Thrust reversers failed in deadly jetliner skid

[rsmith6621]

CHICAGO - The reverse thrusters that should have slowed a Southwest Airlines jetliner before it slid off a runway at Midway Airport and into the street didn’t immediately kick in when the pilots tried to deploy them, federal investigators said Saturday after interviewing the crew.

http://www.msnbc.msn.com/id/10388519/

I heard from a US Air mechanic today tha the aircraft landed at 160kts....normal landing speed is 125-35 depending on load.....if true it would be typical....British Air in SEA last year filed a complaint with the FAA against SWA for taxing often at 35kts or greater to and from the gates....they do this to achieve quick turnarounds....Oh British filed the complaint after their 777 was damaged by SWA when their Tail hit the 777 wingtip grounding the 777 for three days.

[The Velveteen Ocelot]

1. Interesting... But what would cause both reversers to fail?

A lot of people have observed that SWA tends to taxi pretty fast, but that doesn't mean they landed too fast. Their landing speed would be determined by weight, maybe a little faster if there were gusty crosswinds. They'd have been carrying extra fuel because of the weather at MDW, but they were coming from BWI, so they'd have burned off a couple of hours worth of fuel, and they were carrying only 98 passengers. That means they wouldn't have been especially heavy -- MDW is weight-restricted because of the short runways. Pilots who fly there regularly know they have to be careful, and SWA goes there all the time. I doubt those pilots, knowing there's no margin for error at MDW, would have flown an excessively fast approach, especially in a snowstorm.

[phaseolus]

2. hydraulic or electrical control problems?

I don't know anything about planes, but that would be my first guess why a thrust reverser wouldn't open... though you would think on something critical like a passenger plane there would be redundancy to prevent failures like this. (Are there?)

[The Velveteen Ocelot]

4. Just about every system on a large jet is required to have redundancies.

I'm familiar with McDonnell-Douglas aircraft, not Boeing, and I know for sure that on an MD-80 or a DC-9, for example, the reversers operate separately -- that is, if one fails, the other wouldn't. (Although on a slippery runway you wouldn't want just one reverser to operate.) Also, there are hydraulic accumulators to operate them in the event of a hydraulic system failure. I should think the 737 has something similar. The photo looked like it was an 800- series, which has more electronic systems. Guess we'll just have to wait for the NTSB to finish its investigation.

[SIU_Blue]

5. FYI, Aircraft was a -700 series.

N471WN, a -7H4

:D

[theHandpuppet]

9. From what I've read this was a new plane, too

Purchased new by Southwest only 18 months ago.

[SIU_Blue]

3. I don't pretend to be an expert on the 737...

I'm only a private pilot. I wonder if the linkage between the thrust reversers in the NGs and the cockpit levers are seperate or linked. I assume that for redundancy purposes they are separate, but it seems to me that the only way both reversers could fail is A. they are linked, or B. pilot error. I guess we'll have to wait for the NTSB report.

[awoke_in_2003]

6. I work on a 737-800

flight simulator, and yes, the thrust reversers are separate. The only way to fail both at the same time is by having a total hydraulics system failure (not easy to achieve, either). From my years of dealing with pilots, I would lean on pilot error in almost all accident cases.

[SIU_Blue]

7. "I would lean on pilot error in almost all accident cases."

So would statistics. However, you have to remember, that pilot error is a very broad category. Any failure of any system onboard an aircraft immediately becomes "Pilot error" if the appropriate corrective action is not taken. Even if there is a mechanical failure, it can be construed as pilot error. At this point, I'd say it's best not to speculate.

[Jakey]

11. Air/Ground sensing...

of some sort is used to determine when thrust reverser deployment is available. Weight on both main gear and radio altimeter readouts ring a bell but I've been away from it for too long to recall Boeing's favored parameters. Anyone else?

[Tesha]

35. Pretty sure throttle position factors in as well. (NT)

[snooper2]

47. Reverse Thrusters engage based on Wheel Spin...

When the main landing gear hit the runway and begin to spin, that is the activator for the reverse thrusters. With a slick runway, the wheels appear to have "slid" instead of start spinning immediately, thus, crash...

[wakeme2008]

8. Extra speed would be normal..

If the pilot thought that he would have to do a go-around he would come in a little fast. And while I do not know 737, I would have to believe they have a safe guard that would keep reverse thrusters from deploring if one side fails...

[Jakey]

10. Sorry but...

Extra speed would be normal...If the pilot thought that he would have to do a go-around he would come in a little fast.

this "technique" is never even considered in a professionally run cockpit. You must ALWAYS anticipate a requirement to execute a missed-approach or "go around". Computed approach speeds for aircraft weight, configuration and meteorological conditions (wind) are more than adequate to safely conduct such a procedure and additional speed will be recognized by most aviators as a BAD technique and BAD approach management.

[natrat]

12. extra speed and snow? nt

[The Velveteen Ocelot]

14. You wouldn't fly a faster approach. You always plan for a go-around

but fly the normal calculated approach speed (ref+5, probably). Assuming they were following their normal procedures, they shouldn't have been flying faster than that. I can't think of many things more dangerous than flying an unstabilized approach into MDW in a snowstorm -- I doubt they would have done that.

[sendero]

13. Wouldn't the deployment of a single..

... thrust reverser make the plane turn uncontrollably?

[Jakey]

15. Only if...

improperly co-ordinated with other means of directional control, to include aerodynamic directional control (rudder), differential braking and nose wheel steering (when available). It is, however, more problematical with the presence of runway contamination (ice,snow,water) and would need to be applied much more judiciously under those conditions.

[DemoTex]

17. Single thrust-reverser operation is allowed (one T/R on MEL, engine-out).

If a thrust reverser(s) (T/R) is/are inoperative, maintenance has the option of "pinning" one or both T/Rs (in the retracted position) and operating the aircraft according to any restrictions or limitations required by the MEL (minimum equipment list). The aircraft can be dispatched under the MEL with one or both T/Rs pinned closed.

However, it is up to the captain as to whether the flight goes. Personally, would never put my signature on a dispatch release with one or both T/Rs on MEL to a destination with relatively short, contaminated runways. Although T/Rs cannot be a part of the calculation of landing distance, it would not be the safest operation to land on an icy runway without all the bells and whistles to get that 150,000 pound sled stopped.

Jakey is correct too. If one T/R is inoperative (MEL, engine failure or shut down), it is OK to use a single T/R for deceleration. However, that single T/R use must be appropriate to runway conditions. Long, slippery runways .. keep it stowed. Short slippery runways .. deploy the single T/R very gingerly.

Single T/R operation is much easier in an aircraft with tail-mounted engines (DC-9, MD-80, Learjet) because the center of thrust is nearer the centerline of the aircraft. Single T/R operation with wing-mounted engines (B-737, Airbus A-320) results in more yaw.

[DemoTex]

16. Performance data for landing distance CANNOT include thrust-reversers.

The deceleration benefits of thrust reversers (T/Rs) are merely gravy when it comes to calculating landing distance or accelerate-stop distance (rejected takeoff). The two key elements for valid landing distances are 1. maximum braking immediately after touchdown, and; 2. full spoiler extension immediately after touchdown. Again, T/Rs are gravy.

However, if the SWA B-737 bounced (as was reported Friday) and the crew did not realize that the aircraft was airborne again, any effort to deploy the T/Rs would have been futile. A "squat" switch prevents the T/Rs from being deployed unless the main landing gear is on the ground with the strut compressed. That I remember from my 8000+ hours flying Boeing 737-200/300/400s.

[The Velveteen Ocelot]

18. OK, now it's making sense.

I'm familiar with the reversers on the DC-9/MD-80s, which can be deployed in flight, not that you ever would do that; they are not dependent on a ground shift mechanism like the 737 apparently is. So if that device failed, either on account of a bounce or for some other reason, there goes your reverse thrust capability. And you are definitely correct about landing distance -- reverse thrust is never considered; only weight, spoilers and anti-skid braking. So theoretically they should have been able to land the airplane safely even without the TRs, but if the last third of the runway had poor braking action, or if they were landing a little fast, now there's trouble. At MDW in a snowstorm, you can't afford to have anything go wrong.

Is that an MD-80 in your sig pic?

[DemoTex]

21. Yep ..

I was a MD-80 captain.

[Buns_of_Fire]

23. Slightly off-topic, but I just gotta know...

You mentioned that the reversers on the DC-9/MD-80 can be deployed in flight (and that you obviously wouldn't do that, which, even as a non-pilot, I totally agree!).

So, if I understand this correctly, does this mean that they could actually be deployed at, say, 30,000 feet at cruise speed? No lockouts, no fail-safes, no "squat switches," no nothin'?

I felt a cold shiver run up my spine, reading that...

[DemoTex]

24. DC9/MD-80 T/Rs cannot normally be deployed in flight.

Again, the air-ground sensor switch on the main landing gear. However, that switch could fail. "Popping" the T/R levers in flight, with a failed "squat" switch, could be much more hazardous to one's health than putting the landing gear lever to the "up" position on the ground.

Having said that, it is my understanding from my old Braniff buds (who flew DC-8s) that the T/Rs on the DC-8 are designed for deployment in-flight for descent augmentation. I have never flown the four-motored DC-8, so I don't know that for a fact.

BTW: I have a DC-9 type-rating, although I never flew the straight 9s. I only flew the MD-80/82, which is a streched-out and tricked-up DC-9. All MD-80 drivers hold DC-9 type-ratings. There is no MD-80 type-rating.

[Buns_of_Fire]

25. Thank you. I feel (sort-of) better now.

T/Rs activated in flight for "descent augmentation" -- yes, I can imagine that would markedly speed up the descent process. Talk about hitting the emergency brakes... Thanks again!

[DemoTex]

26. I'll check on that DC-8 info tomorrow with one of my Braniff buds.

I do remember what he said about the DC-8 with two (of four .. I think that was the limitation, and those two had to be inboard engines) reverse buckets out in-flight. "That old 8, with two buckets out, shook like a dog shittin' peach pits."

[The Velveteen Ocelot]

27. The DC-9 TRs don't have a lockout mechanism.

They are not connected to ground shift, which is on the nose gear (though the ground spoilers are), and mechanically can be (but were not intended to be and are prohibited by limitation) deployed in flight, but only with the throttles at idle. The MD-80s could be different; I'm not as familiar with them (I'm a DC-9 instructor).

[The Velveteen Ocelot]

28. On the DC-9 it's physically possible with the throttles at idle, but

it is prohibited by limitation. It would never be done deliberately, and really can't be done accidentally -- you have to try. The TRs are held closed in flight by accumulator pressure to the stowed side of the actuators, by safety latches and by an overcenter linkage. I have heard of only one instance where a DC-9 TR accidentally deployed in flight due to a failure of a latch (it was Eastern Airlines, I think, many years ago), and the pilot quickly shut down the engine and returned to the airport. It just doesn't happen. No need for cold shivers.

[DemoTex]

29. Crazy design, that!

I'll have to get back into the bowels of the MD-80 book. I am almost positive in my previous statements, because I remember a bounce (no-T/R) and squat-switch failure scenarios in my MD-80 dungeon .. er, I mean simulator .. experience.

[Jakey]

19. Looks like...

some question is developing as to "maximum braking immediately after touchdown", given the following from the same story...

He (Robert Benzon, National Transportation Safety Board investigator in charge) said the pilots told investigators they began applying the brakes manually as soon as they noticed that the plane wasn’t slowing properly.

...and it is yet to be divulged what degree of auto-braking, if any, (which I believe this aircraft is equipped with) had been selected. Perhaps a setting of less than maximum (I believe there are 3 levels) was selected (or, doubtfully, none at all?) and the application of manual maximum braking was just too late to prevent the overrun.

On edit: looks like 4 levels of auto-brake (1,2,3 & MAX) See nice illustration at http://www.surclaro.com/pfaa (737 main panel)

[maseman]

20. They also had a 13 knot tail wind

Not only did they have the poor visibility and short runways, compound that with a 13 knot tailwind and it is just another variable to make matters worse. I have about 1,000+ flight hours (none in large jets like the 737 or DC-9, but in twin props) and all pilots know the more negative variables working against you the less chance you have for perfection.

These guys were landing at night at a short runway airfield with terrible visibility and a tailwind. Plus the runways had some amount of substance on them regardless of the reports. One report said the first two-thirds were fine but the last third was pretty slippery. With a 6,500' runway you need all the thirds to cooperate.

[Macman44]

22. I did some research

and the runways MDW was favoring had at least a gusting 13kt tailwind. Plus the instrument approach has a fairly steep glideslope. So coupled with a tailwind and carrying a little extra speed for the missed approach, he came sizzling in. The ceilings were reported to be at minimums. Not to mention that the runways have a displaced threshold. So its possible he ate up a good portion of the runway before the pilot "planted" the plane on the runway and tried to deploy the T/R. The slats, I believe, are coupled with the T/R and are raised when the reversers are deployed. Braking action has yet to be reported but it could have been poor. We could "hanger fly" this accident all day and I will withold my final opinion until the NTSB report comes out.

[DemoTex]

30. L/E slats and Kruger flaps go down with the trailing-edge flaps on B-737.

Leading edge slats/Kruger-flaps/and trailing-edge-flaps have absolutely nothing to do with T/Rs, brakes, or shit-house-mice. L/E devices don't retract until the flap handle is moved to the "up" position (B-737 and MD-80).

I am type-rated in both the Boeing 737 series and the MD-80 (DC-9) aircraft. You really need to do some more research .. please!

[DemoTex]

41. If they knowingly landed with a 13 knot tailwind, they have big problems.

The maximum tailwind component for a Boeing 737 (and MD-80) is 10 knots. If the crew knew that they had more than 10 knots tailwind component, they were in violation of the aircraft limitations and risk charges of dangerous and reckless operation, among others. Not good.

[Jakey]

42. Sequence reports?

I'd be interested in viewing the source of this allegation. Anyone have a link to credible weather data for MDW at the time of the crash?

[DemoTex]

43. The tower gives the current wind with the landing clearance.

For example, "Delta 123 cleared to land runway 31, wind 270 at one-two, gusting two-two." The last wind-speed information conveyed by the tower to the SWA crew will be critical in the investigation. Also, the tower is required to make a special weather observation (to include wind direction and speed) immediately after an accident.

Hourly reports, ATIS reports, etc., are all interesting, but it is that last real-time conveyance from the tower to the aircraft that is germane. That will be a matter of record on the ATC tapes. If there was cockpit discussion of the wind conditions, it will be on the cockpit voice recorder. Data harvested from the flight management computer system will probably give investigators the winds aloft for any point on the final approach. However, that last conveyance of winds from the tower to the aircraft is legally controlling with respect to crosswind or tailwind limitations.

[Jakey]

44. Excellent points DT...

and perhaps the NTSB will issue some preliminary data in that regard relatively soon. However, until that time, I'd be curious to see where this tailwind component info is coming from.

[Jakey]

46. Braking Action reported "fair to poor"

If the following report is accurate, this accident is beginning to look like the "perfect storm" for a runway overrun...

The plane needed to land within about the first 2,170 feet of the runway to stop on the snow-slicked surface, which was rated "fair" to "poor" for aircraft-braking ability by the pilot who landed several minutes in front of Flight 1248, transportation officials said.

http://www.chicagotribune.com/news/local/chi-0512110438dec11,1,5398944.story?coll=chi-news-hed

Whether the crew was informed of this report or had inquired about braking action reports will be a subject of high interest to accident investigators.

[ronstearneylaw]

31. Any SWA pilots?

I represent the family of the boy that was killed and am also a pilot. If anyone knows someone who is a pilot for SWA or has acces to their Pilots Operating Handbook, I would sure like to look at it. Now on to the topic at hand, I do not believe there was a failure, here's why. If you listen carefully to what the NTSB said (and remember they are only repeating what the crew told them), the captain said he couldnt deploy the reversers and that the copilot had to do it. The reversers only reached level 3 at the time the airplane struck the fence. There are three possibilities: the reversers were spooling down, they were spooling up, they were maxed out. The copilot said that he had to deploy the revesers. Well, the reversers are on the throttles and both pilots use the same controls. What was the problem?

Anyone who knows the speed that the airplane landed, or any info, please let me know. Thanks.

[DemoTex]

32. You are way out of line, friend.

Take you ambulance-chasing to another site. Please. I know more about the Boeing 737, the MD-80, and accident investigation (I am an ALPA-trained accident investigator) than almost anyone else on this board. Get it? Adios, loser.

Captain Mac

[PeterPuck]

38. check this out...

looks like the poster is the lawyer in the article... what a sleaze bag

http://www.cnn.com/2005/US/12/12/midway.accident.ap/index.html

[MaineDem]

39. Thank you

What a sleezy thing to do. Your response is perfect.

[The Velveteen Ocelot]

33. A new DUer, I see. This isn't the place to be trolling for information.

Do your own investigation, and don't assume anything. And don't expect other pilots to help you hang this crew. (And no, I don't work for SWA.)

[Jakey]

34. I'd get right on it but...

I'm a bit tied up right now composing my keynote address to the Congress of Neurological Surgeons.

Oh, and have your people contact my people inre that pesky expert witness retainer? They're in DKOS from 10-11 CST and DU from 12-4 CST. While I agree that 30% of gross might sound high, my Googling billables don't come cheap, and, after all, you DO want the best for your client.

Looking forward to a profitable alliance, I remain

Litigiously yours,

Jakey
CEO
Dewey, Cheatem and Howe

[jackstraw45]

37. I'm Abraham Lincoln....

Anyone know anything about fighting Civil Wars?


:eyes:

[tedzbear]

36. So far it sounds like SWA is going to be held responsible...

...either because of mechanical malfunction or pilot error.

BTW, don't 737-700s use "fly by wire" instead of old fashioned hydraulics???

[Jakey]

40. You misconstrue the concept...

BTW, don't 737-700s use "fly by wire" instead of old fashioned hydraulics???

"Fly by wire" implies electronic manipulation and/or management of aircraft systems, to include those dependent on hydraulics for operation.

[tedzbear]

45. Thank you.

I was wondering how electrical wires could deliver enough power to move the flaps up and down against 500 mph wind.