Turbinator
New member
Just imagine the pilots reaction:
News Copilot mistakes rudder trim for door lock, rolls 737
- Thread starter Ark
- Start date
Blacklight
New member
- Joined
- Jun 15, 2008
- Messages
- 259
- Reaction score
- 0
- Points
- 0
- Location
- In space ?
- Website
- www.myspace.com
I don't think this happened at all. It would have been all over CNN or the AP if it did. There would be passenger interviews, lawsuits, etc....
Turbinator
New member
According to this logic; there is no Sudan–SPLM-N conflict going on, and there has been no second Freedom Flotilla this August. I could go on, and on, and on. If the western media chooses not report on a world event, it does not mean it did not happen. This point perfectly demonstrates the ability of media to control a populations perception of the world.
.
Last edited:
Linguofreak
Well-known member
This happened on a flight from one Japanese city to another and didn't result in any deaths. Therefore, it doesn't seem too important to the international media.
That said, I was rather doubtful until I saw the Aviation Herald article.
ky
Director of Manned Spaceflight
This Is a credible news story
CBS NEWS: Cockpit error sent 737 into Pacific nose dive
ABC NEWS: Pilot Error Plunges Plane 6,234 Feet
CBS NEWS: Cockpit error sent 737 into Pacific nose dive
ABC NEWS: Pilot Error Plunges Plane 6,234 Feet
RisingFury
OBSP developer
Somewhat off topic...
You know, guinea pigs are much smarter than you give them credit for.
Box o' piggies:
The survivors of three litters about one week old in the photo.
This must have been the copilot:
You know, guinea pigs are much smarter than you give them credit for.
Box o' piggies:
The survivors of three litters about one week old in the photo.
Hi All,
New to Orbiter and the Forum. I would have liked my first post to have been a little more space related, but I am a 737-800 Captain for a US Airline. Cant help but throw my 2 cents in. Yes, the two knobs are very different. I would like to think that anybody qualified on the airplane could not get the two confused. But even if he did, I find it hard to believe that he would hold it in for that long. As soon as you move it, even just a little, you feel it. But let's see if we can figure this out.
The 737 series are not FBW, just good ol' fashioned cables and hydraulics. But, and this is an interesting point, the rudder is NOT connected to the autopilot system. Why Boeing designed it that way, I dont know. With the autopilot engaged, the yaw damper takes care of rudder input.
I assume (take that with a grain of salt) that at FL410, autopilot is engaged in LNAV and VNAV with Path indicated (Altitude Hold). Of course I'm also assuming that the CDU's for the FMC's are either Honeywell or GE. Not sure how ANA has their avionics configured.
At any rate, if rudder trim is applied, the plane will start to yaw slightly and roll slightly. No big deal. The autopilot will apply aileron and some yaw damper input to hold course and altitude. As more rudder trim is applied, obviously, the more opposite inputs the autopilot will try to make in order to maintain course and altitude. The autopilot cant do this all day. It will alert the crew that it's having a hard time with a series of beeps and the green VNAV and LNAV indications on the MFD's turn yellow with horizontal lines. Unless the crew does something to alleviate the condition that the autopilot is attempting to correct, the autopilot will disconnect.
It might be a little dramatic at 41,000 when the autopilot clicks off and all those control inputs are nulled...except for the rudder trim.
Consider this (again just an idea), Captain wants back in (we have different procedures in the US). FO applies rudder trim instead of the door switch knob. Captain's pulling on the door trying to get in. FO's looking at the door. Autopilot's making corrections, unknown to FO who has head turned around, but still cranking in trim. Captain's still pulling on the door. Now the autopilot beeps at him. FO turns around to see what's going on and while he's trying to figure it out (the plane is still straight and level, what could be wrong), the autopilot clicks off with an annoying alarm and WHAM you'll get tossed around a bit as all those control inputs return to normal... with the exception of the rudder trim.
At 41000, you'll lose some altitude and could end up with an unusual attitude recovery...again, not sure...never tried it....never will.
Anyway, thanks for listening. I'll post something soon on the meet and greet thread
David
oh and
New to Orbiter and the Forum. I would have liked my first post to have been a little more space related, but I am a 737-800 Captain for a US Airline. Cant help but throw my 2 cents in. Yes, the two knobs are very different. I would like to think that anybody qualified on the airplane could not get the two confused. But even if he did, I find it hard to believe that he would hold it in for that long. As soon as you move it, even just a little, you feel it. But let's see if we can figure this out.
The 737 series are not FBW, just good ol' fashioned cables and hydraulics. But, and this is an interesting point, the rudder is NOT connected to the autopilot system. Why Boeing designed it that way, I dont know. With the autopilot engaged, the yaw damper takes care of rudder input.
I assume (take that with a grain of salt) that at FL410, autopilot is engaged in LNAV and VNAV with Path indicated (Altitude Hold). Of course I'm also assuming that the CDU's for the FMC's are either Honeywell or GE. Not sure how ANA has their avionics configured.
At any rate, if rudder trim is applied, the plane will start to yaw slightly and roll slightly. No big deal. The autopilot will apply aileron and some yaw damper input to hold course and altitude. As more rudder trim is applied, obviously, the more opposite inputs the autopilot will try to make in order to maintain course and altitude. The autopilot cant do this all day. It will alert the crew that it's having a hard time with a series of beeps and the green VNAV and LNAV indications on the MFD's turn yellow with horizontal lines. Unless the crew does something to alleviate the condition that the autopilot is attempting to correct, the autopilot will disconnect.
It might be a little dramatic at 41,000 when the autopilot clicks off and all those control inputs are nulled...except for the rudder trim.
Consider this (again just an idea), Captain wants back in (we have different procedures in the US). FO applies rudder trim instead of the door switch knob. Captain's pulling on the door trying to get in. FO's looking at the door. Autopilot's making corrections, unknown to FO who has head turned around, but still cranking in trim. Captain's still pulling on the door. Now the autopilot beeps at him. FO turns around to see what's going on and while he's trying to figure it out (the plane is still straight and level, what could be wrong), the autopilot clicks off with an annoying alarm and WHAM you'll get tossed around a bit as all those control inputs return to normal... with the exception of the rudder trim.
At 41000, you'll lose some altitude and could end up with an unusual attitude recovery...again, not sure...never tried it....never will.
Anyway, thanks for listening. I'll post something soon on the meet and greet thread
David
oh and
FSXHD
Heterotroph
- Joined
- Jan 17, 2010
- Messages
- 106
- Reaction score
- 0
- Points
- 0
NGX Simulation of Events
Howdy folks!
Tonight I had some time to try this in FSX with the PMDG NGX, one of the most realistic addons ever developed for a consumer flight simulator.
Intro
For the test aircraft, I chose the -800 variant with winglets and a shiny chrome livery (PMDG has not released their -700/-600 series yet). I took into account many factors to try and get this right. I calculated that the fuel (including reserves/alternate) from Okinawa to Tokyo would be about 19500 pounds. I loaded up 16000 lbs. in the simulator in order to achieve a realistic amount of fuel that would be present during the test as I would not be flying the whole flight. I put the weight equivalent to 130 passengers and approximately 4000 pounds of baggage/cargo. My plan was to fly from Paine Field (Boeing Factory) out to the coast and Pacific Ocean where the tests would be done and back.
At the Ramp
Takeoff
Climbing
The Knobs in Question
As you can see, the knobs are indeed close together and have somewhat the same shape, but they are still obviously different. The PIC during the mishap must of had an extreme lapse in mental consciousness to make the mistake he made.
The Test
The 737 was on VNAV PTH and heading hold modes with the FMC controlling the speed through VNAV. The test was at FL410 as stated in the articles. I altered the trim in two degree intervals. As David said earlier, the rudder trim is not connected to the autopilot system. Because of this, the trim can be handled to a point, but once the deflection limits of the ailerons were reached, the airplane began to roll. The autopilot could handle the trim difference until about 2 degrees. Once past this, a shallow bank angle was maintained of around 7 or 8 degrees. Rudder trim was at about 2 or 3 degrees.
As the trim was increased, the bank angle increased.
Once the trim got to about 5-7 degrees the airplane started a continuous roll.
I paused to get a shot of the rudder trim window. It is listing 7 degrees left.
The plane lost altitude quickly and I recovered around 36,500 feet (-5500ft), very similar to the -6300ft lost in the real flight.
After this I flew her back to Paine Field. Here is a landing shot for kicks.
The results seem very consistent with what happened in the real flight. The PIC during the event must of held the trim knob for quite a while cause it took about 10 seconds to get to critical position where the continuous roll occurred. I'm not sure exactly what happened, but the PIC has some questions to answer.
If you have any questions, feel free to ask!
By the way, here is the full album on imgur as I didn't post all of the shots - http://fsxhd.imgur.com/ana_737ngx_roll_test
Regards,
Kevin B
Howdy folks!
Tonight I had some time to try this in FSX with the PMDG NGX, one of the most realistic addons ever developed for a consumer flight simulator.
Intro
For the test aircraft, I chose the -800 variant with winglets and a shiny chrome livery (PMDG has not released their -700/-600 series yet). I took into account many factors to try and get this right. I calculated that the fuel (including reserves/alternate) from Okinawa to Tokyo would be about 19500 pounds. I loaded up 16000 lbs. in the simulator in order to achieve a realistic amount of fuel that would be present during the test as I would not be flying the whole flight. I put the weight equivalent to 130 passengers and approximately 4000 pounds of baggage/cargo. My plan was to fly from Paine Field (Boeing Factory) out to the coast and Pacific Ocean where the tests would be done and back.
At the Ramp
Takeoff
Climbing
The Knobs in Question
As you can see, the knobs are indeed close together and have somewhat the same shape, but they are still obviously different. The PIC during the mishap must of had an extreme lapse in mental consciousness to make the mistake he made.
The Test
The 737 was on VNAV PTH and heading hold modes with the FMC controlling the speed through VNAV. The test was at FL410 as stated in the articles. I altered the trim in two degree intervals. As David said earlier, the rudder trim is not connected to the autopilot system. Because of this, the trim can be handled to a point, but once the deflection limits of the ailerons were reached, the airplane began to roll. The autopilot could handle the trim difference until about 2 degrees. Once past this, a shallow bank angle was maintained of around 7 or 8 degrees. Rudder trim was at about 2 or 3 degrees.
As the trim was increased, the bank angle increased.
Once the trim got to about 5-7 degrees the airplane started a continuous roll.
I paused to get a shot of the rudder trim window. It is listing 7 degrees left.
The plane lost altitude quickly and I recovered around 36,500 feet (-5500ft), very similar to the -6300ft lost in the real flight.
After this I flew her back to Paine Field. Here is a landing shot for kicks.
The results seem very consistent with what happened in the real flight. The PIC during the event must of held the trim knob for quite a while cause it took about 10 seconds to get to critical position where the continuous roll occurred. I'm not sure exactly what happened, but the PIC has some questions to answer.
If you have any questions, feel free to ask!
By the way, here is the full album on imgur as I didn't post all of the shots - http://fsxhd.imgur.com/ana_737ngx_roll_test
Regards,
Kevin B
Hi again,
Kevin, really nice job. Im surprised the autopilot didn't kick off....course, having never tried something like this, wouldn't quite know what to expect. It must have been pretty insidious at night.
Thanks,
Dave
Kevin, really nice job. Im surprised the autopilot didn't kick off....course, having never tried something like this, wouldn't quite know what to expect. It must have been pretty insidious at night.
Thanks,
Dave
Ark
New member
- Joined
- Jan 31, 2009
- Messages
- 2,200
- Reaction score
- 0
- Points
- 0
More like "It's not working, better twist harder!"
Turbinator
New member
Apology by the executives, and an animation showing the flightpath:
http://www.reuters.com/video/2011/09/30/japan-airline-apologises-for-plane-flop?videoId=222108745
It is interesting to see that once the plane starts to roll, there is a brief moment where it resists and even slightly rolls back. Then, all of a sudden, it goes almost completely over. This seems to perfectly fit the earlier description in this thread of the behavior of the autopilot, and the resulting consequences of that behavior.
.
http://www.reuters.com/video/2011/09/30/japan-airline-apologises-for-plane-flop?videoId=222108745
It is interesting to see that once the plane starts to roll, there is a brief moment where it resists and even slightly rolls back. Then, all of a sudden, it goes almost completely over. This seems to perfectly fit the earlier description in this thread of the behavior of the autopilot, and the resulting consequences of that behavior.
.
Last edited:
Linguofreak
Well-known member
One thing that I haven't seen mentioned yet in this thread is that the incident happened in the middle of the night, so the crew very well may not have had a good external horizon reference.
As I've learned more about this, my opinion of it has gone from a near-impossibility most likely resulting from gross negligence to something that could very plausibly be caused by a fairly minor mistake.
As I've learned more about this, my opinion of it has gone from a near-impossibility most likely resulting from gross negligence to something that could very plausibly be caused by a fairly minor mistake.
