From Aviation Week

 

Fire On Takeoff Downed B-2

Feb 26, 2008

David A. Fulghum/Aerospace Daily & Defense Report

 

One of the pilots of the B-2 stealth bomber that crashed Feb. 23 reported a fire at takeoff from Andersen Air Force Base, Guam, which was followed quickly by loss of control of the bomber, according to a senior Air Combat Command (ACC) official.

 

The stealth bomber rolled uncontrollably to the right and fell between the taxiway and the ramp at 10:45 a.m. Guam time just after passing the control tower. It was attempting a takeoff toward the seaward end of the runway. The two pilots ejected with one being hospitalized. A dark plume of smoke arose from the crash site and civilians outside the base reported a second explosion about 30 minutes after the initial impact.

 

The aircraft can lose one or even two of its four General Electric F118-GE-100 17,300-pound thrust engines and still take off, so it’s unlikely that engine failure was to blame, says a retired U.S. Air Force pilot who has flown the B-2. Moreover, early suggestions that the aircraft struck birds or stalled in a steep takeoff climb also have been dismissed as unlikely. Also, the weather was reported as clear.

 

The bomber Spirit of Kansas, tail no. 890127, was the second in a four-aircraft flight that was ending its deployment and taking off for return to home base at Whiteman Air Force Base, Mo. They were being replaced by six B-52s as a forward-based, heavy-bomber force in the Pacific. The loss cuts the number of combat-coded B-2s to 15 from 16 out of the total force of 21. The force has a minimum aircraft requirement of 19 airframes.

 

The other three B-2s later returned to Whiteman, where the wing commander has declared a “safety pause” for the fleet, ACC officials say. During the pause, procedures are being reviewed with the pilots and training is at a standdown. However, if the stealthy bomber is needed for an operational mission it is cleared to fly.

 

The aircraft that crashed rolled off Northrop Grumman’s line in 1989 and had accumulated 5,176 flying hours at the time of the crash. Early testing indicated that the aircraft would remain structurally intact for about 40,000 flying hours. Analyses also posited that the rudder attachment points would be the first structural failure item.

 

This is the first B-2 crash, but another of the Air Force’s heavy bomber fleet, the B-1, has suffered a number of seemingly similar in-flight emergencies. A pelican-sized bird ripping through fuel and hydraulic lines downed the first, a fuel-fed fire on takeoff destroyed another, a fire in the instrument panel over Kentucky struck another and a fourth was abandoned by its crew about 100 miles short of Diego Garcia in the Indian Ocean while returning from a bombing mission in Afghanistan.

Complete with video footage!

 

From Aviation Week's ARES Blog

 

Ever in search of instant gratification, we've edited together the USAF's various videos and reconstruction animations of the B-2 crash at Guam in February. The crash was caused by moisture in the air-data pressure sensors - a known problem with a known fix that was never written down, but instead passed (or not) by word of mouth. You can read Dave Fulghum's Aviation Week story here, and the USAF accident investigation report here.

 

In the video your first see B-2 tail number 88-0331 make a normal take-off from Andersen AFB. It is followed by the incident aircraft, tail number 89-0127. You can see the bomber pitch rapidly nose-up. Because of the skewed sensors, the flight control computer thinks the B-2 is at its 140kt rotation speed, but it's actually at 130kt. The sensors were also indicating that the bomber was nose-down, so commanded the rapid pitch-up. The combination of slow airspeed and high angle-of-attack caused the aircraft to stall and roll to the left. As the wingtip hits the ground, the pilots eject.

 

The video also shows the USAF's reconstruction of the accident from two viewpoints, behind and ahead of the aircraft. Along the bottom are instrument representations showing altitude, airspeed, attitude and stick position. Then you get to see video of the crash from a different angle - again the normal take-off by aircraft 88-0331 followed by the dramatic pitch-up by accident aircraft 89-0127.

Seems it was 'moisture', rather the opposite of the originally reported fire , that brought down the Spirit of Kansas.

More photos of the crash have been released:

 

 

 

Well, there $2 billion worth of junk.

 

Based on the reporting it makes me wonder if we rely too much on computers. Perhaps there should be an override switch in the cockpit. But, all things considered, it probably wouldn't have made a difference in this case. ???

 

But then again in the future when every aircraft is a computerized UAV..... I feel safer already.

 

 

Steve M

Well, the computer was given bad information so it made bad decisions. That human complacency led to a situation where the computer was given bad information (by not doing needed maintenance on the probe) leads us right back to human fallability.

 

I don't happen to like the idea of not having a manual override but the highly unstable planes can't fly without the computer anyway so a manual override becomes useless if a human can't control the thing manually anyway. In this case maybe the pilot could have managed a nicer crash or even a decent landing but I don't know if it would have in the end justified the expense of providing an over-ride capability everywhere something automated could go wrong.