NTSB Releases Prelim Info On Collings B-17 Crash


The NTSB has published its initial factual findings on the crash of the Collings Foundation B-17 Nine-O-Nine on Oct. 2. The Boeing B-17 crashed shortly after departure from Bradley International in Connecticut, killing seven of the 13 aboard, including pilot Ernest McCauley and copilot Michael Foster.

According to the NTSB’s initial report: “On the morning of the accident flight, an airport lineman at BDL assisted the loadmaster as he added 160 gallons of 100LL aviation fuel to the accident airplane. The lineman stated that the accident airplane was the first to be fueled with 100LL fuel that day. According to preliminary air traffic control (ATC) data provided by the FAA, shortly after takeoff, at 0950, one of the pilots reported to ATC that he wanted to return to the airport. At that time, the airplane was about 500 feet above ground level (agl) on the right crosswind leg of the airport traffic pattern for runway 6. The approach controller verified the request and asked if the pilot required any assistance, to which he replied no. The controller then asked for the reason for the return to the airport, and the pilot replied that the airplane had a ‘rough mag’ on the No. 4 engine. The controller then instructed the pilot to fly a right downwind leg for runway 6 and confirmed that the flight needed an immediate landing.”

Once handed back to the tower, the pilot was informed that the wind was calm and was cleared to landing on Runway 6. The pilot acknowledged and, according to the NTSB factual, the B-17 was just 300 feet above ground (AGL) on the midfield right downwind. “The tower controller asked about the airplane’s progress to the runway and the pilot replied that they were ‘getting there.’”

The B-17 contacted approach lights 1000 feet short of the runway and made ground contact 500 feet before the threshold. “It then veered right off the runway before colliding with vehicles and a deicing fluid tank about 1100 ft right of the center of the runway threshold. The wreckage came to rest upright and the majority of the cabin, cockpit, and right wing were consumed by postimpact fire. The landing gear was extended and measurement of the left and right wing flap jackscrews corresponded to a flaps retracted setting,” the NTSB reports. Overall control continuity was established by NTSB investigators at the scene.

Examination of the left-side engines (Numbers 1 and 2) suggest they were still making power at the time of impact while the investigators found that the Number-3 engine’s propeller had one blade that was “impact damaged and near the feather position. The other two blades appeared in a position between low pitch and feather.” Moreover, the report confirms that “… all three propeller blades on the No. 4 engine appeared in the feather position.” The NTSB notes that both engines on the left wing and the inboard right-side engine had been overhauled at the previous annual, around 270 hours prior to the accident. The Number 4 engine had 1106 hours since major overhaul at the time of the crash.

In a statement on the Collings Foundation website, the organization says, “Our thoughts and prayers are with those who were on that flight and we will be forever grateful to the heroic efforts of the first responders at Bradley. The Collings Foundation flight team is fully cooperating with officials to determine the cause of the crash of the B-17 Flying Fortress and will comment further when details become known.”

As we reported previously, the Collings Foundation Nine-O-Nine, registered N93012, entered service in April 1945. It was purchased and restored by the foundation in 1986 after having spent time as an air-sea rescue aircraft, nuclear test subject and fire bomber. N93012 did not see combat but was named and painted to honor the original Nine-O-Nine, which completed 140 combat missions without an abort or loss of a crewman and dropped an estimated 562,000 pounds of bombs in WWII. The first Nine-O-Nine flew with the U.S. Army Air Forces 91st Bomb Group’s 323rd Squadron. It was retired after the end of the war and scrapped. 

Marc Cook
KITPLANES Editor in Chief Marc Cook has been in aviation journalism for more than 30 years. He is a 4000-hour instrument-rated, multi-engine pilot with experience in nearly 150 types. He’s completed two kit aircraft, an Aero Designs Pulsar XP and a Glasair Sportsman 2+2, and currently flies a 2002 GlaStar.

Other AVwebflash Articles


  1. If you have already lost #4 and #3 is losing power necessitating feathering #3 prop, those pilots were faced with getting the airplane back to Runway 6 turning into what appears two dead engines. That would be consistent with a max altitude gain of 500ft with a loss of 200 feet by mid-field on a right downwind due to two engines on the right side failing. Those right hand turns would have to be very gentle trying to keep what little airspeed they had up, G loads low, plus doing it with a boot-full of left rudder at the same time…basically cross-controlled trying to get lined up with 06. It also appears the right wing contacted the ILS lights which would further swing the airplane to the right. I would not be surprised if the NTSB finds out the two left engines, while making power, they were not making full power and were spooling down as 3 and 4 had already done.

    It was quite a feat of airman-ship to nurse a low and slow, heavy bomber doing right turns into two dead engines with no more than 500 feet of altitude to begin with and make it back to the airport. Although short of the runway, it was still under control when they contacted the ILS lights. Right side dead engines, right side contact with ILS lights…not much to help you turn left or even stay straight under those circumstances.

    I believe that they were using their CRM skills to the utmost, staying very busy yet focused on getting that airplane back on the ground without a loss of control. And they succeeded but for those ILS lights aiding in turning the airplane into the deicing hangar. While the progress was slow, what other choice did they have?

    They were busy is an understatement. Lousy ending in spite of heroic efforts to keep flying the airplane. I believe these efforts displaying exemplary flying skills allowed for some to survive. With all the GoPro cameras, surveillance video, survivor recollections, and what evidence can be collected from the remains of the cockpit, engines, and instrumentation, the cause will be found for the failure of two engines and rapidly losing power on the remaining two.

  2. Rule# 1 on an “engine out” in a multi engine aircraft is “clean up the airplane” and don’t put the landing gear down. unless you have the have the runway made. Landing “gear up” does very little damage to the air frame.

      • I don’t think that the pilot was actually aware of the seriousness of the problem(s) onboard. This idea is supported by the pilot’s lack of urgency on the radio and his actions to fly a “pattern” instead of land immediately. It looks a lot like he’s thinking that he had a lot more available than he actually had so “why worry” about cleaning up or aiming at the first available runway because it sure looks like he did not realize that it was going bad in such a hurry.

  3. I do not want to Monday Morning QB this. I would just like to add a few thoughts having flown King Airs to C172s in and out of BDL. There is always that moment of indecision of “try for the field” or land off field. The environment around BDL is there are flat large tobacco fields surrounding the airport especially on the right hand traffic side of 6. One possibility. Another possibility was landing on 33, the wind was calm and it was CAVU so 33 was an option that would have saved several miles of flight and one turn since they were already on the crosswind for 6. They could have declared an emergency used the crosswind of 6 as the downwind for 33 and probably had enough energy to land maybe even hot on 33 which is still 6847 ft.

    I agree with Jim H. that to nurse a B-17 with one maybe 2 dead or dying engines was a feat of airmanship! We will never really know since to my knowledge there was no recorder but I am sure all the various cameras in the area will be examined.