Boeing 737 MAX

[Revoltingest]

It sounds like a recipe for a disaster no matter how one looks at it and something technical schools will use as an example of ridiculous overconfidence for decades to come, if true. One of those things a lecturer throws out to new students as an example of what not to do.

There was a post on a pilot's forum regarding the Indonesia crash I ran into where they said triple redundancy is what they'd expect.

Having 3 AOA sensors is one more than was typical in the past.

 

[Jumi]

It isn't well known, but very complex control systems which play tag team with human pilots are extremely difficult to design & optimize. This is why testing is so necessary. When problems arise, the solutions look obvious in hindsight. But what will be the next problem, which has a currently "obvious" solution? We don't know. The current pretty good safety record of air travel is the result of a century of experimentation.

A staggering amount of work goes into things that seem to work without a hitch to someone who doesn't know what goes into it.

I still marvel at how reliable these complicated contraptions are.

Yes, there is very little risk in flying these days. At least using airliners with good track records.

Having 3 AOA sensors is one more than was typical in the past.

But if one failing is enough to cause a disaster it's worse than having just two. Depending if the two other sensors can also cause the same kind of failure, it could be worse than having one.

 

[Revoltingest]

But if one failing is enough to cause a disaster it's worse than having just two. Depending if the two other sensors can also cause the same kind of failure, it could be worse than having one.

That seems unlikely to me.
I favor having more sensors, & having the system vote for the majority.
(Of course, there are great complexities lying behind how to "vote".)
This offers the higher probability of avoiding error.

I believe we've reached a stage where in an emergency, the pilot
will be more of a problem than a solution. This is because people
are slow & unreliable under stress. A capable automatic system
can act with greater predictability, reliability & speed.

 

[Bob the Unbeliever]

I still marvel at how reliable these complicated contraptions are.

I am still amazed that the US was able to get men from the earth, to the moon, and back, using computers dumber than an Atari 800... and electronics so primitive that Large Scale Integration was too extreme to trust fully--- better stick to discrete transistors and whatnot. Think a Japanese Transistor Radio of the 1960s vintage. (which I have disassembled down to every part).

Okay, okay, there was that one mishap on the launching pad, and a near-mishap that other time.. but, the success ratio? Was incredible.

Still amazes me to this day. I have a toaster oven that has computerized controls, and is "smarter" than the Apollo systems were...

... of course, the worst that would happen with my toaster oven failure, would be burned toast... so there's that. (and to be sure, I'd have to go in, and disable several safety features to make it an actual fire hazard, under normal use...)

 

[Revoltingest]

I am still amazed that the US was able to get men from the earth, to the moon, and back, using computers dumber than an Atari 800... and electronics so primitive that Large Scale Integration was too extreme to trust fully--- better stick to discrete transistors and whatnot. Think a Japanese Transistor Radio of the 1960s vintage. (which I have disassembled down to every part).

Okay, okay, there was that one mishap on the launching pad, and a near-mishap that other time.. but, the success ratio? Was incredible.

Still amazes me to this day. I have a toaster oven that has computerized controls, and is "smarter" than the Apollo systems were...

... of course, the worst that would happen with my toaster oven failure, would be burned toast... so there's that. (and to be sure, I'd have to go in, and disable several safety features to make it an actual fire hazard, under normal use...)

Burned toast is still a tragedy.

 

[Shad]

We've miscommunicated then as I was still talking about the thing where people assumed they must be incompetent or (your post?) falsifying maintainance logs. I was talking about how good Ethiopian airlines track record was. Your replies were about the design.

Yup. No problem though as I believe we understood each others points.

I understood what you were talking about. It just wasn't a reply to what the issue in the plane, lack of redundancy in handling sensor input.

If I remember correctly Boeing claimed to have issues training pilots for every problem with their computer systems anyways.

 

[Shad]

IMO, manual backup systems are over-rated in optimizing
safety, reliability & survivability for some planes......
- High performance aerodynamically unstable planes
(which no pilot could fly without fly by wire anyway).
- Large planes with large control surfaces...manual control
can only apply enuf force if augmented by hydraulics. I expect
that any failure which causes loss of all electrical controls would
also cause loss of power to hydraulic pumps.
Of course, this is highly speculative, & can only be confirmed
by extensive experience.

Note that there's a cost in flying anything which doesn't add to
function. I recall reading that saving 1 kg of extra weight would
save something on the order of magnitude of $1M over the life
of an airliner.
Some background....
How much of an improvement would a 1% weight decrease on an airplane be to the industry?

I do not agree as fly by wire will always have the issue of no backups if the computers fail.

 

[Revoltingest]

I do not agree as fly by wire will always have the issue of no backups if the computers fail.

I expect that it won't even be an option if airliners move towards
aerodynamic instability to save fuel & improve performance.

 

[Shad]

I expect that it won't even be an option if airliners move towards
aerodynamic instability to save fuel & improve performance.

Yah a problem of how the military design filters down into commercial design.

 

[Revoltingest]

Yah a problem of how the military design filters down into commercial design.

Problem?
Nah.
Advantage.

 

[Shad]

Problem?
Nah.
Advantage.

In warfare an EM could wreck the US air power. Something both Russia and China knew and design for.

 

[Revoltingest]

In warfare an EM could wreck the US air power. Something both Russia and China knew and design for.

You mean an EMP?
If those things become a risk, we have bigger problems.
For one, what powers the hydraulics for those mechanical
controls you like? Electric motors.
And the instrumentation would also be lost.
The trick there is to work to avoid war & terrorism.

Edit...
Some motors are powered by the engines.
But try operating those without electricity.

 

[Shad]

You mean an EMP?

Yes.

If those things become a risk, we have bigger problems.

It is already a risk. Any nuke creates EMP.

For one what powers the hydraulics for those mechanical
controls you like? Electric motors. And the instrumentation
would also be lost.

The flight control surfaces.

The trick there is to work to avoid war & terrorism.

My point was more about issues that are possible rather than a regular issue.

 

[Revoltingest]

It is already a risk. Any nuke creates EMP.

Are you advocating airliners be designed to be resistant to nuclear warfare?

The flight control surfaces.

Those don't provide any power.
Instead, power is needed to move them. Hydraulics are used
to augment force input by pilots in a mechanical system.

My point was more about issues that are possible rather than a regular issue.

It would be financially impossible (& perhaps entirely impossible) to address every
possibility. And of course, that wouldn't even be knowable. Instead, designers
address what is likely. And that is learned primarily thru experience. This will
determine if mechanical backup systems are worthwhile...until they aren't.

 

[Shad]

Are you advocating airliners be designed to be resistant to nuclear warfare?

I was pointing out a design philosophy which filters down to commercial lines

Those don't provide any power.
Instead, power is needed to move them. Hydraulics are used
to augment force input by pilots in a mechanical system.

Hydraulics are not powered system. I said flight controls as what I want a backup for. Only some instruments would be lost but not all. Gyros for example.

Sorry I had to step away so didn't read your question.

It would be financially impossible (& perhaps entirely impossible) to address every
possibility.

Sure. However flight controls are possible.

And of course, that wouldn't even be knowable. Instead, designers
address what is likely. And that is learned primarily thru experience. This will
determine if mechanical backup systems are worthwhile...until they aren't.

I think it was cut to save costs.

 

[Revoltingest]

Hydraulics are not powered system. I said flight controls as what I want a backup for. Only some instruments would be lost but not all. Gyros for example.

I'm missing something here.
If hydraulics aren't powered, how can they move control surfaces?

Note:
The 737 uses 3000 psi hydraulics.
(5000 psi systems are the modern norm.)
It takes a lot of horsepower to move the control surfaces.
So pilots control the flow which is powered by hydraulic pumps (PCU).
The "mechanical backup" is actually levers & cables connected to
the hydraulic system. They provide feedback for "feel".

Sure. However flight controls are possible.

But there will be flight control situations which weren't predicted.
These will be discovered thru experience.

I think it was cut to save costs.

Save 1 kg....save a million bucks or so over the life of the plane.
And this is primarily in fuel, which will have AGW consequences
in addition to financial, eg, ticket prices, profit.

 

[Shad]

I'm missing something here.
If hydraulics aren't powered, how can they move control surfaces?

As hydraulics use pressure and liquid combined with raw strength of the user to move parts. Power can augment those systems making it easier but is not required.

But there will be flight control situations which weren't predicted.
These will be discovered thru experience.

Sure I think that is to be expected.

Save 1 kg....save a million bucks or so over the life of the plane.

Which is a bottom line cost savings not about safety.

And this is primarily in fuel, which will have AGW consequences
in addition to financial, eg, ticket prices, profit.

Sure. However regarding costs are those worth the increase safety measures? Would you pay more for a ticket on a plane that doesn't turn into a flying brick the moment windows 10 crashes?

We could recover costs on fuel by doing a number of things. Such as standardizing weight classes. If you are obese you pay more for your ticket as you weight is a greater factor than someone that is a male at 180. We could limit cargo passengers can bring. Removal of services such as entertainment, food and access to power for devices.

A lot of planes use hydraulics for flight surface control The fly by wire system links the control stick to the hydraulic systems so there is a gap between the two with no nonelectric backup. So for those planes the measure would be linked the stick to the existing hydraulic system with an new hydraulic system. Rather an old system design as those common in early flight and WW2.

 

[Revoltingest]

As hydraulics use pressure and liquid combined with raw strength of the user to move. Power can augment those systems making it easier but is not required.

Wrongo pongo.
Pilots not only lack the strength to move the control surfaces,
their mechanical connection isn't strong enuf to do that. It only
controls valves in the hydraulic system control lines, & provides
feedback "feel".

See my edit above for more info.

Which is a bottom line cost savings not about safety.

This is unrealistically cynical.
It is all about both.
While I worked only on military systems, most co-workers
had worked on civilian systems too. Everyone considered
safety an important concern. This is evidenced by the great
general safety record for airline travel, & by the effort to
investigate each crash, & make improvements to all systems.

Consider: If safety were poor, there would be no customers.

Sure. However regarding costs are those worth the increase safety measures? Would you pay more for a ticket on a plane that doesn't turn into a flying brick the moment windows 10 crashes?

I wouldn't fly on any plane I thought to be unsafe.

We could recover costs on fuel by doing a number of things. Such as standardizing weight classes. If you are obese you pay more for your ticket as you weight is a greater factor than someone that is a male at 180. We could limit cargo passengers can bring. Removal of services such as entertainment, food and access to power for devices.

Cargo is already limited.
As for the other suggestions, if they made an airline more
competitive, I suspect that some would've already adopted them.
(Lose the entertainment, & planes would become an unacceptable hell.)

Saving money on fuel isn't a matter of taking this or that measure,
& then considering the job done. Designing an airplane is about
far more integration of many considerations than that.
You'll see much more carbon fiber...monolithic construction of
fuselages & wings....more efficient engines....higher pressure
hydraulics....improved aerodynamics....etc. There will be
continuous improvement wherever possible, in both safety
& economics.


About safety. It cannot be perfect.
You wouldn't want perfection because you couldn't afford it.
The question with cars, planes, buildings, etc is how to optimize safety & cost.
I could design a car which is safer than anything on the road now.
But no one would buy it...too large, heavy, slow, & expensive.

I could make air travel at least 50% safer with just a few quick policy changes.
But you wouldn't like that either.
- No flying at nite, during inclement weather, or in low visibility.
- Make every pilot training program around the globe meet top US standards.
- Close dangerous airports entirely, eg, Congonhas in Brazil, Aspen & Yellowstone here.
It's a long list. Many remote places would become inaccessible.
- Cut air traffic 25% at the airports which remain open.
- Allow no small aircraft at the same airports as airliners.
- Remove every other row of seats in planes.

A $500 ticket would balloon up to $5000 or so.
Everyone but the very wealthy would find my policies unacceptable.
But I'd be putting safety ahead of financial concerns.
Or would I?
Increased car traffic would increase the total transportation death rate.

Lo! People would think....
"Gee...let's not demand total safety."
"There should be reasonable compromise."
"Balance cost, convenience, & safety."
"I miss being visiting Junior in NYC for $100 on Spirit....even though I hate Spirit."

 

[LuisDantas]

For good or worse, it is simply not possible to achieve total coverage of all legitimate concerns without seriously impacting the range of options that we have in modern life.

We could perhaps consider going back to a more self-made lifestyle. But that would have a lot of consequences, and I don't think that it can work with 7 billion people walking the Earth.

 

[Jumi]

That seems unlikely to me.
I favor having more sensors, & having the system vote for the majority.
(Of course, there are great complexities lying behind how to "vote".)
This offers the higher probability of avoiding error.

This would have been my expectation as well, but it seems not to be the case. The software is confused by one sensor and doesn't do checks you and I would expect it to.

The problem with the Lion Air flight was the MCAS went to work when it shouldn’t have. The 737 MAX was climbing normally, but due to a faulty sensor the digital flight data recorder detected a hard-to-believe 20-degree difference in the angle of attack between the left and right sides. Over the next 10 minutes, the pilots repeatedly tried to pull the plane’s nose back up, but the MCAS kept forcing the yoke forward, pushing the plane down. Ultimately, the plane crashed into the Java Sea, killing everyone aboard.

I believe we've reached a stage where in an emergency, the pilot
will be more of a problem than a solution. This is because people
are slow & unreliable under stress. A capable automatic system
can act with greater predictability, reliability & speed.

That is what I'd expect from a capable system. This one, it seems, had a fatal flaw. As I read about it, the FAA and Boeing were aware of the issue, but unable to go ahead with the software update due to government shutdown in the US. Whether that's true, I'm uncertain.

Sadly an event like this will probably cause pushback for it is good in general and the only solution in the long run.