View Full Version : ILS telling lies

13th Jul 2006, 15:41
The ILS can tell lies (http://www.youtube.com/watch?v=0qaDVaKQSkQ&search=Air%20New%20Zealand)

Loose rivets
14th Jul 2006, 04:38
All together now....

"Go around!"

"Go around!"

"Go around!"

Capt. Okay, now we,ll do it again, this time number two, keep in tune!"

14th Jul 2006, 07:34
This is why we have a "Glide Slope Verification Altitude".

Ignition Override
14th Jul 2006, 07:41
Yes. An instrument 'internal failure', as happened with a doomed DC-9 in Switzerland, can show you on course even when the Loc "CDI" has failed. There might be no 'NAV' flag, or such.

At DTW Metro (DTW), we had a false glideslope appear a few miles before we intercepted the actual glideslope, but this appeared on the ADI but not on the HSI's secondary glideslope. Fortunately, the ceiling and visibility were not near minimums.

But next time? :ouch:

14th Jul 2006, 08:16
This is why we have a "Glide Slope Verification Altitude".

....And it's why it is unfortunate that many ILS procrdures don't have one:eek:

Loose rivets
14th Jul 2006, 18:50
Is it me, or did it seem to take an eternity to realise that they had an unacceptable situation?

Maybe the timescale was distorted for the film, but there we soooooo many clues.

14th Jul 2006, 21:33
Loose Rivets, the time scale may be a bit distorted, but you also have to take into account several factors before getting too critical of the crew. First of all, watching the video you are fully briefed as to the scenario unfolding, so you are seeing things that confirm exactly what you expect to see. The crew see some different things but they also confirmed what they expected to see i.e. an on slope indication, a good capture of the slope by the autopilot and a good ident. They did the right thing I think by getting the energy state under control first and this obviously was a distraction to the events unfolding. I think the majority our first instincts would be to get to an energy state where we can start configuring. Once this was under control and the Landing Checklist completed then they got back to cross checking the approach. As soon as they became aware of an anomoly they called for a go around. There were other contributing factors to the situation; black hole effect on the approach and this was all going on about 2am when none of us are at our sharpest even in the best of situations.
Maybe things were not done perfectly, but you must give credit to the crew and the airline for doing their best to make the event known to others. I think many do trust the ILS too much and if this sows a seed of doubt that helps prevent another close call then kudos to them.

14th Jul 2006, 22:22
Wizofoz - what part of the world do you fly in, boet? Every ILS approach I've ever seen has a published altitude for crossing the FAF. Third box from the left on Jeppesen's briefing strip at the top of the approach plate.

15th Jul 2006, 00:33
Suggest that anyone who is interested gets hold of a copy of, and reads, the report. ANZ (and I knew some of the investigators) put a lot of work into the investigation and the reports highlighted a number of tech problems with ILS technology which the tech-o community were aware of .. but the fliers .. not aware.

While it is often easy in hindsight to see the clues, at the time everything is quite dynamic and nowhere near as straightforward.

However, a very useful and educational incident to read up on ...

One thing which impressed me was the level of company support given to the crew .. ANZ had a good approach to "just culture" when it came to investigations.

Trash Hauler
15th Jul 2006, 05:54
It also highlights how our mind's work. We "weigh up" the information available to determine the best course of action. In this case the 'good' outweighed the 'bad' in the early stages of the approach. The crew had all the indications that everything was normal - no flags, localiser and glideslope captured and autopilots engaged. In conducting the DME crosscheck the mind would more likely conclude the DME was in fact in error not the ILS. 3 positive indications to one negative.

It took the combination of the higher initial descent rate, the outside picture and DME crosscheck to bring the doubts to a level sufficient to initiate go around.


Loose rivets
15th Jul 2006, 08:04
Firstly, I did feel pangs of guilt about my initial quip. I certainly would not have made light of anything resulting in the slightest of accidents. Also, I'm very aware of being a dinosaur, sitting in the comfort of my home. However....

While agreeing on a possible distorted time-line, I am still...let's say puzzled, by three obviously experienced guys not reacting positively to the situation sooner. It was seemingly a clear night, so perhaps they were quite sure of there being a safe margin from any obstacle. It would have been a logical reason to allow more time to sort things out. But things had gone wrong very quickly.

So much time was spent on ‘flight management': ‘Preoccupied' was the word used. That sounds familiar...NI and a wrong airfield I believe.

"Like a strong tail wind / heavy." Power reduction seemingly not controlling speed as expected, but I'm not too clear on the extent of course. But speed brakes as well? So it must have been a determined attempt to kill the speed.

At this stage I hope that I would have dredged up from my memory banks the training on false glide-slope. But that doesn't happen in the modern world does it? Perhaps the fact that the system was ‘Unmonitored', would be a clue that it was being serviced, so anything would be possible. But then, those old secondary lobes, when they did escape, were ridiculously steep. Couldn't be that...could it? ( I will read the report saspo.)

The DME factor I'll leave at the moment, I don't know if there were two separate units working to compare. But the unit speed of change is a very valid clue to their operation.

Then there was a ‘miss-mash' of lights, or words similar. Then island lights too close. Then ‘Where is the airport?' That belies the argument for good visual reference.

I do agree on the psychology issue. I recall going into AMS one hazy day, and both localizers were full scale...in opposite directions. The first sensation was of disbelief.

I don't know why I have taken a particular interest in this incident. It makes a good CRM discussion topic perhaps, but I think it's a bit more important than that. I just have a gut feeling that this had the makings of a much more serious incident, several things all coming together, and rather too long for positive recognition of the seriousness of the situation.

15th Jul 2006, 08:29

Have a look at Amsterdam as an example. The box is there alright, with "No published alt" printed in it.