Co-pilot's sidestick from an Airbus A321.

Control feedback is important to pilots. Through the yoke or stick, we can feel what’s happening with the plane; for example, are we pulling or pushing too hard on the stick? Are the controls mushy, indicating that we’re close to a stall? Are they fluttering, possibly because of misbalanced control surfaces or icing? Are they bouncing round in turbulence? This stuff is so important that sophisticated, fly-by-wire airliners, which have no direct connection from the yoke or stick to the control surfaces, actually simulate some of it.

I do most of my flying as the only pilot in my small plane, so I never considered how important control feedback is for feeling what the other pilot is doing. It turns out that lack of that kind of feedback is one of the major links in the accident chain that brought down Air France 447 in 2009, as described in detail in annotated CVR transcript in this Popular Mechanics article.

Throughout the incident, one of the pilots was panicking and pulling all the way back on his stick. It wasn’t the thunderstorms that brought the plane down, but a mush from 37,500 ft all the way down to sea level. At any point before the last few thousand feed, the jet would have recovered if the pilot had simply released the stick. The psychology of stress and panic are complicated, and I’m not writing this to condemn the poor pilot who did that; however, the point of requiring two people in the cockpit is that if one is incapacitated (whether by panic or illness), the other can take over.

In this case, however, the other pilot wasn’t aware of the problem, because a crucial piece was missing: Airbus designed the A330 with asynchronous control sticks — that means that the pilot in the left seat had no easy way of knowing that the pilot in the right seat was holding the stick all the way back, because the other stick remained in a neutral position. At one point, when he did push his stick forward, the Airbus simply averaged the two inputs — to me, that makes about as much sense to as averaging two wheels on a ship, when one wheelman turns 15 degrees to the left of an iceberg, and the other turns 15 degrees to the right.

There were many other causes in the accident chain, of course (for example, the pitot tubes initially iced over, so the flight computer didn’t have good airspeed readings, and it shut off the system meant to prevent pilots from stalling the plane), but the glaring one seems to be Airbus removing, by design, a cockpit-resource-management tool that almost every other type of aircraft has: the ability to see and feel the other pilot’s control inputs.

I’d be interested to hear from anyone who knows Airbus’s reasoning behind designing asynchronous control sticks and (in some control regimes) averaging the inputs.

(Photo credit: Flying China Man blog)

You can have enough time to fly, or enough money to fly, but not both.

Changes in work

Consulting work has been wonderful crazy busy for me and pretty-much every other consultant I know, whether in IT, business, aid work, development, or what-have-you. Canada barely got brushed by the recession that devastated the US in 2008-09 (with record-high property prices in most of our big cities, it may be that our bubble just hasn’t burst yet). The problem is that my customers are mostly in Canada and the UK now — the US economy isn’t so great, remember — so I don’t have work excuses to fly down to Boston, NYC, Washington, etc. like I used to.

Recency Lost

I’ve never taken such a long break before. After six months away from flying, bad things start to happen:

• You can’t fly IFR, because you have to have done six hours real/simulated IMC and six approaches within the last six months (the infamous 6-6-6).
• You can’t carry passengers at night, because you have to have done five takeoffs and landings at night within the last six months.
• You can’t carry passengers during the day, even other pilots, because you have to have done five takeoffs and landings (any time) within the last six months.

Recency Regained (part 1)

So now it’s a matter of crawling my way out of the hole, milestone by milestone, until I can get back to my regular IFR/cross-country kind of flying. I started by calling my flying buddy Mike Hopkinson and asking for an intervention, and he complied by texting me last Saturday to remind me to get to the @#$% airport and then meeting me after his shift on dispatch. I uncovered the plane, made some stupid mistakes trying to start it (yes, it does help to check the fuel cutoff), recharged the battery that I drained, then did my normal post-maintenance checks:

• Long, thorough preflight, including control movements.
• Confirm that the engine compartment is clean (I even took a picture for before/after comparison).
• Do a 5-minute lean ground-runup, and watch for anomalies.
• Do a high-speed taxi down the runway to check what happens as the controls become effective.
• Shut down and search for stains, leaks, cracks, or anything else that shouts DON’T FLY!
• Start up again and fly circuits, always within gliding distance of the runway.
• Shut down and repeat the stain/leak/crack checks.

The plane passed with flying colours (so to speak — my paint scheme is drab), and I managed five touch-and-go landings in a crisp cross-wind, progressing from “what-the-hell-was-that?” on the first to “now I’ll gently lower the nosewheel exactly on the centreline” on the fifth. It turns out the the rules make sense — 5 really is the magic number, and now I was legal to carry passengers in day VFR.

Victory lap to Toronto and back

As soon as I was done that and mentioned it on Facebook, my older daughter asked me if I could pick her up in Toronto and bring her home for a break from her studies at U of T. It grated to be a VFR-only pilot, but the weather on Friday co-operated beautifully, so I managed to get 4 hours of proper cross-country flight time (3.6 hours air time) on a beautiful spring day. I was ridiculously nervous beforehand, but on the day of, my flying and radio work through the busy Ottawa and Toronto airspace was fine, and the flight was as boring and uneventful as I want all my flights to be (my rule is that the only excitement in flying should come from the scenery outside the window). No new milestones from that flight, but I feel confident now to start back into the night and IFR work — 2 more milestones to go.

A typical ELT.

I had an unusual experience this morning flying home from Sault Ste. Marie to Ottawa with my dog, Paisley. Three separate high-altitude jets heard an ELT signal at a point along my route about 20 miles south of North Bay.

The most common types of ELTs will start sending a signal automatically when they experience high G-forces (such as when an aircraft has made a forced landing or crashed), but most often, an ELT signal just means that a boat has bumped against a floatplane, someone forgot to turn off an ELT in the back of a courier truck being shipped back for a battery changed, etc. etc. It used to be that satellites monitored the emergency frequency 121.5 MHz, but they don’t do so any more (they’ve switched to the new 406 MHz frequency), so it’s up to us pilots to listen and report.

Joining the search

There was already a plane or helicopter with the callsign “Rescue 333” doing a lower-level visual and ELT search. It also happened, however, that Toronto Centre was already talking with two small planes at lower altitudes who would both be crossing the location where the signal was received: me from west to east, and a Cessna 172 from south to north. Centre enlisted both of us to widen the radio-search area by tuning in 121.5 MHz on our second radios and listening for any signal as we passed through. Every 3-5 minutes, I turned off the squelch on my radio: that creates horrendous static, but it also lets me hear very faint signals.

None of us, including the search aircraft and other high-altitude airliners, heard anything more. I looked as carefully as I could, but from 9,500 ft in a low-wing aircraft it’s hard to see much detail, and I didn’t want to go lower and interfere with the Rescue aircraft’s search patterns.

Ping!

About 20 minutes later, over Algonquin Park, however, I did suddenly hear one very brief ELT wave (the part that sounds like a sheet of metal pinging), as if someone had switched an ELT on to test it then turned it off a second later. It was 8 minutes after the hour. All ELT testing is supposed to take place during the first five minutes of each hour, but I suspect that some floatplane pilot somewhere in the park just had a slow watch. Still, I reported what I heard to Centre, and they passed it on to the Rescue plane.

How to help

The odds are very high that this was a false alarm — most search and rescue calls are — but it was still sobering to think of a brother or sister pilot down there somewhere under the tree canopy, in desperate need of help. If you’re a pilot with two radios, I strongly suggest that you keep the second tuned to 121.5 MHz whenever you don’t need it to get weather, etc. — some day you might actually make a difference.

Where do they look?

Point A to Point B

Let’s say that you filed your flight plan from Point A to Point B and that there are no obvious barriers (mountains, large bodies of water, restricted airspace).  I’m not sure exactly how SAR works, but let’s assume that they’ll search within a 5 nm radius of each airport, and along your route of flight from those circles, allowing for a 10° drift in either direction.  The search area appears in grey below (not to scale):

Point A to Point B search area.

For a 300 nm flight, using my very rusty high school geometry skills, I calculate the search area to be nearly 9,500 nm²!  That’s a lot of forest to search for a little plane with an unconscious pilot strapped into the front seat.

Point A to Point B (via Point C)

Fortunately, we can improve our chances quite a bit, simply by providing a checkpoint halfway through the flight:

Point A to Point B via Point C search area.

Now that Search and Rescue knows that you intended to fly via Point C (represented by the yellow triangle), rather than, say, flying to the north for better scenery, or to the south to buzz your buddy’s cottage, they can anchor their drift lines at an additional point, cutting out a lot of search area (I’m still allowing a 5 nm circle of ambiguity around the checkpoint).  I calculate that the search area is now reduced from nearly 9,500 nm² to just under 5,500 nm², a reduction of over 40% simply by adding one checkpoint in the flight plan.

Point C to Point B

But wait, there’s more!  Let’s say that, when you were over Point C (your midway checkpoint), you actually made a call to flight services and gave a position report.  Now there’s no need to search anything before Point C, because they know you passed it before you made your forced landing:

Search area after making a radio call over Point C.

You’ve reduced the search area from the original 9,500 nm² down to about 2,800 nm² — still a lot, but I’d bet more on your chances now.  Simply choosing a checkpoint, and making a radio call over it, can make a huge difference.

Caveats, etc.

Flight following is an even better option, since ATC will know much more precisely where you disappeared from the radar.  Unfortunately, flight following at lower altitudes is available only in highly populated areas.  In Canada,  below 10,000 ft you’re beyond radar coverage for much of the southern area of the country, not to mention the vast north.  Even in the US, radar coverage can be spotty — I’ve fallen below radar at 7,000 ft when flying IFR in both Maine and New Hampshire, for example, reverting to non-radar reporting procedures.

A checkpoint makes sense only if you can report over it, which means that you need to be able to reach flight services, which is non-trivial at low altitudes away from populated areas.  I plan my VFR flights with checkpoints that I know will be in range of an RCO or DRCO; if that doesn’t work, you can always try relaying your position report through another aircraft (air-to-air range is much further than air-to-ground).

Another interesting option for people who fly a lot in remote areas is the Spot personal messenger (site), which updates your position continuously via GPS and satellite and displays it on a web site. I haven’t tried it yet, but the price looks reasonable. It would be critical to mention the Spot in the remarks for your VFR flight plan, so that SAR would know to go to the site and check your flight path.

Last week I wrote about the summer flying dilemma, the challenge of balancing passenger comfort, speed, and the rising summer cumulus clouds. Pilots face a whole series of dilemmas like that with every flight — in fact, it’s probably fair to say that flying is more about finding balanced solutions to different dilemmas than it is managing the yoke, rudder, and throttle.

Weight vs. time

This week, I had to deal with a different challenge, the fuel and weight dilemma. A few years ago, I never had to worry about this one: my Piper Warrior can carry 665 lb when the fuel tanks are full, and that was enough for my whole family of four, our dog, and luggage, so filling the tanks for every flight was a no-brainer. Now my girls have grown into intelligent and beautiful women, and I’ve gone the other direction by putting on a few extra pounds, so I have to start thinking about fuel and weight.

Trade-offs

Here are the kind of options I consider before every flight:

Additional considerations

You have to consider a few points when looking over these choices:

• There is some stuff I have to carry in the plane — charts, Pilot Operating Handbook, tow bar, a couple of quarts of oil, some emergency supplies, cover, etc. That stuff weighs at least 20 lb even when I strip it to its bare minimum, so out of a 40 lb baggage load, only 20 lb would be available for bags — that’s 5 lb each for four people.
• People rarely fly naked (which is a good thing, given the average pilot’s age). Even in the summer, clothes and shoes add an extra 5 lb per person, and people tend to carry purses or backpacks with food, books to read, etc., adding an extra 5 lb each.
• For safety, you need to have a fuel reserve. The minimum required is 30 minutes (45 minutes at night) VFR, and more, including fuel to get to an alternate airport, IFR. For me, 2:40 fuel means at most a 1:40 flight VFR, and maybe 1:00 IFR, which isn’t enough to be useful for cross-country flying (a fuel stop every hour makes for a long flight).

There are yet other considerations. A lighter plane will take off in a shorter distance, climb faster, and cruise faster (I see a 5-10 knot difference in cruise speed between just me on board lightly loaded, and wallowing along at maximum gross weight). It costs time and money to tanker extra fuel along, and could also be a safety issue on a short runway, if there are trees looming at the other end.

The PIC’s burden

As with the summer flying weather dilemma, there’s no automatic right answer for this: that’s why you’re pilot-in-command — you have to think hard, and trade off one kind of safety or efficiency for another, hoping you’ve found the best balance. Airline pilots worry about this even more than small-plane pilots do: someone once pointed out to me that for a 747, every takeoff is a short-field takeoff.

This week, I got my whole family to Toronto and back safely, IFR both ways, landing with about 1:30 left in the tanks. It was nerve-wracking having to do such careful fuel management, and telling Porter FBO in Toronto the exact number of litres to put in each tank (rather than “fill ‘er up”), but it worked. My fuel burn was exactly as planned, we flew a bit faster for being light, and we got four adults back and forth in a 160 hp airplane.

Next week, it’s just me and the my 12-year-old Border Collie flying to the Soo. I’ve filled the tanks right up to the brim, and will fly the whole thing non-stop for four hours, IFR or VFR, while the old girl naps in the back seat.

On a typical VFR summer day in southern Canada or the northern US — the kind that the weather(wo)men call “a mixture of sun and cloud” with a decent amount of humidity — swirling columns of heated air, called thermals, start rising in the morning, and get higher and higher into the late afternoon as the sun warms the ground.

Clouds are the dividing line

At the top of those thermal columns, like scoops of ice cream on top of invisible cones, scattered cumulus clouds form. At first, those clouds might be only 1,000 feet thick and maybe 2,000 ft above the ground, but as the columns rise, the clouds get both higher and thicker, and some of them may develop into storm clouds during the afternoon (giving your typical cottage thunderstorm or 20-minute shower).

Underneath those clouds the horizontal winds are weaker, but the vertical motion of the thermals can cause a lot of turbulence — while the air may be smooth at breakfast, by early afternoon it can be rough enough to have passengers grabbing for the vomit bags, while the pilot feels like he or she is in a wrestling match with the yoke and rudder pedals.

Above the clouds, the air is normally smooth, but the winds are much stronger.

So where do you fly?

Flying eastbound

Eastbound, the best place to be is above the clouds if you (and your plane) can manage it. Prevailing summer upper winds in this part of the world are from the west or southwest, so on a good summer’s early afternoon you can get a 25-30 knot tailwind and smooth air at 9,500 ft eastbound. What’s not to love?

The problem is that the tops of those cumulus clouds keep getting higher and higher until almost supper time. In Canada, you need oxygen to fly above 10,000 ft for more than 30 minutes, and the tops of summer cumulus clouds could hit that height anywhere between 11:00 am and 2:00 pm, even ignoring isolated storm clouds. As the day goes on, without oxygen and a powerful plane, you hit a point that you can’t outclimb the clouds any more, and you’re forced to descend either into them (IFR) or under them (VFR), giving up some of the tailwind, and throwing you and your passengers back into the turbulence.

The good news is that, with the extra speed from the tailwind, you might be able to finish your trip before that happens. It’s possible to fly a Piper Warrior from Ottawa to Charlottetown in well under four hours with a good tailwind, so you can leave early in the morning, fly non-stop, and land for lunch before the clouds (and turbulence) outclimb you. Your passengers will have to put up with a bumpy descent and landing, but the rest of the flight will be smooth.

Flying westbound

Flying westbound, into the wind, you’re faced with a lose-lose proposition:

1. Fly below the clouds to get weaker headwinds, but subject your passengers to stomach-emptying turbulence as the day goes on.
2. Fly above the clouds to get smooth air, and watch your airspeed drop to (or below) automobile speeds.

First thing in the morning, you have a good chance to find both smooth air and weak headwinds at low altitudes (e.g. 2,500 ft), but that won’t last until lunch. As the cumulus clouds start to form, you’ll have to pick sides, diving down into the turbulence or climbing up into the headwinds. And to throw extra salt into your wounds, the headwinds slow you down and make it more likely that you’ll have to make a fuel stop, slowing you down even more, and pushing your flight later and later into the afternoon and the worst of the turbulence and/or headwinds. If you do decide to go above, even that won’t last: in a low-end plane without oxygen, you’ll eventually have to go down below and start banging your passengers’ (and your) heads against the ceiling of your plane.

Flying north- or southbound

In Canada, it’s safe to say that most cross-country flying is on the east-west axis, since Canadian cities are spread out in a thin line along the US border. In the US, on the other hand, you’re just as likely to be flying on the north-south axis (or something in-between). In that case, if the upper winds are from the southwest (typical in summer, at least east of the Rockies), then flying southbound will be a lot like flying westbound, but with less severe headwinds, while flying northbound will be like flying eastbound, but with less beneficial tailwinds.

Strategies

I fly some passengers who suffer from motion sickness, so I spend a lot of time planning around this stuff. In general, flying eastbound, I’ll go up above and accept a longer flight to get smooth air, since my passengers would rather be comfortable for 5-6 hours than violently ill for 3-5. Still, I can’t protect them from the turbulence when I have to land (fuel stop or destination).

Of course, I can’t stay above all day when the sun’s out — today, coming back from Charlottetown, I started at 2,500 ft in air so smooth you could do calligraphy in the back seat, but by late morning ended up at 10,500 ft over Quebec’s Eastern Townships trying to stay above the rising cumulus layer. Once my 30 minutes above 10,000 ft were up, I had to descend and subject my passengers to 40 minutes of turbulence below the clouds at the end of the flight.

If your schedule can handle it, there are some workarounds to minimize the damage:

• Start as early in the morning as possible, if there’s no fog or low stratus. Some pilots I’ve met like to be preflighting as the sun rises. If your family is willing to wake up at 5:00 or 5:30 am, go for it, and the flight will be more fun (plus you’ll have more of the day at your destination).
• If your route follows a shoreline or a wide river, fly over water instead of land (staying within gliding distance). Water tends to be cooler in the summer, and doesn’t throw up thermals as much. Often, the best break through a line of thunderstorms will be over water, but don’t let yourself get pushed offshore.
• Fly in the evening, especially if you have a well-lit destination for landing after dark. Summer evenings are long-ish, and the air tends to start calming down again around 7:00 pm, give or take, as the ground cools, the thermals die out, and the afternoon cumulus clouds flatten out and disperse.
• Split up the trip. Instead of two three-hour legs with a fuel stop on your westbound flight, fly halfway, land before lunch, and spend the day. The next morning (or that evening), fly the second leg in smooth air again. For example, I could have flown from Charlottetown to Quebec City, spent an afternoon and evening there, and then flown on to Ottawa the next morning.
• Stay on top as long as possible, and descend quickly at your destination. For passengers who suffer from motion sickness, popping ears are far preferable to a long time down low in turbulence.
• Take up soaring. Glider pilots love afternoon thermals, because the rising air lets them stay up almost indefinitely — that’s why you’ll often see gliders happy circling under a cumulus cloud, while pilots of powered planes are impatient to be back on the ground.

In the end, though, turbulence and headwinds are facts of summer flying, and not every day is a typical day — you could drag everyone out of bed at 4:30 or 5:00 am and still be in bumps the whole flight, and you could even end up with headwinds in both directions. Don’t pressure yourself to make the flight perfect and control things outside your control: just do your best with whatever you have to work with.

1. Expand my horizons. This year, I pushed my southern limit a bit by flying all the way south to Dulles Airport. My other extremes flying from Ottawa are Sault Ste. Marie to the west, Baie Comeau to the north, and Port Hawkesbury (Cape Breton) to the east. Maybe this is the summer I’ll fly up to Hudson Bay, just for the hell of it.

2. Seriously consider taking on one or two partners for my Warrior. It’s a lot of work taking care of an airplane, and it could be fun to share.

3. Ace my IFR renewal and my medical. I did my last medical when I was 39, so it was good for five years. My last IFR flight test was in 2007, and was good for two years. They both come due this summer/fall, and the medical will be every two years from now on (since I’m over 40). Recurrent training will help with the first one, and exercise and healthy eating will help with the second.

4. Don’t get stressed. After around 700 hours, I still get stressed (as in can’t-eat, rush-to-the-bathroom stressed) before almost every flight, especially if I have passengers. Some time between starting the pre-flight and starting the engine, it all melts away. I’d like to enjoy the anticipation of the flight as well as the actual flying.

5. Use the blog test for all my flying decisions. Would I be comfortable blogging tomorrow about my decisions and actions today, without making any excuses, modifications, or omissions? If not, time to come up with a new plan.

Support and encouragement from my friends reading this will be very welcome.

On the street

They hooked up London cabbies to an fMRI machine, and observed which parts of their brain were active during different tasks while driving around London in a simulator. Here’s what they found:

Hippocamus

Initial route planning

Retroslenial cortex

Tracking the route in progress (waypoints, etc.)

Anterior prefrontal cortex

Planning diversions during the trip

Right lateral prefrontal cortex

Hazard detection (closed streets, etc.)

Medial prefrontal cortex

Tracking distance to destination

For example, as the cabbies got closer to their (virtual) destinations, the medial prefrontal cortex lit up more and more, like a DME counting down the distance to a VOR. Different parts of their brains performed social tasks like worrying about passengers.

In the air

These map very closely to the tasks a pilot performs, so it’s possible we’d see the same thing if a pilot in a simulator were hooked up to an fMRI: the right lateral prefrontal cortex would light up when watching for traffic or looking at bad weather ahead, the medial prefrontal cortex would show more and more activity as the pilot approached destination, etc.

It makes sense, then, that different people would show different relative strengths based on brain development — some might be very good at planning a route, but lousy at diversions; other people might hate planning, but be great at responding to unexpected problems en route. It’s a good argument against one size fits all for flight instruction.

Or then again, maybe flying is different. If anyone is looking for a great excuse to fit aviation into a grad school research project, here’s your chance …

The incident

I flew through some light snow showers on my way to Kingston with my daughter this morning, so I turned on the pitot heat just before joining the circuit to make sure the pitot blade was clear. At the end of the downwind leg I slowed the engine, reduced power, dropped flaps, verified 70-80 knot airspeed, turned a tight base over the icy water of Lake Ontario, then looked again at the airspeed indicator (ASI).

35 knots. Way below stall speed.

But the plane was flying fine. The nose wasn’t high, the controls weren’t mushy, the stall buzzer wasn’t blaring, the wings weren’t buffeting, and most importantly, the ice floes weren’t spinning and getting larger in the windshield. I gently pushed the nose down enough to speed up 5-10 knots, but still the needle didn’t move. I checked the altimeter and it was behaving properly, showing a slow descent towards field elevation. That meant a pitot failure.

The trickiest part was the turn to final, almost immediately after the failure, when I’d barely had time to process it — it’s easy to lose airspeed in a turn, even with a functioning ASI. After that, it was pretty much a normal approach and landing (no point declaring an emergency when the runway is less than a minute away). The ASI flickered back to life on short final to show that I was 5-10 knots above my normal approach speed. It froze again at some point during the flare and landing (I don’t look at the panel once I’m past the airport fence), then gradually climbed to 90 knots as I taxied in to park the plane.

So now what? The plane is still a useful tool, especially for business and family trips, and I do enjoy the challenge and surprises of Hope Air flights, but there’s nothing exciting about pushing the throttle forward and leaving the ground any more — it feels almost exactly the same as backing the minivan out of the driveway. Here are some of my options:

• learn to fly a taildragger (but once I’ve learned, there’s none that I can fly)
• learn to fly a multiengine plane and get my multi-IFR (ditto)
• learn to fly a floatplane (ditto)
• learn to fly a helicopter (ditto)
• sell my Warrior and buy a share with some partners in a high performance plane like a Saratoga (and pay a lot more for gas)
• upgrade to a commercial license (just for the challenge — I don’t plan to leave my day job)

I know I don’t want to instruct — I used to be a university professor, and while I loved teaching, I don’t want to go back. Any suggestions? What have other people done? After my kids leave for university in a few years, I could spend a couple of months flying around the U.S. and Canada coast to coast, but that’s not practical with the demands of caring for kids in high school and middle school.

• Off the beaten path
• Nec temere nec timide (I also have a blog with a Latin name, Quoderat — so there!)
• pilot in training
• Sulako’s blog

On the way home from Teterboro Airport after a March break family trip to New York City, after 2.6 flying hours mostly in IMC with light to moderate turbulence. My older daughter in the copilot seat whipped out her camera and took this shot on short final on the ILS 32 at Ottawa/Macdonald-Cartier Airport.

We had been racing down the approach at 110 knots to get in ahead of an approaching level 2 cell (though we’d been through a few of those already), and tower turned the lights up to maximum to make sure we wouldn’t have to go missed. The airport was reporting RVR 4000 ft (in other words, you could see less than a mile forward on the runway), but the approach lights were bright enough that I made them out at 800 AGL and started to slow the plane. The runway itself popped into view at 600 AGL, and my daughter took this picture a bit after that.

As ominous as this looks, it would not count as anywhere near a low approach for an ILS. However, it’s worth mentioning that Ottawa was forecasting 3,000 ft ceilings and 5 miles visibility when we left Teterboro.

A lot of the stuff here, though — like the checklists and mixture management, and the caution not to ride the brakes while taxiing — will still be familiar to a student pilot in 2007.

Remember which way the wind’s blowing.

Even if there are blue skies everywhere else, a cold wind will often pull streamers of lake effect weather off the Great Lakes, so if you’re downwind from (say) Lake Ontario or Georgian Bay, expect long streams of clouds with significant icing, snow, and low IFR conditions underneath, topping out somewhere around 6,000-8,000 feet (unless they hit hills like the Adirondacks, in which case the tops can shoot way, way up). This evening, the new photo of the day at Wikipedia (originally from NASA) is a beautiful colour satellite photo of streamers coming off the lakes (click on it for a much larger version), caused by what I’d guess is a low-pressure system centered in northern Quebec a bit east of James Bay. Notice the long fingers reaching out into central Ontario (off Lake Huron and Georgian Bay), Michigan, Ohio, and upstate New York.

My normal flying route from Ottawa to Boston or New York takes me across the St. Lawrence River, and depending on the wind, I can usually expect to have to overfly lake effect weather somewhere along my route. If the wind is from the west, I can expect to find it over Watertown, NY and the relatively flat area of NY state, possibly as far as Lake Champlain; if the wind is from the southwest, it will blow straight up the St. Lawrence towards Montreal, possibly boxing in Ottawa as well; if the wind is from the northwest, it will head towards the Adirondacks, and may throw up cloud too high for me to fly over; it will also blow off Georgian Bay across my route from Ottawa to Toronto, so even if the TAFs are CAVU for Ottawa and Toronto, it might be unflyable in between.

The trick in every case is to make sure that I can stay above the cloud, because it’s pretty nasty underneath it (not to mention inside). For emergencies, as you can see in the photo, there are usually clear spots between the streams for an emergency descent. If the streamers are over your point of departure or destination, on the other hand, forget about it. People on the U.S. side have it a lot harder in the winter, since the cold winter winds usually come from the northwest and blow over them on the southeast sides of the lakes. Aside from central Ontario near Georgian Bay, the Canadian side is usually clearer.

(Source: NASA Astronomy Picture of the Day.)

I flew from Ottawa to Sault Ste. Marie yesterday, then flew a Hope Air patient from the Soo to Kingston this morning, before making the short hop back to Ottawa in the afternoon.

Westbound

Flying westbound to the Soo is a slow prospect in the Warrior. The trip is a bit under 400 nm, but due to the headwinds, it generally takes around four hours. This time, it was all IFR, but at least it was smooth, the icing levels were well above me, and all the thunderstorm activity was about 50 miles to the north of my route. Everytime Toronto Centre gave me a new altimeter setting, it was significantly lower, sometimes requiring an altitude change of a couple of hundred feet, as I flew towards the big low parked over Lake Superior and the nasty weather around it. The IMC on my route from Ottawa, however, had a different cause: the tropical storms and hurricanes had pushed a huge amount of hot, humid air north above them, leaving a stationary front a bit south of James Bay (north of my route). Because the front was there for so long, what had been supposed to be nice and clear became cloudy and hazy, plunging us into unseasonable IMC of the summer variety.

Cold front

A cold front was be blowing through overnight, and I know that I’d have to fly through that cold front from the back side in the morning. For those who don’t fly, you need to understand that cold fonts suck — they mean lots of rain, thunderstorms, turbulence, fog, cloud, and just about anything else you don’t like (in the summer, they even bring tornados). Lying awake in my hotel room at 5:00 am I heard the rain pounding outside, and knew that the front was on its way through.

When I walked out of my hotel in the morning, I found an airplane sitting in the parking lot right outside the door. I wasn’t sure whether to take this as an omen of good luck (the flight will go fine) or a warning (it’s a good day to tow your plane on a trailer) — I guess superstition is too complicated. Here’s a second cellphone photo of the plane, this time from behind, the way I saw it when I walked outside.

Eastbound

IMC and thunderstorms are a bad combination, because when you’re in cloud, there’s no way to see a storm coming. Before I left, weather radar showed that all the activity was well north of my route, though the GFA called for isolated thunderstorms all over ahead of the cold front. I evaluated the situation, and decided to list every way I had to avoid storms:

1. try to stay out of cloud so I can see buildups coming
2. use my Stormscope to watch for lightning strikes
3. ask Flight Services for regular updates from ground-based radar

None of these is 100% reliable: between cloud layers (and the layers go very high ahead of a cold front), it’s often hard to make out buildups ahead, since everything fades to white; the Stormscope is a very blunt instrument and misses intense weather that doesn’t happen to produce lightning; and Flight Services is looking at outdated and and low-res lightning and radar pictures, when I can reach them by radio at all. As a result, I made a couple of rules for my flight:

• At least two of these three methods had to be available to me at every point during the flight, or else I’d abort and land at the nearest airport.
• All available methods had to show a wide, clear path with no storms moving towards it.

It turned out a bit bumpy, with a lot of rain, but nothing more disturbing. Three hours of hand-flown IMC has left me a bit tired now, though. As usual, the weather cleared up just before my destination, so I didn’t get to log an approach.

Northbound

With the tailwind from the southwest, I flew into Ottawa like a rocket: the whole flight from Kingston took about a half hour, and I made it home before the front hit Eastern Ontario. Unfortunately, I then had to wait 35 minutes for a cab to take me the last 10 km home. The wind’s starting to shake the leaves outside my window now, so I guess the weather’s on it’s way in our game of leapfrog.

Recency

Pilots have to do a certain amount of flying in real or similated instrument meteorological conditions (IMC) to keep their instrument ratings current (otherwise, they’re allowed to fly only in visual conditions). In Canada, the recency requirements include six hours and six approaches to minima.

Air mattresses and sailors

Note the phrase real or simulated IMC. Simulated IMC — flying wearing foggles or a hood to block the view out the window — is one of the sad jokes of aviation, since it simulates flying in real instrument conditions about as well as floating on an air mattress in a hotel pool simulates sailing in a storm on the high seas. While flying six approaches wearing foggles (with a safety pilot) meets the legal requirements for IFR recency, I don’t think it does much for actual flight safety, so I use it only as a last resort, mainly in the winter when IMC near the ground almost always comes with icing. As a result, on the rare days when I have no meetings booked with customers, no family committments, and beautiful low rain arrives without any thunderstorms or icing, I rush to the airport, get soaked preflighting the plane, ignore the people standing inside shaking their heads with pity and disbelief, and take off into the clouds.

Getting started

There hasn’t been a lot of rain this summer, but I did manage approaches in IMC in July on actual trips — one into Boston/Norwood, and one into Toronto/City Centre — so if I could manage a quick four approaches today, I’d be current until mid-January (when I’ll probably have to use the hated foggles). I called flight services, and they confirmed with ATC that a mid-day training flight would be OK. At the airport, I fueled the plane (full tanks are always a good idea in low IMC, since diversions can come unexpectedly), holding one hand over the tank opening to keep the rain out, then took off, and within a couple of minutes, there was nothing but white outside my window.

Smith’s Falls NDB 06 full procedure

I’ve developed a nice circuit of approaches around Ottawa. I started by flying to Smith’s Falls/Montague for the full procedure NDB 06 — that has me flying directly to an NDB (an AM radio navigation aid), flying away from the runway for two minutes, doing a funny kind of loop, coming back into to the navaid, then continuing at a pre-determined altitude until I either see the runway or run out of time. I was in and out of cloud bases at the minimum descent altitude of 980 ft MSL (564 ft AGL), but I did see the runway in time that I probably could have landed with a fast dive and some borderline aerobatics. Fortunately, that wasn’t my plan today, so I started climbing again for the missed approach and called back in to Ottawa Terminal.

Ottawa/Carp VOR/DME B with 21 DME arc transition

The second approach was the VOR/DME B (B is pronounced “bravo”) approach into Ottawa/Carp. For this one, instead of a full procedure (flying away from the airport, then reversing and coming in), I flew something called a DME arc, which is my very favourite IFR procedure. DME is an old-fashioned (pre-GPS) instrument that tells how far my plane is away from a UHF transmitter: in this case, from the Ottawa VOR (an FM radio navaid) on the Quebec side of the Ottawa River. To fly a DME arc, you simply turn 90 degrees from the DME at a pre-determined distance, and adjust your course so that the distance on the DME stays the same. If you flew long enough, you’d do a complete circle around the DME transmitter. There are all kinds of convoluted procedures for flying a DME arc, including messing with VOR radials, etc., but mine is easy: just turn the plane a few degrees away from the DME if the distance is getting too low, or a few degrees towards the DME if the distance is getting too high. It works a charm, and requires almost no work. I flew the 21 nm DME arc until I intersected a VOR radial that would take me over the Carp airport from the south, then followed the radial, lowering my altitude in steps: down to 1,400 ft once I was on the inbound radial, 900 ft (518 ft AGL) at 11 DME from the VOR (1.8 DME from the airport), and then climbing back up to 2,900 ft at 9.2 DME when I was over the airport. I saw the Carp airport at the last second, and again, could have made it in with some aerobatics, but this was a circling approach (I wasn’t lined up with a runway), and it might have been a bit too exciting. More on circling approaches later. Note also that at both airports, I was barely able to see the runway at around 500 ft AGL.

Ottawa/Gatineau VOR/DME 09

The third approach in my circuit is a straight-in VOR/DME 09 approach for Ottawa/Gatineau, on the Quebec side of the Ottawa River. The transition from the Carp approach is trivial: the Carp missed approach heads for the Ottawa VOR, and the Gatineau approach starts at the VOR, so just … fly to the VOR. For the Gatineau approach, I fly away from VOR on a predetermined radial, again, with a series of step-downs at various DME distances: 2,900 ft to the VOR, 2,400 ft until 5 DME, 1,300 ft until 11 DME, then 760 ft (551 ft AGL) until 14.3 DME, when I’m right over the airport and begin the missed approach. The weather was a little better here, and I broke out of the cloud bases around 1,200 ft MSL (1,000 ft AGL) and was able to see the runway clearly. That led me to think that I could expect about the same ceiling across the river at Ottawa/Macdonald-Cartier.

Ottawa ILS 07 circling 04 … no, straight-in 07 … no, circling to a taxiway … I mean 04

My last approach was back into Ottawa. All of the previous approaches were non-precision: the navigation aids guided me to the airport horizontally, but did not guide my altitude (except through step-down fixes). Ottawa has two more advanced navaids called ILS, that can provide a precision approach: not only does it help me stear left or right, but it tells me precisely what altitude I should be at during each stage of my descent. I have to fly around to the west side of the airport to get in line for the ILS 07, which I’ll be sharing with big transport jets. ATC tells me that there are three big jets on the way in, so I have a couple of choices: fly out 15 miles and have a nice, easy approach behind them, or turn in tight almost right over the final approach fix and rush down ahead of the jets. When I was an IFR student, I would always have taken the easy one because approaches seemed so hard to set up, but the tight one sounded like better practice (and is more realistic for a small plane at a busy airport), so I took that.

Terminal gave me an immediate turn and descent. I turned sharp (30 deg bank) and dropped the plane at 1,000 fpm to show them that I was capable of taking this approach without messing up their traffic flow. That was good enough, and they kept turning me in and dropping me until I joined the ILS right outside the final fix. I knew enough not to fly an approach at 90 knots in these circumstances, so I pushed the throttle forward and whizzed down the ILS at 120 kt. Satisfied that I wasn’t going to be in the way of the jets, Terminal turned me over to tower.

An ILS approach straight-in to a runway has very low minima: normally, you can fly right to 200 ft above the ground before you have to see the runway. However, I was planning on doing a circling approach to the runway near my parking spot (remember circling approaches from Carp?), so I needed legally to see my runway at 506 feet above the ground (880 ft MSL) and be able to stay at that altitude until I was lined up for my final approach. No dice. While the airport was reporting better conditions over the control tower, there was low cloud over the approach to 07, and I was coming down through 1,200, 1,100, 1,000 ft with no sign of either runway 07 or 04, though I could see a bit of ground straight down. At 950 ft I began a rapid call to the tower asking to cancel the circling and land on 07, but right at that moment — as I passed through 900 ft, 20 ft above circling minima — I broke out and saw the whole airport. In the middle of a sentence, I switched back with “correction: 04 in sight, continuing with circling approach”. Unfortunately, in those 2-3 seconds, I forgot that airports always look different in IMC, and I actually lined up with a taxiway instead of 04 and descended below the 880 ft circling minima. Fortunately, I caught the mistake before tower did, and — with some of the stupid borderline aerobatics I had been smug about avoiding at Smith’s Falls and Carp — sidestepped a half mile to the actual runway 04 and did a smooth landing on the wet pavement.

Lesson learned: circling approaches near minima really are a dumb idea, and it’s hard to make good decisions in a fraction of a second at the end of an approach. When the ceiling was close to circling minima, I should just have planned on the straight-in landing and an extra 10 minutes taxiing.

Update #2: I think the comments are better than my posting, so you might want to skip ahead and read them first (especially the third one).

Update #3: She got back behind the yoke and did great.

After learning the basics of handling an airplane and working in the system, flying is largely about two things: weather, and personal physical limitations. Weather (which I’ve discussed in other postings) is something that you can learn by studying, but personal physical limitations are things that you stumble upon, sometimes fatally, but always unexpectedly.

Finding limitations

Some limitations are obvious — we all know that we cannot live more than a few days without water, that we cannot jump off tall buildings and fly to the ground, and that we will not survive a collision standing in front of a speeding train. Other limitations are trickier, because they are not the same for every person, or change for different times and circumstances. After four hours at 10,000 feet without supplemental oxygen, is your flying impaired? Never? Always? Sometimes? It’s a crap shoot. Put two pilots side by side, and one might be able to solve differential equations in her head, while the other can no longer remember his girlfriend’s last name. Maybe tomorrow it will be the other way around. Ditto for spatial disorientation, fatigue, and the other physical limitations that sometimes sneak up and kill pilots. How long can you hand-fly in IMC and moderate turbulence? Once I did eight hours with no problem; another time, I was physically trembling after four, and had to ground myself and call in help to get a Hope Air patient back home. Good flying is learning to fly safely and well as the pilot you are, not trying to force yourself to become the pilot you wish you were.

Aviatrix and the smoke

All of this is leading up to the latest posting from my favorite blogger, a Canadian commercial pilot who goes by the pseudonym Aviatrix. In her posting Effects of Smoke, she describes the end of a long series of events that led to her decision to resign her job and quit flying altogether being fired from her job and deciding to take time to consider her future in aviation. Flying low, through dense forest-fire smoke (she reported less than a mile visibility) without supplemental oxygen, her sense of judgement became impaired and she had trouble flying the plane, to the point that her first officer had to take over control to make a safe landing.

There’s good news and bad news here. The good news, obviously, is that she lived to blog about her experience. The bad news is that she blames herself and has lost some of her confidence to fly. She never even considers the smoke itself (hypoxia or CO poisoning) as a cause, probably because her first officer seemed unaffected by it, and even now she does not want to admit any physical weakness in the face of the male-dominated world of northern flying. The thing is, personal limitations are kind-of random. For example, consider another male-dominated world, firefighting, where the participants are much more physically fit than the average pilot. Here’s an excerpt from an article about firefighting hazards:

Over 50% of fire-related fatalities are the result of exposure to smoke rather than burns. One of the major contributing factors to mortality and morbidity in fires is hypoxia because of oxygen depletion in the affected atmosphere, leading to loss of physical performance, confusion and inability to escape.

Compare that to Aviatrix’s posting. Sound familiar? Huge, muscular, testosterone-soaked male firefighters exposed to smoke sometimes become as helpless as babies, and their buddies have to pull them out. It’s not a girl thing.

Next steps?

As far as I can tell, without knowing Aviatrix personally, she’s a highly-skilled and experienced pilot who ran hard into a physical limitation for dealing with hypoxia. Does that mean she’s now unsafe to fly? Of course not. If anything, it means that she can become an even better pilot, because she’ll know herself more thoroughly than she did before. She’ll know the early signs of hypoxia from bitter experience, and will (perhaps) grab a small portable oxygen bottle from her flight bag or, if up high and not in smoke, choose a lower altitude before things deteriorate. She’ll cancel a flight that she might have decided to fly before, and will be more confident about her decision. And her passengers will have a much better chance of a safe landing than they would with most of the bold, young male northern pilots who yet to discover how non-negotiable their own personal physical limitations will be.

I think I’m safe in wishing Aviatrix luck from all of her readers. I, for one, would not hesitate to get onto a plane where she was PIC — odds are, she’ll be the safest pilot at any base — and I hope to read more of her flying stories soon.

• Flying four adult-sized people plus luggage plus full fuel in a 160 hp Warrior is legal, but it’s a huge challenge. You have to treat every takeoff as a short-field takeoff, and have to be bang on the numbers to make the thing leave the runway and climb at all on a hot day (I’ve posted before about how flight training fails to prepare pilots for heavily-loaded, underpowered planes). Expect to see frequent negative climb rates above 7,000 ft (4,000 ft if there are any mountain waves): just hold Vy and be patient.
• Deviating around thunderstorms in a slow plane is also a huge challenge because of the distances involved. I guessed wrong and deviated south when all the airliners were deviating north, and ended up giving my family a grand tour of the Eastern townships of Quebec (which we couldn’t actually see, but no matter). I’m looking forward to having radar images available for Canada.
• Bilingualism is a good thing, except in a busy circuit at an uncontrolled Quebec airport. I do speak some French, and I managed to understand that an incoming pilot behind me was bound and determined to land on 05 when I was already in the circuit for 23 and Unicom was insisting over and over that 23 was the preferred runway. Unfortunately, I didn’t understand enough French to realize that he was cutting me off after I switched 180 degrees and joined the downwind for 05 to accomodate him. He was behind me, but when I turned base I saw him right in front of me on a straight-in final (which isn’t even legal for VFR in Canada at an uncontrolled airport). Life’s too short to argue with morons, especially in another language, but I have another good reason to prefer controlled airports. The other pilots at the airport were decent, no matter what language they were speaking.
• With fuel prices so high, I’m glad to fly a plane that doesn’t burn too much of it, even if our trips are sometimes a couple of hours longer. I found some good prices in Quebec, though, sometimes in unexpected places.
• I always worry about emergency landing spots when planning a flight over completely deserted areas, such as the interior of the Gaspé peninsula (where the St. Lawrence River meets the Gulf of the St. Lawrence). No need, really. It turns out that the interior is criss-crossed with, literally, hundreds or thousands of wide logging roads. I imagine that the surfaces are rough, and it might take a while for someone to find us, but even though I flew for over an hour out of range of towns, farms, roads, airports, etc., if I had lost an engine I would always have had a choice of literally dozens of easy, straight landing spots right under the plane. They might not have been nice on the landing gear, but they’d be a lot better than trees or plowed fields.
• It’s very hard to resist cheating when you’re IFR in actual IMC, have a VFR-only GPS and ATC offers you a direct routing that will shave 15 or 20 minutes off your trip (and vectoring isn’t possible, because you’re below radar). I won’t say whether I resisted successfully or not, but if I hadn’t, it would have been easy to verify my position periodically using VOR/DME fixes.
• I have a decent amount of actual IMC now, all of it hand-flown, and much of it in rough conditions. In a simple plane like a Warrior, I don’t think there would be any real benefit in having an autopilot, because the plane is draggy and unresponsive, giving me lots of time to fold maps, talk on the radio, etc. without going off course or tilting the wings. I know that things would be very different in a retractable.
• A long, non-stop flight is always tempting, especially when you’re flying back home westbound and might not have serious headwinds. The upper wind forecasts suggested that I could do the return flight (503 nm) in only 4:10, while my Warrior holds enough fuel to fly over 5:30 lean of peak at 75% power (8.5 gph from 48 gallons usable). When the actual winds indicated 4:30-4:40, however, I decided that I didn’t want to be one of those morons who runs out of fuel, even if the GPS said that I would have minimal legal VFR reserves, so I added a fuel stop. It turned out that I did have enough fuel, but family bladders appreciated the stop all the same.
• Flying an underpowered plane over hills makes me appreciate that the air moves in three dimensions: in addition to headwinds, tailwinds, crosswinds, etc., there’s always an updraft or downdraft. I was high enough to avoid the rough stuff, but at 8,500 feet over the mountains (3-4 thousand feet still make a mountain) I was constantly aware of the gentle waves (5-10 minute cycles) that were adding or subtracting about 5-10 knots of airspeed. I notice that flying over the Adirondacks as well. Some people just hold a constant airspeed and ride up and down on the waves, while others hold constant altitude and let the airspeed climb and drop. I chose the latter, since I was in an area of no radar coverage, and didn’t want to crowd the IFR altitudes.
• I had my first experience flying a significant amount of IFR in class G airspace (green on the map), but it wasn’t actually uncontrolled. Because I was going to be crossing a controlled airway, ATC kept talking to me and never said that I was uncontrolled, even though I was flying no-radar. I could legally have flown about an hour of my route IFR without a clearance, though, as long as I could have crossed the airway VFR.
• My family has almost 4 years experience with the airplane now, and they’re all extremely light packers and unfussy travellers. How many teenaged girls can pack enough for a week in a 10 lb suitcase? That alone is probably enough to justify the expense of flying.

• Almost every landing is really a crosswind landing to some extent, so don’t think of The Crosswind Landing as a special procedure.
• Don’t set up the slip until you’re well below treetop/roof level, and preferably, not until you’re in the flare. The winds at 100 ft (or even 50 ft or 20 ft) may have little or nothing to do with the winds you’ll be landing in. I’ve often had a ferocious crosswind on the turn from base to final that faded to almost nothing by the time I touched down (even the windsock is too high to get the touchdown winds for a 172 or PA-28).
• In the flare (for any landing), think of the controls differently — use the ailerons to slide the plane left/right so that it stays over the centreline, and use the rudder to keep the nose pointing straight forward. Don’t worry about a slip per se — it will just be the natural result of these inputs.
• (Advanced) If the crosswind is very strong, back off on the rudder a bit and flare in a crab, then kick the rudder hard to straighten the nose just before the wheels touch. Don’t try this until you’re pretty comfortable (and probably never in a taildragger, unless you enjoy ground loops).

By the way, I always land with full flaps, even in a strong crosswind. The only time I don’t use flaps is when I’ve seen any trace of icing during flight — if, for some reason, there’s undetected ice on the stabilator/horizontal stabilizer, dropping flaps can cause a tail stall [video], which is almost never recoverable close to the ground.

I spent a bit over an hour today flying a spanking-new Cirrus SR22-GTS, courtesy of Easy Air Share, which is setting up a fractional ownership program here in Ottawa. I flew in the left seat with Floyd, a production test pilot from Cirrus, in the right seat. Here, as a counterpoint to a previous Cirrus posting, are some of my initial impressions:

Best feature:

electric elevator and aileron trim — I used it religiously to relieve control pressures.

Second best feature:

pilot-side door.

Most overrated feature:

the glass panel (it’s cool, but it’s more information than I really want when I’m flying, especially VFR).

Hand flying:

slightly more slippery than a Warrior, but after a bit of practice, it was no problem holding heading and altitude — the stuff people say about needing the AP to fly is BS.

Instrumentation:

I missed the analog altimeter and gyro compass (there was a backup altimeter at the bottom), but I got used to the tape display fast enough.

Landing characteristics:

slightly higher approach speed, but otherwise almost identical to landing a Cherokee or 172.

Power:

I’d like to be able to choose my own RPM on the CS propeller; the engine flew nice and smooth lean of peak, though.

Castering nosewheel:

no big deal, no big problem.

Sidestick:

no big deal, no big problem, except that it requires very small inputs compared to a Cherokee or 172.

Leg room in the back seat:

(tested on the ground) very spacious.

Autopilot:

smooth and well integrated.

Four-point harness:

extremely comfortable.

Sound system:

nice, but not a big deal for me.

Chute:

no opportunity to observe.

Would I mind owning a share of an SR22? Not at all. It feels like a great machine for long cross-country trips, and I know my family would love every second in it. But I couldn’t help thinking how much more fun that same flight would have been in my trusty old Warrior, slow speed, and all. I love working on computers; I don’t know if I love flying them.