Narrative:

After an extended delay on the ground for maintenance we finally obtained an aircraft for our departure to mynn. It had been raining heavily off and on with tropical WX conditions in the area. It was difficult keeping the aircraft dry during passenger boarding. The first officer was wet from the heavy rain. While he boarded passenger, I copied the clearance as requiring a departure frequency of 125.5. I performed a normal engine start and we commenced the normal 'before taxi checks' and 'before takeoff checks' over the normal aircraft noise without intercom since the first officer was having trouble with intercom and radio communications. I elected to proceed thinking the trouble would remedy itself since the problem he was having appeared to be intermittent and I was having no radio or intercom difficulties. The airport was busy, we were running very late, and there was at least a 10 min wait to the hold short. Radio traffic volume was very high, and my first officer had a strong language accent. Therefore, I performed all radio calls. We were cleared into position and hold for runway 12, but were further delayed for sequencing traffic. Finally, tower cleared us to change runways to 9L to 'get us out,' and then cleared us for takeoff from runway 9L as we were in the process of taxiing from runway 12 to runway 9L. After takeoff, tower advised us to turn on our transponder and contact mia departure. I verified to tower that it was on, which it was, and we switched to 125.5. After repeated attempts to contact departure (it sounded as if we were broadcasting due to appropriate normal side tone in my headset, but there were no other voice calls heard) I tuned in 119.45 and could finally hear departure, so I believed that departure frequencys had been changed to 119.45 during our delay at the hold short. Just as I was about to check in with departure control, and passing approximately 2800 ft in the initial climb, the first officer (by this time he was on the intercom) reported smoke in the cockpit. The smoke was beneath the first officer's instrument panel, and seemed to hover low without billowing. It was lightish gray in color, but not thick. However, it was acrid to the smell and burning in the nose, so I called for the emergency electrical smoke or fire checks. Next, the avionics panel dimmed and I saw the avionics master switch in the 'off' position. I don't know why this occurred, but I turned it back on and the radios remained on for the remainder of the flight. We both donned masks, and I contacted approach who by this time was frustrated with us not checking in. The intercom worked normally with the oxygen masks on. I advised that we had smoke in the cockpit and that we were declaring an emergency. They gave us vectors back to the field and set us up for the ILS runway 27R. It was raining and we saw the field at approximately 1000 ft AGL. We received normal vectoring service and were switched to a different frequency, which we did without difficulty. At this point, I was using both VHF radios -- 1 with approach and 1 with tower. Both worked normally. When we activated the oxygen supply to the flight crew, we visually inspected the passenger cabin and could see no smoke in the passenger cabin. I called for the continuation of the electrical smoke and fire checklist, which we completed to the point where the checklist calls for securing the avionics master switch. I elected not to do this because we were in actual nighttime IMC and moderate rain. While on final approach, we briefed the passenger that we were commencing an emergency landing and evacuate/evacuation. We landed uneventfully, and the crash crew was there to meet us. There were no lingering fumes or signs of fire after shutdown. The passenger were evacuate/evacuationed without injury. After landing and passenger evacuate/evacuation, I made a logbook entry bout the 'smoke in the cockpit,' but I did not speculate as to its source. I made no entry in the logbook about the radio problem, thinking that the 'smoke in the cockpit' write-up would suffice in giving maintenance proper notification to perform a thorough inspection of all aircraft's system. On the day following the incident, I debriefed with the carrier's director of operations. We discussed the nature and intensity of the smoke and the evacuate/evacuation procedure, but did not at that time discuss the problems with the intercom at the inception of the flight. Likewise, during my first contact with the FAA poi, our conversation centered mostly on the smoke and evacuate/evacuation procedures. A few days after the incident, I spoke with the director of operations again and we discussed the entire incident again, and the issue of the intercom and radio irregularities were discussed. Callback conversation with reporter revealed the following information: the reporter stated the maintenance people signed off the smoke report as being caused by water vapor condensation from the air cycle machine. The reporter said the FAA primary operations inspector asked if the signoff was satisfactory and the reporter stated it was not satisfactory as it did not explain the acrid odor present with the smoke and the lights dimming. The reporter stated in high humidity conditions it is normal to get some condensation in the cockpit and passenger cabin, but in this case it was only present in the cockpit. The NTSB was involved due to the passenger being evacuate/evacuationed from the aircraft.

Google
 

Original NASA ASRS Text

Title: A BE1900 ON INITIAL CLB AT 2800 FT DECLARED AN EMER AND DIVERTED DUE TO SMOKE AND ACRID SMELL IN THE COCKPIT.

Narrative: AFTER AN EXTENDED DELAY ON THE GND FOR MAINT WE FINALLY OBTAINED AN ACFT FOR OUR DEP TO MYNN. IT HAD BEEN RAINING HEAVILY OFF AND ON WITH TROPICAL WX CONDITIONS IN THE AREA. IT WAS DIFFICULT KEEPING THE ACFT DRY DURING PAX BOARDING. THE FO WAS WET FROM THE HVY RAIN. WHILE HE BOARDED PAX, I COPIED THE CLRNC AS REQUIRING A DEP FREQ OF 125.5. I PERFORMED A NORMAL ENG START AND WE COMMENCED THE NORMAL 'BEFORE TAXI CHKS' AND 'BEFORE TKOF CHKS' OVER THE NORMAL ACFT NOISE WITHOUT INTERCOM SINCE THE FO WAS HAVING TROUBLE WITH INTERCOM AND RADIO COMS. I ELECTED TO PROCEED THINKING THE TROUBLE WOULD REMEDY ITSELF SINCE THE PROB HE WAS HAVING APPEARED TO BE INTERMITTENT AND I WAS HAVING NO RADIO OR INTERCOM DIFFICULTIES. THE ARPT WAS BUSY, WE WERE RUNNING VERY LATE, AND THERE WAS AT LEAST A 10 MIN WAIT TO THE HOLD SHORT. RADIO TFC VOLUME WAS VERY HIGH, AND MY FO HAD A STRONG LANGUAGE ACCENT. THEREFORE, I PERFORMED ALL RADIO CALLS. WE WERE CLRED INTO POS AND HOLD FOR RWY 12, BUT WERE FURTHER DELAYED FOR SEQUENCING TFC. FINALLY, TWR CLRED US TO CHANGE RWYS TO 9L TO 'GET US OUT,' AND THEN CLRED US FOR TKOF FROM RWY 9L AS WE WERE IN THE PROCESS OF TAXIING FROM RWY 12 TO RWY 9L. AFTER TKOF, TWR ADVISED US TO TURN ON OUR XPONDER AND CONTACT MIA DEP. I VERIFIED TO TWR THAT IT WAS ON, WHICH IT WAS, AND WE SWITCHED TO 125.5. AFTER REPEATED ATTEMPTS TO CONTACT DEP (IT SOUNDED AS IF WE WERE BROADCASTING DUE TO APPROPRIATE NORMAL SIDE TONE IN MY HEADSET, BUT THERE WERE NO OTHER VOICE CALLS HEARD) I TUNED IN 119.45 AND COULD FINALLY HEAR DEP, SO I BELIEVED THAT DEP FREQS HAD BEEN CHANGED TO 119.45 DURING OUR DELAY AT THE HOLD SHORT. JUST AS I WAS ABOUT TO CHK IN WITH DEP CTL, AND PASSING APPROX 2800 FT IN THE INITIAL CLB, THE FO (BY THIS TIME HE WAS ON THE INTERCOM) RPTED SMOKE IN THE COCKPIT. THE SMOKE WAS BENEATH THE FO'S INST PANEL, AND SEEMED TO HOVER LOW WITHOUT BILLOWING. IT WAS LIGHTISH GRAY IN COLOR, BUT NOT THICK. HOWEVER, IT WAS ACRID TO THE SMELL AND BURNING IN THE NOSE, SO I CALLED FOR THE EMER ELECTRICAL SMOKE OR FIRE CHKS. NEXT, THE AVIONICS PANEL DIMMED AND I SAW THE AVIONICS MASTER SWITCH IN THE 'OFF' POS. I DON'T KNOW WHY THIS OCCURRED, BUT I TURNED IT BACK ON AND THE RADIOS REMAINED ON FOR THE REMAINDER OF THE FLT. WE BOTH DONNED MASKS, AND I CONTACTED APCH WHO BY THIS TIME WAS FRUSTRATED WITH US NOT CHKING IN. THE INTERCOM WORKED NORMALLY WITH THE OXYGEN MASKS ON. I ADVISED THAT WE HAD SMOKE IN THE COCKPIT AND THAT WE WERE DECLARING AN EMER. THEY GAVE US VECTORS BACK TO THE FIELD AND SET US UP FOR THE ILS RWY 27R. IT WAS RAINING AND WE SAW THE FIELD AT APPROX 1000 FT AGL. WE RECEIVED NORMAL VECTORING SVC AND WERE SWITCHED TO A DIFFERENT FREQ, WHICH WE DID WITHOUT DIFFICULTY. AT THIS POINT, I WAS USING BOTH VHF RADIOS -- 1 WITH APCH AND 1 WITH TWR. BOTH WORKED NORMALLY. WHEN WE ACTIVATED THE OXYGEN SUPPLY TO THE FLC, WE VISUALLY INSPECTED THE PAX CABIN AND COULD SEE NO SMOKE IN THE PAX CABIN. I CALLED FOR THE CONTINUATION OF THE ELECTRICAL SMOKE AND FIRE CHKLIST, WHICH WE COMPLETED TO THE POINT WHERE THE CHKLIST CALLS FOR SECURING THE AVIONICS MASTER SWITCH. I ELECTED NOT TO DO THIS BECAUSE WE WERE IN ACTUAL NIGHTTIME IMC AND MODERATE RAIN. WHILE ON FINAL APCH, WE BRIEFED THE PAX THAT WE WERE COMMENCING AN EMER LNDG AND EVAC. WE LANDED UNEVENTFULLY, AND THE CRASH CREW WAS THERE TO MEET US. THERE WERE NO LINGERING FUMES OR SIGNS OF FIRE AFTER SHUTDOWN. THE PAX WERE EVACED WITHOUT INJURY. AFTER LNDG AND PAX EVAC, I MADE A LOGBOOK ENTRY BOUT THE 'SMOKE IN THE COCKPIT,' BUT I DID NOT SPECULATE AS TO ITS SOURCE. I MADE NO ENTRY IN THE LOGBOOK ABOUT THE RADIO PROB, THINKING THAT THE 'SMOKE IN THE COCKPIT' WRITE-UP WOULD SUFFICE IN GIVING MAINT PROPER NOTIFICATION TO PERFORM A THOROUGH INSPECTION OF ALL ACFT'S SYS. ON THE DAY FOLLOWING THE INCIDENT, I DEBRIEFED WITH THE CARRIER'S DIRECTOR OF OPS. WE DISCUSSED THE NATURE AND INTENSITY OF THE SMOKE AND THE EVAC PROC, BUT DID NOT AT THAT TIME DISCUSS THE PROBS WITH THE INTERCOM AT THE INCEPTION OF THE FLT. LIKEWISE, DURING MY FIRST CONTACT WITH THE FAA POI, OUR CONVERSATION CTRED MOSTLY ON THE SMOKE AND EVAC PROCS. A FEW DAYS AFTER THE INCIDENT, I SPOKE WITH THE DIRECTOR OF OPS AGAIN AND WE DISCUSSED THE ENTIRE INCIDENT AGAIN, AND THE ISSUE OF THE INTERCOM AND RADIO IRREGULARITIES WERE DISCUSSED. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE RPTR STATED THE MAINT PEOPLE SIGNED OFF THE SMOKE RPT AS BEING CAUSED BY WATER VAPOR CONDENSATION FROM THE AIR CYCLE MACHINE. THE RPTR SAID THE FAA PRIMARY OPS INSPECTOR ASKED IF THE SIGNOFF WAS SATISFACTORY AND THE RPTR STATED IT WAS NOT SATISFACTORY AS IT DID NOT EXPLAIN THE ACRID ODOR PRESENT WITH THE SMOKE AND THE LIGHTS DIMMING. THE RPTR STATED IN HIGH HUMIDITY CONDITIONS IT IS NORMAL TO GET SOME CONDENSATION IN THE COCKPIT AND PAX CABIN, BUT IN THIS CASE IT WAS ONLY PRESENT IN THE COCKPIT. THE NTSB WAS INVOLVED DUE TO THE PAX BEING EVACED FROM THE ACFT.

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.