Jump to content


Could the 737 and 777 crashes have be caused by reprogramming from the ground?


62 replies to this topic

#1 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 13 March 2019 - 09:30 PM

This morning a 777 cargo pilot, a friend of Mrs., told her he thought the 737 crash and the earlier MH370 disappearance could have been caused by reprogramming from the ground. That does not make it authoritative of course, just the view of one pilot of 290,000 commercial pilots in the World, but I looked it up, and he could be right. Below is an article by TheGuardian.
This pilot changed from flying passenger to cargo jets to make more money (that is interesting), so he has flown a range of different planes and in his view Airbus software is the Biz, but Boeing excels in equipment quality. He reckons 50% of all pilots just push the auto button as soon as they can.

https://guardianlv.c...rom-the-ground/

"Evidence that the aircraft could have been controlled remotely are self-evident. The United States has deployed drone aircraft all over the world to perform both surveillance and fire missions controlled by pilots half a world away from the targets being surveilled and attacked by their drones. Drones are designed from the ground up to be controlled remotely and so, it turns out, is the Boeing 777, which is equipped with an emergency intervention system that would allow a remote operator to land the aircraft from the ground simply by manipulating the autopilot."

Would SPECTRE be willing to make a few billion by shorting Boeing stock and diving their planes into the ground? If not SPECTRE maybe SMERSH, it has got to be SMERSH
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#2 baw1064

    formerly of the public sector

  • Members
  • PipPipPip
  • 5111 posts
  • LocationEarthquakes, tsunamis, and volcanos--oh my!

Posted 14 March 2019 - 12:43 AM

Would the flight data recorder be able to tell if a malicious command were triggered remotely? Or would this, if true, be another instance of hackers covering their tracks?
“Unless someone like you cares a whole awful lot, Nothing is going to get better. It's not.” --Dr. Seuss

#3 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 14 March 2019 - 06:10 AM

According to one source the pilot of the downed plane only had 200 hours experience on this plane. Because it is still a 737 in principle, a short conversion course is all that is required and this is appealing to developing countries because this saves a lot of money. A full course costs $500,000, although this may be the cost to train a pilot from scratch.

To accommodate the much larger engines, the MAX 8 front wheel had to be raised 9.5" and the 849 lbs heavier engines also had to be pushed forward, changing the flight characteristics and low speed stability, hence the attention in the new software to stall prevention named the 'Maneuvering Characteristics Augmentation System', MCAS.

This tallies with an Aviation Week report, that the MCAS software was installed to mitigate the poor low speed handling of the new MAX 8 and MAX 9.

Government supervisory authorities have a long history of being friendly to plane manufacturers. Michel Asseline, the French A320 test pilot who crashed into a wood, on a display flight - with passengers on board, insisted the software prevented him from applying power on a low pass by. Many people, including myself, believed him. But Europe wanted to protect their investment, even to the extent of tampering withe black boxes - so it is claimed. Michel's career was ruined. https://www.smithson...-cover-up/17336
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#4 baw1064

    formerly of the public sector

  • Members
  • PipPipPip
  • 5111 posts
  • LocationEarthquakes, tsunamis, and volcanos--oh my!

Posted 14 March 2019 - 12:47 PM

You're changing the focus here. The original thesis was someone on the ground hacking the autopilot and the crash being due to sabotage. The plane handling significantly differently than its predecessors and therefore the issue being insufficient pilot training is a completely different topic!

I'm not sure how the Airbus anecdote applies here. Whether or not it actually happened, I can appreciate the possible motive. But in this case, many of the passengers were European, and I imagine their investigators will be playing a role in the inquiry. Why would they want to falsify flight data to help Boeing?
“Unless someone like you cares a whole awful lot, Nothing is going to get better. It's not.” --Dr. Seuss

#5 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 14 March 2019 - 10:01 PM

View Postbaw1064, on 14 March 2019 - 12:47 PM, said:

You're changing the focus here. The original thesis was someone on the ground hacking the autopilot and the crash being due to sabotage. The plane handling significantly differently than its predecessors and therefore the issue being insufficient pilot training is a completely different topic!

I'm not sure how the Airbus anecdote applies here. Whether or not it actually happened, I can appreciate the possible motive. But in this case, many of the passengers were European, and I imagine their investigators will be playing a role in the inquiry. Why would they want to falsify flight data to help Boeing?
Yes there is a bit of thread drift. Threads do drift. I never said European investigators would falsify data to help Boeing! The Airbus anecdote details how the EU were under similar financial pressure when the A320 went down. There was huge incentive to clear the plane of any fault. In this case it was easy to pin all the blame on the pilot but it is widely believed he did nothing wrong. The plane was stable, so it was down to that version of software or the pilot.

With the 737, after two similar crashes, there is now little margin to do anything except to fess up the plane was at fault. This may be corrected with revised software.

But we need to be careful. The FAA has a long history of being cozy to plane manufacturers. When there are tens of billions of dollars at risk I believe regulatory and government organisations (everywhere) are put under extreme pressure. The 737 MAX has 5000 planes on order at a cost of 200m each making a total of one trillion dollars business so far.

The solution for Boeing could involve costly training of all pilots from developing countries. Two hundred hours flying training from scratch and relying on sophisticated automation is not working out so well. Boeing's dividend it going to take a hit, so what, shrug it off. Just do the right thing.
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#6 LFC

    Fiscal Conservative

  • Members
  • PipPipPip
  • 31119 posts
  • LocationPennsylvania

Posted 15 March 2019 - 08:44 AM

View PostGeorge Rowell, on 14 March 2019 - 10:01 PM, said:

Yes there is a bit of thread drift. Threads do drift.

Not here at TRS. We ALWAYS stay on track. [insert time for TRS reader to think "WTF is LFC smoking now? Old tires?"]
" 'Individual conscience' means that women only get contraceptives if their employers, their physicians, their pharmacists, their husbands and/or fathers, pastors, and possibly their mayors, Governors, State Secretaries of Health, Congressmen, Senators, and President all agree that in that particular case they're justifiable." --D.C. Sessions

"That's the problem with being implacable foes - no one has any incentive to treat you as anything more than an obstacle to be overcome."

"The 'Road to Serfdom' is really all right turns." --Progressive Whisperer

""The GOP ... where every accusation is also a confession." --Progressive Whisperer

#7 HockeyDon

    Mind blown...

  • Members
  • PipPipPip
  • 3745 posts
  • LocationNew Britain, CT

Posted 15 March 2019 - 09:16 AM

Based on your participation in the "Legalize weed" thread, I'm sure it's safe to assume.
Well, fuck.

How can I be expected to distinguish BS from reality when so much of my reality is utter BS?!

"There seems to be a lot of people dying of ignorance while living in the information age." my sister-in-law.

#8 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 15 March 2019 - 09:42 AM

View PostLFC, on 15 March 2019 - 08:44 AM, said:

Not here at TRS. We ALWAYS stay on track. [insert time for TRS reader to think "WTF is LFC smoking now? Old tires?"]
Thanks LFC, I was freaking out :). I never made consistency to small details my hobgoblin anyway. You get loved, hated or misunderstood but never bored.
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#9 LFC

    Fiscal Conservative

  • Members
  • PipPipPip
  • 31119 posts
  • LocationPennsylvania

Posted 15 March 2019 - 10:12 AM

View PostGeorge Rowell, on 15 March 2019 - 09:42 AM, said:

Thanks LFC, I was freaking out :). I never made consistency to small details my hobgoblin anyway. You get loved, hated or misunderstood but never bored.

We've had thread drift here that makes this guy look like he was bobbing around in his backyard pool.

Posted Image
" 'Individual conscience' means that women only get contraceptives if their employers, their physicians, their pharmacists, their husbands and/or fathers, pastors, and possibly their mayors, Governors, State Secretaries of Health, Congressmen, Senators, and President all agree that in that particular case they're justifiable." --D.C. Sessions

"That's the problem with being implacable foes - no one has any incentive to treat you as anything more than an obstacle to be overcome."

"The 'Road to Serfdom' is really all right turns." --Progressive Whisperer

""The GOP ... where every accusation is also a confession." --Progressive Whisperer

#10 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 15 March 2019 - 10:58 AM

View PostLFC, on 15 March 2019 - 10:12 AM, said:

We've had thread drift here that makes this guy look like he was bobbing around in his backyard pool.

Posted Image
LFC. That is simply GREAT!!!
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#11 LFC

    Fiscal Conservative

  • Members
  • PipPipPip
  • 31119 posts
  • LocationPennsylvania

Posted 15 March 2019 - 11:37 AM

View PostHockeyDon, on 15 March 2019 - 09:16 AM, said:

Based on your participation in the "Legalize weed" thread, I'm sure it's safe to assume.

Posted Image
" 'Individual conscience' means that women only get contraceptives if their employers, their physicians, their pharmacists, their husbands and/or fathers, pastors, and possibly their mayors, Governors, State Secretaries of Health, Congressmen, Senators, and President all agree that in that particular case they're justifiable." --D.C. Sessions

"That's the problem with being implacable foes - no one has any incentive to treat you as anything more than an obstacle to be overcome."

"The 'Road to Serfdom' is really all right turns." --Progressive Whisperer

""The GOP ... where every accusation is also a confession." --Progressive Whisperer

#12 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 15 March 2019 - 09:56 PM

In the crashed jet the stabilizer screw controlled by the software was hard to the end. The surface of the stabilizer is much larger than that controlled by the stick. The natural reaction to a stabilizer change would be to correct the trim using the stick. Eventually the stabilizer will win because of it's larger control area.

The stabilizer action might be relatively slow, I checked out one that took 40 seconds to go from one extreme to the other. The unsuspecting pilot will slowly apply more and more displacement until the stabilizer eventually won.
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#13 baw1064

    formerly of the public sector

  • Members
  • PipPipPip
  • 5111 posts
  • LocationEarthquakes, tsunamis, and volcanos--oh my!

Posted 15 March 2019 - 11:13 PM

So is this best described as a software issue, a human/machine interface issue, or a training issue?
“Unless someone like you cares a whole awful lot, Nothing is going to get better. It's not.” --Dr. Seuss

#14 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 16 March 2019 - 01:02 AM

View Postbaw1064, on 15 March 2019 - 11:13 PM, said:

So is this best described as a software issue, a human/machine interface issue, or a training issue?
It looks like a bit of everything.

It has been pointed out that when the stability issues started to manifest a new airworthy certificate would normally have been applied for. This is a time consuming and expensive procedure and it did not happen.

The first officer on the MAX 8 had only 200 hundred flying hours. Considering the challenges with the MAX 8 that should have been 1000's of hours. In the states a minimum of 1500 hours is required.

The software is clearly faulty for a whole host of reasons, including.
1) It never alarmed the pilot that it was using the trim to prevent a stall. That is frankly amazing.
2) It apparently did not detect that the pilot was using the stick to change the attack, it just carried on until it met the end stop.
3) it failed in it's primary objective :)

The length of the time required to turn the stabilizer position was slow, unavoidable due to the worm drive, and the hapless pilot was compensating for the changes. He should have realized that the settings were seriously wrong. That is a training and familiarity issue.

Following the previous crash Boeing should have made absolutely sure that all pilots knew about the software changes, that they must have known by then had several issues.

I am sure we will hear a lot about the design compromises that went into the MAX conversion, but the jet is probably still viable. Lots of jets have known issues that pilots need to learn. As for stability, well modern fighter jets are deliberately designed unstable to get maneuverability and are computer stabilized. It is no big deal. However, the public have now lost confidence in the plane and Boeing.

It is vital for Boeing to restore credibility with an education campaign. With all their miss-steps so far that will not be easy, but with one Trillion dollars in orders at stake you can bet they are going frantic from every perspective to sort this out.

My own view is too much money has already been spent on the plane and the benefits (green and otherwise) are too significant to be lost. For the cost of developing another plane to take it's place, by Boeing or Airbus, tens of thousands of lives could be saved in hospitals and improved healthcare. That will dwarf the loss of life in present and future accidents. The public are unlikely to buy into that argument but it is still a relevant factor for government policy makers. The US gov and the FAA will no doubt be having many gut wrenching meeting behind closed doors. For my part I am glad the decisions are up to someone else.

Maybe it is time to start educating the public that it is most safe to have 'pilots' whose main job is to monitor the computers rather than attempt to fly the machine.
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#15 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 17 March 2019 - 02:35 AM

Back to pilot training. US airlines started lowering the requirements to become pilots and crashes started happening.

"Pilots were being taken on with as little as 350 hours of total time, assigned to the first officer’s seat of sophisticated regional jets.”

Then, regional airlines started having crashes, most famously the Colgan Air disaster in Buffalo, New York, in 2009. Forty-nine people were killed due to errors made by the pilot and co-pilot. After that, Congress and the Federal Aviation Administration set up new rules requiring co-pilots to have 1,500 hours of flight time — instead of the previous requirement of 250 hours — and for pilots to have 1,000 hours as a co-pilot.

Just like that, the (accident) pipeline was broken."
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#16 Traveler

    Rambling Member

  • Members
  • PipPipPip
  • 13369 posts
  • LocationPhilly Area

Posted 18 March 2019 - 12:22 PM

This article pretty well nails what happened.

Quote

The safety analysis:
  • Understated the power of the new flight control system, which was designed to swivel the horizontal tail to push the nose of the plane down to avert a stall. When the planes later entered service, MCAS was capable of moving the tail more than four times farther than was stated in the initial safety analysis document.

  • Failed to account for how the system could reset itself each time a pilot responded, thereby missing the potential impact of the system repeatedly pushing the airplane’s nose downward.

  • Assessed a failure of the system as one level below “catastrophic.” But even that “hazardous” danger level should have precluded activation of the system based on input from a single sensor — and yet that’s how it was designed.
[..]
Both Boeing and the FAA were informed of the specifics of this story and were asked for responses 11 days ago, before the second crash of a 737 MAX last Sunday.
That is breathtakingly stupid way to design a system and its interface. SO the more you fight it, the more it fights you. At the end it takes 80kg of stick force to fight the stick. How long would you like to curl 180 lb?

Then let's get into the AOA sensor issue. These things freeze up. There have been many false stall events, that are easily overridden in other aircraft. That is why there are at least on two every plane. Except the Max. So now you have a faulty system based on one faulty sensor. See below:
Posted Image




Boeing and the FAA have a LOT to answer for.

I posted this before I read past the paras cited. It is even worse.

Quote

One current FAA safety engineer said that every time the pilots on the Lion Air flight reset the switches on their control columns to pull the nose back up, MCAS would have kicked in again and “allowed new increments of 2.5 degrees.”



“So once they pushed a couple of times, they were at full stop,” meaning at the full extent of the tail swivel, he said.

Peter Lemme, a former Boeing flight controls engineer who is now an avionics and satellite-communications consultant, said that because MCAS reset each time it was used, “it effectively has unlimited authority.”

Swiveling the horizontal tail, which is technically called the stabilizer, to the end stop gives the airplane’s nose the maximum possible push downward.

“It had full authority to move the stabilizer the full amount,” Lemme said. “There was no need for that. Nobody should have agreed to giving it unlimited authority.”

I was also wrong about the AOA sensor- It is actually worse.

Quote

Like all 737s, the MAX actually has two of the sensors, one on each side of the fuselage near the cockpit. But the MCAS was designed to take a reading from only one of them.
This is because they classified failure as merely hazardous.

This article should be required reading for all reporting on this.
"Men occasionally stumble over the truth, but most of them pick themselves up and hurry off as if nothing had happened."-- Winston Churchill
"Anyone who has the power to make you believe absurdities has the power to make you commit injustices" Voltaire

#17 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 18 March 2019 - 10:51 PM

View PostTraveler, on 18 March 2019 - 12:22 PM, said:

This article pretty well nails what happened.

That is breathtakingly stupid way to design a system and its interface. SO the more you fight it, the more it fights you. At the end it takes 80kg of stick force to fight the stick. How long would you like to curl 180 lb?

Then let's get into the AOA sensor issue. These things freeze up. There have been many false stall events, that are easily overridden in other aircraft. That is why there are at least on two every plane. Except the Max. So now you have a faulty system based on one faulty sensor. See below:
Posted Image




Boeing and the FAA have a LOT to answer for.

I posted this before I read past the paras cited. It is even worse.


I was also wrong about the AOA sensor- It is actually worse.

This is because they classified failure as merely hazardous.

This article should be required reading for all reporting on this.
Indeed, someone has a lot to answer for. I suspect a chain of groups.

YOU GET WHAT YOU PAY FOR!

Software engineering does not teach you coding, that you pretty much teach yourself. The emphasis is on getting it right. Specifications, design, implementation and testing. It is all about methodology. What I find so appalling with these reports is the total unsuitability of the software for the system, it is as if they gave the job to an apprentice. The problems I see are:

1) There were two AOA sensors but the computer could only sample one at a time. OK, the AOA sensor rotates and it measures the difference between a calibrated reference position and the actual new angle of attack. That may involve waiting or it may not, probably not, but are we saying that with such a key sensors the computer has to switch from one channel to another? Why not monitor both?

2) If one AOA sensor showed a fault condition surely the software checked the other AOA sensor. This implies that either a SINGLE AOA sensor was allowed to cause an emergency exception or BOTH AOA sensors went wrong at the same time. That is highly unlikely.

3) The action of the modifying the stabilizer was started without issuing an audible warning. It just went into emergency mode behind the pilots back, secretly. That is preposterous!

4) The stabilizer was pushed to the limit, reportedly 4x more than the operation required. Beyond the point when the pilot could regain control. That is another clear error.

5) The software ignored the pilots actions, it fought the pilot and just kept applying more and more stabilizer. Was there no check to see what the pilot was doing, no look at the stick displacement?

To me this looks like first cut 'proof of concept' software, end of story. It does not look like a single part of the AOA stall exception routine has been checked or signed off.


The flight control software was almost certainly outsourced. That is OK, but I wonder if the highly competitive Third World programming community set the same standards we do or even have the the same simulation facilities. One particular aspect is regression testing. Every software change no matter how small should be followed by regression testing. In essence chnaging innocent looking code should be followed by a test of the complete code, and test of all the exceptions and subroutines. That means simulating all flight inputs in all combinations. That can take days, weeks, months or even years. It is the methodology and record making that is the expensive part, not the code. You get what you pay for.
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#18 baw1064

    formerly of the public sector

  • Members
  • PipPipPip
  • 5111 posts
  • LocationEarthquakes, tsunamis, and volcanos--oh my!

Posted 19 March 2019 - 12:03 AM

Apparently, Airbus has had the same issue, In 2014, an A321 got faulty data on its angle of attack and kept trying to lower the nose and fighting the pilots (sound familiar?), losing 4000 feet of altitude before the pilots had the good sense to disconnect the safety system and then were able to restore control.

https://avherald.com...rticle=47d74074

A sanity check (wait, something doesn't make sense here!) is something humans are pretty good at, if they've had prior experience. But it's pretty hard to do a good job programming this kind of judgement into a computer.
“Unless someone like you cares a whole awful lot, Nothing is going to get better. It's not.” --Dr. Seuss

#19 George Rowell

    Advanced Member

  • Members
  • PipPipPip
  • 840 posts
  • LocationIn Perth now

Posted 19 March 2019 - 12:50 AM

View Postbaw1064, on 19 March 2019 - 12:03 AM, said:

Apparently, Airbus has had the same issue, In 2014, an A321 got faulty data on its angle of attack and kept trying to lower the nose and fighting the pilots (sound familiar?), losing 4000 feet of altitude before the pilots had the good sense to disconnect the safety system and then were able to restore control.

https://avherald.com...rticle=47d74074

A sanity check (wait, something doesn't make sense here!) is something humans are pretty good at, if they've had prior experience. But it's pretty hard to do a good job programming this kind of judgement into a computer.
I checked that out - yes sounds familiar. It does make you wonder how they do regression testing. if you have 1000 inputs requiring 100mS stabilization and you need to check in all orders then that is 1000 factorial x 1/10 seconds. That is clearly a no-go.

Re - training, like I pointed out earlier, when US airlines dropped the requirement to 350 hrs flight experience a spate of accidents happened. Those accidents stopped quickly when the FAA said make that 1500 hours.

I have just read that the same thing happened the day before but the pilot turned off the MCAS. It just gets more and more amazing. You have got to ask if the crew left that incident in the log and if so why the new crew did not read it and be prepared.
A doctor knows a little about a lot. A specialist knows a lot about a little. In time the doctor knows less and less about more and more and the specialist knows more and more about less and less until ultimately the doctor knows nothing about everything and the specialist knows everything about nothing.

#20 HockeyDon

    Mind blown...

  • Members
  • PipPipPip
  • 3745 posts
  • LocationNew Britain, CT

Posted 19 March 2019 - 07:31 AM

View PostGeorge Rowell, on 18 March 2019 - 10:51 PM, said:

Software engineering does not teach you coding, that you pretty much teach yourself.

Add to that, neither teaches real-world application/use. I deal with that all the time with both our ERP/MRP and our CAD (computer aided design) software.
Well, fuck.

How can I be expected to distinguish BS from reality when so much of my reality is utter BS?!

"There seems to be a lot of people dying of ignorance while living in the information age." my sister-in-law.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users