Friday, March 13, 2009

In a Flap

The customer called me at noon for a mostly daytime flight today. I picked up a strawberry chicken sandwich at a deli and ate it in the truck on the way to the airport. It was slightly messy, but definitely good.

The before takeoff checklist requires me to verify all the gauges are green: oil temperature, oil pressure, cylinder head temperature, fuel flow, EGT: every needle sitting in the proper range. As I add power, my eye flicks again to the two three-fold engine gauges. When I flew this airplane two crew the non-flying pilot's line was "green six ways" and I still think that as I start to add power for the takeoff roll. Airspeed alive, rotate, positive rate, insufficient runway remaining, gear up. The engines are at full power now, and I'll leave them that way through five hundred feet of climb. I'm not going to mess with them until I have at least a few seconds of glide time in which to think.

I'm never conscious of having just looked at my engine gauges, but my wheels were barely off the pavement before I knew that the needle on my right engine oil temperature gauge had fallen to zero. Perhaps the motion was rapid enough that I caught it in my peripheral vision. Presumably I'm descended from the cavemen who saw the sabre-toothed tigers sneaking up, and not from the ones who missed that slight movement of a stalking cat, so this sort of thing gets my attention. It's training not evolution that resulted in absolutely no reaction from me though. That's not steely astronaut nerves. It's just knowing that there is no emergency that causes oil temperature to suddenly zero. It's just the gauge. I continue the flight, tapping the instrument from time to time, but it never comes back to life.

The route took us over Texarkana, a town name that reminds me of the names of the American republics in Robert Heinlein's future America in his novel Friday. It's actually a town on the border of Texas and Arkansas, mostly the latter. I don't know what it has to recommend itself apart from its cool name and accessible airport. While we were on frequency there, a departing jet reported a flap problem. The flaps were stuck at nine degrees, I think the pilot said. "Do you want any equipment?" asked the air traffic controller, after confirming that the filed destination was unchanged. The pilot gave a rambling answer, saying that he'd discussed the issue with company, and that he wasn't declaring an emergency, and he didn't expect any problems, but under the circumstances, yes, he'd like trucks. For those of you not familiar with aviation euphemisms, both "equipment" and "trucks" here refer to the sort that have flashing lights all over them, and sirens.

After a while he changed his mind and said that yes, he was declaring an emergency, but that his situation hadn't changed, he was going to be fine. I don't think the paperwork is any different whether you use the e-word or not.

I'm sure he arrived safely, after a somewhat slower than usual flight and a somewhat faster than usual landing. I hope his passenger briefing was more succinct than that radio call. I'm not saying I'm any expert on emergency PAs, nor that I don't make the same mistakes, but you can hear some cringeworthy cabin announcements from pilots who are more worried about scaring the passengers than about the upcoming landing. The rule I'm trying to cement in my head is keep it brief. If you say there's nothing to worry about once, some passengers start to worry. But if you say it twice, everyone starts wondering just how worried they should be. Phil has an amusing account at his blog of a flapless landing.

A reader sent me the clip above of a propeller malfunction at fifty feet after takeoff. Not only did the propeller go into beta uncommanded, but pulling the prop control back to minimum didn't fix the problem, so he had to shut down the engine. All done correctly by a brand new captain, I'm told.

There's nothing in his PA that shouldn't be. It doesn't matter whether the engine is surging, on fire or the Mormon Tabernacle Choir jumped out of it and started dancing on the nacelle, "experienced a slight abnormality" and "precautionary shutdown" are reasonable things to say. The bullshit might be less obvious if he had left off the "slight." You don't want it to come out like, "Nothing is wrong, we're just going back to land because uh, I forgot my wallet." I think it's mostly the repetition of "here" that undermines his message. He sounds like that not because he can't talk, but because talking is occupying only a small part of his attention. Passengers hear that and think that the rest of his attention is taken up with "Oh my God, we're all going to die!" A Dash-8 climbs like a bat out of hell on one engine, and he's in VMC, so while I'm sure he's focused on landing the plane, he's also thinking about the inevitable paperwork, fending off idiotic comments from his FO, wondering if this is going to screw up his whole work schedule for the month (answer: no, a newly upgraded captain already has the worst possible schedule), and hoping his SOPs are letter perfect on the tape.

I approached my destination just after dark. The shape of clouds that had not been topped very high at sunset obscured some of the stars. I was focusing on my descent checks and noticing flashing on the engine nacelles. As the sun goes down the strobes reflect off the nacelles and I almost always think, "was that lightening?" One the sunset was so bright and orange reflecting off the bottom of the nacelle that I did a double-take, looking for flames. The strobes can become distracting, so I switched them off, but I had a little suspicion already that it wasn't just the strobes this time. I turned slightly and at that moment the clouds right in front of us lit up the sky, right to the back of the airplane with its blacked out windows. You can't count the seconds between the flash and the bang in the air, because you hear the lightning instantly on every radio frequency. There was little radio interference, and winds at the airport below were light, both consistent with my belief that the lightning was a good ways off. Terminal passed me off to tower and I joined "downwind" (in quotes because I had been required to stay high until very close to the airport, so we were still plumetting towards circuit altitude, not flying a level circuit). I land, and taxi in past someone washing his car on the ramp.

27 comments:

Steve said...

Love your posts - I look forward to them every day.

Anonymous said...

In the video above, what surprised me the most was the smoothness of the landing: no lurching of the camera was evident at touchdown.

I hope he has a long career in aviation.

Anonymous said...

How long would you expect pressure to take to fall if the oil pump shaft broke? That happens from time to time either as a root cause of failure or as a result of FOD jamming itself into the pump.

I suspect if something actually was wrong, it wouldn't change your actions except that you'd eventually shut the engine down when you were at a safe altitude or it no longer was making thrust.

Louise said...

I know what you mean about keeping it brief and not repeating yourself. I once failed the blood drop test when giving blood - the drop of blood floated when it should have sunk, or sank when it should have floated. They took a blood sample to send off and check that I wasn't anemic. I was told by 5 different people that this was nothing to worry about, most of them telling me several times. By the time they'd finished reassuring me I thought I was on my death bed. Less is definitely more.

Anonymous said...

Interesting video clip. I'm always impressed that the Dashs can find room to tuck their long legs away. I was watching for the moment of failure, but there was just a dissolve to the feathered prop. Knowing nothing about turboprop controls .. ( Is "beta" reverse pitch? Are take-offs done with flat/fine pitch and the failure leading to un-commanded coarse/beta pitch?) .. I'd expect this to be a rather dramatic failure with take-off power, but maybe it was gradual. Well handled anyway, the single engine landing looked routine.

I'm glad you are aware of Heinlein's work. His stuff has been a favorite of mine, growing up with "Podkayne of Mars", the classics like "Stranger in a Strange Land", and maybe my all time favorite, "The Moon is a Harsh Mistress". Friday was a small departure from his "future history" line, but much more fun for me than the later Heinlein's novels. His cantankerous Lazarus Long would fit well in Texas, what with all the guns, radical libertarian politics and kilt-wearing. Well, maybe not the kilts.

Anonymous said...

Aviatrix,
The resulting asymetrical thrust from one engine going into beta at full thrust must be great, with one engine full throttle and one in full reverse can the rudder alone still keep the plane straight?

Mark

Anonymous said...

Hey, readers... Aviatrix offered me a chance to write a couple of answers to the Dash malfunction questions.

In this kind of controllable prop, you have feathered (about 90 degrees from the rotation) the flight range (coarse to fine), flight beta (also called flight idle, ultra fine pitch), the ground beta range, aka Disc (ultra fine to essentially zero), and the reverse range (beyond essentially zero, where the engine actually increases power and a host of automated things kick in).

It would look like this from the top, if the prop on its hub rotated clockwise:

Feathered: |
Coarse: /
Beta: -
Reverse: \



The prop on takeoff is here: -
At fifty feet, the engine thought it went to here: \
Which forced it back toward here: | to /

And a fight starts. The engine is pulsing from where it is commanded to be by the condition levers in the cockpit to the coarse condition by the beta-lockout system... the thing that failed. It failed by misreading what the prop was doing, and came awake at a bad time.

The answer is to command the props to the coarsest flight condition in the cockpit. At that point, the failed system obviously knows it can't be anywhere NEAR the beta range, and shuts off.

It didn't shut off the pump in this video, and he had to shut down the engine.

The "good" engine has plenty of horsepower to get her up and out of any situation by itself.

As far as actual full reverse in flight? You can't control the plane. It is one of the most insane things I've ever seen in the simulator. It is an instant departure from controlled flight, a huge fight to get the affected engine shutdown, and a call to maintenance for a new seat cushion. Severe injuries would be a certainty in the back, and if the FA was up- she'd probably be killed. It is that extreme.

Anonymous said...

Ah, thank you anonymous. That makes sense. It was a prop pitch "fight" and not anything as drastic as reverse or even beta inflight. That would be a loss-of-control scenario.

I've heard MU-2's and the like on approach/landing, and when the props go into beta it must be like holding up 12-foot diameter plates. It slows down realfast.

Anonymous said...

Sarah wrote [i] Is "beta" reverse pitch? [/i]

Sarah, “beta” is one of those terms which generates a lot of misinformation.

You will hear “beta” variously described as: reverse pitch, reverse thrust, zero pitch, flat pitch, zero thrust pitch, and similar definitions. The trouble with these definitions are that they are all simplified to the point of inaccuracy.


The truth is, “beta” is not a blade pitch angle, nor a pitch angle range, although it is related to a range of blade angles. Beta is actually a range within the function of the engine and prop controls. On the airplane I fly (L-382) the engines always turn at a constant RPM (unless a low speed ground idle , LSGI, setting is selected)

In the Alpha range (normal flight range) engine power is controlled by the position of the power lever which commands the fuel control to deliver more or less fuel to the engine for more or less power , and blade angle is controlled by the governor, which commands more or less blade angle to maintain 100% RPM while he power increases or decreases.

By contrast, in Beta range, control reverses effectively; the prop governor is bypassed and the prop blade angle is selected by the power lever position, and the fuel control varied the amount of fuel delivered to the engine to maintain 100% RPM at the blade angle selected by the power lever.

To summarize:
Alpha range, engine power is selected by the pilot, and blade angle is automatically selected by the prop governor.

In Beta range, the prop blade angle is selected by the pilot and the engine power is automatically selected by the fuel control.

Although you will find neutral thrust and reverse thrust within the beta range, beta range is neither correctly defined as reverse thrust nor as neutral thrust. On the L-382, if you have all 4 engines at 100% RPM and you have the power lever in the upper end of beta range you get a great deal of forward thrust, and you'll very soon find yourself moving very fast, unless you're dragging the brakes constantly.

Now this description is based on the L-382 which has engines which run at 100% RPM all the time (unless LSGI is selected) The Mitsubishi MU-2 also has powerplants which operate this way, and I suspect that the definition of Beta for that airplane would be virtually identical.

The Dash-8 has a different type of engines, ones which will operate through a range of RPM, rather than either 100% RPM or LSGI (about 70-75%RPM) I've not flown the Dash-8, nor other airplanes with similar engines so I won't claim any special knowledge of that airplane, but from doing a little reading, it appears that on those engines, Beta has a similar definition, with RPM added as a variable, rather then just power and blade angle.




Using the Dash-8 as an example, the Beta Metering Valve activates with the Power Levers just slightly above Flight Idle, maintaining the blade angle at 17.5; this actually occurs in flight. On the ground, pulling the throttles past the Flight Idle gate will continue to push the prop farther into Beta, REDUCING the blade angle more and more until it reaches the "Disc" position, which has the blades at 1.5 degrees, almost flat relative to the motion of the airplane. During this stage the engine is idling, only putting out enough power to keep the RPM from decaying into the restricted low-RPM range. This "Discing" is the big drag referred to above, and in this position the prop actually throws some air SIDEWAYS (go stand next to one in Disc sometime if you don't believe me!). Passing another stop into the "Reverse" range, the prop angle goes negative, decreasing all the way to -12 degrees; simultaneously the ECUs will increase engine power, which DEFINITELY blows A LOT of air forward (again, go stand in front of one sometime!).

Here is a description of Beta in a Dash-8 which comes from the PPRUNE tech log archives.

http://www.pprune.org/archive/index.php/t-324680.html

[i]System-wise, the Beta valve actually controls regulated bleed-off of oil pressure in response to a blade-angle feedback signal from the prop, doing so to maintain a specific blade angle commanded by a linkage to the Power Lever. In the extreme lower range of Beta an additional valve (Reverse Valve) opens to allow unmetered oil pressure into the fine pitch side of the prop pitch change mechanism, which is then regulated by the Beta Metering Valve as described above.

Operationally, in the "Alpha" or Governing range of the prop, the PCU or governor varies blade angle to maintain set RPM; in the "Beta" range, the governor is bypassed and the Power Lever directly controls blade angle to adjust drag for ground (or sometimes flight) operations.

Again, the example above is for the PW 121 in the DHC-8, which is a free-turbine engine. Similar principles apply to most others. Hope this helps! [/i]

So unless the author completely fabricated that description, Beta on the Dash-8 is pretty much analogous to Beta on the L-382, Prop pitch is controlled by the governor in Alpha range and by power lever position in Beta range.


It's not as simple a concept as “flat pitch” or reverse range” or “neutral thrust”, but I've always preferred an accurate description that takes a little more effort to wrap my mind around, than a description that is simple, easy to understand and incorrect. YMMV

Anonymous said...

Sorry, I hit the publish button a little too quickly, everything from "Using the Dash-8 as an example" to "hope this helps" is a quote from the referenced forum, except for:

Here is a description of Beta in a Dash-8 which comes from the PPRUNE tech log archives.

http://www.pprune.org/archive/index.php/t-324680.html

which is my comment and should have been placed at the beginning of the quote, instead of inserted in the middle.

Sorry for the confusion.

Anonymous said...

WOW! Since A^2 took this to a new level, I will too.

I have only flown the PW120A, 123, and 123C engines on ye Dash. Never a screamin' demon Allison powered workhorse!

Our "Flight Beta," or "FLT IDLE," is the stop at which the gates prevent going into "Disc", and that mechanical stop is augmented by the beta-lockout system. It prevents a PLA (power lever angle, or position) below FLT IDLE- into the "Ground Beta" range. This is the same system that failed. It has it's own prop feathering pump and metering system.

HOWEVER. When you select the PLA below the bottom of DISC, into the "REVERSE" range, the ECUs schedule an increase in fuel to accommodate an increase in reverse power. PLA is irrelevant to blade angle.

So, PLA controls prop blade angle within a certain range (depending on prop condition selection), and past that- the ECUs schedule fuel and the governors come alive to prevent a whole number of things... Ng overspeed, prop overspeed, etc.

I have had the distinct pleasure of flying with a Marine Herc pilot as the FO.

What a fantastic airman, and a heroic drinker.

I called him "Major XXXX" the entire trip, even as I bought the beers.

Excellent post- is it airspeed and altitude?

Anonymous said...

@Aviatrix - No suprise that you'd read Heinlein. Your persona and inner airplane/techno-geek suggest you'd find his writing entertaining.

@Sarah - Two thumbs up for The Moon... Though, I have all of his works, The Moon is a Harsh Mistress and The Cat Who Walks Through Walls are my favorites. Friday is another good tale, though I must confess that the cover art was what caught my attention back then...

Anonymous said...

WOW ^2 anonymous. Thanks for the details on alpha/beta/PLAs and what the levers actually do on the Dash8 and the mighty Hercules.

I'll probably never have use for this knowledge (unless both pilots have the fish & are food-poisoned ) but I like to know as much as I can. Which is surprisingly little...

Anonymous said...

anonymous,

Thanks for adding details of the Dash's engine/prop controls. My knowledge of free turbines is pretty spotty.

is it airspeed and altitude?
No nothing that clever (although I might adopt that) It's just my initials

Anonymous said...

Sorry to hijack, Aviatrix-

No prob, Asquared. The Pratt is a relatively simple and tough engine. It is TINY, too. Take off the RGB and AGB, and it's less than two feet long. Another blogger would call that 'major cool.'

Since it's a free turbine, we can't NTS. There's nothing to sense! Instead, a system compares torque output between the two. If conditions are met and one drops to an arbitrary level (depends on the engine), it automatically feathers the sick one and uptrims the good one.

NOT a fun system failure, by the way. Torque jumps through the roof and twists the "turbomachinery" all to hell if it feathers a good engine.

Aviatrix said...

I don't consider it a hijack at all. It's expanding on the information in the post. I'm delighted to have the information. Maybe I should get you guys to guest post.

Anonymous said...

Anonymous,

I was wondering if there was an NTS system or something similar. For those wondering what NTS is it's Negative Torque Sensing. NTS is a system on the Allison 501 and (other turboprop engines) the which mechanically senses when there is negative torque on the reduction gearbox (ie when the prop is driving the engine rather than the engine driving the prop) and it opens a valve in the prop control which increases the pitch until the negative torque situation is alleviated. It's not exactly an auto feather system, because it doesn't feather the prop completely, but in an engine failure it will reduce prop drag significantly until the prop can be feathered.

Anonymous said...

I thought you'd get a kick out of that.

There's a myth about free turbine turboprops- if you hold the propeller and start the engine, it won't rotate. The idea is the core of the engine is designed to blow air aft across another set of turbines, which are on a shaft inside the core's turbine shaft (like a paper towel roll over a broomstick). The broomstick goes all the way through to the rest of the magic, and that's what turns the prop.

Hold the prop, the aft turbine stages won't turn, and you have the paper towel roll screaming along just happy as a clam while the broomstick stays put.

NOT TRUE. The amount of torque and the amount of air 'thrust' across the free turbine stage is INCREDIBLE.

I've seen a mishap firsthand (and ferried that damned thing later) where this was proven wrong. That early morning, the slices of swiss cheese lined up so that a brand new procedure, a brand new FO, and a sleepy captain fired up an engine that had the prop tethered (never tied that prop before, the FO assumed ground handling removed it).

It was one hell of a sight to see- once the first "SLAP" occurred, two of us went running with the "KILL ALL ENGINES" hand signal.

Anonymous said...

There's a myth about free turbine turboprops- if you hold the propeller and start the engine, it won't rotate.

Hmmmm, do you mean that it can't be held stationary, or that it can't be held stationary by a human grabbing it?

It's my understanding (and I may have been misinformed) that the ATR, which also uses a 100 series P&W engine, has a prop brake, and the prop can be held stationary by the brake, while the rest of the engine is running.

Anonymous said...

How cool to hear that you're at least a part-time Heinlein (Hine-line), reader! I've actually pondered that very point, and figured there was an excellent chance!

As a Q400 Captain, I thought the PA in the embedded video was excellent. I think the repetition of "here" serves to make the announcement sound slightly more conversational, and thus indicative of an inconvenience, and not a precursor to a BRACE command. I consciously affect a very slightly 'bored' tone when announcing such things that may be alarming in appearance to pax.

Signed,

Formerly Flydaddy, not blogging any longer due to creeps.

Anonymous said...

Yeah, the myth was by hand. I've been hearing it for years- and I have never had any "takers" for 100 of my dollars so I could watch them prove it!

You're exactly right- the PW127 on the ATR has a prop brake on the right engine, in lieu of an APU.

Dagny said...

"I've heard MU-2's and the like on approach/landing, and when the props go into beta it must be like holding up 12-foot diameter plates. It slows down realfast."

I just wanted to say, that is VERY correct. ;)

Callsign Echo said...

YAY Heinlein and YAY technical explanation of beta range...

Anon and A Squared's comments prompted me to read further. I realize I had a total misconception of what that meant.

The sheer genius of aircraft systems never ceases to fascinate and amaze me.

Moon is a Harsh Mistress is one of my favorites. Starship Troopers is probably my all-time favorite...just seeing the movie title in Blockbuster makes me want to throw things. What a horrible to do to a provocative and innovative story. And way to water down Carmencita's character. They took a smart, independent, talented, beautiful, powerful woman and turned her into a two-dimensional static eye candy. Her role did not benefit by being expanded.

Oops, threadjack, sorry.

Great post!

jinksto said...

Too late now Call Sign Echo... I'm with you in hating that movie! What a horrible horrible thing to do.

Anonymous said...

Interesting reading. I think I found the actual plane tail number for this incident.

N333EN

Enter that into:

http://www.myairplane.com/databases/aircraft_report/

Check all the Scroll to the bottom of the incident

Anonymous said...

Triple 3 is a good machine. Each one of our Dash 8s has a personality. Some fly "crooked," others haul ass during single engine taxi (ECU engine trim is whacked), some land purty, some land crashy, one had an unexplained anomaly that months later proved to be a magnetized flap drive shaft due to a lightning strike.

This particular plane's gremlin was an overzealous APU bleed air valve. When on the ground and operating the air conditioning, it would pull so much air out of the compressor stage of the APU that it would kill it if you weren't careful or aware of it.

She is a great machine. My favorite 300 in the fleet.

Now that we've gotten official confirmation that we're losing 20% of our airplanes- flying that will be replaced by other airlines' utlra inefficient RJs- I need something to help the sting of a $35K paycut for myself, and a 100% paycut for up to 90 of our pilots.

Unknown said...

this was not a propeller going into Beta...it was a failure of one of the components in the system, that detects excursions to Beta (of either the controls or components), and limits the propeller speed to 1000 RPM...no engine shutdown was required or necessary