Narrative:

I was the captain on a flight originating at ZZZ. We were in ZZZ to have our aircraft serviced for autoplt issues, associated with a 'rvsm' avionics upgrade installed previously. This was to be a test flight demonstrating the autoplt's ability to smoothly 'capture' the selected altitude in the altitude selector. I pre-briefed with my first officer what we were looking for and how we would go about the test flight. Since the automatic-pilot had not displayed any erratic behavior previously (outside of altitude capture), no mention was made to any min altitude to select basic autoflt automation. After a normal takeoff, at approximately 2000 ft MSL, I selected 'pitch' mode and heading select on the autoplt in order to easily scan for traffic and assist in the level off at our assigned altitude of 3000 ft MSL. Upon selection of the autoplt, the nose aggressively pitched up. I attempted to disengage the autoplt, but pitch control was extremely difficult and trending worse. I disconnected the main trim and with assistance from the co-pilot held significant forward pressure on the yoke to arrest the significant climb rate and capture our assigned altitude while using the standby trim to return normal yoke pressures and trim. We overshot our altitude by a maximum of 200 ft returned to the airport for a normal landing. Upon return to the maintenance facility, we reported the event. Some background study was done on the autoplt amplifier, the serviced component. It was discovered that collins avionics released this 'loaner' amplifier into service after a previous trim runway event. The company had signed off the amplifier as, 'could not duplicate.' we had no idea that our aircraft would display a completely new failure mode after servicing, and our pre-flight briefing reflected this. We should have assumed that for the test flight, the 'test' of the serviced component should be a complete and comprehensive look at all functions at a safe and pre-determined altitude. All modes should have been considered as suspect, not just previously squawked modes. Callback conversation with reporter revealed the following information: the reporter stated the aircraft was on a test flight for a failure to capture a selected altitude. The reporter said the test flight was planned and a discussion highlighted the method and procedure. The reporter stated the crew was not prepared for a rapid pitch up when the autoplt was engaged and momentary pitch control was lost. The reporter said the autoplt amplifier installed in the aircraft was tagged as a serviceable unit and had a previous report of a runway trim problem and maintenance action was 'could not duplicate report.'

Google
 

Original NASA ASRS Text

Title: A WESTWIND 24 AT 2000 FT THE AGGRESSIVELY PITCHED UP. WHEN AUTOPLT PITCH MODE AND HEADING SELECT ENGAGED MAIN TRIM WAS IN A RWY CONDITION.

Narrative: I WAS THE CAPT ON A FLT ORIGINATING AT ZZZ. WE WERE IN ZZZ TO HAVE OUR ACFT SERVICED FOR AUTOPLT ISSUES, ASSOCIATED WITH A 'RVSM' AVIONICS UPGRADE INSTALLED PREVIOUSLY. THIS WAS TO BE A TEST FLT DEMONSTRATING THE AUTOPLT'S ABILITY TO SMOOTHLY 'CAPTURE' THE SELECTED ALT IN THE ALT SELECTOR. I PRE-BRIEFED WITH MY FO WHAT WE WERE LOOKING FOR AND HOW WE WOULD GO ABOUT THE TEST FLT. SINCE THE AUTO-PLT HAD NOT DISPLAYED ANY ERRATIC BEHAVIOR PREVIOUSLY (OUTSIDE OF ALT CAPTURE), NO MENTION WAS MADE TO ANY MIN ALT TO SELECT BASIC AUTOFLT AUTOMATION. AFTER A NORMAL TKOF, AT APPROX 2000 FT MSL, I SELECTED 'PITCH' MODE AND HEADING SELECT ON THE AUTOPLT IN ORDER TO EASILY SCAN FOR TFC AND ASSIST IN THE LEVEL OFF AT OUR ASSIGNED ALT OF 3000 FT MSL. UPON SELECTION OF THE AUTOPLT, THE NOSE AGGRESSIVELY PITCHED UP. I ATTEMPTED TO DISENGAGE THE AUTOPLT, BUT PITCH CTL WAS EXTREMELY DIFFICULT AND TRENDING WORSE. I DISCONNECTED THE MAIN TRIM AND WITH ASSISTANCE FROM THE CO-PLT HELD SIGNIFICANT FORWARD PRESSURE ON THE YOKE TO ARREST THE SIGNIFICANT CLB RATE AND CAPTURE OUR ASSIGNED ALT WHILE USING THE STANDBY TRIM TO RETURN NORMAL YOKE PRESSURES AND TRIM. WE OVERSHOT OUR ALT BY A MAX OF 200 FT RETURNED TO THE ARPT FOR A NORMAL LNDG. UPON RETURN TO THE MAINT FAC, WE REPORTED THE EVENT. SOME BACKGROUND STUDY WAS DONE ON THE AUTOPLT AMPLIFIER, THE SERVICED COMPONENT. IT WAS DISCOVERED THAT COLLINS AVIONICS RELEASED THIS 'LOANER' AMPLIFIER INTO SVC AFTER A PREVIOUS TRIM RWY EVENT. THE COMPANY HAD SIGNED OFF THE AMPLIFIER AS, 'COULD NOT DUPLICATE.' WE HAD NO IDEA THAT OUR ACFT WOULD DISPLAY A COMPLETELY NEW FAILURE MODE AFTER SERVICING, AND OUR PRE-FLT BRIEFING REFLECTED THIS. WE SHOULD HAVE ASSUMED THAT FOR THE TEST FLT, THE 'TEST' OF THE SERVICED COMPONENT SHOULD BE A COMPLETE AND COMPREHENSIVE LOOK AT ALL FUNCTIONS AT A SAFE AND PRE-DETERMINED ALT. ALL MODES SHOULD HAVE BEEN CONSIDERED AS SUSPECT, NOT JUST PREVIOUSLY SQUAWKED MODES. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE RPTR STATED THE ACFT WAS ON A TEST FLT FOR A FAILURE TO CAPTURE A SELECTED ALT. THE RPTR SAID THE TEST FLT WAS PLANNED AND A DISCUSSION HIGHLIGHTED THE METHOD AND PROC. THE RPTR STATED THE CREW WAS NOT PREPARED FOR A RAPID PITCH UP WHEN THE AUTOPLT WAS ENGAGED AND MOMENTARY PITCH CTL WAS LOST. THE RPTR SAID THE AUTOPLT AMPLIFIER INSTALLED IN THE ACFT WAS TAGGED AS A SERVICEABLE UNIT AND HAD A PREVIOUS RPT OF A RWY TRIM PROB AND MAINT ACTION WAS 'COULD NOT DUPLICATE RPT.'

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.