Tuesday, January 26, 2010

When Check Pilots Get Bored

Some weird aviation stories have been coming out of India lately. Six years ago if you had a commercial licence and the right to work in India, you had a jet job. Airlines were expanding so quickly that just maybe not everyone got the training or screening they might have in other times.

The first story I see is of a check pilot surreptitiously disabling the aircraft in which he was riding, in order to see what the pilot would do. The check pilot is an observer, a passenger in the aircraft. He shouldn't be so much as turning on a light switch, let alone pulling important circuit breakers, literally behind the pilot's back.

So the first officer is flying the airplane when all of a sudden the autopilot cuts out--I assume Boeing is clever enough to have designed the autopilot such that the disconnect warble is on a different circuit breaker and is still audible, but that I don't know for sure. At the same time, the flight director disappears, so the pilots has to control the airplane himself and decide where it should go. The pilots do notice the problem and the commander takes control, but with an excessive sink rate. The aircraft was recovered and landed without damage.

Oddly to me, the analysis in the newspaper piece appears to claim that the lacl of an EGPWS put the airplane in this situation. I've watched numerous people land 737s by hand, and the only time I heard the GPWS alert was when the runway was in a steep valley and the airplane had to pass over a cliff to remain on the approach path. The pilots didn't react to the warning except to silence it, because they were in control of the airplane and were expecting that warning on that approach. Perhaps in the Jet Airways case a functional EGPWS would have pointed out the problem with the approach a little sooner, but according to the article they were on the ILS. Surely the fly up indication on the integrated HSI would have been given that information.

It's true that ICAO standards call for such aircraft to have a functioning EGPWS, and that they probably would have prevented some accidents, but when the ILS is working properly, staying on slope will cover that territory for you.

I suppose stupid check pilot stunts follow from flight instructor practices in much simpler aircraft. An instructor, in visual conditions, might turn off the master, pull the mixture or shut off the fuel to one engine, or remove a light bulb from a gear position indicator. It's good to verify that a student's reaction is the same as that rehearsed when the failure is simulated or simply described. I remember an instructor of mine who shut off the fuel to one engine in cruise while I was learning. When the engine faltered I ran a quick cause check, and discovered the fuel valve position. I immediately corrected it, and the engine restarted--I'm not sure it had entirely stopped--so I returned to normal flight. The flight instructor was going "but ... no... wait ..."

"What?" I asked, wondering what I had missed. "The fuel valve was switched off." It honestly didn't occur to me until that moment that I was supposed to leave it off and feather the propeller as if I didn't know what had caused it. (I've done that sort of thing more than once. If an instructor fakes a failure by a means other than briefed, I treat the fake, not the failure).

I've also been with an instructor in an airplane with electrically operated landing gear who routinely pulled the circuit breaker after the gear was confirmed down, to guard against accidental retraction. I didn't really like it, but I can see the argument for it. I know of a case in that same type where there was a n accident because a student raised the gear on the runway. And I know of another where there was a problem getting the gear down because the gear CB had been routinely pulled and wouldn't stay set. Circuit breakers are not switches.

Here's a digest article describing other Indian airline mishaps lately. I hope it is just a combination of statistical anomaly and sudden focus of the national news media on everything aviation related, rather than a real symptom of problems. Right after I wrote my next sentence as "the same goes for the number of times American Airlines has been in the news lately," I read Captain Dave's blog on the Jamaica runway overrun. There's so much that can go wrong in that moment in the dark, there is not room for stupid pilot tricks. As Dave says, Fate is out hunting you. No need to go looking for it.


Anonymous said...

Re: circuit breakers are not switches.

Well, Call me a smartass, but sometimes they can be.

As for check pilots failing items without a proper briefing first - this should be a violation of some sort. I think you're right about the inadvertant carry-over from instruction on simpler craft.

nec Timide said...

Speaking of instructor behaviour in light aircraft... I flew once as a student pilot with an instructor in the right seat, and one as a passenger in the back. The one in the back started messing arround with the trim wheel with his bare foot. After the third time, and my third complaint he said "you have runnaway elevator trim, what are you going to do about it?" This even though this PA28 didn't have electric trim. I elected to disable the runnaway by forceful application of my pencil.

A Squared said...

well, not to argue for the propriety of a noncrewmember pulling breakers in flight (that's a bad idea) but having spent a significant portion of my airline career flying aircraft which didn't have GPWS (with or without the "E") flight directors or autopilots, I'm just not seeing the horrible calamity of hand flying from raw data. In fact, now that I am flying an airplane with all the magic, I often turn it all off (not the EGPWS) because, well, I like flying airplanes.

Granted, pulling breakers without the flightcrew's knowledge is bad, but it seems that the times of India is stretching to make this something it's not. As far as the "excessive sink rate" I'd have to have a bit more information on how much in excess of 1000fpm it was and how it came to be before I'd call it a dangerous situation. Was it 1100 FPM or 3000? Did the pilot intentionally, momentarily, increase the descent rate because he was a little behind on his GS intercept, or were they actually plunging out of control?

fche said...

I've also been with an instructor in an airplane with electrically operated landing gear who routinely pulled the circuit breaker after the gear was confirmed down, to guard against accidental retraction.

What was this person's theory as to overshoots?

jinksto said...

I dunno. If I'm a passenger I want every one of the blinky lights doing the blinky thing every time.

My pilots training and experiences tell me that you can fly just fine with an ASI and a window but that doesn't change my wanting all of the gizmo's working when I'm riding in the back with a hundred of my bestest friends.

That said, I agree that the article gets a little dramatic. However, I suspect some level of cultural difference here. I'm not an anthropologist by any means but I work with Indians daily in a technical field and have noticed that this oversimplification (which adds an unintended dramatic effect) can happen even when the author has a very detailed knowledge of the subject. I suspect that it seems worse when the author... a journalist... doesn't have training in the field.

Wayne Farmer said...

Re: "I know of a case in that same type where there was an accident because a student raised the gear on the runway."

That accident also happens in one of my favorite films, the 1951 "No Highway in the Sky". James Stewart plays an aeronautical engineer who is convinced that the British plane he's traveling on as a passenger is going to break apart and crash. You can watch the film for free on Hulu.com; the whole film is good, but for just the particular indicent, see 00:54:00 through 1:03:00.

nec Timide said...

You can watch the film for free on Hulu.com;

Only if your IP address resolves to the US though.

A Squared said...

I dunno. If I'm a passenger I want every one of the blinky lights doing the blinky thing every time.

SO what if they're not doing the blinky thing? Suppose that the #1 radio altimeter took a dump, and took with it the autopilot, flight director and EGPW? (which seems from the article a plausible scenario)

Who would you rather have up front?

Crew A, who always uses the magic, all the time, and has not hand flown a approach from raw data (which if you don't know is a very different perishable skill set than matching the pitch bars up to the command bars of the flight director) since their respective instrument rating checkrides.


Crew B, who occasionally turn off the magic, just to make sure that they still can fly raw data?

Aviatrix said...

Anonymous 00:24, yes, and if there is a need for them to be that is what should be installed.

nec Timide, I suffered jammed elevator trim on a flight test once, right after my first power reduction. I had to fly untrimmed for a few minutes while I got clearances and tracking sorted out, then discovered I had accidentally dropped a pencil into it.

A Squared,
1) I too was puzzled by the extreme emergency suggested in the article by the removal of the EGPWS. This blog entry actually started off a lot more along the lines of "what is wrong with people that they can't land an airplane without a piece of equipment I've never had?" but as the article wasn't entire clear as to what all had been taken out -- EGPWS, autopilot, flight director, anything else? -- I toned it down.
2) Deliberately turning off the magic in a well-briefed scenario is quite different from a non-crew member doing it behind your back. Depending on the operation, in Canada it would likely still be illegal. Ops manuals, which once approved by Transport Canada are legally binding, generally forbid disabling safeguards for practice during passenger flights.

fche, Presumably the CB-pulling instructor reset the relevant CB in the case of an overshoot. I don't remember if I did any overshoots with that instructor.

A Squared said...


I don't think we disagree here. My diatribe was directed at the article, not at your blog entry. As a semantic point, I would say that turning off a radalt or GPWS would be "disabling safeguards" while electing to not use a hand-fly without a flight director would not. I do the latter, but not the former. Well, actually we do on occasion inhibit the EGPWS, as it doesn't play well with operations from frozen lakes and such, but in normal, airport to airport operations I wouldn't do that. FWIW, my ops manuals encourage the use of the automation, but are very careful not to mandate it.

I think that, perhaps like you, I'm bit taken aback by the attitude that operations without full automation are dangerous. A few year back, on an aviation forum which I frequent, someone opined (and found a fair amount of agreement) that anyone who would hand fly an ILS vs letting the autopilot do it was a dangerous cowboy. It kind of rubbed me the wrong way, as at the time I still at the airline with no autopilots, so I had no choice but to be a "dangerous cowboy" on every ILS.

On a not entirely unrelated note, on another aviation forum, a more recent discussion arose about how to deal with AC giving you an immediate and unexpected hold at your current position. There was a lot of discussion on which FMS function to use, an some commented that it might take a bit to recall how to access that function on their FMS as they don't often use it, and others said that their FMS's doesn't have a function that allowed a hold at present position, and might might require input of an artificial point along the airway, and one went as far as to say that it might be 10 miles or so before they would be able to comply....Jesus H Christ folks! ATC is having a meltdown and they need you to stop, right now! If you can't get your FMS to do it for you in about 2 seconds, hit the HDG button on your autopilot and turn the heading bug to the right until you've turned 180 degrees, or turn George off and turn the airplane by hand. I think that with some the reliance on the FMS has become so complete, that they've lost sight of the fact that it's possible to fly without it.

nec Timide said...

...then discovered I had accidentally dropped a pencil into it.

Small world. But, in case I wasn't clear enough, I solved my problem by applying the pencil to the foot, not the trim.

Verification word: menduck?

gmc said...

Is hand flying becoming a lost art? All through the Airbus training course, pilots are REQUIRED to use the bells and whistles all the time and disengagement of said bells and whistles has actually led to failures of flight tests.

This is a bit of a can of worms. A pilot who, on an initial conversion flight test, disconnects the goodies to hand fly the approach - something he /she can easily do - may be displaying good airmanship, but perhaps inadequate knowledge of the New Stuff on the Airbus - which is the core focus of this training program.

It was (is?) routine for newly trained Airbus pilots to arrive on the line actually nervous about turning off the electronic training wheels.

Judging from A Squared's story about folks forgetting there are "other" ways to steer an aircraft into a hold! makes the point that emphasis on automation can go way too far.

Strange as it may seem, a fully manual approach, is a skill that fades rapidly indeed and it doesn't surprise me that the pilots in the article momentarily devolved into an unstabilized approach. I've seen it happen (but not this badly).

(Aside - did anyone else note that the article refers to landing the airplane as "the autopilot's job that can be done by the pilot when required..." WTF!!!!)

Dr.ATP said...

While I agree that check pilots should generally sit quietly and observe, I have had occasion to do otherwise.

I was a check airman for a charter company. One of of the pilots "decided" to try to fly the ILS at cruise power (in other words, he forgot to reduce the power). Approaching glide slope intercept I just *knew* he was going to lower the landing gear, even though we were about 30 knots above the maximum speed. I didn't want to say anything, but I didn't want to leave landing gear parts long the final approach course, either, so I pulled the landing gear breaker.

He somehow managed to pull the power and slow down before putting the gear down; when we got below gear speed I reset the breaker. He never knew.

Chris said...

I don't know about you Aviatrix, but if I was flying with an instructor who simulated an engine failure by switching off the fuel, or even pulling the mixture, I would be letting them know that was their first and only opportunity to do that with me. What's wrong with pulling the throttle back to idle? On more than one occasion cutting off the fuel has resulted in turning a practice emergency into a very real one.

A Squared said...


You may not have noticed that the incident Aviatrix mentioned took place in a multiengine aircraft in cruise.

Chris said...

A Squared - you're right, I did miss that.

What I said still applies for <3000'AGL.

jinksto said...

A Squared, I think we are in agreement. I actually want:

Crew C: the ones that have practiced flying from raw data (even unbriefed) without me in the back.

I completely agree that pilots need to practice flying the airplane manually. Heck I feel safer with "hands on the stick" landings anyway. But, as you know landing is a high workload phase of flight, increasing that workload unexpectedly with folks who want to get home for dinner shouldn't be done.

I can walk across my living room in complete darkness but if someone flips the light switch off mid-travel I'm likely to trip over the futon unless I stop and think about where I am.