Hi Scott , great job , a lot of info on a short video . I am a retired 25000 Hour Regional Airline Captain ! Most all my airplanes were cable driven controls . I was always directly connected to my Control surfaces . Now days new technology is taking the pilots out of the Loop ! And the thing that is sad for people in the air and on the ground , safety is moving to the engineers and in many cases have no Flying experience flying . They are trying to Phase out the pilots . We are in this New era and there are going to be accidents to make it all work . We are in If man were ment to fly era again ! I retired from aviation in 1997 at the ripe Old age of 47 and I haven’t flown since I drive my motor home everywhere. The Decline in aviation safety started in 1985 With deregulation . I left aviation because I didn’t think it was right for me To support unsafe industry ! My Dad Captain Howard Wirth a 747 Captain for TWA taught me at a very young age that True Safety is no accident! After becoming an A/P Mechanic and working On aircraft first, I thought I could make This theory work ! I new all my airplanes Very well and have a spotless record as an aviator ! But I was totally connected to all my aircraft and I trusted the maintenance done on them ! But when deregulation dropped in and cut pay to all the proficient mechanics , ATC controllers , and pilots they left And we were left with inexperience, it scared me . I left aviation for my guitar . Now days it’s almost like aircraft accidents is a form of entertainment ! I’m writing this because I’m living this Transition that I’m seeing ! To shorten Up this whole explanation in a few words In my era and my Dads era it’s was if there is a question we always did the safe thing ,and we’re not questioned ! Now days you better Not creat waves or you will get a black mark on your record ! My days were True Safety Is No Accident and You Make your Own Luck ! ( God Help All The Innocent People on the Ground and In the Air . Captain Mark H Wirth PS Now days if you climb on an aircraft It’s no one’s fault but your own if there’s a problem ! Your the pilot in command of Your own life !
We talk a lot about doing a “root cause analysis” in my field-medicine-when a serious incident occurs, but that language often locks everyone into the mindset that there is only ONE root cause, when in fact there may be several. We should really be doing “an analysis of root CAUSES” or even more broadly, “an analysis of causation.” I have read the NTSB reports on these two accidents several times, and so I find your analysis insightful: if no upset, then no radical rudder inputs, so no hardover, so no mishap. Great second look at cases that were settled (incompletely) long ago. Great work!
There is a ginormous elephant in the room here. The word ”error” implies culpability. That is a very serious implication in the roman law tradition of your culture. ”Error” is the derived from the latin word ”Errandus”: To go astray. How can you go astray when there isn’t even a defined Straight ’n narrow? It’s a bit like saying you got hit in the head by a meteorite because you stood in the wrong place. Factually correct as it may be still not contributing one bit… Still, awesome video and very important! Thank’s Scott!
Another great, informative, debrief Scott. Your prevailing theme of 'what happened', and 'what can be learned', is what makes these recaps so valuable, as always. Thanks, Scott
I have a strange connection to Flight 427 as did many Pittsburghers. That evening, I happened to be driving inbound on the Parkway West and less than 5 miles from the accident. However, I was ahead of it and only made aware when the emergency vehicles were Outbound. I also knew one of the investigators who was a physician and a US Air pilot who was assigned to investigate. At the time, I lived at a place called Lake Latonka and several of my neighbors were US Air pilots. Needless to say, it was a very sad place for quite some time. Thanks for the deeper dive much like Fate is the Hunter. A final note, for those of us who have experienced real wake turbulence in a light aircraft (me in 2020 runway 04 SRQ) flying a Cessna 162 Skycatcher, you remember cleaning your drawers out. I've since have declined more than one take-off clearance on intersecting runways with Commercial equipment.
The process of understanding the result of ANY mishap in ANY environment is exactly as you describe. I am not a pilot. I am a process development chemist who has experienced / investigated a few catastrophic events. As humans we reach for the simple answer. The simple answer is probably never the whole story. Personally I would not look for blame but rather what needs to be done to prevent it happening again. You can only accomplish that by knowing everything that happened. I’m probably not making complete sense. What I am trying to say is to resist the suction caused by the bullshit storm caused by the oversimplified conclusion. E.g. there was a spark and then a loud boom!
Much appreciated, Scott, and one of the reasons I subscribe. My instructors drilled into me that it's rarely ONE cause, it's always a catastrophic chain, which can encompass human and non-human (e.g. mechanical) errors. However, one of my own comments got eaten by YT which was the Airline Pilot's Union reply to the NTSB findings (and incorporated in the final report). It very much wanted to focus on the mechanical aspect in a maybe understandable effort to absolve "their" people.
Scott you are right on the money.... unfortunately we raising System Managers in the cockpit, instead of Stick and Rudder Pilots. If the airplane can't be flown by the Flight Director...and the situation hits the fan....You're in for a long day.
Scott, as usual, well researched. Not mentioned however, is that the B737 was certified with only one rudder PCU. During the design and certification phase, Boeing had to assure the FAA that reversal of the power control unit was not possible as there was no back up. Initially Boeing refused to accept that there had been a malfunction, this refusal occurred over a number of years while the aircraft continued to fly without corrective action. It was only after persistent follow up and testing by ALPA and industry advocates that Boeing relented and the PCU was redesigned. The airlines do minimal training in recovering from upsets. Most of the training is recognizing the "approach to a stall". Full stalls, as in light planes, are not conducted in air carrier aircraft. There is a dramatic difference in performance and recovery techniques depending on hi altitude incidents and low altitude deviations. The crossover speed between rudder effectiveness and aileron authority was not mentioned in the AFM. On the B727 and B707 there was additional roll authority when the flaps were extended through the use of supplemental inboard roll devices. You do an excellent job, no flame suit required!!
Back in the 80s and 90s I flew the 737-200, 300, and 400 for about 20 years before going on to the 757 and 767. Out of the three I felt the 200 was the best handling. To me it was like the 300 didn’t have enough wing for the job. The roll rate was not as good as the smaller 200. I also found that the 300 would be easier to upset in flight behind, not only larger aircraft, but also aircraft of equal size. One time while in trail of an MD-80 on final I felt a little shake and noticed the airspeed beginning to bleed off rapidly. We were at vref for flap 5 at the time. The plane began a slow roll to the right and I realized I was going to need more airspeed to bring it under control. Since I was on final and cleared for the approach I had no need to maintain altitude and was able to trade it for airspeed. I pushed the nose over a little while at the same time increasing engine power. It was almost like recovering from a stall entry. Another time something similar happened behind a Fokker F-100. It’s amazing that aircraft of that size, compared to you, can create such upsetting conditions. If you ever have to trade altitude for airspeed it’s easier to explain why later since you’re still around to do so.
Not many had a chance to fly the B737-100, but it had the best handling of the bunch. It was a "stick and rudder" airplane, no fancy electronics, navigation, just the old "steam gauges" and very simple SP77 autopilot. To me the Wiz Bang gadgets helped in efficiency, navigation, and reduced the workload at cruise, but they also took away an element of the pure fun factor of just flying and being in harmony with the aircraft. Scott, Juan, and Dan look to recover that basic simple joy of flying with their own private aircraft.
In the '90s I was on a flight with a co-worker from Sacramento to LA on a Southwest airlines 737 and we were on approach to LA and the airplane suddenly dipped the left wing straight down and then the right wing was straight down and then the plane righted itself. The pilot came on the intercom and apologized and explain that they were caught in a wake from a 747 that was ahead of them.
Hi Scott, I appreciate the research you've done and the conclusions you're drawing here - on the other hand the 737 PCU valve hardover/reversal story opens up several cans of worms for a reason. As others have commented, the B737 was certified with a single PCU valve which was designed to be redundant in and of itself, and ultimately it turned out that this was not sufficient. Rewinding to when Boeing was designing these valves in the '60s, the manufacture of these valves was subcontracted to Parker Hannifin based on tolerances specified by Boeing, and the valves were tested to the extent of the knowledge held at the time. As I understand it, the "thermal shock" scenario came from a discovery (late in the investigation) by a Boeing engineer that when the holes drilled for the transfer of hydraulic fluid were located towards the far end of the specified tolerances, the behaviour of the valve could be affected when certain conditions prevailed - thermal shock being one of them. It's important to note that the only PCU valve tested with the thermal shock scenario was that recovered from the wreck of USAir 427. The Eastwind aircraft was tested with the PCU in place on the ground, and if I recall correctly, it was discovered that the PCU could hardover and reverse without requiring the thermal shock condition to be present. The PCU valve from the wreck of UA585 was too badly damaged by impact for any dynamic testing to be performed on it - all they could do was disassemble it and check it was within tolerances (which it was). Furthermore, because the United B732 was such an old airframe, the FDR with which it was fitted only recorded very basic parameters; specifically it did not record rudder pedal input. This is one of the reasons that the initial NTSB report on UA585 was forced to conclude that there was insufficient evidence to determine probable cause, in spite of the fact that the departure from controlled fight was consistent with a rudder hardover. One particular can of worms regarding this story when it comes to USAir 427 in particular - and possibly a reason why the response regarding your views on pilot input received the pushback it did - is that Boeing initially went to considerable lengths to try and prove that the upset was *solely* caused by inappropriate handling on the part of the flight crew, and as I'm sure you're aware, that rarely goes down well among the piloting community. This was the second accident in which the B737 rudder was considered suspect, so it's hardly surprising that Boeing's efforts were regarded as a damage control effort, if not an attempt at a cover-up. You correctly state that the circumstances of UA585 were not conducive to the "thermal shock" scenario which was proven to have been a possible factor in USAir 427, but that does not take into account the fact that thermal shock was probably not the only condition in which the original PCU valve, when manufactured close to the tolerance limits, could behave abnormally. What allowed the NTSB to finally conclude that a rudder hardover and reversal was a likely factor in UA585 was the flightpath of the aircraft, which was consistent with that scenario, and regardless of the crew's handling of the situation, the PCU valve's design fell short of the redundancy requirements necessary for certification. Line pilots may not have been trained in the significance of crossover speeds, but the crossover speed would certainly have been considered as part of the certification, and because the PCU valve's flaws were not understood or taken into account at that time, the "safe" airspeeds determined at certification for those phases of flight did not take that into account. I absolutely agree with you on your point regarding the tendency to wish for a single "smoking gun" cause, and how focusing on that can be detrimental to a full understanding of all the factors involved in an accident. However, as anyone who has ever tried to ride one of those "funfair" bicycles where the steering is reversed can attest to, even if one is logically aware of the fact that the controls are reversed, the reality of experiencing a behaviour to which you have become accustomed for a significant time suddenly no longer making sense is a very serious problem. In other words, if a rudder pedal input resulted in a reversal and hardover even for a few seconds, the disorientation that would cause would certainly drastically hamper any attempt at recovery because you'd instinctively know that at least one of your controls has malfunctioned, but have no idea which one, nor would you know that the malfunction was transient. As such, I find it difficult to consider the inability to recover as an "error" on the part of the flight crew - even if the hardover/reversal was triggered by an unusual control input due to circumstance - because the aircraft not only behaved outside of the parameters which they were trained to expect, it behaved outside of the parameters for which it was certified. This is in turn significantly different from the situation regarding AA587, which to my mind was also a systemic set of failures - in that case it was a training programme (AARP) which did not take the differences between types in the fleet into account combined with a particular pilot who came out of that training with an inappropriate understanding of the risks (the training left him exceptionally nervous around wake encounters), which led to inappropriate application of the techniques (namely unusually aggressive handling of the rudder controls). A final systemic problem was that other flight crew noticed this nervousness and excessive use of rudder, but at no point was an attempt made to understand and remedy why this pilot was reacting and behaving the way he was. Ironically, one outcome of this accident was an effort on the part of some AA pilots to treat the A300's rudder controls as a single "smoking gun".
The AA program was called AAMP and the FOs flying in 587 in no way reflected the program. I went through the AAMP program and it was excellent and I believe it WAS type specific. The thread that is common in all three of these accidents is the crew did not fly the airplane first. They did not understand the need to reduce AOA. The AA program was talking about crossover speed shortly after 427. Rudder doublets: ‘reversals’ were not taught. I think you discount the ability of the crew to fly the plane. Review the comments of the USAir Flight Managers.
turricaned tc: Very well said and reasoned, it appears to me after re-reading much of the 427/585/517 NTSB accident report (AAR-99/01) that the author here has never actually read the entire document or understood its findings as his sequence of events description is so at odds with that of the report which, btw was the longest, most thorough in NTSB history I believe? He describes the duration of the jam/hardover phenomena as lasting a few seconds while the Eastwind upset data shows clearly it persisted 13 seconds, keeping in mind that that the entire 427 accident sequence occurred in a total of ~ 25 seconds. Furthermore, contrary to the author's assertion, there is adequate data from the Eastwind incident including the live crewmembers and FDR data. It should be noted that unlike the United and USAir upsets, the Eastwind upset occurred in calm air without a crew input to the rudder. Instead the crew reported feeling an initial bump coming from the pedal itself......there were obviously more than the one "thermal shock" failure mode displayed by the 737 rudder control system. Indeed the board found that the PCU's servo valve secondary slide could jam against the PCU housing out of its neutral position and display no adverse rudder performance as the primary slide would respond normally (referred to as a latent failure) and control the rudder as usual UNLESS a rapid rudder input was made by the crew at which point the servo valve primary slide could over-travel, mis-port the hydraulic fluid and cause a rudder hardover AND reversal. Depending on where the servo valve secondary slide jammed (the NTSB believed 427's jammed at 100% off of its neutral position) only a relatively ordinary, but quick rudder input could cause a reversal. Additionally, they found that if the pilot maintained the rudder input (427's flying pilot could be heard on the CVR grunting during this period as he strained while applying rudder pressure) the rudder reversal and hardover would persist. Only when the pilot released the rudder, which the pilot in this case would believe was the correct input (right rudder), for the left yaw and roll - would the jam/reversal relent according to the board's findings. I simply can't understand why such an accomplished, articulate pilot as the author is, would make a mess of this accident's causes and the crew's actions which I feel he roundly mis-characterizes. However, I agree with his call for better, more aggressive unusual attitude/upset training which can only benefit pilots and their passengers.
Hi Scott, I congratulate you on your rational approach. This is proper forensics and this is how we help keep people alive. As you say, no one WANTs to blame pilots, but refusal to look at human actions and human factors denies us the opportunity to become better and save lives. Your perpective is correct and interesting because I've observed a related phenomenon regarding ATC/Tower responses to pilots errors or pilots in bad situations. In most cases, the ATCs have reacted properly -- asking for example if the pilot would like to declare an emergency, etc. (Sometimes pilots don't recognize a danger early enough.) But in some other cases, (see Air Safety Intitute, Gear 74 and VASA videos), the ATCs, instead of focussing on the actions necessary to resolve the problem, they have instantly become emotional and shifted into "scold reflex", berating the pilots for their errors rather than prioritizing a solution. In one shocking example, a Cessna had inadvertently flown into the Bravo. He was trying to get out but didn't know the safe vector and altitude out. He was in the Bravo for 15 whole minutes (!) before the ATC even noticed! Then, the ATC immediately began to scold the pilot, declare her authority, and refuse to clear him through or offer a vector and altitude out. Shocking. This "scold reflex" not only occasionally affects ATCs where it only prolongs the danger, but the scold reflex can also affect us viewers sometimes. We are uncomfortable "blaming" the pilot. But is not blame. Our goal is to solve a problem, learn from our mistakes and save lives. Thank you, Sir, for the fine video.
The main factor i think was the finding of very old and dirty hydraulic fluid in flight 427 controls system as well of other USAair airplanes too.. Dirty or old fluid can jam valves, sending fluid in different direction as intended. And that is what happened here mostly. As an auto mechanic i was- before a pilot, i had to convince a lot of dummies to change the dirty transmissions, steering and differentials dirty fluids they had. Dirty fluids jam valves, springs and systems.
Spot on analysis Scott, well put together. Recovery upset training is something that needs to be addressed more fully for pilot training. Don't worry about the flaming from the crazies that don't have a clue 😂
Thank you, Scott. My takeaway for top-of-mind awareness for upset recovery after this presentation is unload and get out of there. Crossover speed is best achieved by pushing the stick forward and adding thrust. With windshear upsets, especially with jumbo aircraft it's paramount to keep ahead of the power curve. Studying what happened to Delta flight 191 crash at D/FW Airport in 1985 drove this point home in my mind. The FO was making the approach and windshear removed relative wind from the nose of the aircraft. The approach was fine for normal days in calm air. The captain announced to push the engines way up too late, and the flight landed off airport. Indicated airspeed numbers must be increased to ensure positive control in gusty conditions. Drivers must stay ahead of the power curve and know how to divide the difference. When will they ever learn stall recovery 101? Point the nose down and add full power!!! Unload and get out of there!!! An aviator's job is to overcome mechanical failure without adding to them. Driverless vehicles scare me! Complacency leads to autopilot error compounding pilot error. Remembering Flight 191 has kept me from making the same mistake.
Good talk Scott...I see what you did there with Rio Linda..LOL. As I indicated earlier on the other video, the crew on 427 left the autopilot on while trying to control the plane. If they had disconnected the autopilot at the get go , they would have had more authority initially. Until Boeing redesigned the rudder valve, at UAL we would do training on hardovers at every PC and the importance of maintaining the cross over speeds as you slowed.
Scott, great video as always. However, Boeing bears responsibility in faulty design of the rudder control valve on both UA 585 and US 427. And probably damn near 100%, no different sadly from the 737 max & elevator trim issues. Airbus does on AA 587. Pilot error is a secondary cause, not the primary. NTSB got these right, although took them years (w/ 737), due to lack of CVR/FDR data on UA 585. Notice how no planes crashed after the BOEING redesign of the rudder hydraulics(!!), thanks to NTSB and FAA. And no A300's either after AA 587. Love Boeing, but they don't always get it right. Neither does NTSB every time, but they eventually did here. Pilots can never be blamed for bad engineering, and I say this as both an engineer and pilot.
Totally agree with your points. I think a lot of what you said here applies in spades to the 737 MAX accidents. Yes, those airplanes had flaws that set the crews up for emergencies with the failures of just one sensor, but there were definitely human factors involved in those two accidents and with proper responses, both airplanes could (and probably should) have recovered. Lion Air basically proved that point the day before their accident, with the very same airplane.
“Too Late For a Few”…Maybe “Just in Time for Many to Come”, Scott, Great Job !!!😢👍👍✈️ Keeping Us All Consious of The Human Factors IS A GOOD THING,…God Bless You,….& RIP for All Who have paid the Ultimate Sacrifice !!!!😳😢😢😢😢😢
In flying, the test is administered first. The education comes afterwards. It's up to us to make sure that learning takes place and is not furred up by prejudice and our tendency to look right off the bat for the usual suspects. As pilots, we have a strange, split path in front of us: to have loads of confidence in our skills and our machines (because it would be dangerous to be so full of doubts and fears that we couldn't do our jobs....see "helmet fire"), but also pay very close attention to where things went south for other flights and other pilots. We have to know what's possible while denying that it could happen to us. We pay very close attention to accident reports. We also read them thinking, "Yeah, but I wouldn't have made that mistake."
I have no experience as a pilot, but I say this with lots of experience with logic and common sense...you sound like my Dad talking to me where after the conversation was over, I said, "Dammit, he's right (again)." 🤣
Or the pilot that never ever crashed,he is the guy that knows and he is the guy you should listen to. don't ask the guy who crashed how to fly!even if he is giving advice for free.🤣
This made me think of the United 232 / Sioux City incident - you CAN control almost everything flying that still generates lift and has some response to control inputs - however, figuring out howto might be too complex for any human or take too much time. Kudos to everyone who tries to tame a heaver than air chunk of metal.
"For those of you in Rio Linda", Yeah! You suddenly became my favorite aviation content creator! Will you talk a bit about crossover angle of attack, is it still a thing, if not did it become crossover air speed?
Thanks, I appreciate it! Maybe I should... and you got it perfectly right. Crossover is ALL about AOA... I alluded to that briefly in one of my comments. But folks generally like to talk about 1G flight and think that way. Speed is the proxy and measurement we can use. AOA is not displayed in airliners.
The crew reacted correctly according to the training they had. That's really sad that it made it worse than better. Most pilots would be caught off guard big time!
Thanks for the video Scott, and for including the note at the very end about the need to unload the wing because g-load increases the crossover speed. Most pilots instinctively "pull" in stressful moments like this, but that just raise the angle of attack and the g-load and makes it harder to recover.
I really appreciate your calm, non-judgemental, dissection and presentation of events. You are the precise opposite of a blowhard (and there are too many of those on RUclips). I'm a GA pilot and painfully aware that there is a great deal that I simply I don't know. For example, I've no idea what a snap roll is - much less how to recover from one!
Bob, A snap-roll is an aerobatic maneuver. Basically it's a spin going forward/ level, instead of down toward the ground. Recovered in a similar manner as a Spin.
Scott, great video. Thanks for continuing to talk about this subject. We have to get better as a pilot community of sharing information and getting important lessons, whether from the airlines or GA, disseminated to the aviation community. We are seeing so many accidents continue to repeat themselves; circling approaches to land, too slow on final, poor or no pre-flight, etc, etc. Hopefully, between your channel, Blancolirio, Gryder, and others, we can reduce the numbers of these kinds of accidents. We will always have mishaps and death related to flying. Its about effectively managing that risk to the lowest possible denominator and making sure you are ready, capable, and prepared to fly. The rest is left up to a higher power.
Scott, this all goes back to the old saying "you don't know what you don't know ". Simple stupid I know, but those critics of your educated story about what happened need to remember we are only human. I say Amen to your professionalism and response.!!!!
Epistemology. What an idea. Thank you, Sir. Thanks, Scott. Thanks, for sharing your knowledge, experience, insight and time. Scott Purdue, doesn't need to prove anything. Here's a man who genuinely cares about people and people who fly. He doesn't need to do this. He doesn't have to do this. Maybe Scott isn't always correct but maybe, just maybe, about flying...He is? If a surgeon says your appendix needs to come out like, RIGHT NOW, do you send him or her an email about how you don't agree? No, because you're too busy vomiting, scared and feeling like there isn't enough morphine in the world. You just say ok... So, here's one of those moments. Shut up, listen and then send a nice gift basket when you're back home, safe and sound with your family, And NOT dead. There's no reason to be a "douche" with an ego, full of "keyboard bravery" and overwhelmed with the uncontrollable impulse to explain after every Flywire production, how whatever Scott was speaking about, reminds you of the time YOU, shot down Manfred. Thank You, Scott. You're THE Best.
I made a comment at your previous remarks re 427 where I laid the blame at the PCU failure, and I hear your researched POV now but as I see it it’s little bit like saying the crews of the MAX disasters were at fault, technically they encountered a catastrophic condition they knew nothing about and failed to recover, a bit harsh I feel, I still blame the externally encountered upset. Had the crews recovered fully they would have been heralded as living hero’s, it is perhaps unfair to expect a crew to have recovered any of these encounters. As far as GA, better pilots than either of us have lost their lives to far less catastrophic events than rudder hard over or MCAS.
Well done Scott, There will unfortunately always be altruistic ego driven decisions which prevent us from discovering the true reasons for these and many other accidents both GA as well as commercial. Thank you !
Really interested to hear your thoughts on using rudders on large aircraft Scott , I agree with a previous poster , rudders only for engine out and crosswinds
A lot of people who opine here on YT airplane accident videos are morons who know next to nothing about airplanes or flying. Don't let 'em get you down and thanks again for another nice video. I had an FAA Air Safety Inspector geek on the JS one day. He was their number-one go-to guy for all things Airbus because he was also a former USAir Airbus pilot. He told us that the FO on the American Airbus JFK crash DID NOT use stop-to-stop rudder inputs as was so widely circulated. Rather that his relatively minor inputs (to the flight computers through the pedals) resulted in FULL stop rudder movements by the computers. He then showed us (graphically on his laptop) just how close you are to the failure limits of the rudder in various speed/flap configurations. Needless to say, any significant rudder inputs in those regimes could be dangerous. At about that same time at Brand X, we were drilled on new unusual attitude recovery techniques. None of them called for rudder whereas previously we had been told that it was OK to use some rudder to help roll the plane toward 60 degrees of bank to assist in lowering the nose toward the horizon in extreme nose-high attitudes. Airline pilots generally only use rudder to help taxi, for cross-wind takeoffs and landings, and during V1 cuts. 99 percent of the remaining time their feet are flat on the floor. I always found it hard to believe that a guy would suddenly jam in full and opposite inputs. But, perhaps you're right.
It’s what I learned from the accident when I was at AA. As I remember the airplane was just past the speed crossover point and beyond the rudder travel enhancement point. Below a certain speed the rudder does act as you related. Still, rudder was NOT the appropriate control input at all.
It is easier to blame the mechanics because to blame the human element is to admit that we can be at fault too. We just don't do that readily. Good sermon material once again.
It wasn't the mechanics, it was Boeing's faulty/non-redundant rudder control system design and Parker-Hannifin's manufacturing which brought these aircraft down. Would you also blame the MAX crashes on the pilots?
They did end up replacing the valve design did they not? Since then this hasn’t happened again has it? Different circumstances caused the valve to malfunction but the valve a factor was it not? Why redesign valve if not the issue?
The valve is totally the issue. I’m honestly really irritated he’s doubling down incorrectly here. It’s frankly disrespectful to the crew that perished. They were not at fault in the slightest. P
Sten Mulin was ambushed by the Airbus flight control system imo it's completely unfair he got blamed for AA587. only other aircraft ever to lose a tail that I know of was a '52 bomber. flying too damn fast thru a colorado rotor interestingly enough.
@@FlyWirescottperdue No. He had no idea what the flight control system was going to do to his rudder at 250kts. Not even his boss knew. nobody knew but airbus until _after_ the accident by the NTSB report. would you expect to go full deflection in 1.2"? Not to mention the pedal force had just been cut to less than half right then as well. I respectfully disagree. 1.2" to full deflection w/half the pedal force you had before. Ambushed. Sten Mulin was framed. Here is a quote from the factual report that might help shed some light on what I am saying.. "Regarding the rudder pedals, the A300 fleet standards manager stated that, before the flight 587 accident, American Airlines did not teach its pilots during training that rudder pedal movement would become restricted as airspeed increased. The fleet standards manager also stated that he did not know that the rudder pedal movement would become restricted because the pedals are not normally pushed to the stop in flight. In addition, the fleet standards manager stated that, before the flight 587 accident, he did not think that any pilot would have thought that full rudder could be gained from about 1 1/4 inch of pedal movement and 10 pounds of pressure (above the breakout force) at an airspeed of 250 knots."
Regarding USAir 427.... you might want to explain why the *forward cargo door* _departed the airframe_ and landed in a _local garden center_ as *USAir427* flew overhead _on final_ that day. _( hint: The _*_pressurized_*_ forward cargo doors open INWARD.)_
I understand your explanation that in automatic mode that the actuator for the rudder on the 737 would only hard over for a few seconds. That is fine and dandy except that when the pilot is actually trying to command at the opposite direction as he would normally do the rudder never releases period causing you to go completely out of control as happened in the Pittsburgh accident. I've read the research on it because when they first tested the rudder actuator it function just fine. Until the pilot would try and come and a Rudder deflection in the direction he wanted the plane to go instead of the direction the plane was going. It would stay hard over as long as he was trying to introduce command. All the way to the ground and well out of the area of turbulent air. I understand that turbulent air is what started that issue but it didn't continue all the way to the ground. I was called specifically before they found the issue with the rudder actuator with the chief investigator of that crash and it was part of the reason why they went back and looked at it after they realized that there was something that they were missing. The Denver incident was just a few seconds long. Obviously those few seconds are long enough to invert an aircraft and head it towards the ground. Something I am also familiar with because we had a problem with the F-111 and the takeoff and land configuration kicking on while in fight causing the exact same issue period and it only occurred if you were doing a light check while in flight. And it was exceedingly rare and it took us a while to figure out what the circuitry issue was there. We finally figured it out and solved the issue. My reason for getting involved in that investigation was merely because the effects of the flaw or identical and it was entirely possible that the 737 had a similar system in it. And at the end of my conversation with the chief investigator I said something cause that aircraft hard over the rudder and stay there. And it was for a hell of a lot longer than a few seconds that you were talking about. Seem strange how if it was only a small flaw within that actuator that could not have been solved without replacing virtually everyone of those actuators in the fleet. None of this removes your point about turbulence causing major issues within a certain distance of aircraft. Why the SR22 flew into that turbulence is beyond me. He should have been warned off!
Or there was no turbulence and the pilot of SR22 was flying in an unusual attitude without being aware of it. you can"t see turbulent air so how does the pilot know he hit wake turbulence?some pilot react in strange ways and add strange control inputs and loose control like the thousands of pilots who have lost control on the dreaded base to final turn. they didn't know that they had overstepped the mark till they bit the dust.
Well said. Any type of control reversal and jam at or near crossover speed, close to the ground would take quick and expert thinking to recover from without doubt.
Outstanding analysis and explanation, thank you. How many more lives have to be lost before Boeing admits their approach to controls is fundamentally wrong (and has been for decades) and the Airbus way with flight envelope protection is the safest solution.
Scott do you think the current training requirements are sufficient that modern airline pilots are unlikely to input snap roll inputs when startled? On another note, someday perhaps you can make a video on windshear escape. I'm curious if your viewpoint is that the airliner escape strategy is also appropriate in the Bonanza or other light airplanes.
When I was learning to fly Cessna textbook said 5miles separation (large) and small aircraft not just (heavy) but large and heavy air traffic controllers try to get to many on the ground I think Saratoga at (RDU) n534z in with fatal crash was ten falte ass hole controllers felt they would given a lackluster runway IFR at night n534z and best runway was given to faster aircraft I have ben known to get a bit short with those boys . Your input would be apricated Thank You Good Podcast. PS PATCO was around back in the day'
It's presumptuous to say "a lot of folks have ... a less-than-perfect understanding that rudder hardover played" in US Air 427. I think the controversy, such as it was, came about from your description in the first video of the cause of the crash. You essentially said the cause was pilot error -- and that may well be, at least in part, a valid opinion. The problem is you presented that opinion as fact. That didn't make sense to a lot of people WITHOUT also stating that the NTSB clearly and unambiguously determined that the crash was caused by the servo fault.
Agreed. And, with respect, it’s a pretty shitty thing to place much blame on pilots who have a key flight control doing the OPPOSITE of what it’s supposed to. That is setting a bar for recovery that no pilot can be expected to meet. 3.2 Probable Cause The National Transportation Safety Board determines that the probable cause of the USAir flight 427 accident was a loss of control of the airplane resulting from the movement of the rudder surface to its blowdown limit. The rudder surface most likely deflected in a direction opposite to that commanded by the pilots as a result of a jam of the main rudder power control unit servo valve secondary slide to the servo valve housing offset from its neutral position and overtravel of the primary slide.
The rudder of an aircraft is more effective at lower speed than are the ailerons for lateral control. As speed increases the ailerons/roll spoilers gain authority such that at a certain speed (crossover) they're able to counter rudder effects in roll control.....basically.
@@dutchflats this speed isn’t something that is published and would vary by aircraft and perhaps configuration. How does one determine this speed for a particular aircraft? Also when using aileron to counter roll, at least in light aircraft we add rudder to counter adverse Yaw. Fly wire mentioned unloading the aircraft (push). What impact does this have on crossover speed, if any?
@@imaPangolin We had published speeds for a given flap/configuration setting below which we couldn't bank the aircraft more than 15 degrees called floor speeds as well as minimum flap retraction speed, but they weren't crossover speed, a term which I only heard after the 427 accident although everyone probably knew it existed intuitively. We tend to use less rudder in large jets during ordinary operations than do GA aircraft, applying gentle/subtle rudder pressure (if at all) other than during takeoff and landing. In addition, rudder travel is progressively limited automatically based on your airspeed, as your speed increases less rudder deflection is available. I believe Fly Wire was getting at reducing G load to help the aircraft accelerate better so as to more quickly get above "cross over speed" and increase aileron effectiveness. With G on the plane you're losing energy to the turning motion which could be better used as airspeed. BTW, in aircraft like the A320 flap speeds, best lift over drag speed, stall speed, etc. are displayed on the PFD speed tape automatically and are calculated by the flight management system based on configuration and weight.
@@dutchflats I’m a large jet pilot as well. But I never knew this term and now some things make sense that prior I thought was only related to stall speed. We have a minimum maneuver speed for various flap configurations that really receive no explanation in training. Just don’t go below that speed. This speed is always above Vref by quite a margin. I’m pretty sure after this video and discussion that this is related to this “cross over speed”. I’m frankly a bit stunned at my ignorance.
@@imaPangolin Not as stunned as the guys flying 427 were I'm sure. Same with us regarding maneuvering speed, I'm retired now so some of this stuff is starting to escape me. Nowadays you just don't fly slower than the yellow stripe. One of the airplanes I flew had an actual angle of attack indicator which I loved, but haven't seen it on any other transport category jets?
Another training element taught because of these hardovers, is an emphasis of unloading the wing as the first move when encountering an upset. Unloading lowers crossover speed and increases aileron effectiveness. Even though 427 was slightly below crossover, unloading would have given them more control… if they had known. As far as untrained control inputs, examining the Nov 2001 NYC Airbus crash, wake vortices were encountered and the pilot flying put in the trained inputs of knocking down the wing with rudder and applied multiple full scale reversals until the vertical stab sheered. This training completely nullified Va protection and resulted in that crash. I don’t know what the training for the crews of 427 were for wake encounters, but I am making a general assumption that these two events not being that far apart in time may have had similar training. Good video and thank you for making it.
Thanks Flyin and Drivin... gotta correct one thing about AA587. I went through the AAMP program the same way the FO did.... AND THAT WAS NOT A TRAINED REACTION. Period, he made it up on his own. And was told about it, still did it anyway. Repeat, the reaction of the FO was not a standard maneuver, old wives tale or whatever it was it was not taught.
@@FlyWirescottperdue I appreciate the correction. It was definitely a garbage technique. Now it’s unfortunately a training point for others. I always encourage my students to ask “why?” and this is a learning tool for exactly that scenario. Use the proper controls for the proper movement. I had a disagreement with some old timers who insisted on inappropriate rudder use with their excuse being “well in the F-4 they ruddered those for roll all the time.” A. I doubt that. B. We’re not flying F-4’s. If you wouldn’t mind checking out my other channel: Spread Aviation. Thank you sir.
Very good. I was a FO at Usair on the 737 at the time of the accident and the way we flew the B737 changed to take advantage of the crossover.
Well said, Scott. We all need to learn from those who paid for the information with their lives. Thanks for the video.
Hi Scott , great job , a lot of info on a short video . I am a retired 25000 Hour Regional Airline Captain ! Most all my airplanes were cable driven controls .
I was always directly connected to my
Control surfaces . Now days new technology is taking the pilots out of the
Loop ! And the thing that is sad for people in the air and on the ground ,
safety is moving to the engineers and in many cases have no
Flying experience flying . They are trying to Phase out the pilots . We are in this
New era and there are going to be accidents to make it all work . We are in If man were ment to fly era again !
I retired from aviation in 1997 at the ripe
Old age of 47 and I haven’t flown since
I drive my motor home everywhere. The
Decline in aviation safety started in 1985
With deregulation . I left aviation because I didn’t think it was right for me
To support unsafe industry ! My Dad Captain Howard Wirth a 747 Captain for
TWA taught me at a very young age that
True Safety is no accident! After becoming an A/P Mechanic and working
On aircraft first, I thought I could make
This theory work ! I new all my airplanes
Very well and have a spotless record as an aviator ! But I was totally connected to all my aircraft and I trusted the maintenance done on them ! But when deregulation dropped in and cut pay to all the proficient mechanics ,
ATC controllers , and pilots they left
And we were left with inexperience, it scared me . I left aviation for my guitar .
Now days it’s almost like aircraft accidents is a form of entertainment !
I’m writing this because I’m living this
Transition that I’m seeing ! To shorten
Up this whole explanation in a few words
In my era and my Dads era it’s was if there is a question we always did the safe thing ,and we’re not questioned ! Now days you better
Not creat waves or you will get a black
mark on your record ! My days were
True Safety Is No Accident and You Make your Own Luck ! ( God Help All
The Innocent People on the Ground and
In the Air . Captain Mark H Wirth
PS Now days if you climb on an aircraft
It’s no one’s fault but your own if there’s a problem ! Your the pilot in command of
Your own life !
We talk a lot about doing a “root cause analysis” in my field-medicine-when a serious incident occurs, but that language often locks everyone into the mindset that there is only ONE root cause, when in fact there may be several. We should really be doing “an analysis of root CAUSES” or even more broadly, “an analysis of causation.” I have read the NTSB reports on these two accidents several times, and so I find your analysis insightful: if no upset, then no radical rudder inputs, so no hardover, so no mishap. Great second look at cases that were settled (incompletely) long ago. Great work!
There is a ginormous elephant in the room here. The word ”error” implies culpability. That is a very serious implication in the roman law tradition of your culture. ”Error” is the derived from the latin word ”Errandus”: To go astray. How can you go astray when there isn’t even a defined Straight ’n narrow? It’s a bit like saying you got hit in the head by a meteorite because you stood in the wrong place. Factually correct as it may be still not contributing one bit…
Still, awesome video and very important! Thank’s Scott!
Another great, informative, debrief Scott. Your prevailing theme of 'what happened', and 'what can be learned', is what makes these recaps so valuable, as always. Thanks, Scott
Thank you kindly
I have a strange connection to Flight 427 as did many Pittsburghers. That evening, I happened to be driving inbound on the Parkway West and less than 5 miles from the accident. However, I was ahead of it and only made aware when the emergency vehicles were Outbound. I also knew one of the investigators who was a physician and a US Air pilot who was assigned to investigate. At the time, I lived at a place called Lake Latonka and several of my neighbors were US Air pilots. Needless to say, it was a very sad place for quite some time. Thanks for the deeper dive much like Fate is the Hunter. A final note, for those of us who have experienced real wake turbulence in a light aircraft (me in 2020 runway 04 SRQ) flying a Cessna 162 Skycatcher, you remember cleaning your drawers out. I've since have declined more than one take-off clearance on intersecting runways with Commercial equipment.
Excellent. Thanks again for sharing.
The process of understanding the result of ANY mishap in ANY environment is exactly as you describe. I am not a pilot. I am a process development chemist who has experienced / investigated a few catastrophic events. As humans we reach for the simple answer. The simple answer is probably never the whole story. Personally I would not look for blame but rather what needs to be done to prevent it happening again. You can only accomplish that by knowing everything that happened. I’m probably not making complete sense. What I am trying to say is to resist the suction caused by the bullshit storm caused by the oversimplified conclusion. E.g. there was a spark and then a loud boom!
Thanks for sharing Thomas, you make perfect sense.
Thank you for your lucid, mature analysis and opinion.
Thanks again for another excellent video, Scott.
Fantastic summery Scott , Thanks
Thanks for bringing all your experience and knowledge to me…someone who flew a desk all his life and now in his 70s learning to fly a Tomahawk.
Much appreciated, Scott, and one of the reasons I subscribe. My instructors drilled into me that it's rarely ONE cause, it's always a catastrophic chain, which can encompass human and non-human (e.g. mechanical) errors. However, one of my own comments got eaten by YT which was the Airline Pilot's Union reply to the NTSB findings (and incorporated in the final report). It very much wanted to focus on the mechanical aspect in a maybe understandable effort to absolve "their" people.
Scott you are right on the money.... unfortunately we raising System Managers in the cockpit, instead of Stick and Rudder Pilots. If the airplane can't be flown by the Flight Director...and the situation hits the fan....You're in for a long day.
Scott, as usual, well researched. Not mentioned however, is that the B737 was certified with only one rudder PCU. During the design and certification phase, Boeing had to assure the FAA that reversal of the power control unit was not possible as there was no back up. Initially Boeing refused to accept that there had been a malfunction, this refusal occurred over a number of years while the aircraft continued to fly without corrective action. It was only after persistent follow up and testing by ALPA and industry advocates that Boeing relented and the PCU was redesigned. The airlines do minimal training in recovering from upsets. Most of the training is recognizing the "approach to a stall". Full stalls, as in light planes, are not conducted in air carrier aircraft. There is a dramatic difference in performance and recovery techniques depending on hi altitude incidents and low altitude deviations.
The crossover speed between rudder effectiveness and aileron authority was not mentioned in the AFM. On the B727 and B707 there was additional roll authority when the flaps were extended through the use of supplemental inboard roll devices.
You do an excellent job, no flame suit required!!
Truth🎯
Back in the 80s and 90s I flew the 737-200, 300, and 400 for about 20 years before going on to the 757 and 767. Out of the three I felt the 200 was the best handling. To me it was like the 300 didn’t have enough wing for the job. The roll rate was not as good as the smaller 200. I also found that the 300 would be easier to upset in flight behind, not only larger aircraft, but also aircraft of equal size. One time while in trail of an MD-80 on final I felt a little shake and noticed the airspeed beginning
to bleed off rapidly. We were at vref for flap 5 at the time. The plane began a slow roll to the right and I realized I was going to need more airspeed to bring it under control. Since I was on final and cleared for the approach I had no need to maintain altitude and was able to trade it for airspeed. I pushed the nose over a little while at the same time increasing engine power. It was almost like recovering from a stall entry. Another time something similar happened behind a Fokker F-100. It’s amazing that aircraft of that size, compared to you, can create such upsetting conditions. If you ever have to trade altitude for airspeed it’s easier to explain why later since you’re still around to do so.
Correctamundo 💯
Not many had a chance to fly the B737-100, but it had the best handling of the bunch. It was a "stick and rudder" airplane, no fancy electronics, navigation, just the old "steam gauges" and very simple SP77 autopilot. To me the Wiz Bang gadgets helped in efficiency, navigation, and reduced the workload at cruise, but they also took away an element of the pure fun factor of just flying and being in harmony with the aircraft. Scott, Juan, and Dan look to recover that basic simple joy of flying with their own private aircraft.
Another great video, Gunny. Well thought out and carefully delivered. Thank you
Excellent, Scott. Your focus on our false sense of infallibility as pilots is valuable and may save lives.
In the '90s I was on a flight with a co-worker from Sacramento to LA on a Southwest airlines 737 and we were on approach to LA and the airplane suddenly dipped the left wing straight down and then the right wing was straight down and then the plane righted itself. The pilot came on the intercom and apologized and explain that they were caught in a wake from a 747 that was ahead of them.
Hi Scott, I appreciate the research you've done and the conclusions you're drawing here - on the other hand the 737 PCU valve hardover/reversal story opens up several cans of worms for a reason. As others have commented, the B737 was certified with a single PCU valve which was designed to be redundant in and of itself, and ultimately it turned out that this was not sufficient.
Rewinding to when Boeing was designing these valves in the '60s, the manufacture of these valves was subcontracted to Parker Hannifin based on tolerances specified by Boeing, and the valves were tested to the extent of the knowledge held at the time. As I understand it, the "thermal shock" scenario came from a discovery (late in the investigation) by a Boeing engineer that when the holes drilled for the transfer of hydraulic fluid were located towards the far end of the specified tolerances, the behaviour of the valve could be affected when certain conditions prevailed - thermal shock being one of them.
It's important to note that the only PCU valve tested with the thermal shock scenario was that recovered from the wreck of USAir 427. The Eastwind aircraft was tested with the PCU in place on the ground, and if I recall correctly, it was discovered that the PCU could hardover and reverse without requiring the thermal shock condition to be present. The PCU valve from the wreck of UA585 was too badly damaged by impact for any dynamic testing to be performed on it - all they could do was disassemble it and check it was within tolerances (which it was). Furthermore, because the United B732 was such an old airframe, the FDR with which it was fitted only recorded very basic parameters; specifically it did not record rudder pedal input. This is one of the reasons that the initial NTSB report on UA585 was forced to conclude that there was insufficient evidence to determine probable cause, in spite of the fact that the departure from controlled fight was consistent with a rudder hardover.
One particular can of worms regarding this story when it comes to USAir 427 in particular - and possibly a reason why the response regarding your views on pilot input received the pushback it did - is that Boeing initially went to considerable lengths to try and prove that the upset was *solely* caused by inappropriate handling on the part of the flight crew, and as I'm sure you're aware, that rarely goes down well among the piloting community. This was the second accident in which the B737 rudder was considered suspect, so it's hardly surprising that Boeing's efforts were regarded as a damage control effort, if not an attempt at a cover-up.
You correctly state that the circumstances of UA585 were not conducive to the "thermal shock" scenario which was proven to have been a possible factor in USAir 427, but that does not take into account the fact that thermal shock was probably not the only condition in which the original PCU valve, when manufactured close to the tolerance limits, could behave abnormally. What allowed the NTSB to finally conclude that a rudder hardover and reversal was a likely factor in UA585 was the flightpath of the aircraft, which was consistent with that scenario, and regardless of the crew's handling of the situation, the PCU valve's design fell short of the redundancy requirements necessary for certification. Line pilots may not have been trained in the significance of crossover speeds, but the crossover speed would certainly have been considered as part of the certification, and because the PCU valve's flaws were not understood or taken into account at that time, the "safe" airspeeds determined at certification for those phases of flight did not take that into account.
I absolutely agree with you on your point regarding the tendency to wish for a single "smoking gun" cause, and how focusing on that can be detrimental to a full understanding of all the factors involved in an accident. However, as anyone who has ever tried to ride one of those "funfair" bicycles where the steering is reversed can attest to, even if one is logically aware of the fact that the controls are reversed, the reality of experiencing a behaviour to which you have become accustomed for a significant time suddenly no longer making sense is a very serious problem. In other words, if a rudder pedal input resulted in a reversal and hardover even for a few seconds, the disorientation that would cause would certainly drastically hamper any attempt at recovery because you'd instinctively know that at least one of your controls has malfunctioned, but have no idea which one, nor would you know that the malfunction was transient. As such, I find it difficult to consider the inability to recover as an "error" on the part of the flight crew - even if the hardover/reversal was triggered by an unusual control input due to circumstance - because the aircraft not only behaved outside of the parameters which they were trained to expect, it behaved outside of the parameters for which it was certified.
This is in turn significantly different from the situation regarding AA587, which to my mind was also a systemic set of failures - in that case it was a training programme (AARP) which did not take the differences between types in the fleet into account combined with a particular pilot who came out of that training with an inappropriate understanding of the risks (the training left him exceptionally nervous around wake encounters), which led to inappropriate application of the techniques (namely unusually aggressive handling of the rudder controls). A final systemic problem was that other flight crew noticed this nervousness and excessive use of rudder, but at no point was an attempt made to understand and remedy why this pilot was reacting and behaving the way he was. Ironically, one outcome of this accident was an effort on the part of some AA pilots to treat the A300's rudder controls as a single "smoking gun".
The AA program was called AAMP and the FOs flying in 587 in no way reflected the program. I went through the AAMP program and it was excellent and I believe it WAS type specific. The thread that is common in all three of these accidents is the crew did not fly the airplane first. They did not understand the need to reduce AOA. The AA program was talking about crossover speed shortly after 427. Rudder doublets: ‘reversals’ were not taught. I think you discount the ability of the crew to fly the plane. Review the comments of the USAir Flight Managers.
turricaned tc: Very well said and reasoned, it appears to me after re-reading much of the 427/585/517 NTSB accident report (AAR-99/01) that the author here has never actually read the entire document or understood its findings as his sequence of events description is so at odds with that of the report which, btw was the longest, most thorough in NTSB history I believe? He describes the duration of the jam/hardover phenomena as lasting a few seconds while the Eastwind upset data shows clearly it persisted 13 seconds, keeping in mind that that the entire 427 accident sequence occurred in a total of ~ 25 seconds.
Furthermore, contrary to the author's assertion, there is adequate data from the Eastwind incident including the live crewmembers and FDR data. It should be noted that unlike the United and USAir upsets, the Eastwind upset occurred in calm air without a crew input to the rudder. Instead the crew reported feeling an initial bump coming from the pedal itself......there were obviously more than the one "thermal shock" failure mode displayed by the 737 rudder control system. Indeed the board found that the PCU's servo valve secondary slide could jam against the PCU housing out of its neutral position and display no adverse rudder performance as the primary slide would respond normally (referred to as a latent failure) and control the rudder as usual UNLESS a rapid rudder input was made by the crew at which point the servo valve primary slide could over-travel, mis-port the hydraulic fluid and cause a rudder hardover AND reversal. Depending on where the servo valve secondary slide jammed (the NTSB believed 427's jammed at 100% off of its neutral position) only a relatively ordinary, but quick rudder input could cause a reversal.
Additionally, they found that if the pilot maintained the rudder input (427's flying pilot could be heard on the CVR grunting during this period as he strained while applying rudder pressure) the rudder reversal and hardover would persist. Only when the pilot released the rudder, which the pilot in this case would believe was the correct input (right rudder), for the left yaw and roll - would the jam/reversal relent according to the board's findings.
I simply can't understand why such an accomplished, articulate pilot as the author is, would make a mess of this accident's causes and the crew's actions which I feel he roundly mis-characterizes. However, I agree with his call for better, more aggressive unusual attitude/upset training which can only benefit pilots and their passengers.
thank you for so many learning opportunities through your productions.
Hi Scott, I congratulate you on your rational approach. This is proper forensics and this is how we help keep people alive. As you say, no one WANTs to blame pilots, but refusal to look at human actions and human factors denies us the opportunity to become better and save lives. Your perpective is correct and interesting because I've observed a related phenomenon regarding ATC/Tower responses to pilots errors or pilots in bad situations. In most cases, the ATCs have reacted properly -- asking for example if the pilot would like to declare an emergency, etc. (Sometimes pilots don't recognize a danger early enough.) But in some other cases, (see Air Safety Intitute, Gear 74 and VASA videos), the ATCs, instead of focussing on the actions necessary to resolve the problem, they have instantly become emotional and shifted into "scold reflex", berating the pilots for their errors rather than prioritizing a solution. In one shocking example, a Cessna had inadvertently flown into the Bravo. He was trying to get out but didn't know the safe vector and altitude out. He was in the Bravo for 15 whole minutes (!) before the ATC even noticed! Then, the ATC immediately began to scold the pilot, declare her authority, and refuse to clear him through or offer a vector and altitude out. Shocking. This "scold reflex" not only occasionally affects ATCs where it only prolongs the danger, but the scold reflex can also affect us viewers sometimes. We are uncomfortable "blaming" the pilot. But is not blame. Our goal is to solve a problem, learn from our mistakes and save lives. Thank you, Sir, for the fine video.
Whoa! Rio Linda? Keep up the good work, Scott. Megadittos to you.
Haven’t heard those comments in a while….sadly.
Thanks, Scott. This was a major mystery.
Thanks!
You bet! I appreciate it.
The main factor i think was the finding of very old and dirty hydraulic fluid in flight 427 controls system as well of other USAair airplanes too.. Dirty or old fluid can jam valves, sending fluid in different direction as intended. And that is what happened here mostly. As an auto mechanic i was- before a pilot, i had to convince a lot of dummies to change the dirty transmissions, steering and differentials dirty fluids they had. Dirty fluids jam valves, springs and systems.
Spot on analysis Scott, well put together. Recovery upset training is something that needs to be addressed more fully for pilot training.
Don't worry about the flaming from the crazies that don't have a clue 😂
Very well said! And as usual very well analyzed. Thanks so much Scott.
Looking forward to this Scott. You are one of my 4 fav aerospace channels
Great job. You’re right on the money Scott!
Thanks for the update. 🙏
Thank you, Scott. My takeaway for top-of-mind awareness for upset recovery after this presentation is unload and get out of there. Crossover speed is best achieved by pushing the stick forward and adding thrust. With windshear upsets, especially with jumbo aircraft it's paramount to keep ahead of the power curve. Studying what happened to Delta flight 191 crash at D/FW Airport in 1985 drove this point home in my mind. The FO was making the approach and windshear removed relative wind from the nose of the aircraft. The approach was fine for normal days in calm air. The captain announced to push the engines way up too late, and the flight landed off airport. Indicated airspeed numbers must be increased to ensure positive control in gusty conditions. Drivers must stay ahead of the power curve and know how to divide the difference. When will they ever learn stall recovery 101? Point the nose down and add full power!!! Unload and get out of there!!! An aviator's job is to overcome mechanical failure without adding to them. Driverless vehicles scare me! Complacency leads to autopilot error compounding pilot error. Remembering Flight 191 has kept me from making the same mistake.
Awesome comments Will!
👍
Scott, these are very informative! Keep them coming!
Good talk Scott...I see what you did there with Rio Linda..LOL. As I indicated earlier on the other video, the crew on 427 left the autopilot on while trying to control the plane. If they had disconnected the autopilot at the get go , they would have had more authority initially. Until Boeing redesigned the rudder valve, at UAL we would do training on hardovers at every PC and the importance of maintaining the cross over speeds as you slowed.
Scott, great video as always. However, Boeing bears responsibility in faulty design of the rudder control valve on both UA 585 and US 427. And probably damn near 100%, no different sadly from the 737 max & elevator trim issues. Airbus does on AA 587. Pilot error is a secondary cause, not the primary. NTSB got these right, although took them years (w/ 737), due to lack of CVR/FDR data on UA 585. Notice how no planes crashed after the BOEING redesign of the rudder hydraulics(!!), thanks to NTSB and FAA. And no A300's either after AA 587. Love Boeing, but they don't always get it right. Neither does NTSB every time, but they eventually did here. Pilots can never be blamed for bad engineering, and I say this as both an engineer and pilot.
Love these discussions thanks for sharing the knowledge with us all Scott.
Totally agree with your points. I think a lot of what you said here applies in spades to the 737 MAX accidents. Yes, those airplanes had flaws that set the crews up for emergencies with the failures of just one sensor, but there were definitely human factors involved in those two accidents and with proper responses, both airplanes could (and probably should) have recovered. Lion Air basically proved that point the day before their accident, with the very same airplane.
Tremendous video. Thanks 🇺🇸
Bless Jose Ponce, he perished that day on Flight 427. With many others.
Excellent explanition.
“Too Late For a Few”…Maybe “Just in Time for Many to Come”, Scott, Great Job !!!😢👍👍✈️ Keeping Us All Consious of The Human Factors IS A GOOD THING,…God Bless You,….& RIP for All Who have paid the Ultimate Sacrifice !!!!😳😢😢😢😢😢
Great...! Video and information sir.
“those of you in Rio Linda”...,
hahaha... (R.I.P., Rush Limbaugh)...,
Yep, I caught that Scott.
Thank you Ser!
Excellent!
In flying, the test is administered first. The education comes afterwards. It's up to us to make sure that learning takes place and is not furred up by prejudice and our tendency to look right off the bat for the usual suspects. As pilots, we have a strange, split path in front of us: to have loads of confidence in our skills and our machines (because it would be dangerous to be so full of doubts and fears that we couldn't do our jobs....see "helmet fire"), but also pay very close attention to where things went south for other flights and other pilots. We have to know what's possible while denying that it could happen to us. We pay very close attention to accident reports. We also read them thinking, "Yeah, but I wouldn't have made that mistake."
Wise words…
I have no experience as a pilot, but I say this with lots of experience with logic and common sense...you sound like my Dad talking to me where after the conversation was over, I said, "Dammit, he's right (again)." 🤣
Or the pilot that never ever crashed,he is the guy that knows and he is the guy you should listen to. don't ask the guy who crashed how to fly!even if he is giving advice for free.🤣
Situation awareness, upset recovery training might help, low altitude problems can only go so far.
Thanks Scott.....!!!
Oooh, I dig your "Fly Wire" + F-14 sticker on your fin. I LOVE the upside bit!
One correction… it’s an F15E;)
@@FlyWirescottperdue Ya know...I 1st thought it twas an F-15, but then I thought, you were Navy...Navy = F-14.
Do you, did you know one Kathy?
Love the Limbaugh, Rio Linda reference.
This made me think of the United 232 / Sioux City incident - you CAN control almost everything flying that still generates lift and has some response to control inputs - however, figuring out howto might be too complex for any human or take too much time. Kudos to everyone who tries to tame a heaver than air chunk of metal.
Awesome, thank you for the information.
You have a sharp memory, reciting facts like an investigator.
Awesome! Thanks I get it.✋🏻👍
Thanks for the video.
Well said.
Good video.
"For those of you in Rio Linda", Yeah! You suddenly became my favorite aviation content creator!
Will you talk a bit about crossover angle of attack, is it still a thing, if not did it become crossover air speed?
Thanks, I appreciate it! Maybe I should... and you got it perfectly right. Crossover is ALL about AOA... I alluded to that briefly in one of my comments. But folks generally like to talk about 1G flight and think that way. Speed is the proxy and measurement we can use. AOA is not displayed in airliners.
The crew reacted correctly according to the training they had. That's really sad that it made it worse than better. Most pilots would be caught off guard big time!
Truth 🎯
Very Good! Thanks!
Thanks for the video Scott, and for including the note at the very end about the need to unload the wing because g-load increases the crossover speed. Most pilots instinctively "pull" in stressful moments like this, but that just raise the angle of attack and the g-load and makes it harder to recover.
Bravo.
Big part of these is the startle effect.
Absolutely right... and training gets us through that!
Completely agree.
Well I get the points you were trying to make. The book on Flight 427 is pretty interesting.
absolutely, Capt well said
"The common denominator - easy for you to say" I was expecting the "easy for you to say" after that sentence, lol. I've been conditioned!
I really appreciate your calm, non-judgemental, dissection and presentation of events. You are the precise opposite of a blowhard (and there are too many of those on RUclips). I'm a GA pilot and painfully aware that there is a great deal that I simply I don't know. For example, I've no idea what a snap roll is - much less how to recover from one!
Bob, A snap-roll is an aerobatic maneuver. Basically it's a spin going forward/ level, instead of down toward the ground. Recovered in a similar manner as a Spin.
Scott, great video. Thanks for continuing to talk about this subject. We have to get better as a pilot community of sharing information and getting important lessons, whether from the airlines or GA, disseminated to the aviation community. We are seeing so many accidents continue to repeat themselves; circling approaches to land, too slow on final, poor or no pre-flight, etc, etc. Hopefully, between your channel, Blancolirio, Gryder, and others, we can reduce the numbers of these kinds of accidents. We will always have mishaps and death related to flying. Its about effectively managing that risk to the lowest possible denominator and making sure you are ready, capable, and prepared to fly. The rest is left up to a higher power.
Respect.
For those of you in Rio Linda, I love it!
Hey, I’m not far, less than 5 miles, but thank the golden microphone for making it famous.
Thanks
Scott, this all goes back to the old saying "you don't know what you don't know ". Simple stupid I know, but those critics of your educated story about what happened need to remember we are only human. I say Amen to your professionalism and response.!!!!
Epistemology. What an idea.
Thank you, Sir.
Thanks, Scott. Thanks, for sharing your knowledge, experience, insight and time.
Scott Purdue, doesn't need to prove anything.
Here's a man who genuinely cares about people and people who fly.
He doesn't need to do this. He doesn't have to do this.
Maybe Scott isn't always correct but maybe, just maybe, about flying...He is?
If a surgeon says your appendix needs to come out like, RIGHT NOW, do you send him or her an email about how you don't agree?
No, because you're too busy vomiting, scared and feeling like there isn't enough morphine in the world. You just say ok...
So, here's one of those moments. Shut up, listen and then send a nice gift basket when you're back home, safe and sound with your family, And NOT dead.
There's no reason to be a "douche" with an ego, full of "keyboard bravery" and overwhelmed with the uncontrollable impulse to explain after every Flywire production, how whatever Scott was speaking about, reminds you of the time YOU, shot down Manfred.
Thank You, Scott.
You're THE Best.
I made a comment at your previous remarks re 427 where I laid the blame at the PCU failure, and I hear your researched POV now but as I see it it’s little bit like saying the crews of the MAX disasters were at fault, technically they encountered a catastrophic condition they knew nothing about and failed to recover, a bit harsh I feel, I still blame the externally encountered upset. Had the crews recovered fully they would have been heralded as living hero’s, it is perhaps unfair to expect a crew to have recovered any of these encounters. As far as GA, better pilots than either of us have lost their lives to far less catastrophic events than rudder hard over or MCAS.
And I disagree... that is why I harp on training so much. You don't learn how to fly by Osmosis.
Well done Scott, There will unfortunately always be altruistic ego driven decisions which prevent us from discovering the true reasons for these and many other accidents both GA as well as commercial. Thank you !
Another great video!
Really interested to hear your thoughts on using rudders on large aircraft Scott , I agree with a previous poster , rudders only for engine out and crosswinds
A lot of people who opine here on YT airplane accident videos are morons who know next to nothing about airplanes or flying. Don't let 'em get you down and thanks again for another nice video.
I had an FAA Air Safety Inspector geek on the JS one day. He was their number-one go-to guy for all things Airbus because he was also a former USAir Airbus pilot. He told us that the FO on the American Airbus JFK crash DID NOT use stop-to-stop rudder inputs as was so widely circulated. Rather that his relatively minor inputs (to the flight computers through the pedals) resulted in FULL stop rudder movements by the computers. He then showed us (graphically on his laptop) just how close you are to the failure limits of the rudder in various speed/flap configurations. Needless to say, any significant rudder inputs in those regimes could be dangerous. At about that same time at Brand X, we were drilled on new unusual attitude recovery techniques. None of them called for rudder whereas previously we had been told that it was OK to use some rudder to help roll the plane toward 60 degrees of bank to assist in lowering the nose toward the horizon in extreme nose-high attitudes.
Airline pilots generally only use rudder to help taxi, for cross-wind takeoffs and landings, and during V1 cuts. 99 percent of the remaining time their feet are flat on the floor. I always found it hard to believe that a guy would suddenly jam in full and opposite inputs. But, perhaps you're right.
It’s what I learned from the accident when I was at AA. As I remember the airplane was just past the speed crossover point and beyond the rudder travel enhancement point. Below a certain speed the rudder does act as you related. Still, rudder was NOT the appropriate control input at all.
It is easier to blame the mechanics because to blame the human element is to admit that we can be at fault too. We just don't do that readily. Good sermon material once again.
It wasn't the mechanics, it was Boeing's faulty/non-redundant rudder control system design and Parker-Hannifin's manufacturing which brought these aircraft down. Would you also blame the MAX crashes on the pilots?
Ha, Ha, Ha I went to Rio Linda High School......and learned to fly at Rio Linda Airport.
They did end up replacing the valve design did they not? Since then this hasn’t happened again has it? Different circumstances caused the valve to malfunction but the valve a factor was it not? Why redesign valve if not the issue?
The valve is totally the issue. I’m honestly really irritated he’s doubling down incorrectly here. It’s frankly disrespectful to the crew that perished. They were not at fault in the slightest. P
@@floatplane Spot-on assessment ✔
Sten Mulin was ambushed by the Airbus flight control system imo it's completely unfair he got blamed for AA587. only other aircraft ever to lose a tail that I know of was a '52 bomber. flying too damn fast thru a colorado rotor interestingly enough.
Daniel, no he actually did the wrong thing in 587.
@@FlyWirescottperdue No. He had no idea what the flight control system was going to do to his rudder at 250kts. Not even his boss knew. nobody knew but airbus until _after_ the accident by the NTSB report. would you expect to go full deflection in 1.2"? Not to mention the pedal force had just been cut to less than half right then as well. I respectfully disagree. 1.2" to full deflection w/half the pedal force you had before. Ambushed. Sten Mulin was framed.
Here is a quote from the factual report that might help shed some light on what I am saying..
"Regarding the rudder pedals, the A300 fleet standards manager stated that, before the flight 587 accident, American Airlines did not teach its pilots during training that rudder pedal movement would become restricted as airspeed increased. The fleet standards manager also stated that he did not know that the rudder pedal movement would become restricted because the pedals are not normally pushed to the stop in flight. In addition, the fleet standards manager stated that, before the flight 587 accident, he did not think that any pilot would have thought that full rudder could be gained from about 1 1/4 inch of pedal movement and 10 pounds of pressure (above the breakout force) at an airspeed of 250 knots."
Regarding USAir 427.... you might want to explain why the *forward cargo door* _departed the airframe_ and landed in a _local garden center_ as *USAir427* flew overhead _on final_ that day. _( hint: The _*_pressurized_*_ forward cargo doors open INWARD.)_
I understand your explanation that in automatic mode that the actuator for the rudder on the 737 would only hard over for a few seconds. That is fine and dandy except that when the pilot is actually trying to command at the opposite direction as he would normally do the rudder never releases period causing you to go completely out of control as happened in the Pittsburgh accident. I've read the research on it because when they first tested the rudder actuator it function just fine. Until the pilot would try and come and a Rudder deflection in the direction he wanted the plane to go instead of the direction the plane was going. It would stay hard over as long as he was trying to introduce command. All the way to the ground and well out of the area of turbulent air.
I understand that turbulent air is what started that issue but it didn't continue all the way to the ground. I was called specifically before they found the issue with the rudder actuator with the chief investigator of that crash and it was part of the reason why they went back and looked at it after they realized that there was something that they were missing. The Denver incident was just a few seconds long. Obviously those few seconds are long enough to invert an aircraft and head it towards the ground. Something I am also familiar with because we had a problem with the F-111 and the takeoff and land configuration kicking on while in fight causing the exact same issue period and it only occurred if you were doing a light check while in flight. And it was exceedingly rare and it took us a while to figure out what the circuitry issue was there. We finally figured it out and solved the issue. My reason for getting involved in that investigation was merely because the effects of the flaw or identical and it was entirely possible that the 737 had a similar system in it. And at the end of my conversation with the chief investigator I said something cause that aircraft hard over the rudder and stay there. And it was for a hell of a lot longer than a few seconds that you were talking about. Seem strange how if it was only a small flaw within that actuator that could not have been solved without replacing virtually everyone of those actuators in the fleet.
None of this removes your point about turbulence causing major issues within a certain distance of aircraft. Why the SR22 flew into that turbulence is beyond me. He should have been warned off!
Or there was no turbulence and the pilot of SR22 was flying in an unusual attitude without being aware of it. you can"t see turbulent air so how does the pilot know he hit wake turbulence?some pilot react in strange ways and add strange control inputs and loose control like the thousands of pilots who have lost control on the dreaded base to final turn. they didn't know that they had overstepped the mark till they bit the dust.
Well said. Any type of control reversal and jam at or near crossover speed, close to the ground would take quick and expert thinking to recover from without doubt.
Thank You for your excellent insight & observations. Spot-on ✔
Outstanding analysis and explanation, thank you. How many more lives have to be lost before Boeing admits their approach to controls is fundamentally wrong (and has been for decades) and the Airbus way with flight envelope protection is the safest solution.
WHOA !!!
Scott do you think the current training requirements are sufficient that modern airline pilots are unlikely to input snap roll inputs when startled? On another note, someday perhaps you can make a video on windshear escape. I'm curious if your viewpoint is that the airliner escape strategy is also appropriate in the Bonanza or other light airplanes.
Good idea!
Always nice to hear lack of nuance and knowledge in laymen being stomped out by someone who knows more
These breakdowns are just so interesting
When I was learning to fly Cessna textbook said 5miles separation (large) and small aircraft not just (heavy) but large and heavy air traffic controllers try to get to many on the ground I think Saratoga at (RDU) n534z in with fatal crash was ten falte ass hole controllers felt they would given a lackluster runway IFR at night n534z and best runway was given to faster aircraft I have ben known to get a bit short with those boys . Your input would be apricated Thank You Good Podcast. PS PATCO was around back in the day'
It's presumptuous to say "a lot of folks have ... a less-than-perfect understanding that rudder hardover played" in US Air 427. I think the controversy, such as it was, came about from your description in the first video of the cause of the crash. You essentially said the cause was pilot error -- and that may well be, at least in part, a valid opinion.
The problem is you presented that opinion as fact. That didn't make sense to a lot of people WITHOUT also stating that the NTSB clearly and unambiguously determined that the crash was caused by the servo fault.
Agreed. And, with respect, it’s a pretty shitty thing to place much blame on pilots who have a key flight control doing the OPPOSITE of what it’s supposed to. That is setting a bar for recovery that no pilot can be expected to meet.
3.2 Probable Cause
The National Transportation Safety Board determines that the probable cause of the USAir flight 427 accident was a loss of control of the airplane resulting from the movement of the rudder surface to its blowdown limit. The rudder surface most likely deflected in a direction opposite to that commanded by the pilots as a result of a jam of the main rudder power control unit servo valve secondary slide to the servo valve housing offset from its neutral position and overtravel of the primary slide.
8:57 if you already know about the accident
So this is very similar to American 587 (Airbus A300 accident departing JFK)... excessive rudder inputs..
Would you please do a video on cross over speed.
The rudder of an aircraft is more effective at lower speed than are the ailerons for lateral control. As speed increases the ailerons/roll spoilers gain authority such that at a certain speed (crossover) they're able to counter rudder effects in roll control.....basically.
@@dutchflats this speed isn’t something that is published and would vary by aircraft and perhaps configuration. How does one determine this speed for a particular aircraft? Also when using aileron to counter roll, at least in light aircraft we add rudder to counter adverse Yaw. Fly wire mentioned unloading the aircraft (push). What impact does this have on crossover speed, if any?
@@imaPangolin We had published speeds for a given flap/configuration setting below which we couldn't bank the aircraft more than 15 degrees called floor speeds as well as minimum flap retraction speed, but they weren't crossover speed, a term which I only heard after the 427 accident although everyone probably knew it existed intuitively.
We tend to use less rudder in large jets during ordinary operations than do GA aircraft, applying gentle/subtle rudder pressure (if at all) other than during takeoff and landing. In addition, rudder travel is progressively limited automatically based on your airspeed, as your speed increases less rudder deflection is available.
I believe Fly Wire was getting at reducing G load to help the aircraft accelerate better so as to more quickly get above "cross over speed" and increase aileron effectiveness. With G on the plane you're losing energy to the turning motion which could be better used as airspeed. BTW, in aircraft like the A320 flap speeds, best lift over drag speed, stall speed, etc. are displayed on the PFD speed tape automatically and are calculated by the flight management system based on configuration and weight.
@@dutchflats I’m a large jet pilot as well. But I never knew this term and now some things make sense that prior I thought was only related to stall speed. We have a minimum maneuver speed for various flap configurations that really receive no explanation in training. Just don’t go below that speed. This speed is always above Vref by quite a margin. I’m pretty sure after this video and discussion that this is related to this “cross over speed”. I’m frankly a bit stunned at my ignorance.
@@imaPangolin Not as stunned as the guys flying 427 were I'm sure. Same with us regarding maneuvering speed, I'm retired now so some of this stuff is starting to escape me. Nowadays you just don't fly slower than the yellow stripe. One of the airplanes I flew had an actual angle of attack indicator which I loved, but haven't seen it on any other transport category jets?
The phrase that comes back watching this is "bounded rationality".
Another training element taught because of these hardovers, is an emphasis of unloading the wing as the first move when encountering an upset. Unloading lowers crossover speed and increases aileron effectiveness. Even though 427 was slightly below crossover, unloading would have given them more control… if they had known.
As far as untrained control inputs, examining the Nov 2001 NYC Airbus crash, wake vortices were encountered and the pilot flying put in the trained inputs of knocking down the wing with rudder and applied multiple full scale reversals until the vertical stab sheered. This training completely nullified Va protection and resulted in that crash. I don’t know what the training for the crews of 427 were for wake encounters, but I am making a general assumption that these two events not being that far apart in time may have had similar training.
Good video and thank you for making it.
Thanks Flyin and Drivin... gotta correct one thing about AA587. I went through the AAMP program the same way the FO did.... AND THAT WAS NOT A TRAINED REACTION. Period, he made it up on his own. And was told about it, still did it anyway. Repeat, the reaction of the FO was not a standard maneuver, old wives tale or whatever it was it was not taught.
@@FlyWirescottperdue I appreciate the correction. It was definitely a garbage technique. Now it’s unfortunately a training point for others. I always encourage my students to ask “why?” and this is a learning tool for exactly that scenario. Use the proper controls for the proper movement. I had a disagreement with some old timers who insisted on inappropriate rudder use with their excuse being “well in the F-4 they ruddered those for roll all the time.” A. I doubt that. B. We’re not flying F-4’s.
If you wouldn’t mind checking out my other channel: Spread Aviation. Thank you sir.
Make it so the rudder always works. Like in a Cub.
Kinda hard to do that in an airplane like this... lots of different issues than a Cub.
@@FlyWirescottperdue Stab trim off, yaw damper off, hydraulics off, manual reversion: not enough time for all that.
Scott please consider running for president.
HAHA, no way!
Do you review your completed video before you upload? You had a link blocking your intro
RUclips pouts stuff on top of videos... I have no control over that.