THE NOSE AND REACHED FOR THE THROTTLES. THEY WERE RETARDED TO THE IDLE STOPS AND RETARD WAS ON THE FMA. NORMAL PWR WAS APPLIED AND A NORMAL UNEVENTFUL RECOVERY WAS ACHIEVED. I HAD TO TURN THE AUTOTHROTTLES OFF TO GET OUT OFF RETARD MODE. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: RPTR CAPT IDENTED ACFT AS MD83. DURING THE LNDG PHASE, PRIOR TO THE FAR, THE AUTOTHROTTLES RETARDED CORRECTLY. WHEN THE GAR WAS INITIATED, RPTR SAYS THAT THE FMA SHOWED GAR AND AUTOTHROTTLES ADVANCED. RPTR CAPT ADMITS THAT HE USES A TECHNIQUE WHICH MAY HAVE CONTRIBUTED. WHEN IN A DSCNT, LNDG, FOR EXAMPLE, HE SPINS THE ALT SELECTOR UP TO A HIGHER ALT SO AS TO ELIMINATE ALT WARNINGS. THUS, DURING A GAR, WHEN SETTING THE GAR ALT, IT IS NECESSARY TO SPIN THE ALT KNOB BACK DOWN TO THE ASSIGNED ALT. SO, RPTR THINKS THAT MAY HAVE CONTRIBUTED, BUT DOESN'T KNOW HOW OR WHY. ALSO, RPTR FOLLOWED UP WITH DISCUSSION WITH HIS COMPANY AND OTHER PLTS AND NOBODY HAD HEARD OF A SIMILAR EVENT ON THE MD83. BECAUSE THE AUTOTHROTTLES GO TO RETARD ON LNDG WHEN 75 FT FROM THE GND, AND THAT FUNCTION IS TIED TO THE RADAR ALTIMETER, RPTR THINKS A RADAR ALTIMETER MALFUNCTION MIGHT HAVE CAUSED THE IDLE POS. SUPPORTING THAT CONTENTION WAS THE FMA SHOWING RETARD. WITH THE THROTTLES AT IDLE AND IN A CLB, THE AIRSPD DROPPED RAPIDLY AND THE FLC GOT A STICK SHAKER PRIOR TO ADVANCING THE THROTTLES.

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.