Narrative:

The crew was composed of myself, an instructor/check airman, and an experienced, very competent FMS first officer. This was the return leg of our first pairing of the day. Our return to phnl was flight phko to phnl on our departure from hnl the WX was poor enough to close the airport for about 10 mins due to it being below minimums for the lda runway 26L approach. It had improved considerably during our trip to koa, however, and was reported as approximately 800 scattered 2500 and higher broken with strong surface winds from the southwest. Cruise was at FL240 with the PNF (me) in VOR raw data and the PF (first officer) in the map mode, VNAV and LNAV, proceeding direct to sakki. We received clearance to cross sakki at 6000 ft at 250 KTS. As this restr had been previously programmed into the FMC, the airplane remained in VNAV to initiate the descent. Just after the descent was begun ATC advised slow to 250 KTS. Of course this made us immediately high on profile. First officer deployed speed brakes to regain the profile. Shortly after that center issued instructions for us to hold at sakki 'as published' maintain 7000 ft. The flag went up for both pilots as we were not aware of a published pattern for sakki. The airplane was in VNAV and the first officer, a good and rapid programmer, was entering the hold into the FMC. Company aircraft was ahead of us and cleared to hold at 6000 ft. Air carrier B was cleared to 8000 ft. I checked the en route chart and found no published pattern. The first officer checked the local chart (I believe after he had programmed the hold) and found no published pattern. I asked the controller if she wanted us to hold on the lny 278 radial. She replied to hold on the lda inbound course, left turns. The other aircraft inbound to the fix also queried the controller on the holding instructions as there was no published pattern and were ultimately given the same clearance. The hold was not for WX, but for traffic build-up from the previous closure. As we entered the hold I had visual contact with both company and air carrier B. On looking back inside I saw the altimeter pass through 6800 ft, I said '7000 ft' and arrested the descent with the yoke, stopping at approximately 6700 ft and began climb back to 7000 ft whereupon the controller asked for confirmation of 7000 ft. I acknowledged 7000 ft as by then the excursion had been corrected. 3 ATC clrncs in rather rapid succession, coupled with the puzzlement caused by a confusing hold clearance caused a rapid rise in workload. When the workload increases there is little time for monitoring FMA's. The airplane either was not in VNAV or the first officer had not changed the altitude in the FMC. I can say with certainty that both pilots have a high level of altitude awareness and always set the altitude alert and confirm it upon receipt of clearance. But not this time. I had turned away to write down the clearance and then went immediately down to the flight bag to check the chart for the pattern. The first officer went right to the CDU to program the hold. The normal sequence of events was broken. Thus, no one set the altitude window. When the airplane exited the VNAV mode, for whatever reason, altitude protection was lost. The lesson here is one of prioritization. First, set the altitude window before anything else is accomplished, including chart reference and CDU entries. Supplemental information from acn 294792: center directed us to 'hold at sakai as published, 7000 ft' -- we both checked our en route charts and the lda runway 26L approach, but no published holding was found -- I entered 7000 ft into the FMC, as sakai (250 KTS/7000 ft) -- we questioned center on the holding, and she changed it to hold south on the localizer, left turns. Set this in the FMC hold page. MCP was in a speed descent -- not VNAV. Event occurred because of distraction about the 'published' holding pattern and altitude change, and that 7000 ft was not set in the mlp altitude window, and not being in the VNAV mode. Care must be taken to insure the MCP altitude window is always set on the assigned altitude.

Google
 

Original NASA ASRS Text

Title: 737-400 HAS ALTDEV WHEN DIRECTED INTO HOLD ON DSCNT.

Narrative: THE CREW WAS COMPOSED OF MYSELF, AN INSTRUCTOR/CHK AIRMAN, AND AN EXPERIENCED, VERY COMPETENT FMS FO. THIS WAS THE RETURN LEG OF OUR FIRST PAIRING OF THE DAY. OUR RETURN TO PHNL WAS FLT PHKO TO PHNL ON OUR DEP FROM HNL THE WX WAS POOR ENOUGH TO CLOSE THE ARPT FOR ABOUT 10 MINS DUE TO IT BEING BELOW MINIMUMS FOR THE LDA RWY 26L APCH. IT HAD IMPROVED CONSIDERABLY DURING OUR TRIP TO KOA, HOWEVER, AND WAS RPTED AS APPROX 800 SCATTERED 2500 AND HIGHER BROKEN WITH STRONG SURFACE WINDS FROM THE SW. CRUISE WAS AT FL240 WITH THE PNF (ME) IN VOR RAW DATA AND THE PF (FO) IN THE MAP MODE, VNAV AND LNAV, PROCEEDING DIRECT TO SAKKI. WE RECEIVED CLRNC TO CROSS SAKKI AT 6000 FT AT 250 KTS. AS THIS RESTR HAD BEEN PREVIOUSLY PROGRAMMED INTO THE FMC, THE AIRPLANE REMAINED IN VNAV TO INITIATE THE DSCNT. JUST AFTER THE DSCNT WAS BEGUN ATC ADVISED SLOW TO 250 KTS. OF COURSE THIS MADE US IMMEDIATELY HIGH ON PROFILE. FO DEPLOYED SPD BRAKES TO REGAIN THE PROFILE. SHORTLY AFTER THAT CTR ISSUED INSTRUCTIONS FOR US TO HOLD AT SAKKI 'AS PUBLISHED' MAINTAIN 7000 FT. THE FLAG WENT UP FOR BOTH PLTS AS WE WERE NOT AWARE OF A PUBLISHED PATTERN FOR SAKKI. THE AIRPLANE WAS IN VNAV AND THE FO, A GOOD AND RAPID PROGRAMMER, WAS ENTERING THE HOLD INTO THE FMC. COMPANY ACFT WAS AHEAD OF US AND CLRED TO HOLD AT 6000 FT. ACR B WAS CLRED TO 8000 FT. I CHKED THE ENRTE CHART AND FOUND NO PUBLISHED PATTERN. THE FO CHKED THE LCL CHART (I BELIEVE AFTER HE HAD PROGRAMMED THE HOLD) AND FOUND NO PUBLISHED PATTERN. I ASKED THE CTLR IF SHE WANTED US TO HOLD ON THE LNY 278 RADIAL. SHE REPLIED TO HOLD ON THE LDA INBOUND COURSE, L TURNS. THE OTHER ACFT INBOUND TO THE FIX ALSO QUERIED THE CTLR ON THE HOLDING INSTRUCTIONS AS THERE WAS NO PUBLISHED PATTERN AND WERE ULTIMATELY GIVEN THE SAME CLRNC. THE HOLD WAS NOT FOR WX, BUT FOR TFC BUILD-UP FROM THE PREVIOUS CLOSURE. AS WE ENTERED THE HOLD I HAD VISUAL CONTACT WITH BOTH COMPANY AND ACR B. ON LOOKING BACK INSIDE I SAW THE ALTIMETER PASS THROUGH 6800 FT, I SAID '7000 FT' AND ARRESTED THE DSCNT WITH THE YOKE, STOPPING AT APPROX 6700 FT AND BEGAN CLB BACK TO 7000 FT WHEREUPON THE CTLR ASKED FOR CONFIRMATION OF 7000 FT. I ACKNOWLEDGED 7000 FT AS BY THEN THE EXCURSION HAD BEEN CORRECTED. 3 ATC CLRNCS IN RATHER RAPID SUCCESSION, COUPLED WITH THE PUZZLEMENT CAUSED BY A CONFUSING HOLD CLRNC CAUSED A RAPID RISE IN WORKLOAD. WHEN THE WORKLOAD INCREASES THERE IS LITTLE TIME FOR MONITORING FMA'S. THE AIRPLANE EITHER WAS NOT IN VNAV OR THE FO HAD NOT CHANGED THE ALT IN THE FMC. I CAN SAY WITH CERTAINTY THAT BOTH PLTS HAVE A HIGH LEVEL OF ALT AWARENESS AND ALWAYS SET THE ALT ALERT AND CONFIRM IT UPON RECEIPT OF CLRNC. BUT NOT THIS TIME. I HAD TURNED AWAY TO WRITE DOWN THE CLRNC AND THEN WENT IMMEDIATELY DOWN TO THE FLT BAG TO CHK THE CHART FOR THE PATTERN. THE FO WENT RIGHT TO THE CDU TO PROGRAM THE HOLD. THE NORMAL SEQUENCE OF EVENTS WAS BROKEN. THUS, NO ONE SET THE ALT WINDOW. WHEN THE AIRPLANE EXITED THE VNAV MODE, FOR WHATEVER REASON, ALT PROTECTION WAS LOST. THE LESSON HERE IS ONE OF PRIORITIZATION. FIRST, SET THE ALT WINDOW BEFORE ANYTHING ELSE IS ACCOMPLISHED, INCLUDING CHART REF AND CDU ENTRIES. SUPPLEMENTAL INFO FROM ACN 294792: CTR DIRECTED US TO 'HOLD AT SAKAI AS PUBLISHED, 7000 FT' -- WE BOTH CHKED OUR ENRTE CHARTS AND THE LDA RWY 26L APCH, BUT NO PUBLISHED HOLDING WAS FOUND -- I ENTERED 7000 FT INTO THE FMC, AS SAKAI (250 KTS/7000 FT) -- WE QUESTIONED CTR ON THE HOLDING, AND SHE CHANGED IT TO HOLD S ON THE LOC, L TURNS. SET THIS IN THE FMC HOLD PAGE. MCP WAS IN A SPD DSCNT -- NOT VNAV. EVENT OCCURRED BECAUSE OF DISTR ABOUT THE 'PUBLISHED' HOLDING PATTERN AND ALT CHANGE, AND THAT 7000 FT WAS NOT SET IN THE MLP ALT WINDOW, AND NOT BEING IN THE VNAV MODE. CARE MUST BE TAKEN TO INSURE THE MCP ALT WINDOW IS ALWAYS SET ON THE ASSIGNED ALT.

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.