Continental 737 Off Runway at DEN
Join Date: Jan 2009
Location: alameda
Posts: 1,053
Likes: 0
Received 0 Likes
on
0 Posts
I've experienced crosswind components right up to our limits...29 knots direct.
In smallerplanes I've landed when winds were reported at 60 knots...but so variable as to not be able to calculate the exact crosswind.
In a metroliner, I also landed in a 60 knot wind, with unknown crosswind, but quite bumpy.
I have a feeling that modern pilots are losing the ''art'' of flying, including crosswind takeoffs.
I can't imagine why pilots aren't using traditional crosswind techniques...wheel to keep wing down or at worst level into wind, rudder to keep going straight
In smallerplanes I've landed when winds were reported at 60 knots...but so variable as to not be able to calculate the exact crosswind.
In a metroliner, I also landed in a 60 knot wind, with unknown crosswind, but quite bumpy.
I have a feeling that modern pilots are losing the ''art'' of flying, including crosswind takeoffs.
I can't imagine why pilots aren't using traditional crosswind techniques...wheel to keep wing down or at worst level into wind, rudder to keep going straight
IMHO, many pilots, including some posting in this forum, are forgetting the difference between what they apparently can do and what they should do.
The recognition of this difference is described as good judgment.
Good judgment in this instance would be that you do not fly at the limit condition where risks could be high; instead, by considering a range of variables including the possible overestimation of self-capability allow a suitable margin for ‘safety’.
The recognition of this difference is described as good judgment.
Good judgment in this instance would be that you do not fly at the limit condition where risks could be high; instead, by considering a range of variables including the possible overestimation of self-capability allow a suitable margin for ‘safety’.
Join Date: Jan 2008
Location: Sonoma, CA, USA
Age: 79
Posts: 143
Likes: 0
Received 0 Likes
on
0 Posts
Crosswind Technique
I learned in taildraggers and continue to fly them. The x-wind techniques we learned in the Cub, DC-3 and Beech-18 won't work in certain modern jets because dragging an engine pod is a real possibility.
As for differential power, I used it a lot in the DC-3 and Beech-18. Below 100 kts. IAS there was no danger of back-loading an engine, so in strong x-winds I would retard the downwind throttle and fly the approach with the upwind engine. Sometimes this saved wear and tear on my leg because I didn't need to hold a lot of rudder. Other times it required rudder, engine and a higher than normal touchdown speeds to maintain rudder effectiveness. Landings at LAX during strong Santa Ana winds were a lot of fun.
In gusty conditions, unlike jets, we had almost instantaneous thrust when needed from the retarded engine.
The main landing gear was also in line with the props, so there was no worry about dragging a prop.
Different time; different animals.
I do remember one beautiful landing at LGA while riding in an AA 722. The Captain set her down firmly but smoothly on the right main only with the wing tip about two feet off the ground after a very rough approach. Again, no engine pods to worry about.
As for differential power, I used it a lot in the DC-3 and Beech-18. Below 100 kts. IAS there was no danger of back-loading an engine, so in strong x-winds I would retard the downwind throttle and fly the approach with the upwind engine. Sometimes this saved wear and tear on my leg because I didn't need to hold a lot of rudder. Other times it required rudder, engine and a higher than normal touchdown speeds to maintain rudder effectiveness. Landings at LAX during strong Santa Ana winds were a lot of fun.
In gusty conditions, unlike jets, we had almost instantaneous thrust when needed from the retarded engine.
The main landing gear was also in line with the props, so there was no worry about dragging a prop.
Different time; different animals.
I do remember one beautiful landing at LGA while riding in an AA 722. The Captain set her down firmly but smoothly on the right main only with the wing tip about two feet off the ground after a very rough approach. Again, no engine pods to worry about.
Join Date: Mar 2009
Location: Feriton
Posts: 73
Likes: 0
Received 0 Likes
on
0 Posts
The NTSB has released a factual report on Continental 1404:
Accident: Continental B735 at Denver on Dec 20th 2008, veered off departure runway and burst into flames
Not any shockers here. Interesting bit about a guy driving a pickup nearby:
Accident: Continental B735 at Denver on Dec 20th 2008, veered off departure runway and burst into flames
Not any shockers here. Interesting bit about a guy driving a pickup nearby:
A driver of pickup travelling north on the E-470 had just passed through the toll booth 3nm west of the threshold runway 34R reported an extremely large gust coming from the northeast, that he believed would blow his pickup over. The gust lasted for about 2-3 minutes and stopped as abruptly as it arrived. The driver estimated the gust in excess of 50 (statute) miles per hour. The driver observed, that the temperature gauge of his pickup had shown just above 0 degrees (presumably Centigrade, not specified in the report), but showed 20 degrees after the gust stopped.
Join Date: Feb 2006
Location: USA
Posts: 487
Likes: 0
Received 0 Likes
on
0 Posts
Join Date: Nov 2006
Location: SoCalif
Posts: 896
Likes: 0
Received 0 Likes
on
0 Posts
Back when NCAR, National Center for Atmospheric Research, were testing NexRad in Denver, guess where they saw the most microbursts?
Yep, right where they later built KDEN.
GB
Yep, right where they later built KDEN.
GB
Winglets reduced X-wind limit?
The Denver Post report on the NTSB release of July 16 quotes the NTSB as saying that "the manufacturer and installer of winglets that were on the airplane...had published 'a maximum demonstrated crosswind component of 22 knots for winglet-equipped B-737-500s.' "
Here is the reference to 22 knots in NTSB paperwork:
"In a supplemental type certificate report, Aero Tec (on behalf of Aviation Partners Boeing, the manufacturer and installer of winglets installed on the accident airplane) subsequently published a maximum demonstrated crosswind component of 22 knots for winglet-equipped B-737-500s."
---from the document: http://www.ntsb.gov/Dockets/Aviation...021/421386.pdf - page 20, just below the table of xwind components from CAL's 737 Flight Manual.
Same page goes on to note that Boeing's most recent crosswind guidelines for the B-737 had not been updated since 1996.
Denver Post article: Winds surprised crew on DIA flight - The Denver Post
Manufacturer/installer is Aviation Partners Boeing
Winglets were installed about 2 weeks [correction: 3 weeks] before the accident (Nov. 29, 2008): http://www.ntsb.gov/Dockets/Aviation...021/422045.pdf
(page 2, second item from bottom in section titled "Major repairs and alterations list"
CAL's operational manual listed 33 knots Xwind limit: http://www.ntsb.gov/Dockets/Aviation...021/421403.pdf
Here is the reference to 22 knots in NTSB paperwork:
"In a supplemental type certificate report, Aero Tec (on behalf of Aviation Partners Boeing, the manufacturer and installer of winglets installed on the accident airplane) subsequently published a maximum demonstrated crosswind component of 22 knots for winglet-equipped B-737-500s."
---from the document: http://www.ntsb.gov/Dockets/Aviation...021/421386.pdf - page 20, just below the table of xwind components from CAL's 737 Flight Manual.
Same page goes on to note that Boeing's most recent crosswind guidelines for the B-737 had not been updated since 1996.
Denver Post article: Winds surprised crew on DIA flight - The Denver Post
Manufacturer/installer is Aviation Partners Boeing
Winglets were installed about 2 weeks [correction: 3 weeks] before the accident (Nov. 29, 2008): http://www.ntsb.gov/Dockets/Aviation...021/422045.pdf
(page 2, second item from bottom in section titled "Major repairs and alterations list"
CAL's operational manual listed 33 knots Xwind limit: http://www.ntsb.gov/Dockets/Aviation...021/421403.pdf
Last edited by pattern_is_full; 18th Jul 2009 at 18:32.
Join Date: Mar 2009
Location: Feriton
Posts: 73
Likes: 0
Received 0 Likes
on
0 Posts
Wow, good find, pattern. This seems like a very significant development. It was mentioned in another article I read on the "docket" that 735s are more susceptible to weathervaning because of their short length. And with winglets catching wind also................ Now I am very curious what the crosswind component is of other 737 variants (most of which are considerably longer that the 735) with winglets. Those would be very interesting numbers.
Winglets with a x-wind
Can any CAL pilots tell us if and how a 737-500W handles differently to a 737-500, particularly in a cross-wind?
Re the lower max demonstrated x-wind, I enquired to APB several years ago about the reduction when fitted with winglets and they replied that it was simply that there had not been any greater x-wind conditions present to demonstrate in during certification!
Re the lower max demonstrated x-wind, I enquired to APB several years ago about the reduction when fitted with winglets and they replied that it was simply that there had not been any greater x-wind conditions present to demonstrate in during certification!
Join Date: Mar 2009
Location: Feriton
Posts: 73
Likes: 0
Received 0 Likes
on
0 Posts
An article in the Wall Street Journal, by Andy Pasztor, who seems to have the best info on airline accidents. Several new tidbits, including this:
Data Suggest Crew in Denver Crash Failed to Respond to Wind Gusts - WSJ.com
No mention of winglets though.
The safety board's investigation, which hasn't yet established the probable cause of the accident, also is focusing on whether the design of the Boeing 737-500 model involved in the crash makes it particularly hard to control on the ground in such windy conditions.
Because of its relatively large tail surface compared to its length, pilots and safety experts believe such models may be more likely than other 737 versions to be blown around on the ground by crosswinds.
Toby Carroll, Continental's director of flight safety, told NTSB investigators that the airline was worried enough about the handling characteristics of 737-500s to consider restricting their use in Denver. The main airport there is notorious among aviators for stiff and gusty winds, especially during winter months.
Because of its relatively large tail surface compared to its length, pilots and safety experts believe such models may be more likely than other 737 versions to be blown around on the ground by crosswinds.
Toby Carroll, Continental's director of flight safety, told NTSB investigators that the airline was worried enough about the handling characteristics of 737-500s to consider restricting their use in Denver. The main airport there is notorious among aviators for stiff and gusty winds, especially during winter months.
No mention of winglets though.
Join Date: Apr 2009
Location: Petaluma
Posts: 330
Likes: 0
Received 0 Likes
on
0 Posts
I think short coupling is a non issue. Given that the a/c has sensitive directional moment, it works both ways. May be more sensitive, but not disqualifying for certification, obviously. Winglets are a cruise tool.
Join Date: Jul 2002
Location: Confusio Helvetica
Posts: 311
Likes: 0
Received 0 Likes
on
0 Posts
Here's how I read it.
A. The winglets maximum demonstrated crosswind is weird, but the report doesn't seem to bother with it very much.
B. ATIS had reported winds 270 at 11. With clearance, winds were at 270 at 27. Interpolated estimates put the winds at about 32 kts on the first few kilometers of 34R, giving a 30kt crosswind.
C. The human factors states that the aircraft departed about "2600 feet from the approach end", on a 12,000-foot runway.
D. The five ATPs in the human factors group did two series of simulator tests: the first a series of takeoffs in increasing crosswinds (judging 35 kt to 40 gusting as "slightly difficult"), and the second apparently treating the simulated conditions, with:
The results:
D. The FDR shows that the aircraft passed 90 kts between 18:18:13 and 18:18:14. Around 18:18:14, the last significant pre-departure rudder input occurs; the pedals read about 1.5 degrees left. At the same time, the control wheel reads about 80 degrees right .
In short, the simulations suggest that in those conditions no rudder deflection for three seconds at 90 kts would be sufficient for a negative outcome, regardless of aileron deflection, tiller steering, or the decision to reject or continue. The FDR states that no commanded rudder deflection was sensed for at least three seconds before the overrun. The simulations where the point of departure matched the actual runway departure are those where no rudder was applied after 90 kts.
A. The winglets maximum demonstrated crosswind is weird, but the report doesn't seem to bother with it very much.
B. ATIS had reported winds 270 at 11. With clearance, winds were at 270 at 27. Interpolated estimates put the winds at about 32 kts on the first few kilometers of 34R, giving a 30kt crosswind.
C. The human factors states that the aircraft departed about "2600 feet from the approach end", on a 12,000-foot runway.
D. The five ATPs in the human factors group did two series of simulator tests: the first a series of takeoffs in increasing crosswinds (judging 35 kt to 40 gusting as "slightly difficult"), and the second apparently treating the simulated conditions, with:
1. Removed the feet from rudder pedals at 90 knots.
2. Removed the feet from rudder pedals at 90 knots (rudder trim adjusted to 1.5 degrees left).
3. Removed the feet from rudder pedals at 90 knots. Moved the control wheel full right at 100 knots.
4. Removed the feet from rudder pedals at 90 knots. Tried to maintain directional control using the tiller.
5. Removed the feet from rudder pedals at 90 knots. Waited 2 seconds (until a signal given by an observer with a stopwatch). Resumed rudder inputs and tried to complete the takeoff.
6. Removed the feet from rudder pedals at 90 knots. Waited 3 seconds (until a signal given by an observer with a stopwatch). Resumed rudder inputs and tried to complete the takeoff.
7. Removed the feet from rudder pedals at 90 knots. waited 2 seconds (until a signal given by an observer with a stopwatch). Resume rudder control inputs and tried to perform a rejected takeoff.
8. Began the takeoff roll normally. Removed the feet from rudder pedals at 90 knots. Waited 3 seconds (until a signal given by an observer with a stopwatch). Resumed rudder inputs and tried to perform a rejected takeoff.
2. Removed the feet from rudder pedals at 90 knots (rudder trim adjusted to 1.5 degrees left).
3. Removed the feet from rudder pedals at 90 knots. Moved the control wheel full right at 100 knots.
4. Removed the feet from rudder pedals at 90 knots. Tried to maintain directional control using the tiller.
5. Removed the feet from rudder pedals at 90 knots. Waited 2 seconds (until a signal given by an observer with a stopwatch). Resumed rudder inputs and tried to complete the takeoff.
6. Removed the feet from rudder pedals at 90 knots. Waited 3 seconds (until a signal given by an observer with a stopwatch). Resumed rudder inputs and tried to complete the takeoff.
7. Removed the feet from rudder pedals at 90 knots. waited 2 seconds (until a signal given by an observer with a stopwatch). Resume rudder control inputs and tried to perform a rejected takeoff.
8. Began the takeoff roll normally. Removed the feet from rudder pedals at 90 knots. Waited 3 seconds (until a signal given by an observer with a stopwatch). Resumed rudder inputs and tried to perform a rejected takeoff.
1. The airplane began turning left after the removal of right rudder orrection and it exited the left side of the runway about 5 seconds later. Tire crubbing vibrations and sounds began just before exiting the runway. The airplane left the runway at the 9,000-feet-remaining marker at an airspeed of about 120 knots.
2. The outcome was similar to scenario 1, except the tire scrubbing left turn was more rapid. The airplane left the runway a fraction of a second earlier, bout 300 feet before the 9,000-feet-remaining marker.
3. The outcome was the same as scenario 1.
4. The outcome was similar to scenario 1, except tire scrubbing vibrations and sounds were stronger and began soon after the application of tiller, the turn was less rapid, and the airplane exited the runway several hundred feet beyond the 9,000-feet-remaining marker.
5. All of the participants were able to salvage the takeoff. Three considered this scenario moderately difficult. Two considered it neither difficult nor easy.
6. Three participants were able to salvage the takeoff. Two were unable to rotate before the airplane departed the runway. Four described the scenario as very difficult. A fifth described it as slightly difficult.
7. All participants were able to reject the takeoff with skidding or “fishtailing” on the runway. Three described the scenario as slightly difficult. Two described it as moderately difficult.
8. Three participants were able to reject the takeoff with extreme skidding or “fishtailing.” Two were unable to keep the airplane from departing the runway. All participants described the scenario as very difficult.
2. The outcome was similar to scenario 1, except the tire scrubbing left turn was more rapid. The airplane left the runway a fraction of a second earlier, bout 300 feet before the 9,000-feet-remaining marker.
3. The outcome was the same as scenario 1.
4. The outcome was similar to scenario 1, except tire scrubbing vibrations and sounds were stronger and began soon after the application of tiller, the turn was less rapid, and the airplane exited the runway several hundred feet beyond the 9,000-feet-remaining marker.
5. All of the participants were able to salvage the takeoff. Three considered this scenario moderately difficult. Two considered it neither difficult nor easy.
6. Three participants were able to salvage the takeoff. Two were unable to rotate before the airplane departed the runway. Four described the scenario as very difficult. A fifth described it as slightly difficult.
7. All participants were able to reject the takeoff with skidding or “fishtailing” on the runway. Three described the scenario as slightly difficult. Two described it as moderately difficult.
8. Three participants were able to reject the takeoff with extreme skidding or “fishtailing.” Two were unable to keep the airplane from departing the runway. All participants described the scenario as very difficult.
In short, the simulations suggest that in those conditions no rudder deflection for three seconds at 90 kts would be sufficient for a negative outcome, regardless of aileron deflection, tiller steering, or the decision to reject or continue. The FDR states that no commanded rudder deflection was sensed for at least three seconds before the overrun. The simulations where the point of departure matched the actual runway departure are those where no rudder was applied after 90 kts.
Join Date: May 2008
Location: UK
Posts: 200
Likes: 0
Received 0 Likes
on
0 Posts
CaptainSandL:
Correct.
Whilst you enter test pilot territory the second cross-wind limits exceed that certified, it doesn't necessarily mean that the pilot couldn't handle, or the aircraft wasn't capable of flying in, those conditions.
VMCA dictates the minimum speed to maintain directional control. The rudder starts to become effective around 60 kts.
ECAM Actions.
Re the lower max demonstrated x-wind, I enquired to APB several years ago about the reduction when fitted with winglets and they replied that it was simply that there had not been any greater x-wind conditions present to demonstrate in during certification!
Whilst you enter test pilot territory the second cross-wind limits exceed that certified, it doesn't necessarily mean that the pilot couldn't handle, or the aircraft wasn't capable of flying in, those conditions.
VMCA dictates the minimum speed to maintain directional control. The rudder starts to become effective around 60 kts.
ECAM Actions.
Thanks ECAM,
I have not seen any Vmca data, but as the aircraft was on the ground perhaps Vmcg is more applicable.
Is there any change in Vmcg with a winglet equipped 735 from a non-winglet equiped 735?
I have not seen any Vmca data, but as the aircraft was on the ground perhaps Vmcg is more applicable.
Is there any change in Vmcg with a winglet equipped 735 from a non-winglet equiped 735?
Join Date: Jul 2006
Location: USA
Posts: 14
Likes: 0
Received 0 Likes
on
0 Posts
Bingo!
CaptainSandL, I think you've hit the nail on the head. The Captain on this CAL flight was a good friend of mine, and we've discussed this very thing. If you look at a short-body 737-500 with winglets, pay careful attention to how in a a direct crosswind the winglet actually partially obscures the vertical stabilizer, thereby reducing overall rudder effectiveness. Plus, if the crosswind limit was 33 knots, what would happen if there was a gust exceeding that limit?
X-wind Winglet-Rudder Interaction?
There is a possibility that as speed increases in a strong crosswind, disturbed airflow from the winglet would meet the fuselage centerline at a point that might impair rudder function.
Something to try out in a wind tunnel, especially as there may even be points where the disturbed air meeting the fin would decrease the need for rudder.
The FDR shows:
The second and third rudder pedal excursions are opposite the others, which may be due to correcting an overcorrection, a period of decreased crosswind component, or a winglet diminishing crosswind force on the fin.
Something to try out in a wind tunnel, especially as there may even be points where the disturbed air meeting the fin would decrease the need for rudder.
The FDR shows:
- 4 sec 0 increasing to ~2 deg Right rudder pedal
- 6 sec Left ~3 decreasing to 0
- 3 sec Left ~3 deg
- 2 sec 0
- 7 sec 0 increasing to Right ~10
- followed by more oscillations of mostly heavy positive angle.
The second and third rudder pedal excursions are opposite the others, which may be due to correcting an overcorrection, a period of decreased crosswind component, or a winglet diminishing crosswind force on the fin.
Join Date: Jul 2002
Location: Confusio Helvetica
Posts: 311
Likes: 0
Received 0 Likes
on
0 Posts
While I appreciate all this speculation concerning the rudder being blanked by mach tuck created by the winglets, I suggest those who want to pursue this should take a look at the FDR data -- the .csv files are available if you like.
Both captain (PF) and FO reported that things went bad after 90 kts. Captain asserts he applied full right rudder. FO asserts he felt the right rudder pedal being depressed.
FDR states: (for the record, maximum values recorded are pedal 10.95 degrees (right) and rudder 26.65 degrees (right), at 18:07.7/8, when the aircraft was at ~65 kts.
18:13.0: Ground speed 91 kts.
18:13.2: Heading 344.53, Rudder pedal 3.96 degrees (right), Rudder 10.4 degrees (right)
18:13.7: Rudder pedal -.28 degrees (left), Rudder 3.49 (right)
18:14.2: Heading 343.13, pedal 2.86 (right), rudder 9.01 (right)
18:14.7: rudder pedal .27 (right), rudder 4.12 (right)
18:15.2: Heading 338.55, pedal -.11 (left), rudder 3.43 (right)
18:15.7/8: pedal -.71 (left), rudder 2.16 (right)
18:16.2/3: Heading 332.93, pedal -.77 (left), rudder 1.02 (right)
18:16.7/8: pedal -.71 (left), rudder -3.3 (left)
18:17.2/3: Heading 331.52, pedal -1.1 (left), rudder -3.87 (left)
18:17.7/8: pedal -1.26 (left), rudder -4.31 (left)
18:18.2/3: Heading 333.63, pedal -1.26 (left), rudder -4.25 (left)
18:18.4: vertical acceleration 1.1145 Gs (aircraft left runway)
So in approximately two seconds (dependent on sampling method and rate), the aircraft changed ten degrees in direction. The deviation corresponds to a period where the rudder pedals were centered to slightly left, according to the FDR.
So please tell me how any effect generated by the winglets will either affect the rudder pedals or have a greater influence on the direction of the aircraft than the rudder itself.
Both captain (PF) and FO reported that things went bad after 90 kts. Captain asserts he applied full right rudder. FO asserts he felt the right rudder pedal being depressed.
FDR states: (for the record, maximum values recorded are pedal 10.95 degrees (right) and rudder 26.65 degrees (right), at 18:07.7/8, when the aircraft was at ~65 kts.
18:13.0: Ground speed 91 kts.
18:13.2: Heading 344.53, Rudder pedal 3.96 degrees (right), Rudder 10.4 degrees (right)
18:13.7: Rudder pedal -.28 degrees (left), Rudder 3.49 (right)
18:14.2: Heading 343.13, pedal 2.86 (right), rudder 9.01 (right)
18:14.7: rudder pedal .27 (right), rudder 4.12 (right)
18:15.2: Heading 338.55, pedal -.11 (left), rudder 3.43 (right)
18:15.7/8: pedal -.71 (left), rudder 2.16 (right)
18:16.2/3: Heading 332.93, pedal -.77 (left), rudder 1.02 (right)
18:16.7/8: pedal -.71 (left), rudder -3.3 (left)
18:17.2/3: Heading 331.52, pedal -1.1 (left), rudder -3.87 (left)
18:17.7/8: pedal -1.26 (left), rudder -4.31 (left)
18:18.2/3: Heading 333.63, pedal -1.26 (left), rudder -4.25 (left)
18:18.4: vertical acceleration 1.1145 Gs (aircraft left runway)
So in approximately two seconds (dependent on sampling method and rate), the aircraft changed ten degrees in direction. The deviation corresponds to a period where the rudder pedals were centered to slightly left, according to the FDR.
So please tell me how any effect generated by the winglets will either affect the rudder pedals or have a greater influence on the direction of the aircraft than the rudder itself.
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
Do we have any theories as to why we see left rudder when the a/c is apparently heading rapidly left towards the edge of the runway? I reckon my right foot would have been sticking out of the radome at that point.