What are the problems with FDs?
 

One of the Golden Rules is "one FD off, then sw both off"
and . ."using one on, then both on"

Another SOP as determined by a TCAS RA, includes, FD off . . . .(and the rest of the RA procedure, using the v/s green band . . etc. . .)

Specifically, my question is: As there have been some `watch out` references to the FD by some well seasoned and experienced pilots on here - what lies therein?

What is the supposed dodgyness of FDs if any, apart from those I mentioned above?

Thank you for your time.

As far as I know that's only a golden rule on the Airbus, not on Boeings.. reason being (if memory serves) on the bus if the A/THR is still engaged it will be doing what the remaining FMA says it is doing, and the FD-less pilot (assuming PF) might not get the thrust s/he is expecting. But current bus drivers are more qualified to comment... it's been many years.

MY only golden FD rule is... look through the bloody things! :}

look through the bloody things . . ? . . . .to the artificial horizon beyond . . ?
(and the alt and the spd and all the rest of it?)

Airbus TCAS
 

1-) This is not named as one of "Golden Rules" but "Memory Items" in Airbus terminology.
2-)Both FDs are turned of at the same time.
3-) The reason is :
From Airbus FCTM :
- Both FDs must be disconnected once APs are disconnected:
‐ To ensure autothrust speed mode
‐ To avoid possible confusion between FD bar orders and, TCAS aural and VSI orders.

I have never needed to turn off the FDs when disengaging the AP. There's no reason you can't hand-fly the bars.

Absolutely! The thrust and attitude you set determine what the airplane does, not what some FD is telling you to do. :ok:

It's not a golden rule, but it is Airbus SOP for the reasons listed above. It's not that the FDs are a problem, it's the interaction with the autothrust.

FDs should be off whenever you aren't going to follow them. Examples include TCAS RAs, visual approaches, NPAs once you are below MDA, and hand flown departure procedures with no coding in the FMS.

The reason for

It is just a "common" rule. Clear up what you do not use. It is like the LS bars. You only select them on if you are gonna use them, no?

Besides it clears all your previous selected modes and clears the FMA.

Effing Dee's
 

Actually Nats, there really isn't a 'problem' with the FD's (or at least, that was not the message being conveyed in that Golden Rule).

From what I understand, Airbus insists that whatever you do to one FD, you do to the other. That's pretty simple and also obvious.

If you need to switch off the FD's (for example, TCAS RA), then switch BOTH FD's off, not just the ones on the PF side.

A lot of guys will tell you that FDs in general are flakey and stupid, etc.; but that's another discussion. :)

- what is wrong with raw data and hdg sel or is that forbidden in your airline?

The world wide problem, BOAC, is that automation addiction/dependency is more widespread than aircraft designers would have ever envisaged. Manufacturers and airline managements, push by way of ever-increasing SOP that raw data is the metaphorical road to disaster.

In the early days of EFIS, what became the jocular remark of "I can't fly - but I can type at 80 WPM," is now truer than ever before. Even the selection of HDG mode is seen as a 'downgrading' of automation. The combination of raw data and HDG SEL now sends shivers of alarm among even experienced pilots let alone the newbi seconds-in-command with their new MPL licence.

Natstrackalpha
In early 90s during a non precision approach there has been a fatal accident They were in open descent mode, difficult to say why. Then only no.1FD was switched off leaving the other on, as a result ATHR did not change to speed mode. FWB was a very new concept then, also the crew was inexperienced. They took time before deciding to go around. The a/c hit the ground as engines were spooling up. Since then this practice of announcing FDs off and switching both has been reccommended and followed. But it is not included in the Golden Rules.

After a few sectors with new F/O’s, F/Ds off and A/T off below RVSM. Some of our shorter sectors never make it above FL260. All briefs at the departure point and hand flying from takeoff to touchdown, without F/Ds. New F/O’s need to be reminded that F/Ds provide guidance information only. The command bars are always darting about the PFDs. The new F/O's are forever chasing the pitch and roll bars in an endless game as if trying to tame those rogue bars.

I de-select one mode at a time, pitch off, then on, then roll off then on. This starts the scan process going, then eventually both modes off. The grin on their faces appears when they discover how smooth they handle th ejet without command bars to chase about the PFD. I don't permit the use of the FPV either. If weather permits, raw data and visual approaches are accomplished too. What it comes down to is that we as trainers need to re-acquainting new F/O’sto the 4 fundamentals of flight.

Manual flying leads to manual descent planning calculations to get away from their total reliance of the FMC. But that’s a topic for another thread.

I feel the reason to switch off the FD bars is basically when the pilot is not following the FD commands. This can happen at time of TCAS or any other event.
Both FD are switched off to avoid confusion.

- the point I was trying to make was that one assumes these folk CAN fly without LNAV/VNAV (presumably in HDG SEL/VORLOC and Level change - how else?) - ie "with no coding in the FMS"?? so why not use the FDs if they feel they must. There is no need to switch them off just because the procedure is not in the FMS. I no understand!

Hi Guys
The question Nats has asked is why the FDs are switched off together and put on together. Indeed there is such procedure and it is insisted upon in Airbus. The merits of flying manually with FD etc is different topic altogether. One of the conclusions of the inquiry of the accident was, had the other FD been switched off the ATHR would have changed to speed mode. What Ahramin stated in reply no. 8 is correct. It is because of the interaction with ATHR FDs must be on or off together.

and

This is not the main reason you turn them off during a TCAS RA. I hope most pilots would not be confused if an RA was saying climb but the FD bars showed a descent due to open des mode.

The reason is, if you are in open descent and disconnect the AP and pull back for a TCAS RA no thrust will increase, your speed will bleed off and you wont climb very far. Turning off both FDs puts the autothrust into speed mode so as you pull back the thrust will increase.

Depending on the operator SOP and aircraft type, it may not be true to require both FDs off. Not talking Airbus here, OK?
e.g. an aircraft with a single FD input but dual displays - makes sense for the PF to take-off, say, in GA mode, while PNF stays on raw data. Both FDs on at safe altitude. On approach, PF uses the modes required, PNF monitors in raw data.
Reason being that a spurious FD command may be more readily noticed by the monitoring pilot detecting the flight path deviation.
Also 'looking through" the FD is too hard for this old-timer, though some seem to be able to do it. Unless following the FD commands, it goes OFF. Raw data is actually easier in some situations.

Children of the Magenta?

I have just had sim with grand pa of the magenta line...very scary indeed.

"The reason is, if you are in open descent and disconnect the AP and pull back for a TCAS RA no thrust will increase, your speed will bleed off and you wont climb very far. Turning off both FDs puts the autothrust into speed mode so as you pull back the thrust will increase"

Ding, ding, ding.

Winner here!!

Gosh i love boeing....

Children of the Magenta, indeed, `tis true.

Thank you all for your time and input. My question (as usual poorly written) was intended to see if there were many inherent bugs or bogies with FDs in general.

Specifically however, the A320 FDs - so the feedback you have given me on the A320 systems is invaluable and I thank you very much and I will have to get yet even deeper into the AOM and the autothrust system. Especially the autothrust system.

On the magenta line. Once I wondered if I could Heading Sel as BOAC suggested to get my L1011 to work on the minimum of auto, as, bascially, I did not want to f--k about with it. So, we hand flew to cruise and we were very happy. After a major conference in crz to determine that everything is tickety boo as per the plan - then we would select A/P on, and if it did what it was meant to then we would continue to TOD where we we did the same thing in reverse.

I studied, at that time the levels of automention 1,2,3, etc or a,b,c, depending on which school.

Now, with Glass and FBW - as one company puts it - "you will be assimilated" and it is true - in order to properly understand the thing, you have to be a part of it, engrossed though I am there are others who are totally and absolutely technotroids, but you have to be if you are to understand the system like one or two on here do . . . (far more than me (but I am catching up . .!)

Simply put, if I don`t get totally into it then I won`t know why it has just killed me. Not wishing to die and glad of previous experience of different levels of automation to a point where "if it aint doing what we want now then we`ll fly it another way"
downgrade,
dowgrade,
until we are hand flying the thing. Basic stick and rudder stuff Wright brothers?
Well, to a point.

L1011 was a babe to fly hands on, probably the best aircraft ever built.
The airbus will be equally easy and flyable manually, once the techtronic/human decider valve (in my human brain) has disconnected me (as a mere tie connector) from the machine.

FD's OFF
 

Base turn to final, and in a descent. Called for FD's to OFF. Captain turns his FD off, co-pilot leaves his on....Aircraft autothrust selects TOGA, and we begin accelerating into the ground.


That is why you should both turn off, or both leave on the FD's

I will share with you, my recent experience, on B777 climbing out of FL 187 Cleared for FL250, light weight aircraft, about 60 kts head wind, experiencing mod turbulence, so speed was reduced from 310 kts to turb penetration 270 kts, the airplane had a ROC as a result of speed reduction/head wind/light wieght to 6000ft/min

The airplane went into VNAV ALT Mode while passing FL190, which meant the capture phase began at FL190, the pitch mode of the aircraft (as indicated by FD) focussed on capturing FL250, the thrust lever changed to spd mode, once the speed dropped below 270 kts, the airplane wouldnt pitch down because the pitch att was stuck to capture FL250, the climb power then became less to maintain 270 kts as shown by the downward trend on speed.

The a/c had to be changed to V/S mode, to reduce pitch, straight away, and since this was anticipated, nothing went wrong, but FD went into capture phase because the FMC logic was such, this could have resulted into a complicated situation, so this is one case, where FD taking inputs from FMC in a case like this, would have managed to bring the airplane to low speed buffet before having a mode reversion to VNAV SPD again, which would have taken lot of time to actually recover, considering the airplane trim was in support of nose up while it was in VNAV ALT.

Seems like a really good reason not to be doing anything but watching what George or Otto is doing when being so dynamic; plus knowing what it should be doing and what you want it to do and knowing how to resolve the two. Seems like you did.


I've seen other scenarios where the A/P was not doing what Freddie in RHS wanted, so they started playing the piano on the MCP trying to correct it, then in desperation disconnected to sort it out. That's quite a healthy outcome; not being afraid to take charge, but it still showed a weak understanding of how to have avoided or corrected the situation via the MPC in the first place.

Therein lies a weakness in training. Cadets are taught how to use the MCP in standard scenarios like trained monkeys, and usually the FD follows their actions. They are not told how to use the MCP in depth and so can choose what to do to achieve the wanted result. Don't learn only one tune but learn all the scales.

Regarding the FD the scan has been diluted to "follow the FD." Many scenarios have been quoted where the FD was wrong or confusing, e.g. FD fly right, LOC deviation left. Scanning the ND for S.A. & W/V, ADF needles etc. resolved what was happening, but the dogmatic following of FD in such circumstances caused confusion with PF and a little consternation within me as PM that the PF was not using all the information available but was tunnelled vision on the FD.

If the automation is not doing what you want it to do, take the auto pilot out and hand fly it. Everything gets very simple, very quickly. This holds true for Airbus or Boeing.

Automation is ment to reduce workload, the moment it starts to increase it, its time to start ditching it. The problem is recognising that moment.

RAT 5
Airbus goldenrule 2.
2.Use the appropriate level of automation at all times.
•The appropriate level of automation depends upon the situation and the task. Pilot judgment prevails, including selecting manual flight.
•Understand the implication of the intended level of automation.
•Select the intended level.
•Confirm the expected aircraft reaction
Glden rule 4
4. Take action if things do not go as expected
. By PF changing the level of automation.
. By PF reverting to manual flight.
. By PNF taking action,
.Question.
Challenge.
Take-over.

　

All very fine if you have a compliant captain who is more than willing to hand over control meekly when his 240 hour support pilot grabs the controls.
But if the captain asserts his command authority backed by the Regulations, and tells the lad to bugger off and handle the radios, then the mantra of Question, Challenge, Take Over is yet another motherhood statement written by the gurus of the CRM tribe. I have yet to see any official guidance from a Regulator or any other authority on the specific method a first officer is supposed to fall back on to physically take command of the aircraft.

Centaurus
Didn't you read comments on Air India express crash at Mangalore? So many guys blamed cockpit gredient and Asian culture and suggesting that FO should have been more aggressive. This perticular accident it was so necessary that it was better to loose the licence than life. However life is rarely in black and white more often it is shades of grey. That is where real problem arises and I agree with you there. However a culture needs to be built in an Airline where captain should feel that willful violations of SOPs will be met with resistance. I got the impression that it already exists in western society. It is becoming more difficult as the level of experience of the FO is progressively brought down where the experience gradient has become steep.

Crash axe or fire extinguisher, which ever comes to hand first.

To Agg Karan:

Interesting experience that you share with us. Please research the following link, you will see that history repeats itself and it is the sharing of experiences like you just did, that may prepare others to react properly and timely when it is their time to deal with a like occurrence.

Use Google to search for
Transport Safety Board of Canada AVIATION INVESTIGATION REPORT
A04O0020

or use this direct link

Transportation Safety Board of Canada - Aviation Investigation Report A04O0020

Safe flying to you!

To people who still think that certain things happen only to Airbusses and not to Boeings, a few quotes to capture their attention ...

Quote
The autopilot pitched the aircraft up to 29.5º in an attempt to maintain the ALT CAP profile and, as a result, the airspeed decreased further. This attitude was maintained until the stick shaker activated and the flight crew took recovery action.

Other finding:
The autopilot was functioning as designed during the occurrence.
Unquote

Proper understanding and use of automation is required on all types of aircraft.

Automation
 

Thank you for the appropriate example.:)

de facto
Any comments from you on Boeing on EMIT's blog no.32 or love is blind?

You might notice that defacto was posting in reply to the peculiar FD problem on airbus aircraft. And yes, in that regard the boeing is much more simple. Doesn't matter at all if one FD is on, both, or none, in both manual and automatic flight.

The ALT CAP thing is usually not a problem with proper training, its demonstrated once during the initial typerating and then each pilot uses one of several methods that prevents it. However, it is an autoflight issue and not a FD one.

Denti,

With regards to "it" being an autoflight thing and not an FD thing - the flight director is the visual depiction of what the pilot should do to follow the same path as the autopilot would follow.
So, the FD would have seduced a slavish follower down the same primrose path to a stall. The only giveaway would have been the removal of the pitch bar from the display - and an Airbus would also have done that (re. the AF447 report, FD disappearing and re-appearing)

As far as autothrust is concerned - if you descend in FLCH and do not follow the FD pitch commands (such as during a TCAS CLIMB event), the autothrust will not increase thrust to keep speed constant, it will still indicate FLCH or THR HLD in the FMA (it "thinks" that the FD pitch mode is taking care of the speed).

A problem noted by the Canadians was that pilots routinely do not know all the possible autoflight/FD symptoms, such as diasappearance of bars, amber lines through FMA modes, etcetera, which may also differ from mod to mod status, because one "never" operates on those edges of the envelope.
However, abiding by the simple rule, if the aircraft does not do what you want it to do, grab control (physically do it yourself with stick, rudder and throttle) solves all problems.

Indeed, and it might even work on an airbus.

The ALT CAP thing is something we encounter a lot, especially in out -700s. Those are quite often light weight and climb rates in excess of 4000fpm up into the mid 300s are normal, which can preclude the reduced rate in the last 1000ft before level off since ALT CAP becomes active too early. Simple thing to do if one doesn't want to disconnect the autopilot is to push the yoke a bit, CWS will work just fine. Disconnecting the AP works just as well of course.

Yeah Denti, I recognise the situation.

Another simple rule of thumb - climb rate not more than altitude difference, i.e.4.000 ft to go, V/S no more than 4.000 ft/min, 3.000 ft to go V/S 3.0000 and so on and switch to V/S mode if climb rate of VNAV (or similar mode) does not respect that schedule.

If that's happened on an A320 series you've got something else wrong. The autothrust can't 'just' select TOGA it has to be in response to something else. So, what's the other bit of this story.....

However, abiding by the simple rule, if the aircraft does not do what you want it to do, grab control (physically do it yourself with stick, rudder and throttle) solves all problems.

Proper understanding and use of automation is required on all types of aircraft.

Wonderful. I think this debate is going round in circles and has ended up where it was some years ago prompted by other threads. 1st point is a training/experience issue, but sadly due to lack of training it might increase problems. But that should not be the case.

2nd point is also a training issue and has not been addressed by the XAA's or companies. I've been involved for many years in TQ training. The change that the customers have introduced is minimum sim training time to pass LST due to self payment versus training to a higher standard than LST. I do not know what standard they train manual flying to on the line, but the reports of incidents on here lead me to suspect not very high.

In some companies there has definitely been a philosophy to reduce the amount of stick & rudder use on the line due to cock ups. This is the attitude of "if they don't know what they are doing, or don't know how to to, then don't let them do it." IMHO this is an accident looking for a place to happen. Sometime, something will go wrong. The sharp end jockeys need to be able to sort it out. The pax expect this. I wonder if the XAA"s do? Not everything is in a QRH. Often some lateral out of the box thinking is required to save the day. The sort of thing a 'cabin attendant' could not sort out....if you get my drift. The sort of thing we should be educated in every 6 months, not the routine boring trained monkey stuff. After 30 years if I can't handle a V1 cut I should be driving the crew bus, if there was one. I've done it 60 times in a sim. The OPC/LPC is in need of a reality check. You know you are going to get a SE G/A from a SE ILS. There is no surprise. You know you are going to get an RTO. Surely there needs to be some reality/surprise introduced into the prof' checks. The 6 month merry-go-round is not such a sure method of maintaining standards. It's a tick in the box and I've seen some lax application of pass criteria in my time.

The more the companies decide that we need to be just button pushers the more they will train to that idea. Then they will write such comprehensive SOP's as to make us robotic trained monkeys. Then they can pay us peanuts to sit in the sun all day. It's a great conspiracy, and it's happening in an airline near you.