Jump to content

Schiphol Crash


Recommended Posts

  • Aerosoft

According the the first (yet very official report) the cause of the crash at Schiphol was a faulty left Radar (Radio) altimeter. It jumped from +1950 feet to -7 feet. This caused the Auto Throttle to power down and that aircraft to pitch up for the flare. The pilots might have been a bit slow to pick on the problem (though the same issue happened twice before on that aircraft) and did not disable the AP fast enough to prevent the crash. Of course it is unsure if there WAS enough time.

As this was seen as such a major issue it was announced very fast and Boeing has been advised to inform companies who operate the 737-800 models to make sure all pilots know that any problem with the radar altimeter (in this case an inconsistency between left and right) makes a auto land impossible.

Now my question... how can a radar altimeter show a negative value? The B737 manuals I got here clearly state it shows altitude above ground and not altitude above ground corrected with actual altitude (the runway they were landing on is at -7 feet).

Link to comment
Share on other sites

Now my question... how can a radar altimeter show a negative value? The B737 manuals I got here clearly state it shows altitude above ground and not altitude above ground corrected with actual altitude (the runway they were landing on is at -7 feet).

That was one of the questions I had after hearing the official report yesterday. But that's what they're going investigate on now, though the further investigation can take up to a year.

Link to comment
Share on other sites

According the the first (yet very official report) the cause of the crash at Schiphol was a faulty left Radar (Radio) altimeter.

Well, after the radio altimeter failed, apparently the crew failed to notice during the following 100 seconds that the throttles were at idle, letting airspeed decay to 40 kts below Vapp. The real cause of the crash seems to have been human factors of some kind - the aircraft was perfectly flyable, and the crew certainly had enough time to react.

Tom

Link to comment
Share on other sites

There were two things puzzling me in the report:

First the fact that the RA is linked to the autothrottle at all. That is something I wouldn't even have expected from Airbus. Seeing a Boeing with such a link surprises me.

Second, the report states that the left RA failed at least two times during the previous flights, which were still in the flight recorder database. No measures to replace the RA seem to be taken.

Cheers,

Emacs

Link to comment
Share on other sites

  • Aerosoft
Well, after the radio altimeter failed, apparently the crew failed to notice during the following 100 seconds that the throttles were at idle, letting airspeed decay to 40 kts below Vref. The real cause of the crash seems to have been human factors of some kind - the aircraft was perfectly flyable, and the crew certainly had enough time to react.

Tom

Yes but that's the cause with the majority of accidents. In this one there must have been a lot of conflicting information to handle for the crew, but clearly they did not manage to distill the right conclusion and failed to act in time. You can be sure that in the end it will be clear the crew was trying to debug the problem and forgot flying the aircraft, it seems to be a very human thing. But it's hindsight to say that just flicking the AP/AT off was the solution.

I've been in some big sims where others had fun throwing these kinds of conflicts at me. Even when there are two different warning system lighting up it becomes hard to keep flying the aircraft. Nobody seems to know in detail what the system does with this problem, most certainly none of the 737 pilots I spoke yesterday knew exactly what would be the result of this problem. Why the left radio altimeter commanded the AT even though it was not in line with the other altimeters. In my laymans eyes that is a dangerous setup. Before the throttles are closed you would expect the system to be sure it is not doing so at near stall speed at 1900 feet AGL.

Link to comment
Share on other sites

  • Aerosoft
There were two things puzzling me in the report:

First the fact that the RA is linked to the autothrottle at all. That is something I wouldn't even have expected from Airbus. Seeing a Boeing with such a link surprises me.

Second, the report states that the left RA failed at least two times during the previous flights, which were still in the flight recorder database. No measures to replace the RA seem to be taken.

Cheers,

Emacs

Radar altimeters are the only way for an aircraft in autoland mode to determine when to start the flare. There is no other option.

And radar altimeters are not on the list of equipment that will ground an aircraft. The fact it failed twice before and was still used is of course questionable. But I'll bet you the crew also did not know a single radio altimeter could command the AT system.

Link to comment
Share on other sites

But it's hindsight to say that just flicking the AP/AT off was the solution.

I'm not saying that at all, please don't put words in my mouth.

Why the left radio altimeter commanded the AT even though it was not in line with the other altimeters. In my laymans eyes that is a dangerous setup. Before the throttles are closed you would expect the system to be sure it is not doing so at near stall speed at 1900 feet AGL.

A very reasonable question. I have no doubt that the Dutch Safety Board and Boeing will be looking into this very carefully. (Although the plane was well above stall speed when the retard to idle occurred).

But it seems to me that for some reason the crew had become too reliant on automation, allowing the failure of a single instrument to develop into an accident. If that is true, then focusing on the radio altimeter failure is sidetracking from a very important (and much more difficult) discussion about how flight crews behave in an automated environment. We are talking about three pilots, presumable well qualified and professional, from an airline with a good reputation for safety, who somehow failed in a very basic way to fly the airplane. How on earth could that happen? And what is to prevent it from happening on another flight, today or tomorrow?

Tom

PS: My guess is that -8 is the ground reading for the radio altimeter. It is calibrated to show the height of the main gear AGL in a pitch-up attitude.

Link to comment
Share on other sites

Radar altimeters are the only way for an aircraft in autoland mode to determine when to start the flare. There is no other option.

Well, what about the computer call out something like "retard, retard", instead of cutting the power by its own?

Cheers,

Emacs

Link to comment
Share on other sites

Now my question... how can a radar altimeter show a negative value? The B737 manuals I got here clearly state it shows altitude above ground and not altitude above ground corrected with actual altitude (the runway they were landing on is at -7 feet).

Hi Mathijs,

Radar Altimeters are designed to show the height above the ground of the main gear when the aircraft is at it's landing attitude. It is not uncommon for them to show a slight -ve value when horizontal on the ground.

I've been in some big sims where others had fun throwing these kinds of conflicts at me. Even when there are two different warning system lighting up it becomes hard to keep flying the aircraft. Nobody seems to know in detail what the system does with this problem, most certainly none of the 737 pilots I spoke yesterday knew exactly what would be the result of this problem. Why the left radio altimeter commanded the AT even though it was not in line with the other altimeters. In my laymans eyes that is a dangerous setup. Before the throttles are closed you would expect the system to be sure it is not doing so at near stall speed at 1900 feet AGL

Most (if not all) AT settups use a single RA for height information. The real question is why it took 100 seconds before the throttles were manualy opened to full and how a trained crew allowed the speed to decay to VREF-40 knots in that time. There will be lots of questions asked regarding the human factors in this case. The faulty Radalt should not have led to the downing of this aircraft.

Link to comment
Share on other sites

  • Aerosoft
I'm not saying that at all, please don't put words in my mouth.

A very reasonable question. I have no doubt that the Dutch Safety Board and Boeing will be looking into this very carefully. (Although the plane was well above stall speed when the retard to idle occurred).

But it seems to me that for some reason the crew had become too reliant on automation, allowing the failure of a single instrument to develop into an accident. If that is true, then focusing on the radio altimeter failure is sidetracking from a very important (and much more difficult) discussion about how flight crews behave in an automated environment. We are talking about three pilots, presumable well qualified and professional, from an airline with a good reputation for safety, who somehow failed in a very basic way to fly the airplane. How on earth could that happen? And what is to prevent it from happening on another flight, today or tomorrow?

Tom

PS: My guess is that -8 is the ground reading for the radio altimeter. It is calibrated to show the height of the main gear AGL in a pitch-up attitude.

Mmmmm that's not a bad suggestion. So if the radar altimeter would be blocked directly under the transmitter it would show something like that. Otherwise it would show plus something when the aircraft was on the ground.

And there have been many aircrashes caused by a crew that were trying to explain and debug instruments and forgetting to fly the aircraft. I would say that a third person in a two men crew enviroment could very well have confused the situation.

Link to comment
Share on other sites

I would say that a third person in a two men crew enviroment could very well have confused the situation.

Almost certainly. I would imagine this accident was ultimately due to a CRM breakdown not helped by the extra FO in the jumpseat.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. Privacy Policy & Terms of Use