I Would Have Gone With Stephen Hawking

Before making an approach to an airport, every pilot wants to know the wind strength and direction, what clouds they will encounter, the altimeter setting, and what runway is in use. At some airports the pilot figures out this information herself by flying over the field. At others a flight service specialist simply recites the information to each arriving pilot. At busy airports, a recorded message called ATIS recites the information on a dedicated frequency. The information on the ATIS is typically updated once an hour, or when conditions change. At some airports a controller just reads the information into a recording machine, so you can hear his or her voice, fast or slow, annoyed or cheerful, with occasional minor stumbles and the tower background noise. At others an automated voice reads the ATIS.

Here's a live human reading the ATIS:

Here's the ATIS read by the English-language automated voice Nav Canada uses (in Montréal, the Frech ATIS is on a different frequency):

Today I've just picked up an ATIS that uses an automated voice, and my non-pilot co-worker says, "I like him. He reminds me of David Suzuki." It's pretty obvious that "he" is a robot, and I've never thought of David Suzuki as sounding robotic. For foreigners not familiar with Suzuki, here's a clip. (I ask you to please refrain from commenting here on Malthusian economics, environmental sustainability or Suzuki as a person. The clip is there just so you can hear his voice).

I don't hear the resemblance, except perhaps that he has the earnest careful way of speaking, slowing down to make points. But not more so than any other public speaker. It made me realize, though, that I always assumed the robot was a white guy. I'll try and picture him as different ethnicities now, and wearing different funny hats and ties.

Here's a British automated ATIS voice, not as frightfully British as I was expecting.

Bad Technology Day

I've been struggling with my phone lately, not getting text messages, not having it ring when I'm called. An update to the OS did the trick, even though the OS update notes were all for apps I don't use. I heard someone in the air having an equally frustrating time with technology.

"We're struggling with the FMS right now. Can we have direct BOXON?"

The FMS is the flight management system, the magic box that tells the flight director and the autopilot where the airplane is supposed to go. I don't know the clearance that the pilots were given initially, but they were apparently unable to persuade their FMS to accept the waypoint or procedure, so they chose one it would accept to fly towards it while they continued the battle.

I hear another pilot given a hellish multi-stage missed approach clearance including direct to a random sounding lat-long. It's because there is a large swathe of NOTAMed active military airspace blocking access to the normal airways no doubt. Programming a lat-long into the FMS while on approach sounds fun. Not. I screw it up on the ground from time to time. I have to create a user waypoint for the lat-long, can't enter it in the flight plan or direct-to screen. The pilot reads the clearance all back and life goes on. But a while later she comes back to confirm the lat-long. Yep, it's correct.

"The FMS won't take lat-long. Can we go direct [fix]?" she says. Negative, that's in restricted airspace.

"How about [fix]?" That's approved.

We cancel IFR and slip underneath it all.

LOST

II might as well post this now, because I won't have a chance to update it if the mystery of Malaysian Airlines 370 is solved in the next few days. I don't have any special insights about the disappearance of an entire Boeing 777, but as a pilot I have to wonder about it.

First off, despite what movies like Airport '77 depict, an airplane that crashes into the sea doesn't slip whole and undetected under the waves. With considerable skill, or even questionable skill and some luck, it could alight on the surface of the water without substantial damage. Handled as if it were being landed on land, an airplane can skim over the surface of the water the way a skipped pebble or a snowmobile does. (Unlike the snowmobile, an airplane doesn't immediately sink, because the fuselage is full of air, and buoyant). No sane person would deliberately ditch an aircraft without a substantial emergency, because if something goes slightly wrong, the picture is substantially different. The video below is of a hijacked Ethiopian Airlines aircraft that was out of fuel. The captain was trying to ditch it while fighting with the hijackers. Even that fuselage stayed afloat for a reasonable amount of time, and many of the deaths were not due to crash injuries, but because passengers inflated their life vests before exiting the aircraft, impeding evacuation.

There's a post being shared on Facebook that purports to show the aircraft recovered, but you can tell from the still that it's actually a Lion Air landing accident from last year. Don't click that one because it's probably a virus.

When an airplane comes into contact with it, water is really quite a bit like ground. Do it in landing configuration, descending towards the water at a shallow, about three degree angle, then touching down wings level, slightly nose up and it is quite survivable. Do it at high speed, or inverted, or any other way that doesn't involve control at a low descent rate, and the result is catastrophic. Swiss Air 111 hit the water 20 degree nosedive, and at a bank turn angle of 110 degrees, and disintegrated into millions of pieces. Only one of the passengers was visually identifiable. Alaska Airlines 261 was a similar story. Many of the millions of pieces that make up an airliner and its contents float, so the debris is visible on the surface and drifts ashore. Jet fuel is oily, and floats on water leaving a visible slick. Also both those crews were in contact with air traffic control during the sequence that led to the crash.

When an aircraft suddenly disappears from radar and stops transmitting, one good guess is that it broke up in midair. But when that happens the pieces are in the water in roughly the area where the radar track stopped, and the last week has seen ships, helicopters satellites and and airplanes tasked to search that area with no results. Plus substantial other evidence exists that the airplane did not explode, crash, alight or otherwise make contact with the water in the area they've been looking for it. SATCOM, an automated system on board the aircraft, and one a very diligent Boeing pilot of my acquaintance says he doesn't know how to turn off, continued sending transmissions. It's a passive system that pings about once an hour, sending what's basically a "ready to transmit" signal but no data links were opened, because the companies involved had not subscribed to that level of service from the satellite operator.

Here's what I have gleaned of the timeline. Times are Malaysian local.

• 12:41 a.m. Take off
• 1:07 a.m. The ACARS data transmission system stopped communicating. ACARS is a two-way data transmission system in the cockpit. The pilots can use it to request weather data and the airline will use it to send them connecting flight information, and even to update the flight management system. It automatically transmits some data on phase of flight and things like fuel status and engine performance, so the airline can detect a need for maintenance before it is urgent. It can also be used for communication with ATC, but as far as I know is generally only used with ground controllers to get the complex initial clearance for a flight. It might be used in flight for a complex rerouting, but I don't know. Check the comments to see if one of my more knowledgeable readers does. The ACARS system can be turned off from the cockpit.
• 1:21 a.m. The crew is 'handed-off' by Malaysian ATC. That is they are told to contact the next controller on a specific frequency. They reply normally to this call, implying that they will comply, but do not call the Vietnam ATC.
• 1:21 a.m. The transponder is turned off. The transponder is a simple device that works in concert with ATC radar. Before take-off ATC assigns the airplane a four digit code, which the pilot sets on the transponder. When it receives a ping from radar, it sends back a signal encoded with the aircraft altitude and the four digit code. that the pilot has set on the device. The ATC computers match the code with the aircraft type and destination in its flight plan, plus calculate and display a speed based on consecutive returns. If an aircraft does not have a working transponder, ATC may still see the aircraft on radar, but it is a "primary target." The transponder can be turned off from the cockpit. Transponders malfunction pretty frequently, and an aircraft like the B777 probably has two, for backup and to facilitate changing codes without an interruption in radar track.
• 1:22 a.m. Vietnam ATC expect a check in from the airplane but instead the radar track shows an almost 180 degree turn.
• 1:38 a.m. Civilian radar lose primary target. Without the positive identification of the transponder reply, controllers only know that their radar is hitting something there. Based on the speed and location, they can sometimes deduce that the target is a train, ship, or flock of birds, and they can be pretty confident that a blip going the speed of an airliner, in the location where an airliner transponder just stopped transmitting, is that airliner. Primary and secondary surveillance radars may have different coverage, so an airplane that could normally be tracked by its transponder signal may not be visible as a primary target.
• 2:15 a.m Malaysian military radar detects an unknown aircraft, possibly MH370, on a track that suggests it made another turn.
• 2:30 am Another possible military radar contact with the missing jet. I'm not sure if it shows the jet on the same track or not.
• 8:11 a.m. Final SATCOM ping received from the aircraft.

The aircraft had about eight hours of fuel on board at the speed it was going, and more if they reduced the power settings. I don't know the radar structure of the airspace they were in to say where they could get to without the aircraft being detected by an agency that would recognize and report its significance.

I don't know what it all means. I have a friend who is livid at the apparent lack of cooperation among the various military agencies who almost certainly have more information on the track of the missing airplane, but don't want to divulge their capabilities to one another. He also suggests that if the SATCOM signals were received by multiple satellites their source location could be triangulated. He also implied that a team of children with expertise at Where's Waldo should be put on the case, but I think the former suggestion was the more serious one.

Here is a link that answers some basic questions I didn't think people would ask, but people have some odd questions about airplanes. Someone asked me if this could be a case of pilot incapacitation, with the autopilot or random chance making those turns. I'd say no. The autopilot wouldn't be programmed to make those turns, and if the autopilot wasn't on, I would expect significant altitude changes. It's an odd incapacitation that would cause you to turn off two separate systems, one fourteen minutes before a normal call to ATC.

This one gives a pretty good explanation of tracking technology, and addresses the question of "how can we apply technology so this can't happen, and why haven't we done so already?"

Because I don't like to think of airplanes crashing if it's not necessary, I imagine that the pilots have stolen the aircraft, kidnapping the passengers. Although probably worse. It sounds like some cellphones were still switched on and even logged into social media accounts. Given the attachment some people have for their phones, it seems unlikely that that many people could be taken alive without anyone managing to get a message of some sort away. Despite my recent rewatching of the television Series LOST, I don't believe that the airplane has been spirited away to a mysterious island guarded by a smoke monster.

The facts that there will be wreckage if an airplane crashes, and that they haven't found wreckage, don't mean that the airplane didn't crash. It's a really BIG ocean, and it takes a long time to search. I liked this picture of Malaysian SAR crews. Of course it's probably posed, but I'm always attracted to shots that depict a group of people engrossed in something, not looking at the camera. It suggests expertise, focus, teamwork, and an important task. I also like the simplicity of the way the women's hijabs work with their flight suits.

I'll give the final word to The Onion, a satirical publication that doesn't know the phrase "too soon."

Update: They've pulled some location information out of the satellite data, but don't have enough satellites receiving the signal to resolve the position ambiguity.

Autocow

I went to an agricultural fair recently. You know the sort I mean? With cows and pigs, and games where you try to throw a dart to win a giant stuffed dog, brightly coloured spinny rides with loud music, and cotton candy and barrel racing. I was describing the events of the day to a friend, when I realized I remembered all the animal pictures (you were supposed to "collect" them on a checklist, but I didn't bother) displayed inside a maze made of straw bales, and I could recite biographical information on the various cows I saw. Why did my brain choose to store that? That storage capacity would be much better occupied by information on how to ensure my autopilot doesn't fly me into a tree today. I got it to work well once, but ever since when I have tried to engage it, it is excessively pitchy.

Perhaps if my autopilot had clear operating instructions printed on the side of a cow I would have as firm a grasp of its use. Mind you, if only my autopilot had clear operating instructions, I might have a firm grasp of its use. It has vague operating instructions, without pictures, and leaves the rest to experimentation. For example, I don't really know if there is a tree-seeking mode, or if there is something specific I must do to disable it. That's the sort of thing I'd expect an autopilot instruction manual to be clear about, if I hadn't read this one.

It has approximately five on/off switches. There's a large rocker switch for roll, square push on-push off buttons for heading, altitude, and something else not clearly specified on its switch, but probably pitch, plus a separate wing leveller. That's five, or it's six if you count the red pushbutton on the yoke, which doesn't work quite the way I expected.

Roll is the basic mode. There is a left-right roll knob, like balance on an old stereo system. You're supposed to centre that knob, toggle the heading lock button off, and engage the roll rocker switch. The airplane should now roll up to a 30 degree bank angle in response to turning the roll knob. If you toggle the heading lock on, bank should be restricted to twenty degrees.

Pitch only works when roll is already engaged. You're supposed to disengage the altitude on/off push button and centre the pitch "command disc," a vertically-mounted edge-on wheel with a flat spot that presumably represents the centre position. It tells me "with the airplane in level flight," implying that this isn't something I can do on the ground before takeoff, to rotate the altitude selector DN/UP knob (looks just like the roll knob) until the trim UP/DN indicator is level. The trim UP/DN indicator is a tippy line in a circle. The instructions next say to calibrate the altitude indicator to match the altimeter by rotating the knurled altitude indicator dial. The altitude indicator is a wide vertically mounted edge-on white plastic wheel with numbers on it representing altitudes, and intermediate lines between them. One edge of it has knurled ridges, so you can change which altitude is displayed, but it doesn't go that far. When I was at 8000' it was displaying 13,500' and I could only coax it down to 12,000'. Just two more steps: rotate the altitude selector knob to select the desired altitude and finally push the altitude preselect button to ON.

The pilot can trim the airplane with either the trim crank or the electric trim on the yoke, but trimming is automatic when the pitch section of the autopilot is engaged. There is a pitch trim warning light which should illuminate if the pitch is out of trim for four seconds while the pitch section of the autopilot is engaged. To me this could mean inoperative trim, runaway trim or possibly some balance problem with the trim surface such as ice accretion. You shouldn't use an autopilot in icing conditions.

Wing Leveller: The manufacturer has given this a cutsie name, shared by a passenger briefing system and a kind of torque converter, obscuring for the pilot what this function is supposed to do, but the hints that it is approved for use during take off and landing and is automatically disengaged by selecting the roll rocker switch suggest to me that it's a simple wing leveller. The pilot is advised that to make turns using this mode, she should press and hold the the off button on the control wheel, make the turn and then release the off button to re-engage the wing leveller. That would be the first 'momentary off' yoke-mounted autopilot disconnect button I've ever encountered. It looks to me as though the wing leveller mode is needed because the rest of the system is so crazy that it can't be momentarily engaged so the pilot can take her sweater off.

The instructions do summarize this in a way that sounds simple:

• trim airplane
• engage wing leveller
• ensure directional gyro and artificial horizon are functioning properly
• engage roll section
• engage pitch section

My issue has been that there is so much pitching that I can't troubleshoot it with someone in the back.

There is also an array of navigational functions that the autopilot can be selected to follow. There's heading, to follow the heading bug, OMNI and NAV to track the selected radial on the VOR receiver with approach or en route precision respectively, and there's a LOC and BC mode, which I've had no chance to check. There's also a glide slope coupler, and curiously, the instructions imply that the landing gear must be extended in order to use this mode. Or perhaps they just figure that extending the landing gear is such a good idea they will include it in the instructions.

There is also a procedure to ground check the systems before flight.

Turn everything off. Engage the roll rocker switch and rotate the roll knob full left and right. The yoke should follow. Centre the knob.

Rotate the pitch command disk full DOWN and full UP and see that the control column similarly moves fore and aft. Then centre the disc.

Engage the pitch function and hold the press-to-test for the trim warning indicator. After four seconds it should illuminate.

Ensure autopilot disengaged before takeoff.

I'm going to need a pretty big cow for all this, aren't I? I wish I were as resistant to stress as Highland cattle claimed to be on their info card.

Update: I've since found some better instructions, and seem to have better control of it now.

Trial by Rain

Morning comes as early as ever and I'm checked out and ready to go in the breakfast room. We are going back to Slave Lake to try to get that work done today before the clouds roll in. Asked me what I want 'to go' for lunch, I specify a peanut butter and jelly sandwich on whole wheat. I get a funny look for having the dietary preferences of an eight year old, but it won't go bad in the sun, and I can eat it with one hand while flying.

I fax off a request for "photo blocks," clearance into airspace about 12,500' where we must work under IFR or "controlled VFR," and by the time I call to get my transit code for Calgary airspace, they have an approval for me. We launch and soon reach the work area. It takes the controller two tries to get me to do a full readback on the CVFR clearance. Oh, I get it, CVFR is VFR in normally IFR airspace, so I have to do IFR-style readbacks. This can mess a girl up. I'm on oxygen up at 16,000' flying in really straight lines overhead Slave Lake and environs. This s so much easier at altitude. I even get the autopilot to fly one line, to make sure I know how it works, but the operator says he wants it hand-flown. I think the autopilot could be a very good tool to reduce my fatigue and thus increase overall quality, but I've never had one of those jobs where "you push the button and you take the money," as a reader once explained it to me. I get to fly the airplane.

The pattern we are flying is centred on the town of Slave Lake, and I still have the ADF tuned to the Slave Lake NDB -- I haven't used it since I left -- so every time I turn around it points back to town. It's a fine, clear day here and there is no smoke, mist or other obscuring phenomena interfering with good photos. From the air you see the somewhat peanut-shaped body of water that is Lesser Slave Lake, and the town of Slave Lake right at its eastern tip. We're too high to be usefully monitoring the Slave Lake mandatory frequency, but I listen to 126.7 most of the time just to hear who is coming and going across the north. It's a normal day of medevacs, helicopters and other working aviators. We finish all the north-south lines on either side of the Athabaska River and move onto the east-west ones. These were the ones assigned to the other airplane, but they are still in Edmonton getting the camera fixed, so we'll get a head start on their work. We fly until I have fifty minutes of fuel remaining and then head down for landing.

It's funny doing aerial work this high, right above the airport of landing. It's right there but it's going to take me 15-20 minutes to get there, even at a higher rate of descent than I would use for passengers. I check with him that he's used to rapid descents and he is. The wind direction favours runway 10, landing over the lake, so I'm going to make a big descending spiral along the north shore of the lake and back onto final. I start down at about 1500 fpm. I feel my sinuses squeak a little from the air pressure equalization, and there's a little pressure in my temples, but I wiggle my jaw and let the air leak through my head. Pilot Eustachian tubes must be more flexible than the general public's. At the same time I'm trying to reduce power in stages so as not to shock cool them. Passing about 7000' there is a mighty whack of turbulence which I realize is from the southeast winds whipping over the cliffs along the north shore of the lake and being forced up. It was quite a brutal hit. I'm just turning over the shore, but there will be more turbulence everywhere below. And now my camera operator says some terrifying words. "Oh my NECK!"

I've never had someone hurt on board my aircraft. Although I wasn't doing anything unreasonable, and we knew to expect turbulence descending into high winds, was there something I could have done to alleviate this? "Did you hit your head?" I ask. He says, no, it wasn't the turbulence that did it, his neck just started hurting, a spasm like when you wake up and have slept the wrong way. There's nothing I can do but fly well and put the airplane on the ground, so I do, land, short backtrack and roll off at the taxiway. His neck pain has gone away as quickly as it came. As I'm rolling to the pumps I'm mentally running along the streets of Slave Lake looking for a chiropractor or massage clinic. I hope he's not being macho about this, and the feeling of thinking someone has been hurt in the airplane you were flying is like having a pet or child left in your care be sick. Even if it had nothing to do with you, you feel responsible and want to fix it. But he's moving around normally, talking on the telephone, so perhaps it was just a transient ouch, coincident with the turbulence.

Now there's another issue, a perfectly ordinary one, a fuel transfer. There's a big tractor-trailer fuel truck parked in front of the pumps. I narrate what the operator probably already knows, as it's a common enough occurrence. "This can take thirty minutes or more before the pumps are available to us. I'll find out more as soon as I can." I park where I am physically next in line for the pumps, but the truck can still get out, stick chocks around the nosewheel and go do the pee, flight following, close with flight services, file new flight plan routine. They say fifteen minutes for fuel. I eat half of my peanut butter sandwich and then start just the left engine to pull around for the pumps. As soon as I start to apply power to taxi forward I know exactly what I have done. Damnit! Shut down, jump out, remove chocks, restart cranky hot left engine and this time do pull up to the pumps, close enough for the hose to reach but at an angle that will allow me to scootch out without hitting the small helicopter that is also waiting now.

I shut down there and while they are fuelling, chat to the helicopter folks. "You see my stupid?" They didn't, and say they would have come over and pulled the chock for me had they seen. They are doing photo work too. Very low level work with a special camera that hinges down off one of the skids. When my airplane is fuelled I check the caps, and start that left engine again to pull off the pumps, getting a thumbs up from the helicopter guys to indicate that I have correctly judged my wingtip clearance as I pull past. I shut down again and add some oil before the operator comes back aboard and we blast off to do some more work.

There's a split now in the fuel flow. It's a digital fuel flow meter, visible to the operator in the back. I acknowledge that I'm running the left engine richer but while the temperature gauges wouldn't be against it, that engine is noticeably rougher if I lean any more. I feel frustrated that my engine handling, on what is after all a familair engine, is not up to scratch, but I'd rather burn more fuel than an engine. We get a few long east-west lines done, and then a cloud shield that has been slowly moving from the west all day reaches the western end of our lines. We don't do partial lines, so that ends the work here and the operator wants to go land southwest of here, to have the best chance of staging for our next day's work. "Whitecourt," he says. I call the controller and request clearance out of controlled airspace for landing at Whitecourt. She thought we were landing at Slave Lake. So did I. It's a keep you on your toes sort of operation. She gives me a descent and I pick up Whitecourt weather on the way. It's 6000' broken, raining, with 6 sm vis. That's to be expected, given that we're flying towards the system that just put us out of work here.

The operator doesn't like that. Can we go somewhere else? Sure. I suggest Edmonton. We can see from the shape of the cloud that Edmonton doesn't have the rain yet. He wants to go further west than Edmonton, something southwest of Whitecourt? I'm running a map of Alberta in my head and there's not a lot out there. All I can think of is Grande Cache, which I point out that they built because there wasn't anything there , and we don't want to be stuck in the mountains if the weather is bad and they want to go back east tomorrow. Whitecourt it is. I've never been there, but he has.

The controller clears me down to 6000', which seems odd to me, because we should be in uncontrolled airspace out here by now, but whatever, it's nice to have someone looking out for us. Through 7000' it gets ridiculous bumpy again, so we tell her we're staying up until we get closer. She clears me for an approach into Whitecourt, at which point I realize that she thinks we're IFR. I would have liked to accept the IFR approach, into an unfamiliar airport in worsening weather, but I'm not sure that's legal given that we're actually on a CVFR flight plan. I let her know that, having another of what is going to be a very common confused controller conversation on the topic, and then am released VFR into Whitecourt. There's a NOTAM I don't have, rats, I didn't explicitly ask for it with the the weather and should have. There's a forest fire area inside their control zone. I end up having to stay high over the NOTAMed area than doing a left 270 to intercept final, and then I go to turn on the windshield wipers and darnit, this airplane doesn't have any. I bitch about that, as the operator helps me find the airport. He says "We only fly on nice days!" Point taken. This is crummy weather, but it's perfectly safe VFR visibility and this is the worst he normally flies in. I land and taxi to the pumps.

The weather really is miserable here, it actually got colder as we descended westbound into the front, the wind hasn't let up and it's raining big time. I huddle in the airplane and make my postflight calls from the cockpit. The briefer who takes the call is in Edmonton, so as I close my flight plan I add, "The weather is nasty and it's heading your way." She doesn't hear that and asks me to repeat, so I admit it was just irrelevant pilot whining. I whine some more to a helicopter pilot who is part of the forest fire fighting efforts nearby. Once I'm fuelled--gotta love fuel service, it's actually been a while since I fuelled my own airplane in nasty weather--I taxi to parking and put everything away for the night. Walking back across the apron I'm hailed by the pilot, putting blade covers on his helicopter. He asks me if I've called a taxi yet. I haven't. "Call two please, I need one in about half an hour."

In the terminal there are no taxi company numbers or cards by the payphone, but there's a phone book, one of the northern kind that has twenty or so different small communities all in one book. I leaf through the cab section finding "Sparky's" with a Whitecourt phone number. He can be there in ten minutes, and then can go back for the other guy. I don't think it's a two-cab town.

In fact it's not a fully one-cab town. The vehicle arrives has a spiderweb-cracked windshield, no seatbelts and an inoperative right side sliding door. It's the sketchiest cab I've ever been in, and recall that I've been to Cambodia. The driver looks like one of the Yukon prospectors who didn't strike gold. He appears sober, so we ride into town and send him back for helicopter guy, whose name I forgot to ask, so I give the company name from the side of the helicopter.

We're at the Super-8, across the street from Boston Pizza. Ahh good times, old times. You know, you can be nostalgic for anything. I'll bet there are Cambodians my age who manage to have moments of nostalgia for the days that they woke up at five a.m. to try and grow rice under conditions that rice doesn't grow. I alert the operator that I am going to order my "high-maintenance pizza" and he wants to know what that is. "Are you sure you want to hear it twice?" He does. It's a small Rustic Italian, whole wheat crust, double sauce, no cheese, add pineapple, and extra sauce on the side for dipping, please. He rolls his eyes, but the waitress accepts it cheerfully.

It's seven p.m. and we're discussing the next day's plans. He'd like to be over the mountains and working in the Vancouver area at 9:30 a.m. I do napkin math. So take off at nine from an airport near there, maybe Abbotsford, at nine, so have to land Abbotsford at 8:30, assuming unfavourable winds leave here at 5:30, so get to the airport at 5:00 a.m. so allowing for taxi and hotel stupidity, a 4:00 a.m. wake-up call, which means I have to be in bed by eight, the restaurant isn't too busy, so we should be done in time for that. Yup I can do it.

"You can go to bed at eight, just like that?"

"That's my job." I could go to sleep right now, but I'm going to eat my pizza first. You may have noticed that I missed a step in my math. Neither of us did, at that point.

I say I'll have my leftover half peanut butter sandwich for breakfast, and he suggests that I can have leftovers too. That's before he watches me inhale the pizza. I was going to have a dessert too, but have to get to bed for eight.

It's absolutely pouring now. I'm glad I through that big coat in the plane with me. It seemed pretty silly on a 20+ degree day leaving home, but now it's perfect. This is the company coat (logo covered over with a generic patch) that I snared on my first day at Victory Airways, electing to take one size too big rather than wait an uncertain period for someone to get me one the right size. Pilot decision-making at its finest. The hood keeps me dry.

I go straight to bed and encourage sleep by running gently through the alphabet picturing an animal for each letter of the alphabet as it curls up and goes to sleep. Somewhere around giraffe I realize the error in my napkin-based planning, but I will fix it in the morning. I never got past giraffe. Think about it for a while before you scroll down for the horrifying truth. In what position would a giraffe sleep? Standing up? There's no way the head would balance up there. Standing with the head drooping down and lying on the ground? Lying on its side with the head and neck stretched way out where another animal might step on it? Curled up like a cat? I somehow always had sleeping gorillas, gerbils, goats and geckos. This is the first time I've had to contemplate nap time for the poor giraffe.

Sleep is very very important to me, and I won't compromise my duty rest for anything. I know how quickly a sleep debt can build up and how severely it can impact my behaviour. I'll forever wonder if the giraffe would like to get much more sleep, but just can't get comfortable. More on sleep postures and sleep behaviour here.

Always Watch Your Autopilot

Once I'm clear of Calgary's airspace, I climb to 7500' and then level off and trim every which way while the speed stabilizes. The GPS is set direct to destination, the trackbar to the track I am following to get there and the heading bug to the heading I'm flying to hold it in northwest winds. I turn on the autopilot, and press the Engage button. Unlike Star Trek the airplane doesn't flash and disappear into a rainbow rubber band of light. Instead it rolls into a thirty degree right bank and drops into a 300 fpm descent. I retake control, climb back to 7500', disable altitude control and switch it from nav mode to heading mode. When I reengage it, it follows the bug straight ahead, and left and right when I move it. I go back to nav mode and watch it for a bit. I'm a few hundred metres off the centre of the track and apparently when you're an autopilot this calls for drastic measures to regain track. I guess when you're an autopilot that's your whole job, so you take it pretty seriously. Or when you're an autopilot designed in the seventies, you only know one strategy for getting on track: you bank thirty degrees towards the track until you're within some logic of centred and then you roll level.

It's kind of interesting that the autopilot, which was designed in the 1970s works with a brand new GPS unit. There's no computer in the autopilot and no mechanical parts in the GPS. The truth is, they don't talk at all. They communicate entirely through an intermediary, the HSI ("horizontal situation indicator," the second stupidest aviation abbreviation after DME). The autopilot also talks to the flight director, which is part of the attitude indicator, but I don't believe the GPS participates in that conversation. The GPS tells the HSI how far off course we are and then the autopilot rolls the airplane into a bank to regain the course, with the flight director telling the autopilot when it has reached the bank limit. I have to set the heading bug and track bar manually.

After I get tired of correcting the autopilot, I turn off the altitude hold feature and keep altitude myself while the autopilot holds the course. This doesn't work that well because it doesn't take much force to disconnect and doesn't have a disconnect alarm, and every once in a while I turn the yoke inadvertently and disconnect it, so that it drifts off course. But I persevere. I'll have to get used to this.

Spring is coming to the prairie but there are still snow curls on the ground, highlighting all the ditches and gullies. As I approach the destination airport, I start looking at water features on the ground for confirmation that the runway I am anticipating is the correct one. I get caught by this every time: when the snow is off frozen ponds look just like open water from the air, and the ripples are frozen right in. So they're only useful to determine landing direction if I have a time machine to the day they froze. I take a guess and then compare the GPS ground speed to my airspeed to confirm that I don't have an untenable tailwind.

I land and taxi slowly in, then text the flight follower. He tells me whom to call for a pick up. Fuel is closed for the night, so I park in front of the terminal and plug into an outlet there. I don't know anything about what kind of power I can draw from it, so I only activate the cabin heater to keep the computers warm overnight. The engines will have to settle for tents and plugs. Shiny new engine plugs which keep the heat in in the winter and the birds out in the spring and summer.

I see the uv damage lines on the tire again as I chock. They're just on the surface, but they bug me because they aren't usually there. I unload my baggage, and a box of parts which must be everything they took out while doing the work. The client picks me up and laughs at the parts box. He offers to leave it in his truck so I don't have to keep it in my hotel room.

Supper is something generic at a local restaurant. I use a trick I learned from another pilot to avoid eating more than I need: I ask to have a takeout container brought with the meal. Then right away, before I eat any, I put half the food in the takeout container. That protects me from unconsciously eating it all even though its more than I need. My other strategy for this month is the iTunes tactic: whenever I'm tempted to buy a chocolate bar from the hotel vending machine, I'm going instead to buy a new song from from iTunes. It costs exactly the same and I'll have the at least as much fun picking it out. And they I'll put it on repeat and get up and dance to it.

It's the end of the day so I just check in and veg out. The TV comes on at the weather channel. A lot of hotels do that. If they don't have their own dedicated advertising channel shilling for the pay movies, they have the default channel be weather. It's useful to most people and completely inoffensive. It will never come on in the middle of a violent TV show or a sexy movie. Unless you're offended by pressure charts or footage of a woman walking down the street carrying her cat in a front slung baby carrier, you're going to be happy or bored watching this. And then they toss off a casual mention that a Chinese oil tanker has just taken a dump on the Great Barrier Reef. I literally scream. I don't know what caused the accident. Probably a stupid autopilot. I don't want to read more about it. I hate horrible things that I can't do anything about, and as a consumer of petroleum products I have to acknowledge some responsibility, too. Not too many years from now global climate change will result in easy passage through the once treacherous Northwest Passage and countries that didn't even know it existed before will be plying the arctic waters. It's a completely different ecosystem, but just as vulnerable as the coral reefs. I'm glad I'm not immortal. The future has amazing 3-D movies but dead barren seas.

Autopilot Disconnect

Sometimes I read so many manuals that the words scramble themselves and I can't tell what they mean any more. But sometimes it's not me, it's them. The truth is, my autopilot is so mad, that even the manual disconnects unexpectedly.

The Century Flight Systems, Inc. Lateral Guidance System contains. Track interception angles a completely automatic, analog computer that directs the autopilot in both VOR and ILS navigation. The system contains a five position mode selector switch which mounts in the instrument panel are 45° and an automatic 15° crosswind correction capabilities is provided. The complete capture, intercept and tracking sequence is accomplished automatically without monitoring or multiple switching.

What? It's some sort of scrambled puzzle. Here are the pieces:

1. The Century Flight Systems, Inc. Lateral Guidance System contains
2. .
3. Track interception angles
4. a completely automatic, analog computer that directs the autopilot in both VOR and ILS navigation.
5. The system contains a five position mode selector switch which mounts in the instrument panel
6. are 45° and an automatic 15° crosswind correction capabilities is provided.
7. The complete capture, intercept and tracking sequence is accomplished automatically without monitoring or multiple switching.

Reordering 1-4-5-2-7-3-6 we get an almost coherent statement:

The Century Flight Systems, Inc. Lateral Guidance System contains a completely automatic, analog computer that directs the autopilot in both VOR and ILS navigation. The system contains a five position mode selector switch which mounts in the instrument panel. The complete capture, intercept and tracking sequence is accomplished automatically without monitoring or multiple switching. Track interception angles are 45° and an automatic 15° crosswind correction capabilities is provided.

The only problem is that "an automatic 15° crosswind correction" is a capability, not capabilities, plural. Can anyone solve the puzzle to make it an entirely readable paragraph in natural English? Perhaps the answer will result in the key to keep the actual product online for an entire flight.

Please, if you assemble or install autopilots or write equipment manuals, go easy on the liquor, and on the cut and paste options.

The autopilot seems like a skilled copilot in that it is usually very good at holding headings and altitudes, but it is actually a really poor copilot in that it it will not tell you if it is feeling unusual control forces and will never remind you of any other task. It is just a super fixated control manipulator, with no actual piloting skills. Narcolepsy is not the least of its problems.

Hold On

My mission, determined in the previous installment, is to learn how to program a hold into the GNS430 in no more than double the time it takes to calculate the hold and set it up manually. Or at least to figure out how to make the HSI behave like an HSI and allow me to fly the hold manually. I've tried wandering around the menus and scrolling down to the secret parts I can't see right away, but nowhere have I found a choice like "Select Hold" or "Hold at waypoint." I can visualize what the screen might look like, with an option to select my holding point, inbound track, turn direction (default right), desired time (default one minute) and an optional second point if it's a DME hold on an airway. But no such screen is in evidence.

Back to the pdf manual (damn being a pilot is so glamourous) to look for holds. There's nothing in the manual specifically about defining your own hold, but I found this advice online. There is also an iPod application on the GNS430; author Max Trescott generously offered me a promotional copy of the app, but sadly iTunes won't let Canadians do that. Thanks to everyone for your various suggestions. I need to go back to my friend the OBS key. They should have labelled that key HLD. It would make a lot more sense.

So I reposition the simulator to try out an enroute hold. And here's a problem I've never had in the real aircraft. I moved the simulator window slightly, and my HSI vanished. Just poof, gone. Blank spot in the simulated panel. The simulator, in other words, is not without bugs. I exit and restart.

This time, the hold was easy. Using the OBS key you can hold anywhere you have a waypoint.

• fly direct the waypoint
• before you reach it, hit the OBS button to disable waypoint sequencing
• bug the outbound heading of the assigned hold
• cross the fix, turn appropriately, and set the trackbar to the inbound course
• fly the hold just like you would if it were a VOR hold and you didn't have a GPS

It's cool because you can fly a hold at any fix at all, without messing about trying to identify the fix using multiple nav aids, but if it's not a published hold (based on a comment, I'm guessing the American for that is "charted hold") the GPS doesn't do the timing for you.

That technique may sound kind of lame, as it's just pushing a button then flying a hold manually, but by comparison flying published holds on the GPS is not all that whiz bang either. The flight plan names the fix then has the word "hold" on the next line. About five seconds (yeah, real heads up there, Garmin) before crossing the fix it tells you the entry type (only "TEARDROP ENTRY" not any instructions on the outbound heading), or even reminding you if it's left or right turns. The timer is kind of a neat gimmick because it starts the clock at the correct moment, passing abeam the fix outbound and turning towards it inbound, but it's not like you push a button and then sit back and watch the airplane fly the hold. There are cockpit systems that do that, even calculating compensation for wind.

Also, the little kids doing the announcements for the Paralympic Winter Games opening ceremonies did a great job. I didn't realize how small they were until they introduced the Governor-General and they all walked along holding hands.

Confusion is Only Simulated

I knew that Garmin made simulator modules for their products, but I had to search to find it. The FAQ "expert assistant" search returned the answer "yes" to the question of whether there was a simulator available, but didn't actually return a link. It's easy to install though and the 150% view makes it nicely usable on the laptop. I selected the C/V volume knob to turn it on, and it started up fine, running through the same integrity checks as the real unit. I verified that all the flags were as they should be and then noticed a MSG light, so pressed the key to find the message. "Main processor error." I've no idea where that came from. Who would simulate a hardware flaw? It only came up the once and never again, so as long as the real units don't do it, it should be good.

It's well-designed as a simulator. The screen consists of a photographic representation of the physical knobs of the unit with superimposed arrows so you know where to click in order to turn the ones that need to be turned. For the aircraft interface they depict only an HSI and generic simplified autopilot controls. Really simple: NAV HDG ALT, an altitude increase/decrease rocker switch that instantly puts the airplane at the altitude you select, plus a SPEED slider that instantly puts the airplane at any speed from 0 to 600 knots. This takes aircraft performance and configuration completely out of the equation, you just set the speed you would have for the phase of flight, and it acts as both a pause key and a fast forward. The only non-intuitive part is using the OBS and heading bug knobs on the HSI. You need to click on the appropriate knob then hold and drag straight horizontally, maybe all the way across the screen for a large change. It's easy to do once you figure it out, though.

My first surprise was that the menu screens don't wrap all the way around. That is, if you're exploring the suboptions of the NAV menu by turning the inside right knob, when you get to the seventh screen you can't go back to the first one by turning the knob one more click. You have to go all the way back through the way you came. Weird.

I set up a flight plan from Saskatoon to Prince Albert, because aside from having a NSFW name (don't google for images) PA has at least one of every sort of approach. I dial the speed up to around mach one so that by the time I've followed the instructions for selecting a full procedure ILS runway 08 approach and admired all the pretty menus, I'm almost there. I slow down to a more realistic speed for my aircraft, activate the approach and start the descent. It counts down to the beacon, and I set the heading bug to the outbound intercept, but nothing happened. It starts counting up how far away from the beacon it is. I activated the approach again and it started turning right toward the airport, not left on the outbound PT. I let it, to see what it would do, and it turned all the way around. Ack, the logic must have told it to fly to the beacon and then outbound and with the beacon behind and just a hair to the right, that made it do the right 270.

It tracks outbound and I fly the procedure turn then toggle the CDI to VLOC while inbound on the reversal. I notice that the glideslope is flagged off and recheck the frequency. Everything looks okay, but no glideslope indication appears and the distance counts down to the FAF, then counts up again as the map depicts reaching the runway and overflying it. Back to the manual.

This time I program in a sample flight plan from the manual, it's KFDK-KLYH. I finally figure out what I've been doing wrong: I have to make sure that the OBS key is not selected. The OBS option suppresses automated sequencing of waypoints. Silly me, I thought it enabled the OBS.

The example in the manual tells me to select VOR 03 but the simulator database only has VOR04: the manual is from 2009 but the simulator database only goes up to 2006 and the approach has changed. I choose the VOR 04 -- and the transition as told. It's loaded but not activated. How do I tell if it's activated by looking at the GPS display? What happens if I activate it twice by accident? I'm not sure. The waypoints sequence and I try the simulated ILS approach, but the glideslope doesn't work for this, either. I guess its just not simulated.

I set my location back to PA and try to do a VOR/DME arc. It's a 14 DME arc that runs from XEXEX to XETUL (the latter is in line with the approach). I'm starting at the YPA VOR, not the way you'd do a real one, but I want to see how this works. I can see how to go direct XEXEX, but I aim manually for the arc, west of XEXEX, trying to set the obs to a radial perpendicular to the arc where I will intercept it, as I would without the GPS, but it doesn't work that way. I have to set it along the arc and it does the conversion itself. Even when I set the CDI to VLOC I can't use it the familiar way. The main nav screen doesn't tell me my DME from the VOR. I'd prefer that this do a better job or enhancing rather than completely replacing old habits. It directs me to turn left onto the approach before reaching the fly-by waypoint at XETUL. I turned as directed and then had to correct back to the right in order to get on the radial by PUVER. That was no good. I don't trust it so I'm doing things myself and we're fighting and making a mess. Try again, this time from the south.

I go way out so there's lots of room and select vectors so I won't be asked to go to XIBEV. I fly direct BETIM. It's irritating that it doesn't tell me the ONE piece of information I want setting up for a DME arc: my DME! I'm 16 nm from BETIM, which is 3.5 nm from YPA and it's a 14 DME arc, so I have a few miles. If BETIM to YPA were a straigt line I'd be expecting to intercept at 10.5, from BETIM but they are not, so it should be before then. Yet, crap! I'm 9.1 from YPA and there's no arc intercept depicted and no turn instruction. I guess "vectors" was vectors to the final approach for the VOR/DME 08 and not for the arc. Try again.

What you have to do, is what I did before, set up for the DME arc, choosing one of the endpoints XEXEX as the initial fix, then go into the flight plan where you see the sequenced waypoints XEXEX, dme arc, XETUL, PUVER, BETIM all one below the other. Activate the cursor, scroll down to dme arc, and select direct to. This is going to work. I pull up the NRST screen that shows me the GPS distance from the VOR and it's showing 15.0 as I follow the directions through the arc. If I'm on the arc, the GPS distance should actually be less than the DME because DME includes slant range, and the depiction on the NAV map shows the same: I'm tracking outside the arc. It's good enough, I suppose but not really clear guidance for flying the arc.

And then it's straight down the approach to the missed. I always have to go missed in this simulator, because there's no way to see the runway. The missed instructions are straight ahead to 3100 then left to YPA VOR. I climb instantaneously with the little rocker switch, check to make sure the CDI is in GPS and not VLOC mode (I'm going to do that as part of every missed approach so that I don't forget on the occasions that it's an ILS approach) and then cancel the SUSP by hitting the OBS button. The simulator turns left as the missed approach instructions specify, but that was chance, because I've since seen it turn right, so remember it is not flying the missed, just going to the holding point. It is entirely up to the pilot to fly the missed as published.

When I did this on the sample approach from the manual, approaching the holding point it gave directions to fly the TEARDROP hold entry. I didn't actually fly the hold there, just repositioned and went back to Canada.

Canadian plates don't usually depict a published hold after the MAP, you just know that you can hold there, on the inbound track, and ATC can direct you to hold anywhere. Back in simulated Canada (don't you wish the real GPS had a transporter function?) I intend to enter and fly the hold manually, because after the missed I'm right on top of it, and too close to scramble into menus. Left turns, my outbound course just right of my current track to the VOR, makes it a parallel entry. I watch a messy station passage on the map view and I make the first right, using the CDI button to pop the unit into VLOC mode, verifying that the YPA frequency is the active one, but the flag doesn't flip. Map view shows I'm clearly past the VOR. Weird. I start the timer anyway and fly a minute outbound. Right turn inbound, flying an intercept heading on the VOR, but map view shows me cross the selected radial with no change to the CDI. Obviously VLOC mode doesn't do what I thought it did. I'll take a vacuuming break and read up on this later.

IFR Procedures with the G430

The GNS430 is installed and IFR certified, so it is approved to work with the autopilot. I program in what it's supposed to do, and then it will give the autopilot directions on how to proceed. My job then is to monitor that it is giving the right instructions and that the autopilot is carrying them out. It's like you could give your car's cruise control instructions from your dashboard GPS on where the speed limit changed and where your exit was, and it would automatically exit and slow down.

The IFR procedures are all stored on a data card, and yes the cards must be physically replaced every 28 days, which I just know is going to be a pain and a half someday soon when I need to escape IFR from somewhere where overnight delivery service takes three months. I hope it won't lock up and refuse to work at all just because the card has expired. [Update: Paul says the updates can be downloaded over the Internet.] Canadian procedures only change every 56 days, unless by NOTAM. Question: if a NOTAM has temporarily amended a procedure can I modify it in the device? I'm guessing no. I'd have to hand fly it.

To load a procedure into the flight plan, I must first have a flight plan (or it will settle for an active "direct-to" command to an airport). I press PROC and it pops up a menu of types of procedures. I use the large right knob to Select Approach (or Select Arrival?, Select Departure? as appropriate). Enter selects, and it presents a choice of procedures depending on the destination airport in the flight plan or direct-to. Assuming you're selecting an approach, it drills down to allow you to specify the initial approach fix (IAF) or that you will be receiving vectors to final. Once you have selected the procedure with the enter key you can choose to load or activate it. Typically you would load it but not activate it until cleared for it, but it looks like you can activate it any time without restrictions. To do that after loading, you press PROC and then choose Activate Approach.

About 30 nm from the runway the GPS switches to terminal mode, i.e. 1.0 nm CDI scaling each side. It's a gradual transition, so if you were sitting there one dot off centre on the 5 nm scaling it doesn't suddenly whip over to full scale deflection. If you want, you can press FPL and right knob scroll through the list of waypoints to review the approach procedures. Seeing as the autopilot is flying, you could do that the way you look at nav radios as you give the approach briefing.

Approaching the IAF it displays your next desired track (DTK) as the outbound from the nav aid. It's up to you to make sure the OBS is set correctly and to decide which procedure turn to fly. It doesn't direct you left or right but just depicts your position relative to the inbound track. About a minute after passing the final approach fix (FAF), the nav aid you'll fly toward on the way back in towards the runway, it directs Start procedure turn, but again it's up to you to choose and fly the reversal. This could be done presumably by hand flying or by directing the autopilot with the heading bug. Question: Is it always about a minute after the FAF? what about those facilities in tight valleys or over residential areas where you have to fly a longer outbound leg? Halfway through the reversal, the GPS sets your desired track to the inbound and leaves you to intercept it. I'm not entirely clear on how I should be using the heading bug here. The GPS-autopilot interface is quite mysterious, given that they are technologies invented close to fifty years apart in time.

The unit does not accept a barometric pressure setting, and does not display altitude on any of the screens I've looked at in the manual so far, so it would appear that the GPS provides directional guidance only, and that the pilot is responsible for instructing the autopilot in step down altitudes and rates of descent on a non-precision approach. (There is a mode to reach a target altitude at a specified location, but presumably that is GPS altitude and is intended for VFR use). The GPS directs the autopilot to hold a heading and the pilot directs the autopilot to descend to an altitude, while the pilot uses the throttles to instruct the engine regarding power input. It's like CRM between a human and two machines: you have to be very clear who is responsible for what.

On a precision approach, the GPS is not approved to maintain either lateral or vertical guidance: you must switch to the VHF navigation signal from the ground-based ILS. The CDI can be set to automatically switch from indicating deviation from the GPS track to the localizer track, so long as you are tracking within 1.2 nm of the localizer, starting 2.0 to 15 nm back from the FAF. If it isn't set to switch automatically, or if you have a less than 2 nm gate, you can activate it manually by pressing the CDI button. In either case, the correct ILS frequency must be set in the VLOC radio, for the switch to occur. The unit will put the ILS frequency up on standby for you, but you have to toggle it from standby to active yourself.

While you are on an approach, the missed approach point (MAP) is the current waypoint. As you reach the MAP, "SUSP" appears above the OBS key and the pilot is expected to either land or conduct the missed approach. To land I guess you just land: the autopilot is still on and presumably the last heading instruction the GPS issued to the autopilot is still valid, so the airplane continues to the runway and the pilot can disable the autopilot and land at whatever point they prefer. To initiate a missed approach, hit the OBS key (and switch the CDI back to GPS from VLOC if it was an ILS approach) then the unit should give directions into the hold. It is not entirely clear from the documentation whether it will take you step by step through an elaborate missed, but at any rate if the missed called for an immediate hundred degree turn away from the mountain, I would be doing that before messing with the GPS for guidance. It looks like the GPS is a fair weather helper that is not willing to help you with the highest workload portion of the flight. It will tell you the correct entry procedure for a hold, but I'm thinking that unless it is a published hold that the database knows about, it would be quicker to work out the entry myself than to input the hold instructions.

Question: does the approach I fly have to be the LAST WPT, such that I can't pre-include the alternate as part of the flight plan? It appears that that is the case.

It looks like the easiest things to forget are: taking care of the vertical profile; activating the approach; and ensuring the correct CDI source for the phase of flight. Anyone who has any IFR experience with autopilot-linked Garmins is very much encouraged to give me pointers here. I am, for example, stymied as to how I define a hold that is not depicted on the plate as part of an approach, if say there's someone ahead of me on the approach and I just need to hold at the NDB and wait for them, or if ATC suddenly assigns a hold while they sort themselves out.

Tomorrow's post (not skipping a day this time) will be me trying it all out in the Garmin-supplied simulator.

Garmin 430 Overview

I'm training on a new GPS NAVCOM next week. (That is a GPS receiver that has an integrated communications radio and simulates traditional navigation instruments). This one talks to the radio navigation instruments and can combine their knowledge with its. The GPS can drive my HSI CDI so that I'm looking at the traditional instruments, but navigating by GPS information. This is good, because the traditional instruments have a better display and are right in front of me, so more comfortable to use. The system is complicated, but fortunately it's a Garmin, and I've used a lot of Garmin GPS receivers, so the logic isn't completely foreign; I'm working my way through the first online manual now.

Equipment like this has a lot of context sensitive keys. For example if you push the middle of the bottom right knob, you activate or deactivate the GPS cursor. If the cursor is deactivated, turning that knob changes screens, but if the cursor is active, turning that knob moves the cursor from field to field. But not to all fields. There's a separate cursor for changing VHS standby frequencies¹, and two separate frequency swapping buttons. I like that, as I have been caught out on units that had one cursor, messing up my standby frequency just before I needed it, because I was hunting for the approach page. The section of the screen that acts as what would be labelled NAV on the analog radios is designated VLOC here. That represents VOR plus LOCalizer², i.e. this radio is used to tune VOR or ILS frequencies. They needed a more specific term than NAV, because the whole thing is for navigation.

The 29 possible display pages are in logical groups, so you first use the bottom right knob, cursor deactivated, to select a group, i.e. NAV, WPT, AUX or NRST, then you turn the inside right knob to select an individual page within that group. I need to familiarize myself with which information is available on which pages. Setup is under AUX. Airport information is under WPT. If I get lost³, I can press and hold the CLR key to get back to the default NAV page.

Hmm, on page 17 it says that to stop navigating a flight plan I select Delete Flight Plan. Single use flight plans? That seems a little drastic. That option is usually Stop Navigation. It isn't until page seventy-five in the full Pilot's Guide that it confirms "Deleting a flight plan does not delete the waypoints contained in the flight plan from the database or user waypoint memory." You're really just deleting it from RAM and it's still stored, ready to be selected again.

Some of the pages can be accessed without going through the menu system. The approaches and all their associated pages are accessed with one touch, the PROC key. This is good. All those other pages are just pretty information screens. The approaches are going to actually tell me--or the autopilot⁴--where to go. The unit will tell me if I select an approach that has not been approved as standalone GPS. For such approaches I have to navigate primarily by the traditional nav aids, and monitor the GPS only for situational awareness. If traditional nav aids are available, I'll be monitoring them anyway, but it's a matter of what is driving my CDI⁵. I press the CDI button to select that, or "An ‘Auto ILS CDI’ setting provides automatic switching to ‘VLOC’ once established inbound on the final course segment of an approach." Nice, but dangerous, in the way that autothrottles are: more than one pilot accustomed to the airplane doing things for him has forgotten to do them himself when the occasion called for it. I'll have to make verifying that that AUTO CDI switch occurred a part of my routine.

Ever memorized something because you were told to, without being given enough context to use it or even understand it? I hate doing that, but I did so years ago in order to know the answers to questions I was told would be on the exam. I guess they were. I had to know the CDI scaling for different phases of flight. I don't remember them now, so I don't know if they are the same as on this 430, but the numbers here are:

Phase	Scaling	Location
ENR	5.0 nm	enroute
TERM	1.0 nm	<30 nm from destination or 1-5 nm from departure
APR	0.3 nm	<2 nm from destination when on a GPS approach

I think those are full scale deflection each side. I'll have to confirm that with the next manual. I'll make the procedures for flying an approach on this unit a post unto itself, for another day, because despite all the bells and whistles, that's what I'm really here to learn. This whole week will be Garmin 430 posts.

---

Notes

Anyone know if there is an html tag that will automatically number footnotes, and superscript the index, like in LaTeX? Ideally it would provide internal page links to the footnote and back to the text, too.

1. If you press and hold the com swapping button for about two seconds, you get the emergency frequency of 121.5. I wonder how long it takes before the unit malfunctions and you risk getting 121.5 any time you swap frequencies. It's very common for older radios to double-flip when you try to swap them, leaving you talking to the same controller who just told you to switch frequencies. Embarrassing, but having that happen on 121.5 would be more so.

2. This makes VLOC a deeply recursive acronym, as VLOC is short of VOR/LOCalizer, and VOR is short for VHF Omnidirectional Range, and VHF is short for Very High Frequency, but no one ever thinks about what VOR stands for as they are saying it. It's usually pronounced Vee-Oh-Are, but there are a few people who say "vore." I suspect they are all students of the same eccentric flight instructor.

3. Lost in the menu screens, not lost flying the plane, nor can I get Lost the TV series on the little screen. It is in colour, though.

4. I'll believe that the autopilot works in this plane when I experience it myself.

5. The CDI is the little needle that tells me which way I need to correct to be on course.

eAPIS, or Praise Be to Customs Expediters

I learn that my next task is not to return to the field with this airplane, but to take a different airplane to Kansas to get the autopilot repaired. Apparently every avionics technician in Canada has washed their hands of this terrible autopilot, and there's a superb shop in New Century, Kansas (I'm guessing that town just celebrated its centennial) where they know how to fix the most reluctant autopilots. Airplane number two is currently sitting in a maintenance shop at another airport, where a different engineer and apprentices are wrestling with the installation of the electronic tachometers. The PRM will drive me there and then I will take the airplane to Kansas and get an airline flight home for my time off. Meanwhile the next shift of pilots will come here to take this airplane back to the jobsite.

So the downside is that I won't get to see the look on the customer's face when he sees that his complaint about the carpet has been addressed in full, but I've never landed in Kansas, as far as I remember. I don't even know much about it. I understand that Kansas is very flat, has powerful cyclonic storms, is located somewhere south of Nebraska and north of Arkansas, and they make Cessnas there. And wheat. I had also heard a rumour that circles in Kansas were officially not quite the way they are in the rest of the world, but that seems to have been Indiana, and not made it into law. I prepare to go to Kansas.

You might think I'd start by increasing my geographical precision beyond the Nebraska-to-Arkansas approximation, but getting there will be the easy part. It's getting permission to get there that will hold me up. I ask the boss if he has started the paperwork with our broker for the flight, or if I should contact them myself. He says that as this is not a commercial flight, I shouldn't bother with the broker, just do it as a private flight to get the work done on the airplane. I start by choosing an airport of entry, one close to a straight line from here to Kansas, with 24 hours customs service, and fairly close to the border so that if anything happens requiring a diversion, I'm not forced to land at a US airport with no customs. I call the airport to find out their procedures. This shouldn't be necessary, but experience has taught me that things are not done the same way at every station and the best way to avoid delays or disapproval is to respectfully find out exactly how they want it done. I reach an amazingly friendly and helpful customs agent who e-mails me the form I need and a link to the appropriate website along with instructions for exactly where to go and what to do at his airport.

As well as filing the usual flight plan and customs arrival intentions, I now have to complete eAPIS paperwork. Except, as the e implies, it's not paper, but electronic. This is where I find out about the work my customs expediter has been doing on my behalf. Mind boggling. Once I see how much work it will be I go to bed, planning to do it in the morning. The airplane won't be ready to go until midday.

Never Know Where I Spent the Night

So I did my weekly paperwork. My weekly expense/pay record is an Excel spreadsheet. Actually it used to be an Excel spreadsheet, but I couldn't find my MS Office installation CDs after I bought my new computer, so now I use Open Office. I'm too cheap to buy whatever the latest release is. I have to keep track of where I am each night and how much I spent on things like fuel and extension cords and Febreeze. The "where" is always in the form of a four letter airport identifier. Concise, unambiguous and easy for the boss to match with the logbook if he needs to check up on us.

If I've been staying in the same place for a few days I'll update the spreadsheet just by selecting the cell where I last typed an identifier and copying the information down into the subsequent cells using what I think is called the drag handle. This works fine to make a column of seven CYZF entries or four times CYXE. But when I'm working out of some little airport that has numbers in the identifier, Excel tries to get clever. If I'm in Killarney, Manitoba and have logged CJS5 into my spreadsheet, when I drag it down it autoincrements, and one finds that on subsequent nights I have stayed at CJS6, CJS7 and CJS8. I often don't notice until the increment reaches 10 and the five digit code gets my attention. I would be annoyed at the software for trying to outsmart me, if it weren't so funny. I laugh every time and just change it back.

I also manage to annoy my coworker by accidentally implying that (a) she doesn't communicate with me, (b) she was somehow responsible for my losing my battery clip and (c) that I left an apple core in the cockpit. Fortunately I found the battery, properly disposed of the apple core and I think she's forgiven me for the miscommunication. And I got the printer to behave, but I still hate it. Hates it, nasty printer.

Cavalry Charge

Did you hear about the A330 that encountered severe turbulence, pitot icing, unreliable airspeed indications, and a cascade of system warnings while crossing the Atlantic towards France? The pilots did a great job getting that airplane to Paris in one piece.

No, I haven't entered an alternate reality where Air France 447 reached its destination. I'm reading a company memo from Hugh Houang, the Air Caraïbes flight safety officer, describing technical difficulties they had with that airline's A330-200 aircraft in August and September last year, and the inadequacy of the Airbus checklists for dealing with them. The memo is in French, but someone who speaks Airbus will probably be able to read it with minimal trouble. Airbus system and warning names are all in English. Below is a highlights summary, with my own commentary in parentheses.

It first documents the flight of F-OFDF, en route from Fort de France, Martinique to Paris, France:

• 22:11 - They encountered adverse conditions. Prescribed weather deviation procedures didn't help flight conditions, so they returned to FL350 at 22:14.
• 22:22 - They reduced speed and power settings, and disconnected the autothrust in accordance with the SEVERE TURBULENCE checklist.
• 22:22 - In seconds, the air temperature rose from -14 to -5, indicating the temperature of the ice, as opposed to of the outside air, a sign of severe pitot icing. The displayed calibrated airspeed dropped from 270 kts to 85 kts (yikes! that's too slow for me!); the flight directors and autopilot disconnected and there were a cascade of warnings and alarms, my favourite of which was CAVALRY CHARGE. (An Airbus speaker explained that this apparent Napoleonic reference is the name of the distinctive audio signal associated with autopilot disconnect). The stall warning sounded, as did the CRICKET (another type of annunciator) and the master warning came on. (I'm skipping many of the alarms listed. Suffice to say, this cockpit looked and sounded like a pinball machine).
• 22:23 Over the next two minutes the temperature dropped again, the airspeed came back up and the altitude jumped up to 34500'. They soon recovered their flight directors and autopilot.

For 86 seconds the crew had no reliable airspeed, mach or altitude indications. (This may seem like a very short time, but 178 seconds was the average time a pilot with that information but without the training on how to use it took to destroy a simulated small airplane. And those test subjects weren't in severe turbulence. All the training in the world doesn't help a crew keep the airplane under control if they don't have the data). The crew concentrated on flying the airplane, using GPS data, and trying to complete the Unreliable Speed Indication checklist. They were helped by the fact that they had already completed the Severe Turbulence checklist, but there was not time during the incident to complete its recommendations. The manual strongly suggests to the pilot flying that the stall alarms are inappropriate. (The picture given by the message RESPECT STALL WARNING AND DISREGARD "RISK OF UNDUE STALL WARNING" STATUS MESSAGE IF DISPLAYED ON ECAM is of someone flying an airplane where the stall warnings are saying that the airplane is stalling, the ECAM screen is telling the pilot to disregard the stall warnings, and the manual is telling the pilot to disregard the ECAM and heed the stall warnings, but the pilot doesn't believe that the airplane is stalling. It's like having your chief pilot, training manager and captain all on the flight deck, telling you do different things).

The next part of the memo then analyzes the event and the warnings in terms of the Airbus protections offered by ALTERNATE LAW, NORMAL LAW, and DIRECT LAW, detailing what was and lost or changed in response to the various alarms, such as the F/CTL ADR DISAGREE. He points out that the checklists contradict each other when the unreliable speed indication checklist says, RELY ON THE STALL WARNING THAT COULD BE TRIGGERED IN ALTERNATE OR DIRECT LAW. IT IS NOT AFFECTED BY UNRELIABLE SPEEDS, BECAUSE IT IS BASED ON ANGLE OF ATTACK, while the icing checklist warns UNDUE STALL WARNINGS MAY MAINLY OCCUR IN THE CASE OF AN AOA DISCREPANCY. (AoA is angle of attack, measured by vanes outside the aircraft, which in severe icing can also be unreliable).

(With the autothrust selected off and the crew confidence that they were not in a stall, no stall recovery inputs were made. Which is good, because the result of stall recovery inputs when you are not in a stall could be an overspeed, and in alternate law, if I understand M. Houang correctly, the high speed protection warnings are reduced).

In response to the two incidents (the second one is not described, but the partial aircraft ident "PTP" is given), Air Caraïbes quickly replaced the pitot tubes on its fleet with a different kind, with drainage designed especially for the heavy precipitation and severe icing encountered.

In October, Air Caraïbes officials met with Airbus representatives, who understood that the stall warning messages were contradictory and that the checklists were difficult to complete rapidly. They said they'd think about modifying the checklists.

And then M. Houang says he hopes the memo has answered everyone's questions, and wishes everyone good flights.

I expect a lot more Airbus pitot tubes to be replaced now, whether or not the similar events preceding the loss of AF447 is determined to stem from the same cause.

Pretending to Fly

Reader Mark Richards asked me recently about the flight simulator game that I use to practice my instrument procedures with, and I know Sir Lukenwolf is following hopping around the north in a flight simulator, too. Mine is Microsoft Flight Simulator 2002. I know any real sim enthusiast probably has much newer software, but I expect that sim hobbyists hold in fairly high priority the realism of the simulated scenery and audio. I surmise this because I've owned several versions of MSFS and each update has much more detailed scenery than the last. The amount of required processor power also increases with the version, which is why I don't own a more recent version.

Almost every airport I've ever used is in the database. That's improved over the years, too. Unfortunately a lot of Canadian nav aids are missing from my version, but you can download patches (thank you, fanatical Canadian simmers) to add them in. You can also download aircraft to fly, and customized panels, but I just find an airplane that will fly at the normal approach speed I use, and use it, even though the power settings might be different. I don't really need to practice the manipulation of the controls so much as I need to practice the focus and pacing and remembering of all the details of the procedures.

Probably eighty percent of what I use MSFS for could be accomplished with pencil, paper and a FlightSafety poster of the aircraft layout, if I'd just sit down and spend the time concentrating. The simplest advantage of the game is that it's a video game, so holds a person's attention better, because there's the payoff of 'really' intercepting the glideslope. It also has the advantage of showing you your track, so if you got confused in strong winds and ended up outside of protected airspace, you can look at the log afterward and see the point at which you turned at the wrong time and your hockey stick procedure turn became a tennis racket.

Mouse on screen is such an awkward interface for tuning instruments that I just use the keyboard, often pausing the game in order to do so, and at the same time reaching my actual arm to where the real control is in the cockpit. Usually that doesn't really matter, because the instrument adjustment details vary from airplane to airplane anyway. I often trim by turning on the autopilot and then turning it off again, or just fly by telling the autopilot where to go, as I would on a single-pilot IFR ride. I have the throttle and flaps and brakes all programmed into the game controller, which is labelled all over with DYMO tape, but I go for the keyboard if it's not working out. The biggest interface peeve I have is when the same keyboard command does an action and reverses that action, whereas in the airplane the commands are different. The biggest example is the gear. The G-key, or whichever joystick button I map to it, both raises and lowers the gear. That means that if I hit the G, or touch the G-mapped yoke control, a second time, in what to me is a simulation of placing my hand on the gear lever to confirm it is selected down, or if I accidentally double touch the G at first extension, I have the gear up instead of down. I don't know why they did it that way. Why not have a separate command for gear up and gear down?

Another drawback is that because it's not real, you're free to do stupid things, self-vectoring yourself through IMC for an ILS approach instead of doing a procedure turn, continuing a botched approach if the VOR is at full deflection instead of doing the missed, or flying an approach you made up yourself. There's the risk that practicing these things and having them work out will transfer psychologically to the real airplane, making me think I can be an idiot in the air, too. The game is hilariously forgiving about things like landing crooked and runway excursions, and no effort has gone into placing dread in your soul when you crash.

A typical session for me is to start up at or near an airport I'm expecting to work at soon. I set the weather to the lowest minima for the approaches, brief and fly an instrument departure, intercept an airway, fly a hold, then return to the airport for an NDB approach. The weather will be too low, so I'll miss, and then come back for the approach that uses the minima that should get me in if I fly it right. If it's a tricky approach with a dogleg at the beacon, or a circling, I will probably do it first with the scenery turned on, just to see how it works, but then I'll close the scenery window and just do it by the instruments. It's fun to have the weather turned on and watch the raindrops on the windshield. That's fun in real life, too. Some of the scenery is actually good enough to be fun to watch, even though it's only a game. Unfortunately, of an hour I spend with the simulator, ten or fifteen minutes always ends up spent mucking about with the software, just because there are so many options to play with and cool things to look at.

I once paid money for a Twin Otter add on but I found it frustratingly unusable. The designer had tried to make the controls look as much as possible like the actual airplane, which required me to load different pages and wait in order to do something simple like toggle bleed air, which in real life you can do without looking, just by raising your hand. Overdoing the reality detracts from the experience. But you've got to love the "squeak squeak!" tire sound when you touch down on the water in a float plane. I'm sure they've fixed that by now, but it was funny in the first version with float planes.

Once I was flying a simulated winter flight way up north between something like Alert and Cambridge Bay, and I was really startled to see the northern lights. They were done well, not overdone, and it might have taken an hour of night arctic flying for them to appear. I'm still not entirely sure I wasn't making it up. Perhaps someone will confirm that there really are northern lights in the game.

As for the question Mark asked about Microsoft ending support for the product, I don't see it really relevant to me. My CD-ROMs of the product won't stop working. If it doesn't run on future versions of the Microsoft OS, I can probably find some clever hack to work around that, like I use for Commander Keen. I don't know its value for money because it was given to me by a friend who is a Microsoft employee. He would give me a more current one, but this one is good enough for what I do.

It Was Right Next to the "Fly To Calgary" Button

The Aviation Herald reports a little excitement on a Westjet Flight departing Abbotsford BC.

Incident: Westjet B736 at Abbotsford on Jan 20th 2009, navigation system flies different route than pilots

The crew of a Westjet Boeing 737-600, registration C-GWSB performing flight WS-456 from Abbotsford,BC to Calgary,AB (Canada) with 104 people on board, noticed during an RNP (Required Navigation Procedure, precision RNAV navigation) departure, that the airplane was not following the programmed standard instrument departure route. The airplane was below minimum safe altitude. The crew contacted Victoria Terminal and was issued radar vectors away from high terrain. The airplane continued to the destination for a safe landing.

Subsequently it was established, that the engine out procedure had been inadvertently activated in the flight management system.

Now you see why a pilot might want to shoot the FMS?