UK AAIB report into G-MEDG incident, Khartoum, March 2005

[The Nr Fairy]

Searched for other pertinent threads - couldn't find any.

The UK AAIB have issued a formal report into an incident involving Airbus A321-231, registration G-MEDG, during an approach to Khartoum Airport, Sudan on 11 March 2005. The report can be found here (http://www.aaib.dft.gov.uk/publications/formal_reports/5_2007_g_medg.cfm).

Surprised no-one's posted this already. I don't do big iron, but in the summary, when I read "The minimum recorded terrain clearance achieved during the recovery manoeuvre was 121 ft" I gawped.

[pilotbear]

bad place to go to, have done it many times. Anything can and does happen, you CANNOT rely on the navaids, and should always calculate a descent path to follow and monitor it.
have also read the report:ugh:

[Shore Guy]

Learn something new every day, or try to….
I was not aware that the protection boundaries of EGPWS were increased with GPS position sourcing……
“The EGPWS manufacturer expressed concern about the sub-3 nm CFIT risks.
The company advised of the need for a direct link between the GPS and the
EGPWS, the benefits of which are concentrated in this risk area.
A direct data link from the GPS would allow alert/warning boundaries to
operate up to ¼ nm from the runway instead of the 1 nm in use on non-GPS
aircraft. Using an FMS as the EGPWS positional sensor, even if the FMS is
using a GPS feed, does not provide uniform results across the large spread of
FMS products being flown and is subject to the vagaries of the many different
radio aid environments at airfields. Other solutions that feed GPS data to the
EGPWS via other systems also increase the chances of data being corrupted and a common mode sensor failure proliferating errors across the aircraft systems.”

[yetanotherdawn]

PB, was not the calculation of a descent path and the subsequent attempts to follow it how this whole incident kicked off in the first place? It might appear that the actual execution of this intent was a little awry after the pilot perceived that the autopilot was not doing what he expected - the, "What the f*** is it doing now?" scenario I would imagine. You always need to get the ducks in row in that place and many other African destinations, as you point out you never really know what navaids are going to work, but if the crew were not familiar with the place they are not going to be pre-prepped for the range of potential disasters lurking there.

Other safety recommendations that could have been considered are to improve the disemmination of nav data to the various agencies outside of Sudan, more stringent efforts to ensure that the ILS and other navaids work more reliably and ensuring that the controller is presented with and uses met information that is current and not, from personal experience, up to 4 hours old. That's always assuming there is any money left after funding their various military adventures of course.

I wonder how much fuel was left in Port Sudan after 3 approaches.

[747-436]

I wonder how much fuel was left in Port Sudan after 3 approaches.

As they came from Amman I would imagine they were tankering fuel for the return so they may well have had quite a lot of fuel left.

[Dan Winterland]

Quote Shoreguy "I was not aware that the protection boundaries of EGPWS were increased with GPS position sourcing……"

The EGPWS gets it's info from the FMGC1 position. At that stage The ANP should be less than RNP of.3NM and that is the only parameter (as far as I understand) that the EGPWS could use. I think what they are getting at is that on some other aircraft types, EGPWS takes it's position directly from the GPS, or in the case of one aircraft I've flown, from it's own dedicated GPS - and these are more accurate.

The Terrain Clearance Floor (TCF) parameters are from 400' AGL at 5NM, reducing to ground level at the runway threshold. Not very much! You can see the problems with an approach in GPS Primary when the aircraft doesn't downgrade the status until .3NM.

Managed NPAs are very clever, but you really have to know what you're doing.

[twistedenginestarter]

Managed NPAs are very clever
The Autopilot even does NPAs now? What are you boys paid all that money for?

[212man]

Out of interest, I went and loaded the approach into our FMS trainer, see screen shot below. Certainly in this case, it can be seen that HASAN to KTM (MAPt) is 4.4 nm in the FMS, despite the plate showing 5d. It can also be seen that the company DA (MDA+50) will be reached approx 0.5nm before the MAPt as the encoded crossing altitude is 1498 ft.

http://i202.photobucket.com/albums/aa50/S92ctc/HSSS.gif