Gulfstream IV jet ran off a runway while taking off

[Evan]

Yet, I don't think there's many procedures anywhere where pilots calculate and brief on a past-V1-but-still-safe-to-abort speed....

While anything over 80-100kts is probably unsafe-to-abort I want to reaffirm my advocacy for a third V-speed above V1 (whenever the conditions exist) that is 'still-possible-to-unsafely abort speed', beyond which the crew will know that an abort probably results in an overrun.

I know, I know, conservative pilots will flame me for adding complexity to a system that ain't broken (with the rare exception) but since V-speeds are commonly calculated by the avionics anyway, it would not add workload and would also never be encountered by pilots (with the rare exception).

[3WE]

...I know, I know, conservative pilots will flame me for adding complexity to a system that ain't broken (with the rare exception) but since V-speeds are commonly calculated by the avionics anyway, it would not add workload and would also never be encountered by pilots (with the rare exception).

It would appear discussion worthy for this crash given the comment that they could have stopped on-pavement after they apparently realized the locks were in place...

And, as I stated above, there's at quite a few takeoffs being done on crazy-long runways where you usually CAN stop anytime before V2...so do we really calculate and brief that you can still stop safely after V1 on the really long ones?

...it's just the fundamentals behind the procedures that it's generally (though not always) better to continue.

The Les Abend article in last month's Flying magazine discussed the Concorde crash, and restated that an abort probably would probably have resulted in a catastrophic over-run.

[3WE]

While anything over 80-100kts is probably unsafe-to-abort.

Ummm, what happened to your propensity to recite actual checklists from specific aircraft FCOMs and QRHs?

In broad, fundamental terms, I think those numbers are off a bit.

If you are right, then there really is no need for V3P.

[Evan]

Ummm, what happened to your propensity to recite actual checklists from specific aircraft FCOMs and QRHs?

...meaning I don't think you can really call a high-energy reject entirely "safe". I tend to think this is the reason for the 80kt callout (often 100kt these days) and the general rule to continue after 80kts unless engine failure or unsafe to fly.

[3WE]

...the reason for the 80kt callout (often 100kt these days)...the general rule to continue after 80kts unless engine failure or unsafe to fly...

Interesting...

You have brought up an existing, "added-complication" to "the critical, high-work-load, takeoff procedure and decision tree."

Below 80(100) knots, abort for little stuff.

Above 80(100 knots) only abort for big stuff.

After V1, don't abort even if you lose an engine. (assuming it's a routine failure where the engine 'simply' stops producing power and not_a situation where it's slinging blades, destroying all your hydraulics and initiating an un-stoppable catastrophic fire).

Is it worth it to also know the true, "you-ain't-gonna-be-able-to-stop speed" for use on LONG-ASS runways?

(or a variation, of a simple Yes/No- that on THIS long runway, you can stop anytime up to V2 (once you take flight, you probably do need to ride things out (with some prayer)).

Is it that much more workload?

A control problem is a fairly bad thing to have, so maybe aborting after V-1 for serous control problems is a valid consideration.

(Of course a "Flight controls free and correct check" is another additional procedure that should be considered for this particular case)

/ass-hat, pontificating, parlour-talk from a non ATP dude flying a computer keyboard.

[Evan]

(Of course a "Flight controls free and correct check" is another additional procedure that should be considered for this particular case)

Yes, but we are talking about something a procedurally perfect airplane would do, not a procedural lazy prone-to-human-error pilot...

After the a/c exceeds Vr+10, in ground mode, if there is significant runway left a new bug pops up on the PFD (let's call it Vrwy) and an amber line appears between Vr and Vrwy. That's your last-ditch margin to get airborne. If you can't, you reject at or before Vrwy and still stop on the runway surface. If you exceed Vrwy on the ground, you're going offroading.

[3WE]

Yes, but we are talking about something a procedurally perfect airplane would do, not a procedural lazy prone-to-human-error pilot...

After the a/c exceeds Vr+10, in ground mode, if there is significant runway left a new bug pops up on the PFD (let's call it Vrwy) and an amber line appears between Vr and Vrwy. That's your last-ditch margin to get airborne. If you can't, you reject at or before Vrwy and still stop on the runway surface. If you exceed Vrwy on the ground, you're going offroading.

As we consider the various aspects of this, it should be remembered that V1, Vr and V2 happen in awfully fast succession.

...I think I'm sticking with my much-less complex "Denver Briefing"...OK PNF, it's a long-ass runway, so if the damn thing won't rotate or the controls feel wrong, or we see an engine literally explode and take out half the wing before V2, we'll go ahead and abort...you call the tower.

I'm just wondering how long until the guys who actually do this tell us how stupid we are, so we know to abort.

[Evan]

As we consider the various aspects of this, it should be remembered that V1, Vr and V2 happen in awfully fast succession.

But V2 never happens if you can't get off the runway (in my proposed scheme, the bug disappears after Vr+10 until you get out of ground mode). So you have a crew barrelling down a runway after Vr and the thing won't lift off. They need to know how long before they no longer have any chance of stopping either. For an A380, it's probably V1 but for a G5 on a 14,000ft strip I think that would be valuable information...

[3WE]

But V2 never happens if you can't get off the runway

Que?

V2 is the pre-calculated weight-& atmosphere-adjusted velocity at which you should lift off (assuming your cowboy, seat of the pants technique of rotating and attitude control are well honed)...liftoff may not happen, but V2 does indeed happen...

And, if you haven't lifted off and there isn't a good reason, you should be aware that something big is wrong, BUT cue the fundamental concept that it's generally bad to abort (unless we're on a 3-mile long runway)...

Again, V1, Vr and V2 come in fairly rapid succession (I believe that 3 seconds is a rough, but realistic estimate). If you aren't flying by V2 should you really suddenly focus on V3p at that time- to keep trying so you can be sure you arrive at the piano keys with red hot brakes and tires that are going to blow?

If we're gonna rewrite the books, I don't see why you want to scream on past V2 on up to V3p when you can still stop just fine from V2.....

[Gabriel]

As I understand it, V1 is the speed above which the takeoff cannot successfully be aborted.

That's a misconception. V1 is a speed where the take off can be successfully aborted AND where the take off, in the event of an engine failure at that speed, can be successfully continued.

Below V1, it must be possible to abort the take off;
Above V1, it must be possible to continue the take-off in the event of an engine failure.

But nothing is said about what happens with an aborted take-off above V1 or with a continued take-off after an engine failure below V1.

That you can stop from V1, doesn't mean that you can't also stop from a faster speed. And that you can continue the take-off after an engine failure at V1, doesn't mean that you can't also continue the take-off after an engine failure at a slower speed.

If you read carefully my simplified version of the definition, you will see that I wrote "a speed", not "the speed". That is because, unless the plane is operating for a runway with the minimum length that it can legally operate for it's weight, selected take-off thrust and weather conditions, there will be a range of speeds that meet the requirements of V1.

If you read the FAR 25 (14 CFR part 25), you will see that it says that V1 "is a speed selected by the operator that...." and start to enumerate the requirements.

To make it easy to visualize, say that you have a runway 50 miles long.
You will be able to successfully abort the take-off from any speed below VR (and, with 10 miles, much more than that, but VR is one of the upper bounds for V1 in the FAR). At the same time, you will be able to continue the take-off after an engine failure from any speed that gives you enough control authority to arrest the asymmetric thrust (Vmcg, minimum control speed on ground), which is the lower bound for V1 in the FAR.

Any speed between these two is a valid V1. Select a V1 in the middle of the range and you can successfully abort the take-off from a speed greater than V1 (but below VR), and also continue the take-off with an engine failure below V1 (but above Vmcg);

Presumably "successfully" in this case means without causing an accident.

For the sake of V1, "successfully" means with the plane stopped within the limits of the runway (or the stopway if there is one) for the aborted take-off scenario, and with the airplane at V2 and 35ft AGL over the end of the runway (or of the clearway if there is one) in the continued take-off scenario.

In this case during the takeoff sequence (per the report), the "V1" call was made and about 1-1/2 seconds later the "rotate" call was made. The phrase "lock is on" is heard on the CVR about 1 seconds after the "rotate" call - that presumably represents the first time the pilot(s) identified the fact the gust lock was engaged.

But here's the good part, from page 48 of the report:

The airplane performance study determined that, if the flight crew had initiated a rejected takeoff in accordance with the G-IV AFM at the time of the PIC’s first “lock is on” comment or at any time up until about 11 seconds after this comment, the airplane could have been stopped on the paved surface.

So basically they're saying that up to about 13 seconds after the "V1" call, the takeoff could have been aborted safely.[/quote]
Yep, that's not unusual.

Or in other words, the aircraft's speed and distance down the runway when "V1" was called did not conform to the meaning of "V1".

Nope. What they are saying is that there was much more runway than needed for the take-off.

IMHO taking that into account, the concept of a "V3" speed where an abort could be made with a possibly-survivable crash doesn't pertain to this accident. Instead, the pilots presumably thought the takeoff could not be aborted safely when the gust lock was found engaged since that happened after the V1 call which by definition means a safe abort is no longer possible. [Well, I hope it is clear by now that this statement is not correct] Yet in this case per the NTSB data, a safe abort *was* possible... and if the pilots had known that, the outcome quite likely would have been different.

I wonder what they were doing or thinking during the 11 seconds after the first "lock is on" comment.

The SOP is that if you judge that the plane is unable to fly, you abort at any speed, with the idea that if you are going to crash, it is preferable (but no guarantee) to do so while slowing down rather than while speeding up.

Evidently, the plane is not able to fly with the control lock on. So an abort even past V1 would have been appropriate.

The only "logical" (from an SOP POV) possibility that I can think is that they judged they would be bale to release the locks and the plane would be able to fly so they preferred not to risk an abort past V1, but then they fought and couldn't release the lock, so at a much later time they judged that the plane would not fly and changed their mind and decided to abort even if they were a speeding rocket by then.

[Gabriel]

While anything over 80-100kts is probably unsafe-to-abort I want to reaffirm my advocacy for a third V-speed above V1 (whenever the conditions exist) that is 'still-possible-to-unsafely abort speed', beyond which the crew will know that an abort probably results in an overrun.

That information is and has always been right there in the airplane manual (the manufacturer one, not necessarily the operator's one).

The manual will give you the range of speeds where you can select your V1 from for any given take-off. The lowest limit of the range is the minimum speed where an angine can fail and you can continue the take-off and achieve 35ft and V2 over the edge of the runway. The higher limit is the maximum from which can stop the plane on the runway. If we call them V1L and V1H, the calls and thoughts would be:

V1L: Now we can continue the take-off if an engine fails.
V1H: Now we cannot stop on the runway if we need to abort.

[Gabriel]

(Of course a "Flight controls free and correct check" is another additional procedure that should be considered for this particular case)

Are you sure that that is the exact procedure that applies for this specific airplane type, model, year and color scheme?

[Gabriel]

Que?

V2 is the pre-calculated weight-& atmosphere-adjusted velocity at which you should lift off (assuming your cowboy, seat of the pants technique of rotating and attitude control are well honed)...

No. That's Vlo (V lift-off).

V2 is the speed that you would achieve at most over the limit of the runway if an engine fails at Vef (V1 minus reaction time) and the speed that you can use during the initial climb in the event of an engine failure and it will ensure minimum climb gradient, obstacle clearance, and flight characteristics.

That said, in those rare events that the engine fails to fail at V1 (or at any other speed), with the plane accelerating twice as fast, V2 might be achieved and exceeded with the plane still on the ground.

[3WE]

Are you sure that that is the exact procedure that applies for this specific airplane type, model, year and color scheme?

I do not know.

But, much like the 35,000 ft stall incident, this is just one of those broadly-applicable fundamental things that would have prevented the crash (even if you are an idiot about the specifics of V speeds).

[Evan]

I wonder what they were doing or thinking during the 11 seconds after the first "lock is on" comment.

The SOP is that if you judge that the plane is unable to fly, you abort at any speed, with the idea that if you are going to crash, it is preferable (but no guarantee) to do so while slowing down rather than while speeding up.

The SOP is to be "go-minded" after V1. In most cases I believe this is what you are calling V1L because, if the plane is safe to fly, it is safer to lift off and return than to reject at high speed.

I think they were locked into being "go-minded" after the V1 call and in the seconds of realization that the locks were on they couldn't overcome this ingrained mentality. I think it's just human nature and dogmatic thinking.

If, however, the PFD at that moment was showing a line indicating the range between V1L and V1H, pilots might be more situationally aware that they can still reject after V1L and MUST reject by V1H. In a light jet on a very long runway that range would allow them time to think rationally.