Boeing 737 MAX - software wouldn't fix faulty airframe

SamR

Joined Mar 19, 2019
5,031
Will the GPS used in aeronautic avionics now give elevation (over terrain?) as well as Lat/Lon ground position?
 

Tonyr1084

Joined Sep 24, 2015
7,852
Could always maintain a mechanical linkage between the controls and the control surfaces. And *always* allow the pilot to override any automation simply by mechanically flying the airplane. After all, that is what a pilot is trained to do right? If not, we have a bigger problem
It is actually impossible to fly that aircraft and many others using mechanical linkages and has been for many years now.
The control surfaces are so large that mechanical linkage would be impossible due to the forces acting upon them. AT BEST a pilot MIGHT be able to execute a slow roll or yaw. Nose up or down would be even harder because of the balance of a plane is dependent upon stationary loads. People move around during a flight. A heavy person in first class going to the back of the plane to alleviate themselves could cause a nose up condition. The drinks cart is heavy, and moving that around would have huge affects on flight dynamics too.

Mechanical linkage works well for small aircraft at lower speeds because of the reduced surface area and forces acting upon them. Not so on a large aircraft. And I'm not referring to BIG aircraft like the 747 or other wide bodies. Something as small as a Cesna Citation MIGHT be mechanically controllable, might be "Barely" controllable mechanically. That's why they have hydraulics. My experience with aircraft hydraulics, my recollection from the 70's and 80's is that they typically ran at 3000 PSI. Can a human apply that kind of force to a large wing control surface?
 
Last edited:

Alec_t

Joined Sep 17, 2013
14,280
At age 16 I once put a Slingsby Cadet training glider (with only mechanical linkage) into an over-steep bank. It took all my strength with two hands on the control column to get the plane level. I wouldn't fancy trying that on a 737 :).
 

Ylli

Joined Nov 13, 2015
1,086
OK, OK, perhaps the term 'mechanical linkage was not the best. What I mean is that the PIC should have the ability to override the automation and take control of the airplane. After all the PIC is responsible for the operation and safety of the aircraft, he/she should have the ability to control that aircraft.
 

Tonyr1084

Joined Sep 24, 2015
7,852
Boeing is trying to take the "Mistake-ability" of a PIC out of the equation. Unfortunately for the MAX, the computer made the mistake and there was no way to override it. Well, probably WAS a way to override it - but it may be as complicated as turning around to a breaker panel and pulling the breaker that controls the MCAS system. If you've ever seen any of those breaker panels - they can be packed with tens and tens of breakers. Maybe even as many as a hundred or more breakers. Finding the exact breaker when the plane is in a nose dive would be challenging to me. Don't know how challenging it would be to a pilot, but there's not only the main breaker panel, there are other breakers located about in the cockpit of a DC-10 or MD-11. Not familiar with other aircraft - but I can imagine the sheer quantity of breakers to be familiar with.

I remember in my junior high school shop class - there was a big red button on the wall. If someone got into trouble with power anyone could hit that button and kill all power. Perhaps Boeing might do well to install an override button for such emergencies as when the pilot feels there's no chance of recovery of the aircraft without human intervention. Then again, how many pilots would simply hit that button just because they don't trust the system - then go on to make a fatal error?
 

jpanhalt

Joined Jan 18, 2008
11,087
I have read that there's a philosophical difference between Boeing and Airbus regarding the PIC. Boeing still wants pilots to be pilots and gives then a yoke. Airbus just wants pilots to make suggestions and gives them a joystick that is not even in front.

The 737MAX did allow pilots to take command. Unfortunately, not all were trained how to do that, and of course, other deficiencies showed up with further investigation.,

Airbus:
1581970042182.png

Boeing :
1581970114859.png
 

SamR

Joined Mar 19, 2019
5,031
The argument was made long ago to automate aircraft launch and recovery on carriers. The pilots were not at all bashful about their refusal of the idea. The prominent objection was along the lines of "I refuse to put my life in the hands of some 18-year-old pimple-faced kid tasked with working on my avionics"!
 
The problems are due to Boeing outsourcing their software development to $9/hr shops where no experience is necessary, fresh interns can write avionics code even though there are no aircraft manufacturers in the country. Does that seem suspicious? There are other corporations "saving money" this way, the fake claims of having qualified programmers are not kept in check. MCAS is a prime example of where you need top notch in-house software developers, instead of firing them all and outsourcing.
durgasoft ad1.jpg
 

nsaspook

Joined Aug 27, 2009
13,079
The problems are due to Boeing outsourcing their software development to $9/hr shops where no experience is necessary, fresh interns can write avionics code even though there are no aircraft manufacturers in the country. Does that seem suspicious? There are other corporations "saving money" this way, the fake claims of having qualified programmers are not kept in check. MCAS is a prime example of where you need top notch in-house software developers, instead of firing them all and outsourcing.
View attachment 199367
Nope. The software didn't malfunction. The MCAS software specification and implementation wasn't the problem. The flight control specifications for X hardware given to the software engineers to implement was defective/unsafe.
 
Any S/W developer with proper avionics experience would refuse to write substandard code- instead of making a buck. Relying on a single sensor input in a safety-critical system is a software developer fail. The hardware for two sensors and two processors exists on the Max. Even the AoA Disagree annunciator doesn't work despite it being an extra cost feature purchased by some airlines. How does a non-functioning blinky light make it through S/W dev and test?
Engineers follow a Code of Ethics which software engineering does not yet have.

Really, we can argue but Boeing's management team is corrupt and ultimately the aircraft has dozens of errors in specifications, hardware, software.
 

nsaspook

Joined Aug 27, 2009
13,079
Any S/W developer with proper avionics experience would refuse to write substandard code- instead of making a buck. Relying on a single sensor input in a safety-critical system is a software developer fail. The hardware for two sensors and two processors exists on the Max. Even the AoA Disagree annunciator doesn't work despite it being an extra cost feature purchased by some airlines. How does a non-functioning blinky light make it through S/W dev and test?
Engineers follow a Code of Ethics which software engineering does not yet have.

Really, we can argue but Boeing's management team is corrupt and ultimately the aircraft has dozens of errors in specifications, hardware, software.
There's no argument here. Boeing's management of the MAX was abysmal but blaming software developers and engineers is like blaming them for the VW diesel cheating scam because they didn't all quit when given the specifications. Proper pilot training about exactly how the system worked and what do do quickly would have most likely saved lives here. The capability existed to eliminate the effects of MCAS auto-trim but it had to be done quickly and was done quickly by some pilots by flipping the manual trim power cutout switches early and permanently for the rest of the flight.
 
On this and many, many other forums, there has been a boatload of discussion about how these issue will cause Boeing to lose so much money to declare bankruptcy.

I don't think that will happen, ever. Boeing is such an integral part of the military industrial complex, that US government will bail them out.
Remember Lockheed and the L1011.

Having said that, it might force Boeing to separate its commercial and military units.

But the US government might insist that, whoever acquires the commercial unit remains a fully owned American company. Commercial aircraft production is one of those strategic industries where the US still is a top-tier player.
 

Tonyr1084

Joined Sep 24, 2015
7,852
Remember Lockheed and the L1011.
You mean the industrial espionage of the DC 10??? #2 engine configured different. Also different is the front ram air scoops for the air conditioning. DC 10 has two on the left and one on the right whereas the L1011 (also a rip off of the CD 10 name) has two on the right and one on the left. Or I might have that reversed, but the 1011 is opposite of the 10 with the air conditioners. Same exact wing span. Slightly different length but similar configurations as well. Other differences (I'm not aware of) may be electrical wiring, but even the ADG (Air Driven Generator) is mounted in the same location on both planes.
 

402DF855

Joined Feb 9, 2013
271
Will the GPS used in aeronautic avionics now give elevation (over terrain?) as well as Lat/Lon ground position?
AFAIK GPS cannot give terrain information but only X, Y, Z, and T for the receiver's location within the satellite constellation. Clearly, further information can be derived from those (e.g. velocity), but terrain would require a database matching the terrain over the area of travel, which would obviously need to be updated frequently.
Any S/W developer with proper avionics experience would refuse to write substandard code- instead of making a buck.
IME that's not really how avionics are developed. Both S/W and H/W are strictly overseen and directed by the systems engineers who are supposed to determine in fine detail how the system should work. Add to that the demanding and extensive level of review and testing required not only by regulators but the airframe manufacturers as well, and the S/W developers don't have a lot of freedom to make mistakes.

I've been away from commercial avionics development for a while, and haven't followed the MAX debacle closely, but have been shocked at the end result. Boeing/MD, Airbus, Honeywell, Rockwell, et al, have historically had a fairly strong track record I'd say, and problems like we are seeing with the MAX are supposed to be mitigated by the rigorous development process. Not sure where the blame lies, but the system is designed to deny culpability on the part of the code monkeys. (These comments apply to commercial avionics; military is whole other thing.)
 

Reloadron

Joined Jan 15, 2015
7,501
The argument was made long ago to automate aircraft launch and recovery on carriers. The pilots were not at all bashful about their refusal of the idea. The prominent objection was along the lines of "I refuse to put my life in the hands of some 18-year-old pimple-faced kid tasked with working on my avionics"!
Actually many of those pilots are not much older than 18 years old themselves at maybe 23 or 24 years old. Actually they do place their lives in the hands of those writing the software, at least when a FA 18 Hornet or Super Hornet is launched from the deck of an aircraft carrier. The FA18 Flight Manual mentions "Immediately after the end of the catapult stroke, the aircraft will rotate to capture the 12° reference AOA (hands-off). To avoid PIO with the FCS, do not restrain the stick during catapult launch or make stick inputs immediately after catapult launch. The pilot should attempt to remain out of the loop but should closely monitor the catapult sequence". As the aircraft leaves the catapult the pilot is relying totally on his flight computer.

OK, back to the Boeing 737 .... :)

Ron
 

jpanhalt

Joined Jan 18, 2008
11,087
Isn't that what regulators are supposed to do?

Many years ago (>40), after several annuals at a Beechcraft dealership (before I owned the airplane) and two at my local shop, I did an annual inspection with an IA who would work with me. We discovered a slightly oversize close-tolerance bolt in a wing attachment fitting of a 1967 Beech Musketeer B19 (https://en.wikipedia.org/wiki/Beechcraft_Musketeer) . It was "factory." (Maybe sort of like Northrup vs. Boeing rivets. :) ) There was a lot of paperwork involved, but we did it right. Of course, a Beechcraft B19 is nothing in complexity compared to a Boeing Max, but there are regulations and procedure in place that must be followed, and it is the FAA's responsibility to do that.
 
Last edited:
Top