Announcement

Collapse
No announcement yet.

Gulfstream IV jet ran off a runway while taking off

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #46
    Originally posted by Evan View Post
    That's a bit overly dogmatic isn't it? V1 is the critical engine failure decision speed. A control surface response issue does not apply.
    V1 is not "just" the critical engine failure.
    V1 gives you enough room to complete a take-off with one engine failed.
    But it also gives you enough (many times barely enough) room to STOP in the remaining runway, and here it doesn't matter why you stop. Try to stop after V1 and chances are that you'll overrun.

    That's why the mindset is trained to be "GO" after V1. There only 2 or 3 reasons to abort after V1. Two of them are "unsafe" and "unable to fly". But that'n quite fuzzy, not like there is an "abort" horn, so it might take a bunch of seconds to judge if the plane is unsafe or unable to fly.

    Please tell me you don't go flying with a control surface issue.
    I try not to. That's why I included this in my previous post:
    In any event, I'd ask what happened with the "locks ==> off, controls ==> free and correct, stall barrier ==> test".

    --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
    --- Defend what you say with arguments, not by imposing your credentials ---

    Comment


    • #47
      Originally posted by Gabriel View Post
      That's why the mindset is trained to be "GO" after V1. There only 2 or 3 reasons to abort after V1. Two of them are "unsafe" and "unable to fly". But that'n quite fuzzy, not like there is an "abort" horn, so it might take a bunch of seconds to judge if the plane is unsafe or unable to fly.
      There are an assortment of scenarios that would cause an aircraft to fail on rotation. Loading/COG issues, bad v-speed calculations, jammed control issues, pilot's lunchbox jammed under the control column (or was that an Airbus sidestick issue?) etc. etc.

      What kills me is the 168kts.

      I'm hoping one of our resident professionals might be lurking here and would comment on what a pilot might be expected to do after Vr when the thing refuses to lift off and the opposite threshold is not getting any further away...

      Nothing against you Gabriel. Does V1 even apply to a single engine piston aircraft on a 5000'+ runway? Certainly engine failure is a no-go decision.

      Comment


      • #48
        Evan, your black and white thinking continues to amaze me.

        You are unable to put yourself in the seat of a G4 on a dark night where every other takeoff in your entire life offered elevator control and that it's not that strange that the pilots might burn a a few valuable seconds of really good acceleration after V1 knowing SOMETHING- but not sure WHAT is wrong and that the training that has always saved them in the simulator is to continue...wait no, I guess we better stop...

        And V1 is the engine failure speed and does not apply to other problems...where your mind is supposed to instantly know to stop- even if it's after V1 and you are likely to run off the end where all planes have a bad tendency to crash and burn....
        Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

        Comment


        • #49
          Originally posted by 3WE View Post
          Evan, your black and white thinking continues to amaze me.

          You are unable to put yourself in the seat of a G4 on a dark night where every other takeoff in your entire life offered elevator control and that it's not that strange that the pilots might burn a a few valuable seconds of really good acceleration after V1 knowing SOMETHING- but not sure WHAT is wrong and that the training that has always saved them in the simulator is to continue...wait no, I guess we better stop...

          And V1 is the engine failure speed and does not apply to other problems...where your mind is supposed to instantly know to stop- even if it's after V1 and you are likely to run off the end where all planes have a bad tendency to crash and burn....
          Like the Concorde. Oh no, wait, they continued didn't they...

          Comment


          • #50
            Originally posted by Evan View Post
            ...what a pilot might be expected to do after Vr when the thing refuses to lift off and the opposite threshold is not getting any further away...
            Repeating, Vr is after V1

            If the plane refuses to lift off, I guess the pilots course of action would be to apply brakes and run off the end.

            Which is pretty much what they did.

            Now, how fast can you diagnose "unflyable"?...what if they pulled back on the control wheel and it came back like it always did?- but the nose didn't lift off? Remember-at that point they ARE GOING TO CRASH because they are already past V1 and this runway was not a long one.

            I think I'm going to try to fly...mess with the trim...pump the wheel in and out...maybe put out more flaps...

            The more I think about this- those guys only chance was to try and fly...the option to stop passed before they tried to rotate.
            Les règles de l'aviation de base découragent de longues périodes de dur tirer vers le haut.

            Comment


            • #51
              Originally posted by 3WE View Post
              If the plane refuses to lift off, I guess the pilots course of action would be to apply brakes and run off the end.
              For example: http://aviation-safety.net/database/...?id=19620603-0
              "I know that at times I can be a little over the top." -ITS

              Comment


              • #52
                There are an assortment of scenarios that would cause an aircraft to fail on rotation. Loading/COG issues, bad v-speed calculations, jammed control issues, pilot's lunchbox jammed under the control column (or was that an Airbus sidestick issue?) etc. etc.
                Exactly, and most of the times the plane will still fly and that results the safest course of action.

                Originally posted by 3WE View Post
                I think I'm going to try to fly...mess with the trim...pump the wheel in and out...maybe put out more flaps...
                Exactly what I was thinking word by word.
                Pull up. Nose doens't go up? Firewall the throttles. Pull harder, ask for help to the MP, add nose-up trim, add more flaps. Still nothing? Abort.

                Most of the times that a plane fails to lift-off it's because a load / GG / low calculated VR / mistrim / misconfig, and most of these can be solved with more speed and/or unusual control inputs.

                As 3we said, at what point you judge "usafe/unable to fly"? It's not a balck and wite decision like "UAS ==> 5° and CLB".

                --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
                --- Defend what you say with arguments, not by imposing your credentials ---

                Comment


                • #53
                  Originally posted by Not_Karl View Post
                  Good example. This plane and 130 lives could maybe have been saved if they had continued the take-off while troubleshooting the problem. Pull up at 4 hands while the flight engineer manually turns the trim wheel.

                  --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
                  --- Defend what you say with arguments, not by imposing your credentials ---

                  Comment


                  • #54
                    Originally posted by Evan View Post
                    But I see I spoke too soon about a nice long runway. I checked the link Leftseat posted and RWY 29 is only 7011'. That's pretty tight for a G-IV considering it requires 9,800' at MTOW (according to the Gulfstream flight planning manual). Even at 11,000lbs lighter it required 6,900'. Add a rolling, derated takeoff to that (one report is saying flaps were at 10) and I guess there really is no time to dilly dally after a failed rotation attempt.
                    Well, they were only going to Atlantic City which is about an hour away, and factor in that they had a 1,000ft stopway + it looks like another 1,000ft of grass after that before falling into that ravine. Maybe tankering was involved, but for a short hop like that in a G-IV, 7,000ft + 1,000ft stopway should be plenty, G-IVs routinely operate into much shorter runways.

                    Still, how do you get to 168kts in this scenario unless you COMMIT to flying despite the fact that it won't fly and only pull the throttles going into the overrun?
                    That's the million dollar question here, 168 kts seems like well past V1/Vr, but maybe Gabriel's scenario is what got them there. They committed to the takeoff but then realized they literally could not, and by the time they decided to call it off it was far too late.

                    Comment


                    • #55
                      Originally posted by ErwinS View Post
                      Strange. On the pictures of the wreckage the reversers are not deployed.
                      I'm not certain here, but isn't it very possible that they were deployed, but after the crash & fire they dropped back into a stowed position?

                      Comment


                      • #56
                        Originally posted by Leftseat86 View Post
                        I'm not certain here, but isn't it very possible that they were deployed, but after the crash & fire they dropped back into a stowed position?
                        Could be but normally the are locked out when deployed.
                        “The only time you have too much fuel is when you’re on fire.”

                        Erwin

                        Comment


                        • #57
                          Originally posted by Gabriel View Post
                          Exactly, and most of the times the plane will still fly and that results the safest course of action.
                          Well let me just present an opposing point of view: Nothing black and white here... that's just my point. After V1 is NOT always go. On takeoff pilots need to be extremely vigilant and prepared to abort if anything indicates an issue that could make the aircraft unsafe to fly. Before V1, not a hard decision, you reject. After V1, you have to think very fast, I understand that. Reject immediately after Vr and you will almost certainly overrun and destroy the aircraft but perhaps everyone will walk away. It is a fairly controlled crash unless there is a structure or a terrain hazard there (which shouldn't be there if at all possible). But continue and you are playing all or nothing: you might end up in the air and land safely, you might lift off, roll over and kill everyone or you might have to reject far later at a much higher speed with far less of a chance of survival.

                          Now in this scenario, assuming they rotated and got little or no response, there are two assumptions to make: 1) the v-speed calculation is off and liftoff will occur at a bit higher speed or 2) there is something wrong with weight/balance or flight control. Making assumption 1 you continue knowing if you are wrong you will probably kill everyone on board and possibly people on the groiund as well. Making assumption 2 you reject knowing that the plane is probably lost but the crash will be controlled and survivable. Rejecting at 130-140kts earlier on the runway might have left then in the grass short of the ditch, sans landing gear and LOC antenna. It seems to me that assumption 2 is, overall, the safer assumption to make.

                          The reports are saying that, after Vr, the crew discussed a flight control issue. That leans towards scenario 2. Yet they obviously continued.

                          I think, with VIP pax bound for Atlantic CIty and a very expensive private jet there may be a greater resistance towards the second option and a tendency to put all your chips on lady luck. The caveat here is that these pax were apparently in the gambling mood to begin with.

                          On my next A320 flight however, I'd rather crawl out of the wreckage in the airfield than end up in a smoking hole.

                          Comment


                          • #58
                            One problem IMHO is that scenario #2 isn't nearly as clear as you make it out to be.

                            If CG location is a little too far forward or the plane is slightly overloaded, the plane will still fly, it'll just take a little longer to get airborne and not climb as fast. If it's very hot out or the field is at high altitude that in itself could result in a crash, but neither was the case with this accident.

                            In case of a control problem, it's much clearer that rejecting is the correct choice. But put yourself in the pilot's seat: you just completed a checklist that included a flight control check, and presumably you convinced yourself they're working right otherwise you never would have attempted the takeoff. With no hard evidence, it's going to be awfully tough to "undo" that convincing in just a second or two.

                            On a related note: I'd love to hear from some of the pro pilots here... how often in simulator training do you practice rejecting a takeoff after Vr? I bet the answer is either "never" or very close to never.
                            Be alert! America needs more lerts.

                            Eric Law

                            Comment


                            • #59
                              Originally posted by Evan View Post
                              Well let me just present an opposing point of view: Nothing black and white here... that's just my point. After V1 is NOT always go. On takeoff pilots need to be extremely vigilant and prepared to abort if anything indicates an issue that could make the aircraft unsafe to fly. Before V1, not a hard decision, you reject.
                              Not evn that. The "official" procedure is:
                              Below 80/100kts (depending on the operator) the decision is not hard: You reject for any caution or warning alarm, strange noise or vibration, open door or window, deflated trire, whatever...
                              Between 80/100 and V1 you ONLY abort for the same reasons that you would above V1 + engine failure. Nothing more. The decision is easy if there is an easily identifiable engine faiulre. A "unsafe/unable to fly" condition can be hard to detect before you actually attempt to fly, what happnes after VR.

                              The thing is that history show that there were A LOT of BAD ACCIDENTS AND LOSS OF LIFE due to high speed aborts, many times even before V1. So the airplane industry (the manufactureres together with the FAA and other agencies) gathered together to study the situation and prepare some gidelines, very much like they did with the stall 20 years later.
                              The resutl was a proposed culture that, in most cases pasr 80/100kts, a take off is safer than an abort, and that in case of doubt, your best bet is to go. For example, you had an engine failure CLOSE to V1, what shall you do?
                              The more practical results were the procedure described above, where most of the causes were taken out of an abort past 80/100, the change in the definition of V1 that changed from "decision speed" to "decision TAKEN speed" where TAKEN means that the pilot started the first action to effectivelly abort the take-off, and that, in case of doubt, go. ANd this saved a lot of lifes. Probably it costed some lives that would have been saved in the "old way" (where the culture was "better to stay on the ground thatn fly if there are dome doubts), but the ballance is highly possitive.

                              After V1, you have to think very fast, I understand that. Reject immediately after Vr and you will almost certainly overrun and destroy the aircraft but perhaps everyone will walk away. It is a fairly controlled crash unless there is a structure or a terrain hazard there (which shouldn't be there if at all possible). But continue and you are playing all or nothing: you might end up in the air and land safely, you might lift off, roll over and kill everyone or you might have to reject far later at a much higher speed with far less of a chance of survival.
                              And yet, the study that I mentioned above shows otherwise.
                              Sure, once you are in the air is a full monty bet, but one that you win 99% of the times (99% is an invented figure, just to show the concept). A high speed abort is a more fuzzy bet, with more things between "let's go bak to the ramp" and "all died", however the dead toll showed to be higher on average.

                              And the problem is not so much to think fast, but that the information that the pilot has can very likely be insufficient to make the right decision.

                              Take for example the MD-80 that took off without flaps in Detroit and Madrid. Detroit continued the take-off and Madrid aborted (after loss of control and uncommanded re-land). In both cases almost everybody died, but in both cases everybody would have survived and the plane would have resulted undamaged (or perhaps with a scratch under the tail) if the pilots would have followed today's stall/escape procedure. Madrid could also have safely aborted if the pilot had not kept pulling up in the middle of the stall, but they had a very unusual 11000 ft runway.

                              Now in this scenario, assuming they rotated and got little or no response, there are two assumptions to make: 1) the v-speed calculation is off and liftoff will occur at a bit higher speed or 2) there is something wrong with weight/balance or flight control. Making assumption 1 you continue knowing if you are wrong you will probably kill everyone on board and possibly people on the groiund as well. Making assumption 2 you reject knowing that the plane is probably lost but the crash will be controlled and survivable. Rejecting at 130-140kts earlier on the runway might have left then in the grass short of the ditch, sans landing gear and LOC antenna. It seems to me that assumption 2 is, overall, the safer assumption to make.
                              I don't agree. It's easy now in hindsight, but story and statistics tell otherwise. I, for one, if suspected a CG issue, would not abort past Vr in a tight runway. If the plane rotates slow or don't rotate at all, that's a nose-heavy situation. Stability is not compromised and colntrollability can be fixed with more speed, more force and trim (what would have saved the lives in that 707).

                              Remember that the stopway is part of the take-off calculatiojn, so if V1 and the weight was limited by accelerate-stop distance, they would have ended at the edge of the grass aborting at V1. Past that, the situation worsens VERY VERY quickly, You have less and less runway to stop from a faster and faster speed. You can do an Excel simulation.

                              The reports are saying that, after Vr, the crew discussed a flight control issue. That leans towards scenario 2. Yet they obviously continued.
                              Just inventing but:
                              - Pilot tries to rotate, nothing happens.
                              - Pilot tries harder and says "something odd with the elevator". Nothing happens.
                              - Pilot keeps trying, adds trim, firewalls the throttles and says "Hey, help me pull up, it's an bit nose heavey", the nose roses a bit.
                              - The nose doesn't raise enough to lift off. Pilots starts to abort saying "The elevator is stuck".
                              There you have your "control problems discussion".

                              Here you can have a good 4/5 seconds of acceleration at up to 10kts per seconds past Vr (not V1). So they could be doing something like Vr+30/40 / V1+40/50 and 350/400m past the V1 point on the runway. So IF they were at V1, they would stop 350/400m past the paved surface what would pot you already in the ditch or its edge (and that assuming "paved braking action" on the grass, wich is obviusly ridiculous), add 40kts more of speed and the very small braking action on the grass and you are hitting a ditch at 80kts.

                              --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
                              --- Defend what you say with arguments, not by imposing your credentials ---

                              Comment


                              • #60
                                Originally posted by elaw View Post
                                If CG location is a little too far forward or the plane is slightly overloaded, the plane will still fly, it'll just take a little longer to get airborne and not climb as fast.
                                So we can rule that out then?

                                I'm thinking of Teterboro. I can't recall the details but it involved tankering which caused (or added to) a forward center of gravity that inhibited or prevented rotation. I don't recall if they rejected at that point or continued to attempt to get it off the ground before rejecting.

                                Comment

                                Working...
                                X