Cargojet slide off the end of 08L at YVR

[lostaviator]

This approach seemed rushed considering they just discovered the issue on the base turn. One little loop south of final, then requested direct the IF for a 80 degree intercept (which they blew through). The controller clearly saw an issue with their request for direct the IF and he even asked them if this was ok. Sharp intercept and faster than normal ground speed. No way they got through the entire flap checklist in that amount of time. Multiple pilots talking on the radio. Seems messy and rushed.

[VFS]

From the audio recordings, kinda funny the skipper had to jump in and help make 30% of the radio calls, while being paid only 30% of what a 767 skipper should be paid.

[Nauclerus]

then requested direct the IF for a 80 degree intercept (which they blew through).

They didn't blow through the final, they requested direct to EMKAM when they were already south of the LOC on a radar vector heading of 050. It looks like they intercepted and then maintained the LOC with no deviations. They also leveled off at 2000 feet well before intercepting the glideslope from below at a speed consistent with their final approach speed. From that point to the threshold the approach appeared to be stabilzed.

From the audio recordings, kinda funny the skipper had to jump in and help make 30% of the radio calls, while being paid only 30% of what a 767 skipper should be paid.

I'm not sure what your point is but It's typical airline SOPs for the Pilot Flying to take the radios while the PNF is reading/doing the Emergency/Non-Normal Checklists.

If the trailing edge flaps were greater then 20 degrees (which it appears they were) it would take about 2 minutes to read and complete the Leading Edge Slat Asymmetry Checklist. The only additional actions required are setting the ground proximty swith to override, setting the approach speed to VREF30+20, and then checking the required landing distance in the QRH.

[boeingboy]

ADSB data shows it departing the runway at over 100 knots. You don't go nearly 2000 feet off the runway at 50 knots.

This is the ADSB Data I'm seeing. (FlightRadar24, Google Earth)

Time, Altitude, Ground Speed, Distance

09:27:45, 1050 feet, 160 knots, 2.93 nm (from threshold)
09:27:55, 900 feet, 162 knots, 2.49 nm
09:28:04, 750 feet, 165 knots, 2.03 nm
09:28:14, 625 feet, 166 knots, 1.58 nm
09:28:23, 475 feet, 164 knots, 1.15 nm
09:28:32, 325 feet, 166 knots, 0.73 nm
09:28:38, 250 feet, 172 knots, 0.45 nm
09:28:44, 125 feet, 173 knots, 0.17 nm
09:28:54, 25 feet, 169 knots, -8000 feet (from end of runway)
09:29:01, 50 feet, 159 knots, -5967 feet
09:29:09, 50 feet, 147 knots, -4119 feet
09:29:15, 50 feet, 136 knots, -2542 feet
09:29:25, 50 feet, 120 knots, -532 feet
09:29:31, 50 feet, 109 knots, +516 feet (from past end of runway)
09:29:36, 0 feet, 76 knots, +1485 feet
09:29:38, 0 feet, 53 knots, +1661 feet
09:29:41, 0 feet, 27 knots, +1887 feet

How accurate is ADSB data close to the ground?

I don't think i would trust those numbers. I highly doubt a Boeing 767 - with any flaps - could stay airborne and flying below 120 kts.....never mind 109 kts.

[Inverted2]

Good thing the ground was soggy or it would have went straight into the shopping mall.

[PostmasterGeneral]

Those ARFF guys on the radio need to learn when to shut up.

[digits_]

This is the ADSB Data I'm seeing. (FlightRadar24, Google Earth)

Time, Altitude, Ground Speed, Distance

09:27:45, 1050 feet, 160 knots, 2.93 nm (from threshold)
09:27:55, 900 feet, 162 knots, 2.49 nm
09:28:04, 750 feet, 165 knots, 2.03 nm
09:28:14, 625 feet, 166 knots, 1.58 nm
09:28:23, 475 feet, 164 knots, 1.15 nm
09:28:32, 325 feet, 166 knots, 0.73 nm
09:28:38, 250 feet, 172 knots, 0.45 nm
09:28:44, 125 feet, 173 knots, 0.17 nm
09:28:54, 25 feet, 169 knots, -8000 feet (from end of runway)
09:29:01, 50 feet, 159 knots, -5967 feet
09:29:09, 50 feet, 147 knots, -4119 feet
09:29:15, 50 feet, 136 knots, -2542 feet
09:29:25, 50 feet, 120 knots, -532 feet
09:29:31, 50 feet, 109 knots, +516 feet (from past end of runway)
09:29:36, 0 feet, 76 knots, +1485 feet
09:29:38, 0 feet, 53 knots, +1661 feet
09:29:41, 0 feet, 27 knots, +1887 feet

How accurate is ADSB data close to the ground?

I don't think i would trust those numbers. I highly doubt a Boeing 767 - with any flaps - could stay airborne and flying below 120 kts.....never mind 109 kts.

Is ground effect a big factor on that size airplane? Enough to keep it floating for a while?

[Blackdog0301]

Listening to the audio, I immediately take notice of the FO's slow and messy radio calls. I'd have to think that he was at least relatively low experience based on that alone. But with 20k lbs of gas remaining, this also seems extremely rushed. With that much gas, and a complex checklist such as a slat/flap disagree/asymmetry, I'd be headed out over the water to do the checklist at a slow and methodical pace while also burning off some of that gas and reducing the weight. This screams inexperience to me.

[altiplano]

then requested direct the IF for a 80 degree intercept (which they blew through).

They didn't blow through the final, they requested direct to EMKAM when they were already south of the LOC on a radar vector heading of 050. It looks like they intercepted and then maintained the LOC with no deviations. They also leveled off at 2000 feet well before intercepting the glideslope from below at a speed consistent with their final approach speed. From that point to the threshold the approach appeared to be stabilzed.

From the audio recordings, kinda funny the skipper had to jump in and help make 30% of the radio calls, while being paid only 30% of what a 767 skipper should be paid.

I'm not sure what your point is but It's typical airline SOPs for the Pilot Flying to take the radios while the PNF is reading/doing the Emergency/Non-Normal Checklists.

If the trailing edge flaps were greater then 20 degrees (which it appears they were) it would take about 2 minutes to read and complete the Leading Edge Slat Asymmetry Checklist. The only additional actions required are setting the ground proximty swith to override, setting the approach speed to VREF30+20, and then checking the required landing distance in the QRH.

Trailing edge wouldn't have been greater than 20 when they recognized the problem and started the CL though. Downwind/base? That's more than a 2 minute CL if I was running it.

Plus you have to pull landing performance data.

How does CJT get their performance?

Just listened to the audio, seems rushed.

[JungleRiot]

Listening to the audio, I immediately take notice of the FO's slow and messy radio calls. I'd have to think that he was at least relatively low experience based on that alone. But with 20k lbs of gas remaining, this also seems extremely rushed. With that much gas, and a complex checklist such as a slat/flap disagree/asymmetry, I'd be headed out over the water to do the checklist at a slow and methodical pace while also burning off some of that gas and reducing the weight. This screams inexperience to me.

Crew fatigue should definitely be considered a contributing factor, since the crew seemed to be doing a redeye. I did a flap asymmetry scenario in the sim a few months back, and the entire checklist takes about 30 minutes, including landing performance numbers.

Diverting to settle would also have been another good option since 34R is 12,000'

[JungleRiot]

Listening to the audio, I immediately take notice of the FO's slow and messy radio calls. I'd have to think that he was at least relatively low experience based on that alone. But with 20k lbs of gas remaining, this also seems extremely rushed. With that much gas, and a complex checklist such as a slat/flap disagree/asymmetry, I'd be headed out over the water to do the checklist at a slow and methodical pace while also burning off some of that gas and reducing the weight. This screams inexperience to me.

Crew fatigue should definitely be considered a contributing factor, since the crew seemed to be doing a redeye. I did a flap asymmetry scenario in the sim a few months back, and the entire checklist takes about 30 minutes, including landing performance numbers.

Diverting to settle would also have been another good option since 34R is 12,000'

[VFS]

then requested direct the IF for a 80 degree intercept (which they blew through).

They didn't blow through the final, they requested direct to EMKAM when they were already south of the LOC on a radar vector heading of 050. It looks like they intercepted and then maintained the LOC with no deviations. They also leveled off at 2000 feet well before intercepting the glideslope from below at a speed consistent with their final approach speed. From that point to the threshold the approach appeared to be stabilzed.

From the audio recordings, kinda funny the skipper had to jump in and help make 30% of the radio calls, while being paid only 30% of what a 767 skipper should be paid.

I'm not sure what your point is but It's typical airline SOPs for the Pilot Flying to take the radios while the PNF is reading/doing the Emergency/Non-Normal Checklists.

Yea I am aware, that’s clearly not what took place if you listen to the recordings. Typical airline sop’s is for the failure to be identified, the PF to call for the appropriate memory items or checklist and to state “I have atc” or some version of that.
Not for the constant switching of the person responding to ATC during the same set of instructions or questioning. That was my point.

[pelmet]

Diverting to settle would also have been another good option since 34R is 12,000'

Not sure about typical 767 fuel burns but if the slats are locked out partially, it will be difficult to get a fuel burn estimate for the flight which could also have restricted altitude and airspeed. Might not be able to figure it out until at cruise altitude, possibly only at 20,000 feet.

[Nauclerus]

I don't think i would trust those numbers. I highly doubt a Boeing 767 - with any flaps - could stay airborne and flying below 120 kts

I'm skeptical as well but as I mentioned earlier there is a video that pans from the threshold to the aircraft that appears to show the main tire tracks start digging into the ground at about 500 feet past the end of the runway.

Trailing edge wouldn't have been greater than 20 when they recognized the problem and started the CL though. That's more than a 2 minute CL if I was running it. Plus you have to pull landing performance data.

The QRH seems to indicate the problem can be recognized with flaps greater than 20, it's the first choice.

B767-300 QRH
LEADING EDGE FLAP ASYMMETERY
Messages: LE SLAT ASYM
Condition: The leading edge slats are not symmetrically extended
Note: Do not use FMC fuel predictions with the flaps extended
1 GND PROX OVRD Switch .... OVRD
2 Choose one:
- Indicated flap position is greater than 20:
- use current flaps and VREF 30 + 20 for landing
-> go to step 12
12 Check the Non-Normal Configuration landind Distance tables in the Performance Inflight-QRH chapter.
13 Checklist complete except for deferred items.

Completing the above in around 10 minutes doesn't necessarily seem unrealistic. Maybe they were already familiar with that section of the QRH facilitating its timely completion ?

TSB Report #A23F0232: C-GAZI, a CargoJet B767-338, was operating flight CJT2384 from Vancouver International Airport (CYVR), BC, to Hamilton/John C. Munro International Airport (CYHM), ON. During the cruise phase of the flight at FL350, the crew received multiple EICAS messages. The crew started working down the list and got a LE SLAT DISAGREE. As the checklist reached "changing flap configuration", the crew noted a STANDBY BUS OFF message. Figuring that is what caused it, they started that checklist. In the meantime, the crew requested a descent out of RVSM airspace, since the autopilot and autothrottle were INOP. A PAN PAN was declared. The crew also talked to MCC and Dispatch, finished the checklists and continued to destination without further incident.

Typical airline sop’s is for the failure to be identified, the PF to call for the appropriate memory items or checklist and to state “I have atc” or some version of that. Not for the constant switching of the person responding to ATC during the same set of instructions or questioning. That was my point.

The transmisions to ATC went ... FO - FO - C- FO - C - C - C - FO - FO ... So once, hardly "constant switching". Doesn't seem unreasonable to me.

[altiplano]

Trailing edge wouldn't have been greater than 20 when they recognized the problem and started the CL though. That's more than a 2 minute CL if I was running it. Plus you have to pull landing performance data.

The QRH seems to indicate the problem can be recognized with flaps greater than 20, it's the first choice.

B767-300 QRH
LEADING EDGE FLAP ASYMMETERY
Messages: LE SLAT ASYM
Condition: The leading edge slats are not symmetrically extended
Note: Do not use FMC fuel predictions with the flaps extended
1 GND PROX OVRD Switch .... OVRD
2 Choose one:
- Indicated flap position is greater than 20:
- use current flaps and VREF 30 + 20 for landing
-> go to step 12
12 Check the Non-Normal Configuration landind Distance tables in the Performance Inflight-QRH chapter.
13 Checklist complete except for deferred items.

You're not calling for FLAP 20+ on the STAR.

The second choice is indicated flap position is less than 20 and it leads you to a longer checklist.

And I'll ask again, how does CJT get their performance data for an abnormal landing? QRH? ACARS? Did they do it here?

[Jean-Pierre]

Good thing the ground was soggy or it would have went straight into the shopping mall.

This photo give context about how far they overran.

FB_IMG_1732113271664.jpg (16.02 KiB) Viewed 3676 times

[digits_]

Good thing the ground was soggy or it would have went straight into the shopping mall.

This photo give context about how far they overran.

FB_IMG_1732113271664.jpg

Holy cow. It's amazing that didn't end in a big fireball.

[Inverted2]

Those ARFF guys on the radio need to learn when to shut up.

Imagine if that was in Montreal. Everything those Kay-Bekkers say is 3x as long as English.

[Nauclerus]

You're not calling for FLAP 20+ on the STAR. And I'll ask again, how does CJT get their performance data for an abnormal landing? QRH? ACARS?

It's been a few years but it was common when landing 08L off that STAR for ATC to slam dunk you high to final when traffic was light. A lot of pilots would start selecting flaps abeam the airport on the downwind in anticipation. Maybe it's changed in the last few years ? It does seem a bit early for Flap 20 but the point being they might have been there earlier than usual. It seemed to happen around the time they were given the base turn.

I'm not sure if Cargojet uses ACARS for landing data ? If using a QRH (book or ipad) it wouldn't take that long to get a distance, a minute or so at most. But I get it it's the middle of the night and you're tired, so maybe an extra minute or so then.

[co-joe]

...
I'm not sure if Cargojet uses ACARS for landing data ? If using a QRH (book or ipad) it wouldn't take that long to get a distance, a minute or so at most. But I get it it's the middle of the night and you're tired, so maybe an extra minute or so then.

Using OPT or a similar program would be quick, but even finding the right page in the PI chapter of the QRH can take several minutes, nevermind using a chart you've likely looked at once several years ago, and then having the other guy verify the data. Odds are they either got the data electronically, or just skipped this step.

Any more info on the fact that the plane had gone mud bogging on its previous flight and had major brake work done? I don't imagine an autobrake or anti skid MEL would have helped the situation at all...

[altiplano]

You're not calling for FLAP 20+ on the STAR. And I'll ask again, how does CJT get their performance data for an abnormal landing? QRH? ACARS?

It's been a few years but it was common when landing 08L off that STAR for ATC to slam dunk you high to final when traffic was light. A lot of pilots would start selecting flaps abeam the airport on the downwind in anticipation. Maybe it's changed in the last few years ? It does seem a bit early for Flap 20 but the point being they might have been there earlier than usual. It seemed to happen around the time they were given the base turn.

I'm not sure if Cargojet uses ACARS for landing data ? If using a QRH (book or ipad) it wouldn't take that long to get a distance, a minute or so at most. But I get it it's the middle of the night and you're tired, so maybe an extra minute or so then.

Sorry man, but I think you're reaching.

I've flown big jets on that arrival plenty and NEVER would do that.
You're not fastest going down on an unrestricted descent by slowing for flap 25. Placarded speed for that is 180 KIAS. Speedbrakes and maybe flap 5 at most but usually not until base.

So they are less than Flap 20 when they get the problem and the checklist has some extra steps.

In the QRH perf inflite you have like 30 or 40 abnormal landing distance charts to flip through to find the right one.

The base distance when you enter the chart for leading edge asymmetry with good reported braking action is ~ 8500' or 7800' unfactored depending on what flap you ended up at. That's right at about MLW. But on a 9900' runway in mixed snow and rain with a who knows how old braking action report I would be spending the time to plug in the numbers and do the math and interpolations to make sure I got the right distance. Then I'd look again. For me, that's more than 1 minute to make sure I got the right page and double check it and talk about it with my partner.

Tired in the middle of a dark dirty night and dealing with an abnormal with 20,000 lbs. of fuel? Take the time. We get paid by the minute.

[daedalusx]

You're not calling for FLAP 20+ on the STAR. And I'll ask again, how does CJT get their performance data for an abnormal landing? QRH? ACARS?

It's been a few years but it was common when landing 08L off that STAR for ATC to slam dunk you high to final when traffic was light. A lot of pilots would start selecting flaps abeam the airport on the downwind in anticipation. Maybe it's changed in the last few years ? It does seem a bit early for Flap 20 but the point being they might have been there earlier than usual. It seemed to happen around the time they were given the base turn.

I'm not sure if Cargojet uses ACARS for landing data ? If using a QRH (book or ipad) it wouldn't take that long to get a distance, a minute or so at most. But I get it it's the middle of the night and you're tired, so maybe an extra minute or so then.

Sorry man, but I think you're reaching.

I've flown big jets on that arrival plenty and NEVER would do that.
You're not fastest going down on an unrestricted descent by slowing for flap 25. Placarded speed for that is 180 KIAS. Speedbrakes and maybe flap 5 at most but usually not until base.

So they are less than Flap 20 when they get the problem and the checklist has some extra steps.

In the QRH perf inflite you have like 30 or 40 abnormal landing distance charts to flip through to find the right one.

The base distance when you enter the chart for leading edge asymmetry with good reported braking action is ~ 8500' or 7800' unfactored depending on what flap you ended up at. That's right at about MLW. But on a 9900' runway in mixed snow and rain with a who knows how old braking action report I would be spending the time to plug in the numbers and do the math and interpolations to make sure I got the right distance. Then I'd look again. For me, that's more than 1 minute to make sure I got the right page and double check it and talk about it with my partner.

Tired in the middle of a dark dirty night and dealing with an abnormal with 20,000 lbs. of fuel? Take the time. We get paid by the minute.

Absolutely, 100%.
Take the time.
Even if the NNC is completed.
Do a nice slow briefing, try to anticipate what else could possibly happen (issues with antiskid, issues with ground switch, spoilers not coming up after td, missed approach procedure if you float or come in unstable, etc etc)
Unless you're on fire or running out of fuel, slow down.
And no, you're not calling FLAPS 20 downwind coming over YVR for the 8s whatever the time of day it is or if you're cleared unrestricted ... jeez

[airway]

One suggestion I would have for ARFF is to try to make radio calls shorter and more to the point. Think about if it is really necessary to make a call at all. Plan your call out before you hit the transmit button.
In this incident the accident aircraft had to wait at least once to make an important radio call because ARFF was making a call.

For example:

RED 14: "North ground this is RED 14 + 1"
GROUND: "RED 14 + 1 go ahead"
RED 14: "North ground this is RED 14 + 1, we are just on the main apron, on Juliet Alpha heading towards our standby location which will be Ummm Quebec, Ahhh, we can enter the runway at any location which is reasonable though, Ahhhh, but requesting Quebec, RED 14+1"

This can all be said with:

RED 14: "North ground, RED 14 + 1 on Juliet Alpha, request standby location on Quebec"

No need for the first 2 transmissions in this situation.



.

[digits_]

One suggestion I would have for ARFF is to try to make radio calls shorter and more to the point. Think about if it is really necessary to make a call at all. Plan your call out before you hit the transmit button.
In this incident the accident aircraft had to wait at least once to make an important radio call because ARFF was making a call.

For example:

RED 14: "North ground this is RED 14 + 1"
GROUND: "RED 14 + 1 go ahead"
RED 14: "North ground this is RED 14 + 1, we are just on the main apron, on Juliet Alpha heading towards our standby location which will be Ummm Quebec, Ahhh, we can enter the runway at any location which is reasonable though, Ahhhh, but requesting Quebec, RED 14+1"

This can all be said with:

RED 14: "North ground, RED 14 + 1 on Juliet Alpha, request standby location on Quebec"

No need for the first 2 transmissions in this situation.



.

Good luck with that. How many real calls do these guys attend? How concise and to the point are the radio calls of a new PPL student during his first 10 flights?

I think the easiest solution is to not have the plane switch away from ATC during the emergency. But no solution will be perfect.

[airway]

One suggestion I would have for ARFF is to try to make radio calls shorter and more to the point. Think about if it is really necessary to make a call at all. Plan your call out before you hit the transmit button.
In this incident the accident aircraft had to wait at least once to make an important radio call because ARFF was making a call.

For example:

RED 14: "North ground this is RED 14 + 1"
GROUND: "RED 14 + 1 go ahead"
RED 14: "North ground this is RED 14 + 1, we are just on the main apron, on Juliet Alpha heading towards our standby location which will be Ummm Quebec, Ahhh, we can enter the runway at any location which is reasonable though, Ahhhh, but requesting Quebec, RED 14+1"

This can all be said with:

RED 14: "North ground, RED 14 + 1 on Juliet Alpha, request standby location on Quebec"

No need for the first 2 transmissions in this situation.



.

Good luck with that. How many real calls do these guys attend? How concise and to the point are the radio calls of a new PPL student during his first 10 flights?

I think the easiest solution is to not have the plane switch away from ATC during the emergency. But no solution will be perfect.

The 767 stayed on 08L tower frequency until a few minutes after they were stopped, then they were switched to a discrete frequency by tower so they could talk directly to ARFF. The problem in YVR is usually the 08L tower controller is also the North Ground controller and everyone can hear all the transmissions anyone makes. That's why the 767 had to wait until the ground communication stopped. At night sometimes there is 1 controller doing north and/or south ground, tower on 1 runway, and clearance.






.