777 strikes approach lights on departure 9/15

[derg]

Hmm.

I know it's easy to say "I'd have done things differently", especially when you know there were mistakes made, but even in a 319 I'd set the parking brake, take a real "jaundiced, mistrustful" look at the performance data and think "is this really worth it?" before accepting a departure from T1.

Now a 767? Nope. 330? Forget it. 777-300? Having never flown one, I seriously doubt I would even consider doing that, especially without a static takeoff.

You may have "the numbers" for that, but there may be other considerations like PKOF on the 76 or APU on the 330/320 for improved performance. Generally too much stuff to worry about during a critical phase of operation.

 

[MikeD]

It's not just the abort, it's also crossing 35' on the DER and first/second segment climbs. These guys under rotated or grossly neglected to verify the intersection takeoff (probably apparent). .

The abort is where the "extra credit" the runway gives you, is what I'm referring to in this case. If your TOLD data checks out, making the DER crossing generally isn't an issue. Something went wrong indeed, I'm just curious what their driving factors would've been for this crew. Time will tell as the info comes out.

 

[Alchemy]

Wonder what the reason was for an intersection takeoff? Not much is generally gained from that in any normal ops for a large/heavy jet, apart from runway behind you right from the get go, in this case a significant amount of runway.

There are really only two reasons to accept an intersection: you're trying to save time or to acommodate an ATC request/construction. With that said, the performance most airlines operate with these days is just going to give you a lower thrust setting when you use a longer runway, so your stop margin doesn't really improve most of the time, unless you use airmanship and judgement to force a higher thrust setting.

The 747-8 produces an EICAS master caution if you advance thrust to takeoff power at a position other than they runway you've programmed into the FMC. The warning is even produced spuriously in China at airports that don't comply with the WGS-84 datum. I wonder if the 777 has that. Even if it did, I'm not sure if it works for intersection departures as it might just check for lateral displacement from runway centerline.

 

[MikeD]

[

I know it's easy to say "I'd have done things differently", especially when you know there were mistakes made, but even in a 319 I'd set the parking brake, take a real "jaundiced, mistrustful" look at the performance data and think "is this really worth it?" before accepting a departure from T1.

Agreed. Hence why I'm curious what's discovered about that.

Now a 767? Nope. 330? Forget it. 777-300? Having never flown one, I seriously doubt I would even consider doing that, especially without a static takeoff.

Are rolling takeoffs common for these jets? In all our jets, it was always static takeoff for engine/AB checks, etc..

You may have "the numbers" for that, but there may be other considerations like PKOF on the 76 or APU on the 330/320 for improved performance. Generally too much stuff to worry about during a critical phase of operation.

Agreed. That's why I wonder what the major gain(s) would've been, for the runway decided to be left behind.

 

[derg]

The abort is where the "extra credit" the runway gives you, is what I'm referring to in this case. If your TOLD data checks out, making the DER crossing generally isn't an issue. Something went wrong indeed, I'm just curious what their driving factors would've been for this crew. Time will tell as the info comes out.

IF, they had performance data for the intersection.

Hell, even for the long haul flights, we'd have full-length wet, dry and maybe something like 27L Y11 or Y12 (using CDG as an example because I've flown that a billion times) but rarely anything as restrictive as T-1 in MIA without a crap-ton of other variables for improved performance.

 

[MikeD]

There are really only two reasons to accept an intersection: you're trying to save time or to acommodate an ATC request/construction. With that said, the performance most airlines operate with these days is just going to give you a lower thrust setting when you use a longer runway, so your stop margin doesn't really improve most of the time, unless you use airmanship and judgement to force a higher thrust setting.

The 747-8 produces an EICAS master caution if you advance thrust to takeoff power at a position other than they runway you've programmed into the FMC. The warning is even produced spuriously in China at airports that don't comply with the WGS-84 datum. I wonder if the 777 has that. Even if it did, I'm not sure if it works for intersection departures as it might just check for lateral displacement from runway centerline.

Cool info on the 747-8. Agree on the basics for an intersection departure, and I can see it easily in a piston/turboprop plane. Even lighter jet. Just seems a larger risk for a large/heavy jet, with not a ton of gain.

 

[derg]

Agreed. Hence why I'm curious what's discovered about that.

Me too!

Are rolling takeoffs common for these jets? In all our jets, it was always static takeoff for engine/AB checks, etc..

Rolling takeoffs are preferred except when necessary for performance. I don't specially make the SAC ALERT departure, but I'll generally get lined up, as we roll slowly, then spool the engines then go to the FLX or TOGA detent. More of a modified rolling takeoff. I'm generally very conservative until I know the engines have spooled up symmetrically because I've heard too many stories of "Pratt" having a very different attention span than "Whitney" and heading off into the weeds.

 

[MikeD]

IF, they had performance data for the intersection.

Hell, even for the long haul flights, we'd have full-length wet, dry and maybe something like 27L Y11 or Y12 (using CDG as an example because I've flown that a billion times) but rarely anything as restrictive as T-1 in MIA without a crap-ton of other variables for improved performance.

Agreed, assuming they had the numbers for the remaining length at the intersection.

And yes, I can see the Y-11/12 at CDG, because you have what, ~11,500-12,000 still remaining?

 

[MikeD]

Rolling takeoffs are preferred except when necessary for performance. I don't specially make the SAC ALERT departure, but I'll generally get lined up, as we roll slowly, then spool the engines then go to the FLX or TOGA detent. More of a modified rolling takeoff. I'm generally very conservative until I know the engines have spooled up symmetrically because I've heard too many stories of "Pratt" having a very different attention span than "Whitney" and heading off into the weeds.

A slow-roll rolling takeoff makes sense. Yeah, fun is having a rolling takeoff and getting an asymmetrical afterburner lightoff, and the bucking bronco begins.

 

[derg]

Agreed, assuming they had the numbers for the remaining length at the intersection.

And yes, I can see the Y-11/12 at CDG, because you have what, ~11,500-12,000 still remaining?

Pretty much. Generally they'll queue the narrowbodies on K7 and the heavies and "seeeew'pairs" on Y11 and Y12.

Personally, if there's any question about performance, I'm going to roll with the most reasonably conservative approach (kick out the illegals and stop OBAMAcare… wait, wrong conservative) and be the unpopular guy and say we're going to wait for full length if I feel it's necessary or involves too much FMS downtime, aircraft configuration changes and other considerations. Every time you touch something, there's a small probability of error, and the more you dick with soething, that microscopic percentage becomes a large, whole number very quickly.

If we get too focused making something work for ATC and overlook configuration, performance and situational awareness, bad stuff happens.

 

[Alchemy]

Cool info on the 747-8. Agree on the basics for an intersection departure, and I can see it easily in a piston/turboprop plane. Even lighter jet. Just seems a larger risk for a large/heavy jet, with not a ton of gain.

Agreed, and as a heavy jet operator we turn down intersection departures regularly even at places like ORD because 1: We're often in a 400+ ton airplane, if that isn't justification to use all the pavement, I don't know what is 2: We don't want to go heads down and take the time to set everything back up. It's not like the ERJ where I could just whip out the performance book and 10 seconds later say "yep we're still good". 3: Even if the stop margin doesn't improve from full length because of the planned reduced thrust, it saves engine life and if there is an engine problem on takeoff, there will be a less violent yaw displacement.

 

[typhoonpilot]

That would be a pretty huge mistake to go off of T1 if they meant to go off of full length.

I don't operate out of Miami so a couple of questions:

1) Is T1 the normal departure default point used for departures off of RWY 09? Or is full length the normal point?

2) Is there any construction or closure of Taxiway S that is affecting the use of full length?

Some things that are probably useful for speculation:

At most foreign airlines the crew compute their own takeoff and landing performance. It is not like the U.S. majors where central load planning do all the computations. That means the crew get the final zero fuel weight and, after calculating the takeoff weight, input that into the Onboard Performance Tool (OPT). A lot can go wrong with the use of an OPT. In this case one of the things that could have gone wrong is the use of full length data when intersection T1 was used. Another thing is input of the wrong takeoff weight. 100 tonne errors are not unheard of. I'm actually working on a paper about that subject at the moment. B777 crews are quite used to takeoff weights in the 200+ tonne range. It is usually only on ultra long haul flights that 300+ tonne weights are achieved. That means a mistake of typing in 250,000 tonnes instead of 350,000 tonnes can go unnoticed because it is a "reasonable", and actually more normal, weight for the aircraft. That is one of the primary human factors issues that caused the Emirates Melbourne tail strike in an A340-500.

From my understanding, you can't tail strike a 777 normally. They didn't tail strike it as well it seems like....

You are correct, in the B777 the Primary Flight Computers calculate if a tail strike is imminent and decrease elevator deflection, if required, to reduce the potential for a tail strike. The feature is so good that Boeing has removed the tail skid from new build B777-300ERs.

They hit the lights and continued on. Wonder if the nonunion, lack of a safety culture contributed to this...

Ignorant comment Seggy. I can't speak for Qatar from personal experience, but I know they train to a high standard with all the modern tools for safety and CRM incorporated. Union or non-union has nothing to do with it.

Procedurally if they had no indications of a tail strike and pressurization was normal there would be no requirement to return to the departure airport. If, OTOH, they received a TAIL STRIKE caution the checklist requires landing at the nearest suitable airport. There have been B777-300 tail skid contacts in the past where the tower reported it to the aircraft. Again, even with that, unless the TAIL STRIKE EICAS comes up, the written procedure from Boeing is to continue as normal.



Typhoonpilot

 

[Alchemy]

That would be a pretty huge mistake to go off of T1 if they meant to go off of full length.

I don't operate out of Miami so a couple of questions:

1) Is T1 the normal departure default point used for departures off of RWY 09? Or is full length the normal point?

2) Is there any construction or closure of Taxiway S that is affecting the use of full length?

Some things that are probably useful for speculation:

At most foreign airlines the crew compute their own takeoff and landing performance. It is not like the U.S. majors where central load planning do all the computations. That means the crew get the final zero fuel weight and, after calculating the takeoff weight, input that into the Onboard Performance Tool (OPT). A lot can go wrong with the use of an OPT. In this case one of the things that could have gone wrong is the use of full length data when intersection T1 was used. Another thing is input of the wrong takeoff weight. 100 tonne errors are not unheard of. I'm actually working on a paper about that subject at the moment. B777 crews are quite used to takeoff weights in the 200+ tonne range. It is usually only on ultra long haul flights that 300+ tonne weights are achieved. That means a mistake of typing in 250,000 tonnes instead of 350,000 tonnes can go unnoticed because it is a "reasonable", and actually more normal, weight for the aircraft. That is one of the primary human factors issues that caused the Emirates Melbourne tail strike in an A340-500.




You are correct, in the B777 the Primary Flight Computers calculate if a tail strike is imminent and decrease elevator deflection, if required, to reduce the potential for a tail strike. The feature is so good that Boeing has removed the tail skid from new build B777-300ERs.



Ignorant comment Seggy. I can't speak for Qatar from personal experience, but I know they train to a high standard with all the modern tools for safety and CRM incorporated. Union or non-union has nothing to do with it.

Procedurally if they had no indications of a tail strike and pressurization was normal there would be no requirement to return to the departure airport. If, OTOH, they received a TAIL STRIKE caution the checklist requires landing at the nearest suitable airport. There have been B777-300 tail skid contacts in the past where the tower reported it to the aircraft. Again, even with that, unless the TAIL STRIKE EICAS comes up, the written procedure from Boeing is to continue as normal.



Typhoonpilot

1) Full length is normal. Granted I usually operate from the western U freight area, west of T1, so full length is almost as convenient. I have never even been offered T1 in the dozen or so times I've operated out of there, I don't recall it being advertised on the ATIS either, although I'm no MIA expert.

2) I operated out of MIA on both Sunday and Wednesday. I just reviewed our NOTAMs from both flights and see no mention of construction on either flight's paperwork.

 

[Alchemy]

Also, I will mention this in a purely speculative tone:

MIA tower has a habit of clearing aircraft for takeoff well before they are close to the end of the runway. This isn't abnormal, per se, in the US but it did take me by surprise the other night when tower cleared me for takeoff while I was still half a mile from the holding point at the end of the runway (It was also about 4 AM local though and clearance/ground/tower were all being conducted on tower frequency with one controller).

MIA ground sometimes issues non-standard taxi instructions. I regularly hear "transition to S, taxi to western U" when I vacate runway 9 after landing. It's not a big deal, but it leaves some ambiguity as to when an aircraft is expected to transition from taxiway T to taxiway S. Maybe that is actually an OK taxi instruction per the FAA, it's just not very common.

These probably have no relevance to the incident, just things I've noticed.

 

[Screaming_Emu]

View attachment 32792

Hmm.

I know it's easy to say "I'd have done things differently", especially when you know there were mistakes made, but even in a 319 I'd set the parking brake, take a real "jaundiced, mistrustful" look at the performance data and think "is this really worth it?" before accepting a departure from T1.

Now a 767? Nope. 330? Forget it. 777-300? Having never flown one, I seriously doubt I would even consider doing that, especially without a static takeoff.

You may have "the numbers" for that, but there may be other considerations like PKOF on the 76 or APU on the 330/320 for improved performance. Generally too much stuff to worry about during a critical phase of operation.

TEM.

Numbers may say you can do something. But I've landed, had one brake be inop, the other tricked the anti skid, and we used WAY more runway than planned. Thankfully we had way more than we needed.

We aren't the last flight out of 'Nam. There just isn't a reason to expose yourself to more risk than is necessary.

 

[Minuteman]

In TYOOL 2015, how are there not webcams pointed at every runway of of the 20-busiest airports in the western world?

 

[jtrain609]

Pretty much. Generally they'll queue the narrowbodies on K7 and the heavies and "seeeew'pairs" on Y11 and Y12.

Personally, if there's any question about performance, I'm going to roll with the most reasonably conservative approach (kick out the illegals and stop OBAMAcare… wait, wrong conservative) and be the unpopular guy and say we're going to wait for full length if I feel it's necessary or involves too much FMS downtime, aircraft configuration changes and other considerations. Every time you touch something, there's a small probability of error, and the more you dick with soething, that microscopic percentage becomes a large, whole number very quickly.

If we get too focused making something work for ATC and overlook configuration, performance and situational awareness, bad stuff happens.

Do guys go to SFO at all? If you call clearance to ask them which runway to program they'll tell you both. When they're departing the 28's and I tell them, "Sure we can take 28R for departure, but it'll be about 3 minutes to reprogram and run a checklist" they either get REALLY angry or just keep you on 28L.

 

[bp]

Wow. Well, I'm sure the BA crew in Vegas was glad they took the whole runway...

 

[Nick]

Absent some more information I am not going to enter the [good] discussion that is going on here, but I was just thinking.

Imagine you're way back in economy class, and watching the "Nose Cam" forward view on your personal entertainment as the plane takes the runway (quite a few new widebodies on various non-US carriers have this; perhaps Qatar 777-300s have them).

The centerline lights up on your little TV screen as the nose lights come on and the plane starts to accelerate down the runway. Cool...a trip begins.

Then as you're watching the runway go by, you see what, to a layman, would maybe kinda look like the end of the runway go by, because the stripes end and it looks pretty dark except for those little light bulbs going by...really fast.

Then there is a *BAM* right below your seat somewhere which is just a few rows from the aft end of the coach cabin...and even a slight thud is felt.

Off you go, into the night . . . across that big, dark ocean!


Maybe they don't have that entertainment option, and maybe the light-pole impact wasn't loud enough above the noise of the engines and wind, but . . . what if??

I think I'd be ringing that flight attendant call button!

 

[mikecweb]

TEM.

Numbers may say you can do something. But I've landed, had one brake be inop, the other tricked the anti skid, and we used WAY more runway than planned. Thankfully we had way more than we needed.

We aren't the last flight out of 'Nam. There just isn't a reason to expose yourself to more risk than is necessary.

You say that now....