Narrative:

How cold is too cold? My MD83 is certified for -54 degrees C on the ground and -76 degrees C en route. However, it did not operate very well at -41 degrees C in fai. Leg #1 sea-anc --no abnormalities observed. Leg #2 anc-fai -- takeoff weight 134.0, landing weight 129.5, cruise altitude FL280, block time 52 mins, fai WX: 1/4 freezing fog, temperature 041 degrees C, average RVR 1200+ ft. Approach: CAT IIIA runway 1L. Nothing abnormal observed. Landing: autoplt disconnected during flare mode at or just prior to touchdown producing a firm touchdown. Rollout was done manually. Maintenance was notified of unsuccessful autoland dfgc -- 2. During the autoland, the flare and pitch attitude looked proper, but at disconnect the nose aggressively pitched down. Analyzing the event showed sufficient trim available and no other apparent problems -- so I thought. Leg #3 fai-anc -- total ground time in fai was 33 mins. Block to air 46 mins. Fai WX: 1/4 freezing fog, temperature -41 degrees C with no precipitation reported. During taxi out, all flight controls were checked and cycled numerous times. Elevator trim was reset and moved freely. Elevator check showed proper indications with full and unrestr movement in both directions. However, forward elevator movement required more force than normal and a slight delay to produce the 'elevator power on' annunciator light. Aft elevator movement appeared normal. Takeoff RVR's were 1000/1000/1000 ft. The performance data/gross for runway 19R weight produced a rotation speed, vr of 130 KTS. During the takeoff roll at rotational speed the elevator feel was very heavy (ie, dead) and non responsive to normal inputs. Proper takeoff pitch attitude was difficult to obtain but was accomplished at a much slower than normal rate to 10-13 ft anu. Acceleration past vr was so rapid, aborting the takeoff was not an option. My primary concern, besides runway length, was with all this additional elevator back force, that the aircraft would suddenly react and over-rotate to an excessive pitch attitude. Not knowing my exact liftoff point, the aircraft became airborne around 180+ KTS. Luckily, I knew we had a 10000+ ft runway with no obstacles. The aircraft never felt out of trim(ie, nose heavy/nose light). During cruise to anc, autoplt pitch control was fair, but at the time appeared acceptable. During our autoplt off visual approach to anc runway 6R, elevator feel and response appeared normal because of minimal use and using stabilizer trim for primary pitch control. But once we started to slow and configure for landing, the elevator became stiff and non responsive to normal control force inputs. A normal VFR landing was accomplished using additional speed and stabilizer trim for additional pitch control. Maintenance inspections in anc revealed nothing mechanically wrong with the elevator system. Something in the elevator control system had to be too cold to function properly yet appear to check normal. Note: within the next 18 hours this exact event recurred. Autoplt disconnect on autoland approach/landing, ineffective elevator takeoff and landing. This all occurred in the same environmental conditions, -40 degrees C, same airport, fai, same aircraft type, MD80, but a different tail number. At the time of this report maintenance inspections on both aircraft revealed nothing mechanically wrong. Callback conversation with reporter revealed the following information: the reporter stated the aircraft was an MD83 pwred with P&west JT8D-219 engines. The reporter said on landing at fai the autoland appeared normal until the autoplt disconnected during flare mode and pitched the nose aggressively down. The reporter said the control check was normal and after 33 mins the aircraft was ready for departure to anc. The reporter stated the controls were again checked and were free and unrestr. The reporter stated that during takeoff roll the elevator force was heavy and would not rotate at the calculated vr of 130 KTS but was made at 180 KTS with 10-13 degrees nose up. The reporter said the elevators were stiff and non responsive when slowing down on approach to anc. The reporter stated the departure from fai was at -41 degrees C. The reporter said the next day when at maintenance control center at ZZZ reporting on this incident another incident was being reported of limited pitch control out of fai on a company MD80. The reporter said the aircraft had a different tail number and the report was almost identical with autoplt disconnects and heavy elevator forces in the same environmental conditions. The reporter said fortunately this second aircraft was equipped with a digital flight data recorder and recorded a sticking or bypassing valve. The reporter did not identify the valve function.

Related reports:

Google
 

Original NASA ASRS Text

Title: AN MD83 OPERATING IN EXTREME COLD CONDITIONS RPTS LIMITED PITCH CTL ON TKOF AND LNDG.

Narrative: HOW COLD IS TOO COLD? MY MD83 IS CERTIFIED FOR -54 DEGS C ON THE GND AND -76 DEGS C ENRTE. HOWEVER, IT DID NOT OPERATE VERY WELL AT -41 DEGS C IN FAI. LEG #1 SEA-ANC --NO ABNORMALITIES OBSERVED. LEG #2 ANC-FAI -- TKOF WT 134.0, LNDG WT 129.5, CRUISE ALT FL280, BLOCK TIME 52 MINS, FAI WX: 1/4 FREEZING FOG, TEMP 041 DEGS C, AVERAGE RVR 1200+ FT. APCH: CAT IIIA RWY 1L. NOTHING ABNORMAL OBSERVED. LNDG: AUTOPLT DISCONNECTED DURING FLARE MODE AT OR JUST PRIOR TO TOUCHDOWN PRODUCING A FIRM TOUCHDOWN. ROLLOUT WAS DONE MANUALLY. MAINT WAS NOTIFIED OF UNSUCCESSFUL AUTOLAND DFGC -- 2. DURING THE AUTOLAND, THE FLARE AND PITCH ATTITUDE LOOKED PROPER, BUT AT DISCONNECT THE NOSE AGGRESSIVELY PITCHED DOWN. ANALYZING THE EVENT SHOWED SUFFICIENT TRIM AVAILABLE AND NO OTHER APPARENT PROBS -- SO I THOUGHT. LEG #3 FAI-ANC -- TOTAL GND TIME IN FAI WAS 33 MINS. BLOCK TO AIR 46 MINS. FAI WX: 1/4 FREEZING FOG, TEMP -41 DEGS C WITH NO PRECIP RPTED. DURING TAXI OUT, ALL FLT CTLS WERE CHKED AND CYCLED NUMEROUS TIMES. ELEVATOR TRIM WAS RESET AND MOVED FREELY. ELEVATOR CHK SHOWED PROPER INDICATIONS WITH FULL AND UNRESTR MOVEMENT IN BOTH DIRECTIONS. HOWEVER, FORWARD ELEVATOR MOVEMENT REQUIRED MORE FORCE THAN NORMAL AND A SLIGHT DELAY TO PRODUCE THE 'ELEVATOR PWR ON' ANNUNCIATOR LIGHT. AFT ELEVATOR MOVEMENT APPEARED NORMAL. TKOF RVR'S WERE 1000/1000/1000 FT. THE PERFORMANCE DATA/GROSS FOR RWY 19R WT PRODUCED A ROTATION SPD, VR OF 130 KTS. DURING THE TKOF ROLL AT ROTATIONAL SPD THE ELEVATOR FEEL WAS VERY HVY (IE, DEAD) AND NON RESPONSIVE TO NORMAL INPUTS. PROPER TKOF PITCH ATTITUDE WAS DIFFICULT TO OBTAIN BUT WAS ACCOMPLISHED AT A MUCH SLOWER THAN NORMAL RATE TO 10-13 FT ANU. ACCELERATION PAST VR WAS SO RAPID, ABORTING THE TKOF WAS NOT AN OPTION. MY PRIMARY CONCERN, BESIDES RWY LENGTH, WAS WITH ALL THIS ADDITIONAL ELEVATOR BACK FORCE, THAT THE ACFT WOULD SUDDENLY REACT AND OVER-ROTATE TO AN EXCESSIVE PITCH ATTITUDE. NOT KNOWING MY EXACT LIFTOFF POINT, THE ACFT BECAME AIRBORNE AROUND 180+ KTS. LUCKILY, I KNEW WE HAD A 10000+ FT RWY WITH NO OBSTACLES. THE ACFT NEVER FELT OUT OF TRIM(IE, NOSE HVY/NOSE LIGHT). DURING CRUISE TO ANC, AUTOPLT PITCH CTL WAS FAIR, BUT AT THE TIME APPEARED ACCEPTABLE. DURING OUR AUTOPLT OFF VISUAL APCH TO ANC RWY 6R, ELEVATOR FEEL AND RESPONSE APPEARED NORMAL BECAUSE OF MINIMAL USE AND USING STABILIZER TRIM FOR PRIMARY PITCH CTL. BUT ONCE WE STARTED TO SLOW AND CONFIGURE FOR LNDG, THE ELEVATOR BECAME STIFF AND NON RESPONSIVE TO NORMAL CTL FORCE INPUTS. A NORMAL VFR LNDG WAS ACCOMPLISHED USING ADDITIONAL SPD AND STABILIZER TRIM FOR ADDITIONAL PITCH CTL. MAINT INSPECTIONS IN ANC REVEALED NOTHING MECHANICALLY WRONG WITH THE ELEVATOR SYS. SOMETHING IN THE ELEVATOR CTL SYS HAD TO BE TOO COLD TO FUNCTION PROPERLY YET APPEAR TO CHK NORMAL. NOTE: WITHIN THE NEXT 18 HRS THIS EXACT EVENT RECURRED. AUTOPLT DISCONNECT ON AUTOLAND APCH/LNDG, INEFFECTIVE ELEVATOR TKOF AND LNDG. THIS ALL OCCURRED IN THE SAME ENVIRONMENTAL CONDITIONS, -40 DEGS C, SAME ARPT, FAI, SAME ACFT TYPE, MD80, BUT A DIFFERENT TAIL NUMBER. AT THE TIME OF THIS RPT MAINT INSPECTIONS ON BOTH ACFT REVEALED NOTHING MECHANICALLY WRONG. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE RPTR STATED THE ACFT WAS AN MD83 PWRED WITH P&W JT8D-219 ENGS. THE RPTR SAID ON LNDG AT FAI THE AUTOLAND APPEARED NORMAL UNTIL THE AUTOPLT DISCONNECTED DURING FLARE MODE AND PITCHED THE NOSE AGGRESSIVELY DOWN. THE RPTR SAID THE CTL CHK WAS NORMAL AND AFTER 33 MINS THE ACFT WAS READY FOR DEP TO ANC. THE RPTR STATED THE CTLS WERE AGAIN CHKED AND WERE FREE AND UNRESTR. THE RPTR STATED THAT DURING TKOF ROLL THE ELEVATOR FORCE WAS HVY AND WOULD NOT ROTATE AT THE CALCULATED VR OF 130 KTS BUT WAS MADE AT 180 KTS WITH 10-13 DEGS NOSE UP. THE RPTR SAID THE ELEVATORS WERE STIFF AND NON RESPONSIVE WHEN SLOWING DOWN ON APCH TO ANC. THE RPTR STATED THE DEP FROM FAI WAS AT -41 DEGS C. THE RPTR SAID THE NEXT DAY WHEN AT MAINT CTL CTR AT ZZZ RPTING ON THIS INCIDENT ANOTHER INCIDENT WAS BEING RPTED OF LIMITED PITCH CTL OUT OF FAI ON A COMPANY MD80. THE RPTR SAID THE ACFT HAD A DIFFERENT TAIL NUMBER AND THE RPT WAS ALMOST IDENTICAL WITH AUTOPLT DISCONNECTS AND HVY ELEVATOR FORCES IN THE SAME ENVIRONMENTAL CONDITIONS. THE RPTR SAID FORTUNATELY THIS SECOND ACFT WAS EQUIPPED WITH A DIGITAL FLT DATA RECORDER AND RECORDED A STICKING OR BYPASSING VALVE. THE RPTR DID NOT IDENT THE VALVE FUNCTION.

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.