What I find most interesting about this story is that... No one did anything wrong. No one made any mistakes, made a poor decision, broke the rules... Everyone behaved properly at all times and it was the actual system that let everyone down.
I’d be interested to see if the Embraer’s company has SOPs on doing high speed aborts. I’ll bet they do and I’ll bet that they don’t want you aborting for a bird strike in the high speed regime. If that is the case then the pilot flying was in the wrong.
@@stevenbeach748 bird strike is most dangerous on the runway if you take off and that engine fails when you are just starting to climb you might crash sully is an example of a double engine bird strike at low altitude one that turned out well most the time losing both engines right on take off leads to everyone dead because you don't have enough height to make it back to a runway
@@stevenbeach748 Pilot's decision on whether the aircraft is safe to fly is absolute. No person or company is going to question a pilot's decision to reject for safety before V1.
@@TWEAKLET Part 25 certified airplanes (this one) has to be able to take off on one engine after the engine fails on the runway. The plane must be able to continue to accelerate to rotation speed on the remaining engine and climb away from the ground at a 2.4% minimum climb gradient. Bird strikes, especially one bird, are no big deal and should not be a reason to abort at high speed.
@@MentourPilot Since my poor eyesight blocked my high school goal of becoming a pilot in our Air Force, your channels get me as close to being a pilot as I can get. 👍
@@stefanlaskowski6660 poor eyesight SUCKS. Its the most important sense and its so bad for too many people. Its awful. I also have poor eyesight and i like being in the nature outside of cities and having bad eyesight just sucks. Its so humiliating to know that i would have just died if i had poor eyesight in the stone age. I don't want to be dependant on some stupid glasses... my passion is about nature so imagine a tiny rain that would just refresh normal people but it basically blinds me. Or maybe have a surprise branch touch your face? Glasses either fall off or get scratches. Oh and don't even think about being able to see anything while swimming I HATE THEM SO MUCH Thanks for letting me tell you about it even thought you didn't let me and it was just my idea to comment this
@vandecayear10 I totally agree with you 💯 and sometimes I feel like I can now pilot a plane just by the way he explains every single step throughlly.Going back to this video it reminds me of the Tenerife disaster of 1977,although in that case the weather played a huge part as the pilots of both planes and the air control tower guys could not visually see the planes as the airport was engulfed with a lot of fog reducing visibility significantly and although a lot of lives were lost that day,it also brought upon a lot of significant changes to help avoid such issues again 😊.
I loved how the 'highly optimized ATC workflow ' was called out as a contributing factor to the near-accident. Taken out of context, that phrase may seem like a compliment ! Loved how you laid out the ATC operation context so that the rest of the story made sense ☺️
Highly optimized usually leaves little to no room for even small errors. We have seen that in the global supply chain. You always need to be on the lookout. I believe there was a pilot waiting for takeoff in the Linate disaster that noticed on his TCAS that an aircraft ahead of his had not gotten/gained any altitude and questioned his own takeoff clearance. Maybe this 777 should have waited to see the smaller jet ahead off him get airborne.
@@Stettafire in practical terms, yes, but in strict definition, no. "To optimize" means to make something more optimum, or closer to perfect. "Highly optimized" means very efficient. But in reality, pushing busy, intricate actions closer to 'perfect' when human actions and interpretation is highly involved quickly reaches the point of work overload and conflicting, simultaneous, even out of sequence, actions. In short, if *everything* goes *exactly* right, then no problem. But *any* hiccup means the whole thing collapses.
@@MentourPilot What software do you use for all these renders? I live 15 minutes from Toronto Pearson and this all looks very realistic! You can even see Mississauga downtown in the background. Truly outstanding!
@@MentourPilot can you please do a video on the famous Air France flight 447 that crashed into the ocean ? The one that was out into a stall by one of the copilots. If you have already too kindly lead me to the link . Thank you
I’ve always found the fact that they’re using comms on the runway that can block out another to seem like such an archaic system for such a high tech and safe industry.
I suppose it's actually rather difficult when you need to ensure prompt live communication from potential massive numbers of pilots to ATC and so driect calling of course wouldnt be practical, and theres no real way technically as far as i know that could prevent transmissions overlapping on an open frequency that you need to handle that. But I wonder if something like whatsapps voice notes would probably work. levels of urgency could be hardcoded into the message - Mayday calls would always be trasmitted live and overide any other queued messages. Pan-Pan level 2 priority, then after that could just go to playing out in the order received. From the users perspective anytime the frequency is open they can talk live and would notice no difference from normal radio comms. When there is a transmission coming through if you transmit it will just queue in and play like you started your transmission as soon as theirs ended, minimal delay. I could see in very busy airspace that it could lead to radio communication being delayed by a matter of some seconds, but to be honest its the same now anyway but with the additional risk of an overlap
agreed! I feel like critical messages (rejected takeoffs, takeoff rolls, etc.) should be transmitted twice? or would that just congest the air? through all of my Mentour Pilot binges I have noticed this quite frequently.
The RIMCAS system having to wait for a 50 knots threshold from a decelerating airplane is pretty dangerous in my opinion. The simple fact that the airplane is decelerating should indicating a rejected takeoff and immediately trigger the RIMCAS system.
@@Mountain-Man-3000 Why even use the speed as an airborne indicator. The only indicator for airborne should be the WOW sensor, at least from my point of view.
@@TrevorSmithy there *is* a button. It's automated - their transponder. Which worked as intended. The problem lies in RIMCAS programming, and the decisions and assumptions made in programming it.
Ah the “combined position.” This concept is probably familiar to many people in other professions. We certainly know of it in health care. This is where you keep cutting staff / costs until you achieve a really really good balance sheet. Ideally, some executive should get a promotion or a big bonus. I believe that the process can continue until a complete disaster happens. If you are really really lucky this will only cost a huge amount of money and prestige. But quite often it also costs lives.
I was looking for this comment and slightly disappointed to not find it among the top 3. Indeed, I can confirm that the process CAN continue (provided some executive keeps getting a promotion/big bonus) until a complete disaster happens. But then, you just roll with what @Jonathan D has suggested
Well, how else are stockholders supposed to accumulate more "record profits"??? There's only so much "planet earth" to go around and only so much of any market to fit on it. Investors don't like to invest without those magic words, "record profits" or they don't like buying the stock. AND of course, that favorite motto of them all, "If you're not growing, you're dying." ;o)
Very true. In this case it could certainly have led to disaster. In other cases it leads to early burn out and people not wanting to work under such ridiculous workloads and an even higher loss of, or failure to even find, experienced reliable staff. This kind of stupidity is seen in soooo many places these days. You know tick boxes to make sure that things are done right can be very useful, but if you don't give your staff enough time to actually check and do thinks right you make a right ****** of everything.
17:10 trucking instructor once told me "if something happens you are not prepaired for or havnt been trained for, go with your first instinct. Its usually right". thats what this pilot did, nobody got hurt, nothing damaged. Win in my book.
@@biscuit715 Been a while since ive seen the video, dont remember what it was about. BUT while a plane is taking off there are speed limits they have to reach by certain points. V1, Rotate, and V2. If something happens. before V1 the pilots hit the brakes. Something happens after V2 the pilots MUST continue takeoff as there is not enough runway to stop and at V2 the plane is already nose up and starting to fly. Anything happens between V1 and V2 is up to pilots discretion as above V1 they don't have enough room to stop but are still to slow to fly. Lose an engine after V1 but before you rotate, you dont have enough room to stop or enough power to take off properly so it is left up to the pilots descrition
It seems insane to me that an aircraft will report itself airborne before V1. There's no guarantee that it will actually become airborne up until that point.
Peter said that is for TCAS. TCAS is completely disabled for aircraft that are not airborne, I think. If the plane is about to be airborne, TCAS should be on.
@@thewhitefalcon8539The problem is using one sensor for two different systems, each of which increases safety by moving the sensor's decision point in opposite directions. This can't possibly work right.
@@KaiHenningsen Yes, but they can't retrofit all the planes to update their TCAS to send out a "maybe" on the ground state. Do the best you can with what you've got. Nobody is supposed to RELY on the runway incursion system.
@@knutaune724Piper Cub: “Hold my Beer” The first time I watched a cub take off I was taxiing a 172 on the parallel taxiway and I thought at any minute it would stall, lose control, and perhaps crash into me.
A simple suggestion for a solution would be a rule that a call-out for a rejected take off must be repeated until ground control acknowledges it. Along with that would be to establish an SOP to include a pre designated runway exit instruction (in case of rejection) to enable a safe exit from the runway at busy airports.
that may block the frequency for even more important situations such as mayday or fire... no? however repeating a couple of times seems to me (who knows very little) could be advisable. they may implent such a directive in case of relying on visibility
acknowledging communications works both ways. ATC is required to acknowledge the pilots. In my opinion the pilots who rejected should have done a better job making sure their extraordinary intentions were heard. They knew the airport was running tight departure. They knew that there was a takeoff clearance given and acknowledged on their runway. Then again, we are talking about seconds so perhaps this happened too fast for a repeat call to come out.
The Canadian approved procedure should never have been approved for use. Until the first a/c has its wheels off the ground , there’s no clearance for the second a/c to take off. It’s really very simple .
The pilots of the 777 definitely made the right call. Would they have had enough runway to take off? Maybe. Would they have had enough vertical clearance with the tail of the Embrear? Maybe. Would you want to gamble almost 500 lives on a maybe? Definitely not.
In 1971, there was a near disaster in Sydney when pilot of a 727 took off on a wet night over a DC-8 which had been taxing towards it because of a misheard communication and poor visibility from the tower. The 727 continued the takeoff because the its captain judged he could not stop in time. The 727 struck the tailfin of the DC-8, and while damaged, was able to keep flying and land safely.
This is why you shouldn't have a computer in control during anything but regular boring operations, because you can't program a computer to decide the outcome in drastic decision making conditions.
I 100% agree. The 777 would almost immediately have been in a position where they could not abort their own take-off - whatever happened - simply because they could not stop before hitting the Embraer. Given hindsight, imagine the 777 had also had a bird-strike (which might have knocked out an engine).
+ there was no guarantee that the 777 itself wasn't gonna have any problem with its takeoff, especially when the aircraft right in front of them on the same runway had to reject their takeoff which could've been (it kinda was) caused by the runway environment and not the aircraft in front.
I have a suggestion for another video: how passengers can be inadvertently left aboard planes after they land. The same airport, Pearson International in Toronto, had an incident a few years back where a woman who was asleep in her seat was missed by the cabin crew when they disembarked the passengers. The other passengers left the plane and then the cabin crew and pilots left and none of them saw this woman still asleep in her seat. Eventually, she woke up and found herself in a dark, unheated plane; it was winter and after dark. The plane was no longer connected to the terminal - it had been towed away to a parking area - and the doors were all closed. The woman tried to phone the airport but her phone battery was almost dead and she wasn't able to clearly explain her situation before the phone died. If I remember correctly, she finally got noticed because she managed to figure out how to open one of the doors and then shouted for help through the open door and someone driving a baggage carrier heard her and came to the rescue. I don't know how frequently such things happen but I thought it might make a good topic for a video if it is not a one-in-a-billion fluke. After all, any of us might find ourselves in such a situation and might benefit from advice on what we should do if it happens.
I knew a guy who fell asleep on his flight, the plane landed, some people got off but others stayed who were continuing one. But he was supposed to get off at that airport. Well, he wakes up either enroute or at the next destination because someone didn't check if everyone was off who was supposed to get off had actually gotten off.
Or he could do a story about the time they forgot a suitcase in Rome, and it flew all the way back to Copenhagen before flying back to Rome again! Scary stuff, the suitcase didn't have any phone or other ways to communicate!
Hugh, stop picking on Canada. Yes, someone missed someone who was asleep. Who wants to wake someone sleeping? I, pearsonably, would want to sleep by Pearson airport all the time.
Hugh Mungus Explain what you mean by "The woman tried to phone the airport" ? The WoMan either phoned the airport or she didn't - There is no TRIED. Whether the call was successful or not is another matter.
Also, I back the Boeing's decision to reject 100%. As you said, at this point, there's no time for complex decisions. There'd be 2 thoughts in my head "are we passed V1" and "can we safely reject"
When shit hits the fan and you are *below* V1, reject. It's that simple, it's what the entire purpose of determining V1 is for (and what the bigger system of V speeds is made for).
and also they were seeing the aircraft in front of them, but because they themselves were in motion they couldn't possibly know if they were going to a stop or just taking more time to become airborne. Must have crossed the Boeing's captain that they could become airborne right when they were under them I think
My step father was an Air Traffic Controller in Canada for a number of years. He's always the most tense one in the family when we travel by air because he says he knows better than any of us where things can go wrong.
For all the incredible safety systems involved in the aviation industry, marking something "airborne" when it's clearly not airborne is absolutely unfathomable to me
This is the root cause. Presumably cause of commercial pressure (as Mentour alludes to). And there is no indication this root cause was addressed (least, from the conclusions Mentour gave?).
@@PaulJakma I would make an argument that the root cause was having one controller manage two opposing runways. But again, are you going to require two ATC controllers every day? Thats probably the difficult balance of human safety vs practicality. I think the bulletin was used because you can make an argument that pilot decision is the safety redundancy in a scenario like this.
First off all, this video shows how absolutely useless RIMCAS actually is, if TCAS is deemed so much more important that airborne is given of before (and long before) actually being airborne. Secondly after the radio incident on Tenerife, i find it unaccceptable that radio messages are still able to be blocked. Seems as if for take-off rejection should be a rule about read back and or sending on secondary frequency. Thirdly the ATC assumed the Embraer to go airborne soon. That does not equal being airborne. And since assumption is the mother of all f..k-ups we should reprimande him at least a little bit. The biggest blame however is on the system in place where a useless RIMCAS gives false trust and a FAA organization still has not managed to deal with radio blocking.
@cessnadriver6813 in an ideal world yes but what specific regulation do you want to add and how will that affect airport operation and industry as a whole? Especially when ATC is dealing with their own labor shortage.
I went into this video going "wow, I haven't heard of this incident before!" Probably because nobody was injured. Sometimes it's nice to hear of almost-incidents that had happy endings.
That operations flow seems dangerous. It seems to me there's nothing spectacularly unlikely about that sequence of events, it could easily happen again.
It really isn't all that dangerous as long as the aircraft holds its T/O roll until the preceding aircraft is airborne (since you are responsible for visual separation from the preceding aircraft after all and you have no idea if they will reject the T/O or not like in this case). It only takes additional 30 seconds (or not even) to wait on the runway but still can get way more departures out this way over a long period as mentioned in the video on a clear VFR day.
That seems to be a pattern with a lot of incidents and accidents in North America. Their standard practices of pre-clearing aircraft for landing (and canceling later if the runway ends up being occupied for whatever reason) is another one that comes to mind, which is another thing that isn't done in europe. It seems they put too much emphasis on "efficiency" over safety over there. I know some US airports have a ton of traffic, but I'd rather be late than burning to a crisp on the runway because another aircraft slammed into mine.
It is still amazing that the communication systems still are subject to interference if used at the same time. Especially when you consider larger airports with potentially dozens of planes using it at once.
It is one reason why there are multiple frequencies, e.g. one for Ground, one for Clearance Delivery and one for the Tower, so that the number of aircraft trying to share a frequency is reduced. Pilots learn what overlapping transmissions sound like and should reissue any communication that they think might have been overlapped.
@@staticbuilds7613 We all get it, but using such old technology, especially in a large modern airport seems... like it needs an upgrade to the 21st century?
@@plektosgaming Most of the radios are 21st century. The reason this happens is because radios work on radio waves. You have one signal per frequency. You said "we all get it" but you clearly did not
There have been other incidents where the radio transmission to inform ATC of a rejected takeoff was stepped on by another transmission on frequency. It's relatively likely to happen, because it's an unexpected call at a time when ATC is likely to talking to other aircraft in sequence. The F/O doesn't want to wait for another aircraft to read back a takeoff clearance before making their call, because it's important that the tower is alerted to a rejected takeoff as soon as possible, before that other aircraft begins their takeoff roll. The radio is not a very reliable way to convey this information in a timely manner. It seems like the RIMCAS should also raise an alert if the aircraft's ADS-B ground speed drops by more than (let's say) 20 knots after accelerating above 50 knots. If the aircraft is slowing down significantly after initiating the takoff roll, it's not taking off, regardless of the air/ground state logic, and the controller should be alerted immediately. It also wouldn't be a bad idea to hit the transponder IDENT button on a rejected takeoff to help draw the controller's immediate attention, at least at airports with radar-equipped towers.
I would have thought a rejected takeoff transmission would be repeated if the tower hasn't acknowledged it - I did comms for a while, and important messages were always ack'd or readback to avoid them being missed. (edit: a word)
@@VosperCDN yeah, but non-comms people don't like that and don't want to take that extra step. comms people tend not to make it into leadership positions, so leadership tends to sympathize with the non-comms side of things and you get situations like this which then leads to an emphasis on instruction (treated as a new instruction) until it is forgotten and things repeat. same things go with backup frequencies. having a second frequency that the tower monitors in case your messages aren't going through due to high traffic volume allows one to get important messages through even when there is too much chatter going on. this means an extra radio for the tower, and another thing to remember and keep track of. if you are streamlining workflow this seems like an easy win to cut, but it always leads to problems at some point.
There was a suggestion to FAA from a collection of bush pilots asking for Full Duplex ATC & Unicom... Everyone transmits on freq. A, listens on Freq. B, except tower. Tower hears A directly and transmits on B directly. When tower not transmitting, the repeater puts out what it gets from A onto B. Everyone hears what ATC does... Which is only ATC when ATC is talking. FAA said it was too expensive for the majority of Aircraft to be refit. that was late 1980s.
@@WilliamHostman Wouldn't that have exactly same issue though? Per your explanation, everyone would still transmit on the same frequency, and for that case it genuinely doesn't matter if you listen on the freq A, freq B or you get a transcript via email - both sources still transmit on A. The only difference for your explanation between current and suggested, would be that if ATC would be transmitting, it would not retransmit planes to other planes. Which would technically be the same situation, assuming that tower is able to overpower the planes anyway.
@@tipakA you'd know you were "stepped on" in FDx, whe in simplex, you don't. The 1O of the smaller would have heard the heavy and known he didn't get heard, and be aware he needed to repeat.
What always amazes is me is the stopping on the runway. When I went through training, the ATC sternly talked to me about not stopping on the runway after my first solo landing. In either case, unless I literally can't move my aircraft, my priority would be to slow the aircraft down to ground maneuvering speed quickly, then get the hell of the runway, not sit there and cause a hazard.
I agree completely. I realize that the 190 crew was unaware that their "reject" call wasn't heard by ATC. I'm sure they were anticipating an "exit runway when able" call from ATC. That being said, I wouldn't have sat there stopped on the runway, response from ATC or not, completely vulnerable with an A/C that's totally capable of taxiing clear. I'm not going to criticize the 190 flight crew; I'm just saying that I personally (been doing this gig for 3 decades) would've taxied clear and held on the proper side of the hold line and then attempted to communicate with ATC. Unlike on a runway, traffic on a taxiway is moving nice and slow...that's where I want to be sitting with the parking brake set.
I totally agree, but it was in a matter of seconds, I'm pretty sure by the time the 777 decided not to take off, the 190 was still slowing down. All of this must have happened in 30 to 45 seconds, I guess. So it's difficult to slow down enough to a point where you still have time to turn 45 degrees and exit. It takes a while to do that while on a plane and without doing anything drastic like taxying outside the runway.
@@paulw4310 100% I would reject, without knowing it was a reject the most dangerous situation would be trying to fly over them while they rotate and climb into your underside or lose control from your wake. Also agree that my training and experience is to keep the active as clear as possible. Never come to a complete stop and exit at the first opportunity once slow enough. I haven't flown anything as bigger than a seneca, so maybe the rules are different for 190s and 777s. I have flown in/out of some busy airports, but I cannot think of a time that I started to roll prior to seeing the prior aircraft land or take off and totally clear of the active. Even if I had been given clearance, I think I would hesitate and wait.
If you have a rejected takeoff sometimes you have to stop on the runway. When landing you are required to exit the runway as soon as possible unless you have a problem.
I'm an air traffic controller. It's possible that there is an aviation channel out there where the host has a better grasp of ATC equipment, procedures and environment, and the ability to clearly explain it, but I haven't come across it yet. The amount of research Mentour puts into his subjects is clearly evident. As to the comments about the radio issues of multiple aircraft and/or ATC stepping on each others transmissions, this is a real problem in our ATC environment during busy periods. However, usually, transmissions don't overlap precisely in length so there is some indication to the controller that he/she missed a call and they can solicit that missed call. It was bad luck in this scenario that they apparently fully overlapped. Also, full duplex communications in ATC have the potential to wreak havoc on communications as it would allow multiple transmissions at the same time (imagine a party line phone call or zoom meeting) with no ability to distinguish any of them. Additionally, the current system, in the event of a stuck mic (stuck PTT) on an a/c, because ATC radios are more powerful than a/c radios, allows ATC to override the stuck mic transmission and still transmit to other a/c under its control, important from a safety perspective. And, in the event ATC needs to make an emergency/urgent transmission while another a/c is transmitting, we can "drown" out that transmission and make our urgent transmission by "stepping on" the currently occurring transmission. Finally, airlines' financial need to optimize efficiency does increase the risk of something going wrong, but that can be found in many aspects of the operation (for example, the decision to fly with less than full fuel on every flight (see Mentour's Singapore airline video). We can absolutely make the system safer by applying greater separation and increasing delay in the system (one in, one out?), flying with full fuel, etc., but that would come at increased cost to the consumer, which has not proven to be popular. In this case, the multiple layers of aviation security, which included the 777 pilots awareness of the situation ahead of them and recognizing that the preceding departure they were to follow had not gotten airborne and that they were closing on them instead of having increasing separation, and taking the appropriate action. worked. Apologies if this was too long or previously addressed.
18:26 Strewth, that plane is practically invisible behind the airport building. The re-enactments in these videos do a great job of recreating the situations, and giving a clearer idea of what is going on. I like how the incident reports maintain such a neutral tone; the 747 pilot isn't commended for aborting his take-off, better than that, it simply isn't mention. It is amazing how much potential carnage a single bird can cause.
I love that you broke this down Petter because it opens up so many opportunities to make aviation safer. How a system can show you as "airborne" before you're even at your V1 decision speed is ridiculous to me
I think its a matter of the system being generalized as possible. In short, its designed to work for every aircraft. From an a380. All the way down to a c152. So 50 kts makes sense. WOW could also work, but not all planes have that sensor. This is just my 2 cents as an air traffic controller. Ultimately since speed is really the only route to use it should be plane specific. still wouldnt be perfect but a lot better.
@@zachfield5336 The 50kts threshold is specific to the Embraer and it's the transponder which emits the airborne signal, the RIMCAS only interprets it. So it's perfectly feasible to have the transponders fallback to emitting airborne signal at V1 instead of a static value, especially curious that the Embraer system was programmed at 50kts. Apparently the threshold used is to activate TCAS, so when RIMCAS was developed, they just inverted the TCAS activation signal. So by increasing the threshold for "Airbourne", you increase the danger of incorrect warning suppression on TCAS.
@@DryBones111 Makes sense I guess. I just talk to the planes, im not an aerospace engineer. But i dont really see why TCAS and RIMCAS couldnt be operational at the same time.
Before I used to love watching NatGeo’s air crash investigations, but now I can’t stand them. I got spoiled by Peter’s incredible explanations, context of systems and how it all comes together in easy to understand pieces of information. The standards are quite high and I’m glad I’m constantly following this series in your channel
I agree... petter dont give us huge fireballs, burned bodies, teddybears sadly lying on the ground as a horrific witness to the disaster... He gives us the WHAT, WHY and the AFTERMATH, that hopefully ensure that this particular incident never happens again... Love from Sweden 💖 Å Ebba Blitz är obeskrivligt het...
Same. And when they do their cliff hangers or other guess what incredible thing happened? Most of the time I already know what the problem is because of these videos. Just so frustrating to keep watching such long boring story when it could have been told faster and with more interesting details
As a teacher I have to appreciate your way of explaining everything so well around aviation that even an amateur could understand! It definitely is an inspiration! :)
I think the second plane rejecting the take-off was a good decision, they had no idea why the first plane was not taking-off and in different scenario they might easily get into more troubles by taking off. Also if they got in any sort of troubles after committing to take off ,there would have been still plane in front of them severely limiting their options.
I can’t wait for you analyse the recent LATAM airplane accident in Peru, where a firefighters drill, ended up with them entering the runway where a plane was just getting ready to take off, and crashing with the plane. It was a miracle that only two people sadly passed, but the Peruvian management of these events really make you wonder about their professionalism
Sounds like poor planning and lack of coordination. Both have human error involve but this double take off incident has a lot to do with over reliance on technology.
This incident genuinely scares me because it shows that all the safety measures put in place to prevent another Tenerife can fail in what sounds like the exact same way with a masked radio transmission. While the training, safety, professionalism and culture in the industry is clearly very good and constantly improving, this story suggests that the necessity of automated systems gives a sense of security that can turn out to be false
(Haven't read the full report, probably mentioned) - Granted this was all a matter of seconds, but transmitting in the blind about a reject takeoff and not getting an acknowledgment from the tower should have been some high alert first officer making sure it was heard. I work in a semi-related field with critical radio transmissions all the time...and the number of times we get "I told you I was x status" - never came through. No ack = didn't happen.
@@gerbsvizsla Fog played a major role, too - and the lack of a Ground Radar in Los Rodeos at that time - also the overcrowding of this small airport due to the closure of Gran Canaria. It´s a complex story. @Mentour Pilot made a very good video about it in last December, please watch it here on this channel.
One easy fix would be to revoke the "AIRBOUNE" status if an aircraft starts decelerating... ...or to stop cost cutting in the ATC tower, that could also work.
In Canada at that airport, they do have an alternate runways, not sure why the rimcas system did not sound after a certain time, in refence to the closeness, of the three airplanes. They were on the same runway, actually three planes. Two of the airplanes did decide to abort there flight. It was perhaps due to the system that there was no incident involved.
With respect, I'm glad that situations are analysed even without a bad ending for those involved. This situation where a workflow causes the chance of an incident to rise is something that happens in many situations but these are very professional pilots who are trained to react appropriately in any case. Glad no one got hurt too
In most modern industries "near misses" and "accidents" are both investigated in the same manner (although with perhaps less urgency for the former) Happily, there are relatively few full blown accidents and investigators keep their skills sharp investigating near misses. You learn just as much from a near miss as you do with an 'accident'.
@@KnugLidi Systems are made to be reliable but only if they were implemented properly and within tolerance. The TCAST and RIMCAST system are working as designed but how they work don't scale up properly with the increase traffic flow of North America airports. Also, implementation of the systems miss out other assumptions, such as aborted take off and increase in traffic flow.
I mean... sometimes things like this just absolutely baffle me. How, in an industry focused almost solely on safety, does it get decided that "airborne" should mean _anything at all_ except for *actually in the air???* It seems like _someone, sometime_ should have audited that procedure, and thought, "Actually, you know what? Let's make it only say "airborne" when it's, I dunno, airborne?"
It's because ADS-B is also used to avoid in air collisions, and the possibility that the "in the air" sensor could have a false negative. RIMCAS was added later, and utilized the same ADS-B data which is used for TCAS. Maybe a future version of ADS-B could have a data flag for "I might be airborne, or on the ground, I'm not sure yet." so it could activate both avoidance systems during the takeoff roll. But it would have to be backwards compatible with older aircraft.
For plane it's the "turn on" of TCAS point, so for plane engineers it's safer to turn on TCAS earlier. Why RIMCAS engineers use "airborne" parameter as a single and true status of airborne airplane - this is the question.
@@brianorca That does make _some_ more sense... but it's still objectively the wrong call. Ground collisions are _vastly_ more likely than airborne collisions, and most of the worst collision disasters have been on the ground.
@@barefootalien Maybe they should find another term to replace 'airborne' when the aircraft is still on the ground but is in a critical phase of its take-off roll.
To me, the biggest surprise is how the Pilot Applied Visual Separation Rules in North America can allow for the air traffic controller to give take-off clearance to an aircraft, before ATC can visually confirm that the preceding aircraft is actually airborne. We're talking about a couple of seconds here. And while I understand that there are economic incentives for shaving off a few seconds, from a safety perspective, those extra seconds seem quite well invested.
Brings true tears to my eyes for the 777 Captain. Placing the safety of human life above and beyond what the dollar allows. Indeed, no time for complex decisions. The Captain of Embraer, ditto, same. Outstanding decisions by both. Definitely needs the same attention to safety as Sully on the Hudson. A wonderful presentation by our Mentour. Beautiful illustrations. Brilliantly thought out and presented. Pilot, thoughtful, illustrator and producer, to say the very least.
I find it incredible that die aviation industry is still primarily using a form of communication where sometimes messages just don't come through and nobody even realises that a message was just lost. Baffling!
aviation is in fucking shambles and it's cheaper to blame individual pilots and controllers than to actually make comms and management not a total shit show
They could use separate frequencies for transmit and receive to allow full duplex communications, but nobody would want to pay to replace all the radios.
@@jonathanma2741 You know we've had group calls for ages right? I understand AM radio has _some_ advantages compared to more sophisticated digital communications, but it seems the disadvantages far outweight them. At the very least they should use a dual system, so if the digital communication gets disrupted for whatever reason you can still fallback to AM (and this should happen automatically), but the normal workflow should be handled by the superior system.
I'm not in aviation but I find your videos interesting. You explain enough without over explaining and you are presenting these in an objective manner. This is what happened, this is what's normal, this is what was a bit off
Hi Petter, I really appreciate that you talk about the lessons of near-misses as well as accidents. It would be great if you could cover the serious near-miss of Air New Zealand Flight NZ60 at Samoa on 29 July 2000. This was an issue with a erroneous ILS glideslope, flying over water at night. When this "failure" happened, the instruments indicated they were perfectly on glideslope no matter what their altitude was! The flight descended as low as 340 feet, about 9 kilometres from the destination runway. This event has been discussed in Air NZ training but I don't think any other aviation commentator has covered it. There were a lot of holes in the swiss cheese, but fortunately they didn't all line up this time. Factors include the water "black hole" effect at night; over-reliance on technology; missing a checklist item while busy; an unusual "failure" mode of the ILS; the fact that GPWS is suppressed when landing; challenges of flying into remote island airports. The NTSB in the USA seemed to hold the opinion that this incident was primarily an issue of pilot error but others placed more emphasis on the erroneous and misleading ILS indications. Your opinion on all that would be absolutely fantastic. Cheers from New Zealand!
After viewing any of Peter’s videos the comments offer an equally valuable contribution to his presentations. Thank you all for your participation and insights into this experience. I encourage all to offer, in my point of view as I am someone who loves to fly but found it proper to discontinue my private pilot training when I experienced too many flaws existing at the time most of which have been addressed currently but the learning process is never finished. I apologize I did not mean to write a book here. Respect.
Hello! I’m actually having my solo flight next week! I’m nervous, but this is an amazing video! I’ve truly forgotten all my fears after watching this!😊
This is such a good channel. I live in Toronto but didn't know about this incident. I love the way you break down the events that combined to cause these kinds of incidents and make it easy to understand for people who aren't involved in aviation. Great visuals as well!
We in aviation are so good at learning from our mistakes and making sure that any future mistake hasn't already been made,… except for this. There's NEVER a good reason to make or allow for assumptions. An Embraer pushing 50kt is automatically airborne? A 777 at 100kt?? Hell, I can keep a 172 on the ground at 90kt if I push hard enough! This is a perfect example of why systems MUST only give ACTUAL instrument data, not projections or assumptions. Another fahn-TAHSS-tick video, Mentour. Keep your blue up and your brown down.
I can't believe that it was considered acceptable for them to have 2 aircraft taking off simultaneously on the same run way, that just sounds like a recipe for disaster. What happens if the following aircraft was more focused on the instruments and missed the aircraft in front stopping
@@lukepevensie Tenerife was in cloud with visibility ranging von 100 to 900 meter because clouds move over the runway. The controller couldn't even see the runway.
I agree. My first thought was what about if a takeoff has to be aborted. I had an aborted takeoff last week (student pilot) when a plane that was entering the downwind instead cut across the crosswind, opposite the direction I’d have turned, and we’d have reached the same spot in the air at the same time. Dumbass made no calls whatsoever. The field I was on wasn’t long, and there are 100’ tall trees 33’ from the end of it. I was almost at Vr and had to abort. The thought of another plane starting behind me makes my heart race.
RIMCAS seems to prioritize capacity instead of safety. In my experience, when modelling complex systems (which is part of what RIMCAS is doing) it is always best to keep everything as closely to reality as possible, so considering an aircraft "airborne" when it in fact isn't, just to decrease the number of "false alarms", is asking for an accident to happen. In other words, it should IMHO not be up to the ADS-B in the aircraft to fake the signal that it has become airborne in order to allow for its "customer" (RIMCAS) to squeeze the time between takeoffs - that consideration should be a concern of the receiving system. Any optimization should be done there. Also, when determining whether the rules are safe or not, one has to take into account that mistakes will happen, like the 737 stepping on the Embraer's transmission in this case. Here, there was one more slice of Swiss cheese that saved the day. I just wonder how many lives will be lost before this way of doing things is deemed unsafe and revised.
Not a pilot but that seems like a pretty smart fix; If "airborne" was just based on sensors that directly measure that factor like the weight-on wheels switch and radio/standard altimeters, which seems like the original design, and it was up to the receiving systems to apply their own additional checks to suit their individual use cases, that seems like the safer scenario. If RIMCAS didn't disable until the aircraft was 100% positively airborne, and TCAS activated if there was any reasonable chance that it could be airborne, their overlap would seem to offer an additional redundancy covering what seems like a fairly critical edge-case
I know I'm oversimplifying in saying this but wouldn't it be better if the RIMCAS/ADS-B airborne status were tied to the V speed settings instead of some arbitrary number such what was saw in this video? For example, pilots calculate their VR speeds and feed it into the FMC, that VR speed data is then somehow given to the ADS-B/RIMCAS as a better estimate on when the aircraft is airborne.
@@SVnerd or at least if a set number needs to be used it should accurately reflect the stall speed of the aircraft, there is no way in the world the embrair would be able to stay airborne at 50kn, but that should be a last resort
Seems like the best thing would be to no takeoff clearance until the entire runway is clear. Waiting until the plane ahead has cleared the end of the runway seems like it would only cost a few seconds. Starting a takeoff on an occupied runway just seems like asking for trouble.
I'm a military heavy pilot. I'm surprised we don't have an automated way of detecting radio heterodyne (blocked radio transmissions) and alerting the callers.
I still marvel at how dangerously bad VHF can be but also at how difficult it is technically to come up with an alternative - to find a part of the spectrum we could use globally and get everyone to switch over at once. Impossible!
Yes, indeed - bingewatching here! I am absolutely fascinated. The visuals and explanations are phenomenal. Thank you so much. I would also like to comment that both your warnings to sensitive viewers and your respectful handling of fatality reports shows a great deal of kindness,& properly honors the departed individuals. The fact that you make a point of showing what was learned from a tragedy makes one feel that the lives lost were not in vain.❤
In my aviating days I witnessed a number of incidents like this were a takeoff clearance was provided while runway occupied. In one case a KC135 aborted while an F4 was rolling behind it. The F4 in full ad just clearwd the tankers tail. My personal pre take off checklist was, heads up, runway clear ahead, left and right, then roll.
@@cupofjoen The speed activation thing is a backup in case of faulty weight on wheels switches that never detect the plane getting airborne, and thus don't activate the in-air TCAS system.
The answer is already in the video: "Increasing the 50/100 knots limit significantly increases the false warning rate which in turn make the entire system much less reliable"
@@dvdraymond having a speed activation alone seems very poor for RIMCAS - I can understand it activating TCAS but having it turn off RIMCAS seems dangerous (especially without another confirmatiory system - like change in altitude).
I don't know how many of your videos I have watched where something goes wrong because the air traffic controller can't hear an important message because it came in at the same time as another one from a different aircraft. There should be a warning system in place, to alert the traffic controller that another message came in from 'X' aircraft, so that they can contact them right away after they're done with the first message that came in. Or better yet, the system should record what is transmitted and alert the traffic controller that they have a message that came in while they were handling another communication, so that they can just press a button and listen to it. I don't know what the difficulty of doing this would be, but I think it would prevent a lot of problems.
thats not how it works,. the messages are not heared because they are not received. in essence, the more powerful signal destroys the weaker signal. there is no technical way for the sender, the overpowering sender and the receiver to know about what happened. there is no message to record, there is no warning to alert of a missed message. surely there are ways to transmit information in a way where you can notice, record and replay messages. thats how the phone system works. however, that system is way more complex, and expensive, and error prone, which is why the old radio waves system is still in use and while not perfect has many advantages. you can more or less literally build a receiver out of the lint in your pocket and a ripped of button (not really, but its really basic and not sophisticated technology), and as long as you have power, it works, no other or moving parts needed... and you are right, this has certain drawbacks that need to be taken in account. one of them is you can not know if your message has been received. this is where "over" and "over and out" comes from (making sure the whole message is heard and signaling others they can now talk) and why its important to acknowledge receipt of a message like "heard" or reading it back. its also why radio chatter needs to be kept at a minimum. you only can hear one sender, and when someone else is sending, nobody else can be talking. or they could, but only the strongest sender is being heard. while the video did not make mention of this, and i dont know the rules, that the air traffic controller did not acknowledge the firsts plane abort message might have been where you could have saved the day here. the pilots could have repeated the message because it was not acknowledged.
Still ridiculous why aircrafts are still using analog radio for communication where only one person can speak at the same time while everybody has a smartphone which can handle easily two or three persons speaking at the same time. Digital radio would also allow to identify each aircraft which is sending by adding its call sign to the digital data
@@catwiesel_81 "there is no technical way for the sender, the overpowering sender and the receiver to know about what happened" There is an easy way to detect this. If somebody is sending something his receiver must verify that he receives (nearly) the same signal which the sender is sending and if not create a warning that somebody else used the frequency for sending at the same time. "surely there are ways to transmit information in a way where you can notice, record and replay messages. thats how the phone system works. however, that system is way more complex, and expensive, and error prone, which is why the old radio waves system is still in use and while not perfect has many advantages. you can more or less literally build a receiver out of the lint in your pocket and a ripped of button (not really, but its really basic and not sophisticated technology), and as long as you have power, it works, no other or moving parts needed..." I never heared that an aircrew build an analog radio using "lint in the pocket" or "ripped of buttons" while in the air.... That may work on a ship there you have plenty of time but not in the cockpit of an aircraft. And our digital phone systems like 2G,3G,4G and LTE are working very good with hundred millions of users even the used hardware comes from many different companies - the only scenarios where it comes to problems is either power outtage or overload by to many users in the same cell.
There are a few ways to mitigate stepped on transmissions. The foremost being a response including the aircraft callsign and "roger" or "acknowledged" etc...this should alert a station that broadcast at a similar time that their transmission was either missed or stepped and and prompt another call. (Happens a lot). When things are moving quickly things get missed and stuff like this happens but there are procedures in place to prevent this so its not a constant safety issue under normal conditions (former pilot, current ATS)
@@elkeospert9188 works great until you surpass the incredibly limited range of cellular. Be it vertical distance or otherwise, you just simply don't have options for utilizing a cellular signal at 20+ thousand feet as I understand it. Ever get out in the ocean on a boat? You lose cell signal pretty quick. Plus what airline wants to pay a mobile carrier to handle their mission critical safety communications. Radios have very few failure points. You need a working transmitter and receiver. No towers, no fiber runs, no data centers, and importantly, much less latency.
What! I’d praise the 777 pilot, his decision was not “understandable”, it’s praiseworthy! There could be many reasons why the front aircraft aborted, like something on the runway caused damage to its tires etc. there’s no reason to brute force that takeoff.
on the 777 imagine if the T/O calculations were wrong or they lost one engine power after V1 and needed more runway space for the rotation, they made the better choice
@@Phvpark It seemed the Boeing didn't know by number where the Embraer was either. (The systems thought it was airborne and the Boeing wouldn't have had a number.) Was there time for the math even if the number was known? I don't think so. "It looks like we can stop -- let's do it" seems to be it. A new safety regulation was written in brown pants -- not blood, thank God! Do jet pilots practice takeoff roll panic stops? If not, should they? Is there some rough mnemonic for the normal panic stop distance vs. speed? Would a pilot remember it as such a rare thing to need? All we can say is whew, and don't crowd takeoffs like that when you don't have a near bulletproof collision avoidance system.
When I was a programmer I would explore every possible combination of parameters to insure that there would never be conflicts. "It's unlikely that would ever occur" was not good enough. It could not occur, period. We avoided the kind of engineering that would risk building a seawall too low to prevent a tsunami from flooding a nuclear power plant. "Unlikely" means "possible".
Excellent and thorough explanation. One issue that wasn’t mentioned is that the tower was designed for two Local Controllers when traffic requires and management did not staff it in accordance with that workload. We spend millions on design and sight lines and place highly trained professionals with excellent eyesight in towers in the sky just so they can use visual separation to safely move traffic. The second aircraft should never be cleared for takeoff until the tower controller visually observes the first one rotate or can ensure separation will occur. Giving the flight crew the responsibility to maintain this separation from another aircraft that can be as much as two miles away just because you want to staff one local controller is not ensuring safe operations.
Okay, serious props to you for your youtube channel allowing you to be able to write off the top tier beautiful airport and scenery addons for flight sim, and maybe even the gaming computer to run it.
I notice a theme in some incidents - that two or more people transmitting simultaneously on a frequency can result in information being lost. Wondering if anyone has looked at a digital packet switching system with, say, time division multiplexing could allow all the signals to get through. The tower would then have heard both transmissions.
I think there's just a reluctance to change. Consumer handheld radios have a 'busy lockout' function for a long time which stops you transmitting if a carrier from another transmission is detected on the same frequency, that would at least let you know that your transmission didn't get through so you can resend it.
Yes! This risk must be addressed. That could be done just at the receiving end, with upgraded tower equipment. Multiple incoming analog voices could be computer-separated using voice recognition or other clues, and then each stepped-on call automatically and immediately replayed (of course not conflicting with newer incoming calls).
The different messages are not arriving from the same azimuth so an array antenna could separate them. This is somewhat complicated by the very low frequencies, and therefore long wavelengths assigned to this fonction. Fundamentally it does not seem appropriate in the 2020s to restrict aircraft safety operations to using 1950s technology. This issue also arises in the conflict between 5G telecommunications, which uses 21st century electronics, conflicting with the radar altimeter which is likewise stuck using 1950s technology. There is no reason that the aircraft could not transmit the information using BOTH 1950s AM analogue transmission and 21st century digital transmission. It could even send a text message through the 5G network.
@@jamescobban857 you're right in one aspect, that technology could help both messages get through. This problem happens whilst aircraft are in the air too with the potential for similar dangerous issues. However if both transmissions get through the controller may not be able to tell who called or what they said due to the garbliing of the voices coming through at the same time. There would be no certainty that it would help much. In the air a controller may have up to 30 aircraft on frequency at any one time so this problem can be much bigger in en route sectors. CPDLC, (electronic messaging) does help reduce transmissions and is probably the way to go regarding this "call blocking" issue, even on the ground. The only minor error here was by the Embraer crew that reported rejecting the take off. They should have expected a reply from the controller. I realise they were busy in the cockpit, but when no response was heard, they should have repeated the message.
The animation is excellent, but I was a little startled to see trees in full green leaf in Toronto in February. I guess this shows how well I was taken in by the realism! ;-)
Wow! I've been up the control tower, and this is so accurate. Even the layout of the tower is correct, down to the staircase location. Super impressive video, I subscribed!
It really suprised me that this takeoff clearance situation was allowed in 2020. In the video, when you said "gave clearance and 777 was accelerating" I was immediately thinking "but if the Embraer needs to abort and stop" ... and then they had to abort and stop. But there is so much more puzzeling: 1) To me, in context with takeoff clearance, airborne should be defined as the plane being a couple of feet above ground and in a clear lifting situation. 2) A controller should not be allowed to look away and turn his/her attention until the planes are safely in the air and no longer related to traffic on the runway. 3) Having to watch and control traffic on opposite sides just feels like a stressful situation, prone to cause such situations. 4) And considering that all this is done to safe money ... I guess it is history that will repeat again and again. No amount safed can repay the potentially catastrophic outcome of such conflicts. So, I am happy that this was no longer permitted in Europe. And I fully understand the 777 pilot's decision to perform a hard stop. I wondered if they could have taken off in good time anyway. But there would have been the risk that they misread the situation and the other plane would have attempted a lift off anyway, in which case they might have collided in mid-air. Anyway, happy it did end well and always good that aviation looks into every incidence, even the ones that did not cause damage or loss of life.
Actually, it's just north america region that's built different. Not only in EU, but in pretty much every other part of the world (asia, africa, australia, etc), dual takeoff clearance is not allowed. ATCs are not supposed to give another TO clearance before the 1st aircraft reported airborne or identified on app radar. Heck even virtual skies such as VATSIM/IVAO which implements standard ICAO rules does not allow that. Regarding your #2, sometimes it's a bit dilemma when you have runway on both sides. As mentioned, the controller went to check arrival traffic on 05. If he keeps his eye on 06L, then he would have no idea whether the landing traffic landed successfully, vacated runway, or going around, etc. This is even more important when there's 2+ tightly separated traffic inbound ("highly optimized", quoting the report). The other points... yeah, money takes priority over people's life (and sadly that's how the world works)
#2 is wrong, there should be enough controllers to cover all runways in use, if there aren't, the excess runways need to be shut down until enough controllers are available to cover them all. the fact that someone is being expected to stand up and look behind them to make sure nothing crashes over there as well as in front of them means some manager or executive or administrator belongs in prison
@@0x73V14 I 100% agree to that. And I probably phrased #2 improperly. Often when I listen to Peter's analysis I get the feeling that the controll towers are understaffed. But to me this one has beaten most others. There should be one controller for each side. And yes - management responsibility.
While flying keeps getting safer, I’ve heard that the incursion incidents are on the rise. Is that true? If so, I’d love a video that describes why this might be so and what the industry is doing to better handle the situation. Perhaps more appropriate for the other channel but this video reminded me. As always, awesome content!
Over what time scale, we just had a global pandemic, which a huge drop in flights, so since things have opened up, of course we're getting more runway incursions, simply because things are busier,
I’ve usually heard it as part of news coverage when some accident or near-accident is being reported, but to Anne’s point, there’s never enough context or data shared, so that’s why I’m not sure if it’s even a real thing. If it’s true I’d assume it has to be as a % of the number of flights attempted, otherwise if there’s more flights now than in the year 2000 (for example), then there’s likely to also be more incidents. With more news outlets and content creators emphasizing viewership and clicks, sound bites over context, it’s really hard to know if you’re getting accurate information.
@@ellicel Yeah pretty much this. We'd need long term percentages per year to tell if in total runaway incidents have occurred more often accurately. I don't doubt it is the case in some airports given different environmental effects and management skill levels. Honestly this question is so localized it likely is better to look at individual airports statistics over national statistics to get a better feel for where it tends to happen the most at.
It basically would have worked. if this story tells me summin, its that we shud be doing double or triple or more take offs at once. The system is simple, first you taxi all the planes to one end of the runway, then you send the first out. The first one goes onto the runway and starts to take off, if after 30 seconds they haven't started taking off they have to immediately turn around and go to the back of the queue, and the next plane taxi's on. The plane that failed section A has the pilot reprimanded and they lose 5 pilot points (if you haven't heard of my pilot points system I'll post it below). Anyway once a plane doesn't fail this, and starts going off, the next plane immediately taxis onto the runway as soon as the pilot is out of the way. This crew then also starts their 30 seconds. By the time they've done their checks in 30 seconds, then the other plane will be at least like 30% of V1. So plane 2 starts immediately because it's effectively safe. Once the first plane takes off it immediately banks to the left to prevent backwash, similarly the second plane banks to the right, third to the left, etc etc. So what happens if a plane aborts take off? Well it depends. If the pilots abord after V1, then the other planes just continue like normal, since that plane is going to go off the end of the runway anyway and not pose a collision risk. So the other planes just carry on like normal. Why waste revenue when there's no chance of a collision? Similarly there's a new requirement for any aborts, the plane must immediately taxi off the runway we could even develop shallow angle off-roads for this. Anyway, we could simply have a computer system model this. It just checks whether the aborted plane will be able to make it off in time +5 or +10 seconds. If a plane can't get off for some reason, then the other plane simply uses these, and only in this extreme situation do we pause take offs. Just think of how much time and revenue a system like this could save. And as this video proves, if anything it'd be better * pilot points are an idea of mine to prevent pilots causing errors like this. E.g. a pilot who cancels take off and delays other planes, has pilot points removed, e.g. 10 points would be removed. Pilot points are awarded per successful *in the air* hours E.g. a pilot would earn 0.1 pilot points for each air hour, but could also gain points by e.g. not leaving their company hotel while in another country, reporting other pilots for violations, etc. They'd lose points for: customer complaints, not having as many customer compliments as the average, incorrect uniforms (e.g. not tucking in your shirt, etc). Simulator hours do not count, since they don't generate revenue. Then at the end of each financial year, the pilots with the least points would face disciplinary action. The pilots with the most would be able to spend it on things like gift vouchers, trade them for time off (at the companies discretion), use them when they're ill and can't fly, etc etc. All of this would massively increase profits for airliness,would
It's nice to hear about what happened from someone in the industry. I live about four hours away from Toronto and heard about this event on the news and enjoyed the play-by-play of what happened that you gave us. It reminds me of how effective the Swiss cheese analogy for incidents is.
I see what you did there, subtly showing a KLM 747 taking off when explaining that you cant start your take off roll unless the preceeding aircraft is already airborn in Europe... 😎
I noticed that too!! Carefully setting everyone up to think 'Tenerife' even before they were fully aware of the situation as it unfolded. Masterfully done.
Perhaps the Embraer FO could have re-iterated his aborted takeoff message when he didn't receive immediate confirmation from the tower. Nothing wrong with saying something that important twice.
During a rejected takeoff, the FO is kinda busy to be repeating calls. He's got a checklist that needs to be executed correctly, *right now.* What's needed is a more effective ground communication system that is robust to this kind of interference. Not at all sure how that would be done, unless they went to some kind of digital system that both preserved communication even under low signal strength, *and* inserted key action words into the ground traffic management system display.
Non-pilot but it seems like there would be a pretty consistent set of actions involved in rejecting a takeoff: spooling back the engines (I'd assume back to idle unless reversers were used which could be another flag), applying brakes, and probably some other aircraft-specific ones. Could the ADS-B be reprogrammed to detect that a set number of these actions are being performed, maybe even triple checked by a continuous reduction in airspeed, and broadcast an "aborting" status? That would seem to offer some degree of redundancy to the radio call without increasing the FO's (likely already high) workload, although I lack the experience to know whether there would be too many potential side-effects
@@lairdcummings9092 if you have 3-5 antenna, you can determine pretty precisely where each signal is coming from. Which maybe doesn't prevent the smaller signal from being stepped on, but it does tell you that more than one signal is being transmitted at the same time and who did it.
Reminds me of an incident I remember in the news somewhere in Japan years ago where the controller cleared an plane for take off but the pilot of a hellicopter thought it was clearance for him to take off and he flew over the runway causing the plane to reject their takeoff. Meanwhile, the controller had cleared another plane to land on the same runway and didn't realize what was going on until the landing plane was too low in it's pilots opinion to safely abort the landing...
@@MentourPilot thank you for taking the time to respond! It just proves how in touch with your audience and professional you really are. I can appreciate the amount of effort and dedication involved maintaining a relevant multi-platform production at your level. You deserve all your success sir, great work!
In modern high tech manufacturing, many tools on factory floors send out warnings to engineers in charge of their maintenance. Therefore, the fact that nuisance warnings increase the risk of the engineer ignoring them is not lost on me. However, I agree with other comments read that RIMCAS at least at the time of this incident was perhaps leaning a bit too much towards capacity rather than safety. It was fortuitous that the hole in the last slice of Swiss cheese didn't line up, and disaster was averted, when the 777 captain aborted take off on spotting that the Embraer had rejected takeoff. Looking at the state of automated warning systems in general, not just in aviation, these warnings are a great tool and simplify things for us humans, but we still need to be very alert. On another interview about a general aviation accident involving a highly experienced pilot in the USA, an NTSB official remarked that pilots need to always be vigilant and have the mind set as though they are operating chain saws. Petter, I had read about this incident elsewhere, but your narration had my pulse racing, even though I knew the outcome and your description always maintained a calm and even tone. Thanks again for a clear and succinct description of RIMCAS and its advantages and pitfalls , that is easily understood by someone who is not involved in aviation.
I'm hooked to these videos and appreciate the clarity of the technical explanations to the point that I have a greater understanding of what takes place in the cockpit. Thanks Peter
I find it quite strange that at the same time the ATC feels the need to push throughput while at the same time the tower was not fully staffed because of moderate to light traffic. If such techniques are required, the tower should be fully staffed.
If you open all the positions in the US often the controllers aren’t happy because it cuts into their break time. In the US the average controller works on position about 4.5 hours a day maximum. Some facilities more, some less.
Peter, you set this one up with perfection and had me on the edge of my seat! I didn't think it was possible but you're getting even better with each video.
I wonder if it wouldn't be possible to just update the radios at this point. I understand it's a massive undertaking as all planes and airports need to be updated. But the fact that important transmissions can be blocking each other without any kind of warning that it happened, is quite dangerous.
There are always tradeoffs. A digital radio system that can avoid this issue is much more complex than simple analog radio, which means there's a higher risk of failure. Analog radio is dead simple.
This situation shows me yet another time, that rules here in EU, even tho they sometimes seem to be crazy, are completely logic. It's not that other are worse, it's just us being extra cautious, which could avoid this particular situation.
Great explanatory video, as always. I'm so used to safety being a top priority in aviation but a situation where an airport has a "safety system" that receives/emits false information combined with an Air Traffic Controller being required to turn his back on an active runway beggars belief.
Re: 'Masked' Radio Transmissions. One of the peculiar advantages of the legacy, very old fashioned, Amplitude Modulation (AM) mode still used in the VHF ATC radio band is that simultaneous transmissions are audible as beat notes (caused by the slightly different frequencies) and garbled noises. This is quite different than Frequency Modulation (FM) which can have a 'Capture Effect' in that the weaker transmission can be 100% totally inaudible. Those involved with ATC communications should be aware of this radio communications detail and listen-out for such situations.
I have been binge watching your entire catalog of videos for the past few days! Loving all the factual and unbiased breakdowns of these events. With that said, nothing has made me laugh like the way the graphic of that bird went up and across the screen after it hit the side of the plane 😆 HUGE thumbs up to whoever put that together 👍
I loved the "how the human factor saved the day" comment at the end. Sometimes the human factor can indeed save the day too! Well done video, as always! Hats off to you and your production crew :)
Hello Petter, I just would like to say that I've been a very nervous flyer the few times that I've been a passenger in an airplane before, mainly due to having watched documentaries that focused more on sensationalism and drama, and I think that your videos have done great job with explaining these accidents and incidents and how the industry has learned from them in a much more logical and compassionate way, because of that I think that I'll feel more safe about being on an airplane if I ever have to, so I just wanna say thanks for that :)
I'm always amazed at quality of the recommendations that come from those reports. The industry is held to extremely high standards while also understanding that humans have limitations that have to be realistically managed.
I'm a bus driver, and for years I have heard about how bus company operators can't wait for autonomous vehicles. I have always pushed back against those sentiments, highlighting that a human operator can't be hacked, and can process a situation and take action more quickly than a computer. I know autonomous aircraft were not involved in this incident, but I can't help but think how this may have played out if the human element wasn't present. Regardless of vehicle type, highly trained and skilled human operators are essential to the safe movent of our passengers and cargo. As critical as this incident was, I commend the actions of the crews of those aircraft, and I'm glad we are all here to learn from this incident.
Years ago I remember an incident with a "drive by wire" bus that collided with a stop - bus driver said the bus went out of control however stopping the engine cleared all the data from the system (handy eh?). Dr Hannah Fry also thinks autonomous technology will never work.
you seem to be missing the consideration of how much faster and accurate computers are than humans. think about all the problems and crashes and incidents that are caused by human error, which computers could have avoided
But autonomouse vehicles are safer because they not only obey laws, they can also predict what the other vehicle is doing and calculate at a higher rate. The problem will always be humans interference. If you had a course with 100% autonomous vehicles vs one with 100% human drivers you would see how safe automated vehicles can be.
@@Interitus1 exactly! I think that the automotive insurance industry which makes billions in profit also has a hand in slowing down autonomous vehicles. because from their point of view, the safer the roadways are, the less reason for them to exist
@@Interitus1 autonomous vehicles lack perception - a human trait that allows us to comprehend a situation and not kill people. We've seen self-driving cars drive over cyclists and kill them. Please go and read up on the Turing Test. If the autonomous vehicle fails to understand the situation it hands control back to the completely unprepared driver. Which kills people.
As a software developer/designer, the way the aircraft sends Airborne flag and the way RIMCAS interprets the potential collision events seems very inadequate to me. First thing which causes red flag to me is the fact that the aircraft sends false information on purpose. It's showing airborne when it's still very far away from being actually airborne. There must be better ways to trigger that information more reliably. The second thing, that immediately caught my eyes was the fact that the system uses only actual speeds and not acceleration/ deceleration information. There should be some information/warning when accelerating aircraft suddenly goes into high deceleration. That is a clear indication of some issue and should trigger a warning. Either way, amazing video as always and with a happy ending on top of that. Thank you.
And there are probably reasons for both of those where changing it would make things worse, not better. The early switch to an airborne state is so that RIMCAS deactivates, but more importantly, that TCAS activates. It also enables the type of operation as we have just seen, If you don't do it, a jet would not even be able to line up and wait until the previous aircraft was actually in the air, losing even more time. And in a world where time = money, that's not a good thing. And as long as you are airborne, it doesn't matter what happens to your acceleration, RIMCAS will ignore you. This is once again because it would otherwise generate too many false alerts and controllers would start to ignore it. Systems like RIMCAS need to find a fine balance to ensure it doesn't alert too often and doesn't miss too much. And that is never an easy feat. Yes, it is easy to create a system that would alert on each and every small thing, but as humans, we tend to ignore alerts when they happen too often. Car alarms are a fine example of this imho. 20 years ago, all high end cars had car alarms and boy, did we know it. You couldn't walk on a parking lot without hearing one go off somewhere. After a while, no one looked up anymore when yet another car alarm went of and as a result, very few cars still have that type of alarm these days cause we all ignore it.
I'm not a software developer, just a structural engineer who occasionally has to program as part of his job, but I'm going to push back on the "way the aircraft sends the Airborne flag" as being bad. I totally agree with you that the way RIMCAS interprets that flag is defective, though. The way the Airborne Flag (which I'm going to call "AB" hereafter so I can use "airborne" to refer the the actual state of the aircraft), makes sense. Remember, it was _invented_ for TCAS, and then only many years later used for RIMCAS. Because it was invented for TCAS, what they appear to be going for is that you don't want an aircraft on short final to get TCAS alerts from aircraft on the ground near the runway, because then that would always happen and the pilots will start tuning out TCAS. You absolutely cannot have that, because the only way TCAS works is if pilots *INSTANTLY* do *EXACTLY* what TCAS says without hesitation. If they have to start filtering for false warnings themselves, you might as well throw out the whole system. (I imagine false warnings could also happen during lower overflights of an airport, and maybe in the pattern, but those will be fewer than alerts on short final from, e.g., the aircraft that will likely be sitting on the end of the runway.) The real rule for it appears to be something like this: Set AB=True if the aircraft is Almost Certainly airborne, but it Must Never be set to AB=False if the aircraft is, in fact, airborne. If this is what you're trying to do, the it seems pretty damn reasonable to me to have the actual software use the following, with WOW being the "Weight on Wheels" switch, a boolean that's True if the switch is squatted and False otherwise, and V being the measured airspeed in knots: if WOW == False || V > 100 then AB = True Then, the TCAS system on the aircraft can crack open each ADS-B packet as it comes in. The first thing it'll do after decoding it is look at the AB flag. If AB=False, then the TCAS system can yeet that packet into /dev/null and move on to the next one. Yes, the AB flag might not be correct, but if it's incorrect it's going to mistakenly be reporting AB=True. For TCAS, it doesn't care--the circumstances where this incorrectness are likely to come up are ones that (1) won't matter, because the place you have a low aircraft that may come into conflict with a mistaken AB=True is when it's landing and the other aircraft is taking off, where TCAS won't alert because they're not closing, or (2) it will matter, but the incorrectness in the airborne state is likely to result in a correct TCAS alert, just a bit early--this would be where an aircraft is about to take off, but hasn't actually left the ground, but the airspeed rule causes AB=True. If, under those circumstances, you get a TCAS alert it's pretty likely that the aircraft, when it truly becomes airborne, will be on a collision course with you! So this simple rule for generating the AB flag, and the simple "delete packet and move on" rule, both work for what they were designed for. I get what you're saying in that you can probably improve the correspondence between the airborne state and AB=True, but I question if the increase in complexity in generating that flag is worth it. I was able to describe it in a single line. If you start ladling in more data, now that gets more complex to both program and understand. Yeah, you can start considering the vertical airspeed and the acceleration/deceleration, etc., but I think there are problems with that: 1) These other data sources are much more complex themselves. The WOW state and V I gave above are both quantities measured directly off their sensors and piped in with very little processing--the WOW is "is there +5V between these two wires," and V is probably pretty close to "here's a 16-bit value of the airspeed in knots at this exact point in time" updated every half second or so. Stuff like acceleration, vertical airspeed, etc., all have to be computed from a time series of other things, or from laggy sensors. For example, the old Vertical Speed Indicator "steam gage" instrument worked by having a calibrated leak into a chamber with the current outside air pressure. If you had a certain rate of climb, the air would leak out at a certain rate as the pressure inside that chamber dropped, and that moved the needle to indicating your rate of climb. If you climbed faster, the pressure differential generated would be higher, so the air would leak out faster, and the needle would point to a higher value. However, it would take a bit of time for the leak to reach its equilibrium, so the VSI needle would noticeably lag your inputs. Right before you initiated the climb, there would be no air leaking, so it would point to zero. Then, you started the climb and it would take a bit of time for the air leaving the chamber to reach the equilibrium leak rate, so there'd be a second or two before the needle would "catch up". This is useful for checking that you're climbing at reasonable rates, but you can't use it to make immediate control inputs--something new pilots have to learn--and I'm not sure you want to use it for this AB flag, either. 2) Now, to be clear, I'm not saying that you couldn't work through all of the potential issues in (1)--I'm sure you could come up with a bunch of if-then statements for each potential sensor (OK, if WOW=True || (V>100 && VSI>50 for the last two seconds), and, and, and..). However, this means that you're going to have a really complicated set of rules that now generate your AB flag. I was able to state how it's currently generated in a single statement above; now, you'd need a f***ing flowchart to even hope to understand how the aircraft generated the AB flag that it's reporting. I know for sure it would make it difficult to understand for me sitting in a chair watching RUclips, but I wonder how complicated it's going to make writing and debugging the AB flag function. I can see this exploding combinatorically as you consider more sensors, increasing the chances of a subtle bug from a really weird combination of all these inputs. Taking all this together, I argue that the extra complexity in generating a "better" AB flag could very well not be worth it. It does what it was designed for very well, and is very easy to understand. I do think that the Embraer setting it as low as 50 kts is probably silly, and question that--but note that this setting, does, in fact meet the "almost certain/must never" rule above! Both 100 kts and 50 kts are round numbers, which means somebody just pulled them out of their fourth point of contact as "good enough" so it's hard to say one is silly and the other isn't on other than an emotional level. The real problem is that the people specifying RIMCAS didn't think through that the AB flag cannot give them the information they want with the way it is currently generated--as I argue above, it's pretty simple to understand--and probably should not consider it in their data flow. If you want to consider all the if, then, the other thing stuff, they might need to generate it in their own software. And it sound like they have a lot of that already, based on this video talking about the velocity rules. You always have to consider what something somebody else built was designed for if you want to use it yourself, and you cannot assume you know what it does just by the name! Given what TCAS was trying to do, naming their flag "Airborne" made sense, but the details of how it was implemented means it doesn't make sense if you're looking at a ground system. You have to go back and look at what they did yourself, because the name may make sense for their application, but not yours.
@@Hans-gb4mv I completely agree that you have to strive for as minimum false or pointless alarms as possible, but that doesn't change the fact that the system is evaluating false information. As you said, RIMCAS will ignore you when you're airborne, even though you're still very much on the ground and that's the issue. An example solution would be to trigger TCAS above certain speed (for example 50 knots), but start sending airborne after you're trully airborne. Another options is to start sending airborne as it's doing now, but immediatelly stop sending the flag if you start quickly decelarting within certain timeframe of activating the airborne flag. Or keep using Airborne flag for the purpose of RIMCAS only if the vehicle is actually accelerating. There are many other options how to detect true state of the vehicle without creating false alarms.
This overlapping radio keying seems to be a common factor in these reviews. I think it’s a difficult problem though. If two mics are keyed at once the third receiving radio thinks the weaker one is “noise” and it’s signal is rejected. It’s not clear how one would control which one would be prioritized as “weakness” is a function of distance. Once model for shared channels is the cell phone. In this case the cell tower dynamically controls the power of the cell phone, as well as gives it a time to transmit. However cell calls are private and point to point so this time-shared system works. There is only one human controller, likely they wouldn’t be able to understand two simultaneous messages and make any sense out of it.
@@terrygaryet1856 packet radio and other timesplitting systems exist but are either so old or so new that the tech just hasn’t been mainstreamed. These allow multiple communications on the same frequency, but require special tech to use, and would need an international overhaul to implement
@@terrygaryet1856 put three receivers on the ground that triangulate all transmissions, if two transmissions happen at the same time on the same frequency you can not only know it happened but plot on the controllers screen where the two colliding transmissions came from.
That is true. But there is also another possible resolutions to this without having such a system. Safety critical transmissions must be read back, in this case the Embrearor failed to repeat their rejected takeoff message.
There's a tremendous amount of work in that video, and it shows! Super well made and extremely clear :) From a pilot's perspective, the only tiny mistake I saw was the callsign "JZA" for the WestJet Encore Q400. Again, congratulatons on this excellent video!
Basic ATC (in the US) for successive Cat III departures on the same runway is 6,000ft and airborne. The controller should not have issued the 777 takeoff clearance until they had VISUALLY observed the E190 rotating. Relying on ADSB or ASMGCS for airborne status in the Tower environment on a clear day seems ridiculous to me for this very reason. Also, in my opinion, at that airport configuration, if RY05 and RY06L are both in use, local tower control should always be de-combined. And if the E190 came to a stop on the runway before exiting, they’re wrong. When I’m in the tower, it’s nice to have a little help from technology. But when it comes to runway separation, nothing beats a window and a good pair of binoculars.
I am only an engineer and not involved in aviation. But any system will fail if optimized too much. When safety or design margins become thinner and thinner eventually something with an infinitesimal small probability will happen. Design for inefficiency and redundancy and you have a good chance that it will be able to handle a situation which was calculated not to happen.
In the end, were there any recommendations or operation changes that came from this incident? It's great that they learn what factors caused this, but unlike your other videos where you addressed what was charged to make this industry safer and prevent similar incidents from happening, I don't see it in this video… hopefully I just missed them. Thanks for another great video, captain!
20:00 there was an ATCI mentioning the weird programming, but the programming itself was not changed. this was done because simulations showed that increasing the speed threshold at which airplanes report that they are airborne results in more false positives, thus making the system unreliable and more likely to be ignored.
![ARGENTINA vs. PERÚ [1-0] | RESUMEN | ELIMINATORIAS SUDAMERICANAS | FECHA 12](http://i.ytimg.com/vi/xtI25ENDDSY/mqdefault.jpg)
What I find most interesting about this story is that... No one did anything wrong. No one made any mistakes, made a poor decision, broke the rules... Everyone behaved properly at all times and it was the actual system that let everyone down.
I’d be interested to see if the Embraer’s company has SOPs on doing high speed aborts. I’ll bet they do and I’ll bet that they don’t want you aborting for a bird strike in the high speed regime.
If that is the case then the pilot flying was in the wrong.
@@stevenbeach748 bird strike is most dangerous on the runway if you take off and that engine fails when you are just starting to climb you might crash sully is an example of a double engine bird strike at low altitude one that turned out well most the time losing both engines right on take off leads to everyone dead because you don't have enough height to make it back to a runway
Sometimes you can do everything right and still lose
@@stevenbeach748 Pilot's decision on whether the aircraft is safe to fly is absolute. No person or company is going to question a pilot's decision to reject for safety before V1.
@@TWEAKLET Part 25 certified airplanes (this one) has to be able to take off on one engine after the engine fails on the runway. The plane must be able to continue to accelerate to rotation speed on the remaining engine and climb away from the ground at a 2.4% minimum climb gradient. Bird strikes, especially one bird, are no big deal and should not be a reason to abort at high speed.
You make a great teacher. As a non pilot, I still enjoy your videos very much, because you explain things in a way that even I can understand.
That’s great to hear, that’s what I am trying to achieve
@@MentourPilot Since my poor eyesight blocked my high school goal of becoming a pilot in our Air Force, your channels get me as close to being a pilot as I can get. 👍
@@stefanlaskowski6660 poor eyesight SUCKS. Its the most important sense and its so bad for too many people. Its awful. I also have poor eyesight and i like being in the nature outside of cities and having bad eyesight just sucks. Its so humiliating to know that i would have just died if i had poor eyesight in the stone age. I don't want to be dependant on some stupid glasses... my passion is about nature so imagine a tiny rain that would just refresh normal people but it basically blinds me. Or maybe have a surprise branch touch your face? Glasses either fall off or get scratches. Oh and don't even think about being able to see anything while swimming
I HATE THEM SO MUCH
Thanks for letting me tell you about it even thought you didn't let me and it was just my idea to comment this
I completely agree
@Eetu The Reindeer: Any chance of getting contact lenses, or corrective eye surgery (LASIK, PRK)?
Damn dude, your ability to explain a situation to those of us who have zero aviation knowledge (like me) is really amazing. Very impressed!
It is the only way to explain things
@vandecayear10 I totally agree with you 💯 and sometimes I feel like I can now pilot a plane just by the way he explains every single step throughlly.Going back to this video it reminds me of the Tenerife disaster of 1977,although in that case the weather played a huge part as the pilots of both planes and the air control tower guys could not visually see the planes as the airport was engulfed with a lot of fog reducing visibility significantly and although a lot of lives were lost that day,it also brought upon a lot of significant changes to help avoid such issues again 😊.
Singapore 🇸🇬 ATC Changi International Airport will Never allow this Double Taking off Aircraft like this Incident.
@Angela-Gnow you should thank him by opening your legs for him ...😊
@@ShortCircuit05445 Had me thinking of Tenerife as well!
I loved how the 'highly optimized ATC workflow ' was called out as a contributing factor to the near-accident. Taken out of context, that phrase may seem like a compliment ! Loved how you laid out the ATC operation context so that the rest of the story made sense ☺️
Highly optimized usually leaves little to no room for even small errors. We have seen that in the global supply chain. You always need to be on the lookout. I believe there was a pilot waiting for takeoff in the Linate disaster that noticed on his TCAS that an aircraft ahead of his had not gotten/gained any altitude and questioned his own takeoff clearance. Maybe this 777 should have waited to see the smaller jet ahead off him get airborne.
When I hear "highly optimized," I think "cut corners" and "minimum redundancy."
In short, very little error trapping, and not rubust.
I think it was meant as "congested" or "overloaded"
@@Stettafire in practical terms, yes, but in strict definition, no.
"To optimize" means to make something more optimum, or closer to perfect. "Highly optimized" means very efficient. But in reality, pushing busy, intricate actions closer to 'perfect' when human actions and interpretation is highly involved quickly reaches the point of work overload and conflicting, simultaneous, even out of sequence, actions.
In short, if *everything* goes *exactly* right, then no problem. But *any* hiccup means the whole thing collapses.
@@lairdcummings9092 as the cargo ship getting stuck in the Suez Canal neatly demonstrated
The graphics used in your videos to help explain what is going on are just outstanding. Fantastic job guys.
Thank you! Dom is a magician!
@@MentourPilot What software do you use for all these renders? I live 15 minutes from Toronto Pearson and this all looks very realistic! You can even see Mississauga downtown in the background. Truly outstanding!
@@MentourPilot please do flydubai flight 981
@@MentourPilot So Petter... What do you think of your POS airline's treatment of the Spanish cabin crews?
@@MentourPilot can you please do a video on the famous Air France flight 447 that crashed into the ocean ? The one that was out into a stall by one of the copilots. If you have already too kindly lead me to the link . Thank you
I’ve always found the fact that they’re using comms on the runway that can block out another to seem like such an archaic system for such a high tech and safe industry.
I suppose it's actually rather difficult when you need to ensure prompt live communication from potential massive numbers of pilots to ATC and so driect calling of course wouldnt be practical, and theres no real way technically as far as i know that could prevent transmissions overlapping on an open frequency that you need to handle that.
But I wonder if something like whatsapps voice notes would probably work. levels of urgency could be hardcoded into the message - Mayday calls would always be trasmitted live and overide any other queued messages. Pan-Pan level 2 priority, then after that could just go to playing out in the order received.
From the users perspective anytime the frequency is open they can talk live and would notice no difference from normal radio comms. When there is a transmission coming through if you transmit it will just queue in and play like you started your transmission as soon as theirs ended, minimal delay.
I could see in very busy airspace that it could lead to radio communication being delayed by a matter of some seconds, but to be honest its the same now anyway but with the additional risk of an overlap
The real downside is it would be internet and satellite connection dependent i imagine.
agreed! I feel like critical messages (rejected takeoffs, takeoff rolls, etc.) should be transmitted twice? or would that just congest the air? through all of my Mentour Pilot binges I have noticed this quite frequently.
@@k1ng5urferWhatsApp for ATC? What are you 12?
@@michaelm1573It’s the clarity of a messaging app he is likely referring to. It’s a got a silly name. That doesn’t mean it’s not a useful technology.
The RIMCAS system having to wait for a 50 knots threshold from a decelerating airplane is pretty dangerous in my opinion. The simple fact that the airplane is decelerating should indicating a rejected takeoff and immediately trigger the RIMCAS system.
Why is 50 knots even used as the "airborne" threshold? What commercial aircraft takes off at 50 knots??
Why isn't there just a button in the plane that they push if they reject a takeoff? Seems like that would solve a lot of problems.
@@Mountain-Man-3000 Why even use the speed as an airborne indicator. The only indicator for airborne should be the WOW sensor, at least from my point of view.
Maybe the program only has a data slot for "speed" and doesn't perform derivative operations to get an acceleration number?
@@TrevorSmithy there *is* a button. It's automated - their transponder. Which worked as intended. The problem lies in RIMCAS programming, and the decisions and assumptions made in programming it.
Ah the “combined position.” This concept is probably familiar to many people in other professions. We certainly know of it in health care. This is where you keep cutting staff / costs until you achieve a really really good balance sheet. Ideally, some executive should get a promotion or a big bonus. I believe that the process can continue until a complete disaster happens. If you are really really lucky this will only cost a huge amount of money and prestige. But quite often it also costs lives.
then when someone dies you prosecute the poor bastard who was up for 36 hours with no relief covering 2 or 3 jobs and keep on doing the same thing
I was looking for this comment and slightly disappointed to not find it among the top 3. Indeed, I can confirm that the process CAN continue (provided some executive keeps getting a promotion/big bonus) until a complete disaster happens. But then, you just roll with what @Jonathan D has suggested
Well, how else are stockholders supposed to accumulate more "record profits"???
There's only so much "planet earth" to go around and only so much of any market to fit on it. Investors don't like to invest without those magic words, "record profits" or they don't like buying the stock. AND of course, that favorite motto of them all, "If you're not growing, you're dying." ;o)
Very true. In this case it could certainly have led to disaster. In other cases it leads to early burn out and people not wanting to work under such ridiculous workloads and an even higher loss of, or failure to even find, experienced reliable staff. This kind of stupidity is seen in soooo many places these days. You know tick boxes to make sure that things are done right can be very useful, but if you don't give your staff enough time to actually check and do thinks right you make a right ****** of everything.
@@gnarthdarkanen7464 Shareholders don't respect themselves thus generally not respect others.
The professionalism from all parties involved makes you realize just how tight these airports are run.
dangerously tight
when you dive into those air controls situations, it is almost a miracle that there are not more accidents in the airports on the runways !
lolwhat? 🤣
17:10 trucking instructor once told me "if something happens you are not prepaired for or havnt been trained for, go with your first instinct. Its usually right". thats what this pilot did, nobody got hurt, nothing damaged. Win in my book.
I can't think of many situations where there is an option better than just stopping as hard as you can!
@@biscuit715 Been a while since ive seen the video, dont remember what it was about. BUT while a plane is taking off there are speed limits they have to reach by certain points. V1, Rotate, and V2. If something happens. before V1 the pilots hit the brakes. Something happens after V2 the pilots MUST continue takeoff as there is not enough runway to stop and at V2 the plane is already nose up and starting to fly. Anything happens between V1 and V2 is up to pilots discretion as above V1 they don't have enough room to stop but are still to slow to fly. Lose an engine after V1 but before you rotate, you dont have enough room to stop or enough power to take off properly so it is left up to the pilots descrition
It seems insane to me that an aircraft will report itself airborne before V1. There's no guarantee that it will actually become airborne up until that point.
Peter said that is for TCAS. TCAS is completely disabled for aircraft that are not airborne, I think. If the plane is about to be airborne, TCAS should be on.
@@thewhitefalcon8539The problem is using one sensor for two different systems, each of which increases safety by moving the sensor's decision point in opposite directions. This can't possibly work right.
@@KaiHenningsen Yes, but they can't retrofit all the planes to update their TCAS to send out a "maybe" on the ground state. Do the best you can with what you've got. Nobody is supposed to RELY on the runway incursion system.
Quite Right,
An Aircraft at 50 knots os not flying, nor is a Commercial jet at 100 knots
@@knutaune724Piper Cub: “Hold my Beer”
The first time I watched a cub take off I was taxiing a 172 on the parallel taxiway and I thought at any minute it would stall, lose control, and perhaps crash into me.
A simple suggestion for a solution would be a rule that a call-out for a rejected take off must be repeated until ground control acknowledges it. Along with that would be to establish an SOP to include a pre designated runway exit instruction (in case of rejection) to enable a safe exit from the runway at busy airports.
that may block the frequency for even more important situations such as mayday or fire... no? however repeating a couple of times seems to me (who knows very little) could be advisable. they may implent such a directive in case of relying on visibility
Haha...
acknowledging communications works both ways. ATC is required to acknowledge the pilots. In my opinion the pilots who rejected should have done a better job making sure their extraordinary intentions were heard. They knew the airport was running tight departure. They knew that there was a takeoff clearance given and acknowledged on their runway. Then again, we are talking about seconds so perhaps this happened too fast for a repeat call to come out.
good point, call-out until acknowledment.
The Canadian approved procedure should never have been approved for use. Until the first a/c has its wheels off the ground , there’s no clearance for the second a/c to take off. It’s really very simple .
The pilots of the 777 definitely made the right call. Would they have had enough runway to take off? Maybe. Would they have had enough vertical clearance with the tail of the Embrear? Maybe. Would you want to gamble almost 500 lives on a maybe? Definitely not.
In 1971, there was a near disaster in Sydney when pilot of a 727 took off on a wet night over a DC-8 which had been taxing towards it because of a misheard communication and poor visibility from the tower. The 727 continued the takeoff because the its captain judged he could not stop in time. The 727 struck the tailfin of the DC-8, and while damaged, was able to keep flying and land safely.
This is why you shouldn't have a computer in control during anything but regular boring operations, because you can't program a computer to decide the outcome in drastic decision making conditions.
I 100% agree. The 777 would almost immediately have been in a position where they could not abort their own take-off - whatever happened - simply because they could not stop before hitting the Embraer.
Given hindsight, imagine the 777 had also had a bird-strike (which might have knocked out an engine).
+ there was no guarantee that the 777 itself wasn't gonna have any problem with its takeoff, especially when the aircraft right in front of them on the same runway had to reject their takeoff which could've been (it kinda was) caused by the runway environment and not the aircraft in front.
Right. And what if something goes wrong and they need to reject the take-off? If that happens, it could be an epic disaster.
I have a suggestion for another video: how passengers can be inadvertently left aboard planes after they land. The same airport, Pearson International in Toronto, had an incident a few years back where a woman who was asleep in her seat was missed by the cabin crew when they disembarked the passengers. The other passengers left the plane and then the cabin crew and pilots left and none of them saw this woman still asleep in her seat. Eventually, she woke up and found herself in a dark, unheated plane; it was winter and after dark. The plane was no longer connected to the terminal - it had been towed away to a parking area - and the doors were all closed. The woman tried to phone the airport but her phone battery was almost dead and she wasn't able to clearly explain her situation before the phone died. If I remember correctly, she finally got noticed because she managed to figure out how to open one of the doors and then shouted for help through the open door and someone driving a baggage carrier heard her and came to the rescue. I don't know how frequently such things happen but I thought it might make a good topic for a video if it is not a one-in-a-billion fluke. After all, any of us might find ourselves in such a situation and might benefit from advice on what we should do if it happens.
I knew a guy who fell asleep on his flight, the plane landed, some people got off but others stayed who were continuing one. But he was supposed to get off at that airport. Well, he wakes up either enroute or at the next destination because someone didn't check if everyone was off who was supposed to get off had actually gotten off.
Or he could do a story about the time they forgot a suitcase in Rome, and it flew all the way back to Copenhagen before flying back to Rome again! Scary stuff, the suitcase didn't have any phone or other ways to communicate!
Hugh, stop picking on Canada. Yes, someone missed someone who was asleep. Who wants to wake someone sleeping? I, pearsonably, would want to sleep by Pearson airport all the time.
Hugh Mungus
Explain what you mean by "The woman tried to phone the airport" ?
The WoMan either phoned the airport or she didn't -
There is no TRIED.
Whether the call was successful or not is another matter.
@@andrew_koala2974 Of course there is TRIED, it is really hard to phone the airport without trying, the chances of accidental call are abysmally low.
Also, I back the Boeing's decision to reject 100%. As you said, at this point, there's no time for complex decisions. There'd be 2 thoughts in my head "are we passed V1" and "can we safely reject"
When shit hits the fan and you are *below* V1, reject. It's that simple, it's what the entire purpose of determining V1 is for (and what the bigger system of V speeds is made for).
and also they were seeing the aircraft in front of them, but because they themselves were in motion they couldn't possibly know if they were going to a stop or just taking more time to become airborne. Must have crossed the Boeing's captain that they could become airborne right when they were under them I think
Great that you pointed out at the end that it was the human factor that saved the day. It is often blamed, rarely credited.
That was my point actually. Glad you noticed it!
Agreed very much. Our squsihy bits sometimes do the good work instead of the screwups. Excellent callout.
Retired Air Traffic Controller, I can see this easily happening. You did an excellent job explaining all of the factors involved.
My step father was an Air Traffic Controller in Canada for a number of years. He's always the most tense one in the family when we travel by air because he says he knows better than any of us where things can go wrong.
God Bless you as you may have saved so many souls during your role @Waterwalker
If you, as a former ATC, can "easily see this happening" then that's a problem.
But of course it wasn't the ATC's fault. They were following the rules. It was the rules themselves which sucked.
For all the incredible safety systems involved in the aviation industry, marking something "airborne" when it's clearly not airborne is absolutely unfathomable to me
This is the root cause. Presumably cause of commercial pressure (as Mentour alludes to). And there is no indication this root cause was addressed (least, from the conclusions Mentour gave?).
@@PaulJakma I would make an argument that the root cause was having one controller manage two opposing runways. But again, are you going to require two ATC controllers every day? Thats probably the difficult balance of human safety vs practicality.
I think the bulletin was used because you can make an argument that pilot decision is the safety redundancy in a scenario like this.
First off all, this video shows how absolutely useless RIMCAS actually is, if TCAS is deemed so much more important that airborne is given of before (and long before) actually being airborne.
Secondly after the radio incident on Tenerife, i find it unaccceptable that radio messages are still able to be blocked.
Seems as if for take-off rejection should be a rule about read back and or sending on secondary frequency.
Thirdly the ATC assumed the Embraer to go airborne soon. That does not equal being airborne.
And since assumption is the mother of all f..k-ups we should reprimande him at least a little bit.
The biggest blame however is on the system in place where a useless RIMCAS gives false trust and a FAA organization still has not managed to deal with radio blocking.
Yes, a plane should not begin the take off roll until the plane before has exited the airport perimeter.
@cessnadriver6813 in an ideal world yes but what specific regulation do you want to add and how will that affect airport operation and industry as a whole? Especially when ATC is dealing with their own labor shortage.
I went into this video going "wow, I haven't heard of this incident before!" Probably because nobody was injured. Sometimes it's nice to hear of almost-incidents that had happy endings.
A lot of modern near-misses would have been disasters a few decades ago.
I thought the same thing, and I even live in Ontario, but the incident happened in March 2020 so we all may have been a tad distracted news-wise
That operations flow seems dangerous. It seems to me there's nothing spectacularly unlikely about that sequence of events, it could easily happen again.
It may be dangerous but it is “highly optimised”. Optimised for who though?
It really isn't all that dangerous as long as the aircraft holds its T/O roll until the preceding aircraft is airborne (since you are responsible for visual separation from the preceding aircraft after all and you have no idea if they will reject the T/O or not like in this case). It only takes additional 30 seconds (or not even) to wait on the runway but still can get way more departures out this way over a long period as mentioned in the video on a clear VFR day.
@@c00l00k That's right and that wasn't clearly mentioned in the video, the thread avoidance is guaranteed through the pilot applied visual separation.
Kill hundreds in an avoidable accident < save a few dollars.
That seems to be a pattern with a lot of incidents and accidents in North America. Their standard practices of pre-clearing aircraft for landing (and canceling later if the runway ends up being occupied for whatever reason) is another one that comes to mind, which is another thing that isn't done in europe.
It seems they put too much emphasis on "efficiency" over safety over there. I know some US airports have a ton of traffic, but I'd rather be late than burning to a crisp on the runway because another aircraft slammed into mine.
It is still amazing that the communication systems still are subject to interference if used at the same time. Especially when you consider larger airports with potentially dozens of planes using it at once.
What is this game it looks like Microsoft flight Simulator
It is one reason why there are multiple frequencies, e.g. one for Ground, one for Clearance Delivery and one for the Tower, so that the number of aircraft trying to share a frequency is reduced. Pilots learn what overlapping transmissions sound like and should reissue any communication that they think might have been overlapped.
It's how radio waves work. It's more of a science physics issue than a technology issue
@@staticbuilds7613 We all get it, but using such old technology, especially in a large modern airport seems... like it needs an upgrade to the 21st century?
@@plektosgaming Most of the radios are 21st century. The reason this happens is because radios work on radio waves. You have one signal per frequency. You said "we all get it" but you clearly did not
There have been other incidents where the radio transmission to inform ATC of a rejected takeoff was stepped on by another transmission on frequency. It's relatively likely to happen, because it's an unexpected call at a time when ATC is likely to talking to other aircraft in sequence. The F/O doesn't want to wait for another aircraft to read back a takeoff clearance before making their call, because it's important that the tower is alerted to a rejected takeoff as soon as possible, before that other aircraft begins their takeoff roll. The radio is not a very reliable way to convey this information in a timely manner. It seems like the RIMCAS should also raise an alert if the aircraft's ADS-B ground speed drops by more than (let's say) 20 knots after accelerating above 50 knots. If the aircraft is slowing down significantly after initiating the takoff roll, it's not taking off, regardless of the air/ground state logic, and the controller should be alerted immediately. It also wouldn't be a bad idea to hit the transponder IDENT button on a rejected takeoff to help draw the controller's immediate attention, at least at airports with radar-equipped towers.
I would have thought a rejected takeoff transmission would be repeated if the tower hasn't acknowledged it - I did comms for a while, and important messages were always ack'd or readback to avoid them being missed. (edit: a word)
@@VosperCDN yeah, but non-comms people don't like that and don't want to take that extra step. comms people tend not to make it into leadership positions, so leadership tends to sympathize with the non-comms side of things and you get situations like this which then leads to an emphasis on instruction (treated as a new instruction) until it is forgotten and things repeat. same things go with backup frequencies. having a second frequency that the tower monitors in case your messages aren't going through due to high traffic volume allows one to get important messages through even when there is too much chatter going on. this means an extra radio for the tower, and another thing to remember and keep track of. if you are streamlining workflow this seems like an easy win to cut, but it always leads to problems at some point.
There was a suggestion to FAA from a collection of bush pilots asking for Full Duplex ATC & Unicom... Everyone transmits on freq. A, listens on Freq. B, except tower. Tower hears A directly and transmits on B directly. When tower not transmitting, the repeater puts out what it gets from A onto B. Everyone hears what ATC does... Which is only ATC when ATC is talking. FAA said it was too expensive for the majority of Aircraft to be refit. that was late 1980s.
@@WilliamHostman Wouldn't that have exactly same issue though? Per your explanation, everyone would still transmit on the same frequency, and for that case it genuinely doesn't matter if you listen on the freq A, freq B or you get a transcript via email - both sources still transmit on A. The only difference for your explanation between current and suggested, would be that if ATC would be transmitting, it would not retransmit planes to other planes. Which would technically be the same situation, assuming that tower is able to overpower the planes anyway.
@@tipakA you'd know you were "stepped on" in FDx, whe in simplex, you don't. The 1O of the smaller would have heard the heavy and known he didn't get heard, and be aware he needed to repeat.
What always amazes is me is the stopping on the runway. When I went through training, the ATC sternly talked to me about not stopping on the runway after my first solo landing. In either case, unless I literally can't move my aircraft, my priority would be to slow the aircraft down to ground maneuvering speed quickly, then get the hell of the runway, not sit there and cause a hazard.
I agree completely. I realize that the 190 crew was unaware that their "reject" call wasn't heard by ATC. I'm sure they were anticipating an "exit runway when able" call from ATC. That being said, I wouldn't have sat there stopped on the runway, response from ATC or not, completely vulnerable with an A/C that's totally capable of taxiing clear. I'm not going to criticize the 190 flight crew; I'm just saying that I personally (been doing this gig for 3 decades) would've taxied clear and held on the proper side of the hold line and then attempted to communicate with ATC. Unlike on a runway, traffic on a taxiway is moving nice and slow...that's where I want to be sitting with the parking brake set.
I totally agree, but it was in a matter of seconds, I'm pretty sure by the time the 777 decided not to take off, the 190 was still slowing down. All of this must have happened in 30 to 45 seconds, I guess. So it's difficult to slow down enough to a point where you still have time to turn 45 degrees and exit. It takes a while to do that while on a plane and without doing anything drastic like taxying outside the runway.
@@paulw4310 100% I would reject, without knowing it was a reject the most dangerous situation would be trying to fly over them while they rotate and climb into your underside or lose control from your wake. Also agree that my training and experience is to keep the active as clear as possible. Never come to a complete stop and exit at the first opportunity once slow enough. I haven't flown anything as bigger than a seneca, so maybe the rules are different for 190s and 777s.
I have flown in/out of some busy airports, but I cannot think of a time that I started to roll prior to seeing the prior aircraft land or take off and totally clear of the active. Even if I had been given clearance, I think I would hesitate and wait.
If you have a rejected takeoff sometimes you have to stop on the runway. When landing you are required to exit the runway as soon as possible unless you have a problem.
Absolutely! That’s nuts
I'm an air traffic controller. It's possible that there is an aviation channel out there where the host has a better grasp of ATC equipment, procedures and environment, and the ability to clearly explain it, but I haven't come across it yet. The amount of research Mentour puts into his subjects is clearly evident.
As to the comments about the radio issues of multiple aircraft and/or ATC stepping on each others transmissions, this is a real problem in our ATC environment during busy periods. However, usually, transmissions don't overlap precisely in length so there is some indication to the controller that he/she missed a call and they can solicit that missed call. It was bad luck in this scenario that they apparently fully overlapped. Also, full duplex communications in ATC have the potential to wreak havoc on communications as it would allow multiple transmissions at the same time (imagine a party line phone call or zoom meeting) with no ability to distinguish any of them. Additionally, the current system, in the event of a stuck mic (stuck PTT) on an a/c, because ATC radios are more powerful than a/c radios, allows ATC to override the stuck mic transmission and still transmit to other a/c under its control, important from a safety perspective. And, in the event ATC needs to make an emergency/urgent transmission while another a/c is transmitting, we can "drown" out that transmission and make our urgent transmission by "stepping on" the currently occurring transmission.
Finally, airlines' financial need to optimize efficiency does increase the risk of something going wrong, but that can be found in many aspects of the operation (for example, the decision to fly with less than full fuel on every flight (see Mentour's Singapore airline video). We can absolutely make the system safer by applying greater separation and increasing delay in the system (one in, one out?), flying with full fuel, etc., but that would come at increased cost to the consumer, which has not proven to be popular. In this case, the multiple layers of aviation security, which included the 777 pilots awareness of the situation ahead of them and recognizing that the preceding departure they were to follow had not gotten airborne and that they were closing on them instead of having increasing separation, and taking the appropriate action. worked.
Apologies if this was too long or previously addressed.
18:26 Strewth, that plane is practically invisible behind the airport building. The re-enactments in these videos do a great job of recreating the situations, and giving a clearer idea of what is going on. I like how the incident reports maintain such a neutral tone; the 747 pilot isn't commended for aborting his take-off, better than that, it simply isn't mention. It is amazing how much potential carnage a single bird can cause.
*777
I'm also like, if you control in this configuration why not use camera's on the terminal building at the other side?
*IMO it should be standard procedure if you reject a takeoff and ATC doesn't confirm, you should take the next available exit automatically.*
I love that you broke this down Petter because it opens up so many opportunities to make aviation safer. How a system can show you as "airborne" before you're even at your V1 decision speed is ridiculous to me
I thought the same. Those who developed the system haven't yet heard about 'rejected takeoff'.
I think its a matter of the system being generalized as possible. In short, its designed to work for every aircraft. From an a380. All the way down to a c152. So 50 kts makes sense. WOW could also work, but not all planes have that sensor. This is just my 2 cents as an air traffic controller. Ultimately since speed is really the only route to use it should be plane specific. still wouldnt be perfect but a lot better.
@@zachfield5336 The 50kts threshold is specific to the Embraer and it's the transponder which emits the airborne signal, the RIMCAS only interprets it. So it's perfectly feasible to have the transponders fallback to emitting airborne signal at V1 instead of a static value, especially curious that the Embraer system was programmed at 50kts.
Apparently the threshold used is to activate TCAS, so when RIMCAS was developed, they just inverted the TCAS activation signal. So by increasing the threshold for "Airbourne", you increase the danger of incorrect warning suppression on TCAS.
@@DryBones111 Makes sense I guess. I just talk to the planes, im not an aerospace engineer. But i dont really see why TCAS and RIMCAS couldnt be operational at the same time.
@@zachfield5336 "Ultimately since speed is really the only route to use it should be plane specific. " How about height above ground?
Before I used to love watching NatGeo’s air crash investigations, but now I can’t stand them.
I got spoiled by Peter’s incredible explanations, context of systems and how it all comes together in easy to understand pieces of information.
The standards are quite high and I’m glad I’m constantly following this series in your channel
I agree... petter dont give us huge fireballs, burned bodies, teddybears sadly lying on the ground as a horrific witness to the disaster... He gives us the WHAT, WHY and the AFTERMATH, that hopefully ensure that this particular incident never happens again...
Love from Sweden 💖
Å Ebba Blitz är obeskrivligt het...
Same. And when they do their cliff hangers or other guess what incredible thing happened? Most of the time I already know what the problem is because of these videos. Just so frustrating to keep watching such long boring story when it could have been told faster and with more interesting details
As a teacher I have to appreciate your way of explaining everything so well around aviation that even an amateur could understand! It definitely is an inspiration! :)
I think the second plane rejecting the take-off was a good decision, they had no idea why the first plane was not taking-off and in different scenario they might easily get into more troubles by taking off. Also if they got in any sort of troubles after committing to take off ,there would have been still plane in front of them severely limiting their options.
It's a potentially slow plane up front as well. Good call with an unclear hazard to reject as early as possible.
Yep we know what happened in Tenerife when the KLM crew did not abort the takeoff despite some concern.
@@gnnascarfan2410 in that case KLM should never have started the takeoff roll.
@@hewhohasnoidentity4377 Well the copilot and engineer were too afraid to speak up because the Captain was basically a god.
Thank god a tenerife 77 situation was averted
I can’t wait for you analyse the recent LATAM airplane accident in Peru, where a firefighters drill, ended up with them entering the runway where a plane was just getting ready to take off, and crashing with the plane. It was a miracle that only two people sadly passed, but the Peruvian management of these events really make you wonder about their professionalism
One wonders why they didn't put barriers on all taxiways entering that runway...
Sounds like poor planning and lack of coordination. Both have human error involve but this double take off incident has a lot to do with over reliance on technology.
reasons to not go to shithole countries
This incident genuinely scares me because it shows that all the safety measures put in place to prevent another Tenerife can fail in what sounds like the exact same way with a masked radio transmission. While the training, safety, professionalism and culture in the industry is clearly very good and constantly improving, this story suggests that the necessity of automated systems gives a sense of security that can turn out to be false
Yes. A masked radio transmission played indeed a major role in the Tenerife disaster, too.
and fog NOT a factor
(Haven't read the full report, probably mentioned) - Granted this was all a matter of seconds, but transmitting in the blind about a reject takeoff and not getting an acknowledgment from the tower should have been some high alert first officer making sure it was heard. I work in a semi-related field with critical radio transmissions all the time...and the number of times we get "I told you I was x status" - never came through. No ack = didn't happen.
@@gerbsvizsla Fog played a major role, too - and the lack of a Ground Radar in Los Rodeos at that time - also the overcrowding of this small airport due to the closure of Gran Canaria. It´s a complex story. @Mentour Pilot made a very good video about it in last December, please watch it here on this channel.
@@frollard Yes, better safe than sorry.
One easy fix would be to revoke the "AIRBOUNE" status if an aircraft starts decelerating...
...or to stop cost cutting in the ATC tower, that could also work.
This content is EXACTLY what aviation fans want, and you provide said content perfectly
Same for me and I’ve never been in a plane before
In Canada at that airport, they do have an alternate runways, not sure why the rimcas system did not sound after a certain time, in refence to the closeness, of the three airplanes. They were on the same runway, actually three planes. Two of the airplanes did decide to abort there flight. It was perhaps due to the system that there was no incident involved.
With respect, I'm glad that situations are analysed even without a bad ending for those involved. This situation where a workflow causes the chance of an incident to rise is something that happens in many situations but these are very professional pilots who are trained to react appropriately in any case. Glad no one got hurt too
In most modern industries "near misses" and "accidents" are both investigated in the same manner (although with perhaps less urgency for the former) Happily, there are relatively few full blown accidents and investigators keep their skills sharp investigating near misses. You learn just as much from a near miss as you do with an 'accident'.
@@KnugLidi Systems are made to be reliable but only if they were implemented properly and within tolerance. The TCAST and RIMCAST system are working as designed but how they work don't scale up properly with the increase traffic flow of North America airports. Also, implementation of the systems miss out other assumptions, such as aborted take off and increase in traffic flow.
I mean... sometimes things like this just absolutely baffle me. How, in an industry focused almost solely on safety, does it get decided that "airborne" should mean _anything at all_ except for *actually in the air???*
It seems like _someone, sometime_ should have audited that procedure, and thought, "Actually, you know what? Let's make it only say "airborne" when it's, I dunno, airborne?"
Yes.
It's because ADS-B is also used to avoid in air collisions, and the possibility that the "in the air" sensor could have a false negative. RIMCAS was added later, and utilized the same ADS-B data which is used for TCAS.
Maybe a future version of ADS-B could have a data flag for "I might be airborne, or on the ground, I'm not sure yet." so it could activate both avoidance systems during the takeoff roll. But it would have to be backwards compatible with older aircraft.
For plane it's the "turn on" of TCAS point, so for plane engineers it's safer to turn on TCAS earlier. Why RIMCAS engineers use "airborne" parameter as a single and true status of airborne airplane - this is the question.
@@brianorca That does make _some_ more sense... but it's still objectively the wrong call. Ground collisions are _vastly_ more likely than airborne collisions, and most of the worst collision disasters have been on the ground.
@@barefootalien Maybe they should find another term to replace 'airborne' when the aircraft is still on the ground but is in a critical phase of its take-off roll.
To me, the biggest surprise is how the Pilot Applied Visual Separation Rules in North America can allow for the air traffic controller to give take-off clearance to an aircraft, before ATC can visually confirm that the preceding aircraft is actually airborne. We're talking about a couple of seconds here. And while I understand that there are economic incentives for shaving off a few seconds, from a safety perspective, those extra seconds seem quite well invested.
Brings true tears to my eyes for the 777 Captain. Placing the safety of human life above and beyond what the dollar allows. Indeed, no time for complex decisions. The Captain of Embraer, ditto, same. Outstanding decisions by both. Definitely needs the same attention to safety as Sully on the Hudson.
A wonderful presentation by our Mentour. Beautiful illustrations. Brilliantly thought out and presented. Pilot, thoughtful, illustrator and producer, to say the very least.
The bird probably was hit &?died and could have gotten ingested in the next plane engine had the embrea taken off & not abandoned takeoff.
I find it incredible that die aviation industry is still primarily using a form of communication where sometimes messages just don't come through and nobody even realises that a message was just lost. Baffling!
aviation is in fucking shambles and it's cheaper to blame individual pilots and controllers than to actually make comms and management not a total shit show
They could use separate frequencies for transmit and receive to allow full duplex communications, but nobody would want to pay to replace all the radios.
@@TianarTruegard ok but in situation like that one, the 777 will hear the Embraer is rejecting, but the tower can still only hear the 777
@@jonathanma2741 You know we've had group calls for ages right? I understand AM radio has _some_ advantages compared to more sophisticated digital communications, but it seems the disadvantages far outweight them.
At the very least they should use a dual system, so if the digital communication gets disrupted for whatever reason you can still fallback to AM (and this should happen automatically), but the normal workflow should be handled by the superior system.
@@demoniack81 As I said, duplex communications radio couldn't help the tower to hear two communication at once
I'm not in aviation but I find your videos interesting. You explain enough without over explaining and you are presenting these in an objective manner. This is what happened, this is what's normal, this is what was a bit off
thank you
We should really apreciate his work. I feel like this video (research, graphics and editing) took soo much time. Great work Peter, loved it!
Hi Petter, I really appreciate that you talk about the lessons of near-misses as well as accidents.
It would be great if you could cover the serious near-miss of Air New Zealand Flight NZ60 at Samoa on 29 July 2000. This was an issue with a erroneous ILS glideslope, flying over water at night. When this "failure" happened, the instruments indicated they were perfectly on glideslope no matter what their altitude was! The flight descended as low as 340 feet, about 9 kilometres from the destination runway. This event has been discussed in Air NZ training but I don't think any other aviation commentator has covered it.
There were a lot of holes in the swiss cheese, but fortunately they didn't all line up this time. Factors include the water "black hole" effect at night; over-reliance on technology; missing a checklist item while busy; an unusual "failure" mode of the ILS; the fact that GPWS is suppressed when landing; challenges of flying into remote island airports.
The NTSB in the USA seemed to hold the opinion that this incident was primarily an issue of pilot error but others placed more emphasis on the erroneous and misleading ILS indications. Your opinion on all that would be absolutely fantastic. Cheers from New Zealand!
After viewing any of Peter’s videos the comments offer an equally valuable contribution to his presentations.
Thank you all for your participation and insights into this experience. I encourage all to offer, in my point of view as I am someone who loves to fly but found it proper to discontinue my private pilot training when I experienced too many flaws existing at the time most of which have been addressed currently but the learning process is never finished.
I apologize I did not mean to write a book here.
Respect.
Hello! I’m actually having my solo flight next week! I’m nervous, but this is an amazing video! I’ve truly forgotten all my fears after watching this!😊
Awesome! Best of luck!
Yay, have fun! 🥂
Good luck.
Go luck and congratulations. You're going to do great and have fun.
Good luck. We're all counting on you.
This is such a good channel. I live in Toronto but didn't know about this incident. I love the way you break down the events that combined to cause these kinds of incidents and make it easy to understand for people who aren't involved in aviation. Great visuals as well!
You have one of the best channels on RUclips. In depth explanation while keeping it simple for non airline people. Keep up the good work
I always feel more informed after watching your content, Petter! Much appreciated!
That’s great to hear! Feel free to share it with your friends. It really helps the channel. 💕
@@MentourPilot shared the channel with some friends and they’ve subscribed!
We in aviation are so good at learning from our mistakes and making sure that any future mistake hasn't already been made,… except for this. There's NEVER a good reason to make or allow for assumptions. An Embraer pushing 50kt is automatically airborne? A 777 at 100kt?? Hell, I can keep a 172 on the ground at 90kt if I push hard enough!
This is a perfect example of why systems MUST only give ACTUAL instrument data, not projections or assumptions.
Another fahn-TAHSS-tick video, Mentour. Keep your blue up and your brown down.
I can't believe that it was considered acceptable for them to have 2 aircraft taking off simultaneously on the same run way, that just sounds like a recipe for disaster. What happens if the following aircraft was more focused on the instruments and missed the aircraft in front stopping
phew, that was a close call, ATC would have not notice it until another Tenerife disaster strikes with 3 planes involved
@@lukepevensie Tenerife was in cloud with visibility ranging von 100 to 900 meter because clouds move over the runway. The controller couldn't even see the runway.
I agree. My first thought was what about if a takeoff has to be aborted. I had an aborted takeoff last week (student pilot) when a plane that was entering the downwind instead cut across the crosswind, opposite the direction I’d have turned, and we’d have reached the same spot in the air at the same time. Dumbass made no calls whatsoever. The field I was on wasn’t long, and there are 100’ tall trees 33’ from the end of it. I was almost at Vr and had to abort. The thought of another plane starting behind me makes my heart race.
Canada...
Exactly, now its ok to understand how things happened but there’s a lesson to learn too, maybe time to change rules before it’s too late
RIMCAS seems to prioritize capacity instead of safety. In my experience, when modelling complex systems (which is part of what RIMCAS is doing) it is always best to keep everything as closely to reality as possible, so considering an aircraft "airborne" when it in fact isn't, just to decrease the number of "false alarms", is asking for an accident to happen.
In other words, it should IMHO not be up to the ADS-B in the aircraft to fake the signal that it has become airborne in order to allow for its "customer" (RIMCAS) to squeeze the time between takeoffs - that consideration should be a concern of the receiving system. Any optimization should be done there. Also, when determining whether the rules are safe or not, one has to take into account that mistakes will happen, like the 737 stepping on the Embraer's transmission in this case. Here, there was one more slice of Swiss cheese that saved the day. I just wonder how many lives will be lost before this way of doing things is deemed unsafe and revised.
Not a pilot but that seems like a pretty smart fix; If "airborne" was just based on sensors that directly measure that factor like the weight-on wheels switch and radio/standard altimeters, which seems like the original design, and it was up to the receiving systems to apply their own additional checks to suit their individual use cases, that seems like the safer scenario. If RIMCAS didn't disable until the aircraft was 100% positively airborne, and TCAS activated if there was any reasonable chance that it could be airborne, their overlap would seem to offer an additional redundancy covering what seems like a fairly critical edge-case
I know I'm oversimplifying in saying this but wouldn't it be better if the RIMCAS/ADS-B airborne status were tied to the V speed settings instead of some arbitrary number such what was saw in this video? For example, pilots calculate their VR speeds and feed it into the FMC, that VR speed data is then somehow given to the ADS-B/RIMCAS as a better estimate on when the aircraft is airborne.
@@SVnerd or at least if a set number needs to be used it should accurately reflect the stall speed of the aircraft, there is no way in the world the embrair would be able to stay airborne at 50kn, but that should be a last resort
Better yet, have pilots set a beacon when they reject a takeoff rather than simply report it over an unreliable communication system.
Seems like the best thing would be to no takeoff clearance until the entire runway is clear. Waiting until the plane ahead has cleared the end of the runway seems like it would only cost a few seconds. Starting a takeoff on an occupied runway just seems like asking for trouble.
I'm a military heavy pilot. I'm surprised we don't have an automated way of detecting radio heterodyne (blocked radio transmissions) and alerting the callers.
The tech we use in civil aviation is truly ancient. The HF stuff cracks me up. So glad CPDLC is getting widely adopted now
I still marvel at how dangerously bad VHF can be but also at how difficult it is technically to come up with an alternative - to find a part of the spectrum we could use globally and get everyone to switch over at once. Impossible!
The technology is out there. The airlines are too cheap to implement it.
Yes, indeed - bingewatching here! I am absolutely fascinated. The visuals and explanations are phenomenal. Thank you so much. I would also like to comment that both your warnings to sensitive viewers and your respectful handling of fatality reports shows a great deal of kindness,& properly honors the departed individuals.
The fact that you make a point of showing what was learned from a tragedy makes one feel that the lives lost were not in vain.❤
In my aviating days I witnessed a number of incidents like this were a takeoff clearance was provided while runway occupied. In one case a KC135 aborted while an F4 was rolling behind it. The F4 in full ad just clearwd the tankers tail.
My personal pre take off checklist was, heads up, runway clear ahead, left and right, then roll.
That 50/100 knots limit sounds like a pretty bad design choice.
They should've use the nosewheel pressure sensor instead of speed limit.
@@cupofjoen The speed activation thing is a backup in case of faulty weight on wheels switches that never detect the plane getting airborne, and thus don't activate the in-air TCAS system.
I agree, it's ridiculous. What does 'airborne' mean? In the air. So is it in the air? No. Wtf.
The answer is already in the video: "Increasing the 50/100 knots limit significantly increases the false warning rate which in turn make the entire system much less reliable"
@@dvdraymond having a speed activation alone seems very poor for RIMCAS - I can understand it activating TCAS but having it turn off RIMCAS seems dangerous (especially without another confirmatiory system - like change in altitude).
I don't know how many of your videos I have watched where something goes wrong because the air traffic controller can't hear an important message because it came in at the same time as another one from a different aircraft. There should be a warning system in place, to alert the traffic controller that another message came in from 'X' aircraft, so that they can contact them right away after they're done with the first message that came in. Or better yet, the system should record what is transmitted and alert the traffic controller that they have a message that came in while they were handling another communication, so that they can just press a button and listen to it.
I don't know what the difficulty of doing this would be, but I think it would prevent a lot of problems.
thats not how it works,. the messages are not heared because they are not received. in essence, the more powerful signal destroys the weaker signal. there is no technical way for the sender, the overpowering sender and the receiver to know about what happened. there is no message to record, there is no warning to alert of a missed message.
surely there are ways to transmit information in a way where you can notice, record and replay messages. thats how the phone system works. however, that system is way more complex, and expensive, and error prone, which is why the old radio waves system is still in use and while not perfect has many advantages. you can more or less literally build a receiver out of the lint in your pocket and a ripped of button (not really, but its really basic and not sophisticated technology), and as long as you have power, it works, no other or moving parts needed...
and you are right, this has certain drawbacks that need to be taken in account. one of them is you can not know if your message has been received. this is where "over" and "over and out" comes from (making sure the whole message is heard and signaling others they can now talk) and why its important to acknowledge receipt of a message like "heard" or reading it back. its also why radio chatter needs to be kept at a minimum. you only can hear one sender, and when someone else is sending, nobody else can be talking. or they could, but only the strongest sender is being heard.
while the video did not make mention of this, and i dont know the rules, that the air traffic controller did not acknowledge the firsts plane abort message might have been where you could have saved the day here. the pilots could have repeated the message because it was not acknowledged.
Still ridiculous why aircrafts are still using analog radio for communication where only one person can speak at the same time while everybody has a smartphone which can handle easily two or three persons speaking at the same time.
Digital radio would also allow to identify each aircraft which is sending by adding its call sign to the digital data
@@catwiesel_81 "there is no technical way for the sender, the overpowering sender and the receiver to know about what happened"
There is an easy way to detect this. If somebody is sending something his receiver must verify that he receives (nearly) the same signal which the sender is sending and if not create a warning that somebody else used the frequency for sending at the same time.
"surely there are ways to transmit information in a way where you can notice, record and replay messages. thats how the phone system works. however, that system is way more complex, and expensive, and error prone, which is why the old radio waves system is still in use and while not perfect has many advantages. you can more or less literally build a receiver out of the lint in your pocket and a ripped of button (not really, but its really basic and not sophisticated technology), and as long as you have power, it works, no other or moving parts needed..."
I never heared that an aircrew build an analog radio using "lint in the pocket" or "ripped of buttons" while in the air....
That may work on a ship there you have plenty of time but not in the cockpit of an aircraft.
And our digital phone systems like 2G,3G,4G and LTE are working very good with hundred millions of users even the used hardware comes from many different companies - the only scenarios where it comes to problems is either power outtage or overload by to many users in the same cell.
There are a few ways to mitigate stepped on transmissions. The foremost being a response including the aircraft callsign and "roger" or "acknowledged" etc...this should alert a station that broadcast at a similar time that their transmission was either missed or stepped and and prompt another call. (Happens a lot). When things are moving quickly things get missed and stuff like this happens but there are procedures in place to prevent this so its not a constant safety issue under normal conditions (former pilot, current ATS)
@@elkeospert9188 works great until you surpass the incredibly limited range of cellular. Be it vertical distance or otherwise, you just simply don't have options for utilizing a cellular signal at 20+ thousand feet as I understand it.
Ever get out in the ocean on a boat? You lose cell signal pretty quick. Plus what airline wants to pay a mobile carrier to handle their mission critical safety communications.
Radios have very few failure points. You need a working transmitter and receiver. No towers, no fiber runs, no data centers, and importantly, much less latency.
What! I’d praise the 777 pilot, his decision was not “understandable”, it’s praiseworthy! There could be many reasons why the front aircraft aborted, like something on the runway caused damage to its tires etc. there’s no reason to brute force that takeoff.
or something from the runway surface got sucked into its engine, etc.
on the 777 imagine if the T/O calculations were wrong or they lost one engine power after V1 and needed more runway space for the rotation, they made the better choice
@@Phvpark It seemed the Boeing didn't know by number where the Embraer was either. (The systems thought it was airborne and the Boeing wouldn't have had a number.) Was there time for the math even if the number was known? I don't think so. "It looks like we can stop -- let's do it" seems to be it. A new safety regulation was written in brown pants -- not blood, thank God!
Do jet pilots practice takeoff roll panic stops? If not, should they? Is there some rough mnemonic for the normal panic stop distance vs. speed? Would a pilot remember it as such a rare thing to need?
All we can say is whew, and don't crowd takeoffs like that when you don't have a near bulletproof collision avoidance system.
In situations like this I usually ask myself "How could this happen?" Your clear, thorough explanations answer that question very well.
Dang, this guyshould be a aviation lecture, he can explain every critical system smoothly and clear
When I was a programmer I would explore every possible combination of parameters to insure that there would never be conflicts. "It's unlikely that would ever occur" was not good enough. It could not occur, period. We avoided the kind of engineering that would risk building a seawall too low to prevent a tsunami from flooding a nuclear power plant. "Unlikely" means "possible".
Excellent and thorough explanation. One issue that wasn’t mentioned is that the tower was designed for two Local Controllers when traffic requires and management did not staff it in accordance with that workload. We spend millions on design and sight lines and place highly trained professionals with excellent eyesight in towers in the sky just so they can use visual separation to safely move traffic. The second aircraft should never be cleared for takeoff until the tower controller visually observes the first one rotate or can ensure separation will occur. Giving the flight crew the responsibility to maintain this separation from another aircraft that can be as much as two miles away just because you want to staff one local controller is not ensuring safe operations.
Yes, indeed.
Okay, serious props to you for your youtube channel allowing you to be able to write off the top tier beautiful airport and scenery addons for flight sim, and maybe even the gaming computer to run it.
I notice a theme in some incidents - that two or more people transmitting simultaneously on a frequency can result in information being lost. Wondering if anyone has looked at a digital packet switching system with, say, time division multiplexing could allow all the signals to get through. The tower would then have heard both transmissions.
I think there's just a reluctance to change.
Consumer handheld radios have a 'busy lockout' function for a long time which stops you transmitting if a carrier from another transmission is detected on the same frequency, that would at least let you know that your transmission didn't get through so you can resend it.
For backwards compatibility aviation radio/airband is still based on original analog AM modulation.
Yes! This risk must be addressed. That could be done just at the receiving end, with upgraded tower equipment. Multiple incoming analog voices could be computer-separated using voice recognition or other clues, and then each stepped-on call automatically and immediately replayed (of course not conflicting with newer incoming calls).
The different messages are not arriving from the same azimuth so an array antenna could separate them. This is somewhat complicated by the very low frequencies, and therefore long wavelengths assigned to this fonction. Fundamentally it does not seem appropriate in the 2020s to restrict aircraft safety operations to using 1950s technology. This issue also arises in the conflict between 5G telecommunications, which uses 21st century electronics, conflicting with the radar altimeter which is likewise stuck using 1950s technology. There is no reason that the aircraft could not transmit the information using BOTH 1950s AM analogue transmission and 21st century digital transmission. It could even send a text message through the 5G network.
@@jamescobban857 you're right in one aspect, that technology could help both messages get through. This problem happens whilst aircraft are in the air too with the potential for similar dangerous issues.
However if both transmissions get through the controller may not be able to tell who called or what they said due to the garbliing of the voices coming through at the same time. There would be no certainty that it would help much.
In the air a controller may have up to 30 aircraft on frequency at any one time so this problem can be much bigger in en route sectors.
CPDLC, (electronic messaging) does help reduce transmissions and is probably the way to go regarding this "call blocking" issue, even on the ground.
The only minor error here was by the Embraer crew that reported rejecting the take off. They should have expected a reply from the controller. I realise they were busy in the cockpit, but when no response was heard, they should have repeated the message.
The animation just gets better and better.
Thanks! That’s what we are aiming for! 💕
@@MentourPilot Same feeling, excellent animation 👍
The animation is excellent, but I was a little startled to see trees in full green leaf in Toronto in February. I guess this shows how well I was taken in by the realism! ;-)
Wow! I've been up the control tower, and this is so accurate. Even the layout of the tower is correct, down to the staircase location.
Super impressive video, I subscribed!
It really suprised me that this takeoff clearance situation was allowed in 2020. In the video, when you said "gave clearance and 777 was accelerating" I was immediately thinking "but if the Embraer needs to abort and stop" ... and then they had to abort and stop. But there is so much more puzzeling:
1) To me, in context with takeoff clearance, airborne should be defined as the plane being a couple of feet above ground and in a clear lifting situation.
2) A controller should not be allowed to look away and turn his/her attention until the planes are safely in the air and no longer related to traffic on the runway.
3) Having to watch and control traffic on opposite sides just feels like a stressful situation, prone to cause such situations.
4) And considering that all this is done to safe money ... I guess it is history that will repeat again and again. No amount safed can repay the potentially catastrophic outcome of such conflicts.
So, I am happy that this was no longer permitted in Europe. And I fully understand the 777 pilot's decision to perform a hard stop. I wondered if they could have taken off in good time anyway. But there would have been the risk that they misread the situation and the other plane would have attempted a lift off anyway, in which case they might have collided in mid-air.
Anyway, happy it did end well and always good that aviation looks into every incidence, even the ones that did not cause damage or loss of life.
Actually, it's just north america region that's built different. Not only in EU, but in pretty much every other part of the world (asia, africa, australia, etc), dual takeoff clearance is not allowed. ATCs are not supposed to give another TO clearance before the 1st aircraft reported airborne or identified on app radar. Heck even virtual skies such as VATSIM/IVAO which implements standard ICAO rules does not allow that.
Regarding your #2, sometimes it's a bit dilemma when you have runway on both sides. As mentioned, the controller went to check arrival traffic on 05. If he keeps his eye on 06L, then he would have no idea whether the landing traffic landed successfully, vacated runway, or going around, etc. This is even more important when there's 2+ tightly separated traffic inbound ("highly optimized", quoting the report). The other points... yeah, money takes priority over people's life (and sadly that's how the world works)
#2 is wrong, there should be enough controllers to cover all runways in use, if there aren't, the excess runways need to be shut down until enough controllers are available to cover them all.
the fact that someone is being expected to stand up and look behind them to make sure nothing crashes over there as well as in front of them means some manager or executive or administrator belongs in prison
@@0x73V14 I 100% agree to that. And I probably phrased #2 improperly. Often when I listen to Peter's analysis I get the feeling that the controll towers are understaffed. But to me this one has beaten most others. There should be one controller for each side. And yes - management responsibility.
@@AbWischBar Indeed.
While flying keeps getting safer, I’ve heard that the incursion incidents are on the rise. Is that true? If so, I’d love a video that describes why this might be so and what the industry is doing to better handle the situation. Perhaps more appropriate for the other channel but this video reminded me. As always, awesome content!
Over what time scale, we just had a global pandemic, which a huge drop in flights, so since things have opened up, of course we're getting more runway incursions, simply because things are busier,
Maybe because airports are extremely busy?
I’ve heard also about runway excursions occurring more frequently.
I’ve usually heard it as part of news coverage when some accident or near-accident is being reported, but to Anne’s point, there’s never enough context or data shared, so that’s why I’m not sure if it’s even a real thing. If it’s true I’d assume it has to be as a % of the number of flights attempted, otherwise if there’s more flights now than in the year 2000 (for example), then there’s likely to also be more incidents. With more news outlets and content creators emphasizing viewership and clicks, sound bites over context, it’s really hard to know if you’re getting accurate information.
@@ellicel Yeah pretty much this. We'd need long term percentages per year to tell if in total runaway incidents have occurred more often accurately. I don't doubt it is the case in some airports given different environmental effects and management skill levels.
Honestly this question is so localized it likely is better to look at individual airports statistics over national statistics to get a better feel for where it tends to happen the most at.
It basically would have worked. if this story tells me summin, its that we shud be doing double or triple or more take offs at once. The system is simple, first you taxi all the planes to one end of the runway, then you send the first out. The first one goes onto the runway and starts to take off, if after 30 seconds they haven't started taking off they have to immediately turn around and go to the back of the queue, and the next plane taxi's on. The plane that failed section A has the pilot reprimanded and they lose 5 pilot points (if you haven't heard of my pilot points system I'll post it below). Anyway once a plane doesn't fail this, and starts going off, the next plane immediately taxis onto the runway as soon as the pilot is out of the way. This crew then also starts their 30 seconds.
By the time they've done their checks in 30 seconds, then the other plane will be at least like 30% of V1. So plane 2 starts immediately because it's effectively safe. Once the first plane takes off it immediately banks to the left to prevent backwash, similarly the second plane banks to the right, third to the left, etc etc.
So what happens if a plane aborts take off? Well it depends. If the pilots abord after V1, then the other planes just continue like normal, since that plane is going to go off the end of the runway anyway and not pose a collision risk. So the other planes just carry on like normal. Why waste revenue when there's no chance of a collision?
Similarly there's a new requirement for any aborts, the plane must immediately taxi off the runway we could even develop shallow angle off-roads for this. Anyway, we could simply have a computer system model this. It just checks whether the aborted plane will be able to make it off in time +5 or +10 seconds. If a plane can't get off for some reason, then the other plane simply uses these, and only in this extreme situation do we pause take offs.
Just think of how much time and revenue a system like this could save. And as this video proves, if anything it'd be better
* pilot points are an idea of mine to prevent pilots causing errors like this. E.g. a pilot who cancels take off and delays other planes, has pilot points removed, e.g. 10 points would be removed. Pilot points are awarded per successful *in the air* hours E.g. a pilot would earn 0.1 pilot points for each air hour, but could also gain points by e.g. not leaving their company hotel while in another country, reporting other pilots for violations, etc. They'd lose points for: customer complaints, not having as many customer compliments as the average, incorrect uniforms (e.g. not tucking in your shirt, etc). Simulator hours do not count, since they don't generate revenue. Then at the end of each financial year, the pilots with the least points would face disciplinary action. The pilots with the most would be able to spend it on things like gift vouchers, trade them for time off (at the companies discretion), use them when they're ill and can't fly, etc etc. All of this would massively increase profits for airliness,would
It's nice to hear about what happened from someone in the industry. I live about four hours away from Toronto and heard about this event on the news and enjoyed the play-by-play of what happened that you gave us. It reminds me of how effective the Swiss cheese analogy for incidents is.
I see what you did there, subtly showing a KLM 747 taking off when explaining that you cant start your take off roll unless the preceeding aircraft is already airborn in Europe... 😎
I noticed that too!! Carefully setting everyone up to think 'Tenerife' even before they were fully aware of the situation as it unfolded. Masterfully done.
@@Teverell Yes.
If you let the brakes go with another aircraft on the active you should believe an ICBM is seconds away.
Thank God. It was a miracle. I think everybody involved did a great job to avoid a disaster. Hats off to them. Thanks for the video, well explained.
Perhaps the Embraer FO could have re-iterated his aborted takeoff message when he didn't receive immediate confirmation from the tower. Nothing wrong with saying something that important twice.
shame ATC confirmation to pilot wasnt required
During a rejected takeoff, the FO is kinda busy to be repeating calls. He's got a checklist that needs to be executed correctly, *right now.*
What's needed is a more effective ground communication system that is robust to this kind of interference. Not at all sure how that would be done, unless they went to some kind of digital system that both preserved communication even under low signal strength, *and* inserted key action words into the ground traffic management system display.
Non-pilot but it seems like there would be a pretty consistent set of actions involved in rejecting a takeoff: spooling back the engines (I'd assume back to idle unless reversers were used which could be another flag), applying brakes, and probably some other aircraft-specific ones. Could the ADS-B be reprogrammed to detect that a set number of these actions are being performed, maybe even triple checked by a continuous reduction in airspeed, and broadcast an "aborting" status? That would seem to offer some degree of redundancy to the radio call without increasing the FO's (likely already high) workload, although I lack the experience to know whether there would be too many potential side-effects
Sorry, I said the thing thing, before I saw this comment.
@@lairdcummings9092 if you have 3-5 antenna, you can determine pretty precisely where each signal is coming from. Which maybe doesn't prevent the smaller signal from being stepped on, but it does tell you that more than one signal is being transmitted at the same time and who did it.
Reminds me of an incident I remember in the news somewhere in Japan years ago where the controller cleared an plane for take off but the pilot of a hellicopter thought it was clearance for him to take off and he flew over the runway causing the plane to reject their takeoff. Meanwhile, the controller had cleared another plane to land on the same runway and didn't realize what was going on until the landing plane was too low in it's pilots opinion to safely abort the landing...
I’m so glad I found this account. These videos and your in depth explanations are just incredible.
You and your team are everything that is great about RUclips. Keep up the amazing work!
Aww thank you!
@@MentourPilot I saw this incident on another channel but you did a much better job explaining it here. Great visuals as well.
@@MentourPilot thank you for taking the time to respond! It just proves how in touch with your audience and professional you really are. I can appreciate the amount of effort and dedication involved maintaining a relevant multi-platform production at your level. You deserve all your success sir, great work!
In modern high tech manufacturing, many tools on factory floors send out warnings to engineers in charge of their maintenance. Therefore, the fact that nuisance warnings increase the risk of the engineer ignoring them is not lost on me. However, I agree with other comments read that RIMCAS at least at the time of this incident was perhaps leaning a bit too much towards capacity rather than safety. It was fortuitous that the hole in the last slice of Swiss cheese didn't line up, and disaster was averted, when the 777 captain aborted take off on spotting that the Embraer had rejected takeoff.
Looking at the state of automated warning systems in general, not just in aviation, these warnings are a great tool and simplify things for us humans, but we still need to be very alert. On another interview about a general aviation accident involving a highly experienced pilot in the USA, an NTSB official remarked that pilots need to always be vigilant and have the mind set as though they are operating chain saws.
Petter, I had read about this incident elsewhere, but your narration had my pulse racing, even though I knew the outcome and your description always maintained a calm and even tone. Thanks again for a clear and succinct description of RIMCAS and its advantages and pitfalls , that is easily understood by someone who is not involved in aviation.
I'm hooked to these videos and appreciate the clarity of the technical explanations to the point that I have a greater understanding of what takes place in the cockpit. Thanks Peter
I find it quite strange that at the same time the ATC feels the need to push throughput while at the same time the tower was not fully staffed because of moderate to light traffic. If such techniques are required, the tower should be fully staffed.
Somebody got a big bonus that year for doing just that.
Sometimes they're busy. Usually they aren't..
If you open all the positions in the US often the controllers aren’t happy because it cuts into their break time. In the US the average controller works on position about 4.5 hours a day maximum. Some facilities more, some less.
@bart solari they do whatever they want. At my facility watch tv, eat, go to the workout room or complain to the supervisor they are overworked!😂
Peter, you set this one up with perfection and had me on the edge of my seat! I didn't think it was possible but you're getting even better with each video.
I wonder if it wouldn't be possible to just update the radios at this point. I understand it's a massive undertaking as all planes and airports need to be updated. But the fact that important transmissions can be blocking each other without any kind of warning that it happened, is quite dangerous.
There are always tradeoffs. A digital radio system that can avoid this issue is much more complex than simple analog radio, which means there's a higher risk of failure. Analog radio is dead simple.
@@gwillen Emphasis on the word 'dead'
This situation shows me yet another time, that rules here in EU, even tho they sometimes seem to be crazy, are completely logic. It's not that other are worse, it's just us being extra cautious, which could avoid this particular situation.
With the rules and operating procedures that were in place, an incident like this was absolutely inevitable.
Beautifully assembled and communicated post. Thank you so much from a long-term admirer in the UK
Great explanatory video, as always. I'm so used to safety being a top priority in aviation but a situation where an airport has a "safety system" that receives/emits false information combined with an Air Traffic Controller being required to turn his back on an active runway beggars belief.
😮😅 0:23
Re: 'Masked' Radio Transmissions. One of the peculiar advantages of the legacy, very old fashioned, Amplitude Modulation (AM) mode still used in the VHF ATC radio band is that simultaneous transmissions are audible as beat notes (caused by the slightly different frequencies) and garbled noises. This is quite different than Frequency Modulation (FM) which can have a 'Capture Effect' in that the weaker transmission can be 100% totally inaudible. Those involved with ATC communications should be aware of this radio communications detail and listen-out for such situations.
I have been binge watching your entire catalog of videos for the past few days! Loving all the factual and unbiased breakdowns of these events. With that said, nothing has made me laugh like the way the graphic of that bird went up and across the screen after it hit the side of the plane 😆 HUGE thumbs up to whoever put that together 👍
❤❤❤❤
I loved the "how the human factor saved the day" comment at the end. Sometimes the human factor can indeed save the day too!
Well done video, as always! Hats off to you and your production crew :)
Hello Petter, I just would like to say that I've been a very nervous flyer the few times that I've been a passenger in an airplane before, mainly due to having watched documentaries that focused more on sensationalism and drama, and I think that your videos have done great job with explaining these accidents and incidents and how the industry has learned from them in a much more logical and compassionate way, because of that I think that I'll feel more safe about being on an airplane if I ever have to, so I just wanna say thanks for that :)
I'm always amazed at quality of the recommendations that come from those reports.
The industry is held to extremely high standards while also understanding that humans have limitations that have to be realistically managed.
I'm a bus driver, and for years I have heard about how bus company operators can't wait for autonomous vehicles. I have always pushed back against those sentiments, highlighting that a human operator can't be hacked, and can process a situation and take action more quickly than a computer.
I know autonomous aircraft were not involved in this incident, but I can't help but think how this may have played out if the human element wasn't present.
Regardless of vehicle type, highly trained and skilled human operators are essential to the safe movent of our passengers and cargo.
As critical as this incident was, I commend the actions of the crews of those aircraft, and I'm glad we are all here to learn from this incident.
Years ago I remember an incident with a "drive by wire" bus that collided with a stop - bus driver said the bus went out of control however stopping the engine cleared all the data from the system (handy eh?). Dr Hannah Fry also thinks autonomous technology will never work.
you seem to be missing the consideration of how much faster and accurate computers are than humans. think about all the problems and crashes and incidents that are caused by human error, which computers could have avoided
But autonomouse vehicles are safer because they not only obey laws, they can also predict what the other vehicle is doing and calculate at a higher rate. The problem will always be humans interference. If you had a course with 100% autonomous vehicles vs one with 100% human drivers you would see how safe automated vehicles can be.
@@Interitus1 exactly! I think that the automotive insurance industry which makes billions in profit also has a hand in slowing down autonomous vehicles. because from their point of view, the safer the roadways are, the less reason for them to exist
@@Interitus1 autonomous vehicles lack perception - a human trait that allows us to comprehend a situation and not kill people. We've seen self-driving cars drive over cyclists and kill them. Please go and read up on the Turing Test. If the autonomous vehicle fails to understand the situation it hands control back to the completely unprepared driver. Which kills people.
As a software developer/designer, the way the aircraft sends Airborne flag and the way RIMCAS interprets the potential collision events seems very inadequate to me. First thing which causes red flag to me is the fact that the aircraft sends false information on purpose. It's showing airborne when it's still very far away from being actually airborne. There must be better ways to trigger that information more reliably. The second thing, that immediately caught my eyes was the fact that the system uses only actual speeds and not acceleration/ deceleration information. There should be some information/warning when accelerating aircraft suddenly goes into high deceleration. That is a clear indication of some issue and should trigger a warning.
Either way, amazing video as always and with a happy ending on top of that. Thank you.
totally agree, the deceleration should instantly trigger an alert if anything is behind
And there are probably reasons for both of those where changing it would make things worse, not better.
The early switch to an airborne state is so that RIMCAS deactivates, but more importantly, that TCAS activates. It also enables the type of operation as we have just seen, If you don't do it, a jet would not even be able to line up and wait until the previous aircraft was actually in the air, losing even more time. And in a world where time = money, that's not a good thing.
And as long as you are airborne, it doesn't matter what happens to your acceleration, RIMCAS will ignore you. This is once again because it would otherwise generate too many false alerts and controllers would start to ignore it. Systems like RIMCAS need to find a fine balance to ensure it doesn't alert too often and doesn't miss too much. And that is never an easy feat.
Yes, it is easy to create a system that would alert on each and every small thing, but as humans, we tend to ignore alerts when they happen too often. Car alarms are a fine example of this imho. 20 years ago, all high end cars had car alarms and boy, did we know it. You couldn't walk on a parking lot without hearing one go off somewhere. After a while, no one looked up anymore when yet another car alarm went of and as a result, very few cars still have that type of alarm these days cause we all ignore it.
I'm not a software developer, just a structural engineer who occasionally has to program as part of his job, but I'm going to push back on the "way the aircraft sends the Airborne flag" as being bad. I totally agree with you that the way RIMCAS interprets that flag is defective, though.
The way the Airborne Flag (which I'm going to call "AB" hereafter so I can use "airborne" to refer the the actual state of the aircraft), makes sense. Remember, it was _invented_ for TCAS, and then only many years later used for RIMCAS. Because it was invented for TCAS, what they appear to be going for is that you don't want an aircraft on short final to get TCAS alerts from aircraft on the ground near the runway, because then that would always happen and the pilots will start tuning out TCAS. You absolutely cannot have that, because the only way TCAS works is if pilots *INSTANTLY* do *EXACTLY* what TCAS says without hesitation. If they have to start filtering for false warnings themselves, you might as well throw out the whole system. (I imagine false warnings could also happen during lower overflights of an airport, and maybe in the pattern, but those will be fewer than alerts on short final from, e.g., the aircraft that will likely be sitting on the end of the runway.)
The real rule for it appears to be something like this: Set AB=True if the aircraft is Almost Certainly airborne, but it Must Never be set to AB=False if the aircraft is, in fact, airborne. If this is what you're trying to do, the it seems pretty damn reasonable to me to have the actual software use the following, with WOW being the "Weight on Wheels" switch, a boolean that's True if the switch is squatted and False otherwise, and V being the measured airspeed in knots:
if WOW == False || V > 100 then AB = True
Then, the TCAS system on the aircraft can crack open each ADS-B packet as it comes in. The first thing it'll do after decoding it is look at the AB flag. If AB=False, then the TCAS system can yeet that packet into /dev/null and move on to the next one. Yes, the AB flag might not be correct, but if it's incorrect it's going to mistakenly be reporting AB=True. For TCAS, it doesn't care--the circumstances where this incorrectness are likely to come up are ones that (1) won't matter, because the place you have a low aircraft that may come into conflict with a mistaken AB=True is when it's landing and the other aircraft is taking off, where TCAS won't alert because they're not closing, or (2) it will matter, but the incorrectness in the airborne state is likely to result in a correct TCAS alert, just a bit early--this would be where an aircraft is about to take off, but hasn't actually left the ground, but the airspeed rule causes AB=True. If, under those circumstances, you get a TCAS alert it's pretty likely that the aircraft, when it truly becomes airborne, will be on a collision course with you! So this simple rule for generating the AB flag, and the simple "delete packet and move on" rule, both work for what they were designed for.
I get what you're saying in that you can probably improve the correspondence between the airborne state and AB=True, but I question if the increase in complexity in generating that flag is worth it. I was able to describe it in a single line. If you start ladling in more data, now that gets more complex to both program and understand. Yeah, you can start considering the vertical airspeed and the acceleration/deceleration, etc., but I think there are problems with that:
1) These other data sources are much more complex themselves. The WOW state and V I gave above are both quantities measured directly off their sensors and piped in with very little processing--the WOW is "is there +5V between these two wires," and V is probably pretty close to "here's a 16-bit value of the airspeed in knots at this exact point in time" updated every half second or so.
Stuff like acceleration, vertical airspeed, etc., all have to be computed from a time series of other things, or from laggy sensors. For example, the old Vertical Speed Indicator "steam gage" instrument worked by having a calibrated leak into a chamber with the current outside air pressure. If you had a certain rate of climb, the air would leak out at a certain rate as the pressure inside that chamber dropped, and that moved the needle to indicating your rate of climb. If you climbed faster, the pressure differential generated would be higher, so the air would leak out faster, and the needle would point to a higher value. However, it would take a bit of time for the leak to reach its equilibrium, so the VSI needle would noticeably lag your inputs. Right before you initiated the climb, there would be no air leaking, so it would point to zero. Then, you started the climb and it would take a bit of time for the air leaving the chamber to reach the equilibrium leak rate, so there'd be a second or two before the needle would "catch up". This is useful for checking that you're climbing at reasonable rates, but you can't use it to make immediate control inputs--something new pilots have to learn--and I'm not sure you want to use it for this AB flag, either.
2) Now, to be clear, I'm not saying that you couldn't work through all of the potential issues in (1)--I'm sure you could come up with a bunch of if-then statements for each potential sensor (OK, if WOW=True || (V>100 && VSI>50 for the last two seconds), and, and, and..). However, this means that you're going to have a really complicated set of rules that now generate your AB flag. I was able to state how it's currently generated in a single statement above; now, you'd need a f***ing flowchart to even hope to understand how the aircraft generated the AB flag that it's reporting. I know for sure it would make it difficult to understand for me sitting in a chair watching RUclips, but I wonder how complicated it's going to make writing and debugging the AB flag function. I can see this exploding combinatorically as you consider more sensors, increasing the chances of a subtle bug from a really weird combination of all these inputs.
Taking all this together, I argue that the extra complexity in generating a "better" AB flag could very well not be worth it. It does what it was designed for very well, and is very easy to understand. I do think that the Embraer setting it as low as 50 kts is probably silly, and question that--but note that this setting, does, in fact meet the "almost certain/must never" rule above! Both 100 kts and 50 kts are round numbers, which means somebody just pulled them out of their fourth point of contact as "good enough" so it's hard to say one is silly and the other isn't on other than an emotional level.
The real problem is that the people specifying RIMCAS didn't think through that the AB flag cannot give them the information they want with the way it is currently generated--as I argue above, it's pretty simple to understand--and probably should not consider it in their data flow. If you want to consider all the if, then, the other thing stuff, they might need to generate it in their own software. And it sound like they have a lot of that already, based on this video talking about the velocity rules. You always have to consider what something somebody else built was designed for if you want to use it yourself, and you cannot assume you know what it does just by the name! Given what TCAS was trying to do, naming their flag "Airborne" made sense, but the details of how it was implemented means it doesn't make sense if you're looking at a ground system. You have to go back and look at what they did yourself, because the name may make sense for their application, but not yours.
@CatCube2: You should consider what happens with a bad WOW switch.
@@Hans-gb4mv I completely agree that you have to strive for as minimum false or pointless alarms as possible, but that doesn't change the fact that the system is evaluating false information. As you said, RIMCAS will ignore you when you're airborne, even though you're still very much on the ground and that's the issue. An example solution would be to trigger TCAS above certain speed (for example 50 knots), but start sending airborne after you're trully airborne. Another options is to start sending airborne as it's doing now, but immediatelly stop sending the flag if you start quickly decelarting within certain timeframe of activating the airborne flag. Or keep using Airborne flag for the purpose of RIMCAS only if the vehicle is actually accelerating. There are many other options how to detect true state of the vehicle without creating false alarms.
Such excellent videos. You and Blancolirio are just the best.
Whoa, a story about Toronto! Didn't expect that. Thanks, Mentour!
I can't believe there isn't a technology that alerts you when two comms overlap on the same frequency. Would be of great help.
This overlapping radio keying seems to be a common factor in these reviews. I think it’s a difficult problem though.
If two mics are keyed at once the third receiving radio thinks the weaker one is “noise” and it’s signal is rejected. It’s not clear how one would control which one would be prioritized as “weakness” is a function of distance.
Once model for shared channels is the cell phone. In this case the cell tower dynamically controls the power of the cell phone, as well as gives it a time to transmit.
However cell calls are private and point to point so this time-shared system works.
There is only one human controller, likely they wouldn’t be able to understand two simultaneous messages and make any sense out of it.
@@terrygaryet1856 packet radio and other timesplitting systems exist but are either so old or so new that the tech just hasn’t been mainstreamed. These allow multiple communications on the same frequency, but require special tech to use, and would need an international overhaul to implement
@@terrygaryet1856 put three receivers on the ground that triangulate all transmissions, if two transmissions happen at the same time on the same frequency you can not only know it happened but plot on the controllers screen where the two colliding transmissions came from.
That is true. But there is also another possible resolutions to this without having such a system. Safety critical transmissions must be read back, in this case the Embrearor failed to repeat their rejected takeoff message.
@@Delibro that's also true.
It's a good time to be alive when @MentourPilot has to get his material from incidents where no lives were lost.
❤❤❤❤❤❤
There's a tremendous amount of work in that video, and it shows! Super well made and extremely clear :) From a pilot's perspective, the only tiny mistake I saw was the callsign "JZA" for the WestJet Encore Q400. Again, congratulatons on this excellent video!
The animation, the details and your narration are incredible! Very valuable content, thank you 🙏
Learning from this: If everyone follows procedures, there is no accident, only an incident.
I always look forward to your videos. They are just grade A in quality and information. Thank you!
Thank you! Excellent to hear you think so. 💕
Basic ATC (in the US) for successive Cat III departures on the same runway is 6,000ft and airborne. The controller should not have issued the 777 takeoff clearance until they had VISUALLY observed the E190 rotating. Relying on ADSB or ASMGCS for airborne status in the Tower environment on a clear day seems ridiculous to me for this very reason. Also, in my opinion, at that airport configuration, if RY05 and RY06L are both in use, local tower control should always be de-combined. And if the E190 came to a stop on the runway before exiting, they’re wrong. When I’m in the tower, it’s nice to have a little help from technology. But when it comes to runway separation, nothing beats a window and a good pair of binoculars.
I am only an engineer and not involved in aviation. But any system will fail if optimized too much. When safety or design margins become thinner and thinner eventually something with an infinitesimal small probability will happen. Design for inefficiency and redundancy and you have a good chance that it will be able to handle a situation which was calculated not to happen.
In the end, were there any recommendations or operation changes that came from this incident?
It's great that they learn what factors caused this, but unlike your other videos where you addressed what was charged to make this industry safer and prevent similar incidents from happening, I don't see it in this video… hopefully I just missed them.
Thanks for another great video, captain!
Or maybe nothing was changed. Sometimes there is no easy way to make things safer other than raising awareness.
20:00 there was an ATCI mentioning the weird programming, but the programming itself was not changed. this was done because simulations showed that increasing the speed threshold at which airplanes report that they are airborne results in more false positives, thus making the system unreliable and more likely to be ignored.