Re: Peruvian 757 crash -- possible cause reported

Date:         13 Nov 96 02:41:21 
From: (David T. Medin)
Organization: Rockwell Avionics - Collins, Cedar Rapids, IA
References:   1 2
Next article
View raw article
  or MIME structure

In article <airliners.1996.2359@ohare.Chicago.COM>, writes:
|> I agree that adding more hardware/software to a plane every time
|> there is a particular accident will add weight, complexity, and might
|> not make the plane any safer.
|> I just wonder about gps, however.
|> I own a gps that costs 300 bucks, and can display location, speed, and
|> altitude with reasonalbe accuracy.  Is every airliner equiped with a gps
|> now, and if so why not? This device could also prevent kal-007 type
|> accidents.
|> I am no expert, let me know what the deal is . . .

First, several small Russian airlines I've flown were using a small
general aviation GPS receiver as a primary means of IFR navigation,
supplemented by ground control radar. I was surprised to see
a Magellan handheld receiver sitting on the glareshield (one was a
Yak-40 jet and the other an An-24 turboprop). The pilots were quite
proud of the units, and said that they were far more reliable than the
NDB system used previously, and between GPS and ground-mapping radar,
were used exclusively in place of the old systems.

Anyhow, your $300 GPS may work fine for VFR cross-country flight, but
when you are using the receiver for primary IFR navigation with 100
passengers in back, the rules change a bit. When going into the air
transport category, the rules change even more, driving up the price
to where 300 bucks is a memory.

Things that you might not have considered that are important to an Air
Transport-category craft:

 o The unit must have a reliable means of detecting failure or
 degradation of the satellite constellation members it is looking at.
 Some of the algorithmic means for doing this are not easy, and
 the reliability of some of these means has been covered in
 Av Week (no date references--sorry). Assumption of the satellites
 being inviolate can't be made and short-term problems can't be
 assumed to be caught immediately by the satellite's BITE or the
 ground stations and placed in the almanac

 o The unit must have extensive self-monitoring

 o Sensors will be redundant both for safety and dispatch reliability

 o Sensors must be designed to support other nav systems than a
 simple display, including INS and flight management systems

 o The unit must have some means of detecting interference or
 jamming based upon position reasonableness, resulting in
 flags to the pilot in case of error

 o Integration of new electronics into an existing airframe
 is a tedious effort, and can be very costly. It isn't as
 simple as carrying in a handheld unit. There are compatibility
 issues, power issues, flight deck effects and safety, etc.,
 before regulatory agencies will allow a new system onto the
 flight deck

Remember that an airplane is primarily a collection of spare
parts flying in close formation, held together by paperwork!

       David Medin            Phone: (319) 295-1862
   Rockwell Collins ATD	      Internet:
     Cedar Rapids, IA