This is a fantastic story, and very well recounted.
> presumably because a simultaneous engine failure had been too ridiculous for Boeing engineers to contemplate
This is hard to grasp, esp. given the presence of a RAT. You're giving engineering time to add a device specifically made to handle loss of all engines, but don't spend the time to write the corresponding procedure?
> ...had determined the fuel weight by multiplying the the number of dripsticked liters by 1.77, as indicated by the documentation. However, unbeknownst to the pilots and the fuel crew, this multiplier provided the weight in imperial pounds; the new, all-metric 767 was based on kilograms, and required a multiplier of 0.8
Darn, how many of these incidents will we require until we finally get rid of the metric system once and for all!
> The internal investigation into the incident laid the blame partially upon Captain Bob Pearson and First Officer Maurice Quintal, who should have observed the Minimum Equipment List (MEL) and grounded the aircraft since it lacked functioning fuel gauges
This is an interesting take - and as the pilot of the plan it certainly should rest on you that you're responsible for all to be in working conditions before you leave. However combining that it was apparently a frequent failure, with the corporate culture that's outlined, it's a good cautionary tail about resisting the pressure when something's wrong but "everybody's doing it anyways".
Converting to metric for fuel seems reasonable but measuring fuel in weight (mass) and volume would still lead to errors using only meteric.
Distance and speeds use a mix of feet, statute miles, nautical miles, meters and kilometers depending on the context and country. Eliminating the difference would mean new approved flight manuals for each airplane, and quite expensive changes to flight instruments as well.
There’s a really good fit with feet for altitude in the U.S. that just works well. 1000’ of separation for opposite direction traffic, and 500’ separation between flight rules in the same direction. Directly converting to meters breaks all the simple math involved. Whereas if we keep the math simple, it alters airspace density and capacity.
Anyway, there’s no significant desire or effort to drop any of these units.
> This is hard to grasp, esp. given the presence of a RAT. You're giving engineering time to add a device specifically made to handle loss of all engines, but don't spend the time to write the corresponding procedure?
I imagine the RAT's primary purpose is to provide the power needed to restart an engine if all engines had stopped.
No, it provides power to power flight control systems, linked hydraulics and e.g. some flight instruments. You can't generate the power needed to re-start an engine from a RAT.
> Darn, how many of these incidents will we require until we finally get rid of the metric system once and for all!
Are you serious?
There are 3 countries in the world not using the metric system. The non-metric system is difficult to use because it's using 12 based units with a 10 based numbering system.
Sure switching over is going to be painful but that's true either way, and there are quite a few examples of people switching to the metric system but none, as far as I know, the other way around.
On jul-25 at 11am pdt, charles_f was standing out to comment on an interesting aviation story. Seeing in passing a reference to issues regarding imperial and metric systems confusion, and given the popularity of this kind of debates on the internet, they decided to set course on a "joke".
The joke was thusly constructed: the metric system is commonly used across the world and much more sensical than outdated imperial system, and people commonly announce their discontent for the latter and ask to discontinue its use. Therefore commenting using the opposite point of view as if it were an evidence, and in a serious tone, had a chance to trigger surprise and interrogation in the reader's mind, quickly resolved by the idea that the author wasn't being serious and making fun of this debate, and triggering a chuckle. This type of jokes is commonly referenced as "sarcasm" in the professional jokers world.
But before the joke took flight, the joke funniness sensor wasn't working, as well as the joke and sarcasm announcement systems. The author sent Robert, their left moustachioed fat finger, to dip in the joky tank, and check if it was fun. Robert answered "yo bro yain't starting to finish all your jokes by 'just kidding' like a brainless soccer dad".
Based on this manual reading the author was reinforced in their certainty that the joke would work and maybe make someone else's life a little brighter for a half second (500 time thaus, for those using imperial), and decided to take off.
We all know the result. The joke worked, as some people understood it, but not all detected that it was one.
The author would like to apologize for, and retract the "<joke/>" that they added as a comment to their original comment, since adding a revelation that a joke is, indeed, a joke, undercuts the element of doubt and surprise necessary to maximize its effect and duration. The same goes for lengthy, rational discussions about said joke and should be avoided. (It turns out that this particular type of joke can be delivered in their lengthier form, which get ruined if its actual, underlying agenda is revealed in plain text). As such the author would like to also immediately retract this statement and apologize for it.
I think the GP might be yanking our chains a little bit.
Also you kind of have to include the UK in the list of countries that use imperial measures. We're mostly metric but the roads are in miles/yards and if you tried to take pints (which are larger than American pints at 568ml) out of British pubs in favour of the smaller half-litre it would make the party that did it instantly unelectable. People are nearly always measured in feet and sometimes weighed in stone (14 lb) too. I'm a little surprised we don't measure fuel efficiency in miles to the litre because that would be pretty typical of the British approach to measuring things.
I actually don't think it's a bad thing, a bit of mental maths never hurt anyone and it underlines the fact that units of measurement aren't handed down on a stone tablet as a fact of nature.
> There are 3 countries in the world not using the metric system.
The US uses metric for many, many things. Frequently side-by-side with imperial equivalents. AFAIK it's also the official position of the USG that metric is prefered.
> The non-metric system is difficult to use because it's using 12 based units with a 10 based numbering system.
OTOH, 12 divides evenly by both 4 and 3, which are common divisions.
I've heard that some countries half converted and it's been a disaster for them. They're using grams and meters but are still clinging to on the old base-12 time-keeping system. I hope the US doesn't stumble down the same path.
There are procedures, but they are to general and to emergency-manifold to help. Its basically aviate, communicate, crew-resource management applies kind of rules.
In this case, discover how the plane flies, glide it towards a landable location, while communicating your intentions and location. Do not get panicy, do not get burdened with to many things (the pilot monitoring not telling about people on the runway counts as such).
It makes no sense to capture all the permutations of that.
Just get it down somewhat safe.
They forward-slipped a Boeing 767 to bleed off the remaining altitude! A forward slip is a cross-controlled maneuver that feels unnerving and wrong enough in a Cessna 172, let alone a Boeing 767. And they did it with minimal hydraulic assistance as well...
I believe TACA 110 did the same thing to lose energy on an unpowered approach. It's not normally done because the turbulence induced can cause one of the engines to flame out. This is not a problem when the engines are already not operating, of course.
For the non-pilots here, roughly what is a forward-slip? I understand a little bit about flying, but not enough to grok the explanations/diagrams.
Is it a high-altitude, low-power approach to the runway, where full flaps are used to bleed off more altitude and speed? How is that different from a regular approach?
The ideal approach to landing is what's called a "stabilized approach", in which the aircraft descends to the runway at a constant 3 degree angle and constant descent rate. If the pilot finds themselves with too much altitude too close to the runway, in other words above that glide path, they need to more aggressively descend to meet the glidepath.
One way to aggressively descend would involve just pointing the nose down, but this has the effect of increasing speed (trading altitude for airspeed). You need a way to descend more quickly without increasing speed. Generally, this is done by adding more flaps (increasing drag), but in the case of the Gimli Glider, their hydraulic systems were down, so they could not add more flaps.
A forward slip is a maneuver in which you roll the aircraft such that the top of the wing and side of the fuselage are exposed to the oncoming air (relative wind), and you use the rudder in the opposite direction to keep the aircraft flying straight (with respect to the ground track). When properly executed, the rudder and aileron cancel each other out so you keep flying straight across the ground, but the nose of your airplane is not aligned with the direction of travel, and the top and side of your airplane is exposed to the oncoming wind, significantly increasing drag, and thus descent rate.
You Tokyo Drift the plane, exposing more of the side of the plane to the forward motion of the plane to increase the drag by using the side of the plane as a air-brake.
Instead of the most areodynamic way of flying straight, and the worst aerodynamic way of flying completely sideways, you power-slide the plane so that it's not dead-straight and the extra wind-resistance slows you down.
One of the pilots had sail-plane experience from riding a one-man ice-boat with a sail attached over an icy lake - which is where he learned the maneuver. You can slow your speed by the same trick, it's just not usually done with an airliner.
It is where you fly the aircraft sideways through the air.
It is a way of dealing with the situation where the wind is not coming straight down the runway while landing a light power aircraft. You point the aircraft straight down the runway and add in enough bank to overcome the drift to the side. You use the rudders to prevent the aircraft from turning due to the bank.
Pearson was performing the forward-slip in the less common context of gliding. While flying gliders the problem is that they produce a glide angle that is much too shallow to ever be able to control exactly where you are going to land. Normally you use spoilers (AKA dive brakes) to control glide angle but the spoilers might fail or you might need a steeper approach angle than the spoilers can provide. So you fly sideways to force the side of the glider through the air and produce more drag. You still need to take into account any crosswind while simultaneously adjusting the glide angle while staying lined up with the runway. It's a bit tricky and it takes some practice to maintain the skill. That's why I get to demonstrate a forward-slip as part of my yearly glider check flights.
Pearson did this in an airliner with constantly changing control forces and completely nailed it on his first try. He touched down something like 100 feet past the threshold which is exactly what you want to do in a case where you don't have a lot of braking available. This was one of those times where you wanted the guy that lives aviation and not the nine to fiver.
I think it's fair to describe a forward slip as a maneuver that puts the airplane sideways to use the drag of the fuselage, which is now partially side-on to the airstream, to bleed off speed (energy, really). It's used in situations where you need to bleed altitude quickly without gaining airspeed, which would typically be the outcome of just pointing the nose down—you trade gravitational potential energy for kinetic energy.
I'm also not a pilot but, based on a diagram below, it looks like you roll the plane and apply full rudder in the opposite direction. The opposing roll and rudder keep the plane tracking straight but the control surfaces fighting against each other add a bunch of drag to lose airspeed.
Essentially you fly slightly sideways towards the target landing spot on the ground, which loses speed and/or altitude quicker than you otherwise would.
This is done by applying rudder and ailerons in opposite directions (e.g. full right rudder and enough left aileron to still be heading towards the intended spot).
Typically this would be a bad thing as it feels weird and is inefficient, but in this case the whole point is to be inefficient in order to lose altitude without gaining too much speed.
> Both forward and sideslips involve opposite aileron and rudder input. In a forward slip (typically used to increase drag and decrease altitude), the aircraft’s nose points away from the direction of flight, with the lowered wing facing the direction of flight. In a sideslip (typically associated with the final stage of a crosswind landing), the aircraft’s nose points in the direction of flight, regardless of which wing is lowered.
and
> To understand the difference between a forward and a sideslip, we need to understand the slip in general. A slip, at its core, is an uncoordinated turn. In other words, when the ball of the turn coordinator is not in the middle, you’re in a slip.
So in general, a slip is a situation where you're banking, but not turning, due to opposite inputs to ailerons and rudder (i.e. between stick and pedals).
As I read through all that, I thought I was coming to a decent understanding. However, finally,
> The jargon is unnecessarily confusing – you move forward during a forward and a sideslip, but you don’t move sidewards during a sideslip; you technically move sidewards during a forward slip!?
Flying door handle first, with Déja Vu roaring in the background. (I might be exaggerating a little. Also not a pilot.) It looks like you have to use the rudder and bank slightly so that the plane is still flying "forward" but the nose points away from that direction.
An extremely simplistic way to think of a forward slip is for the pilot to have the rudder pedals hard in one direction, stick/wheel hard in the opposite direction.
Generally to turn left you bank left using ailerons. This angles the lift vector to the left and the airplane will turn that way without using the rudder, but it's not fully coordinated - the plane won't be pointing properly into the turn - it will "skid" a little, creating more drag than necesssary. To coordinate the turn, a small amount of left rudder is used to yaw the plane left just enough.
In a forward slip the goal is to descend more quickly than you otherwise could. If you just point the nose down, the airplane will gain airspeed. If you're trying to descend quickly you'll overshoot where you're aiming. What you want is a lot more drag, which allows you to point the nose down (reducing lift) without gaining airspeed. Normally flaps, spoilers or airbrakes allow you to do this (depending on the aircraft). If you don't have these, as the Gimli glider didn't, what can you do to increase drag? Well, you can use the whole fuselage as an airbrake.
To do this, you can use the rudder to yaw the plane, so it's flying a little bit sideways. Say we yaw it to the left. If the wings have dihedral (they point slightly upwards) then the right wing now has a greater angle of attack than the left wing - the right wing produces more lift and so the plane will bank left by itself without touching the ailerons. If the wings are swept back, as on large jets, yawing left will mean the right wing is longer with respect to the airflow than the left wing, and it will bank left even more. To fly pointing left, but not turn left, you cross the controls: you use the rudder to yaw left, and the ailerons to bank right. The airplane is now flying partly sideways, pointing to the left of the flight path, with the right wing lower than the left wing, but if you balance rudder against ailerons correctly, you fly in a straight line. The fuselage is partly sideways to the wind, creating a lot of drag. The airplane is skidding downwards to its right, and you need to pitch the nose down to maintain airspeed and avoid stalling.
And then you've got to be fairly well coordinated as you come out of the slip so you keep flying in the direction of the runway.
Source: I used to fly gliders. Presumably swept wings increase the amount of aileron needed, but the principle is mostly the same.
That's strange. Not saying slipping a 767 is a walk in the park, but I routinely slip Cessna landings (182s and 206s) because I just need to get down. It is a little weird, but no where near unnerving and for sure not wrong.
It’s perfectly normal in a small plane and the center of gravity doesn’t move much because the plane is small. (Technically center of force vs gravity yaddayadda)
Anything a small plane can do a big plane can (very few exceptions), so they can slip just fine, like they can do a 60° bank just fine - but it becomes VERY uncomfortable for the passengers - who now will feel they’re “outside the plane”.
Its not inherently wrong, but flying cross-controlled feels wrong to me. Probably because I don't do it very often, and I'm aware of the fact that it increases my stall-to-spin risk if I don't properly manage airspeed.
The linked article is not bad but reads a bit too much like a Dan Brown novelization for me taste… “ The mustachioed Captain Pearson pulled out the trusty Boeing handbook, his fingers dashing through the pages…”
The Admiral Cloudberg article is much better, particularly in its explanation of how the plane came to be flown with blank fuel gauges. The first technician noticed an issue with the fuel gauges being blank a few flights earlier and applied a fix he happened to know: disabling one of the two channels, which restored the gauges. This was an acceptable fix - the plane was certified to fly with only one of the two channels active, as long as the lost redundancy was regained with manual checks. It was flown twice in this state (working gauges, single channel, manual backup fuel check).
When this captain handed the plane over to the captain of the incident flight, there was a misunderstanding in the conversation that led to the incoming captain believing the prior captain had been flying with blank fuel gauges and only manual fuel checks. In an unfortunate coincidence, the second technician who was checking the plane at this time disabled both channels to try and troubleshoot the issue again, but failed to return the system to the “one channel” functioning state (apparently because he was interrupted by a request to help with the manual fuel check, ironically).
The second technician’s mistake would have been noticed by the pilots, except that due to the misunderstanding, they were expecting to see blank gauges. Those were the initial conditions that allowed the “pounds, not kilograms” mistake to threaten the flight.
Definitely. What makes Admiral Cloudberg so good is the intense focus on details: both details of technical systems, and details of how humans (mis-)communicate with each other. Those details are where the reasons for failure reside, and so they are the place to focus if you’re interested in these stories from an incident analysis perspective.
With flight 143, the sheer brilliance of the flying after the fuel ran out - plus the too-neat story of “it was all caused by metric / imperial conversion” can easily distract from the real lynchpin of the whole incident, which was as you say why the plane was flown with non-functional fuel gauges when that was contrary to the Minimum Equipment List for the aircraft. Taking off with blank fuel gauges was the core safety rule that was violated: the mistake with the unit conversions was necessary to make that violation dangerous but it wasn’t really the core of the accident.
I'd almost say the two make good companion pieces. DI's is a masterclass in nonfiction storytelling, while Cloudberg's is equally adept in its detailed, high-stakes procedural drama.
It's an incredible enough event that it's worth reading both!
I, for one, quite enjoyed the novelization of it. Especially the part about the space-irony continuum. My sleeping wife did not appreciate being woken up by laughter though.
There is one part in that article that I cant understand. It says:
> ...when flight attendants opened the rear emergency exits, they found that the tail was so high in the air that the slides didn’t touch the ground. Sliding down them was less like schoolyard fun and more like jumping off the second or third floor of a building
But then it proceeds showing a photo of deployed rear exit slides, and judging by human standing there they almost touched the ground, the difference is like half that human at most. Now, while I agree that jumping from 1 meter may be too much for some people, comparing it with 3rd floor jump is totally out of place. What am I missing here?
I am guessing that the slides would swing backwards into a more vertical position if someone were to try to use it, because the bottom wasn’t touching the ground and thus preventing the swinging movement. This would then cause the slide to then be an essentially vertical drop instead.
That was a great article. It seems ridiculous to me that the pilots were punished for this given very specifically the unclear delegation of responsibility when automation was introduced. And if ignoring the MEL was a company culture issue, then it was a company culture issue, and you won't make the company safer by encouraging people not to reveal violations.
The thing that gets me in the originally posted link is that the author decided since this was Air Canada, every single person involved must end their sentences with 'eh'.
And to the person who said they routinely slip Cessna landings... different bird my friend ;) Remember, the slower they get the less the hydraulic assist was working. It's a wonder they were even able to straighten out.
I hope there is some kind of 'case study' that pilots study about this.
It should not be up to the pilot to ground the plane; shouldn't that be up to the airline or chief engineer?
>It should not be up to the pilot to ground the plane; shouldn't that be up to the airline or chief engineer?
Aerospace engineer here. I've worked in airworthiness and maintenance ops, though never on the 767. "Grounding an aircraft" is usually a colloquialism, but there are several people who can declare aircraft to be unserviceable and that includes both pilots and maintainers.
For the uninitiated, what happens is the maintenance organization does what's called a Maintenance Release[0] which is the legal act of transferring the aircraft out of maintenance and certifying it is in compliance with what's called the Approved Type Design. Because of the real world being what it is, there are legal ways to accept an aircraft that isn't in compliance with the approved type design and these take the form of Minimum Equipment Lists (MELs) which are basically a list of pre-approved deviations from the Type Design for which risk management has already been carried out. For example in some aircraft you can take off without life rafts, if your aircraft carries a legal operating restriction to stay within $distance of dry land.
Upon accepting an aircraft for a particular flight, the pilots do their own acceptance checks, the common term for which is "preflight".
Anyways, the long and short of it is that yes, the pilots can and should ground the plane if it's U/S. But maintenance can and should as well. In this case, the pilots were sanctioned because they contravened the MEL for the 767 which was a contributing factor to the incident.
Seriously. I had my personal sites on Dreamhost shared hosting back in the 00s and found DI when it was featured in the Dreamhost newsletter as their site of the month or whatever— I wonder if it's still hosted there?
Yes indeed, we are still on Dreamhost. With almost 20 years of hosting sites there, we get some nice discounts from their referral program--otherwise we'd probably change hosts.
"A crew of engineers from Winnipeg airport clambered into a van and headed for Gimli to assess the damage. During transit, however, their vehicle unexpectedly ran out of fuel, nearly ripping a hole in the delicate space-irony continuum."
I didn't know a space-irony continuum exists, but makes sense. ;-)
OP was on the flight with his parents and had the tickets framed. Also the Captain of the flight is now his stepdad and his dad was head of maintenance for Air Canada at the time.
> The internal investigation into the incident laid the blame partially upon Captain Bob Pearson and First Officer Maurice Quintal, who should have observed the Minimum Equipment List (MEL) and grounded the aircraft since it lacked functioning fuel gauges. Some of the responsibility was also assigned to the maintenance workers, and to “corporate deficiencies.” As a consequence Pearson was briefly demoted, and Quintal was suspended for two weeks. Nonetheless both pilots continued to work for Air Canada, and in 1985 they received the well-deserved Fédération Aéronautique Internationale Diploma for Outstanding Airmanship for their handling of the unusual landing.
Interesting. What kind of demotion did Boeing or Air Canada receive?
Air Canada changed their pilot training, changed several manuals and documents, standardizing the fuel weight units (which was already in process), changed the circuit breaker, established a flight safety organization, and changed fueling procedures.
The story does not say if any person responsible for those parts got any negative consequences, thought one can hope that the hunt for blame was short. It usually doesn't serve to improve security, and in this case there was a long list of consecutive mistakes by a large number of people that allowed for the accident to happen. Among those were also a culture of overriding the Minimum Equipment List, something which Canada had outlawed 5 years before this accident.
> presumably because a simultaneous engine failure had been too ridiculous for Boeing engineers to contemplate
This is hard to grasp, esp. given the presence of a RAT. You're giving engineering time to add a device specifically made to handle loss of all engines, but don't spend the time to write the corresponding procedure?
> ...had determined the fuel weight by multiplying the the number of dripsticked liters by 1.77, as indicated by the documentation. However, unbeknownst to the pilots and the fuel crew, this multiplier provided the weight in imperial pounds; the new, all-metric 767 was based on kilograms, and required a multiplier of 0.8
Darn, how many of these incidents will we require until we finally get rid of the metric system once and for all!
> The internal investigation into the incident laid the blame partially upon Captain Bob Pearson and First Officer Maurice Quintal, who should have observed the Minimum Equipment List (MEL) and grounded the aircraft since it lacked functioning fuel gauges
This is an interesting take - and as the pilot of the plan it certainly should rest on you that you're responsible for all to be in working conditions before you leave. However combining that it was apparently a frequent failure, with the corporate culture that's outlined, it's a good cautionary tail about resisting the pressure when something's wrong but "everybody's doing it anyways".
I love you think not other way around.
Converting to metric for fuel seems reasonable but measuring fuel in weight (mass) and volume would still lead to errors using only meteric.
Distance and speeds use a mix of feet, statute miles, nautical miles, meters and kilometers depending on the context and country. Eliminating the difference would mean new approved flight manuals for each airplane, and quite expensive changes to flight instruments as well.
There’s a really good fit with feet for altitude in the U.S. that just works well. 1000’ of separation for opposite direction traffic, and 500’ separation between flight rules in the same direction. Directly converting to meters breaks all the simple math involved. Whereas if we keep the math simple, it alters airspace density and capacity.
Anyway, there’s no significant desire or effort to drop any of these units.
https://news.ycombinator.com/item?id=36863874
I imagine the RAT's primary purpose is to provide the power needed to restart an engine if all engines had stopped.
https://en.wikipedia.org/wiki/Ram_air_turbine
Are you serious?
There are 3 countries in the world not using the metric system. The non-metric system is difficult to use because it's using 12 based units with a 10 based numbering system.
Sure switching over is going to be painful but that's true either way, and there are quite a few examples of people switching to the metric system but none, as far as I know, the other way around.
On jul-25 at 11am pdt, charles_f was standing out to comment on an interesting aviation story. Seeing in passing a reference to issues regarding imperial and metric systems confusion, and given the popularity of this kind of debates on the internet, they decided to set course on a "joke".
The joke was thusly constructed: the metric system is commonly used across the world and much more sensical than outdated imperial system, and people commonly announce their discontent for the latter and ask to discontinue its use. Therefore commenting using the opposite point of view as if it were an evidence, and in a serious tone, had a chance to trigger surprise and interrogation in the reader's mind, quickly resolved by the idea that the author wasn't being serious and making fun of this debate, and triggering a chuckle. This type of jokes is commonly referenced as "sarcasm" in the professional jokers world.
But before the joke took flight, the joke funniness sensor wasn't working, as well as the joke and sarcasm announcement systems. The author sent Robert, their left moustachioed fat finger, to dip in the joky tank, and check if it was fun. Robert answered "yo bro yain't starting to finish all your jokes by 'just kidding' like a brainless soccer dad".
Based on this manual reading the author was reinforced in their certainty that the joke would work and maybe make someone else's life a little brighter for a half second (500 time thaus, for those using imperial), and decided to take off.
We all know the result. The joke worked, as some people understood it, but not all detected that it was one.
The author would like to apologize for, and retract the "<joke/>" that they added as a comment to their original comment, since adding a revelation that a joke is, indeed, a joke, undercuts the element of doubt and surprise necessary to maximize its effect and duration. The same goes for lengthy, rational discussions about said joke and should be avoided. (It turns out that this particular type of joke can be delivered in their lengthier form, which get ruined if its actual, underlying agenda is revealed in plain text). As such the author would like to also immediately retract this statement and apologize for it.
Also you kind of have to include the UK in the list of countries that use imperial measures. We're mostly metric but the roads are in miles/yards and if you tried to take pints (which are larger than American pints at 568ml) out of British pubs in favour of the smaller half-litre it would make the party that did it instantly unelectable. People are nearly always measured in feet and sometimes weighed in stone (14 lb) too. I'm a little surprised we don't measure fuel efficiency in miles to the litre because that would be pretty typical of the British approach to measuring things.
I actually don't think it's a bad thing, a bit of mental maths never hurt anyone and it underlines the fact that units of measurement aren't handed down on a stone tablet as a fact of nature.
The US uses metric for many, many things. Frequently side-by-side with imperial equivalents. AFAIK it's also the official position of the USG that metric is prefered.
> The non-metric system is difficult to use because it's using 12 based units with a 10 based numbering system.
OTOH, 12 divides evenly by both 4 and 3, which are common divisions.
It makes no sense to capture all the permutations of that. Just get it down somewhat safe.
Is it a high-altitude, low-power approach to the runway, where full flaps are used to bleed off more altitude and speed? How is that different from a regular approach?
One way to aggressively descend would involve just pointing the nose down, but this has the effect of increasing speed (trading altitude for airspeed). You need a way to descend more quickly without increasing speed. Generally, this is done by adding more flaps (increasing drag), but in the case of the Gimli Glider, their hydraulic systems were down, so they could not add more flaps.
A forward slip is a maneuver in which you roll the aircraft such that the top of the wing and side of the fuselage are exposed to the oncoming air (relative wind), and you use the rudder in the opposite direction to keep the aircraft flying straight (with respect to the ground track). When properly executed, the rudder and aileron cancel each other out so you keep flying straight across the ground, but the nose of your airplane is not aligned with the direction of travel, and the top and side of your airplane is exposed to the oncoming wind, significantly increasing drag, and thus descent rate.
Instead of the most areodynamic way of flying straight, and the worst aerodynamic way of flying completely sideways, you power-slide the plane so that it's not dead-straight and the extra wind-resistance slows you down.
One of the pilots had sail-plane experience from riding a one-man ice-boat with a sail attached over an icy lake - which is where he learned the maneuver. You can slow your speed by the same trick, it's just not usually done with an airliner.
It is a way of dealing with the situation where the wind is not coming straight down the runway while landing a light power aircraft. You point the aircraft straight down the runway and add in enough bank to overcome the drift to the side. You use the rudders to prevent the aircraft from turning due to the bank.
Pearson was performing the forward-slip in the less common context of gliding. While flying gliders the problem is that they produce a glide angle that is much too shallow to ever be able to control exactly where you are going to land. Normally you use spoilers (AKA dive brakes) to control glide angle but the spoilers might fail or you might need a steeper approach angle than the spoilers can provide. So you fly sideways to force the side of the glider through the air and produce more drag. You still need to take into account any crosswind while simultaneously adjusting the glide angle while staying lined up with the runway. It's a bit tricky and it takes some practice to maintain the skill. That's why I get to demonstrate a forward-slip as part of my yearly glider check flights.
Pearson did this in an airliner with constantly changing control forces and completely nailed it on his first try. He touched down something like 100 feet past the threshold which is exactly what you want to do in a case where you don't have a lot of braking available. This was one of those times where you wanted the guy that lives aviation and not the nine to fiver.
https://www.aopa.org/news-and-media/all-news/2018/july/fligh...
This is done by applying rudder and ailerons in opposite directions (e.g. full right rudder and enough left aileron to still be heading towards the intended spot).
Typically this would be a bad thing as it feels weird and is inefficient, but in this case the whole point is to be inefficient in order to lose altitude without gaining too much speed.
https://www.youtube.com/watch?v=BBqOyDviE-A
Here are a few key selections from https://pilotinstitute.com/forward-slip-vs-sideslip/
> Both forward and sideslips involve opposite aileron and rudder input. In a forward slip (typically used to increase drag and decrease altitude), the aircraft’s nose points away from the direction of flight, with the lowered wing facing the direction of flight. In a sideslip (typically associated with the final stage of a crosswind landing), the aircraft’s nose points in the direction of flight, regardless of which wing is lowered.
and
> To understand the difference between a forward and a sideslip, we need to understand the slip in general. A slip, at its core, is an uncoordinated turn. In other words, when the ball of the turn coordinator is not in the middle, you’re in a slip.
So in general, a slip is a situation where you're banking, but not turning, due to opposite inputs to ailerons and rudder (i.e. between stick and pedals).
As I read through all that, I thought I was coming to a decent understanding. However, finally,
> The jargon is unnecessarily confusing – you move forward during a forward and a sideslip, but you don’t move sidewards during a sideslip; you technically move sidewards during a forward slip!?
Hmn.
https://youtu.be/lnXrXp_7tyE?t=103
In a forward slip the goal is to descend more quickly than you otherwise could. If you just point the nose down, the airplane will gain airspeed. If you're trying to descend quickly you'll overshoot where you're aiming. What you want is a lot more drag, which allows you to point the nose down (reducing lift) without gaining airspeed. Normally flaps, spoilers or airbrakes allow you to do this (depending on the aircraft). If you don't have these, as the Gimli glider didn't, what can you do to increase drag? Well, you can use the whole fuselage as an airbrake.
To do this, you can use the rudder to yaw the plane, so it's flying a little bit sideways. Say we yaw it to the left. If the wings have dihedral (they point slightly upwards) then the right wing now has a greater angle of attack than the left wing - the right wing produces more lift and so the plane will bank left by itself without touching the ailerons. If the wings are swept back, as on large jets, yawing left will mean the right wing is longer with respect to the airflow than the left wing, and it will bank left even more. To fly pointing left, but not turn left, you cross the controls: you use the rudder to yaw left, and the ailerons to bank right. The airplane is now flying partly sideways, pointing to the left of the flight path, with the right wing lower than the left wing, but if you balance rudder against ailerons correctly, you fly in a straight line. The fuselage is partly sideways to the wind, creating a lot of drag. The airplane is skidding downwards to its right, and you need to pitch the nose down to maintain airspeed and avoid stalling.
And then you've got to be fairly well coordinated as you come out of the slip so you keep flying in the direction of the runway.
Source: I used to fly gliders. Presumably swept wings increase the amount of aileron needed, but the principle is mostly the same.
Anything a small plane can do a big plane can (very few exceptions), so they can slip just fine, like they can do a 60° bank just fine - but it becomes VERY uncomfortable for the passengers - who now will feel they’re “outside the plane”.
Have you actually flown a forward-slip?
It might feel unnerving on the first attempt but once mastered, it becomes a normal procedure one can rely on as a pilot.
Some planes require it for landing, e.g. aerobatic airplanes with a high angle of attack - otherwise the pilot would not be able to see the runway.
Yes, I'm a Private Pilot. I've done them dozens of time. Still feels weird to be flying sideways and cross controlled.
The linked article is not bad but reads a bit too much like a Dan Brown novelization for me taste… “ The mustachioed Captain Pearson pulled out the trusty Boeing handbook, his fingers dashing through the pages…”
When this captain handed the plane over to the captain of the incident flight, there was a misunderstanding in the conversation that led to the incoming captain believing the prior captain had been flying with blank fuel gauges and only manual fuel checks. In an unfortunate coincidence, the second technician who was checking the plane at this time disabled both channels to try and troubleshoot the issue again, but failed to return the system to the “one channel” functioning state (apparently because he was interrupted by a request to help with the manual fuel check, ironically).
The second technician’s mistake would have been noticed by the pilots, except that due to the misunderstanding, they were expecting to see blank gauges. Those were the initial conditions that allowed the “pounds, not kilograms” mistake to threaten the flight.
With flight 143, the sheer brilliance of the flying after the fuel ran out - plus the too-neat story of “it was all caused by metric / imperial conversion” can easily distract from the real lynchpin of the whole incident, which was as you say why the plane was flown with non-functional fuel gauges when that was contrary to the Minimum Equipment List for the aircraft. Taking off with blank fuel gauges was the core safety rule that was violated: the mistake with the unit conversions was necessary to make that violation dangerous but it wasn’t really the core of the accident.
It's an incredible enough event that it's worth reading both!
> ...when flight attendants opened the rear emergency exits, they found that the tail was so high in the air that the slides didn’t touch the ground. Sliding down them was less like schoolyard fun and more like jumping off the second or third floor of a building
But then it proceeds showing a photo of deployed rear exit slides, and judging by human standing there they almost touched the ground, the difference is like half that human at most. Now, while I agree that jumping from 1 meter may be too much for some people, comparing it with 3rd floor jump is totally out of place. What am I missing here?
Dead Comment
And to the person who said they routinely slip Cessna landings... different bird my friend ;) Remember, the slower they get the less the hydraulic assist was working. It's a wonder they were even able to straighten out.
Here's a video on the forward slip: https://www.youtube.com/watch?v=yxy2MnUnfUM
I hope there is some kind of 'case study' that pilots study about this. It should not be up to the pilot to ground the plane; shouldn't that be up to the airline or chief engineer?
Aerospace engineer here. I've worked in airworthiness and maintenance ops, though never on the 767. "Grounding an aircraft" is usually a colloquialism, but there are several people who can declare aircraft to be unserviceable and that includes both pilots and maintainers.
For the uninitiated, what happens is the maintenance organization does what's called a Maintenance Release[0] which is the legal act of transferring the aircraft out of maintenance and certifying it is in compliance with what's called the Approved Type Design. Because of the real world being what it is, there are legal ways to accept an aircraft that isn't in compliance with the approved type design and these take the form of Minimum Equipment Lists (MELs) which are basically a list of pre-approved deviations from the Type Design for which risk management has already been carried out. For example in some aircraft you can take off without life rafts, if your aircraft carries a legal operating restriction to stay within $distance of dry land.
Upon accepting an aircraft for a particular flight, the pilots do their own acceptance checks, the common term for which is "preflight".
Anyways, the long and short of it is that yes, the pilots can and should ground the plane if it's U/S. But maintenance can and should as well. In this case, the pilots were sanctioned because they contravened the MEL for the 767 which was a contributing factor to the incident.
[0] https://www.ecfr.gov/current/title-14/chapter-I/subchapter-G...
It's not up to the pilot; it's up to gravity. No fuel left means the plane is going down somewhere soon.
I don’t think anybody reading this story could have possibly assumed the plane could perpetually fly without fuel…
The website isn't getting much donation https://www.damninteresting.com/damnload/
Dedication.
EDIT: Lol yeah, it was in September 2005— the Dreamhost side is long gone I guess, but DI itself remembers: https://www.damninteresting.com/dreamhost-site-of-the-month/
And yes, looks like based on the donations page (https://www.damninteresting.com/damnload/) it is indeed still hosted there.
I didn't know a space-irony continuum exists, but makes sense. ;-)
https://old.reddit.com/r/aviation/comments/15765zq/40_years_...
OP was on the flight with his parents and had the tickets framed. Also the Captain of the flight is now his stepdad and his dad was head of maintenance for Air Canada at the time.
Interesting. What kind of demotion did Boeing or Air Canada receive?
Air Canada changed their pilot training, changed several manuals and documents, standardizing the fuel weight units (which was already in process), changed the circuit breaker, established a flight safety organization, and changed fueling procedures.
The story does not say if any person responsible for those parts got any negative consequences, thought one can hope that the hunt for blame was short. It usually doesn't serve to improve security, and in this case there was a long list of consecutive mistakes by a large number of people that allowed for the accident to happen. Among those were also a culture of overriding the Minimum Equipment List, something which Canada had outlawed 5 years before this accident.