I ran across this article from 2015. I would love to hear your guys' thoughts on this.

https://www.vanityfair.com/news/2015/11 ... bing-truth

Remember when the USA had a manufacturing industry? Go down to your local Harbor Freight and try to find anything made in the USA. You won't. Why you might ask? It's because Americans want cheap and if you thought that Americans wanted quality made products, you would be mistaken. Same way with your airlines ticket. If you want cheap, you have to get on board with the Airlines ability for cheap which is to maintain their aircraft as cheaply as possible.

To the CEO's of airlines, the added risk involved with sub standard maintenance is just something that if push comes to shove, they will just attribute failures to pilot error because the burden of proof paper trail of sub standard maintenance is hard to prove.

I’m not so sure about the imminent disaster alluded to here. Boeing produced a number of 737s with the engine fire system wired backwards. Fairchild Republic delivered A-10s with foreign objects in a sterile part of the fuselage that caused some accidents thru suspected control jams. Eastern had an L-1011 with all three engines out of oil. AK 262 crashed due to a jackscrew that was improperly maintained. FDX had their 727 reworked in Portugal where the work was as good as anywhere.

Foreign work can be good just as domestic work can be terrible. If this is a problem why all the love for Russian or Chinese built types?

GF

When was the last time we had a fatal accident due to some sort of major overseas (or domestic) maintenance failure?

Agree that the article is a bit overly pessimistic.

Fact is aviation has never been so safe as it is today. We see less accidents/incidents in a much larger fleet of aircraft. What the article doesn't say that even with all the FAA certified staff in the US, mistakes happened. I recall countless of accidents/incidents where maintenance personnel (in the US and elsewhere) did not follow the procedures.

I work in airline maintenance and I am not worried to board any aircraft maintained in China, Jordan, etc. Also funny to that the author calls Singapore a 'developing world' country.

Regards,
Loran

I worry about this and about maintenance in general. The biggest routine issue is that even if something is written up, the deviation is frequently written off as "cannot duplicate". We, in the front office, need far better self discipline to document every and any deviation. Sometimes pilots do not document problems because they go away on their own or because some pilots perform actions to make them go away. Pilots need to be scrutinized about writing up every single defect, whether it goes away or not. Newer airplanes show errors for a reason but pilots frequently dismiss it as a nuisance. My take on it is that if it is a known sporadic thing, either get the manufacturer to rewrite the manual or redesign the system. Dont bet on luck and statistics to make the case NOT to fix it. Furthermore, anything written up should never be dismissed as Cannot Duplicate. Lastly, the FAA needs to greatly decrease the amount of time something can be on an MEL and limit the number of consecutive times the write up occurs. The MEL list needs to be made smaller so that airlines don't fly with entire book length quantities of MEL writeups. The FAA is far too tolerant of airlines. I think this greatly impacts our safety. Of course, the FAA won't ever say anything is a safety issue unless someone dies because the rules are written to always blame the pilot.

But, how do you “fix” deviations that come and go? Loads of it is software that didn’t boot up correctly or had a transient problem. Mechanical problems are usually obvious and don’t come and go without reason.


GF

I have worked in aviation maintenance for about 30 to 35 years now. Started out in the U.S. Navy and now work at one of the major U.S. based freight airlines. I read the post above about foreign maintenance personnel and how it is assumed that they are somehow substandard. I would say this is not the case. In October of 2016, I attended a maintenance school at Boeing in Renton, WA. There were many maintenance people from all over the world in my class including Mexico (Delta), China, Taiwan, Africa, and Russia. They were all fluent in the English language and very knowledgeable about Boeing aircraft and aircraft maintenance procedures. Many of these maintenance personnel were required to attend and pass several Boeing courses in their area of expertise. This, ironically, is something not required by U.S. based airlines. I know because the course I attended was paid for out of my own pocket, it is not required by my company, yet I wanted this training, so I went on my own. The airline I work for has 388 trunk line aircraft. Some of the foreign airlines that had their mechanics going through required training at Boeing had not even a tenth of this fleet. Yet, they must go through these courses to continue working on their aircraft. Also, at my airline we have heavy maintenance and modifications done in foreign nations such as Italy and Singapore. I can attest to the fact that this is done to impeccable standards, the finished product is amazing. Everything done to exact specifications. I have had the opportunity to work in several foreign nations and can say assuredly that at the facilities I was at, the level of expertise was high.

Gregory

GalaxyFlyer wrote:

But, how do you “fix” deviations that come and go? Loads of it is software that didn’t boot up correctly or had a transient problem. Mechanical problems are usually obvious and don’t come and go without reason.
GF

Very good point. Even on new aircraft "Can Not Duplicate" can be a very real thing for flight squawks. What you do to sign off a flight squawk is you confirm the condition or state that the condition can not be confirmed and state the repair or diagnostic steps used to address the flight squawk.

Many years ago, a co-worker deffered a worn main gear tire per the MEL and guess what? About 20 minutes later we were changing the tire with PAX on board because the pilot would not accept the aircraft. So no, as implied in a previous post the MEL does not give carta blanche to just ignore potential safety issues.

GalaxyFlyer wrote:

But, how do you “fix” deviations that come and go? Loads of it is software that didn’t boot up correctly or had a transient problem. Mechanical problems are usually obvious and don’t come and go without reason.


GF

Believe it or not, software is more consistent to diagnose. The nice thing about computers is that they are 100% logical. They never make mistakes which means you can always trace something back to a logical reason. When I write a program, I can put "breaks" in the code where the code execution will pause and I can look at every portion of memory and see the contents of variables.

By the way, computers don't make mistakes -- the software always does exactly what it's programmed to do. When we think the computer is broken, it's just programming being executed that isn't doing what *we* expect, but it is indeed following its program to the letter. As far as didn't boot up correctly or transient problem, those terms don't really exist in software engineering (I'm a computer engineer & pilot, BTW). If the computer doesn't start up correctly there's probably an "unhandled exception" or a mechanical issue such as a power supply not putting out sufficient power or the wrong voltage. An unhandled exception is whenever your code doesn't address the conditional just encountered. For example, you might write your software to do Action A if X < 10 but do Action B if X > 10. But what if X = 10? That's an unhandled exception. Software engineering is super hard because you have to catch millions of logical conditions that are compounded at thousands of layers deep. That's why large companies throw thousands of software engineers at products like Windows and it still has bugs. Aircraft software is simpler, but undergoes considerable testing. One of the difficult things to deal with as an avionics engineer is the extreme limitations of the hardware. There's virtually no memory in most onboard computers and the CPUs are dirt slow by today's PC standards. Again... a "design choice". If you work on Airbus aircraft, you know that it monitors lots of circuit breakers and well as tons of parameters. This data could be used for troubleshooting but not all airlines actually do so. So bottom line is that if airlines wanted to be able to track and diagnose every single fault and anomaly, they could. The technology is certainly there.

Nice explanation for why the fix often is, “shut ‘er down and start it again” and it works fine. I’ve seen it is C-5s when Collins installed the FMS in them. A Collins Engineering team from Cedar Rapids told me to that. I was astounded, but then saw the same process in Collins and Honeywell civil installations.

GF

GalaxyFlyer wrote:

Nice explanation for why the fix often is, “shut ‘er down and start it again” and it works fine. I’ve seen it is C-5s when Collins installed the FMS in them. A Collins Engineering team from Cedar Rapids told me to that. I was astounded, but then saw the same process in Collins and Honeywell civil installations.

GF

That's because they don't want to be on the hook for offering too much...err.. what's really possible. The blame always is squarely on management. Engineers always want to create elegant solutions. This is why the A320 is such a beautifully designed airplane vs. the 737 which is designed/managed by the finance department. Unfortunately the companies they work for want to sell support programs, spare parts, and upgrades without regard to doing state-of-the-art. Ever watch Robocop? Remember how Dick Jones said "we have spare parts contracts... (a few other things).... Who cares if it didn't work!?" (in reference to ED-209).

Trust me, if they wanted to diagnose and fix the real problem, they *could*. However, they are constrained to supporting an official release and using only what the current scope of support contract allows for. In the case of the USAF, they are not under FAA jurisdiction but they are under AFIs that prohibit what a contractor can do unless the SPO gets funding for it. Again, this is all artificial.... Go visit their engineering offices sometime. You'll see all avionics on test benches where they do black box and white box testing as well as automated software testing. No one wants to add any capability to an airplane unless it somehow lowers cost. This isn't a problem of capitalism but rather an indicator of mediocre management teams who myopically only understand cost cutting rather than innovation and progress.

Yup, computers don't make mistakes. They always do what they are instructed to do. https://www.youtube.com/watch?v=-cv2ud1339E

stratclub wrote:

Yup, computers don't make mistakes. They always do what they are instructed to do. https://www.youtube.com/watch?v=-cv2ud1339E

The aircraft in the Mulhouse/Habsheim crash behaved exactly as intended. It was deliberately flown into a low energy situation very close to the ground with no room to recover.

The fact that the aircraft didn't stall and dip a wing, leading to a potentially worse outcome, can be credited to alpha protection in the flight control systems.

As mentioned in the thread about manual handling characteristics, airliners have lots of inertia. They don't change trajectory instantly, and importantly, in this case, there was no energy to spare for doing so. Until the engines spooled up and added energy, the aircraft had to descend or stall.

The engines actually spooled up slightly faster than required for certification. But there was no time left.

greendot wrote:

GalaxyFlyer wrote:

Nice explanation for why the fix often is, “shut ‘er down and start it again” and it works fine. I’ve seen it is C-5s when Collins installed the FMS in them. A Collins Engineering team from Cedar Rapids told me to that. I was astounded, but then saw the same process in Collins and Honeywell civil installations.

GF

That's because they don't want to be on the hook for offering too much...err.. what's really possible. The blame always is squarely on management. Engineers always want to create elegant solutions. This is why the A320 is such a beautifully designed airplane vs. the 737 which is designed/managed by the finance department. Unfortunately the companies they work for want to sell support programs, spare parts, and upgrades without regard to doing state-of-the-art. Ever watch Robocop? Remember how Dick Jones said "we have spare parts contracts... (a few other things).... Who cares if it didn't work!?" (in reference to ED-209).

Trust me, if they wanted to diagnose and fix the real problem, they *could*. However, they are constrained to supporting an official release and using only what the current scope of support contract allows for. In the case of the USAF, they are not under FAA jurisdiction but they are under AFIs that prohibit what a contractor can do unless the SPO gets funding for it. Again, this is all artificial.... Go visit their engineering offices sometime. You'll see all avionics on test benches where they do black box and white box testing as well as automated software testing. No one wants to add any capability to an airplane unless it somehow lowers cost. This isn't a problem of capitalism but rather an indicator of mediocre management teams who myopically only understand cost cutting rather than innovation and progress.

While I agree that the A320 is beautifully (and better) designed - you could argue that even it has fallen under the same mentality, as it took Airbus 27 years (1988-2015) to significantly update the thing. And the wing turns 30 this month...

It seems like the Cold War ended and everyone discovered crack (LSD in Europe) and hip hop, and it was all over for innovation in civil aviation

stratclub wrote:

Yup, computers don't make mistakes. They always do what they are instructed to do. https://www.youtube.com/watch?v=-cv2ud1339E

It didn't make a mistake. It did exactly what it was programmed to do. That was 100% pilot error. Like I said, it's merely not doing what you "expect" it to do. This is largely the result of poor mental ergonomics designed into the man-machine interface. FYI, engineers take many college classes in human factors, ethics, and practice design methodologies that mitigate these known problems. The issues all revolve around simply getting something to market rather than letting the engineers create what they want to create.

greendot wrote:

stratclub wrote:

Yup, computers don't make mistakes. They always do what they are instructed to do. https://www.youtube.com/watch?v=-cv2ud1339E

It didn't make a mistake. It did exactly what it was programmed to do. That was 100% pilot error. Like I said, it's merely not doing what you "expect" it to do. This is largely the result of poor mental ergonomics designed into the man-machine interface. FYI, engineers take many college classes in human factors, ethics, and practice design methodologies that mitigate these known problems. The issues all revolve around simply getting something to market rather than letting the engineers create what they want to create.

It was a breakdown in the logic of the man machine interface. The machine was set up to land and from what I read, by the time the pilot realized this and tried to initiate a go around, it was to late. As I said, the computer did exactly what it was instructed to do.

We certainly have come a long way to getting the man/machine stuff right with human factors analysis, training and CRM training.

This is from the LionAir crash and it makes my point...

https://theaircurrent.com/wp-content/up ... p-hr-2.jpg

Look at the piss poor maintenance response... it was Ops Test OK On Ground. As an airline pilot and engineer, this really pisses me off. I can't tell you how many times I write stuff up only to be dismissed away by this kind of response. As an engineer, I would never accept this as a resolution. Whatever the cause is, it MUST BE FOUND. I realize that maintenance cannot diagnose outside of established manufacturer procedure but it points to a deeper problem. Basically the FAA has failed to regulate, just like the FDA fails to protect your health. The FAA allows airlines to set policy through economic considerations. Otherwise, the NTSB would be allowed to set minimum design and performance criteria. The FAA has a split mandate to regulate air traffic while promoting air travel economically.

If it were me, I would require every logbook writeup to be positively diagnosed to a root cause and then the offending component be replaced. Plus, there should be active campaigns to redesign any part that has a history of at least 2 issues across any operator. The reason I would set such a low bar for action is that manufacturers are not incentivitzed to design high reliability components. Aircraft reliability is abysmal. Every company always has something on an MEL, CDL, or NEF. Imagine if cars had something wrong with them on every trip! Our expectations for aircraft are very low. Industry needs to be forced to progress into using state of the art technologies. For example, rather than using 1970s style circuit boards with large solder joints, all computers should employ modern small circuit board manufacturing techniques that radically reduce the size of components. This would reduce the number of broken solder joints and even reduce the number of wires. Look at your iPad... notice is doesn't even use wires inside? This makes it super reliable since components are radically smaller.

For Boeings 737 design, how about using hundreds of AOA sensors including the same accelerometers andsolid state gyros used in phones and tablets? Why limit the design to just 1 or 2 AOA sensors? This is a big reason why humans are infinitely better than computers in aircraft. Humans have massive parallelism whereas aircraft are designed with huge achilees heels... only few sensors and only a few communication lines. There are many ways to improve aircraft reliability but the FAA is not requiring what they should. They are allowing industry to dictate low standards. In the long run, a better designed aircraft will lower cost.

stratclub wrote:

greendot wrote:

stratclub wrote:

Yup, computers don't make mistakes. They always do what they are instructed to do. https://www.youtube.com/watch?v=-cv2ud1339E

It didn't make a mistake. It did exactly what it was programmed to do. That was 100% pilot error. Like I said, it's merely not doing what you "expect" it to do. This is largely the result of poor mental ergonomics designed into the man-machine interface. FYI, engineers take many college classes in human factors, ethics, and practice design methodologies that mitigate these known problems. The issues all revolve around simply getting something to market rather than letting the engineers create what they want to create.

It was a breakdown in the logic of the man machine interface. The machine was set up to land and from what I read, by the time the pilot realized this and tried to initiate a go around, it was to late. As I said, the computer did exactly what it was instructed to do.

We certainly have come a long way to getting the man/machine stuff right with human factors analysis, training and CRM training.

The aircraft was not “set up to land”. That is one of the persistent myths of that crash. There is no logic in an airliner that forces a landing, as evidenced by the fact that pilots go around routinely.

The pilot flying had slowed down to an unsustainable airspeed while still descending close to the ground. There was simply not enough energy to recover. Alpha protection kept the angle of attack below critical, but without enough energy the aircraft had to descend. A further pitch up would have led to a stall.

I wrote the below years ago on this board, based on the accurate account in the first “Air Disaster” book.


A320 at Mulhouse was an in service AF aircraft doing a flyby at an airshow. So what went wrong?

Some preliminaries:
- Pre flight briefing was not held.
- Flyby altitude was below French regulatory minimums.
- Flyby altitude was below the level of surrounding obstacles (trees).
- Flyby runway was changed at last minute (when in sight of the airfield).

As the plane flew by, speed and altitude were still both decreasing. Engines were at or near idle. The flight control system was compensating by trimming up. But since the plane was just above stall speed, once it reached the max angle of attack the only way not to stall was to decrease the pitch angle and thus descend.

By the time the pilot saw the approaching trees, it was already too late. No amount of thrust would have allowed the plane to clear the trees. On the video you clearly hear the engines spooling up just in time to become woodchippers.

The pilot flying had overconfidence in the plane's control system allowing him to perform a dramatic low speed flyby. In fact, without the control system and envelope protection, the plane might have stalled (it never did). But not stalling is not enough. You also have to avoid hitting the ground.

Nothing was wrong with the plane, and all components performed according to (or beyond) specs. Airbi may have very cool envelope protection systems, but they can't break the laws of physics.

If the pilot had tried the same stunt in a 737 or any other plane, there would still have been a crash.

Old pilot saying: "Use your superior skills to keep yourself out of situations where you might need to use them".

greendot wrote:

Aircraft reliability is abysmal. Every company always has something on an MEL, CDL, or NEF. Imagine if cars had something wrong with them on every trip! Our expectations for aircraft are very low. Industry needs to be forced to progress into using state of the art technologies. For example, rather than using 1970s style circuit boards with large solder joints, all computers should employ modern small circuit board manufacturing techniques that radically reduce the size of components. This would reduce the number of broken solder joints and even reduce the number of wires. Look at your iPad... notice is doesn't even use wires inside? This makes it super reliable since components are radically smaller.

So every car is perfect on the road? No cars with a failed indicator driving around? No under-inflated tires. No tensioner pulley wearing out, but an owner deciding will/can only repair it on the next service?
With all the new cars with fancy ECU's, fault codes are stored for later maintenance even without switching on the CEL.

Imagine if no MEL's of CDL's were allowed: 'Sorry folks, we're not allowed to fly today. A navigation light has burnt out, it will take (x) hours to fix and sign off the books. By the time it's sorted out, we missed our slot and will be right in the back of the queue. Really, we don't need it for a daytime flight, but our plane has be show-room condition with no discretion (researched and documented and analysed, not pilot's whim) is allowed.'

MEMS are taking over slowly in small appliances, but your iPad is hardly certified to the same failure probabilities than an airliner. And if your iPad's accellerometer fails, big deal. Take it in and have it fixed. Nothing that you could've been killed by.

SAAFNAV wrote:

greendot wrote:

Aircraft reliability is abysmal. Every company always has something on an MEL, CDL, or NEF. Imagine if cars had something wrong with them on every trip! Our expectations for aircraft are very low. Industry needs to be forced to progress into using state of the art technologies. For example, rather than using 1970s style circuit boards with large solder joints, all computers should employ modern small circuit board manufacturing techniques that radically reduce the size of components. This would reduce the number of broken solder joints and even reduce the number of wires. Look at your iPad... notice is doesn't even use wires inside? This makes it super reliable since components are radically smaller.

So every car is perfect on the road? No cars with a failed indicator driving around? No under-inflated tires. No tensioner pulley wearing out, but an owner deciding will/can only repair it on the next service?
With all the new cars with fancy ECU's, fault codes are stored for later maintenance even without switching on the CEL.

Imagine if no MEL's of CDL's were allowed: 'Sorry folks, we're not allowed to fly today. A navigation light has burnt out, it will take (x) hours to fix and sign off the books. By the time it's sorted out, we missed our slot and will be right in the back of the queue. Really, we don't need it for a daytime flight, but our plane has be show-room condition with no discretion (researched and documented and analysed, not pilot's whim) is allowed.'

MEMS are taking over slowly in small appliances, but your iPad is hardly certified to the same failure probabilities than an airliner. And if your iPad's accellerometer fails, big deal. Take it in and have it fixed. Nothing that you could've been killed by.

Funny how lighting related groundings went way down with the intruction of LEDs. Seems like faster adoption of superior technologies sure made things better. I remember the days of flying is 707 when we were frequently grounded for analog instrument failures like ADIs that would precess out of control on the ground. Reliability sure went up with glass cockpits. Maybe you haven't been flying recently but some MEL restrictions are you come across now are insanely complex. Have several MEL items and things get extremely complex. At least with jurassic jets, you had a chance of understanding deep systems knowledge. That just isnt the case with new airplanes. Look at that recent AirAsia crash. Had the pilots understood and diagnosed it within a finite amount of time, they might have had a chance. In newer aircraft, 95% of the System is totally abstracted to the pilot. The truth is that it has become almost impossible to really know the systems because it's largely invisible to you. To begin to understand things you have to know how to engineer software and even then, you have to be one of the guys that actually wrote the software to know what it's programmed to do.

So what does this mean? It means that we have pilots that are less capable of solving a problem because of their insufficient knowledge base. Therefore, you have to engineer away any of the human factors related failure modes. This means that the machine has to become radically more reliable. Consider a self driving car. When I was in college, I worked on one of the first self driving cars and I can tell you that everything about it was extremely unreliable and I would never trust it. Advance the clock 25 years and you have self driving cars that are orders of magnitude more sophisticated. They are also no more reliable and what I designed. In fact they are worse because their failure modes have so many secondary and tertiary effects. So what do you think would be the best solution? Do you have a huge airplane sized instruction manual for self driving cars in the event something goes wrong one out of 100,000 times even though that 1 time it mows down 30 kids in a school zone? Do you create a self driving car with an MEL that the high school dropout driver has no prayer of understanding or executing? Do you really want to create a big human factors nightmare waiting to happen?

Of course not.... you engineer more reliable systems. You improve the regulatory environment to require defense in depth and design for compound failures making the whole thing more reliable. You know the engineers who designed Fukushima and various similar nuclear power plants here in the United States always knew that the Fukushima incident was gonna happen. In fact it nearly also happened when a river flooded a nuke plant in the central part of the USA! However, at the time the nuclear regulatory commission did not require high standards for nuclear plants certification. In other words, the engineers wanted to create something that could handle compound failures, such as the loss of back up generators in the loss of cooling water. Thanks to low design requirements, overall reliability was nowhere near the level it should have been at. Bean counters stopped engineers from doing the ethical thing. Then Fukashima happened. Now the regulators final you're doing with the engineers wanted to do over 60 years ago. Now there is regulatory support for higher reliability such as the Westinghouse TPC 4000 reactor that is designed to fail safe for any number of compound emergencies. In fact, it is so reliable that he really only requires gravity to safely shut down. Airplane reliability, in my opinion, is more like Chernobyl level reliability when it comes to what is truly possible.

The FAA needs to evolve much faster and force airlines to innovate and provide more reliable solutions rather than allow endless MEL writeups.

greendot wrote:

This is from the LionAir crash and it makes my point...

https://theaircurrent.com/wp-content/up ... p-hr-2.jpg

Look at the piss poor maintenance response... it was Ops Test OK On Ground. As an airline pilot and engineer, this really pisses me off. I can't tell you how many times I write stuff up only to be dismissed away by this kind of response. As an engineer, I would never accept this as a resolution. Whatever the cause is, it MUST BE FOUND. I realize that maintenance cannot diagnose outside of established manufacturer procedure but it points to a deeper problem. Basically the FAA has failed to regulate, just like the FDA fails to protect your health. The FAA allows airlines to set policy through economic considerations. Otherwise, the NTSB would be allowed to set minimum design and performance criteria. The FAA has a split mandate to regulate air traffic while promoting air travel economically.

If it were me, I would require every logbook writeup to be positively diagnosed to a root cause and then the offending component be replaced. Plus, there should be active campaigns to redesign any part that has a history of at least 2 issues across any operator. The reason I would set such a low bar for action is that manufacturers are not incentivitzed to design high reliability components. Aircraft reliability is abysmal. Every company always has something on an MEL, CDL, or NEF. Imagine if cars had something wrong with them on every trip! Our expectations for aircraft are very low. Industry needs to be forced to progress into using state of the art technologies. For example, rather than using 1970s style circuit boards with large solder joints, all computers should employ modern small circuit board manufacturing techniques that radically reduce the size of components. This would reduce the number of broken solder joints and even reduce the number of wires. Look at your iPad... notice is doesn't even use wires inside? This makes it super reliable since components are radically smaller.

For Boeings 737 design, how about using hundreds of AOA sensors including the same accelerometers andsolid state gyros used in phones and tablets? Why limit the design to just 1 or 2 AOA sensors? This is a big reason why humans are infinitely better than computers in aircraft. Humans have massive parallelism whereas aircraft are designed with huge achilees heels... only few sensors and only a few communication lines. There are many ways to improve aircraft reliability but the FAA is not requiring what they should. They are allowing industry to dictate low standards. In the long run, a better designed aircraft will lower cost.

Well first off if air carriers required every logbook entry to be positively diagnosed and the offending component replaced the air transportation system would grind to a screeching halt. That's why we have MEL/CDL relief. Also not all problems are a component a lot of times there is a short or open wiring problem that can be brutal to find if it is an intermittent problem. The reason your car doesn't have everyday squawks is that airliners have multiple redundant systems and far more parts and equipment than an automobile has so naturally there is bound to be some minor issue on a lot of days in an aircraft. Also not every aircraft system or part is monitored either. Now what do I think of the FAA? Not much actually. A lot of the inspectors I have met do not know the aircraft they are assigned to monitor they most of the time are looking for paperwork violations. They have the nickname the "tombstone agency" for a reason. But I do agree with an airplane that has the same write up multiple times with no solution then that aircraft needs to be taken off the line until the problem is found and when I worked maintenance control we did that on occasions.

stratosphere wrote:

greendot wrote:

This is from the LionAir crash and it makes my point...

https://theaircurrent.com/wp-content/up ... p-hr-2.jpg

Look at the piss poor maintenance response... it was Ops Test OK On Ground. As an airline pilot and engineer, this really pisses me off. I can't tell you how many times I write stuff up only to be dismissed away by this kind of response. As an engineer, I would never accept this as a resolution. Whatever the cause is, it MUST BE FOUND. I realize that maintenance cannot diagnose outside of established manufacturer procedure but it points to a deeper problem. Basically the FAA has failed to regulate, just like the FDA fails to protect your health. The FAA allows airlines to set policy through economic considerations. Otherwise, the NTSB would be allowed to set minimum design and performance criteria. The FAA has a split mandate to regulate air traffic while promoting air travel economically.

If it were me, I would require every logbook writeup to be positively diagnosed to a root cause and then the offending component be replaced. Plus, there should be active campaigns to redesign any part that has a history of at least 2 issues across any operator. The reason I would set such a low bar for action is that manufacturers are not incentivitzed to design high reliability components. Aircraft reliability is abysmal. Every company always has something on an MEL, CDL, or NEF. Imagine if cars had something wrong with them on every trip! Our expectations for aircraft are very low. Industry needs to be forced to progress into using state of the art technologies. For example, rather than using 1970s style circuit boards with large solder joints, all computers should employ modern small circuit board manufacturing techniques that radically reduce the size of components. This would reduce the number of broken solder joints and even reduce the number of wires. Look at your iPad... notice is doesn't even use wires inside? This makes it super reliable since components are radically smaller.

For Boeings 737 design, how about using hundreds of AOA sensors including the same accelerometers andsolid state gyros used in phones and tablets? Why limit the design to just 1 or 2 AOA sensors? This is a big reason why humans are infinitely better than computers in aircraft. Humans have massive parallelism whereas aircraft are designed with huge achilees heels... only few sensors and only a few communication lines. There are many ways to improve aircraft reliability but the FAA is not requiring what they should. They are allowing industry to dictate low standards. In the long run, a better designed aircraft will lower cost.

Well first off if air carriers required every logbook entry to be positively diagnosed and the offending component replaced the air transportation system would grind to a screeching halt. That's why we have MEL/CDL relief. Also not all problems are a component a lot of times there is a short or open wiring problem that can be brutal to find if it is an intermittent problem. The reason your car doesn't have everyday squawks is that airliners have multiple redundant systems and far more parts and equipment than an automobile has so naturally there is bound to be some minor issue on a lot of days in an aircraft. Also not every aircraft system or part is monitored either. Now what do I think of the FAA? Not much actually. A lot of the inspectors I have met do not know the aircraft they are assigned to monitor they most of the time are looking for paperwork violations. They have the nickname the "tombstone agency" for a reason. But I do agree with an airplane that has the same write up multiple times with no solution then that aircraft needs to be taken off the line until the problem is found and when I worked maintenance control we did that on occasions.

It SHOULD grind to a screeching halt. Look what happens when it doesn't... and this is just today... Remember, I'm saying this as an airline pilot.

https://www.bloomberg.com/news/articles ... nd=premium

The system wouldn't come to a screeching halt if systems were designed from the beginning for greater reliability. Do you see airlines requiring this from manufacturers? I just told you about the nuclear industry as an example of what can be done. Do you see airlines doing a proper amount of pre-emptive maintenance? Nope. They wait to fix things only due to (A) FAA mandated inspection periods, (B) it gets written up by a pilot. The question is, do we want to actually advance the industry or keep suffering in the same mediocre paradigms of 1940's aviation?

I do understand much about aircraft because I'm an airline pilot but also an engineer by education. I think many aspects of aircraft are garbage design. The main problem is the lack of redundancy. If you're a pilot, your idea of redundancy is limited to thinking about having 2 or 3 FMSs or backup direct control cables to flight controls. As an engineer, I see that sensors are usually 1 system deep. Electrical systems are likewise only the bare minimum. Look at the recent Boeing crash... what kind of redundancy was there in the software to consider the possibility of faulty AOA sensors? I certainly would've designed the software with this possibility. It certainly IS NOT out of anyone's imagination, particularly that of engineers. Perhaps Boeing decided not to add the safety feature of multiple redundant AOA systems because of bean counters. Like I already said, why wasn't the system designed with accelerometers and solid state gyroscopes like in your cell phone, in addition to systems that measure AOA directly? Like I have said many times, an engineer who really knows the systems would find many, many ways to improve aircraft systems and to improve reliability rates.

I just read this a few minutes ago... it proves my point... why was it allowed to fly? I see this kind of stuff in domestic USA operations weekly. Repeat write ups. Horrible maintenance philosophy from design to execution. We must improve!

https://www.bloomberg.com/news/articles ... nd=premium

Disagree,
Although some time, in some companies, there are “pencil whipped complaints, the way to combat this is not reducing MEL time or MEL qty. To get a needed part and to replace it and test, it’s a time consuming, no easy task, that many times requieres more than 10 days ( for more common “C” MEL’s items)
Operational Test- OK is a normal answer as per AMM, when You had made a test following all requiered steps, and system perform as it should, as per AMM, problem is to follow all proccedures, including, selecting between operational or systems test, as indicated by Manufacturer Trouble/Shooting Guide.

Many, many times, after a system reset vía pulling / pushing related CB, You are not able to repit failure. Them you have two options, as identified by AMM, make a deeper System Test t/s or, after proper system test w/o incidences, close the item.

What really makes differences is when QI or Specialized personnel check all Log Book answer’s daily, and call Mx personnel when any answer deviated from proper proccedures .
In one company I work for, to avoid discussions, when a complaint reiterated, a Mx QI was ó/b, on next flight, to check system behavior. Believe it or not. Many repetitive complaints were closed after that....

Rgds

Sorry, also disagree with Mr. Greendot about preventive Maintenance..
Most of a/c follow a Maintenance Program, proposed by Aircraft Maintenance and approved by regulator (FAA),
In the states, a Maintenance Program for. Commercial Aircraft, consist in:

For Line Maintenance:
Post Flight Checks
Daily Checks
Weekly Checks
Preflight Checks

Heavy Maintenance
Check 1, 2,
Check 3 (called also Capital Check)

This checks are stablished by Manufacturer and includes System Test of all systems
In the Checks, Regulators also include some check and test related to incidents or accidents already happening in regulator’s control area or in other country, when respective regulator issue report about.
System, made by humans, is propense to human errors, but it’s corrected on a regular base.
It’s easy to blame the system, the regulator, the Pilots or maintenance, what it’s not, is to check what had failed in the system and what measures should been taken to correct this situation.

Rgds

Stratosphere, High:
Positively diagnosis of every log book item is possible, of course, but it may demand recourses not currently available to a company: testers, special tools, Ground Support Equipments.... Some times, when failures are very complicated, assistance for Manufacturer is needed, including in site assistance,
To follow that recommendation, a/c will have to be ground in cases when any doubts may arise.
Rgds

Hi:
Recently a plane was sent to Sud America for a check, reason was clearly explained: man-hour there cost 1/3 of the amount that will invoice a local company in the states.
And now a question, are pax ready to pay such a Mx price increase, if Mx is performed at home?
Rgds

Starlionblue wrote:

stratclub wrote:

greendot wrote:

It didn't make a mistake. It did exactly what it was programmed to do. That was 100% pilot error. Like I said, it's merely not doing what you "expect" it to do. This is largely the result of poor mental ergonomics designed into the man-machine interface. FYI, engineers take many college classes in human factors, ethics, and practice design methodologies that mitigate these known problems. The issues all revolve around simply getting something to market rather than letting the engineers create what they want to create.

It was a breakdown in the logic of the man machine interface. The machine was set up to land and from what I read, by the time the pilot realized this and tried to initiate a go around, it was to late. As I said, the computer did exactly what it was instructed to do.

We certainly have come a long way to getting the man/machine stuff right with human factors analysis, training and CRM training.

The aircraft was not “set up to land”. That is one of the persistent myths of that crash. There is no logic in an airliner that forces a landing, as evidenced by the fact that pilots go around routinely.

The pilot flying had slowed down to an unsustainable airspeed while still descending close to the ground. There was simply not enough energy to recover. Alpha protection kept the angle of attack below critical, but without enough energy the aircraft had to descend. A further pitch up would have led to a stall.

I wrote the below years ago on this board, based on the accurate account in the first “Air Disaster” book.


A320 at Mulhouse was an in service AF aircraft doing a flyby at an airshow. So what went wrong?

Some preliminaries:
- Pre flight briefing was not held.
- Flyby altitude was below French regulatory minimums.
- Flyby altitude was below the level of surrounding obstacles (trees).
- Flyby runway was changed at last minute (when in sight of the airfield).

As the plane flew by, speed and altitude were still both decreasing. Engines were at or near idle. The flight control system was compensating by trimming up. But since the plane was just above stall speed, once it reached the max angle of attack the only way not to stall was to decrease the pitch angle and thus descend.

By the time the pilot saw the approaching trees, it was already too late. No amount of thrust would have allowed the plane to clear the trees. On the video you clearly hear the engines spooling up just in time to become woodchippers.

The pilot flying had overconfidence in the plane's control system allowing him to perform a dramatic low speed flyby. In fact, without the control system and envelope protection, the plane might have stalled (it never did). But not stalling is not enough. You also have to avoid hitting the ground.

Nothing was wrong with the plane, and all components performed according to (or beyond) specs. Airbi may have very cool envelope protection systems, but they can't break the laws of physics.

If the pilot had tried the same stunt in a 737 or any other plane, there would still have been a crash.

Old pilot saying: "Use your superior skills to keep yourself out of situations where you might need to use them".

I really appreciate the correction. Some of the stuff I read did allude to exactly what you referred to. On any subject the only thing worse than no information is wrong information so thanks again.

stratclub wrote:

Starlionblue wrote:

stratclub wrote:

It was a breakdown in the logic of the man machine interface. The machine was set up to land and from what I read, by the time the pilot realized this and tried to initiate a go around, it was to late. As I said, the computer did exactly what it was instructed to do.

We certainly have come a long way to getting the man/machine stuff right with human factors analysis, training and CRM training.

The aircraft was not “set up to land”. That is one of the persistent myths of that crash. There is no logic in an airliner that forces a landing, as evidenced by the fact that pilots go around routinely.

The pilot flying had slowed down to an unsustainable airspeed while still descending close to the ground. There was simply not enough energy to recover. Alpha protection kept the angle of attack below critical, but without enough energy the aircraft had to descend. A further pitch up would have led to a stall.

I wrote the below years ago on this board, based on the accurate account in the first “Air Disaster” book.


A320 at Mulhouse was an in service AF aircraft doing a flyby at an airshow. So what went wrong?

Some preliminaries:
- Pre flight briefing was not held.
- Flyby altitude was below French regulatory minimums.
- Flyby altitude was below the level of surrounding obstacles (trees).
- Flyby runway was changed at last minute (when in sight of the airfield).

As the plane flew by, speed and altitude were still both decreasing. Engines were at or near idle. The flight control system was compensating by trimming up. But since the plane was just above stall speed, once it reached the max angle of attack the only way not to stall was to decrease the pitch angle and thus descend.

By the time the pilot saw the approaching trees, it was already too late. No amount of thrust would have allowed the plane to clear the trees. On the video you clearly hear the engines spooling up just in time to become woodchippers.

The pilot flying had overconfidence in the plane's control system allowing him to perform a dramatic low speed flyby. In fact, without the control system and envelope protection, the plane might have stalled (it never did). But not stalling is not enough. You also have to avoid hitting the ground.

Nothing was wrong with the plane, and all components performed according to (or beyond) specs. Airbi may have very cool envelope protection systems, but they can't break the laws of physics.

If the pilot had tried the same stunt in a 737 or any other plane, there would still have been a crash.

Old pilot saying: "Use your superior skills to keep yourself out of situations where you might need to use them".

I really appreciate the correction. Some of the stuff I read did allude to exactly what you referred to. On any subject the only thing worse than no information is wrong information so thanks again.

No worries.

It was a very interesting accident in many ways. I highly recommend the Air Disaster books for more info about this one and others. https://www.amazon.com/Air-Disaster-Vol-MacArthur-Job/dp/1875671110

To throw a little bit of Jet-A into the fire regarding foreign maintenance:
a recent Air Astana aerobatic flight occurred after a C-check at Portuguese MRO, probably MRO OGMA (at least they were working with Air Astana's 190s in 2017)
OGMA appears to be a primarily defense contractor doing overhauls of military planes (including Embraer), located in a NATO country...

Apprentice wrote:

Sorry, also disagree with Mr. Greendot about preventive Maintenance..
Most of a/c follow a Maintenance Program, proposed by Aircraft Maintenance and approved by regulator (FAA),
In the states, a Maintenance Program for. Commercial Aircraft, consist in:

For Line Maintenance:
Post Flight Checks
Daily Checks
Weekly Checks
Preflight Checks

Heavy Maintenance
Check 1, 2,
Check 3 (called also Capital Check)

This checks are stablished by Manufacturer and includes System Test of all systems
In the Checks, Regulators also include some check and test related to incidents or accidents already happening in regulator’s control area or in other country, when respective regulator issue report about.
System, made by humans, is propense to human errors, but it’s corrected on a regular base.
It’s easy to blame the system, the regulator, the Pilots or maintenance, what it’s not, is to check what had failed in the system and what measures should been taken to correct this situation.

Rgds

I'm aware of all those checks but we pilots have a bit of a joke about flying an airplane that came out of any of those checks... basically be ready for everything not working. I've seen fire suppression systems miswired to the wrong engine! That something is FAA approved means nothing. These are just bureaucrats mostly without any technical knowledge. I see "FAA approved" stuff all the time which needs improvement.

I don't see what you point was in mentioning routine maintenance schedules...

Apprentice wrote:

Disagree,
Although some time, in some companies, there are “pencil whipped complaints, the way to combat this is not reducing MEL time or MEL qty. To get a needed part and to replace it and test, it’s a time consuming, no easy task, that many times requieres more than 10 days ( for more common “C” MEL’s items)
Operational Test- OK is a normal answer as per AMM, when You had made a test following all requiered steps, and system perform as it should, as per AMM, problem is to follow all proccedures, including, selecting between operational or systems test, as indicated by Manufacturer Trouble/Shooting Guide.

Many, many times, after a system reset vía pulling / pushing related CB, You are not able to repit failure. Them you have two options, as identified by AMM, make a deeper System Test t/s or, after proper system test w/o incidences, close the item.

What really makes differences is when QI or Specialized personnel check all Log Book answer’s daily, and call Mx personnel when any answer deviated from proper proccedures .
In one company I work for, to avoid discussions, when a complaint reiterated, a Mx QI was ó/b, on next flight, to check system behavior. Believe it or not. Many repetitive complaints were closed after that....

Rgds

So you're giving rationalization of why the current system is inadequate. Ok... problem identified.

By reducing the wiggle room for operators, they will be forced to require manufacturers to increase system reliability.

MassAppeal wrote:

When was the last time we had a fatal accident due to some sort of major overseas (or domestic) maintenance failure?

In the USA? It's been a long time, I think primarily because many onboard systems report system failures to their Maintenance Control on both Airbus and Boeing airplanes. I worked the Airbus Desk at United for over 10 years in Maintenance Control and the systems reporting on Airbus was stellar. Boeing fault reporting got so much better with the B767-300, the B757-200 and 300, the B777-200, and even the B747-400, and of course the B787, there are plenty of differences between the codes but once you know how to break them down? You can pinpoint recurring and Chronic Problems that even the Flight crew cannot readily see. And the airplanes will alert you when the problem becomes a real problem,

greendot wrote:

I worry about this and about maintenance in general. The biggest routine issue is that even if something is written up, the deviation is frequently written off as "cannot duplicate". We, in the front office, need far better self discipline to document every and any deviation. Sometimes pilots do not document problems because they go away on their own or because some pilots perform actions to make them go away. Pilots need to be scrutinized about writing up every single defect, whether it goes away or not. Newer airplanes show errors for a reason but pilots frequently dismiss it as a nuisance. My take on it is that if it is a known sporadic thing, either get the manufacturer to rewrite the manual or redesign the system. Dont bet on luck and statistics to make the case NOT to fix it. Furthermore, anything written up should never be dismissed as Cannot Duplicate. Lastly, the FAA needs to greatly decrease the amount of time something can be on an MEL and limit the number of consecutive times the write up occurs. The MEL list needs to be made smaller so that airlines don't fly with entire book length quantities of MEL writeups. The FAA is far too tolerant of airlines. I think this greatly impacts our safety. Of course, the FAA won't ever say anything is a safety issue unless someone dies because the rules are written to always blame the pilot.

th
While I generally agree with your Premise? I DO disagree about the MEL. It's not the MEL Time limits, It's the chronic monitoring of HOW many times the SAME item has had to be deferred and the frequency OF said deferrals. Dismissing something as a nuisance? Is a call to Problems.
Some airlines need to react sooner when an item becomes Chronic. I spent most of my overnight shifts in Maintenance Control monitoring and documenting Chronic Problems, Then installing directives for the system to get fixed. And? I had the authority to ground an airplane if Necessary.
the Operations guys might not have liked it, But since they didn't have an Airframe and Powerplant License? What were they going to say?
Since I took my days off? Sometimes I would return to work to find a recurring problem and Address it . we had a rule of 3 times in 10 days before reporting the problem as chronic. when it got down to every day or subsequent departures?? Then I could and Would Ground the airplane.
Especially since it was a "big deal" to Override a maintenance controller, Not every airline would or could do that? But that's why The FAA's fault reporting was also a geat source of information for chronic problems as well. On quite a few occasions I got calls from the FAA for fault reports and how or if we'd seen the problem, They also consulted with our system Engineers as resources. The same way we worked together. The aim was to get as much information as you could before going to Boeing or Airbus. To have all of your "ducks in a row" because they can also download information from the aircraft to see what we See..

strfyr51 wrote:

greendot wrote:

I worry about this and about maintenance in general. The biggest routine issue is that even if something is written up, the deviation is frequently written off as "cannot duplicate". We, in the front office, need far better self discipline to document every and any deviation. Sometimes pilots do not document problems because they go away on their own or because some pilots perform actions to make them go away. Pilots need to be scrutinized about writing up every single defect, whether it goes away or not. Newer airplanes show errors for a reason but pilots frequently dismiss it as a nuisance. My take on it is that if it is a known sporadic thing, either get the manufacturer to rewrite the manual or redesign the system. Dont bet on luck and statistics to make the case NOT to fix it. Furthermore, anything written up should never be dismissed as Cannot Duplicate. Lastly, the FAA needs to greatly decrease the amount of time something can be on an MEL and limit the number of consecutive times the write up occurs. The MEL list needs to be made smaller so that airlines don't fly with entire book length quantities of MEL writeups. The FAA is far too tolerant of airlines. I think this greatly impacts our safety. Of course, the FAA won't ever say anything is a safety issue unless someone dies because the rules are written to always blame the pilot.

th
While I generally agree with your Premise? I DO disagree about the MEL. It's not the MEL Time limits, It's the chronic monitoring of HOW many times the SAME item has had to be deferred and the frequency OF said deferrals. Dismissing something as a nuisance? Is a call to Problems.
Some airlines need to react sooner when an item becomes Chronic. I spent most of my overnight shifts in Maintenance Control monitoring and documenting Chronic Problems, Then installing directives for the system to get fixed. And? I had the authority to ground an airplane if Necessary.
the Operations guys might not have liked it, But since they didn't have an Airframe and Powerplant License? What were they going to say?
Since I took my days off? Sometimes I would return to work to find a recurring problem and Address it . we had a rule of 3 times in 10 days before reporting the problem as chronic. when it got down to every day or subsequent departures?? Then I could and Would Ground the airplane.
Especially since it was a "big deal" to Override a maintenance controller, Not every airline would or could do that? But that's why The FAA's fault reporting was also a geat source of information for chronic problems as well. On quite a few occasions I got calls from the FAA for fault reports and how or if we'd seen the problem, They also consulted with our system Engineers as resources. The same way we worked together. The aim was to get as much information as you could before going to Boeing or Airbus. To have all of your "ducks in a row" because they can also download information from the aircraft to see what we See..

The problem is that things get to the point of being "chronic" in the first place. Causes must always be determined, not written off if they cannot be reproduced. I am well versed in how current maintenance practices work but I'm saying that it's insufficient. For example, I've seen MX guys "fix" a non-testing fire detection system by continually pressing the test until it works. The MX manual allowed the mx guy full latitude to considerate functional or not and even he was confused as to how he was to verify 'proper operation' in a vaguely written MEL mx procedure. This whole idea of spotting trends needs to be eliminated from the way mx is done.

Hi: Back to Quotte “This is from the LionAir crash and it makes my point...”
https://theaircurrent.com/wp-content/up ... p-hr-2.jpg” unquotte.
The answers from Maintenance are the ones that those reports deserves. Incompletes, without FM codes, no explained for instance, what any altimeters reads. It’s like “Discrepancies in altimeters... please have a/c ready for next flight in x time”
Some crew are very reluctant to use Fault Reports Manual’s codes, and that way, Mx will have to improve, instead of follow a Boeing’s clear stablished path:Fault Report Manual’s Code - Fault Isolation Manual, Action, Aircraft Maintenance Manual Ops or System Test

Completely agree with You: causes must always be determined, but this is a collective job: proper complaints, with all references, proper actions and answer and proper inspection. Lately You will not find neither proper complaints nor inspections or QA.I have found even MCC that when asked about, answered that: ...”they are avionics, they should know what to do it....”
Let’s start with next complaint entry, clear, with reference code and all available data,.. You will see that if complaint start, “Code 34-11-25 a”, for instance, Mx will reply more less: “Troubleshooting performed IAW FIM Code 34-11-25-a... and that way You will have the certitude that for a Fault that You observe and verified in a FM, Mx had carry the exact task, indicated by Boeing in a FIM.

Rgds

I think the biggest issue in the US is finding capable new blood to replace the aging AMT ranks. Smart kids don't want to go into this field as much as past generations. You have to go to college to find a 'real' job. Not some two year tech school or the military. Then there is the I don't want to work in a field where I would get dirty. You want me to crawl in there? Plus there is the working off shifts.

So we are left with those that really don't want to be here. They got talked into this field, sometimes by a judge. At the first sign of being forced to work midnights they cry hardship and get whisked off to some shop that their seniority can't hold, or they quit. Then there is the drug test issue. So many can't pass. That issue will only get worse as more states legalize pot. I'm personally for the legalization, but due to my profession can not participate. So many just can't get on board that those are the rules.

Dalmd88 wrote:

I think the biggest issue in the US is finding capable new blood to replace the aging AMT ranks. Smart kids don't want to go into this field as much as past generations. You have to go to college to find a 'real' job. Not some two year tech school or the military. Then there is the I don't want to work in a field where I would get dirty. You want me to crawl in there? Plus there is the working off shifts.

So we are left with those that really don't want to be here. They got talked into this field, sometimes by a judge. At the first sign of being forced to work midnights they cry hardship and get whisked off to some shop that their seniority can't hold, or they quit. Then there is the drug test issue. So many can't pass. That issue will only get worse as more states legalize pot. I'm personally for the legalization, but due to my profession can not participate. So many just can't get on board that those are the rules.

You are correct at my carrier we have a AMT trainee program taking kids right out of A@P school. I have met a few and I am not impressed they are horrible to be honest I know I am lazy but I wasn't that way 32 years ago when I go into this biz. These kids don't want to do jack and are totally not interested in anything. I hate to paint them all with the same brush but it applies to most of the ones I have seen. I am glad I am close to getting out and I don't fly much anymore thankfully!!

greendot wrote:

strfyr51 wrote:

greendot wrote:

I worry about this and about maintenance in general. The biggest routine issue is that even if something is written up, the deviation is frequently written off as "cannot duplicate". We, in the front office, need far better self discipline to document every and any deviation. Sometimes pilots do not document problems because they go away on their own or because some pilots perform actions to make them go away. Pilots need to be scrutinized about writing up every single defect, whether it goes away or not. Newer airplanes show errors for a reason but pilots frequently dismiss it as a nuisance. My take on it is that if it is a known sporadic thing, either get the manufacturer to rewrite the manual or redesign the system. Dont bet on luck and statistics to make the case NOT to fix it. Furthermore, anything written up should never be dismissed as Cannot Duplicate. Lastly, the FAA needs to greatly decrease the amount of time something can be on an MEL and limit the number of consecutive times the write up occurs. The MEL list needs to be made smaller so that airlines don't fly with entire book length quantities of MEL writeups. The FAA is far too tolerant of airlines. I think this greatly impacts our safety. Of course, the FAA won't ever say anything is a safety issue unless someone dies because the rules are written to always blame the pilot.

th
While I generally agree with your Premise? I DO disagree about the MEL. It's not the MEL Time limits, It's the chronic monitoring of HOW many times the SAME item has had to be deferred and the frequency OF said deferrals. Dismissing something as a nuisance? Is a call to Problems.
Some airlines need to react sooner when an item becomes Chronic. I spent most of my overnight shifts in Maintenance Control monitoring and documenting Chronic Problems, Then installing directives for the system to get fixed. And? I had the authority to ground an airplane if Necessary.
the Operations guys might not have liked it, But since they didn't have an Airframe and Powerplant License? What were they going to say?
Since I took my days off? Sometimes I would return to work to find a recurring problem and Address it . we had a rule of 3 times in 10 days before reporting the problem as chronic. when it got down to every day or subsequent departures?? Then I could and Would Ground the airplane.
Especially since it was a "big deal" to Override a maintenance controller, Not every airline would or could do that? But that's why The FAA's fault reporting was also a geat source of information for chronic problems as well. On quite a few occasions I got calls from the FAA for fault reports and how or if we'd seen the problem, They also consulted with our system Engineers as resources. The same way we worked together. The aim was to get as much information as you could before going to Boeing or Airbus. To have all of your "ducks in a row" because they can also download information from the aircraft to see what we See..

The problem is that things get to the point of being "chronic" in the first place. Causes must always be determined, not written off if they cannot be reproduced. I am well versed in how current maintenance practices work but I'm saying that it's insufficient. For example, I've seen MX guys "fix" a non-testing fire detection system by continually pressing the test until it works. The MX manual allowed the mx guy full latitude to considerate functional or not and even he was confused as to how he was to verify 'proper operation' in a vaguely written MEL mx procedure. This whole idea of spotting trends needs to be eliminated from the way mx is done.

I see your point totally Greendot. It really is about the money. Airplanes that sit do not make money so the airlines will get away with what they can to keep them flying. Now some carriers are worse than others. When I was a mechanic I was on call for Valujet and we went to work on their airplanes a few times but then us mechanics banded together and refused to do on call MX for Valujet because their airplanes were so poorly maintained. I don't know if it will get to the point you want it too but you can tell the difference.

Dalmd88 wrote:

I think the biggest issue in the US is finding capable new blood to replace the aging AMT ranks. Smart kids don't want to go into this field as much as past generations. You have to go to college to find a 'real' job. Not some two year tech school or the military. Then there is the I don't want to work in a field where I would get dirty. You want me to crawl in there? Plus there is the working off shifts.

Plus the pay sucks. I worked on planes in the Marines, didn’t graduate college, and now I work in a very lucrative non-aviation field that involves getting dirty and working crazy hours and going to random places and as much as I like planes, I’d rather make the money and have a stable healthcare plan.

As far as the rest of your stuff that I didn’t quote, get over yourself.

Dalmd88 wrote:

I think the biggest issue in the US is finding capable new blood to replace the aging AMT ranks. Smart kids don't want to go into this field as much as past generations. You have to go to college to find a 'real' job. Not some two year tech school or the military. Then there is the I don't want to work in a field where I would get dirty. You want me to crawl in there? Plus there is the working off shifts.

So we are left with those that really don't want to be here. They got talked into this field, sometimes by a judge. At the first sign of being forced to work midnights they cry hardship and get whisked off to some shop that their seniority can't hold, or they quit. Then there is the drug test issue. So many can't pass. That issue will only get worse as more states legalize pot. I'm personally for the legalization, but due to my profession can not participate. So many just can't get on board that those are the rules.

all that you say may have some fact in it. BUT!! Anyone who can't work for $35.00HR after 6 months? Really didn't want to work anyway did they?? I worked hand in hand with Teachers who were aircraft mechanics. We had 2 college Aviation instructors who worked in system Engineering helping Troubleshoot. So it's not just a grease monkey job. Nobody can just "turn a wrench" on today's airplanes.. And in my 33 years in the industry? I've found it takes 6-10 years before you can really understand every system you come into contact with. I was a US NAVY Flight engineer before working at United. I had over 3200 Hours flying. and My Training record alone got my college degree an upper hand when I went for my degree when I had to go back to school while at United. There are very few guys who do Nothing. And even the shop guys might be Engineer caliber. Not everybody wants or needs to work the "Line" at an airline. And qualifications or Education? Has Very little to do with it..

Zeke2517 wrote:

Dalmd88 wrote:

I think the biggest issue in the US is finding capable new blood to replace the aging AMT ranks. Smart kids don't want to go into this field as much as past generations. You have to go to college to find a 'real' job. Not some two year tech school or the military. Then there is the I don't want to work in a field where I would get dirty. You want me to crawl in there? Plus there is the working off shifts.

Plus the pay sucks. I worked on planes in the Marines, didn’t graduate college, and now I work in a very lucrative non-aviation field that involves getting dirty and working crazy hours and going to random places and as much as I like planes, I’d rather make the money and have a stable healthcare plan.

As far as the rest of your stuff that I didn’t quote, get over yourself.

I don't think the pay really sucks anymore at US majors. Starting is mid $20 hr and top out is over $50/ hr. Decent medical benefits. Just the hours and shift work suck. Kids just don't want to do this kind of work anymore. They just don't play around trying to fix things as kids like past generations did. There am I 'over myself' now? Never understated that phrase. Just stated my opinion on the kids I see coming into this field. Not a reason for a lame personal attack.

Hi, Starting pay may be below $20 and $50 is not to the reach of every one.
There are a lot of Mechanics doing 2 jobs to live..
In our case, IMHO, problem are Bad complaining procedure and Lack of Comm between pilots and Mechs.

Rgds