SFO Incident

[blues deville]

Good. Let the open discussion and exchange of information continue.

My Airbus experience was short lived and some time ago. But I recall some issues with combining an RNAV waypoint approach to an ILS. Which is probably something Airbus didn't think would be necessary much like a BC approach mode. 

My current aircraft can be set up to fly the Bridge RNAV approach (LNAV/VNAV to landing or transistion to ILS) with a couple of last minute mode selections by either PF or PM depending on autopilot status. The NAV display will always show the active extended runway centre line for reference. 

Another issue with this incident could be related to recency. I know from working in a seniority based scheduling system that I could only hold the better flights during holidays or summertime when the senior crowd was off duty for awhile. Longer more  productive days or great layovers compared to an endless stream of shorter flights. However, these destinations might be places that I'd have flown to a only few times where even the ground agent would say "who are you?" 

This AC crew had a combined flight time of 30,000 hours but when it comes to scheduling they may not have flown to SFO very often or perhaps had never flown this Bridge approach before that night in July.

[Guest]

The 2 SFO topics have been merged.

[blues deville]

2 hours ago, Marshall said:

The 2 SFO topics have been merged.

And the reason for doing that is? Somehow not a great idea. 

[Guest]

blues you do have a point, I have moved most if not all from this topic back to one entitled SFO

 

[blues deville]

1 hour ago, Marshall said:

blues you do have a point, I have moved most if not all from this topic back to one entitled SFO

 

Thanks.

[blues deville]

9 hours ago, Rich Pulman said:

Airbus does indeed have a BC approach mode. There is also a way to display the extended runway centreline. BUT, it does depend on the options installed on the aircraft. Not all operators have the same software. And it shouldn't be a big deal transitioning from an RVAV approach onto an ILS. The FMS would be set up for an ILS approach with a transition/VIA for the RNAV arrival procedure. With the approach push button armed, the system would automatically couple to the ILS when its signal is intersected.

Okay so that sounds very much like what you'd do on my aircraft. 

I thought some of the earlier Airbus (320) models did not have a backcourse mode but there was a workaround procedure. Non?

[Zan Vetter]

AC doesn't permit BC approaches on the airbus. Only place that was a factor was YOW25, where we had to do the NDB instead. Regardless, SFO28R is not a backcourse it is an ILS, which was off the air this night if I'm not mistaken.

[UpperDeck]

On 8/5/2017 at 10:37 PM, Vsplat said:

UD,  I offered my comments in good faith.  Your response is, well, unfortunate. 

You clearly have no idea what I think, and your comments reflect zero insight into the function of the crew of 759. 

You seem to have a strong need to attack, filter and reframe comments into an invalid context. 

Finally, you seem to want the freedom to slander Air Canada employees and spread innuendo at will, up to and including your implication of violation of federal statues and destruction of evidence.  Your reply when reminded of the law, shown it in fact is that you'll take your chances and want to see who else has been convicted. 

 

If you want to act like a troll you will be disregarded as one.

Vs

 

Apologies since the thread appears to now be back on track and I am disinclined to interrupt. I waited awhile to respond to the above post which was the last before the thread was " reviewed". In my opinion, suggestions of defamation and prosecution are thinly veiled attempts to inhibit discussion and should be avoided.

It is clear that no Air Canada employee was " slandered" ( defamed) nor was there ever a statement that any person committed an offense. Let us be clear on that.

To the extent that anyone requires an advocate to present their "defense", those advocates whether self-appointed or otherwise should be guided by the rules of civility required of ALL advocates. Hurling epithets such as "troll" is inconsistent with that standard as are accusations of "slander" and " innuendo".

Enough on that subject.

I previously asked but didn't get a response; is the recording time on the 320 CVR 30 minutes or 120? If only 30, I think it reasonable to conclude it would have been overwritten.

I understand that this occurrence would NOT be within the definition of a reportable " incident" in the US. However, based upon an earlier post, it appears that it would be reportable based on Canadian requirements but would that report be " forthwith" to the Canadian authorities and then " in due course" to the FAA? US counsel advises that an " incident" as defined must be reported to the NTSB but US pilots are advised to NOT  report unless it is clear there was an " incident" as defined. Overflying a taxiway does not appear to be within that definition.

[conehead]

Upper Deck;  in answer your question, I believe the recording time for the CVR used in the A320 is 120 minutes. I'm not 100% certain on that, as it's been a few years since I worked on that type of aircraft.

[J.O.]

Some of the early model A320s were delivered with a 30 minute CVR, not sure on the status of the AC fleet.

[mo32a]

4 hours ago, Zan Vetter said:

AC doesn't permit BC approaches on the airbus. Only place that was a factor was YOW25, where we had to do the NDB instead. Regardless, SFO28R is not a backcourse it is an ILS, which was off the air this night if I'm not mistaken.

The ILS for 28R was functioning when this occurred.

[Vsplat]

19 hours ago, Rich Pulman said:

The FMS would be set up for an ILS approach with a transition/VIA for the RNAV arrival procedure. With the approach push button armed, the system would automatically couple to the ILS when its signal is intersected.

Rich, just confirming something. AFAIK the FMS Bridge Visual is a full RNAV approach, flown in FINAL APP with its own minima on PERF. I haven't (yet) seen anyone string the ILS onto the back of the RNAV to facilitate an automatic transition from managed/managed RNAV to fully coupled.

At present, the ILS can be hard tuned for a manual transition close in, but selecting LS display gets you a warning on APP guidance. 

Is there a specific technique to doing what you describe?

Thanks

Vs

[blues deville]

More about the loss of CVR audio. 

http://www.govtech.com/data/Critical-Evidence-from-SFO-Near-Collison-Erased-Due-to-Outdated-Tech.html?utm_source=skies-daily-news-news-from-the-web&utm_campaign=skies-daily-news&utm_medium=email&utm_term=news-news-from-the-web&utm_content=V1

[Vsplat]

Thanks Rich.  Rats.  I read your post and thought, man that would be an elegant solution to these hybrid situations..

If it's not a thing, maybe it should be.

Cheers

Vs

[ckl]

I used to set up the ILS 28R in the secondary flight plan and get the pilot not flying to activate it when I no longer needed the rnav approach. In fact I always used the secondary flight plan and would often set it up with an alternative approach or runway. Used it more than once in YYZ !

[Vsplat]

I think the ILS is there to back up the side step so you don't crowd 28L traffic.  RNAV approaches are great, but on a non GPS aircraft, the amount of error is enough that we lost the ability to identify the FAF solely based on FMS position.  So the centreline displayed and tracked might be a long way off from the runway itself.

In terms of effort, if the ILS is hard tuned, bringing up the ILS at the end of the LNAV segment is a one button 'LS" selection, but you lose the LNAV/VNAV deviation display when you do that and get some other messages.  The secondary flight plan option would work, but it's a heads down task at at point when I would prefer to have both of us less loaded.  I'd have to look to see if the missed approach is the same for the visual and ILS as well.

Interesting options though.

Vs

[UpperDeck]

On 8/10/2017 at 0:28 PM, blues deville said:

More about the loss of CVR audio. 

http://www.govtech.com/data/Critical-Evidence-from-SFO-Near-Collison-Erased-Due-to-Outdated-Tech.html?utm_source=skies-daily-news-news-from-the-web&utm_campaign=skies-daily-news&utm_medium=email&utm_term=news-news-from-the-web&utm_content=V1

I did a little--- a VERY little---- study of the statutes and regulations, a fairly complex maze. The controllers are part of the FAA as all know. An aircraft operator reports to the NTSB forthwith upon the happening of any incident or accident and must preserve any evidence including any CVR.

The situation at SFO was NOT a reportable occurrence requiring AC or any employee to notify the NTSB or preserve the CRV. I am merely guessing that either the FAA reported the matter to the NTSB some time later or AC did so voluntarily.

The Canadian statutes and regs have a broader definition of a reportable occurrence but the time within which to report the "event" is not "forthwith" and certainly not within a time that would have prevented an overwrite of the CVR.

In Canada, the CVR is " privileged". 

There is a CADORS for the Westjet low approach into St.Maarten and go round that was discussed on this forum. That " incident" was reported.

Given the over-riding importance of using any negative event as a " learning moment", perhaps the discussion should be focused on 1) the definition of an "incident".....is there too much emphasis on an actual accident and damage or injury?; and, 2) the limited use of the CVR to assist in the understanding of processes that may lead to negative results whether that be undue deference in the cockpit or mutual reinforcement despite objective data.

Aviators and the general public will never know what was being said in that cockpit as the aircraft approached SFO. Collectively, we are left with the recall of the pilots themselves and memories are fragile. Personally, I believe that a considerable benefit would be generated were procedures in place to preserve the CVR whenever anything occurred in flight that had some potential to result in loss. I know....I know....take off would satisfy that criterion. Obviously, " anything etc." would have to be defined.

[Vsplat]

Rich, WRT RNP, the answer is, 'it depends'.  There are a number of approaches in North America that can be flown with DME/DME/IRS, others, of course, that require GPS.

Of interest, even an RNP 0.3 approach would have had enough error  to span the difference between the runway and taxiway.  That said, this was a visual approach so that aspect should not normally have been a factor.  I am sure the NTSB will be looking at all of that.

Vs

[mo32a]

17 hours ago, UpperDeck said:

 

The situation at SFO was NOT a reportable occurrence requiring AC or any employee to notify the NTSB or preserve the CRV. I am merely guessing that either the FAA reported the matter to the NTSB some time later or AC did so voluntarily.

 

I disagree. A go around because somebody was slow clearing or spacing was compromised is one thing. If in this case the aircraft had continued to land there would have been a very serious incident. Any controller would report this to his supervisor and a report would be generated.

[UpperDeck]

6 hours ago, mo32a said:

I disagree. A go around because somebody was slow clearing or spacing was compromised is one thing. If in this case the aircraft had continued to land there would have been a very serious incident. Any controller would report this to his supervisor and a report would be generated.

I may very well be wrong but what I read is that the operator is required to report to the NTSB an "incident" forthwith. In the definition of " incident", there is specific reference to actually landing on a taxiway. That would exclude an approach that did NOT result in a landing. That is why I suggested that AC did not have an obligation to report.

I did not read any reference to any obligation upon a controller to report to the NTSB. I would expect the controller to advise a supervisor and it would proceed within the FAA. I understand that a controller can report to a 3rd party for self-protection.

My point was simply that the pilots and AC were not required to report this event to the NTSB.

[UpperDeck]

https://www.ecfr.gov/cgi-bin/text-idx?SID=36418957aeb9385b90ae9a44f4a8d1cd&mc=true&node=pt49.7.830&rgn=div5#se49.7.830_15

I think this link will work. It is the portion of the Federal Code regarding reporting of accidents and incidents. Again, there is no reference to any obligation of a controller to report to the NTSB but regardless, if the event being discussed did not mandate a report by the operator then I doubt there would be a statutory obligation upon a controller to report.

[UpperDeck]

6 hours ago, mo32a said:

I disagree. A go around because somebody was slow clearing or spacing was compromised is one thing. If in this case the aircraft had continued to land there would have been a very serious incident. Any controller would report this to his supervisor and a report would be generated.

mo32a..... 

We know that the NTSB is conducting an investigation. It is my understanding that NTSB field operations have access to a lot of information regarding airport operations. It is possible that the NTSB " self-initiated" the investigation. It is possible that it was reported voluntarily by the operator or an investigation was requested by the FAA.

The relevance of a mandatory report is that it is made FORTHWITH and the CVR must be preserved together with all other potential evidence.

[DEFCON]

Any one, or more of the four aircraft waiting to go could have initiated the reporting process through their own operations centers?

I don't have the rules handy, but my gut tells me that common sense says the incident was serious enough to be reported and apparently someone else did too. It's only an opinion, but to suggest an incident like this is just a business as usual excursion from the norm because it doesn't exactly fit with the list of those incidents that must be reported is disingenuous at best.

The discussion above regarding the various techniques employed to get the aircraft to the threshold tells me there's a big problem with automation dependency. Just think; had a single $300 round faced cross hair ILS indicator been installed, the pilots would have known PRECISELY where the aircraft was in relation to the runway landing zone throughout the entire fancy FMS directed RNAV piggybacked GPS guided blah blah blah VFR approach and thereby avoided the entire event.

.

 

 

 

[Vsplat]

Much of the information that would allow an objective look at this event is lost or otherwise unavailable to us.  That said, it seems to be that perception is a common element in all of the unanswered questions.  Why did the crew believe their approach was correct for so long? When did the ground traffic perceive the threat?  What did the tower controller see and when?  What level of threat was perceived at the time of the event, when was the current level of concern reached and how? (this last question speaks directly to the timing of the report)

We can debate the various technical limitations of the navigation systems, but this was supposed to be a visual approach.  While it is easy to say, it is helpful to me to keep in mind that, at one time, ALL approaches were visual, until we as an industry learned that visual cues were sometimes absent,  insufficient or misleading.  Misleading visual cues have been biting us a lot lately.

Bottom line, this crew perceived their approach as safe for quite a while.  Why was that, and what ultimately changed their mind?  Was it the call from the ground traffic, or did they see something on their own?  Understanding what they saw and processed is going to be the key to mitigating a future occurrence.

 

FWIW

Vs

[JL]

Similar circumstances in an event here in Canada...

https://www.skybrary.aero/index.php/B733_/_DH8D,_Fort_McMurray_Canada,_2014