Through A Glass, Darkly

According to the National Transportation Safety Board (NTSB), in 2006-7 the fatal accident rate for conventionally-equipped ("steam gauge") aircraft was 0.45 per 100,000 flight hours, while that for glass-paneled aircraft was 1.03 per 100,000 flight hours, more than double! [Aviation Week and Space Technology, July 26, 2010, p. 66]. The FAA and the NTSB have squabbling in their mutual genes, and the FAA says that it has done something about the problem by revising the Practical Test Standards (which detail the level of skills that pilots must show to get a new rating) and the Instrument Flying Handbook [H-8083-15A].

Of course both are right. The Instrument Flying Handbook does a great job of explaining how to use glass, but aside from instrument instructors, few pilots read it, although they should. And most pilots, even may professional pilots, forget all about the PTS once the check ride is passed. So maybe the next generation will use these tools more effectively.

The problem is with legacy pilots like me. I got my instrument rating in 1985, while EFIS was nothing but a gleam in some engineer's eye. I used VOR, ILS, DME, and ADF. I lost my instrument rating in 1993 when I became an ATP. To do that I used VOR, ILS, DME, and ADF, supplemented by a VFR-only GPS receiver that had no moving map.

Then one day I found myself going to Santa Monica (KSMO) in an EFIS-equipped King Air. The company had just bought it, and I had read the manual backwards and forwards. But between the early departure, the unfamiliarity, the pace of flying in the LA Basin, and the empty right seat, the odds were stacked against me. I programmed the route before I left, and enroute I managed to load the approach. It was a morning arrival and SMO was predictably under the marine layer. There was traffic everywhere, my airspeed was high at ATC's request, I was descending, and I was busy squinting through the haze looking for traffic and the airport. I was way behind the airplane.

But no matter; it just turned final without me.

The key to this (just barely) success story was that I had spend hundreds or even thousands of hours flying with various IFR-certified GPS units. My habit was to load an approach every time I arrived, whether IFR or VFR. "It's clear and a million!" people flying with me would chide. "It will be winter someday," I answered.

The great chef Escoffier said that one does not know how to cook a dish until one has cooked it 1,000 times.

I say that one does not know how to load an approach until one has loaded an approach 1,000 times.

The key to flying glass safely is practice and training. New pilots get it, because they are new; old timers like me need to make sure that we get it on our own. And, judging by the people at my home field who have glass cockpits, they are not doing so.

The same applies in the glider world. At a recent SSA contest, one of the contestants was selling the latest glass for sailplanes. These are complex units, incorporating the ship's performance, the terrain, and the winds aloft, and after each day's flying I saw him sitting and demonstrating what the unit he was selling could do. He was super-proficient at navigating the pages and punching the buttons.

But he landed out at least once, too.

Tanking Up

Yesterday's training flight brought me a lot closer to total instrument proficiency, although I still have a way to go. My goal is perfection, a goal impossible to reach; the value is in the trying. With the medical problems both my currency and proficiency lapsed, despite some work with various Flight Training Devices (aka simulators).

I did not make things easy for myself. First, all of our club planes are tied up, so I rented a 172XP. I've flown this XP a lot, but never IFR. Unfamiliarity with avionics is a link in a lot of accident chains. In this case I said "Where is the ADF needle?" out loud, but that's the same kind of confusion you see when someone doesn't understand what, say, a Garmin 430 is showing, or how to get it to do what you want. Time spent looking for a needle in a haystack is time not spent managing the airplane. And, with no autopilot, not managing the airplane meant that my heading control was by no means perfect.

Of course it was hard to tell what my heading was: This XP's heading indicator precesses too quickly.

The airplane has no DME and I was using a handheld GPS as training substitute. Between the NOS approach book and the GPS my lap was pretty full. The polite phrase for this disorganized behavior is what the Practical Test Standards now call Single Pilot Resource Management. I'm planning to buy a new GPS, so this area will need some more work. I can use my struggle with it to help my students, especially the ones with their own planes, which tend to have every portable toy available. (The worst I ever saw was the owner of TBM-700 who, despite having every Moving Map, Primary Function Display, and Multi Function Display known to humanity installed in his panel, got lost while being vectored for an ILS.)

The wind was an issue, too, in a good way: a strong westerly flow that distorted my mostly north-south approaches, and made timing in the hold tricky (outbound groundspeed 80, inbound groundspeed 140). That's just good practice.

I flew one approach to an airport just north, planning to intercept the transition for the ILS back home. The first approach went fine, despite the spinning heading indicator and large wind correction angles, because I used the GPS "track" display to make sure that we were moving in the right direction. I started the published missed approach, and tried to tune the VOR for the transition. The NAV flag stayed visible and there was no ID, although I didn't remember seeing a NOTAM in my DUATS briefing.

It was time for plan B. I tuned the ADF to the outer marker for the ILS and headed there. This was a fine plan, except for something else that was not in my briefing. Fire season has begun, although due to El Niño and a late monsoon it is starting late. There were two heavy tankers (P-3s) and two SEATs (Single Engine Air Tankers) working the fire, and they were coming back just a little east of the localizer. I had a good safety pilot and we spotted everyone, but I have done a lot of fire fighting and know that the tanker crews are busy; holding at the marker would definitely be in their way. (I checked: there was no Temporary Flight Restriction last night, although there is one this morning.)

So now it was time for Plan C: hold at the VOR, then head out for the ILS.

---

You just don't get this kind of rapid decision making in any of the simulator scripts I have followed. The simulator has a lot of value, but in any case you have to fly to get the full picture

---

After we were done, Dan (who is not instrument rated) asked me how often I fly approaches in IMC. The answer was embarrassing: the last one was about two years ago. My medical difficulties explained a lot of this, but still...two years? I had filed IFR in that time, and flown plenty in IMC, but always got a visual approach. When I flew King Airs or freight I flew a lot of approaches, but I haven't been PIC of a King Air in a few years. My personal flying is in airplanes that cannot handle ice, so the winter is dicey at best. We tend to vacation to the east; if we took a trip to the Oregon coast there would be plenty of approaches.

Be that as it may, I am current and ready. The trick will be to keep the tank full and stay that way. I wonder if my wife would like a trip to Oregon for her birthday?

Over The Hood

Instrument instruction is its own world, because doing it right requires the instructor to do so many jobs.

First, the instructor has to understand instrument flying. Well, that's impossible. Instrument flying is an infinite series of small realizations that eventually sums to something huge. While you can leave ignorance behind you, mastery never gets any closer. So really the first job of the instructor is to have the desire to understand instrument flying.

Unfortunately, it is often the case that the instructor once had a good start toward understanding instrument flying, but that has stagnated or even backslid. Almost every instrument instructor I know has, at one point, confessed "I thought I was staying sharp by watching, but I was wrong." The cure for this is for the instructor to fly instruments regularly, but that may be difficult on what CFIIs usually earn. In my case, though, I have the two club planes and a day job, and while money is tight at our house (see my last post), I can afford to take some time to at least fly under the hood, barely performing to ATP standards.

Next, the instructor has to be patient. We are teaching something difficult; a lot of pilots start instrument training but don't finish, and a lot of pilots look back over a long career and declare instrument training to have been the most difficult part. We need to remember how difficult it was for us.

And so the iPhone version of XPlane has been kicking my ego around. A couple of the aircraft have EFIS panels. I have a few hundred hours of experience in various EFIS aircraft (not a lot by some standards), but each one is different. For XPlane, the iPhone itself is the yoke: tilt it toward you to pitch up, tilt it away from you to pitch down, and rotate it left or right for aileron control. Hold it rotated to the right and it simulates an aileron roll. This is a little twitchy, so let me just say that I seldom fly it to ATP standards (it includes a few ILS approaches). I am struggling just as hard as my students. (The HUD is a different matter: its velocity vector enables you to fly with extreme precision.)

So, the little simulator in my pocket is great. But it's not a real airplane! And that leads to the hardest part of the instructor's job: safety pilot. Because all the time that the student is under the hood, it is up to the instructor to keep the airplane from hitting something. (The FARs require that the safety pilot has a medical certificate to be sure that the guy keeping watch meets the vision requirements of 14CFR67.) It doesn't matter how insightful and fair your critiques of your student are if he does not live to try it again at a later date.

And now I find that being safety pilot takes up a large part of my instruction time. This was my inspiration from Jim Ralph, the FAA inspector who gave me my initial CFI checkride. Jim was so quick and so funny that it was exhausting to be around him. He had no stomach for BS, whether from the FAA (which he called it the "F'ing A A") or from anyone else. He was a great stick and a great student of every corner of aeronautical knowledge.

Jim's pet peeve (besides Hilary Clinton, who was First Lady when I met him) was midair collisions. "I am not going to die in a midair collision," he said, "and when we're flying I'll bet I see 10 airplanes that you don't see." This was not a statement about his perfect vision: it was a statement about his perfect lookout.

I made Jim very happy once. He had come to watch me give training in GPS approaches for a 135 operator, one of the first to take advantage of the new technology. The Director of Operations was flying the 414, and I was in the right seat. Jim sat in back, and, as is customary, several airport bums were riding along. One of them was a new CFI, working on his CFII, and as we did our procedure turn I said, "Hey, Ben, see how the procedure turn is right over the freeway? You know lots of pilots follow the freeway, so you need to keep your head out of the cockpit through here because there's bound to be lots of traffic." Jim smiled.

Well, Jim was right, he died of cancer, not in a midair collision. But now when I have a student under the hood I act like Jim is in the back seat, critiquing my lookout and spotting more planes than I do. And I do a good scan for traffic.

---

A funny story, at least to me. I was at a local safety meeting, sitting next to one of my instrument students who has his own airplane, a very nice twin. We're upstairs in the classroom, sitting in the same ratty chairs that were there when I moved here in 1991. Some of them are a little wobbly, so I didn't notice that I had set mine down on his foot.

"Ow!" he screamed, "you're on my foot!"

"Hey," I replied, "when I said 'dead foot - dead engine,' I meant it!"

Transitional Times

A nexus of three things got me thinking;

1. aviationmentor's post "Where have all the pilots gone?". One of his points is that the FAA Knowledge Tests are behind the times in requiring knowledge of NDBs;
2. a discussion about NDB approaches with a student;
3. trying to design an extended IFR cross-country program, like  Doug Stewart's or
   Morey's West Coast Adventure

aviationmentor's posts are always thought-provoking and detailed, and he has some good points about the knowledge tests, which are really points about the way we teach flying. I still believe that there is something to be learned from NDBs, but the test questions don't test that, so books and instructors don't teach it, so the point is well taken.  I'll tell you what I think you should learn in a bit.

aviationmentor also points out that NDBs are disappearing quickly, but we should expect that: after all, there was a CONSOLAN station operating on Nantucket in my memory [What's that?? CONSOLAN was a clever low frequency long-range navigation system so obscure now that Wikipedia doesn't mention it, which also means that it wasn't in the Eleventh Britannica.  You can read about it in Bowditch's American Practical Navigator, if you can find an older edition.]

Just like CONSOLAN, lost between Brittanica and Wikipedia, I feel lost between the NDB and the Garmin G1000.  Where have all of the instrument students gone?  That's why I started to think about the multi-day IFR cross country program, as a way to attract students.  I have done this in the past, and it is a great way for people to learn instrument flying in the real world, either as initial or recurrent training.

But lack of a good syllabus isn't the problem.  The problem is that there are no NDB approaches nearby anymore, and the school where I teach has no airplanes that can be considered IFR trainers.  I'm caught in between.  

What makes a good IFR trainer?  A working ADF is nowhere near as important as a good IFR GPS.  You don't need a G1000 all-glass panel, nor even a Garmin 430, just a working IFR GPS, complete with current database.  [The school has one airplane with a King IFR GPS, but it is not working.]  The work of picking a good sequence of airports and routes and approaches and places to stay and eat is just make-work without an airplane that can do the flight.

So what's the use of studying NDBs?  This afternoon I was with a private pilot student who is starting cross-countries.  He had a lot of good questions, and I was trying to emphasize the basics of heading, heading, and heading.  Although I am sure that he will never, ever see an NDB approach, I found myself drawing the following picture.  I'm including it hand-written, not because I am a Luddite but because that's the way I use it.  Part of the pleasure is drawing it.

The picture shows a typical NDB approach from above.  You start over the beacon, and it will take about 2 minutes, or 1/30 hour, to get to the runway.  Imagine a screaming direct crosswind, one you would be willing to land in after breaking out at minimums.  I'll pick 15 knots, because that's what the Cessna 172 POH says the Skyhawk can handle with average pilot technique.

Imagine that you ignore the screaming crosswind and fly without any wind correction.  In  the 2 minutes it would take to fly from the beacon to the runway, a 15 knot crosswind would only blow you 1/2 mile to the side.  Since the visibility should be at least a mile to do the approach, you will see the runway and be able to land.  That's the red line: you land successfully with no wind correction.

If you pay attention to the wind and put in a little crosswind correction, say, 5 degrees, you will go directly to the runway.  That's the green path. 

Notice something: I never mentioned the needle, the relative bearing, the relative azimuth, pulling or pushing the needle, or mental arithmetic.  All you need is to fly a smart heading.

So here's the lesson: be a little aware of the wind and hold your heading.  But keep in mind that this analysis was for 5 miles or so; you need to be more precise about choosing a heading to cross the ocean.

But you still have to hold the heading.  That's the most important thing, and that's what NDBs teach us.

So what's our heading when we teach instrument flying? It's not the relative bearing stuff on the FAA Knowledge Test, but it can't be just the knob-twisitng and button pushing required to get the FMS to fly the approach.  Where's the judgment?