About that B-2 crash in Guam last February...

[Larry Jacks]

sometimes, it's the small things that get you. This accident also shows what happens when there is excessive dependency on automation in any aircraft design.

From the press release (http://www.acc.af.mil/media/archives/story.asp?id=123101589):

Distorted data introduced by a B-2 Spirit's air data system skewed information entering the bomber's flight control computers ultimately causing the crash of the aircraft on takeoff at Andersen Air Force Base, Guam, Feb. 23, according to an Air Combat Command accident investigation report released today.

Moisture in the aircraft's Port Transducer Units during air data calibration distorted the information in the bomber's air data system, causing the flight control computers to calculate an inaccurate airspeed and a negative angle of attack upon takeoff. According to the report, this caused an, "uncommanded 30 degree nose-high pitch-up on takeoff, causing the aircraft to stall and its subsequent crash."

Moisture in the PTUs, inaccurate airspeed, a negative AOA calculation and low altitude/low airspeed are substantially contributing factors in this mishap. Another substantially contributing factor was the ineffective communication of critical information regarding a suggested technique of turning on pitot heat in order to remove moisture from the PTUs prior to performing an air data calibration.

Clogged air data inputs has caused quite a few crashes over the years. It may sound like a trivial problem but it isn't.

Here's a link (http://www.acc.af.mil/accspecialreports/b-2accidentinvestigationboard.asp) to the Air Combat Command accident investigation webpage for this crash. It has a couple videos that I hadn't seen before along with a couple animations based on data recorder information.

From what I can see, the plane over-rotated on takeoff. It didn't appear to stall but instead to get behind the power curve (a condition where the plane is typically in a high angle of attack that causes so much drag that even full power isn't enough to maintain altitude.). When the left wingtip hit the ground, the crew punched out. The plane impacted the ground seconds later.

[peteshimmon]

And to think they could just monitor the
rotation rate of the wheels to get an
accurate take-off speed:)

Glad the crew survived!

[Larry Jacks]

It isn't the speed of the plane across the ground that matters. What matters is airspeed as measured by the pitot tubes. This kind of accident is nothing new. Back in 1963 (http://www.blackbirds.net/sr71/srloc.html), an A-12 Oxcart (the single-seat predecessor to the SR-71 Blackbird) crashed from the same cause:

6926/ #123
Lost on 24 May 1963, Nevada. When water froze in the pitot tube causing an incorrect reading on the TDI(Triple Display Instrument). The aircraft (first A-12 to crash) stalled, then the A-12 entered a inverted spin, crashing 14 miles South of Wendover, UT. The CIA Pilot Ken Collins ejected safely.

Another example is the crash of the X-31A (http://www.check-six.com/Crash_Sites/X-31_crash_site.htm) high maneuverability test airplane back in 1995.

German Federal Ministry of Defense test pilot Karl-Heinz Lang, assigned to the X-31 International Test Organization (ITO) at Dryden, had just completed a series of test points on the 292nd flight of aircraft No. 1 and was headed back to the Edwards area when he noted conflicting air data displays and a Master Caution warning. This was followed by uncommanded and diverging pitch oscillations. The aircraft then became uncontrollable and Lang safely ejected at 18,000 ft. and parachuted to the ground
A Mishap Investigation Board studying the cause of the X-31 experimental aircraft accident concluded that an accumulation of ice in or on the unheated pitot-static system on the aircraft provided false airspeed information to the flight control computers, causing the aircraft to go out of control and crash.

Many modern military planes are uncontrollable without aid from the flight control computers. All of those designs are vulnerable to this kind of accident. Even ordinary planes can crash due to pitot tube blockage when it causes the instruments to display inaccurate information to the crew (as happened in this 727 accident (http://flyingsc.com/learn/res/impossible_climb.html)) or the autopilot to get confused.

In the case of the B-2, earlier crews had learned of the vulnerability and had developed a simple fix - turn on the pitot tube heater to evaporate the water. Unfortunately, the procedure wasn't documented and not everyone knew of it.

[peteshimmon]

Ah well...wheel speed and a rugged wind speed
device when stationary. Suppose the pitot
heater will have to be software monitored.
It seem greater sophistication in all
systems is the trend with many sources of
information for the computers. Electronics is
having more temperature monitoring of
components to try to anticipate failures.

[JMV]

Even ordinary planes can crash due to pitot tube blockage when it causes the instruments to display inaccurate information to the crew (as happened in this 727 accident (http://flyingsc.com/learn/res/impossible_climb.html)) or the autopilot to get confused.
In the 90's, I believe, a 757 crashed because the maintenance crew forgot to remove some sort of tapes that were used during maintenance to cover the ports that provide air data to the flight instruments. The plane was flying at night over water, so the pilots couldn't tell their airspeed or altitude. ATC couldn't help because they got their data from the aircraft's transponder.

[Larry Jacks]

That crash (Aeroperu Flight 603 (http://www.rvs.uni-bielefeld.de/publications/Reports/aeroperu-news.html)) was caused by a similar cause - blockage of the static ports causing inacurrate readings by the flight management computers and instruments. Maintenance crews had taped over the static ports to clean the plane and failed to remove it afterwards. The crew didn't catch the error during preflight.

[cjl]

Ah well...wheel speed and a rugged wind speed
device when stationary. Suppose the pitot
heater will have to be software monitored.
It seem greater sophistication in all
systems is the trend with many sources of
information for the computers. Electronics is
having more temperature monitoring of
components to try to anticipate failures.

The problem is that something like the B-2 actually requires automation - a pure and simple pilot controlled system would be completely unstable, and without the computer it would not be able to fly. Airspeed and angle of attack indicators are absolutely required in the case of the B-2 far more than most aircraft.

[Larry Jacks]

Here's a link to an AvWeb video (http://www.youtube.com/watch?v=8ZB-iziY2Bw) explaining the accident.

[hhEb09'1]

CNN: Air Force: Moisture caused $1.4 billion bomber crash (http://www.cnn.com/2008/US/06/06/crash.ap/index.html?eref=rss_topstories)Both pilots ejected safely just after the left wing made contact with the groundafter it made contact, that's gotta be some reaction timimg

[peteshimmon]

The point about computers being nessessary
with modern aircraft is not missed. It is
just that designers and programmers will now
have to go the extra miles for better safety.
After all you have millions of dollars
investment and lives dependant at times on
single measurements of parameters. I am not
sure in this case but it might have been
possible for the take-off to have been
aborted if software was checking a build up
of speed from the pitot tube mesurement as
to be expected.

Having said that, there are stories of
British aircraft grounded because of concerns
about unvalidated software.

Anyway this is just smart alec comments from
someone safely at home. Presumably folk are
grapling with these problems.

Wonder if they still have to use ADA?

[captain swoop]

Having said that, there are stories of
British aircraft grounded because of concerns
about unvalidated software.

8 Chinook Mk3s. gthey have been grounded since they were purchased by the RAF 7 years ago as the Software doesn't meet UK Airworthiness standards.

They are now having their Cockpits reverted to Mk2 spec.

http://news.bbc.co.uk/1/hi/uk/7434763.stm

[aquitaine]

Heavy aircraft automation has always been part of American design philosophy: Build highly unstable aircraft with sophisticated fly by wire flight control systems.

In contrast Russian/Soviet designs were always stable, yet at the same time extremely maneuverable and do not depend on fly by wire.

[Kaptain K]

Heavy aircraft automation has always been part of American design philosophy: Build highly unstable aircraft with sophisticated fly by wire flight control systems.

In contrast Russian/Soviet designs were always stable, yet at the same time extremely maneuverable and do not depend on fly by wire.

I wouldn't say always! Just since computers made it possible.

Also, stability is a relative thing. There's "civilian" stable and "military" stable.

[Larry Jacks]

Heavy aircraft automation has always been part of American design philosophy: Build highly unstable aircraft with sophisticated fly by wire flight control systems.

If by "always", you mean since about 1974 when the YF-16 prototype began flight testing. It was the first fighter designed to be unstable and use fly-by-wire technology.

In contrast Russian/Soviet designs were always stable, yet at the same time extremely maneuverable and do not depend on fly by wire.

The newest Russian fighter designs incorporate fly-by-wire technology to achieve their remarkable maneuverability.

[Wear Protection!]

Heavy aircraft automation has always been part of American design philosophy: Build highly unstable aircraft with sophisticated fly by wire flight control systems.

If by "always", you mean since about 1974 when the YF-16 prototype began flight testing. It was the first fighter designed to be unstable and use fly-by-wire technology.

In contrast Russian/Soviet designs were always stable, yet at the same time extremely maneuverable and do not depend on fly by wire.

The newest Russian fighter designs incorporate fly-by-wire technology to achieve their remarkable maneuverability.

then why do they try to copy us?

[Larry Jacks]

then why do they try to copy us?

I'm not certain what you're asking. If you're asking why to the Russians "try to copy us" in regards to adopting fly-by-wire technology and dynamic instability in their planes, it's because a fighter that is naturally stable will be less maneuverable and to a degree less efficient than one that is naturally unstable but needs artificial stability to make it flyable. Fighter planes are some of the most highly optimized designs out there. Fly-by-wire can give the plane a competitive advantage. That's why most if not all new fighter designs use it - to not have the technology puts you at a disadvantage. Coming in second place in a dogfight means punching out if you're lucky or dying if you aren't.

Fly-by-wire technology allows some planes to fly that simply wouldn't be flyable without it. The F-16 as designed would be uncontrollable without fly-by-wire, as would the F-117 and many other modern designs. Developing a good fly-by-wire system isn't easy, as evidenced by crashes of the Saab Grippen prototype, the YF-22 prototype, and other planes. Look at planes like the X-47 UCAV and you'll see a plane that would be unflyable without this technology.

Back in the 1940s, Northrop designed the XB-35 and YB-49 flying wing bombers without fly-by-wire technology. The planes flew but the stability wasn't the greatest, so they used autopilots to enhance the stability. It worked but there were still problems (Edwards AFB is named after a YB-49 test pilot who was killed in a crash).

It's possible they could've designed the B-2 so that it didn't need all this advanced technology to operate but the plane would've been less efficient and probably less stealthy. Instead, the incorporated the latest technology available in the 1980s to make the plane work. Unfortunately, that technology makes it vulnerable to pitot and static port blockages. This problem is nothing new - if you note my earlier post in this thread, a pitot blockage caused the crash of an A-12 Oxcart in 1963. This kind of thing has caused several accidents over the years. The B-2 crash is just one of the latest (and most expensive) in a long string of similar crashes.

[HenrikOlsen]

Saab Grippen
Gripen :)