Sunday, October 22, 2006

How to recover from a spiral dive

Because this is a how-to posting, I've asked a few contacts to review it. It has been edited to reflect that feedback. To those reviewers: Thank you.

Yesterday's (Saturday) lesson was more slow flight, as will the next few. I'll post about the lesson itself separately. The new topic for this lesson was Spiral Dives.

In preparation, I read about spiral dives in both From the Ground Up, and in Transport Canada's Flight Training Manual. I was underwhelming by the explanations in both. So (since I'm blogging for me) I need to make notes about spiral dives. If I can explain it, I understand it.

A spiral dive is really quite simple, and this is why I was probably under-whelmed by the manuals. Entry is not as dramatic or as sudden as, for example, a spin. Recovery is simple. The only real drama is that, because speeds are high, there may be some interesting G-forces involved.

In a spiral dive the aircraft is flying "normally", meaning that the wings and control surfaces are all functioning, nothing is stalled, the aircraft is traveling pointy-end first, and so forth. The aircraft has a steep angle of bank, and is descending rapidly. The airplane is flying in a corkscrew pattern in rapid descent.

The issue is that an aircraft in a spiral dive, if left uncorrected or if corrected improperly, will very quickly overspeed, resulting in damage or destruction of the airplane (and occupants).

Because of the potential for airframe damage or destruction, Transport Canada states that it is not a permitted solo flying activity, and even when flying with an instructor it is the instructor who must put the aircraft into the spiral, and then transfer control to the student for recovery.

Recovery is simple:
  1. Recognize you are in a spiral dive.
  2. Close the throttle.
  3. Level the wings.
  4. While leveling the wings keep the nose down (it will want to rise due to the high airspeed).
  5. Once the wings are level, pull back on the stick/yoke. Expect to feel G-forces.

More details.

Recognizing you are in a spiral dive is the critical first step, and because of the risk of overspeed it must be done quickly. You know you are likely in a spiral dive when you are losing altitude, and airspeed is increasing. There is no need to detect the steep bank by referencing the Attitude Indicator (and it may be spinning/useless anyway).

Huh? (This was the first under-explained point in the manuals).

If you are going down (fast) then there are four possibilities:

  1. You're in a wings-level dive. Altitude is decreasing, and airspeed is increasing. But: Because the wings are level and the airspeed is increasing, the nose will want to come up, since a trimmed aerodynamically stable aircraft will want to fly in level flight at a constant speed, and will return to that attitude if left alone. So, left alone, the rate of descent, and the airspeed, will decrease as the aircraft levels and follows the trim.
  2. You're in a spin. Altitude is decreasing, but in a spin the airspeed isn't increasing (except due to the engine, but you're going to close the throttle).
  3. You're in a spiral dive. Altitude is decreasing, and the rate is increasing (you're going down faster and faster). Airspeed is also increasing and moving towards overspeed.
  4. There is one low-probability possibility.... The aircraft is in a wings-level dive, and being held there due to being badly out of trim (the trim is holding the aircraft in the dive). Or the control linkages have jammed. Worry about 1-3 first.

Close the throttle. In a spiral dive or a spin, closing the throttle is the first step in the recovery, so closing the throttle is done automatically (it also does not hurt the recovery options if you are in a wings-level dive, so there is no downside). As soon as there is a realization that you're rapidly descending, close the throttle, even if you are not sure whether you are in a spin or a spiral dive. Closing the throttle reduces the risk of airframe overspeed and engine overspeed.

Level the wings. Be aggressive. A spiral dive must be corrected quickly, so this is not the time for a lazy roll. Even with an excessive forward speed, rolling the aircraft level will not over-stress the airframe. Full airleron deflection is unwise (if faster than Vno), but fortunately it is also not necessary. Coordination using the rudder produces a roll which is both faster, and with no yaw (which means minimized airframe stress).

During the roll keep the nose down using forward stick/yoke. Because of the high airspeed the nose will want to pitch up as the wings level. Keep the nose down. The aircraft is at high speed, and rolling. If the nose is allowed to pitch up then the aircraft will both be pitching up and rolling, and at high airspeed the dual-axis translation results in higher airframe stress.

Pitch up (gently) to a climb attitude. My instruction has been to always return to a best-rate-of-climb attitude after any rapid descent (stall, spin, spiral). The thinking is that you know you have just lost significant altitude, and you may be at a dangerously low altitude. Rather than level off and then start thinking (and your first conclusion may well be "crap, look at those trees right in front of me, I need to get higher now!"), it is a better strategy to recover to a climbing attitude, trade the high airspeed for a rapid increase in altitude, and then apply power as airspeed decreases towards best rate of climb. Once stable in the climb, then there is time to think about other factors (heading, cloud base, etc.). It was pointed out to me that during the flight test it is important to listen to the examiner carefully, and make it clear what the recovery attitude will be. If the examiner insists on level flight, give him/her level flight, so recovery to a climb is not assessed as an error. Or pre-flight clarify that recovery will be to a climb.

What does not work (and what kills pilots): Because the aircraft is descending and in coordinated flight, the natural reaction from a pilot is to pitch up (pull back on the stick/yoke). However, because the aircraft is in a steep bank the effect of the elevator is to tighten the turn, not pitch-up the nose. The risk is that the pilot will panic and pull-back the elevators even harder, possibly damaging the control surfaces... but most certainly not correcting the rapid altitude loss. If no other action is taken the aircraft will auger-in with the the stick/yoke pulled back to the stops and the pilot screaming "WTF is happening? Why won't she pull out?".

In normal flight, entering a spiral dive is inadvertent, never a normal flight maneuver. The typical entry methods are:

  1. Disorientation when flying in clouds, night, or IFR conditions, with a pilot that is not instrument-experienced. In fact, the normal outcome of a non-instrument pilot flying in instrument conditions is a spiral or (sometimes) a spin. Usually within seconds or minutes.
  2. Flying in a steep turn, but allowing the nose to drop. Once the nose drops it is near-impossible to recover using the elevator, and a spiral dive will result unless the wings are leveled. NEVER do a steep turn when in the circuit for landing, as there may not be enough altitude for recovery.
  3. Trying to enter a spin, but the the wings don't stall and a spiral dive results instead.

The critical step in the recovery is recognition that the aircraft is, in fact, in a spiral dive and not just flying in a steep turn, or in level but severe nose-down flight.

The critical action in recovery is first getting the wings level, not first trying to pitch up.

The critical factor in the timing is recover quickly, before the airframe or control surfaces are at risk of damage from overspeed. "Quickly" refers to reaction time, not slamming the control surfaces to the stops.

The DA20 Eclipse can withstand 4.4G - that's a lot of force. Even then, the practical guideline on G forces is that they are limits when it is an expectation that the aircraft will be reused. If it looks like the aircraft is not going to be reused, there are no limits. Do what is required.

Note to self: Even after the reading, pre-flight instruction, recovering 10-15 spiral dives, and the post-flight debriefing, I still didn't understand all aspects of the maneuver. Which is why I tried to explain them on this blog. The benefit (to me) has been excellent. Lesson: Blog about other flight techniques and maneuvers, just to make sure I've learned them. You're welcome to come along for the ride.

Sunday, October 15, 2006

Clogged up and staying down

Business trip last week, which meant a flight there&back, and three days in a meetin room with 30 other people. Somewhere along the way I must have been exposed to a rhinovirus, because I now suffering from a head cold.

Drugs are allowing me to sleep OK, but I know I am not mentally sharp, and my eustachian tubes are plugged.

No flying this weekend. Too bad, because while the weather yesterday was unacceptable (low cloud base and rain), today is not too bad.

I'll embark on a whirlwind of house-cleaning and leaf-raking instead. I'd rather be flying.

Monday, October 09, 2006

Twice in a weekend slowly

This Canadian thanksgiving weekend has been the best weather in a long time, my wife is off travelling, and I am going to be very busy at work in the next few weeks, so I went flying again yesterday (Sunday).

I did the pre-flight checkout under long-distance supervision... instead of the instructor shadowing me at my left shoulder, I did it "alone". I suspected and expected, and confirmed later, that he was watching from the office. Excellent teaching teachnique.

The new checklists work well. I led the pre-flight discussion on the take-off schenarios... speeds to maintain, flap retraction, what to do and where to go if the engine dies.

Take-off was a mixture of good&bad. We had a cross-wind from 9 o'clock, I got too fast before we rotated, the left wing lifted and so we were rolling along on the nose-wheel and right main. I got the nose-wheel up, the left wing lowered, and we somehow got off the ground.

Tracking out from the runway was great. I managed the yaw 100%, and we ended up right of the runway. The crosswind drift took me north.

On Saturday I thought my my radio calls were mushy and lacked confidence. Sunday's radio calls were pretty good.

The flying time was all slow flight skills: entering slow flight, leaving slow flight, climbing turns, descending turns, with flaps, without flaps.

And stalls. One really has to work to make this airplane stall decisively. We stalled power-off, we stalled power-on. We stalled in climbing turns.

The "floating leaf" was fun. The quick explanation is that you reduce power to idle, then raise the nose until you are in a stall. Upon stalling you don't lower the nose and break the stall, but keep the stick hardback, keep the stick centred, and use the rudder to control the wing-drop and yaw. It feels like tap-dancing on the rudder pedals. It develops a light touch on the rudder, confidence in ability to control the airplane, and a sense of how to respond quickly to a changing attitude.

On my first try I lasted about 4-5 seconds (it seemed like a lot longer) before I failed to control some yaw, the left wing dropped, the nose dumped over and we entered a spin. Somehow, without analytic thinking, I centred the ailerons, pulled full opposite rudder, pitched down and stopped the spin before it was established, after about a quarter turn, then pulled nose up. In the midst of all this I applied engine power -- totally not required when you're already 80 degrees nose down and gravity is doing a wonderful job of providing more than enough acceleration.

The approach to the airport was simple, altitude control was constant in the approach and throughout the circuit. Much better than ballooning all over the sky a few lessons back.

The landing was OK... the wind pushed out my base leg and I was a little pear-shaped on the start of the final, but not extreme. The flare was my best yet -- we were straight and level about 4 feet off the runway. As the speed decreased the nose dropped faster than I raised the nose, and we made a somewhat fast three-point landing. We were going fast enough that Bendon took control and pulled the nose off the runway to show me the attitude for flying along in ground effect and shedding speed.

Taxi back (taxiing is getting routine), de-brief, pay the bill. I had nowhere to go and they had to wait for the C150 to return, so we pushed the Eclipse back into the hangar and closed out the day in idle chatter about flight in general.

Sunday, October 08, 2006

Lesson 4 - Slow flight

Today was all about slow flying... so much for screaming about the sky, white scarf flapping in the breeze behind me.

In the pre-flight we covered the theory of flying for range, flying for endurance, entry into and exit from slow flight, and took to the skies.

Pre-flighting was all mine. This instructor is good (or maybe they all do it). Before there was coaching, and this time he didn't say a word. I, on the otherhand, chatted my way through, telling him what I am looking at and what I am seeing. The silence is reassuring, and marks progress.

We have new checklists. Still compliant with the manufacturers' lists, but now slightly re-ordered to be more logical, and also with a few new entries. Early in the list there is a section where we ensure all of the non-flight controls move freely (cabin heat, etc) and I comment that there is no check to make sure the alternate air source is free. I check it anyway.

Startup, and taxi to the run-up area. When I look back at the earlier posts and read about taxiing like a drunken sailor I'm smiling, as my taxiing is much much better. Seat-time does that to you.

Today's take-off was smooth (see last post), I maintained rate of climb quite well, but I didn't counter-act the yaw from the climbing take-off power and so we ended up about 20 degrees left of the runway.

We covered entry into slow flight, turning, recovery from slow flight, flying clean, with take-off flaps and with landing flaps. We used the empirical method determining of determining best endurance speed (the Eclipse's manuals don't illustrate anything below 2400RPM, and under yesterday's conditions the best endurancespeed in this aircraft was about 1650RPM).

Turns without losing/gaining altitude are getting better. My trimming of the elevator was a bit ham-handed, and that was the biggest factor in my barely adequate altitude control.

Radio calls were OK, though clumsy from time to time.

But the weather was wonderful, the sunlight bright, the air quite clear (the brown smudge of smog in the distance was evident, but that barely affected looking down at the trees in their fall splendor).

We had some extra time and a partial fuel load, so the instructor demonstrated stall recovery and spin recovery. I really need to make a video of a spin recovery.

The circuit and setup for the landing went well. I flared a bit high and then didn't fight to keep the aircraft flying, so we landed well but it was a bit of a drop. I'll probably have 5-6-7 landings under my belt before we start circuits.

The bonus was following the lesson. After each lesson there is a de-briefing, which covers the core of today's exercises and takes 5-10 minutes. The next student was solo, and after he was dispatched, I had lots of time to discuss every phase of the flight with the instructor.

The weather is wonderful today, so I'm going to do a few hours of paperwork and yardwork, and then I have another lesson this afternoon.

Airports, airports, airports

Yesterday (Saturday) was airport day.

Up at 4AM, then off to CYOW to put my wife on a plane. She sings with a Sweet Adelines chorus, and they're off the Las Vegas to sing in the international competition. Most amusing part of the trip was when it was about 20 minutes to flight, we were at the front of the line-up, and someone (who happened to be from the same chorus) arrives in a panic for the same 6am flight. Give credit to the Air Canada staff -- they didn't roll their eyes at the time-challenged pax, or fall on the floor laughing. Then again, it was probably the first such occurence of the day, and most certainly the last.

While waiting at the airport the sun had come up, there was now a bit of a breeze, and the fog blew down from the Ottawa river to the 417. What was an idlyllic morning drive into the airport had turned into pea soup for the latter part of the drive home.

Wake up daughter, eat breakfast, and then drive back to the airport to catch a flight for a two-week vacation with her bro in Calgary. The lines at Air Canada were much shorter this time, so we had a chance to sit in the sunshine and chat, before seeing her through to security.

And then, in mid-afternoon, it was off to CYRP for a flight lesson. More about that next posting.

The weather yesterday and today is wonderful... minimal wind, wonderful fall colours, and bright warm sunshine. I'm torn between doing yardwork, or to go book another lesson. Or to drive down to the gliding club and say hello to my friends (and maybe go on a glider flight).