Boeing 737 Max Judged Safe To Fly By Europe's Aviation Regulator (bloomberg.com) 70
schwit1 shares a report from Bloomberg: Europe's top aviation regulator said he's satisfied that changes to Boeing Co.'s 737 Max have made the plane safe enough to return to the region's skies before 2020 is out, even as a further upgrade his agency demanded won't be ready for up to two years. After test flights conducted in September, EASA is performing final document reviews ahead of a draft airworthiness directive it expects to issue next month, said Patrick Ky, executive director of the European Union Aviation Safety Agency. That will be followed by four weeks of public comment, while the development of a so-called synthetic sensor to add redundancy will take 20 to 24 months, he said. The software-based solution will be required on the larger 737 Max 10 variant before its debut targeted for 2022, and retrofitted onto other versions.
Much like the Corona vaccine (Score:5, Insightful)
Re: (Score:2)
I'll stick to the "older" 737 models until they get those sensors finished, thankfully they fly that anywhere I might go and they're the safest and most comfortable thing in commercial use.
Re: (Score:2)
You mean the ones that were the subject of an emergency air worthiness directive from the FAA recently, because their engines keep failing in flight?
Re:Much like the Corona vaccine (Score:4, Interesting)
I'll stick to the "older" 737 models until they get those sensors finished, thankfully they fly that anywhere I might go and they're the safest and most comfortable thing in commercial use.
What sensor are you talking about? There will be NO third sensor. What they will do is synthesize a third virtual sensor using other information they have like airspeed, altitude, etc, and a bit of math.
Re: (Score:2)
So they're going to fix the 737 deathtrap with software? Running on a pair of fossilized 286s, is that it? Please excuse me if I don't jump at the chance to fly in that.
Re: (Score:3)
The really are using 286s and I doubt they have sufficient compute power to do much of anything, let alone anything sophisticated. Plus, they kind of code that needs to be written requires a high degree of skill, care and peer review. But they have demonstrated they can't even get a stall sensor working properly.
Re: (Score:2)
The really are using 286s and I doubt they have sufficient compute power to do much of anything, let alone anything sophisticated. Plus, they kind of code that needs to be written requires a high degree of skill, care and peer review. But they have demonstrated they can't even get a stall sensor working properly.
If you cannot get a 286 to run a non-graphical application looking at the available air data and inertial navigation inputs to approximate a stall warning sensor, you have a serious problem with your programming skills and really don't know what you are doing. A 286, even a slow one, has a LOT of compute power, more enough to do this new task a couple times over without breaking a sweat. Rockwell Collins knows what they are doing here.
Re:Much like the Corona vaccine (Score:4, Interesting)
So they're going to fix the 737 deathtrap with software? Running on a pair of fossilized 286s, is that it? Please excuse me if I don't jump at the chance to fly in that.
You do understand that this computer is the same one flown on many generations of the 737 before the MAX and has shown to be highly reliable design, a characteristic which is highly prised in the industry.
Also understand that a 286, even a slow one, is a LOT of computing power if you are using it correctly (and not doing stupid stuff like running Java JVM's or heavyweight things like windows on it). If you cannot get your real-time computing job done in a 286, you have some serious problem on your hands, something bigger than any modern aircraft. Writing code in assembly or just plain C++ on a Linux platform and running it on a 286 is going to be sufficient for most real-time controller applications I can imagine. Just because it would make a poor performer running bloatware like windows and Excell, doesn't mean it is without computing power. Are there faster CPUs on the market? Sure, but how many of them are able to meet the demanding specifications to be used in an aircraft?
Operating at Extended temperatures at low atmospheric pressures are hard to achieve, your desktop would die quickly at 10,000 FT and -40 F, but not this system, it would run just fine on the tarmac in Alaska in the dead of winter where it's -40 sometimes or departing from Bagdad in the heat of summer in the late afternoon where it's been 120F in the shade for days. I dare you to try that with your laptop.. I'll bet it wouldn't boot in either condition, yet this computer on the 737 has to, and more..
Re: (Score:2)
Actually the pilots complain that the 737MAX FMS is very slow so it doesn't have enough performance after all.
Re: (Score:2)
Actually the pilots complain that the 737MAX FMS is very slow so it doesn't have enough performance after all.
Maybe FMS is slow, but that's not the air data computer. The FMS is s a Motorola 68040 Series CPU running at 30 Mhz clock with 4Mb memory and 32Mb flash. The FMS isn't a "real time" system, has nothing to do with the flight controls but is used to manage flight planning, tune the radios, load waypoints and do various performance calculations, like takeoff and landing distances, weight and balance, fuel consumption. Yea, it's a bear to operate, monochrome character cell screen, no touch, no graphics, and sl
Re: (Score:2)
It was also doing less on those earlier models.
A Commodore PET is supremely reliable and robust, capable of uptimes the envy of all but carrier-grade Linux and better quality BSDs. But I wouldn't want anything highly complex driven by it. It's only reliable and robust because that's exactly what nobody would do
Re: (Score:2)
You're living in dreamland. 1) 286 is criminally underpowered for a modern airframe 2) Boeing engineers can't do anything right, let alone program a 286 as efficiently as possible. Face it: 286 is a museum piece, much like the rest of deathtrap 737.
Re: (Score:2)
And you know this how? I've actually worked on avionics systems, have you?
Modern industrial controllers use much less capable processors that the 286 for some pretty compute-intensive problems.
When you program directly in C or C++ (or any other non-interpreted language) you don't need a huge amount of memory or lots of computing power to get the job done. Don't judge the 286 by how it would perform like a desktop, because that isn't what this system is running. It's likely running on a pared-down Linu
Re: (Score:2)
Yes I have worked in avionics systems. You sound fucking clueless, do you work for Boeing? You have no idea what kind of processing is typical for a modern airframe.
Re: (Score:2)
Yes I have worked in avionics systems. You sound fucking clueless, do you work for Boeing? You have no idea what kind of processing is typical for a modern airframe.
LOL.. You DO realize that the Airbus "fly by wire" system uses less CPU power right? My brief investigation turned up that the primary flight control computers are build around the Motorola 68010 microprocessor and the Secondary control computers are Intel 80186 microprocessor (a scaled down 286 CPU used in the Boeing/Rockwell Collins system on the 737). You knew that Right? Right? I didn't think so.
The Airbus configuration is a whole lot more compute intensive, given that it actually FLIES THE PLANE by
Re: (Score:2)
You realize you're quoting from an out of date reference right? Airbus offers multiple options for the A320 FCS, the Honeywell one uses their 29KII RISC processor. You blathered on about subsystems just translate control demand into actuator motion. What do you suppose generates the control demand?
Ya, you sound clueless because you are clueless, and regurgitating the first google results you stumble across doesn't help.
Re: (Score:2)
6 months? that's awfully generous, especially considering that's the equivalent of 10 days of flight time pre-pandemic.
Re: (Score:2)
Six years for me. :P
Of course it's safe (Score:4, Funny)
With the COVID-19 travel restriction, they all stay on the tarmac and threaten nobody there.
Re: (Score:2)
Just slap a big mask on the front and lets fly! Freedum is FREE!
I will not fly in one of those. (Score:3, Interesting)
Re: (Score:3)
That report you link to is entitled “The annual Boeing Statistical Summary of Commercial Jet Airplane Accidents Worldwide Operations 1959-2018” - note the use of “Boeing” in it.
Its not independent, and I can see several issues with the way it groups the aircraft so as to present a better picture.
Re: (Score:2)
note the use of “Boeing” in it.
Noted. So you'd think that with your newly uncovered controversy a certain large competitor would be all over it and a simple search would find a case of Airbus disagreeing other otherwise providing a retort / counter claim on statistics.
Except that doesn't exist, because despite your knee jerk emotional response Boeing's competitor is actually quite fine with the statistics.
Re: (Score:3)
Generally both companies put out this sort of thing and ignore the other party, and in general these reports mean zero to their actual customers, so yeah, it doesnt mean anything.
It speaks louder that the only airline to have three forced groundings of their aircraft by regulatory bodies in the past 35 years has been Boeing (737 Classic, 787 and 737MAX) - of course those sort of things Boeing doesnt include in these reports, conveniently...
Re: (Score:3)
It is a misleading statistic - the 737NG is a decade younger than the A320ceo family and there have been 1000 fewer of them delivered hence not as much opportunity to actually fly. Put it together with the 737 classic for a better comparison and suddenly the Airbus does better regarding safety, which is not surprising given its enbelope protection and manufacturing according to newer safety standards.
Re: (Score:3)
Precisely. You would have to compare the 737 MAX with the A320 NEO and that wouldn't exactly be favorable now would it.
Re: I will not fly in one of those. (Score:1)
Re: (Score:3)
Given the amount of additional scrutiny, testing and oversight that this plane has been through in the last couple years, I'd say it will easily be the safest plane in the air.
Re: I will not fly in one of those. (Score:3)
I don't know if it's the fastest in the air but it's the safest on the ground.
Re: (Score:2)
Only after those two years. It still does not have enough redundancy on the maligned attitude sensor. It now reads from two sensors instead of one but it doesn't have any way to verify which is right in case of failure. So it's less bad but still not good enough. EASA originally wanted them to add a third sensor (like Airbus has) but Boeing claims they can verify which sensor is wrong with estimates from other sensors (i.e. a software patch) like they do in the 787. But it seems you'll have to wait two year
Re:I will not fly in one of those. (Score:4, Informative)
Re: (Score:2)
It's not an "attitude sensor", but an angle-of-attack sensor. And, after the changes, what is the effect if both sensors fail? The nose-lowering force has been greatly reduced, so now it just means that the pilot has to give a little additional back-force on the stick. This isn't dangerous. The problem with the original MAX was that the pilots were not able to overcome the automatic nose-lowering mode.
I understand there is a lot of fear out there, and I'm a nervous flyer myself. But there doesn't seem to be
This is a management problem, not engineering (Score:2)
Re: (Score:2)
The hardware of the plane is still aerodynamically unstable. No amount of software will fix that.
Re: (Score:2)
Yeah no shit. This was this century's equivalent of the Ford Pinto.
Re: (Score:2)
Yeah no shit. This was this century's equivalent of the Ford Pinto.
Now is that fair? I mean, the pinto kind of paled in comparison to some other cars out there is you didn't hit it in the rear at highway speeds. The AMC Pacer was a rattle trap aquarium with maintenance problems out the wazoo too, it just didn't catch fire when in a rear-end collision.
Re: (Score:2)
It is the least I can do to honor the memory victims, and their families, of the murder that Boeing promoted on passengers of those two flights. For now on, I will fly only in companies that use Airbus airplanes.
Your loss.
This aircraft has been gone through with a fine-toothed comb by both Boeing and the FAA (and now at least one foreign authority, possibly more). ALL the pilots will have recently been through some rigorous training on the issues and demonstrated competency in dealing with it in the simulators. It's likey that the 737 MAX is pretty much the safest thing flying once it starts again.
Personally, I'd fly on one of these over an Airbus. Why? Because the Airbus aircraft in this class have "fly by wire"
Re: (Score:2)
Every modern airliner like this, it makes them safer and it is the reason why any 737 cannot be the safest thing flying. Even the last Soviet designed airplanes had digital fly by wire with envelope protection.
Re: (Score:2)
Re: (Score:2)
This briefing [youtube.com] is a bit more honest.
Anyone have a concise summary (Score:4, Informative)
A few problems I remember:
1) Automated mechanism used only one sensor.
2) No warning of inconsistent/malfunctioning sensors unless you paid for the extra "maybe safe" option package.
3) Mechanism kept repeatedly applying multiple times, even if pilot was putting control force to override, and was cycling mechanism on and off.
4) Pilot training and manuals did not mention the new MCAS system at all, nevermind what to do if it fails.
Re:Anyone have a concise summary (Score:4, Insightful)
Of the fixes they've actually implemented so far. A few problems I remember: 1) Automated mechanism used only one sensor. 2) No warning of inconsistent/malfunctioning sensors unless you paid for the extra "maybe safe" option package. 3) Mechanism kept repeatedly applying multiple times, even if pilot was putting control force to override, and was cycling mechanism on and off. 4) Pilot training and manuals did not mention the new MCAS system at all, nevermind what to do if it fails.
Your number 2 item is just obscene. They wanted an additional $80,000 just to put up a "sensors don't agree'" symbol on the existing display panel when things went wrong. No extra equipment or installation required, just switching on the routine that already existed in the software.
Re: (Score:2)
The "AoA disagree" indicator was supposed to be included all the time, but due to a bug it only worked if you paid for the "dual AoA display" feature. They decided they'd just say "it's not a bug, it's a feature" rather than fix the bug and get the software re-certified. The bug wouldn't have been so much of an issue if they weren't using an AoA sensor reading as an input the the MCAS.
Re: (Score:3)
I thought most of the problem was lack of pilot training, not bad engineering. Boeing tried to sell the 737 MAX on the basis that no additional pilot training was required, for pilots experienced with earlier 737 variants. MCAS was added to make it appear that the 737 MAX would fly just the same as earlier models, which it probably would, until something goes wrong.
As far as I know, the ultimate reason for two 737 MAX crashes was that the pilots were not told about MCAS, and were not told about a potential
Re: (Score:2)
The last crash was after the pilot had been told and had followed recommended workarounds.
Re: (Score:2)
The last crash was after the pilot had been told and had followed recommended workarounds.
I did not know that. So why did the workarounds fail?
As far as I know, the 737 MAX never required fly-by-wire control for typical flight conditions, so I would have thought it would be possible to get manual control by turning off malfunctioning automated systems. I thought the danger condition was nose up and close to stall speed. A standard way of avoiding a stall is to increase airspeed, but this caused a stall in the 737 MAX, due to the off-center thrust of the engines. I presume properly trained pilots
Re: (Score:2)
The EU aviation regulator has forced Boeing to add a synthetic AOA sensor in addition to the two the 737 already has. As far as I understand it works by a software combining inputs from different sensors, flaps setting and the inertial navigation system to calculate the approximate angle of attack for plausibility checks.
Re: (Score:2)
Re: (Score:3)
Although the 737 MAX has two MCAS systems the only way to fail over is to land, turn one off, and turn the other on. Not really my idea of a "failover", but apparently some programmers didn't understand the concept.
The "off" switch for the MCAS wasn't really an off switch, it just disconnected the system from the controls. One of the flights had managed to find it (they thought they were turning off the automatic trim, apparently), turned it off, and recovered at a lower altitude. Now when they turned wh
Re: (Score:3)
It's actually WORSE. There was no specific MCAS cut-off, all you could do is turn off the entire electric trim system, including the thumb switches, leaving only the manual trim wheels. Unfortunately, under some conditions (including when the plane is way out of trim and trying to nosedive), the manual trim wheels required the strength of a gorilla to operate. That's why the pilots kept turning electric trim back on. The thumb switches didn't override MCAS, so recovery required turning electric trim on just
Re: (Score:2)
Re: (Score:2)
You misremember 4. Pilot training manuals did cover the MCAS system. The difference was no one read them because the whole purpose of the MCAS system was to avoid putting pilots through training.
The pilot training manual did not mention your point 3 correctly. I.e. it told how to turn the system off, but wasn't clear on the conditions for when it would turn itself back on.
Re: (Score:2)
Turning off the MCAS system was EXACTLY the same as disabling other systems that messed with the trim tabs, so the pilots "should" have been able to recognize and deal with this issue. In fact, 1 out of three (that we know of) did and successfully landed their aircraft.
Now I'm not going to blame the pilots on the two aircraft that didn't make it, they were not properly trained and Boeing has accepted the responsibility for this (as has the FAA for their lack of oversight), but there was the expectation th
Re: (Score:2)
Software software software... We need a choice (Score:2)
Software isn't magic. If your hardware is at fault (airframe design, engine size and position), then your software will grow more complex than it should, and risk tripping over its own feet.
The best software here is what travel sites can add to give people the choice to avoid flying the max 8 or 800 or whatever identifier the faulty airframe bears these days
Re: (Score:2)
In this case, the hardware (the aircraft) was actually fine, just different than previous models. The MCAS system was added because the larger/higher/more forward engines changed the handling characteristics different than the previous generation of 737s. The idea was that htis would eliminate the need to provide additional training to the pilots. Well, we saw how well that worked out.
Re: (Score:2)
Re: (Score:3)
the hardware (the aircraft) was actually fine, just different than previous models.
Very different = faulty. The comment by the 'thegreatbob' below is spot on, and this LA Times article [latimes.com] explains it well. Boeing was too greedy to spend money to redesign the airframe properly.
Well, we saw how well that worked out.
That's the point I'm making: to workaround the buggy hardware (airframe), Boeing implemented buggy software, which tripped over its own feet, killing people.
Re: (Score:2)
The sensor stopped working, with no backup - a Single Point of Failure. That's been a key factor in previous crashes of both Boeing and Airbus aircraft.
A broken sensor is obviously not fine hardware, but an SPF in and f itself is arguably not fine when you know in advance that it can be catastrophic. So that is arguably a distinct failure.
Re: (Score:2)
Europe before the FAA! (Score:4, Interesting)
Give me a reason and a great price (Score:3)
As a nervous flyer ... I'll jump on one tomorrow with no more worry than any other plane I fly. I just have a primitive nervousness about flying anytime the plane shakes and I just accept it. I also understand statistics. Pre grounding I still would have had almost zero chance of a problem just like any other flight. With some fixes and heavy testing, it'll be even better. Where are we going?
Re: (Score:2)
As a non-nervous passenger who actually loves to fly, I'm not getting on one and not letting my children on one. And of course I understand that the regulating agencies who approve the plane know more about it than I do.
The issue with the 737 MAX is that the airframe is aerodynamically unstable. The engines are in the wrong place. This was done because there was no other place to put the larger, more efficient engines. Unless they would have redesigned the landing gear. Which would have entailed redesigning
Re: (Score:2)
I understand your argument, but Boeing has a history of denial, the 737 has a history of issues, and Boeing's computers are linked to an impressive list of accidents.
You're absolutely right that the risk is very, very low. It was already low when the Comet flew. The Comet was still flown until very recently because it's an insanely good, safe, design. Just one with a propensity to explode if the windows are riveted poorly. Still, there was a lower risk of death from it per mile than from car accidents in th
Talk to me when... (Score:3)
They've got 2 years without an accident at the pre-covid-19 flight levels.