737 Rudder Story
Вставка
- Опубліковано 20 лип 2023
- In this video I explain the rudder issues that beset the 737 in the 1990s. I take you through the accidents, investigations, contrasting opinions, design changes and the many QRH changes that the 737 rudder system has gone through, that all started with two mysterious accidents in the early 1990s for which no conclusive evidence was ever found. I also explain how those events led to a complete reformat of the QRH, the introduction of upset recovery training, and FDR improvements.
Contents
0:46 UA585, Colorado Springs
6:29 The Original Rudder System
26:20 The Second Accident US427, Pittsburgh
27:42 Critical Design Review
30:42 Boeing Yaw Damper Article
33:27 Eastwind 517
34:31 NTSB Intervention
36:43 QRH Change
38:58 Blowdown & Crossover Speed
56:20 Reaction to the NTSB Recommendations
1:02:00 More ADs
1:03:32 First Design Changes
1:11:37 Two More Incidents
1:13:51 US427 Report
1:22:59 Block Speeds Increased
1:26:22 Engineering Test and Evaluation Board
1:35:35 Rudder System Enhancement Program
1:50:11 Evolution of QRH Procedures
1:58:12 FDR Development
2:09:30 Flight Ops Tech Bulletin
2:11:33 Final Thoughts - Наука та технологія
Excellent video- the best I have ever seen on this topic.
(I have well over 12k hours in command on 737 300/400/800, and still found this fascinating.)
Thank you, glad you found it useful
Well, after the APU video, I couldn't help but click on this one!
I used to be in charge of a large datacenter with a 99.999% uptime guarantee, which meant any downtime was very expensive to the company. Not life-and-death at all, but it was still routinely stressful enough that I extracted myself from the industry to preserve my mental health.
It's really interesting seeing some of the parallels between network outages and plane problems; both are complex systems, the operation of both is to a degree "opaque" to the operator-a pilot can't get out and look at the rudder, and a network operator can't see the packets moving on the wire. So you're at the mercy of your system's ability to provide diagnostic information about itself.
I couldn't help but have my mind blown to find out that something like a rudder position display is optional equipment. I can't imagine how hard it must be to troubleshoot a problem _under duress_ without even basic troubleshooting information. I get that the idea with planes is that every possibility should be covered by checklists that do the deep systems-level thinking for you, but it seems like many problems start when the checklist is exhausted and there simply isn't enough information available to begin to understand the problem. Going back to my network operator analogy, when I found myself in those situations, I would have to resort to extremes like "let's unplug it and see what happens".
I remember a few occasions feeling so frustrated by being blind to certain aspects of my system and wishing I had spent more time "on the ground" improving visibility into certain aspects of the network. I also remember on occasion lashing up some spare equipment to help characterize an intermittent event in the system when it happened. It's crazy that Boeing, the NTSB and the FAA were making life-or-death decisions based off of anecdotes and 5 channels of FDR data. I know that airplane hardware is incredibly regulated, but I also wonder how different the outcomes would have been if Boeing, the NTSB and the FAA had come up with a low-cost solution that could be quickly installed on all affected units to passively monitor the PCU that would at least provide diagnostic data in the event of another incident. In the words of the robot Johnny 5, "More Input, more Input!"
Anyway, a very interesting video. I learned a lot!
Some very interesting parallels indeed. At least they eventually got to the bottom of it.
The cross over demo in the sim is fantastic! Thank you as always, and special thanks for showing it in the sim 👍
You are very welcome.
Nice investigative work, Chris. It's a bit creepy to think that this one bit of design failure lasted over ten years. Nice to know that that the safety authorities finally figured shit out.
Hopefully with the advances in FDR it couldn’t last so long these days.
I asked a TRE once if we could simulate a rudder hardover in the sim. I unknowingly encountered "crossover speed" during final as the speed was coming down. What a scary feeling to lose control authority!
If you get chance again, when you're at the limit, try pushing and pulling the stick (so decreasing/increasing AoA briefly). You should see an immediate response, regardless of your speed (the higher the AoA, the more the rudder rules roll, as you lower AoA, the airlerons will gain more authority).
There was a crash where this effect was not understood (I should have info to hand, but I don't) where the rudder was lost hard over, and the pilot coming in, on approach did the natural thing, and didn't concern with keeping speed up and AoA down. They got a touch slow, pull back and with full roll full authority running out, they pull back even more, not realising in that moment they sealed the fate of the aircraft and all those onboard.
For more info see safety pilot at AA video on "control malfunctions & flight anomalies" it's fascinating stuff from 25-30 years ago
Congratulations for this amazing video, Chris. Excellent research and material! Anxious for the next
You are very welcome
Wonderful as always, thanks @ChrisBrady737!
My pleasure, thanks again for watching
Thank you captain for this excellent coverage of the 737 rudder issue 👍
You are very welcome, thanks for watching.
Dear Chris
I truly enjoyed watching your presentation. As a new 737 pilot. I appreciate your detailed presentation of the 737 Rudder Story, it was truly educational with abundance of good information. I was relatively familiare with the above-mentioned two accidents in Colorado Springs and Pittsburgh. However you've presented lot more facts that I was not aware of.
Thank you
Thank you for your kind words, I am glad you found the video interesting.
I am learning the 737 currently, and have found your videos to most informative and "fill in the gaps" with my computer based training.One thing I really wish was there is an easy way to print out the system schematics. I am more of a visual learner in systems and it helps for me to review the systems that way. Thanks for producing these videos!
Great to hear that you are finding the videos useful. Many of the schematics are available in my book. www.b737.org.uk/book.htm
Amazing job Chris 👏👏👏 really interesting . Many thanks !!
You are very welcome
Brilliant history and technical insight! Thank you Chris 🙏
My pleasure Tony, thanks for watching.
We did hard slam reversals on these rudders in the hangar in the 90's, which consisted of one of us in the flight deck and another at the rudder connected by radio. I think it was either an AD or a SB. But I remember also disconnecting the PFCU input rod, applying hydraulics and throwing the spool valve rapidly from side to side with my fingers in addition to the above. We found nothing! We clearly could not replicate the thermal issues which Boeing considered a factor at that time.
Undoubtedly the failure mode was a very rare occurrence and a lot of things had to come together for it to happen. As flightcrew we also used to do a rapid full and free rudder check whilst taxying out during the 90's - never can be too careful!!
Really enjoyed this. My background is aerospace system safety. In aircraft design we are looking to comprehend the safety risks and manage them so they are acceptably low.
Reversed function (one of a number of erroneous function failure conditions) is an important functional failure to be managed as well as loss, partial loss and inadvertent function.
Chris,
Again a GREAT video!
I read myself the NTSB report years ago and its very intersting that they found no specific failed part of the pcu that they can point the finger. Only suspect part despite they tested the hell out from the pcus. Only theories but no hard evidence.
By the way the orginial pcu was far away to call it redundant design. The enhanced one yes and no. Still 1 unit but internally totally separated A and B channels. But not like elevator or aileron when you have two separate units.
Again congratulations for your work!
Thank you Balázs, the design and accident history of the 737 rudder is indeed a fascinating story.
While I though I kinda know this story for some extended, you take us to the whole new level of detail! Tank you!
One of the more recent rudder related issue if you may is ANA140 in 2011 when FO accidentally moved the rudder trim to a "hardover",although the investigation deemed the recovery by FO was insufficient, he still managed to make it in one piece, shows how far we have come in terms of UPRT.
One small question: just wonder if there is any known incident that FFM have been triggered in line operation?
Hi Umi, I am familiar with the ANA140 event where the rudder trim was mistaken for the door unlock. One of the big differences with that event and these is that ANA was at a much higher airspeed so was able to recover relatively easily.
Great Effort, Thank you
Thank you very much, that is very generous of you.
Dear Chris,
It's always a pleasure to see the notification saying that a new video is available. Sharing your knowledge for young pilot is invaluable. Thank you and the video is excellent.
You are very welcome, glad you liked it.
Thank you so much for great videos,coming from 320,I find it very useful and helpful
You are welcome. Thanks for watching.
A key aspect of dual failures is that the first failure mode is often a "latent failure" that exists for several flying hours prior to the incident, and then the second failure mode occurs acutely at the time of the accident itself. This means that the two failures don't need to begin at the same time, and the probability of the double failure increases dramatically. The way to correct this is for engineers to design the control system to intentionally lose some (but not a dangerous amount) of functionality when the first failure occurs, so that the pilots will become aware of the first failure before the second failure has a chance to occur. This is hard to understand, so let me provide an example, when the new RSEP servo valve was rolled out, both of the two linkages connected to the main PCU had jam override couplings where the torque tube meets each rod. This means that, if one valve jams, the rudder pedals immediately require 18lb of extra pressure to move the rudder the same amount. This creates an immediate indication to the pilots that something is wrong, but does not place the aircraft in any immediate danger. In this way, the airline is forced to perform an immediate inspection of the PCU servo valve, which will inevitably catch the single failure before there is any significant chance of a double failure developing.
I watch Mayday. And after watching the one on this issue thought I knew what the cause and fix was. WOW great presentation! Looks like it was way more complicated and had much bigger consequences for the industry. Thanks for all your hard work on this videos!!
My pleasure. Glad you enjoyed it.
At 50:30 .... Boeing Knew from mid 1980's Fight Testing that an Uncorrectable Rudder over Aileron Control crossover Speed in the Normal flight Envelope regime Existed but since the Existing FAA cert requirements Didn't specifically List or Address this exact Control anomaly, Boeing Didn't Reveal it During the Certification process....
I’m a retired virtual airline pilot and now fly virtual GA airplanes for enjoyment in my retirement .. but find these videos very useful as I ponder coming out of retirement to fly virtual 737’s again for Southwest Virtual Airlines .. so thanks 👍
Great to hear that you are enjoying the videos
I upgraded to Captain flying -300/500 in 1992, and was pretty well immersed in the string of events and responses over the succeeding ten years. The company explanation of cold soaked PCU susceptibility to anomalies was never particularly convincing, though it resulted in redesign. The various training in upsets and additional QRH procedures were enlightening, but, as I recall, excessive simulator wear tempered upset training. I don’t recall ever learning about blowdown or crossover speeds. We did implement the 10kt increase in maneuvering speeds…. (cya)
Hi Captain,
Thank you as always for your high quality educational videos, I am hooked to this channel! Would you by any chance have any documents or videos that could perhaps explain the difference between a power control unit and standard hydraulic actuator found on controls such as the flight spoilers? I'm mainly interested in knowing how the mechanism works for delivering the correct quantity of fluids into the actuator for a given flight spoiler deflection before stopping more fluids from entering. Thank you as always, and keep up the great work I look forward to your next series of videos!
Hi, thanks for your kind comments. Unfortunately I don't have anything that specific. However I would describe the main difference between a PCU and an actuator is that an actuator is usually used to move something to a discrete position. E.g. landing gear, flaps, ground spoilers. A PCU is much more sophisticated than a simple actuator because it takes a variable input, such as control column or rudder pedal inputs and gives a proportional output. Hope that helps. Chris
Thank you so much for this elaborate explanation! I always wondered how they fixed the issue as documentaries don’t really do a great job of going into technical depths (naturally)
My pleasure, glad it help your understanding of the issues and fixes.
Another great video, thanks a lot. As a 737 pilot in training I would love to see your take on a professionally executed Walk around. What experiences and knowledge may have added or changed on your way of executing those.
Thanks for the suggestion, I will add it to the list 👍
good work
Thank you
Was just talking to my Capt about the 737 control reversal and there you are have a video for this lol thank you! And this crossover and blowdown also reminds me all the Vmc manevours might also has something to do with this?
Glad my timing was good for you 👍
@@ChrisBrady737 just cruise… how does FBW with side-sticks comply the 25.203 like airbus families since the controls don’t have any force feedback?
The sidestick on the Airbus family still require a positive longitudinal control force to approach the stall so they are compliant.
Chris: It appears we will need an update on the 737 rudder system based on the United Airlines stuck rudder issue of late...
Yes, I am watching that investigation with interest. ATM there is a question about the servo motors performance in cold soak conditions.
Thanks for yet another wonderful episode captain!
As a kid back in 2005 I became fascinated by the mechanics of airliners and stumbled across your website, I distinctly recall reading a whole section dedicated to "Rudder News" back then. It almost feels nostalgic to re-visit the same content in 2023 in the video format.
Ah yes I remember writing that page as a response to the many design and QRH changes to the rudder at the time. It was a nostalgic experience for me to make this video. The page still exists but this video definitely expands on the story somewhat.
got to 1:50:00 and was still interested. well done on keeping technical blurb less than drone on boring. i was also imagining the last seconds of those two accidents and the feeling of woe those pilots must have had.which reminds me, do the 737's now have the auto engine control system which allows directional control in the event of total loss of hydraulics?? by engine power control only.. re: United Airlines Flight 232
Thank you for your kind comments, glad I was able to hold your interest. No, the 737 does not have an engine directional control system. The 777 has a TAC which assists with rudder input if an engine fails but not the venerable 737.
oooh. this is very educational: the US427 is like the F-35 failure upon approach last year 2023 google ", the MA banked sharply with the left-wing down (Tab J-1.22). The MP re-applied full roll stick at approximately 18:08:35L to counter the aggressive MA bank to the left"
When I look at the hardware in a 737, im not thinking about technology but steampunk 🙂. Love the content anyway .
The "crossover" being shown at minute 44 is also known as crossover AoA, and is entirely dependent on AoA. The rudder is the most powerful roll device at high AoA, in all aircraft, on a good day. If you roll airbus at a slow speed and have some way to see (on SIM, or from outside) you'll see the rudder moves as much as the airlerons, even though the pilot didn't command any rudder. I realise we're talking about Boeing here, and it's arguable that airbus isn't so much a plane as it is a computer with wings, but the areodynamic principles remain, and they exploit these PoF with hardwired code into their FCS to do make us of these effects.
These videos are fascinating Chris.
(I got the above information from AA's advanced manoeuvring program, by CPT Vanderburgh, his old presentations are online, even though they don't specifically reference 737s)
It is interesting but actually incorrect, crossover speed is not entirely dependent on AoA, it is dependent on several factors, weight, config, CG, AoA, G, etc
@@ChrisBrady737 that's interesting, because this is not my experience at all.
Try holding the wings level until max roll authority begins to lose it, and push and pull the control, the response is instantaneous... In other words the things you've listed, weight, speed CG haven't changed, only the AoA.
Please try it, but also link me to any material that shows where the elements youve listed can impact this effect, with AoA remaining the same.
Again as I understand it, the ONLY thing to affect this crossover effect, is the AoA, clearly changing weights and CG or configuration is going to affect AoA, but when the same AoA is reached under the new config, in my experience you'll lose it exactly the same place.
@@CameTo i think we are saying the same thing. For a given config, weight, cg, g loading the only remaining variable to affect crossover is alpha. But just like Vs, Vcrossover will vary with the above factors because it is a lift issue.
You can not compare the 737 flight controls to the Airbis fly by wire design. You have turn coordination on the A320 which means automatic rudder (plus automatic stabilizer too) when you roll the plane using the side stick. But only in Normal control law.
On the 737NG you have also automatic rudder called Wheel To Rudder Interconnection system. Only when you do manual reversion. Its using standby hydraulic pcu to move the rudder under 0.45 Mach.
I think it was the fluid temperature variation along with a bit of moisture forming ice crystals. It’s been a long time since I read the reports, but I thought they got it to jam and reverse in the lab with extreme different temps of the pcu and fluid, that could be encountered in the place coming down from altitude.
Non pilot here, just 737 NG enthusiast in x-plane. My question is that it would seem this situation (being close to x-over spd) is closer to happening during an engine out situation?
Engine out or upset recovery situations.
Chris, could the pilots of USAir 427 have recovered their plane looking back in hindsight? or was it impossible considering their altitude of 6000ft. The startle factor must have been enormous.
Worked Piedmont/ Usair ever night we had a a/c in the tail dock doing the AD we never had a failure
Hello Chris....is it safe to say with regards to the FFM and an opposing pressure......
Main Rudder PCU is turned off with A and B Hydraulics out of the picture and Standby Rudder PCU is activated and solely operating the rudder for the rest of the flight....
Or Standby PCU assist the correct working system not jammed to help control the rudder the rest of the flight...
Sorry to ask...got myself in the weeds on this one and the FM for my airlines goes very basic in its explanation.
Thanks for all you do Chris i really appreciate your book, videos and insight on all 737!
Cheers
GUS
Hi Gus,
Your first scenario is correct. If the FFM detects opposing pressure, the rudder will then only be powered by the Standby PCU and the STBY RUD ON light will illuminate.
Great to hear that you like the book.
Chris
@@ChrisBrady737 Thanks so much for clarifying Chris!!!!!!
Have you done a tech guide on ATA22 auto flight?
Not yet. It is not a subject that lends itself to this format but I will keep it in mind.
Hey! Does the objective and condition in the memory item non normal checklist need to be memorized?
No, it is written in the QRH so that you do not have to memorise it.
Another question, Chris - which app do you think is better during Type Rating - 737 Handbook or 737 MRG?
I don’t have either books so I can’t comment.
So if you had a rudder hard over and gotten above the cross over speed and got it under control then what? Stay above 190 kts the whole flight even for landing? Or would you turn off all hyd pumps and land with manual revision?
Please ask any operational or technical questions here: facebook.com/groups/737tech
I wonder if the "blowdown" term was borrowed from the sailboat world. I suspect that every beginner sailboater has been blown down because of too much wind + too much sail (presented to the wind).
What is your 2nd favourite aircraft? (Is it the A220)?
The BAe Hawk!
Chris, please make your book on Apple Books available in Kenya.
Hi, unfortunately I don’t have any control over which countries Apple sells the book. I think they sell in over 50 countries, hopefully Kenya will be one of those soon.
Have you considered a hard copy?
Ooh, okay. Thanks.
Hardcopy would be a bit steep in price once we factor shipping and local taxes.
The link to the hard copy is here:
www.blurb.co.uk/b/11658762-the-boeing-737-technical-guide
Airbus A320s have Rudder and Elevator problems.
Source; Wikipedia
On 28 December 2014, an Airbus A320-216 registered as PK-AXC, crashed into the Java Sea killing all 162 on board.
The cause was initially a malfunction in two of the plane's rudder travel limiter units.
On 5 November 2014, Flight 1829, an Airbus A321-231 registered as D-AIDP, was flying from Bilbao to Munich when the aircraft, while on autopilot, lowered the nose into a descent reaching 4000 fpm. The uncommanded pitch-down was caused by two angle of attack sensors that were jammed in their positions, causing the fly by wire protection to believe the aircraft entered a stall while it climbed through FL310.
On 31 October 2015, an Airbus A321-231 registered as EI-ETJ operated by Russian airline company Kogalymavia crashed in the Hasana area of central Sinai, Egypt while en route from Sharm El Sheikh, Egypt to St. Petersburg, Russia. There were 224 people on board (217 passengers and 7 crew). There were no survivors.
EASA Extends A320ceo Wing Fatigue Cracking Inspections To All A320neo Models
Imagine a wing or even half of a wing breaking off while you are flying along in an Airbus A320-!
There are dozens of ADs for inspections of cracks on all aircraft types.