Narrative:

During climb out of sac on the dudes 8 departure, I overshot the assigned altitude of FL190 by 500 ft (19500 ft) and descended back to FL190. Total time above FL190 was about 1 min. Contributing factor was the autoplt was in use and should have acquired and automatically leveled at FL190, however, it did not acquire. When the altitude warning horn went off at approximately 18200 ft and I realized the autoplt would not level off, the high climb at the time forced the overshoot. At the time we were deviating right of course for a cell that topped at around 25000 ft and were in cirrus clouds and the workload at the time (WX radar use) contributed to late recognition of the problem. It is normal procedure to verify the autoplt is leveling properly and I was late to recognize it was not. I have seen this system not level at the set altitude before when the human pilot manually started the leveloff prior to 'altitude acquire' display, when the autoplt starts to level off, and that may have occurred in this case. Since there was no reported traffic and our TCASII display showed no aircraft, the leveloff and correction was performed with passenger safety from our maneuver in mind rather than abruptly. ZOA asked us our assigned altitude as we descended back to FL190 and we responded FL190 at which point he cleared us to FL230. My normal habit is to do my own leveloff, sticking to that habit will correct the problem. Callback conversation with reporter revealed the following information: for some reason, the autoplt failed to level off the B-737-300. The first officer reporter insists that it was properly programmed and that he did not revert to vertical speed. He may have pushed forward on the yoke, but not enough to disengage the autoplt. Apparently this type aircraft can miss a leveloff when the aircraft is climbing at a very high rate, in this case about 3000 FPM. The crew was distracted by WX and the radar, trying to find a clear path. The first officer clicked off the autoplt and hand flew the aircraft down. ARTCC noticed the overshoot, but apparently was not concerned. Supplemental information from acn 297869: the controller was quite busy. I was operating the radar and was advising the first officer of headings to fly to avoid WX and not monitoring the altitude. We were passing FL180 with about 2500 FPM climb. The autoplt was unable to level with that rate of climb and the first officer wasn't being aggressive enough to level at FL190. My preoccupation with the radar and WX distracted me in monitoring the first officer and his climb rate. Updrafts could have created some of the excessive rate of climb. First officer stated he didn't level more aggressively because of passenger comfort and thought the flight attendants may have been standing.

Google
 

Original NASA ASRS Text

Title: ALT OVERSHOOT ON CLBOUT.

Narrative: DURING CLB OUT OF SAC ON THE DUDES 8 DEP, I OVERSHOT THE ASSIGNED ALT OF FL190 BY 500 FT (19500 FT) AND DSNDED BACK TO FL190. TOTAL TIME ABOVE FL190 WAS ABOUT 1 MIN. CONTRIBUTING FACTOR WAS THE AUTOPLT WAS IN USE AND SHOULD HAVE ACQUIRED AND AUTOMATICALLY LEVELED AT FL190, HOWEVER, IT DID NOT ACQUIRE. WHEN THE ALT WARNING HORN WENT OFF AT APPROX 18200 FT AND I REALIZED THE AUTOPLT WOULD NOT LEVEL OFF, THE HIGH CLB AT THE TIME FORCED THE OVERSHOOT. AT THE TIME WE WERE DEVIATING R OF COURSE FOR A CELL THAT TOPPED AT AROUND 25000 FT AND WERE IN CIRRUS CLOUDS AND THE WORKLOAD AT THE TIME (WX RADAR USE) CONTRIBUTED TO LATE RECOGNITION OF THE PROB. IT IS NORMAL PROC TO VERIFY THE AUTOPLT IS LEVELING PROPERLY AND I WAS LATE TO RECOGNIZE IT WAS NOT. I HAVE SEEN THIS SYS NOT LEVEL AT THE SET ALT BEFORE WHEN THE HUMAN PLT MANUALLY STARTED THE LEVELOFF PRIOR TO 'ALT ACQUIRE' DISPLAY, WHEN THE AUTOPLT STARTS TO LEVEL OFF, AND THAT MAY HAVE OCCURRED IN THIS CASE. SINCE THERE WAS NO RPTED TFC AND OUR TCASII DISPLAY SHOWED NO ACFT, THE LEVELOFF AND CORRECTION WAS PERFORMED WITH PAX SAFETY FROM OUR MANEUVER IN MIND RATHER THAN ABRUPTLY. ZOA ASKED US OUR ASSIGNED ALT AS WE DSNDED BACK TO FL190 AND WE RESPONDED FL190 AT WHICH POINT HE CLRED US TO FL230. MY NORMAL HABIT IS TO DO MY OWN LEVELOFF, STICKING TO THAT HABIT WILL CORRECT THE PROB. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: FOR SOME REASON, THE AUTOPLT FAILED TO LEVEL OFF THE B-737-300. THE FO RPTR INSISTS THAT IT WAS PROPERLY PROGRAMMED AND THAT HE DID NOT REVERT TO VERT SPD. HE MAY HAVE PUSHED FORWARD ON THE YOKE, BUT NOT ENOUGH TO DISENGAGE THE AUTOPLT. APPARENTLY THIS TYPE ACFT CAN MISS A LEVELOFF WHEN THE ACFT IS CLBING AT A VERY HIGH RATE, IN THIS CASE ABOUT 3000 FPM. THE CREW WAS DISTRACTED BY WX AND THE RADAR, TRYING TO FIND A CLR PATH. THE FO CLICKED OFF THE AUTOPLT AND HAND FLEW THE ACFT DOWN. ARTCC NOTICED THE OVERSHOOT, BUT APPARENTLY WAS NOT CONCERNED. SUPPLEMENTAL INFO FROM ACN 297869: THE CTLR WAS QUITE BUSY. I WAS OPERATING THE RADAR AND WAS ADVISING THE FO OF HDGS TO FLY TO AVOID WX AND NOT MONITORING THE ALT. WE WERE PASSING FL180 WITH ABOUT 2500 FPM CLB. THE AUTOPLT WAS UNABLE TO LEVEL WITH THAT RATE OF CLB AND THE FO WASN'T BEING AGGRESSIVE ENOUGH TO LEVEL AT FL190. MY PREOCCUPATION WITH THE RADAR AND WX DISTRACTED ME IN MONITORING THE FO AND HIS CLB RATE. UPDRAFTS COULD HAVE CREATED SOME OF THE EXCESSIVE RATE OF CLB. FO STATED HE DIDN'T LEVEL MORE AGGRESSIVELY BECAUSE OF PAX COMFORT AND THOUGHT THE FLT ATTENDANTS MAY HAVE BEEN STANDING.

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.