Cosmo,

I do agree with you with the mindset of the issuse but what I was trying to convey was that had the AP been engaged during the approach, the PF most probably would not have become so overloaded that he forgot to put the thrust Levers into TOGA...

sharpclassic,
I understand the point, but it's not really addressing the underlying issue - why he could become so overloaded from performing a standard maneuver.

One thing is not "getting in". Of course the approach can be conducted at a greater ease with auto pilot engaged, better overview and capacity to look out, then disconnect and land. But from the start of the go-around all this is history, and the point is now only to get the aircraft away from the ground. It's a new maneuver, a new task and the fact that he didn't land but initiated the go around should have been a reset point. And then a maneuver that is quite simple - pitch, throttle, breathe and take it easy. Only those two things are really important. As long as both engines are operating it will get away from the ground even with landing flaps and the gear down.

Then how could the go-around get so messed up. That is the issue to uncover, rather than to blame it on underuse of automation.

Was it training? fatigue? or... what was the reason? Or perhaps it was none of those and simply the one time in the career where the curtain goes down on an otherwise skilled pilot - Then how come the co-pilot didn't react? Training? fatigue? the same questions over again.. plus some new ones, like culture? hierarchy? etc.

If the reason is found it can, perhaps, be corrected.

I stand corrected. Not flown the 737 but even the 747-100 did auto go-arounds so assumed that more modern aircraft would too.

Back in my box now.

The 737 only does auto go-arounds on condition that you are performing a dual channel approach, on hitting TO/GA one auto pilot is knocked off and the remaining autopilot takes over the go around, on single channel approaches when you hit TO/GA the autopilot flying the approach at the time is knocked off and the PF has to take over the go around....manually that is:}

Actually both AP remains engaged in pitch mode until missed approach altitude is captured or another roll mode is selected. Both provided that sufficient time has passed for the stabilizer has returned to normal trim, otherwise - oh god - the pilot has to fly the aircraft manually! A dangerous beast such a 737, not suitable for modern day Commercial Air Transport, since "Todays aircraft are NOT designed for manual flying as we used to know it." :E

Thanks for pointing out my mistake cosmo,:\ it's true both AP stay engaged in pitch. But selecting another roll mode jus gives you singleAP roll mode operation the pitch still remains in TO/GA dual AP control, everything else you pointed out is dead accurate

BEA Report
 

I don't know if it has been posted before but here is the BEA report.
Some interesting comments but I could not find an English version.

So, if you are expected to use automatics all the time they are available, the only time you hand fly is when something is broken..

9,000 hours watching George fly a fully functioning airplane, and a total of five hours hand flying a cripple in that period.. What could go wroing?

GB

I hand fly every approach in the wigglepig, @ 300-1 we do a pma, not because I need to but our SOP's require it. 300-2 is easy peasy, PMA's are done down to 200-R1200. Don't forget you are pilots people, not monkeys as HR/Accounting/Flt Ops want you to be. Fly the GD Airplane


Sharpclassic - almost 2 miles vis and 300' ceiling, and you want to mandate autopilot approaches?

Obviously I'm not a fan of the 'new' way of thinking that mandates every increasing use of the automation.

Rigpiggy, we all agree manual flying skills are being eroded by automation, and regular practice is important. However, doing a manual approach in marginal CAT1 conditions is probably not the best time to do it.

At the very least a go around costs you 15mins of fuel and time, which in a typical short haul schedule is going to compress the rest of your day and probably delay several hundred passengers.

Explaining to them that you fancied a bit of manual flying as the reason for their delay would be met with a certain amount of irritation. Our Airfrance colleague made a poor decision and showed that he had given even less thought to his go around than he had to his decision to fly a manual approach.

The moral: you can hand-fly a Cat I, but you better not screw it up......

Just had a quick scan at the beginning of the report in french.
14,000 hrs, TRI on type, 3,800 on type. Elected to hand fly with manual thrust . Fully configured and stable at 2,000' agl. Immediately following an A330 which got in. However a previous AF had just missed. A new METAR showed 1000 m vis, scattered 1500', broken 230'. Decided at 200', put power on to FLX/MCT and pitch to 5deg at 150' agl. Reduced flap to config3, engaged AP1 and called for gear up. Watched aircraft level off and start descending. Disconnected AP and pull hard ish. Missed Go around instructions from ATC. Master warnings audible ove the R/T link. Min height of 76' agl.

Seems like a lot of dubious decisions and poor airmanship here. I wonder who was head down and looking in, and head up and looking out. 50 foot height difference between decision and commencing the go around. Not pushing the thrust levers all the way to TOGA. Not reading the FMA and seeing LAND gren still illumintated. Not pitching to at least 10deg nose up one engine inop, or preferably 15 with both engines working. Going for the autopilot before climbing safely away from the ground.

Sometime ago we had a series of events in the poor vis sections of our sim training. The instructor would fail the GO AROUND mode after inducing a low level go around due to another failure or winding the vis/cloudbase right down. Even if you selected TOGA, the FMS was stuck in LAND mode. You had to disconnect the autopilot, ensure TOGA thrust and hand fly away from the runway. Any crew that didn't would find the box on legs heading downhill, with the thrust rolling back. Not pretty.

Lessons for us all here I think.

Sounds familiar... Sometimes longer de-briefing.....

solitaire
 

Just to be clear there's no need to get back in your box, you're not far wrong.

If you opt to fly a coupled approach in a genuinely modern Boeing product such as the rip roaring up to date 747-400 :ooh: or the 777 you will, by default, be flying a multi channel approach. If you hit the TOGA switches with the Autopilot engaged you will do an automatic go-around.If TOGA is pressed the power should run up to Go around power, the aircraft will pitch to fly a 2000' ROC and the aircraft will hold the current ground track..the autopilot certainly shouldn't disengage.

My kind of instructor.

translated notes from the report
 

The leading idea for the captain was to proceed with a 'smooth' go around by slowly applying thrust. By doing so he inadvertently stopped to the second detent he met which is MCT but does not trigger the all go around logic.

Following an earlier event of dubious go around maneuver, Air France had decided :

• To practice a go around exercise on all engines with low level capture.
• To modify how teaching go around maneuvers to ensure that the reading of FMA in the initial phase of the procedure does not occur at the expense of the control of pitch and thrust, the monitoring of the primary parameters remains a priority in all circumstances.

Following the incident of September 2009, the company has reminded all pilots :

• The thrust levers must be forwarded to the TOGA mechanical stop.
• Any alteration of the procedure in order to obtain a smoother go around must be prohibited.
• Pitch control is paramount.
• AP should be engaged only after the trajectory has been stabilized and the FMA is coherent.
• A return to basic principles of flight and appropriate training are necessary.

It seems to me that while its already been discussed that when flying an approach to weather minimums crews should have briefed both the approach AND potential missed, and that the PF should at least have mentally gone over the missed approach (go around) procedures--memory items et al, but the idea that you disengage the AP at minimums (or near) or that you engage it at the start of the missed IMHO is, at best, poor practice, and I would hope, contrary to SOP's. It was for us. You want to fly the approach manually-- disengage at the IAP not the DH(A), and if you miss, you stay in the mode of flight you were in during the approach until you have a stabilized go-around established. I'm just sayin...

Going into Atlanta one day in turbulent IFR conditions at 300 ft my MD80 autopilot shut down. We were still in the clouds so had to do what we could to salvage the approach. We did a recovery and landed but it was not what we planned. Sometimes you don't get what you expect.

So the Capts a TRI on type......... ok :hmm: That's even more worrying. He wanted a "smooth" go-around - ok, once it's going up just come back to the CLB detent if you don't want/need TOGA - the important thing is to make sure it's going UP! This aircraft got within half wingspan of touching down....

I'm an advocate of maintaining handling skills - but there's a time and a place and marginal LVP's is not the time. There are serious questions about decision making and situational awareness here.

I know we can all make mistakes but how can a TRI make such a fundamental error?

A4

Following an earlier event of dubious go around maneuver, Air France had decided :

• To practice a go around exercise on all engines with low level capture.
• To modify how teaching go around maneuvers to ensure that the reading of FMA in the initial phase of the procedure does not occur at the expense of the control of pitch and thrust, the monitoring of the primary parameters remains a priority in all circumstances.

Oh, that's really exciting news. A focus back to flying the airplane FIRST. Geez, where'd we lose focus on THAT?

First and foremost I apologize to all the senior colleagues for rearing up my head, being a just an MD-80 SFO, and being mostly a lurker on this forum. I read the posts here regularly, trying to learn from the mistakes others make but this thread is really starting to miss the point IMHO.

I truly do not understand this Bus vs Boeing discussion developing - the AF guys departed all procedures and common sense, and would have, in my opinion, have done the same, had the airplane in question been a C172. Full use of automation in marginal weather is, at least for me, common sense. I handfly every approach below 10000ft, when the airport environment is calm and the weather is fine, with fatigue within reasonable limits. Adding workload at CDG is just poor judgment. Especially since a possible GA will mean most probably not flying the published procedure, but being vectored half a dozen times within a few minutes, and handed off to several different frequencies, as we have all been in the past. Secondly, Boeing, Airbus, or my relic-class MD, a GA is a GA, throttles in the firewall / GA detent, pitch up. TOGA-tap is just...silly. GAs ate not meant for passenger comfort, but for getting out of a bad place, and fast.

Fly it automatic, fly manually, but fly it. Pitch down and 70ft clearance is not an Airbus issue - that's a very serious proficiency issue.

Just my 2c. Ready for my whipping now.:E

"Fly it automatic, fly manually, but fly it. Pitch down and 70ft clearance is not an Airbus issue - that's a very serious proficiency issue."

My kind of SFO.....right seat me anyday:ok:

Well said Max

Weren't we all taught

"FLY THE B***** AEROPLANE"

And worry about everything else later??

One of the AF particularities is that flying an approach with A/THR engaged is considered as abnormal. Why ?

After the Habsheim disaster, the confidence in the Airbus 320 was “low” among AF pilots.

Despite the fact that the A/THR was one of the most reliable system on this aircraft, some ayatollahs were spreading the idea that this automatism was dangerous and never to be used on landing.

I remember almost killing a captain after using A/THR until the flare.

And now, even if it’s not clearly written in the SOPs, it is well admitted for AF crew that AP OFF means A/THR OFF.

And it’s still difficult to have an FO performed an automatic approach with AP disconnected at MDH if the weather is poor.

Page 11
"The captain explained that a 'smooth' overshoot was linked to the low elevation gain as the missed approach altitude was limited to 2000 feet and to the risk of spatial disorientation. That is why he advanced the thrust levers slowly and believed to be in TOGA"

TOGA
 

TOGA is required for all go-arounds on the 320/330/340 series. After TOGA has been selected, to cycle the G/A flight phase, one can then go back to the CLB gate for a slower rate of climb. Woe betide those who attempt the soft go-around without TOGA being selected.

Prepared?
 

Quote from post 63

... when flying an approach to weather minimums crews should have briefed both the approach AND potential missed, and that the PF should at least have mentally gone over the missed approach (go around) procedures ..



Uhhm, shouldn't you always brief and be prepared for a missed approach, right down to the moment of touching down successfully in the touchdown zone?

Many posters seem to prepare themselves for a Go Around only when the weather is close to minimums, or are they just a little unclear in conveying their ideas?

Did the crew get dismissed?

D and F :8

FMS
 

As an ATCO I have seen the results of the FMS.
I won`t bore you with details but, essentially, the aeroplane ended up where it should not have been. I realised what was happening and, as it was not busy I gently moved the conflicting traffic out of the way...just to see what the clearance buster would actually do.
Eventually I said Kes....XXX what is your position? What is your altitude ?.........What was your clearance?

Oh sorry London, the FMS got it wrong.

Reply: Who is flying this aeroplane.........you or the FMS ?

Dave :confused:

Actually, I was taught to "FLY THE D*** AIRPLANE"

I swear, the training from one side of the pond to the other has got to be standardized (or standardised) ;)

Pilots just need to be pilots. Doing a go around is very simple. It shouldn't require training. Taking off, landing and go arounds are so basic no pilot should need any instruction in these areas once he has soloed. Pilots have been doing it just fine for 107 years without special training.

Arearadar, there is nothing wrong with allowing the FMS/FMGC to fly a missed approach. That's why we brief exactly what "the glass" is going to do and compare it to the paper version. Allowing the autos to do it whilst being carefully monitored is probably preferable to trying to handfly the go-around when a) out of practice and b) in busy airspace with a high workload.

What concerns me here is that there is regular evidence of guys just not understanding what their aircraft is about to do and they're left hanging on by the fingernails. This is simply unacceptable from today's crews in the busy airspace we fly in.

Automation complacency? Too lazy to study? Training deficiency? Take your pick.

A4

A4
 

Not hand flying the go around because of a lack of practice ? :yuk:

So in what case exactly would you hand fly a go around again given the relatively rare occurence of GA's .... which leads to the said "pilot being out of practice"

Do you mean "let the A/P do it cause you -human- can't" ?

GA is one of the critical manoeuvres that saves your life. You should be able to manually execute it whenever needed. NO reason should excuse a pilot for not being able to do it manually.

For all phases of the flight where ground proximity is a factor, it's the pilot job to be prepared and skilled enough to take over automation immediately when things don't go according to plan

GDD

I think you have mis-interpreted my reply to arearadars post. He alluded to the fact that he has seen totally mismanaged go-arounds where the aircraft heads skyward straight through it's missed approach altitude which he put down to "the FMS". The FMS cannot "think" - it will do what it's told to do.

The point I was trying to get across is that, provided the missed approach coding in the FMS has been checked against the plate and the correct MA altitude set in the FCU/MCP then "the FMS" should make the MA a straightforward event - be it AP flown or manually flown (follow the FD's). The initial part of the GA should, of course, present NO problems to ANY pilot on the line but once safely in the climb with the gear up, what's wrong with engaging the AP? What are you trying to prove by hand flying the whole thing?

In busy airspace (e.g. CDG) it's better to let the autos fly the aircraft whilst you monitor...... Why load your colleague up even more than necessary but having him have to monitor you as well as clean up, RT, c/lists etc

There is repeated evidence that GA's are screwed up because some guys out there do not understand what their aircraft is about to do (or is doing). Couple that with trying to handfly and, with some guys, you have the potential for a total breakdown of SA with workload going through the roof...... Level bust...... Flap overspeed etc. How else would you describe the A319 GA which initiated this thread? A total breakdown of technical and non-technical skills.

I'm an advocate of keeping the basic handling skills in tune but there is a time and a place. Many airlines have a pretty restrictive automation policy which can lead to the erosion of these skills...... which has the potential to be counter productive when you actually need them at short notice.

Hope that clears things a bit for you,

A4

A4,

I understood your point and globally agreed with it. I am not an anti automation pilot.

My comment was focused on your mention of "lack of practice" (amongst other valid reasons) as a reason for giving priority to automation. It's simply wrong. I'll leave it to your jugdment to figure out where this kind of logic can lead us if pushed far enough.

And I did not imply that hand flying ability was about trying to "prove" anything. It's merely a case of being skilled enough to take over automation at the right time and with confidence when the AP tend to deviate from the expected plan for whatever reason.

Indeed automation can fail.
Or, can be mis-used. You quote some of the possible reasons:

"Automation complacency? Too lazy to study? Training deficiency? Take your pick."

Why focusing on the "blame" mindset thus forgetting other possible factors such as fatigue or simple mistake due to our human nature ? These automation interfaces can involve a number of parameters each of which being a potencial source of entry error leading to an unwanted behavior of your aeroplane. So error will always bite you one day or another and it's not necessarily a case of poor training or lazyness.
That's why a pilot has to be able to rely on solid handflying skills as an ultimate life insurance

A4

Well said.

Yes, well said and Sharpclassic spot on, there's a time and place, unfortunately where I'm at, the cadets seem to be trained to switch off the automatics at 1000 FT on every approach. :ugh: