You are not logged in. Login Now
 0-9          
 
Author Message
russ
When the news raises a (short) science question, bring it here. Mark Unseen   Jul 3 03:45 UTC 2002

One of my recurrent complaints about "modern" education is that it
is so weak on science that even HS graduates (and many non-technical
college graduates) are unable to appraise and understand the facts of 
things they encounter in the news, in political debate, and even in
every-day life.  This leaves them unable to participate meaningfully
in discussions about the issues or even appraise the sides, and
forces them to make a choice based on charisma or agreement on some
other criterion (such as Political/Theological Correctnesss) rather
than the merits.

Here's my stab at providing a forum for laying out the merits.

First topic:  The DHL/Aeroflot mid-air collision over Europe yesterday.

There has been some question about the Aeroflot pilot's English
comprehension, and how much time he was given to respond to the
request to change altitude.  The first reports said the Aeroflot
pilot had 2 minutes, current reports say he may have had as little
as 1 minute.  How much time would actually be required to avoid a
collision?

My take on it:  Airliners are easily capable of handling 0.8/1.2 G
in maneuvering.  If the Aeroflot pilot had immediately pushed over
to a 0.8 G arc when he received the request to change altitude, he
would have accelerated downward at a rate of 1.96 m/sec^2 relative
to the earth.  The aircraft would have descended about 1 meter in
the first second, 4 meters after 2 seconds, 9 meters after 3 seconds,
and 25 meters after 5 seconds.

Few aircraft are 25 meters tall.  And even after pulling up to a
1.0 G flight path, the aircraft would have continued to descend at
roughly 5 meters per second until the pilot pulled up further to
level flight.  In short, it appears that a collision could have
been avoided by action as late as 5 seconds before impact - 60
seconds should have done the trick easily.  And a 0.8 G or 1.2 G
maneuver is hardly severe, even for air transports.

My appraisal is that the collision was the fault of the Aeroflot
pilot failing to act, and beyond that the failure of the EU to
require TCAS equipment on all airliners flying through its airspace.
(The DHL aircraft had Traffic Collision Avoidance Systems active,
the Aeroflot did not.)
9 responses total.
rcurl
response 1 of 9: Mark Unseen   Jul 3 15:26 UTC 2002

A report in the paper this morning said that the 757 transport plane
had a collision avoidance system that detected the Aeroflot plane and
told the pilot to descend, which he did. The planes collided while
they were both descending to avoid a collision.

(I thought that avoidance instructions also included both planes
turning (to starboard?). This would have helped avoid a collision.)
gull
response 2 of 9: Mark Unseen   Jul 3 17:27 UTC 2002

Re #1: That's what I heard, too...the traffic controllers told the Aeroflot
plane three times to descend.  By the time he finally complied, the 757
pilot had already started a descent requested by his TCAS.
jp2
response 3 of 9: Mark Unseen   Jul 3 19:03 UTC 2002

This response has been erased.

rcurl
response 4 of 9: Mark Unseen   Jul 3 19:30 UTC 2002

Correction accepted.
russ
response 5 of 9: Mark Unseen   Jul 4 04:48 UTC 2002

Corrections:

(1)  The Tupolev airliner was operated by Bashkirian Airlines, not Aeroflot.

(2)  The rate of descent after 5 seconds would be about 10 meters per
     second, not 5.  A prompt reaction would definitely have done the job.
russ
response 6 of 9: Mark Unseen   Jul 13 13:15 UTC 2002

What a difference a few days makes.  As it turns out, the Russian
plane DID have TCAS (I assumed it did not because it is supposed to
make this kind of accident impossible).  What happened:  the TCAS
systems on both planes negotiated avoidance maneuvers; the DHL
plane started to descend and the Russian plane started to climb.
The Swiss controller, out of the loop and acting late, ordered the
Russian plane to descend.  The Russian pilot should have ignored
Swiss ATC and followed the TCAS instructions, and all would have
been well; but instead he descended right into the DHL plane.

Compounding this:  German ATC saw the collision coming several
minutes ahead, but were unable to reach Swiss ATC on the telephone
hotline; the Swiss phones were partially out of order.
rcurl
response 7 of 9: Mark Unseen   Jul 13 17:49 UTC 2002

Do you  know of a reason why they don't  also have a turn for
both planes in the TCAS procedure? This could also have avoided
a collision even with the controller error. 
russ
response 8 of 9: Mark Unseen   Jul 14 01:51 UTC 2002

Re #7:  This isn't knowledge, just deduction:  TCAS works using the
altitude-encoding transponders on the respective aircraft, so there
is good altitude information available to the systems.  The location,
heading and speed info is far less accurate if available at all.
Thus the system can make reliable climb/descend recommendations but
not steering recommendations.

In the days of VOR, head-on collisions were unlikely because the
accuracy of the nav systems wasn't sufficient to run two aircraft
into each other.  Now we have GPS, which is more than accurate
enough to do just that; improved accuracy has ironically made the
system more dangerous.  Offsetting flight paths by 100 meters to
the right of the center of air corridors would eliminate this.
gull
response 9 of 9: Mark Unseen   Jul 16 00:32 UTC 2002

There have still been a lot of collisions over VOR beacons.  It's one of the
more dangerous places to be from a traffic avoidance standpoint, because
planes are converging on it from various directions.
 0-9          
Response Not Possible: You are Not Logged In
 

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss