Having been involved in air force quality assurance in a former life, I find the bits that now come out about the certification process at Boeing interesting.
Used to be the FAA that certifies a plane as airworthy with independent testing and review. It seems that during the years that has more gyrated to Boeing doing the testing and the FAA doing the review of the results.
This CAN work, it is used in other industries, but only if that FAA is very strict and can independently ascertain that indeed the manufacturer is on the ball. But the danger of conflicting interests with 'the butcher certifying his own meat' is always there.
I understand Justice has started an investigation into the practice.
Should be of interest to see how that aspect develops.
Jan
Used to be the FAA that certifies a plane as airworthy with independent testing and review. It seems that during the years that has more gyrated to Boeing doing the testing and the FAA doing the review of the results.
This CAN work, it is used in other industries, but only if that FAA is very strict and can independently ascertain that indeed the manufacturer is on the ball. But the danger of conflicting interests with 'the butcher certifying his own meat' is always there.
I understand Justice has started an investigation into the practice.
Should be of interest to see how that aspect develops.
Jan
If it's a fault that can occur in several different aircrafts, then perhaps it could be the specific design and location of one particular sensor?
Does not have to be the code.
Maybe if there's an air pressure sensor that is located in an unprotected location where you can in some instances get a small vacuum effect from stray air currents, combined with data from the other sensors there would be some few instances where the "brain" is fooled into thinking the aircraft is in a much higher altitude.
Perhaps some other, seemingly unrelated, design improvements where made that in some specific scenarios can increase the possibility of these events to occur?
But I have no clue about anything aircraft related.
However, I have seen similar things happen in completely unrelated equipment, a sensor in the wrong place, or changes to a design that requires new-thinking as to where data should be gathered from, or how it should be processed.
Does not have to be the code.
Maybe if there's an air pressure sensor that is located in an unprotected location where you can in some instances get a small vacuum effect from stray air currents, combined with data from the other sensors there would be some few instances where the "brain" is fooled into thinking the aircraft is in a much higher altitude.
Perhaps some other, seemingly unrelated, design improvements where made that in some specific scenarios can increase the possibility of these events to occur?
But I have no clue about anything aircraft related.
However, I have seen similar things happen in completely unrelated equipment, a sensor in the wrong place, or changes to a design that requires new-thinking as to where data should be gathered from, or how it should be processed.
Last edited:
this guy seems to know whats going on.
Trevor Sumner on Twitter: "1of x: BEST analysis of what really is happening on the #Boeing737Max issue from my brother in law @davekammeyer, who’s a pilot, software engineer & deep thinker. Bottom line don’t blame software that’s the band aid for many other engineering and economic forces in effect.👇🎖🤔"
Trevor Sumner on Twitter: "1of x: BEST analysis of what really is happening on the #Boeing737Max issue from my brother in law @davekammeyer, who’s a pilot, software engineer & deep thinker. Bottom line don’t blame software that’s the band aid for many other engineering and economic forces in effect.👇🎖🤔"
It appears there is only one sensor and best practice usually requires 3 and the computer disregards outlier results coming from one sensor.
Also the MACS system was originally able to adjust the flap by 0.6deg each time it interfered. This apparently was not enough to avoid a stall so it was increased to 2.5deg. 5deg total is enough to put a plane into a nosedive so MACS has only to interfere twice for that to occur rather than 8x.
Starting to look like it could develop into a DC10 type screw up and even 30 years later McDonnel Douglas is mostly famous for their planes dropping out of the sky. This could be very serious for Boeing's civilian operations.
In the 1940s in UK we were very poor at that. The std test for an air plane fuel tank was to fill it with water and drop it from the control tower at Farnborough. This excluded self sealing tanks which were a rather useful thing to have.
nice
He is describing his brother in law who writes software for Boeing. They tried to cut corners on the mechanical R&D and fix it with software, but the mechanics are flawed. They swapped engines to a more economical one without changing the structural components of the plane needed to support this engine.
nice
So he's a mechanical expert as well as a deep thinker? Sorry still not clicking on the link.
Pilot who hitched a ride in cockpit saved doomed Lion Air Boeing 737 Max day before it crashed
Boeing: US orders review of 737 Max licence to fly - BBC News
Boeing must have their legal team on danger money...
Boeing: US orders review of 737 Max licence to fly - BBC News
Boeing must have their legal team on danger money...
My conjecture:
Do we know whether the jackscrew trim electro-mechanical system was altered on the 737 Max?
I think there must be a multiple failure problem. Boeing keep saying all the pilots needed to do was disengage the MACS but clearly that didn't help. Both the Lion Air and the Ethiopian appear to have flown steeply into the sea/ground and had their jackscrews in full dive position. It might be that sensor faults caused the MACS system and the subsequent pilots reactions to cause unusual jackscrew activity that led to an electro-mechanical failure. The engines are below the centre of mass of the plane so cranking them up raises the nose. Something a pilot might try if otherwise unable to stop a dive. The Lion Air hit the sea with engines at high thrust, I understand.
Last edited:
This incident got spot light because of unethical usage mass media.
I this days everyone is expert and freely comment end draw conclusion whiteout proper knowledge. Then they hide behind sentence "That is my opinion".
First you have to know how things are work in order to put a comments.
To know how airplanes work you have to go to school. There no information on internet so someone can become instant expert in a moment.
MCAS system is automatic function design to help pilot in command to fly manually aircraft.
It is active when AP(autopilot) is OFF and flaps retracted.
If not commanded motion of horizontal stabilizer occur or pilot in command detect problem in system hi should put STAB TRIM in CUT OUT position. Every pilot on every type of Boeing aircraft should know that no mater if it is 1 or 40yrs old b737.
Lion air pilots that are flying aircraft recognized this problem and they made a entry in technical log book. Aircraft engineers try to fix this problem. I will spare you of technical details, what they replaced and what tests they performed.
The fact is that pilot in command try to "wrestle" with aircraft, only thing that hi should do is to put stab trim in cut out position.
Boeing is very good aircraft and is have some problem like every aircraft.
I this days everyone is expert and freely comment end draw conclusion whiteout proper knowledge. Then they hide behind sentence "That is my opinion".
First you have to know how things are work in order to put a comments.
To know how airplanes work you have to go to school. There no information on internet so someone can become instant expert in a moment.
MCAS system is automatic function design to help pilot in command to fly manually aircraft.
It is active when AP(autopilot) is OFF and flaps retracted.
If not commanded motion of horizontal stabilizer occur or pilot in command detect problem in system hi should put STAB TRIM in CUT OUT position. Every pilot on every type of Boeing aircraft should know that no mater if it is 1 or 40yrs old b737.
An externally hosted image should be here but it was not working when we last tested it.
Lion air pilots that are flying aircraft recognized this problem and they made a entry in technical log book. Aircraft engineers try to fix this problem. I will spare you of technical details, what they replaced and what tests they performed.
The fact is that pilot in command try to "wrestle" with aircraft, only thing that hi should do is to put stab trim in cut out position.
Boeing is very good aircraft and is have some problem like every aircraft.
Most things do, to a non-expert. ;-)
To come back to certification, that is aimed at avoiding problems in the first place. It's all very nice to crash a plane, then find a software error or mech f*uck up and fix it and dole out blame.
But aeronautical QA tries to set in place processes and procedures to avoid that systems or processes are non-tested when the design is certified.
For instance, who decided that it wasn't necessary to train & tell pilots that there's a new system in their planes called MCAS?
These are the typical things that should be decided by an outfit like the FAA, NOT by the manufacturer, who will always be tempted to cut costs be it for hardware or for training.
These issues are at the root of safe flying.
Jan
Perhaps. But you also need to identify and understand what the problem is before you can take proper action.
Take the example of a riveter working on a plane. You can't realistically check all 1000's maybe 10,000's of rivets. So how to you make sure the rivets are OK and won't pop over the Atlantic?
You certify the riveter. You select a guy who is a stable individual, never been on substance, off the hard liquor, train him to the best you can, run him through some hard checks and then certify him as 'certified riveter'.
And if he gets sick, replace him ONLY with an equally certified riveter, not some Joe who knows a rivet gun and can fall in for the sick guy for the day.
Make that certificate limited validity, say a year, then again put the guy through hard tests before certifying for another year. That's how you build a plane. Or a sub.
And have an independent inspector in the plant with unlimited access, who has the power to shut down the assembly line if, say, the spray painter cannot show a valid certificate as 'certified spray painter'.
And the same way you certify people and processes, certify the design.
I am pretty sure that the so-called 'self certification' which seems to be en vogue these days is the root root cause of these problems. I've seen too much of it.
Jan
You certify the riveter. You select a guy who is a stable individual, never been on substance, off the hard liquor, train him to the best you can, run him through some hard checks and then certify him as 'certified riveter'.
And if he gets sick, replace him ONLY with an equally certified riveter, not some Joe who knows a rivet gun and can fall in for the sick guy for the day.
Make that certificate limited validity, say a year, then again put the guy through hard tests before certifying for another year. That's how you build a plane. Or a sub.
And have an independent inspector in the plant with unlimited access, who has the power to shut down the assembly line if, say, the spray painter cannot show a valid certificate as 'certified spray painter'.
And the same way you certify people and processes, certify the design.
I am pretty sure that the so-called 'self certification' which seems to be en vogue these days is the root root cause of these problems. I've seen too much of it.
Jan
- Status
- Not open for further replies.