It’s time for another edition of Ask Cranky and today, it provides a good excuse to write more about Boeing’s 737 MAX problems. No small talk needed… let’s get into this.
Everyone is bashing Boeing right now, but is there any real difference in the growing pains of the 737 Max vs the computer glitches that caused the pitch down incidents on the Quantas [sic] A330s?
I love a question that makes me do research. I vaguely remembered a Qantas incident on an A330, but I didn’t recall it being all that similar. It turns out, it was more similar than I thought.
Of course, the problem on the 737 MAX is unique. The MCAS automatically pushes the nose down on the airplane when the angle of attack is too high, and no other airplane has an MCAS. Why not? Well, the 737 has been stretched and updated so many times that it’s like a Barbie doll. It can’t stand on its own. The MCAS keeps it flying straight. The A330 has no such issue, but that doesn’t mean there couldn’t have been an example of an airplane forcing its own nose down for other reasons. That happened to a Qantas aircraft back in 2008.
The A330 That Forced Its Own Nose Down
Qantas flight 72 was operated by an A330-300. It left Singapore on October 7, 2008 on a flight to Perth. This airplane had no problem on takeoff as the MAX aircraft did. Instead, it was cruising along at 37,000 feet near the coast of Western Australia when things started to go wrong.
The A330 has three air data inertial reference units (ADIRUs). Just what is an ADIRU? I’ll gladly let Wikipedia explain.
An air data inertial reference unit (ADIRU) is a key component of the integrated air data inertial reference system (ADIRS), which supplies air data (airspeed, angle of attack and altitude) and inertial reference (position and attitude) information to the pilots’ electronic flight instrument system displays as well as other systems on the aircraft such as the engines, autopilot, aircraft flight control system and landing gear systems.
Got that? They basically provide all the most important flight data to the computers onboard so they can make the right decisions. So, yeah, it’s pretty important to have them functioning. On this particular flight, one of the three ADIRUs malfunctioned and started spewing spikes of bad data. Even though the other two ADIRUs were working just fine, this data caused the airplane to become confused. A number of error alerts sounded, the autopilot disconnected, and the pilots took over manual control. Soon after, the airplane’s systems pointed the nose down.
This was no gentle descent. The airplane reached over an 8 degree pitch angle quite rapidly and people were thrown about the cabin. There were more than 100 injuries with 12 being serious. (And yes, the majority of those fools weren’t wearing seatbelts.)
The pilots tried to counteract the airplane, but for 2 seconds, they could not. Then the airplane responded and they went back to normal flight. Two minutes later, however, a less severe pitch down command came in. They recovered once again and diverted immediately. The airplane landed without further incident.
Comparing the A330 and the 737 MAX Incidents
So, here we have an airplane that tried to push its own nose down multiple times after seeing bad data indicating that the angle of attack was too high. Does this not sound a whole lot like the 737 MAX problem that we’re dealing with today? It does, but obviously the outcomes were different. The A330 was never grounded, and as far as I know, there wasn’t a huge negative public reaction to the airplane. What was different?
Let’s start with the obvious. Qantas 72 landed safely and nobody died. Regulatory agencies have an unfortunate habit of being reactive, but if nobody dies, there’s a whole lot less pressure to react at all. But there was much more to it than that.
Unlike the 737 MAX, the A330 family was far from being a brand new airplane. The first A330 was delivered more than 15 year before this incident. In the 28 million plus flight hours between then and 2008, nothing along the same lines had ever happened.
Sure, in 2006, this same exact airplane had a similar ADIRU fault problem off the coast of Western Australia. But in that case, the airplane didn’t adjust flight controls at all, so it was a non-event from a passenger perspective. Another Qantas A330 off Western Australia had a similar issue at the end of 2008 (what the heck is it with Western Australia?!) that again resulted in no change to flight controls.
Even if those had resulted in a nose-down situation, it wouldn’t have been nearly as dire as in the case of the MAX. All of these happened at cruise altitude where it’s a whole lot easier to recover from any upset. Altitude is your friend.
Lastly, this wasn’t some newfangled system on the A330 that pilots weren’t even told about. This was just a malfunction of systems that the pilots were familiar with. The pilots had been trained on how to handle these situations, and in the case of Qantas 72, they did.
You might be wondering, “uh, what about Air France 447 in 2009?” That is the only A330 accident with a major loss of life (except for Afriqiyah Airways putting an airplane into the ground in Libya a year later), and it did involve instrument issues. But that was very different.
That flight from Brazil to Paris crashed into the Atlantic Ocean. This wasn’t a case of faulty data due to computer issues. This was a case of faulty data due to frozen pitot tubes. That still shouldn’t have brought the airplane down, however. The pilots just failed to fly the airplane.
Why the MAX Was Grounded and the A330 Was Not
Unlike the A330, the 737 MAX is a brand new airplane, and it went down right on takeoff soon after being introduced into service. We’ve now had two accidents within a few short months, at least one of which is thanks to the new MCAS system which wasn’t properly explained to pilots. There is no history of safe operation of the airplane to help quell fears.
The 737 MAX delivery schedule is far more aggressive than the A330’s ever was. There are a lot of these airplanes going into service every month… at least there were until the airplane was grounded. So why did the MAX get grounded? There’s no doubt some of it is simply a reaction to all the dead bodies. Pardon the bluntness, but it’s the unfortunate way things work.
But looking at the way this airplane was designed and how training was pushed out, it seems to have been done in a rush and could, to put it mildly, have been thought through better. By grounding the airplanes, the authorities are effectively hitting the pause button to let training and automation catch up to the airplane.
You’re probably wondering what happened in the wake of Qantas 72. Well, first, it was noted that everyone should keep their seatbelts on when seated. That’s obvious.
But the algorithm that told the airplane how to react to the rare “multiple data spike” incident on the aircraft was redesigned to provide more protection. It shouldn’t happen again. Further, there were new training modules and procedures put into place to help make it safer for everyone. I’d recommend a read through the very thorough final report from the Australian Transport Safety Bureau. It’s a good (but LONG) read.
The odd thing is that I expect the resolution to the 737 MAX issue will be similar. Improve training, make the software safer, and then everyone should be satisfied. That’s not to suggest that what happened to the A330 was the equivalent of what is happening to the MAX. The A330 was a proven, safe airplane for many years before this freak incident occurred. The MAX has a long way to go before it gets to that point.
Wouah : 3 posts on the 737 max in less than a week (or so it seems) !…
Well researched, and the conclusion is there : new plane vs established safety record !…
It all boils down to the pushing of the type certificate to (past) its limits …
Well, the 737 has been stretched and updated so many times that it’s like a Barbie doll. It can’t stand on its own. The MCAS keeps it flying straight””””
To me, this sounds like what was done to the DC-8 and DC-9..Is this a fair comparison???
floydk – I can’t say I know exactly what had to be done mechanically when those airplanes were stretched, but certainly both of those airplanes were stretched far beyond the initial design plan. It shouldn’t be a surprise either since it’s that McDD mentality that came over to Boeing after the merger.
The DC-8 stretches didn’t involve changing the relationship between the position of the engines and the wing as radically as the 737MAX does. And the tail-mounted engines on the DC-9 and its successors means engine changes don’t affect the wings at all.
As I understand it, Boeing could have avoided this problem by just making the landing gear longer so the engines didn’t have to be pushed forward so much.
> What is it with Western Australia?
Some people (I’m not one of them) blame this…
I’d got more along the lines that most aircraft from anywhere in Australia fly over WA to get to Asia. Then again, maybe it was a malfunction in the chemtrails manufacturing device ?
AF447 went down due to crew resource management issue. The fact that the CVR and FDR were two miles under water and it took years to recover them did give me pause when flying on that same type during that time frame because quite frankly, we didn’t know what happened. Now I’ve been in favor of the MAX grounding but do say let the investigators do their jobs and everything before their reports is nothing more than “monday morning quarterbacking.”
And, just to stir the pot, if AF447 was a Boeing it would’ve never gone in the drink. With a traditional yoke setup the acting captain in the left seat would’ve very clearly notice the PIC was pulling back on the stick as they are tied to one another. The Airbus sidestick, not so much.
I get your point but…is it true the stall warning was blaring at them for four minutes and not one of the three pilots ever thought to stick the nose down? Which would mean that the yoke vs sidestick debate is kinda irrelevant.
If I remember correctly, the captain was trying to put the nose down, but the first officer was pulling up and the fly-by-wire system just took the sum of their inputs without giving any sensory feedback. So by the time they figured out what was actually happening, they were too close to the ground to do anything about it.
Coincidentally (or not), the Smithsonian Channel’s “Air Disasters” series has been running a show in the Qantas 72 flight this month.
In this day of ‘Fly by Wire’ we all have to be concerned about GIGO. As to AF 477, I’ve always been skeptical that all the Pitot Tubes malfunctioned simultaneously. Just saying
I believe Delta also has had a couple similar faults with their A330s.
I’ve read a couple of stories that mention the that American Airlines’ MAXs have a couple of extra safety features that may not have been on the affected aircraft. I guess safety is more important to some airlines than the size of the bathrooms or the lack of seat back in-flight entertainment.
I have a feeling Boeing will make these added features “standard” in the wake of these accidents and grounding of the plane.
I’ve read that’s happening.
Or you could argue that safety shouldn’t be an optional extra to be added, much like mood lighting in the cabin…
I agree. Safety shouldn’t be optional.
I suggest that everyone visit Leeham News & Analysis, and the post linked below by Bjorn Fehrm. He’s a retired Swedish military pilot with thousands of hours in high-performance jets. Over the last couple years, he’s been a contributor to LN&A (formerly Leeham News & Comment) and he’s posted countless fascinating articles about aircraft performance, dynamics and the trade-offs manufacturers have to make when it comes to designing commercial aircraft.
To paraphrase, he concludes that the MCAS can, in certain instances, result in a phenomenon called “blowback” that prevents the pilots from being able to control the attitude of the plane despite using maximum force on the yoke. His analysis is supported by the data available from both the fatal Lion Air flight, the flight of the same plane the day before that was saved by the actions of a deadheading pilot, what’s known to date about the Ethiopian flight, as well as information from a type-rated 737 pilot who spoke to the Seattle Times about the issue.
With any luck, the MAX’s safety problems will be resolved soon.
Sensor data validation needs to be a big component of the next generation of airliners. I can’t imagine any serious push towards pilotless aircraft without absolute data integrity going in to the flight computers. The “best 2 out of 3” model isn’t going to cut it, and certainly not the MAX “put a critical system on a single sensor”.
For a comparable issue other than the A330, consider the Turkish 737 in AMS. The auto throttle flare-mode power reduction got triggered early by a single false radar altimeter, and the pilots never pushed the power up.
there was a crash of an A330 during testing with 7 lives lost, I believe
I believe you’re referring to an Air New Zealand A320 test flight, not A330
One thing I have not seen very much reporting on is Boeing’s apparent change of design philosophy vis a vis Airbus for the Max 8
Airbus’s philosophy is that the engineers know best so they use automation to prevent a pilot from flying a plane beyond certain limits. This has led to pilots fighting against the autopilot for control of the aircraft.
Whereas Boeing has had the philosophy that the pilot should have ultimate control of the aircraft and can (if needed) fly the plane beyond the recommended limits of it.
This appears to be the first time that Boeing has deviated from their design philosophy.
If the Lion Air and Ethiopian pilots understood that to be the design philosophy, it is no wonder that they would have been surprised by something that happened automatically and that they didn’t understand and/or knew how to turn off.
Automation is a good thing….. until it doesn’t work as expected! :)
Keith – Yeah, though it’s hard to say this was a change in design philosophy by choice. It was really more by necessity because of how much they’ve done to stretch and tweak that airplane. I think they compromised on this because they had no better option. (They just didn’t do a particularly great job of designing that automation.)
But they don’t actually need the MCAS to fly the plane, they added it to ensure the same 737 type rating?
Yeah. I think these kinds of discussions will come up more and more as self driving cars come online as well. Commercial air travel has been exceptionally safe over the last decade, with tons of planes that incorporate a great deal of automation in their design. But when the engineering or code has a blind spot, people tend to be far more critical of an automation failure than a mechanical or driver/pilot error. Obviously, engineering needs to be tight and “smarter” than errors that a global fleet of pilots may be prone to make. And pilots need training that enables them to identify faulty data, understand the systems that support them, and fly the plane when systems fail.
Sounds like these corrections are being implemented for the Max 8. I’m optimistic that the right lessons will be learned at Boeing and future planes will be safer for it.
“Automation is a good thing… until it doesn’t work as expected!” Amen! IMHO some people (including engineers) are WAY too dependent on bits and bytes instead of human ingenuity and thorough training.
What seems clear is that Boeing is manufacturing an airplane that is fundamentally unstable, and obviously they feel cannot safely operate without mcas.
That blows my mind.
The 737 Max issues may be due to lack of pilot training or other cause but I think Boeing/the airlines will still have a lot of trouble convincing people to fly on these planes.
The knowledgeable people commenting on this blog are less than 1% of the flying public. Once this issue is resolved, and the MAX is flying again and a few weeks pass, the only thing that will matter is the price of the airfare. Believe it or not, the Boeing 737 MAX allows the airlines to charge less that other similar models of A/C, that’s why so many have been sold. The majority of the flying public have absolutely no idea what type or brand of A/C they are flying on. To most, a plane is a bus with wings.
Planes are so automated now that pilots don’t have to do much now except punch in some info and the planes systems take over….the A380 had a major problem, the A320 when it came out had issues but when the Max had 2 incidents, they must be grounded….Indonesia has had many issues over the years with crashes, Adam Air comes to mind, airlines landing in harsh story etc….flying needs to go back to more pilot control to keep their minds fresh instead of depending computation….this issue with the Max is also giving 2 airlines the excuse to cancel orders they cant afford or dont want….have other countries that fly the Max had series issues….
“when the Max had 2 incidents”
it was about the A320, not the A330.
The Qantas airbus had a new anti stall system installed and the data entered was mixed up (switched speed for altitude) if I remember correctly. They never did find why. The type of plane Boeing or Airbus and their safety tracks has nothing to do with this… It’s the system that seems to be doing the same crazy unexplained thing without any explanation…
There was an Airbus that crashed during a fly over at an air show in France. They blamed the pilot, but he claimed the plane was not responsive. “Air Disasters” on Smithsonian Channel covered this. Officials blamed for monkeying with data. On the Quantas flight mentioned, that same show ended that they never figured out the problem but that maybe one of the sensors was reporting altitude instead of attitude which sounds strange since it seems that data would be completely different i.e. 10,000ft vs 8 degrees. I am not an expert here though.
I read recently that the 767 has a similar system to the MAX 8.
The real question here is…
Should Boeing have attempted to push the B737 design into a Max! The 737 has been such an awesome aircraft but it’s design is done. Boeing has had years to create and intro a new model. If safety is at the heart of decisions then produce a new aircraft you are proud of Boeing!
If cost wasn’t an issue, and airlines were willing to pay for the additional R&D, retooling, testing etc… costs, then sure. But since they aren’t and this market has become an ever increasing game of bottom line then this is the result you end up with. This is especially true with twin engine platforms.
With that being said – The real fatal flaw in your comment is the idea that a (completely) new design is safer than a redesign. I’m not say that pushing an airframe up to (or past) it’s design limit is safe. But a new airframe has so many more unknown variables that make it much less safe.
Thanks for this article! It is very well written, informative, and to the point. I just saw an episode of “Air Disasters” on the Quantas 72 incident that was also well done, and did a good job of explaining what happened back then.
Thank you for this clear and well researched explanation. I have been scratching my head about the similarities and have not seen this talked about anywhere else. Kudos to you for taking the time to address it. Guess I have a new aeronautics website to follow, Cranky! I look forward to it.
The tone of the original question and several comments implies that people are desperately trying to defend Boeing by attacking Airbus. As an impartial commenter (I don’t live in the US or Europe), I see this as very partisan (saying that a US-made product must be better than a European one) and totally unnecessary. Both major aircraft companies have made good aircraft but teething problems seem to have been common. Boeing apparently made a serious error in the design of the Max which they are trying to fix. You may recall how the 787 had several serious teething issues (with fire) and there are still issues with 787 engines today for some airlines. Equally, there have been issues with some Airbus aircraft. Air travel is safe but not imperfect.
Very well said!
Boeing has not fixed the real problem. The mew engine size and wing placement is not being changed. All Boeing is doing is installing a computer work-around. So from my point of view. the aerodynamics will remain the same. Boeing needs to redesign the wing-engine placement or, even better, design a new modern airframe.
The entire fleet of 737max shud b strictly done scrap… dismantle the ENTIRE manufactured fleets, n tat Donkey MCAS APP , MUST B OVERTHROWN IN DUSTBIN..
THIS PEICE OF SHIT # SO CALLED 737MAXXY## IS NOT EVEN WORTH COMPARING TO BARBIE DOLL..
COZ BARBIE DOLLS GIVES NO NIGHTMARES.. GIVES PEACE, WHN KIDS HOLD THEM …
AND THIS 737MAXXY IS NOTHING BUT FREAK FREAK NIGHTMARE..
ITS BETTER SCRAP THIS MODEL, N TO MOVE ON, MANUFACTURING FOCUSSING ON OTHER SUCCESSFUL MODELS OF BOEING, WAT BOEING HAD SO FAR VRY TRULY ATTRIBUTED TOWARDS SAFETY N QUALITY N UPRIGHTNESS IN AVIATION INDUSTRY..
INSTEAD OF FOCUSING REFOCUSSING N DOCTOR UP, NEW LOGICS, TO PROVE RELIABILITY N SAFETY IN THIS VRY #MAXXY#, ITS BETTER TO SCRAP THE ENTIRE MAX FLEET, DISMANTLE THE EVRY COMPONENTS N PARTS OF EACH FLEET, N USE THEM , BY CUMING UP WTZ NEW MODEL, …
N FOR HEAVEN SAKE, MCAS TO B ABANDONED FOR ETERNITY, ( AS ZORS WAS ABANDONED FROM PLANET KRYPTON IN THE MOVIE SUPERMAN 1))
why doesnt Boeing go back to building the -800NG etc and give customer’s the option on the original or Max -800 etc…..things are to automated and pilots don’t have to think like they use to when flying….planes can basically fly themselves so do away with all this automated stuff and build what Boeing did best….if they want the Max, keep testing, taking time to make sure everything is working properly, the NG’s are fuel efficient and are proven planes…
WHY DOES EVERY GIVE AIRBUS A PASS ? THREE OR TWO PLANES CRASHED BECAUSE THE SPETO TUBES KEPT FREEZING AND ALSO WAS A CONTRIBUTING FACTOR. 447 A TEST FLIGHT ANS I THINK A THIRD.
I NEED PEOPLE TO UNDERSTAND THAT I AM REPLYING AND I CAN NOT SEE MY KEYS HALF THE TIME. I ALWAYS MAKE ERRORS.
I AM FED UP WITH ALL THE OUT CRY ABOUT BOEING AND NOTHING IS SAID ON THE CRASHES AND ISSUES CONCERNING AIRBUS. TWO OR THREE AIRBUS PLANES HAVE CRASHED BECAUSE OF THE SPETO TUBE ISSUES THAT AIRBUS KNEW ABOUT. EVERYONE GIVES THEM A PASS AND I WOULD LIKE TO KNOW WHY? . RECENTLY AIRBUS PLANES HAVE BURNED . THAT COMPUTER GLITCH ON FLIGHT 72 HAS NOT BEEN DIAGNOSED AND CAN STILL BE A ISSUE WITH THEIR PLANES. IT CAN HAPPEN AGAIN. JUST WATCH AIR DISASTERS AND GOOGLE IT. JUST BECAUSE THIS HAPPENED A WHILE AGO DOES NOT MAKE IT OK.
The A330 did have a similar problem which injured several people, some seriously. The reason the aircraft didn’t crash is that it occurred at a higher altitude, sufficient to allow the pilots to get it under control.
Your statement about the 737 MAX not being able to fly straight is pure baloney. It is essentially the same aircraft as the 737 NG, with more powerful engines and the notorious MCAS. Without MCAS, the MAX flies as well as the NG (which has an excellent safety record), but the handling feels different because of the greater takeoff thrust. The purpose of MCAS is to give the aircraft a similar feel to the NG during takeoff so the MAX will require the same type rating. The MCAS was botched, as we all know, but there is nothing wrong with the aircraft’s flight characteristics.
Lion Air dropped into the Pacific, AF447 into the Atlantic Ocean, that’s the only difference. At the Quantas Flight in 2008, passangers didn’t come to harm?! ( I would say not enough, but people got heavily injured).
MCAS and ADIRU ARE absolutely the same issue. No matter, if the trigger was a defect speed measurement device (pitot) or an angle of attack- sensor, both systems lead by ” correct” algorythms false data inputs to false automatic mecanic reactions of the plane, which actually killed people. The lies of the Airbus investigations in 2009/ 2012 , putting responsibilities on the pilots for economic reasons, made the deadly events of two 737 Max, even years later, possible to be repeated