Go Back  PPRuNe Forums > PPRuNe Worldwide > Australia, New Zealand & the Pacific
Reload this Page >

Air Niugini 737 overun at Guam

Australia, New Zealand & the Pacific Airline and RPT Rumours & News in Australia, enZed and the Pacific

Air Niugini 737 overun at Guam

Old 27th Oct 2018, 22:46
  #181 (permalink)  
 
Join Date: Mar 2006
Location: On the water
Posts: 604
I'm not rated on the 737, but fly another modern EFIS jet. Does the 737 not also provide vertical guidance for VOR and NDB approaches? Notwithstanding company SOP or AOM limitations on how the profile should be monitored, one would assume they should have had an indication from their instrumentation as well as profile data on charts and GPWS that they were low on profile.
WannaBeBiggles is offline  
Old 28th Oct 2018, 02:01
  #182 (permalink)  
 
Join Date: Jul 2014
Location: Harbour Master Place
Posts: 576
There is vertical guidance (VNAV) for VOR and NDB, and yes you can cross reference DME/ALT scale. However, how valid is a check 0.25 nm prior to the runway when a DME is rated +/- 2nm? The problem is on a GNSS approach you have no other reference to cross check.

Cross checks cease have much meaning that close to the runway, in reality once you are below the minima you are almost entirely dependant upon the visual picture to judge where you are relative to the runway, monitoring ROD & speed. In fluctuating visibility and passing showers resulting in a loss of vis close to the runway you are in no man's land. The only safe option is a missed approach.

The difficult one is the last couple of hundred feet in rain as the forward visibility deteriorates temporarily with the changes is rainfall intensity, the picture keeps coming and going as you pass in and out of the shower. I would not take long to get 140' low if the sink rate goes to a 1000~1200 fpm as you are continuously changing focus from inside to outside. It's only a bee's d!ck change in attitude or thrust to go from being right where you want to be to be low. A sudden reduction in the visual picture at just the wrong time can end up like this accident quickly.

There was no mention of who was the PF,.
I did note the FO only had 368 hours on type.
Were the runway lights on? The Airport info page also makes the remark " Marking: precision, in fair condition". I wonder what it would have looked like under overcast and rainy conditions, quite grey, and similar to the water, making visual perception relative to the more difficult.
You are also dependant upon the support pilot to call deviations in ROD or speed.

You can see in the image the FO's windscreen wiper is stuck half way across the window, so it was likely on at the time of impact. Captains side not visible out of stowed position, however, without looking at the switch position, you can't conclusively say if it was on or off. But the fact the FO's was on indicates they had probably experienced rainfall during the later stages of the approach.




Lots and lots and lots of ways that this type of approach can go wrong. Lots of questions.
CurtainTwitcher is offline  
Old 28th Oct 2018, 09:31
  #183 (permalink)  
 
Join Date: Jun 2006
Location: Brisbane
Posts: 899
Originally Posted by CurtainTwitcher View Post
There is a difference between the capabilities of the aircraft, the design of the approach and the FMC coding.
Yes, you are correct. My mistake, I hadnít seen that approach chart.

I donít know how itís coded in the FMC, but the chart certainly makes it look like a dodgy approach. I am lucky enough to fly in a part of the world where almost every RNP or RNAV GNSS approach is coded to the runway, or at least very close to it in the case of GNSS.

This approach has an MDA of 420 feet at 2.2NM from the threshold. If you need to get to 420 feet to get visual, it wonít work on a 3 degree slope. You would have to fly it as a dive-and-drive, which is not a recommended procedure in my part of the world. Presumably the surrounding terrain coupled with LNAV GNSS design criteria makes it impossible to get any closer. Flying that approach in my airline would be done at 3 degrees, with autopilot off and flight directors cycled by the MAP (if visual). Not something you would do if you couldnít clearly see the runway.

Even without RNP-AR (which I have no idea if this airline is certified for), Australia has implemented LNAV/VNAV GNSS criteria at many airports, which gets the aircraft quite a bit lower/closer sometimes, and sometimes even lower than RNP-AR 0.10 (which I canít work out).

They are possibly lucky they hit the water. Otherwise they might have hit one of those hills.

Obviously the flight crew should not be fudging the visibily requirements of the approach (if indeed they were), but it would be interesting if procedure design and/or the cost of a better procedure design is a contributing factor. It also could be possible that the terrain would limit even GLS from getting any lower at DA, but at least GLS can remain coupled to the threshold, and could prevent this type of accident. I canít believe how slowly the world is adopting GLS. The FAA has apparently delayed GLS adoption indefinitely. It seems to be very expensive, and I donít understand why that needs to be the case.
Derfred is offline  
Old 28th Oct 2018, 22:46
  #184 (permalink)  
 
Join Date: Jun 2001
Location: sierra village
Posts: 148
Iím looking at the truly minimum rest the crew had prior to that duty day and shaking my head.
lucille is online now  
Old 29th Oct 2018, 02:43
  #185 (permalink)  
 
Join Date: Feb 2017
Location: Anywhere
Posts: 2
Originally Posted by lucille View Post
Iím looking at the truly minimum rest the crew had prior to that duty day and shaking my head.
Really ?
Let's see, 2100 local time arrival after two short sectors from base, followed by a departure at 0850 local the next morning for the two short sectors back to base. Crew hotel about 5-10 minutes from the airport.

So what would you consider to be an adequate rest period in Pohnpei for such an arduous duty ? 18 hours, 24 hours...........a few days perhaps ?? Just asking
Charley Farley is offline  
Old 29th Oct 2018, 09:00
  #186 (permalink)  
 
Join Date: Mar 2005
Location: Darwin and PNG
Posts: 873
Swiss cheese holes had been lining up for years.

Most of us said this was going to happen and it has, one only needs to search this forum to find the evidence.

And yes, Iím an ex PX Captain and it totally saddens me, like most others that the airline that was once regarded as being very safe particularly given the operating environment.

Seroius actions need to be taken immediately by the airline without politics involved to recover from the fallout from this accident if they are to rebuild their reputation let alone survive.

Ageing fleet, grubby aeroplanes and crappy inflight catering (compared to what it was 20 years ago) are only going to make things worse if they want to rebuild their commercial reputation.


Last edited by Duck Pilot; 29th Oct 2018 at 20:57.
Duck Pilot is offline  
Old 29th Oct 2018, 09:28
  #187 (permalink)  
 
Join Date: Mar 2005
Location: Darwin and PNG
Posts: 873
How far can you reduce the RNP down to using the FMS is the 737?

When I was flying the Dash 8 with PX it was drilled into the SOPs that the FMS RNPs were set to 0.03 if I recall correctly? And we were encouraged/taught to manually set up the VNAV profile for every approach which were all pilot calculated, not published or approved. Lots of potential for error in that little excercise, I always monitored the shit like a hawk as it was fought with danger in my opinion.
Duck Pilot is offline  
Old 29th Oct 2018, 10:32
  #188 (permalink)  
 
Join Date: Mar 2002
Location: Seat 0A
Posts: 7,769
Duck, I wouldn't worry, the proper FMS aeroplanes are nothing like the Dash 8. That thing is a dinosaur from what I have seen in the descent profile department.

If the 737 is anything like what I fly, you don't select the Required Navigation Performance, in terms of accuracy of tracking; the aeroplane will track as accurately as it can at all times, which is pretty damn good and well within any RNP. It's called the ANP, Actual Navigation Performance. Now, if the ANP drifts out to the RNP (which is either coded into the approach or set by the crew) then you get a warning. As alluded to in a post by Derfred I think it was, the 737 is deadly accurate: at Mildura, Virgin did a Zero/Zero landing on the runway using only a bog-standard cheap charlie RNP LNAV approach (old GPS NPA), similar to that published at Chuuk. As also pointed out by Derfred, this particular approach is a curly one because the MAPt is hit well before the MDA if you are on the 3į slope to the runway. You either accept that on a 3į slope (that could be flown in VNAV/AP) you will never get to the MDA, or you duck below the 3į to get down to the MDA first before the MAPt.

Last edited by Capn Bloggs; 30th Oct 2018 at 05:42. Reason: MAPt/MDA in wrong order...corrected.
Capn Bloggs is offline  
Old 30th Oct 2018, 05:08
  #189 (permalink)  
 
Join Date: Jun 2001
Location: sierra village
Posts: 148
Originally Posted by Charley Farley View Post
Really ?
Let's see, 2100 local time arrival after two short sectors from base, followed by a departure at 0850 local the next morning for the two short sectors back to base. Crew hotel about 5-10 minutes from the airport.

So what would you consider to be an adequate rest period in Pohnpei for such an arduous duty ? 18 hours, 24 hours...........a few days perhaps ?? Just asking
According to the report. Capt had 5 hours sleep, and F/O had 6. How so?
lucille is online now  
Old 30th Oct 2018, 05:37
  #190 (permalink)  
 
Join Date: Jun 2006
Location: Brisbane
Posts: 899
Originally Posted by TBL Warrior View Post


Just saying...
We have done a 10hPa altimiter error identical to your example in the simulator, and an EGPWS terrain caution was issued at around 200í RA.
Derfred is offline  
Old 30th Oct 2018, 05:43
  #191 (permalink)  
 
Join Date: Mar 2002
Location: Seat 0A
Posts: 7,769
I have heard of a nifty feature where a warning pops up if the crew don't set the QNH by Transition... That would be handy for NPAs...
Capn Bloggs is offline  
Old 30th Oct 2018, 09:53
  #192 (permalink)  
 
Join Date: Jul 2011
Location: Hong Kong
Posts: 56


78,77,73 logic same for Terrain warning...
Originally Posted by Derfred View Post


We have done a 10hPa altimiter error identical to your example in the simulator, and an EGPWS terrain caution was issued at around 200í RA.
Not possible according to EGPWS logic, unless flap was not set to landing or gear not down....
TBL Warrior is offline  
Old 30th Oct 2018, 13:49
  #193 (permalink)  
 
Join Date: Jun 2006
Location: Terra Firma
Posts: 167
Originally Posted by TBL Warrior View Post
Not possible according to EGPWS logic, unless flap was not set to landing or gear not down....
Yes it is possible. In Honeywell's EGPWS system, there is also a 'Terrain Clearance Floor' mode. From their EGPWS manual (my highlighting):

The Terrain Clearance Floor (TCF) function (enabled with TAD) enhances the basic GPWS Modes by alerting the pilot of descent below a defined “Terrain Clearance Floor” regardless of the aircraft configuration. The TCF alert is a function of the aircraft’s Radio Altitude and distance (calculated from latitude/longitude position) relative to the center of the nearest runway in the database (all runways greater than 3500 feet in length).
On short finals the TCF is a sloping plane (100'/nm) starting anywhere from 0.25-2 nm from the threshold increasing to 400'AGL at 4.25-6nm from the threshold. The actual distance depends upon the accuracies of the runway's known position and the aircraft's ANP. Even if you are correctly configured and on a stable 3į descent, you can activate a 'Tow Low Terrain' warning if you are below the nominal 3į path and penetrate the TCF. Derfred's Simex with a 10hPa (300') error perfectly illustrates the TCF mode in action.

Last edited by Bleve; 30th Oct 2018 at 13:58. Reason: Spelling
Bleve is offline  
Old 30th Oct 2018, 20:32
  #194 (permalink)  
 
Join Date: Jun 2001
Location: FNQ ... Still!
Posts: 3,343
The ATSB report into the incident at Kosrae VH-NLK noted the TCF mode activated several times late in the approach. That was a 737-300.
Report AO-2015-066.
Capt Fathom is online now  
Old 30th Oct 2018, 23:53
  #195 (permalink)  
 
Join Date: Jun 2006
Location: Terra Firma
Posts: 167
Originally Posted by TBL Warrior View Post
... there is no reference in any Boeing manuals to ĎTerrain Clearance Floor' ...
That's true, but that doesn't mean TCF doesn't exist, just that the Boeing manuals don't fully describe the Honeywell EGPWS system. Maybe you should ask someone in your technical or training departments about it.
Bleve is offline  
Old 31st Oct 2018, 00:27
  #196 (permalink)  
 
Join Date: Jun 2006
Location: Terra Firma
Posts: 167
Here's some (de-identified) slides from our GNSS training package for my Boeing aircraft:



Bleve is offline  
Old 31st Oct 2018, 00:37
  #197 (permalink)  
 
Join Date: Mar 2002
Location: Seat 0A
Posts: 7,769
Originally Posted by TBL Warrior
Shaggs, there is no reference in any Boeing manuals to ‘Terrain Clearance Floor’
This is what's in mine:
Enhanced Operation
Terrain Clearance Floor
The Terrain Clearance Floor (TCF) alert function adds an additional element of
protection to the basic GPWS by creating an increasing terrain clearance envelope
around the airport runway. TCF alerts are based on current airplane location,
destination runway center point position and radio altitude. When the TCF
envelope is penetrated:
* TOO LOW TERRAIN voice warning is generated twice.
* Amber GROUND PROX message is displayed on PFD.
* Additional TOO LOW TERRAIN voice warning will be generated for every
additional loss of radio altitude of approximately 20%.
TCF is active during takeoff, cruise, and final approach. The TCF alerts add to the
existing Mode 4 protection by providing alerts based on insufficient terrain
clearance including landing configuration.

Last edited by Capn Bloggs; 31st Oct 2018 at 00:45. Reason: Bleve's post above.
Capn Bloggs is offline  
Old 31st Oct 2018, 12:50
  #198 (permalink)  
 
Join Date: Apr 2002
Location: Oz
Posts: 244
How does Boeings "Integrated Approach Navigation (IAN) " interact with "Terrain Clearance Floor (TCF)"?

And a further question...Is this TFC a standard fit, or a Boeing "add on"?

Asking as I have had never heard of TFC in the Boeing until reading it here....well aware of the VSD and also aware of the various modes of the EGPWS, but TCF??.
Square Bear is offline  
Old 31st Oct 2018, 13:52
  #199 (permalink)  
 
Join Date: Mar 2002
Location: Seat 0A
Posts: 7,769
Another function of the EGPWS: "Terrain Awareness":
Terrain conditions are displayed and annunciated as follows:
* Terrain caution - Displayed in solid amber on ND, amber flashing GROUND
PROX on PFD, and voice warning CAUTION TERRAIN, CAUTION
TERRAIN (TERRAIN AHEAD, TERRAIN AHEAD - option)
Qantas 737s have this, as mentioned in the report into the Canberra terrain incident in 2004.
Capn Bloggs is offline  
Old 31st Oct 2018, 14:34
  #200 (permalink)  
 
Join Date: Jun 2006
Location: Terra Firma
Posts: 167
Originally Posted by Square Bear View Post
... a further question...Is this TFC a standard fit, or a Boeing "add on"? Asking as I have had never heard of TFC in the Boeing until reading it here....well aware of the VSD and also aware of the various modes of the EGPWS, but TCF??.
As far as I'm aware, it's not an 'add on', but a 'standard' part of the 'E' (ie Enhanced) bit of Honeywell's EGPWS. If you have EGPWS with the Terrain Alerting Display option enabled, you have TCF.
Bleve is offline  

Thread Tools
Search this Thread

Contact Us Archive Advertising Cookie Policy Privacy Statement Terms of Service

Copyright © 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.