PPRuNe Forums - View Single Post - Adam Air B737-400 fatal crash January 2007
Old 3rd Aug 2008, 09:14
  #316 (permalink)  
pacplyer
 
Join Date: Nov 2006
Location: Asia
Posts: 183
Likes: 0
Received 0 Likes on 0 Posts
punkalouver,

I only flew the classic 737 briefly, so my display and nav switching knowledge is lacking on this model. Also I'll use ADI instead of glass names.

Loss of Nav:
Reads kinda like without a warn/fault light on either of the IRU's the F/O was confused about which one deserved to be switched to ATT. This freaked the Captain because he wanted heading info punched into his IRU from the one the F/O just killed!

But yelling to go back to NAV was a real systems knowledge failure on the part of the Captain. ATT is a one-way-ticket, as others have pointed out; once you go there the horse has galloped out of the barn and it can only give you reduced presentation attitude info for the rest of the flight.

Some great earlier posts in this thread. Looks like the CAM recording has disappeared off those links on the net, so there must be some damage control going on?

Looks like these guys were handed a very unenviable situation. One or two bad IRU's (since it sounds like all this airline did was swop them when there was a write up) possible improper differences training (all-too common today,) possible inadequate abnormal checklist distribution.

Loss of A/P: (I'm assuming the A/P was lost as a result of the #2 MSU selected to ATT by the F/O)
But assuming a functioning SAI, (can we on this ship, PK?) both guys, if they're any good, should be doing a cross scan at all three attitude displays: Cpts, F/O's and the SAI before platform alteration occurs. #2MSU is switched to ATT:A/P lets go. A/P horn is cancelled. When the bank angle warning goes off, your primary horizon reference can no longer be trusted. When the altitude alerter tone goes of at ~250' off assigned, you're probably already doing unintended steep turns and losing altitude with increased load factors, who knows. You've had two clues that you're losing it; it's time to abandon the Captain's ADI and go back to the SAI (the F/O's ADI is not usable for 30 secs, right?).

Hard core instrument pilots flying junk call this: maintaining critical triangles of agreement.

But, because of good automated instrument comparator systems today, this old stick and rudder skill (cross-cockpit scan) is sorely lacking in todays air line pilot. Not all the guys I flew with in later years were even capable of flying without a flight director, let alone hand flying raw data across the cockpit at altitude for very long. If you don't keep current hand flying you will pay the price one night when "Outto" decides he's had enough of this flakey maintenance at this place and drops out when ATT is selected.

Ideally, before you start changing anything, you should have an idea which of the three horizon displays is presenting good info. You should then verify that your chosen primary horizon ref (for example: the Captain's ADI) is selected to it's normal display source: (e.g: The #1 IRU?) Instrument switching (which they never even got to) can be tricky, and before you get to restoring displays or even altering platforms as they did, the handing pilot has got to have faith that his interim source of wings level info is working for him.

But you should realize that failures can be progressive, and that you have to keep up the critical triangles of agreement going. Compare both remaining horizons with heading raw data change.

The no checklist way:
Normally, you just vote the odd horizon out. In this case, knowing the ship history and the airline, that's not good enough. The Pilot Flying (Capt) should be verifying partial panel behavior. He should instruct the co-pilot to get a flashlight out and dig out the wiskey compass taking readings of it for a minute in S&L and make sure his selected attitude reference is not changing heading on him (sounds like it did.)

Again, today's modern airline pilot almost never has to do this. But in this case, I wouldn't feel confident in choosing either of these IRU platforms to supply either the attitude reference OR the nav function.

Sounds to me like: This flight was beyond the abilities of most auto-pilot dependent line pilots.

But the colleague that you criticize for hand flying all the time (John Wayne) would have made it to the bar alive maybe.

The above are all just my contrarian opinions only.

Last edited by pacplyer; 3rd Aug 2008 at 14:54.
pacplyer is offline