Air New Zealand vol 60 : Un ILS vers nulle part

This is a YouTube video titled "Air New Zealand vol 60 : Un ILS vers nulle part" uploaded in September 2021. It features a detailed aviation narrative by Jean Claude Van Day explaining a critical incident involving an Air New Zealand Boeing 767 approaching Apia Airport in Samoa. The video discusses technical aspects of the Instrument Landing System (ILS), the challenges faced during the approach due to faulty ILS calibration, and how the pilots managed to safely land the aircraft despite the malfunction. The content is a monologue focused on aviation safety and technical explanation rather than a conversation.
Questions this video answers
July 29, 2000
Air New Zealand Boeing 767
165 passengers
Apia Airport, Samoa (Western Samoa)
The glide-path (glide) transmitter
Foreign hello, my friends. Jean Claude Van Day when you trust trust, you become confident. And in aviation, when you become confident, you're not far from swallowing your birth certificate. So this story takes place on July 29, 2000. It's an Air New Zealand Boeing 767. It leaves Auckland to go to a remote place where there's not a fly on a map of the Pacific. This is what we call, at least until 1997, Western Samoa. The state of Samoa, from its capital to Pierre, is made up of volcanic islands. This means that the terrain can be quite high on these islands. They are sometimes as wide as they are high. And in fact, the approach equipment at Apia Airport consisted of an ILS and instrument landing system. In fact, these islands were often covered in fog and they are by the sea relief. Humidity coming in from the sea, you can quickly imagine you can have very low clouds or even fog that completely covers the island. The plane had 165 passengers on board, plus a certain number of crew members when it began the approach. When it approached the island, the 767 was in total darkness. It was almost midnight local time. And on top of that, there was no moon at that time, so it was completely dark. Add to that the clouds that I already mentioned. So the plane follows a sort of arc. An arc materialized by what is called udme. It is a distance measuring device. So it will follow an arc that will gently lead it to intercept the ils, the instrument landing system. If you ask pilots which instruments they trust the most for an approach, in order, the king of instruments is at the top of the scale, which is the ils. Then comes the VOR approach to the dme. And the last approach, in terms of level of confidence or comfort for pilots, is the NDB approach. That day at pia, they had an ILS approach. So it really is the Rolls Royce of approaches. The 767 has three autopilots that intercept the Runway centerline, the glide path, you know, it's an approach where the ILS provides two pieces of information. It provides information on the Runway centerline, so it allows you to catch the Runway centerline. And it also provides information on the glide path. It allows the aircraft system and the pilot to know where they are on the perfect glide path, which is generally 3 degree glide when there are no other obstacles. Like here we have a Runway facing the sea, 3 degrees glide path and good glide. So the pilots intercept what is called the localizer or the lock. I'll call it the lock. It's a System that gives the axis of the Runway, right? It lets them know if they are on the right or the left. There they are exactly on the axis, and then they wait for what is called the glide to arrive. The glide is the part of the ILS that indicates the glide path. The glide arrives, the autopilot is on it, and then the aircraft dives and starts to follow the autopilot by going, you know, like on an electromagnetic rail until the entrance to the Runway. That's more or less what was supposed to happen in theory, but it didn't happen like this. So to understand, let's go back over two or three theoretical considerations. How does the ILS work? In fact, if you want to know a little bit about all the components of the ILS, I'll provide you with ICAO Annex 10. In the description of the video, it's an NX. If you look at the first part, it's 500 pages long. It gives you a little bit of detail about these systems, its navigation aids. So in reality, the ILS has several systems. It's a set of systems. It's not a system. It's not like the vor, when we say vor, it's a system or the ndb, a single beacon. The ILS is a little more complicated than that. So at the end of the Runway where the plane lands, there are antennas, several antennas that are distributed across the entire width of the Runway that define what is called the lock, the axis of the Runway. On the side of the Runway, there's another antenna that defines the glide path, the descent path. There is also distance measuring equipment, a dme. Sometimes there are markers as well. There is also a monitoring system at the tower that sends an alert to the air traffic controller if there is a fault, if there is a problem with the ils, there are also mandatory redundant systems. This is what OCER says. There are systems, redundant transmitters and amplifiers. If, for example, the glide path stops transmitting, there is an immediate switch to a redundant system that will resume the signal. What does the signal look like? Let's take a look at the glide path. Now, the glide path was made in the 1940s. It was technology from that era, but it has of course evolved. It is made on the following principle. We have a descent path that is represented by a beam, a beam that is actually made of a mixture of 150 Hz and 90 Hz waves. So when pilots arrive at an airport, they will enter the so called ILS frequency. It is indicated on the ILS frequency charts. In reality, what they enter into the navigation radio is only the lock frequency. They never enter the glide path frequency. If you go to pilots who are on approach or ask them what the glide path frequency is, they don't know. They don't need to know. It's not even indicated on the charts. In fact, they enter the lock frequency. If you look at the ACI appendix that I mentioned, there are tables that say that for each LCK frequency there is a corresponding frequency associated with the glide. The lock works on VHF and the Glide works on UHF. There are 40 channels in all. In fact, there are 40 lock frequencies to which 40 glide frequencies correspond. OK, these frequencies are amplitude modulated by. So there is a carrier, and if we take the glide, there is a carrier that is in UHF between 328 and 335 MHz. But it is modulated by this signal of 150 and 90 Hz. So it is an audible signal at the base. We can hear it. We can hear it if we want. If we look for it, we can hear it. So the principle is as follows. When the aircraft is below the glide path, the ils receiving device that is in the aircraft will receive two it will receive 90Hz and 150Hz according to a certain interference pattern. But what it will receive is predominantly 150Hz. The 150Hz will dominate when we are below the glide path. So the more the 150Hz dominates, the lower we are. Okay. Now the opposite above the glide path. If we are above the glide path, the higher we are, the more the 90Hz will dominate and the 150Hz will become less and less present. What is perfection? What is the glide path? When we are in the glide path, we receive both angles, 90 and 150Hz equally. They are mixed in perfect proportion. Excellent. So when the plane perceives, when the aircraft sniffs the radio, this perfect mix of 150 and 90 hertz it will display exactly on the glide path. Okay, here's what happened. First of all, there was work going on at the airport. There was a truck with a probe that had already cut the cables connecting the ILS to the control tower. So the ILS was in unmonitored mode. That means if there was a problem, there would have been no alarm in the control tower. It's already a good start. The pilots knew it. They had read the instructions and they knew that the ILS was not manure, which made them wary of it, and they mentioned it several times. During the descent. The second problem, and this is extremely serious, is that the technicians had put the glide, the glide transmitter, in a so called calibration mode, because they were going to calibrate it that day. So they put it in calibration mode. They forgot it was in calibration mode. What does it do in calibration mode? In calibration mode, what do you have? You have the ils, the glide in particular, which will emit only the carrier part, with a perfect mix of 90 and 150 hertz. And that's it. There is no modulation that will correspond or a mix that will correspond to higher or lower than the glide path. What does that mean? If you put an aircraft equipped with an ILS in the reception field of a glide path that is in this configuration or wherever you put it, it will show you that you are perfectly on the glide path, that you are perfectly on the glide path, no matter what happens, higher or lower for it, you are perfect because it simply receives a perfect mix of 90,750. You see? Very annoying, that. So the aircraft arrives, and then as soon as it receives the glide path, well, the glide path displays that. There you go. It's perfect. You are in the glide path. So what does the aircraft do? Well, it dives. It starts to follow it. Finally following it. It follows nothing at all. Why is the plane descending to the base? Why is this plane descending? So here's why it's descending. Boeing explains this very well. It's that there are two algorithms for autopilots to follow the S. So it has two algorithms. There's an algorithm that's quite simple to understand. It's the algorithm that allows it to maintain the glide path. So the plane is descending. If the ILS says, warning, you're descending too fast, you're too low, what does the autopilot do? It tends to slow down the plane's descent a little. It tends to slow down the rate of descent. On the contrary, if the autopilot senses that it's starting to fly a little above the glide path, it's going to pitch down a little more to catch it. That's simple. But how does it catch it? What does it do when it's in horizontal flight and it has to catch the glide path? That's another algorithm, actually. So the algorithm works like this, very simple. The plane flies horizontally, maintaining an altitude to intercept the glide path. And when it sees that it is in the glide path, the autopilot sends a fixed pitch. This fixed pitch is around 0.5g. So it makes a small fixed pitch dive and then it switches to the glide path maintenance algorithm. So after this fixed pitch descent, it will start to analyze a loop from time to time regularly. It will say, is it enough or not? If it sees that this 0.5g pitch is enough, well, it keeps it. If it sees that it is not enough, it means that it needs to descend too quickly, it will pitch down more, it will increase the rate of descent. On the contrary, if he sees that this O5.5G makes him above the glide path, he will increase the rate of descent. But initially the interception starts with 0.5G forfetcher. So this plane behind New Zealand, this 767, what did it do? It simply arrives. It sees that it is in the glide path. Okay, great. What does it do? It sends 0.5G, it makes the plane pitch down 4fe te and then this forfay. It seems perfect. So the plane starts to descend. The autopilot says to okay, that's great. I don't have to correct anything. I'm in. So it continues to descend. Here are the pilots. There is an anomaly that they will notice at the start. They notice that the plane is going too fast. They have an energy problem. The plane seems to be descending a little too quickly for a glide path of 3 degrees GR. So what do they do to help it? They lower the landing gear to try to slow down the plane a little bit. To slow down the speed of the plane, they lower the air brakes. But they talk about it several times. They say, but that's not possible. There's too much energy. It's falling like a stone. It's descending a bit too much for 3 degrees gradually. QUOTE they're used to doing approaches like that called category three. There are three categories of S. There's category one, which brings you up to 200ft from the Runway, 200ft elevation. There's category two, which goes up to 50ft. And then there's category three, which doesn't even need to have a minimum height of 200ft above the clouds. It's an ILS that's even capable of guiding the plane when it's landed on the Runway, allowing it to maintain the Runway centerline if there's not enough visibility to do so. These people are used to doing Category 3 Israeli new shekels and this is too weird. So the port captain starts to rationalize a little. What does he think? He says, well, we have wind from back. It's true that they approach with a tailwind sometimes there's no choice. Because when you have a Runway and then there's a tailwind in one direction and the conditions are completely imc, so completely overcast, there's no choice. Sometimes the pilots don't have a choice. Sometimes they're forced to land with a tailwind. Since the plane is flying with the wind behind it, its groundspeed is higher. And so to maintain the glide path, it has to descend more quickly. That's the first point. Second point, the plane was very heavy. The plane was heavy. A heavy plane descends more quickly, of course. Why? Because it has to descend quickly to maintain the glide path. What's the connection if a plane is heavy? Well, it has to make the speed. The calculated approach speed will be higher. Obviously more lift is needed. So since it has a higher speed, it has to descend more quickly to maintain the same glide path. Rationalization after rationalization, it continues like this. At one point they are at 400ft, 10 kilometers from the airport. When the radio altimeter indicates 400ft, they are 10 kilometers from the airport. The co pilot looks to the right through his porthole. He sees a village. He sees that they are flying, flying over the village, very, very low. He starts to see the first houses there, little lights like that of civilization. And that's when the pilots think it's the straw that breaks the camel's back for them. It's too much. They put the power back on and they take off again over the ocean. They had a very difficult time because when they were at 400ft with the distance measuring equipment telling them you are about 10 kilometers away, 5.5, 000 nautical miles, you know, pilots trust the ILS with absolute confidence there. They almost tend to think that the ILS is right and the DME is wrong, the distance measuring equipment, the distance is wrong. Maybe he's the one getting wet. Maybe the distance isn't 10 kilometers. Maybe they're one kilometer from the Runway. Maybe they're almost at the threshold of the Runway. Maybe it's this distance measuring equipment that's needed. But in any case, what saved them was that there was also a third pilot who was with them in the cockpit, who took the map, who started comparing the heights with the distances. And at one point he's been doing it from the beginning, he realizes, and it's always good to have a third member in the cockpit, he realizes that they're not following the plane of descent, they're going below. They put the go around back on. And then they're very bothered because you can imagine how lonely they are. They're in the dark above the ocean with a plane that they don't trust because they don't know what's going on. Is it the instruments? Is it the plane? Is it them? Especially since they know that the landing will be done under instrument flight conditions. And it's not easy just to go to an airport. Here we are in the middle of the ocean. So what do they do? They do tests. They take off at 4,000ft above the ocean. They test the instruments and they do various configurations. They try. And then it slowly starts to redevelop, to reconnect with the plane, to redevelop confidence. It realizes that it's the ils, the glide part of the ILS that's messed up, and so they come back to land. This time, what they use, they use only the localizer, the lock, which gives the approach axis of the Runway. And then for the heights, well, they descend in steps. So they descend in small steps using the distance measuring equipment. So they do a DME lock. In fact, they ignore completely the glide path. That was wrong. They landed, everyone arrived safely, safe and sound. They were saved by their vigilance that day. They were also saved by the fact that they were a little far from the terrain, because there, all it would have taken was for there to be a small hill with three, three trees, and they wouldn't have made it home that day. So to tell you this is a little bit like the beginning of the video. Trust in aviation requires constant paranoia. You can't trust anything at all, not even an ils. In the last video, we saw an aircraft approaching Addis Abeba with a vortex that was offset by 20 degrees or 30 degrees in places. There you have it. A vortex can be wrong. Here we see that even an iOS can only trust itself. Thank you, my friends, and see you in the next video.
More from Aldo

La FIN du Royaume-Uni ? L'Ecosse, l'Irlande du Nord et le Pays de Galles veulent partir !

Analyse Présidentielles 2027 : Marine le Pen remet ENCORE le voile sur la table !

Scandale "60 Millions de Consommateurs" : la France mange des pâtes contaminées au Cadmium !

2 dossiers : Accident d'autogyre en Italie & un pilote sauvé par le parachute du Cirrus






