Sunday, December 28, 2008

A Resolution

The aviation blogosphere is a fairly small place, and the closest thing it has to an organization is a Pilotbloggers email list I subscribe to. A bunch of the other members are participating in a mass post today regarding New Year's Resolutions. Well, I don't normally do New Year's Resolutions - I moreso have a master list of long-term goals that get updated occasionally. But I haven't been a very good member of the aviation blogosphere lately - little reading, less linking - so I feel I should post something today.

My initial thought for a resolution is to remain employed as a pilot by the end of 2009. That's not really a good goal, though - there are too many factors outside my control. So that's more of a hope and a prayer than a resolution.

So my official resolution to finish the book by the end of the year. Not published, just finished.

Saturday, December 27, 2008

About That Pilot Shortage...

Since even before I started flying in 1994, I've been hearing about an ever-looming severe pilot shortage. This is a theory that has been promoted most vigorously and infamously by Kit Darby of Air, Inc, but has been repeated at length by various aviation publications and eagerly endorsed by the flight training industry. The idea is that a large contingent of Vietnam-era airline pilots are approaching retirement age at the same time that industry growth creates a need for additional pilots, and not enough new pilots are being trained to fill both needs. The effects will be intense, widespread, and will have far-reaching implications for the airline industry, say the proponents of this theory.

I've seen two large booms in airline pilot hiring in the last 14 years which were, at the time, thought to be the onset of the predicted pilot shortage. Graph 1 below tells the story of boom and bust; it is based on hiring data published by Air, Inc, which despite a minor conflict of interest between collecting/publishing hiring data and selling interview prep materials to aspiring airline pilots remains the industry's most authoritative source of hiring data.
After war, terrorism, and economic turmoil led to the shutdown of Eastern and PanAm and the furlough of thousands of pilots at the other major airlines in the early 90s, steady economic growth paved the way for the airlines' comeback in the middle and later parts of that decade. By 1995 most furloughees had been recalled and major airline hiring increased steadily from 1200 pilots in 1994 to 5000 pilots in 1999. From 1997 to 2000, hiring at national and regional airlines ($100M to $1B in annual revenue) outpaced hiring at the majors, largely driven by the advent of 30-50 seat jets at the regional airlines. 2000 was the height of the hiring boom. Although major airline hiring had eased somewhat, the regionals and nationals more than made up the difference and pushed airline hiring to an all-time high of over 11000 pilots. At the time it looked like the onset of a major shortage. Retirements were just starting to increase and the majors were forecast to grow significantly in the next decade; the regionals frenetically hired pilots to replace those who moved on to the majors, and many were hiring pilots with as little as 500 to 1000 hours. Flight schools had a horrible time keeping instructors around; as soon as they accumulated a few hundred hours, they were off to the airlines. It seemed like a very good time to have one's foot in the door of the aviation industry.

Even if 9/11 hadn't happened, it's likely that the collapse of the dot com bubble economy would've still devastated the airlines. The planes were fairly full within a few months of the attacks; the expense-account business travelers, however, did not come back (or rather, modified their travel purchasing habits), a trend that had already begun in early 2001. The major airlines had built their business models around these travelers in the 90s, and their disappearance devastated the industry. In this difficult, uncertain environment, the major airlines made significant cuts to their flying, which resulted in many thousands of furloughed pilots and the total collapse of hiring between 2000 and 2003. Hiring minimums at the few airlines still recruiting shot skyward.

The second major boom took place only last year. The data in the years leading up to it is more confused than it was in the 90s and requires some explanation. The majors' hiring trended upward after 2003 and was back at moderate levels by 2005. This was driven entirely by the growth of low-cost carriers like Southwest, AirTran, and jetBlue, and freight carriers FedEx and UPS. The "legacy" airlines had no hiring; in 2005, four of the six were simultaneously in bankruptcy and all still had thousands of furloughees on the street. The expanding low-cost and freight carriers hired largely from this pool of furloughees. The regionals and nationals saw fairly low attrition during this time; the large hiring bump from 2002 to 2004 was due to growth thanks to the relaxation of scope in the major airline pilots' post-9/11 concessionary contracts. As these new air regional hiring leveled off in 2005 and 2006.

In 2007 the dynamic changed. By the end of 2006, all the legacies were out of bankruptcy with drastically altered cost structures, and the low cost carriers' growth slowed. Most of the legacies' remaining furloughees were recalled, and by mid 2007 most of the legacies were hiring again for the first time in six years. This time, the majority of civilian newhires came from the regional carriers, and attrition there skyrocketed. This happened at the same time that many regionals were expanding due to the delivery of 70-86 seat jets that were allowed to be outsourced by contracts imposed in the majors' bankruptcies. The result was that the national and regional airlines' hiring doubled between 2006 and 2007, and they suffered a pilot shortage more acute than 2000-2001. The regionals couldn't find enough pilots to fill their classes; hiring minimums fell to the bare legal minimum for the first time in 40 years and handsome signing bonuses were offered to applicants with prior airline experience. There were discussions within the industry that the lack of qualified pilots posed a threat to some regionals' very survival, particularly those with high attrition. This rather acute shortage ended abruptly in the spring of 2008 as high fuel prices brought a halt to hiring at the majors and made the 30-50 seat RJs horrifically costly to operate on a per-seat basis. When fuel prices eased in the latter part of 2008, it was in response to a collapsing economy that posed an equal threat to the airlines; thousands of pilots are again on the street, this time at both majors and regionals. At the same time, retirements are way down thanks to a recent increase in the mandatory retirement age from 60 to 65. Hiring prospects for 2009 look bleak.

With so many surplus pilots during two periods of a single decade, it's easy to dismiss talk of a pilot shortage as nothing but snake oil peddled by the flight training industry, and many pilots do think it was nothing more than the brainchild of Kit Darby. I think it depends on how you define a shortage. The hiring booms clearly had some of the makings of a shortage; neither bubble, however, constituted a truly acute, industry-wide shortage as had been predicted. The boom of 1994-2000 was widespread and prolonged, but was never any sort of threat to the industry. Hiring minimums decreased at both majors and regionals, but neither had much trouble finding qualified candidates. In 2007, the shortage was very acute but limited strictly to the regional airlines; the majors had a huge pool of qualified, motivated candidates to choose from. I believe that these two scenarios are as close to a pilot shortage as we will ever come in the United States. I also believe that the second type of pilot shortage will return with a vengeance several years from now.

My reasoning is based on the supply of new pilots and how it fluctuates in response to varying hiring markets. There isn't any available data on how many people begin flight training each year with a goal of flying for the airlines, but we do know how many commercial pilot licenses the FAA issues every year. Clearly not everyone who is issued a CPL intends to fly for the airlines or even fly professionally, but given that every would-be professional pilot must first earn a CPL it seems reasonable to use this as a gauge of the supply of pilots entering the industry. Refer to Graph 2 below, which charts annual CPL issuances against total airline hiring.
One can see that there is a fairly strong correlation between the strength of the hiring market and the number of CPL issuances. There is also a pronounced 2-3 year lag between the peaks and valleys of hiring and those of CPL issuances. Both conclusions make sense: obviously people find it most worthwhile to invest time and money into flight training when the aviation industry is doing well, and it may be several years between that decision to enter aviation and earning one's CPL. By comparison, the number of Airline Transport Pilot (ATP) certificates issued tracks very closely to the strength of the hiring market with little lag (see Graph 3 below). Once one has the required flight time for an ATP, the training is quite short compared to the CPL. So from these two charts, it is obvious that both pilots and potential pilots take the strength of the hiring market into strong account when deciding whether to invest in training, but the actual supply of new commercial pilots is affected only several years later when the hiring market may be entirely different.By the time the first hiring boom reached its peak in 2000, hiring had been increasing very steadily for six years. Accordingly, the number of CPLs issued had increased since 1997, and there was a rather steady supply of new pilots that prevented the boom from ever becoming a severe shortage. Even if 9/11 and the dot com implosion never happened, I suspect that flight training could've easily kept up with airline growth and retirements. There is a lot of flight training infrastructure in this country that can be utilized when times are good and people choose to invest in flight training (incidentally, the lack of such infrastructure is why the pilot shortage is very real in overseas emerging markets). I seriously doubt that any extended period of growth and retirements will translate into a severe, prolonged pilot shortage.

The boom of 2007 was different. Although airline hiring had increased between 2002 and 2006, it was not the sort of steady growth across all sectors that characterized 1994-2000, and it was a period of instability for the airlines. The woes of the legacies and the plights of their pilots - contracts gutted and pensions terminated in bankruptcy - were very much in the news. These are not the conditions that inspire people to sign up for flight training. Accordingly, CPL issuances reached their post-9/11 nadir in 2006 and posted only a modest gain in 2007. When the regional airlines suddenly had to hire 6800 pilots in 2007 - double that of 2006 and 235% more than Air, Inc had predicted at the beginning of the year - several years of depressed training translated into a dearth of qualified pilots. Hiring minimums plummeted, and those airlines with the highest attrition (i.e., most pilot-unfriendly management) were forced to hire pilots with the minimum legal qualifications - and they still couldn't find enough pilots. If this hiring environment had continued - if high oil prices and the economy hadn't killed of the demand for pilots - it is likely that CPL issuances would've continued the upswing seen in 2007-2008 and eventually satisfied the industry's demand for pilots. In the meantime, though, the shortage would've posed a distinct challenge for the airline industry and a distinct opportunity for pilots.

So what does the future hold? I think 2009 is going to be a very rough year for airline pilots. Although oil prices have fallen significantly, it's anybody's guess as to when and where the economy will bottom out and begin its recovery. All the airlines are forecasting significantly decreased revenue for next year; many forecasts have revenue falling more than in 1987, 1991, or 2001-2002. Accordingly, the airlines have made significant capacity cuts with more probably on the way. To the airlines' advantage, the credit crunch should ensure there are no upstarts to play spoiler by increasing overall capacity like in 2001-2006, but this also means there will likely be no hiring at low cost carriers to absorb the legacies' furloughees. Many regional airlines, too, are facing contraction as their major airline partners try to reduce their exposure to 30-50 seat RJs. Meanwhile, the change to the retirement age means retirements will be significantly below normal until 2012. My guess is that overall hiring for 2009 will be significantly lower than 2002, with a depressed job market for several years after that.

Eventually, though, the shortage at the regionals we saw in 2007 will return, and it will be worse next time. Although CPL issuances will continue to rise for another one or two years, after that you'll see a sharp drop like that in 2003. I think the training slump will be much lower than it was in 2003-2006. Back then, there were still jobs available at the regionals even if the majors were still in turmoil; this time there will be few airline jobs of any description for several years straight. The cost of flight training has increased significantly in the last few years. It will also be harder to secure loans for flight training as long as the credit crunch endures. During the next few years, there will be a lot of people stuck flight instructing, flying freight, etc, but this surplus will only obscure the fact that there are fewer and fewer new pilots being trained.

When the economy begins its recovery and the airlines start adding capacity back into their systems - several years from now, I'm guessing - they will initially have no problem finding qualified candidates for their openings. The majors, of course, will hire from the huge cadre of regional airline captains chomping at the bit to move up to the next level, and the regionals will be able to draw from the pool of flight instructors and freight dogs that formed during the years without airline hiring. If the hiring ramped up slowly and steadily, I would expect the training of new pilots to increase enough by the time this pool is depleted that the boom would never become a shortage. There is a complicating factor, though: the first wave of airline pilots to hit age 65 will retire in 2012 and double or even triple the annual number of retirements. If this happens concurrently with an industry growth period, attrition at the regionals will skyrocket and quickly deplete the pool of experienced pilots. I think you'll see a very similar shortage to 2007, but without several years of preceding hiring activity like 2004-2006 to stimulate new training. You'll see a return of 250 hour commercial pilots to the right seat of airliners (or even less qualified, if the ATA gets it's way on Multi-Pilot License). You'll also see a return of signing bonuses for experienced pilots. This situation will persist for several years until the people it lures into aviation complete their training. It will be a unique opportunity for regional pilots to get better contracts without fear that their flying will be replaced by a lower-cost regional.

In the meantime, those of us at the regionals aren't going anywhere - if we're lucky. I personally don't have a great deal of faith that I'll be employed at NewCo this time next year, in the left seat anyways. If that happens, there's a good chance I'll be heading overseas - where the pilot shortage is very real indeed and still ongoing, and a pilots' services are therefore valued more highly than in the erstwhile "land of opportunity".

Monday, December 08, 2008

Earning My Keep

On Friday I was in San Antonio on a long layover; we had a van time of 4:45pm for a 5:51pm departure to Minneapolis. I slept in that morning and woke too late to do anything interesting like go to the Alamo or Riverwalk. Around noon I met Brent, my FO, in the lobby, and we walked over to Bill Miller's BBQ for their wonderful chopped pork sandwich platter. Over lunch, we had a conversation wherein Brent said something interesting that sparked an idea for a blog post. As soon as I got back to the hotel room I pulled out my notebook and began writing.

I should have given the idea more time to percolate before I started. I wrote several pages of disjointed paragraphs full of half-baked ideas and contradictory proclamations, despaired on my first reading, threw the pages away, and started over with a new preliminary outline. The post I sketched out didn't seem much better than the one in the trash can. I wanted to write a thoughtful essay about the thorny subject of pilot pay. I wanted to move beyond the usual arguments about being overpaid or underpaid and explore the question of what exactly pilot pay is really based on. Is it knowledge? Experience? Certificates? Responsibility for lives and property? Ability of the aircraft to generate revenue? Difficulty of the job? A life spent on the road? Negotiating leverage? Free market value? I doubted whether there's really any subjective answer; you'd certainly get differing opinions from pilots, management, unions, and the traveling public. The framework for an interesting essay was there but I was getting nowhere in fleshing it out. I realized that I've written enough on the subject over the years that I should only return to it if I have something to add. Also, most of my readers are a whole lot less interested in the question of whether I'm paid enough than I am, so any further exposition on the subject ought to at least reward the reader with decent writing. This clearly was not it. Frustrated, I set the notebook aside and did a crossword puzzle instead.

At 4:20 I repacked and changed into my uniform and then met my crew in the lobby; we departed for the airport on time and arrived at the gate just as the inbound crew disembarked. They reported that the ship was "clean" and the ride was good. I checked the radar online; an icy blue smear nearing Minneapolis provided the first clue of things to come. Sure enough, the TAF called for snow to begin falling before our arrival, although visibility and ceiling were respectively forecast to stay above one and a half mile and 1500 feet. Our alternate was LaCrosse, Wisconsin, a small airport about 100 miles southeast of Minneapolis with a similar forecast. Our dispatcher thoughtfully provided us with 800 pounds of holding fuel along with 1500 pounds of contingency fuel. The difference between the two is a fine but important distinction. Holding fuel is considered part of the "minimum fuel load" listed on the release, while contingency fuel provides an additional buffer above the legal minimum.

The departure from San Antonio was uneventful; Brent was flying the leg. This was day three of a four day trip, Brent had several hundred hours in the airplane, and I was comfortable with him landing on a snow-covered runway in Minneapolis. On a leg like this, where there are clues that some Captainly wiliness might be required, I'd just as soon be non-flying pilot anyways. As the western sky grew dim, I set the ACARS up to automatically retrieve the Minneapolis ATIS every time it was updated.

The first two hours of the flight were quiet except for the frequent annunciation of a new ATIS. The snowfall dropped the visibility to three quarters of a mile at first but then it rose and held steady at around one and a half miles; the ceiling stayed pretty high. We could clearly get in but my Captain's antennae were still beginning to twitch. For starters, nearly every ATIS update - specials were coming out every 20 minutes - showed runways 12R and 12L being alternately closed for snow removal. Minneapolis cannot function on one arrival runway without significant delays, as the construction on 30L/12R so convincingly proved in 2006. I decided to climb and slow down a bit to save fuel. We checked on with Minneapolis Center with nary a word of delays. By the time we turned over Fort Dodge IA onto the TWOLF One arrival, I thought that perhaps we'd dodge the bullet yet.

"Newco 1917, got a new route for delays into Minneapolis, advise ready to copy." I groaned and grabbed the clearance printout and a pen. "Newco 1917, cleared to Minneapolis via Sioux Falls, Redwood Falls, SKETR Three arrival." Whoa. I knew Sioux Falls was quite a ways northwest of our position. I pecked "FSD" into the FMS and requested a direct track; sure enough, it appeared on the moving map nearly 90 degrees to our left and over 100 miles away. As soon as I inputted the rest of the arrival, I started crunching the fuel numbers. They weren't pretty; the reroute had decimated our contingency fuel and added nearly 30 minutes to our flight. I texted the new routing and landing fuel to our dispatcher via ACARS.

The flight plan attached to our release normally tells us minimum fuel for each waypoint, but we were no longer on our planned route so I had to work up new numbers. I started with my personal fuel minimum for arrival at MSP. In this case it happened to coincide with the dispatcher-calculated minimum of 5400 lbs, only because of the 800 lbs of holding fuel. If we ended up holding, the dispatcher might amend the release to turn the holding fuel into contingency fuel, reducing our minimum fuel at MSP to 4600 lbs (2400 lbs fuel burn to our alternate plus 2200 lbs reserve). If that happened, however, my personal minimum at MSP would remain 5400 lbs because there's no way I'm going to be shooting an approach in crummy weather at the alternate with less than 3000 lbs in my tanks. After all, the company defines 3000 lbs as a "minimum fuel state" and 2000 lbs as an "emergency fuel state." Working backwards from MSP, I calculated the minimum fuel number for several waypoints on the arrival. Crossing Sioux Falls, we were only 400 lbs above the minimum. As we turned eastbound again, I told Brent to slow down more to take advantage of the strong tailwind and save fuel.

"Newco 1917, holding instructions for delays into Minneapolis...hold southwest of SHONN as published at flight level 350, leg lengths your discretion, expect further clearance 0253 zulu, time now 0205 zulu." I read back the clearance but added "We'll have to run the numbers and get back to you on our fuel state." We were going to arrive at SHONN nearly 30 minutes before our EFC time, with only 500 lbs over the minimum fuel at SHONN. I told ATC we had enough fuel for perhaps 10 minutes of holding before diverting and asked if Minneapolis Approach might take that into consideration. The controller sounded doubtful but said he'd check.

I texted the news to our dispatcher along with a request for updated weather at KLSE as well as KRST. My thinking was that if the weather was good enough at Rochester, we could make it our new alternate; since it's closer to MSP, the decreased fuel burn to the alternate would free up some fuel for holding. Our dispatcher responded with weather for Rochester and Brainerd, about 100 miles northwest of MSP, but not LaCrosse, and asked me to pick KRST or KBRD as my new alternate. Rochester had 3/4 mile visibility, actually below alternate minimums. Brainerd was better but almost as far away as LaCrosse. I texted that I'd just keep KLSE, and the dispatcher responded by informing me that LaCrosse was down so he was changing my alternate to KBRD. Ah...okay then. It didn't save me any fuel but at least the weather was nice there!

We were rapidly approaching SHONN so it was time to tell the passengers something. I wasn't going to mention the possibility of diverting just yet, but they had to be wondering why we weren't on the ground yet. "Folks, from the flight deck...that forecast snow in Minneapolis showed up a few hours ago and is making a pretty good mess of things, with lots of delays going into MSP. We were given a reroute over Sioux Falls which helped, but air traffic control still needs us to hold over a point about 50 miles southwest of MSP for up to a half hour...." I glanced at Brent who was gesturing wildly to me and released the PA button in time to hear ATC cancel our hold. "...Uh, nevermind folks, they just cleared us to come on in, hopefully without much further delay. Flight attendants please prepare for arrival."

We made up some more fuel between SHONN and the airport thanks to our slowed speed, the tailwind, a delayed descent, and an early vector to the downwind. ATC put us on a nearly 30 mile final for runway 12L; 12R was closed for snow removal. We were another 15 miles along when ATC broke us and a preceding NewCo flight off the approach because another aircraft reported nil braking, closing the runway until the snowplows could clear it. The other NewCo flight diverted to their alternate immediately. I told approach we had a little fuel to work with so long as there wasn't much delay in opening 12R. They assured us it'd be open soon and resequenced us onto a 20 mile final.

This time we got almost to the FAF before approach control again broke us off the approach; the snowplows weren't going to have quite enough time to get off the runway. I reconsidered our fuel status. The fuel burn to Brainerd was 2200 lbs, so I wanted at least 5200 lbs on landing. We had a few hundred more than that, enough for six minutes or so. "We can take another go if it's a short final, otherwise we need to go to Brainerd." ATC assured us they'd keep us close in and they lived up to their word. We got the runway in sight a few miles out, and Brent made a nicely firm landing. Despite the heavy use of thrust reversers, the ship was rather slow to decelerate; as soon as I took the controls, I could see why. The runway was quite slick though it had just been plowed. The anti-skid surged in and out rapidly as I mashed down hard on the brake pedals, a rather unnatural act of faith. I slowed to a nearly complete stop before even beginning the turn off the runway, and just as well, for the taxiways hadn't been cleared recently. The centerlines were buried so I stayed in other planes' tracks on the way to the gate. The airport was eerily calm, and I thought that it looked like a ghost town with a two inch layer of chalky dust covering the abandoned wagons and buggies.

No time for such fantasies; we were pretty late getting in, and due out to Missoula in less than a half hour. Almost as soon as our passengers were deplaned, new ones trudged wearily onboard. Brent and I went through our familiar preflight dance efficiently, with the addition of a radio call to the Iceman. Paperwork complete and handed out with a wave to the harried gate agent; door slams shut, made up some time on the turn. We taxied out, got deiced, and took off into a post-apocalyptic orange-hued overcast with snowflake constellations warping past in the momentary glare of our landing lights.

Not until we settled into the reverie of a darkened cockpit in cruise did my thoughts wander back to the previous flight. Pretty uneventful, really - got delayed a bit, decided we had enough fuel, landed. Lots of other planes out doing the same thing, and lots of other snowy nights left in this Minnesota winter. To be sure, the events of the flight drew upon my knowledge, experience, and judgment much more than your average sunny-day flight. Truth is, if every day was sunny and trouble-free, a monkey could do my job. But they are not, so he cannot. It takes an army of well-trained, experienced pilots to keep the fleet flying safety and efficiently in all sorts of conditions. The fact that flying is so routine, so safe - the fact that flights like the last one barely merit mention - is perversely held up as proof that a pilot's job is easy and pilots are overpaid!

So are pilots today underpaid? I think so, but plenty of others disagree. I can tell you this: the current level of pilot compensation is not attracting many new people into the industry. We saw a glimpse of the consequences as recently as this summer, when regional airlines were so short on qualified applicants that they were hiring 250 hour freshly minted commercial pilots into the right seats of RJs captained by 2000 hour freshly minted ATPs. The pendulum has since swung the other way, and there are thousands of qualified pilots on the streets looking for nonexistant jobs this holiday season. There are not, however, that many people taking out $60,000 loans to learn to fly these days, so the pilot shortage will return with a vengeance when the current economic distress subsides. When that happens, I would think twice about putting oneself or one's family on a regional jet on any dark and snowy winter night.

Sunday, November 30, 2008

Out East

Almost as soon as we were through 10,000 feet, I had my atlas out. It was completely clear along our route - a rarity in November - and on the dark, moonless night, every light for hundreds of miles was visible. Every town and road printed on the map was duplicated in orange and white pixels below. From Memphis we headed northeastwards on J42, toward Nashville. It grew steadily brighter until we passed overhead; then the glow of suburbia quickly gave way to the isolated pinpricks of eastern Kentucky's rural hill country. Louisville floated by to our left; to our right, Knoxville illuminated the spine of the Great Smokey Mountains. My First Officer broke the silence to point out his hometown where Tennessee, Kentucky, and Virginia all come together. At Beckley, WV, we made a slight right to cross the Ridge-and-Valley Appalachians just north of I-64. A thin band of light stretching as far as the eye could see marked the Great Valley; the dark band after that, the Blue Ridge. We reentered "civilization" over Charlottesville, VA, the pixels becoming ever thicker and brighter as we approached Washington DC. From 35,000 feet, we could clearly make out the dark National Mall and the floodlit Capitol Building. As we passed over Chesapeake Bay into Delaware, I noted how Washington, Baltimore, Philadelphia, Newark, and New York essentially form one contiguous, enormous metropolis. The Jersey shore, to our right, was much darker but the casinos of Atlantic City shone brightly, prompting my FO to reminisce about his freight dog days flying into ACY. Now New York's glow, visible 200 miles prior, hardened and formed into city and water, then the individual boroughs, and then mile after mile of individual streets and buildings. I picked out the Empire State and Chrysler Buildings as we overflew JFK and LGA. At the Connecticut coast we turned right; the moonless night and dark waters of Long Island Sound made me think of JFK, Jr. By now we were on our descent; the arrival took us over Providence, RI, which until then had always been a fuzzy spot in my geographical knowledge. We went "feet wet" above the spot that the Pilgrims went "feet dry." The vectors to final took us far out over the Atlantic, almost to the tip of Cape Cod. Our 150 minute night tour of the east coast concluded with a nice view of the Boston skyline from the final approach to runway 27.

Prior to flying for NewCo, I had spent very little time "out east." In fact, the furthest east I'd flown was Grand Rapids, MI, and that was on a cross-country flight I did in college! My subsequent flight instructing, freight dogging, and airline flying was all on the west coast. At Horizon, the easternmost destination for the Q400 was Billings, MT. At NewCo, the majority of our destinations are east of the Mississippi. On the east coast proper, we fly to MHT, BOS, JFK, LGA, EWR, PHL, BWI, IAD, RIC, ORF, CLT, JAX, and MCO. When I began flying to these places, my geographic knowledge was sorely lacking. Now all the pieces are starting to fit together.

When I was flying out west, I had no idea just how good I had it. Flying on the east coast in a royal pain. Out west, it's Direct-To everywhere and talk to ATC once every 20 minutes to change frequencies. Out east, it's nonstop convoluted reroutes, last-minute crossing altitude restrictions, and an endless litany of frequency changes. You eventually give up even trying to check in with certain sectors after having your check in stepped on half a dozen times. Airports like PHL, EWR, LGA, and JFK are delay-prone on good days; throw in some bad weather and you're going nowhere quick. Those airports make me thankful that our turd of a contract at least has "Block or Better" pay.* Our hotels at eastern layovers tend to be in decrepit industrial areas near the airport, and the shuttle van drivers are often surly and hurried.

Despite all that, I still enjoy going east. I guess the novelty just hasn't worn off yet. I enjoy the aerial sightseeing when the weather cooperates. The Appalachians, Catskills, and Adirondacks, while less starkly grand than the Rockies, Sierras, and Cascades, have great natural beauty of their own. The human geography, too, is interesting to me: the patchwork quilt-like farmsteads of rural areas, the little towns secluded in dead-end valleys miles from anywhere, the seemingly endless cities of the BosWash corridor. There's a lot of history out east; I enjoy having a birds-eye view of the landscape that famous battles and campaigns were waged across.

On the ground, there's a lot to do if you make the effort. I haven't spent much time in most of the cities we go to, so they invite exploration. The morning after my sightseeing flight up the eastern seaboard, I woke up early and hopped on the subway to downtown Boston. I reemerged at Boston Common, America's oldest park, and started following the red brick path of the Freedom Trail. It was fascinating to see so many important historical sites in the course of a fairly short stroll. The South Meeting House, the Old State House, Old North Church, and Bunker Hill are all legendary places I've heard about since grade school but have never seen for myself. Walking in and around them made the history seem more real, more palpable; I felt new appreciation for the vision and courage of men like Joseph Warren, Samuel Adams, and Paul Revere.

I have a long Philadelphia layover coming up soon. I hope to make it to the Revolution-era sights there. If they're anywhere nearly as interesting as the Freedom Trail in Boston, it should take some of the sting out of having to operate out of PHL.

*"Block or Better" means that we are paid the greater of scheduled or actual block time. Therefore, when we're delayed we get paid extra, but we don't get shorted for being early. Although Horizon has a much better contract than NewCo, they did not have Block or Better; I got paid based only on the historical block time for the leg (usually less than scheduled block), with no additional compensation for overblocking. If Horizon flew into JFK, LGA, or PHL, that'd be a very bad thing to have in the contract.

Friday, November 21, 2008

The Iceman Cometh

Well, there's no escaping it now. Minnesota has had several days of snow, although none of it lasted more than a few hours; it's currently 12 degrees in Minneapolis. When I got done with a trip yesterday morning, my short walk from the bus stop to our apartment froze my fingers and ears so thoroughly that I didn't venture outside again all day. My six month hibernation begins now.

If I could spend the entire winter curled up inside with a warm blanket and hot chocolate I very possibly would, but alas, the need to earn a living saves me from being a total recluse during winter. Being a pilot is a good job for someone who hates long cold winters but lives in Minnesota, as it gets me above the clouds into sunshine rather frequently and also involves the occasional layover in Phoenix or San Antonio. Unfortunately two of RedCo's three hubs are in northern states, so I do spend a lot of time operating in wintry conditions. This rather frequently involves deicing. I deiced for the first time of the season about a month ago. Given that I hadn't done it since last winter, I took that as my cue to open our Deicing/Anti-icing manual and study up.

As I mentioned in my last post, anti-icing systems on jets are beautifully effective in flight but provide no protection on the ground. For that, we use propylene glycol-based deicing fluids somewhat similar to the antifreeze fluid used in your car. Similar fluids have been used in aviation for quite a while, but for a long time they were poorly understood and their use was nonstandard throughout the industry. After some notable accidents involving airliners attempting takeoff with iced-up wings, the FAA got serious about ground deicing. Now every Part 121 carrier is required to maintain and distribute an FAA-approved Ground Deicing/Anti-icing Manual. It spells out in great detail the various ground and flight crew roles and responsibilities, what parts of the aircraft must be "clean" for takeoff, approved methods of removing contamination, the limitations of various anti-ice fluids, and the checks that must be accomplished before an airplane may take off.

As the name of the manual suggests, there are two stages to ground deicing/anti-icing. Deicing is intended to remove any previously accumulated contaminants that are adhering to the airframe. This may include frost, freezing rain, snow, or airframe ice from the last flight. Anti-icing prevents further accumulation of freezing precipitation between deicing and takeoff. In very mild conditions, both of these steps may be accomplished in a single application of deicing fluid, but more often the steps are accomplished with separate coats of different fluids.

The deice fluid used to remove previous contaminants is called Type I fluid. It is fairly thin, slippery, and is dyed red to help deicing personnel see which parts of the aircraft have already been sprayed. Type I fluid is normally heated between 130 and 180 degrees F, and is sprayed out at considerable pressure to aid in knocking contaminants off the airframe. As Type I fluid isn't very viscous, it doesn't adhere to the airframe well. It also has a somewhat limited ability to absorb moisture. For both of these reasons, Type I fluid is used for anti-icing purposes only for short periods or in very light ground icing conditions.

The more commonly used anti-ice fluid is called Type IV. It has thickening agents added and is dyed green. Type IV is always applied cold after a previous application of Type I fluid. It is sometimes diluted with water, not only to save costs but also to (rather counter-intuitively) lower its freezing point. A solution of 75% fluid and 25% water has a freezing point of around -55C; undiluted fluid has around the same -30C freezing point as a 50-50 mix. Type IV fluid's greater viscosity helps it to better adhere to the airframe. It is designed to shear off at around 100 knots airspeed on the takeoff roll, leaving a clean wing. It also has greater moisture absorption capabilities than Type I fluid. All this means that it protects the airframe against contamination for much longer periods than Type I fluid.

The most common apparatus used to spray deicing fluid is a boom truck. The truck drives around the aircraft while the deicer sprays the fluid from a basket on top of the boom. Both Type I and Type IV fluid can be and often are sprayed from the same truck in subsequent applications. A fairly recent development is infrared deicing. Infrared deicers melt existing snow and ice off the airframe as the plane taxies through a hangar-like structure. This arrangement still requires a truck to spray Type IV fluid for anti-ice protection. There are only a few infrared deice installations in the US.

Occasionally ground crews will deice aircraft before the flight crew arrives in the morning if there was frost or snow accumulation overnight but ground icing no longer exists. Usually, though, crews must request deicing. At small stations, the ground crew needs advance warning to make sure the truck has enough fluid and to heat it up. If there are any contaminants adhering to the airframe, or if there's steady snowfall, the decision to deice is an easy one. It's less clear when the temperatures are just warm enough that the snow is melting on contact, or if flurries aren't quite sticking to the airframe. You don't want to deice unnecessarily due to the time and cost involved, but you also don't want changing conditions to make you come back for deicing at the last minute either.

At outstations, I'll inform operations or a member of the ground crew that we need to deice at least 30 minutes before departure. At most of these airports, we get deiced right after pushback from our gate, before we've started our engines. At the smallest stations, the same rampers that just loaded the bags and pushed us back will then hop in the deice truck. At large northern outstations and hub airports, there are dedicated deice pads located near the end of the runways and manned by a virtual army of trucks and staff during winter weather. At Minneapolis, there are four deice pads that can each handle half a dozen aircraft simultaneously! The usual drill is to call "The Iceman," or central coordinator, as soon as you know you'll need deicing. They'll either assign you a pad or dispatch a truck to deice you near the gate when you push back. If assigned a pad, you confirm it with the Iceman when you begin taxiing and then contact that pad on a discrete frequency to be assigned a lane. You usually deice with engines running at deice pads.

Before the crew begins deicing, the deice coordinator will plug into the ground intercom or transmit via radio to confirm what type of fluid we want and that we are configured for deice. This involves ensuring the flaps are retracted, the stabilizer trim is full nose-down (to prevent fluid from entering internal mechanisms), and turning off all the bleed and pack switches. This last step is especially important to prevent passengers from breathing atomized glycol. Unlike ethylene glycol (antifreeze), propylene glycol is supposed to be fairly non-toxic - but it's not exactly pleasant to inhale, either. After deicing is complete, we wait a full minute before selecting bleed sources on, and another minute after that before turning the packs back on.

Once the process is complete, the deice coordinator will tell us the type and amount of fluid used, the fluid/water mixture and freezing point, and the time that fluid application began. The last bit of information is especially important, for it is the time that our holdover time is calculated from. Holdover time is the length of time that the fluid can be expected to provide anti-ice protection in active ground icing conditions. It varies by type and mix of fluid, type and intensity of precipitation, outside air temperature, and a few other factors. We use a series of tables to calculate our holdover time for the given conditions. The tables give us a range of times; for example, with moderate snowfall at a temperature of -5C, Type I fluid has a holdover time of five to eight minutes and undiluted Type IV has a holdover time of 20 to 40 minutes. The Captain chooses a holdover time within this range that most closely corresponds to prevailing conditions, and may adjust the time upward or downward in response to changing conditions. The holdover time begins when the application of the last coat of anti-ice fluid begins; takeoff must be commenced before the holdover time expires.

Of course, in any weather that makes holdover time a factor, there are likely air traffic delays that may hinder a timely takeoff. ATC is usually pretty good about asking pilots their holdover expiration and resequencing aircraft if necessary. Sometimes best efforts aren't enough and the holdover time expires. Different airlines handle this in different ways. Some require the aircraft to return for secondary deicing. Others, including NewCo, permit the use of a Pre-Takeoff Contamination Check. This is a tactile exterior inspection conducted by trained deicing personnel to confirm that the aircraft is still clean and the fluid has not failed. If the aircraft passes this check, it is still allowed to take off so long as the takeoff happens within five minutes of the check. Close coordination between the crew and operations is necessary to ensure there are personnel available at the end of the runway in case holdover time expires.

In any case, shortly before takeoff the flight crew must perform their own Pre-Takeoff Check to ensure the fluid has not become saturated with moisture. Basically you just look at the few airframe parts visible from the cockpit and observe the sheen of the fluid. If it is dull and milky, the fluid has failed and the aircraft must be deiced again. If it's still glossy, then you're set to go. Hopefully you're headed South where you can thaw a bit. Unless they're having a cold snap, that is; the only thing worse than winter weather in the North is winter weather in the South! While Minneapolis and Detroit are true pros at fast, effective deicing, the same is decidedly not true of places that only see winter weather once or twice a year. They simply aren't equipped or staffed for it, and there's always a learning curve.

Portland is one notable exception. They're blessed with fairly mild winter weather but when they do get it, it's in the form of truly nasty ice storms. Accordingly, Horizon puts heavy emphasis on training and equipping their PDX deicing crews for those days. In the end, it's usually all for naught; they soldier on for a few hours before freezing rain closes the airport entirely and grounds a large portion of Horizon's fleet. A few years ago a friend of mine was the FO on the last flight that landed before a particularly nasty ice storm closed PDX for several days. In the last few minutes of the flight, they encountered heavy freezing rain; by the time they landed, every inch of the airframe was encrusted in several inches of ice! The deicing team's last act before conceding to Mother Nature was to deice my friend's CRJ so they could get the main cabin door open to deplane the passengers.

Thursday, November 06, 2008

JungleBus Systems Post: Ice Protection

I recently had a passenger stop at the flight deck during deplaning to say hi to me and the FO. It turned out that he's a Private Pilot with an instrument rating, and flies a nearly-new Cessna 172 with a Garmin G1000 glass cockpit, traffic information system, and weather datalink. We marveled at the advances in general aviation avionics the past few years, and I remarked to him that his cockpit was as sophisticated as ours for a mere fraction of the cost. "Yeah," he said wistfully, "but it's still a single-engine piston, and despite all the goodies I still can't fly it half the winter."

It's a good point. Advanced avionics have brought light aircraft up to transport category standards in many respects and greatly improved their usefulness, but icing remains a significant problem for light aircraft. Despite some advances in technology to make anti-ice equipment lighter and cheaper, only some light twins and very few singles are approved for flight into known icing. A lot of the equipment is of dubious effectiveness, and prudent pilots don't remain in icing conditions for long even in "known ice" airplanes. Transport category jets, on the other hand, have had icing pretty well licked for over 40 years. This is in large part due to their superior performance: icing conditions tend to be pretty localized both by area and altitude, so aircraft with plenty of speed and power can blow through icing too quick for it to be much of a problem. Icing at cruise altitudes is rare for jets; the air is usually too cold to support enough moisture for significant icing. Jets have a significant advantage in the equipment department, too. They enjoy a steady supply of hot air from their engines' bleed valves, which is used to heat the wings, tail, and engine inlets. It's been an effective system since its inception, and has been adapted on most jet aircraft from the DC-8 until today - although the B787 will soon be a notable exception.

The JungleBus is no exception. Its ice protection equipment is pretty standard for a jet, although the operation is more automated than most aircraft. The leading edges of the wings are heated by bleed air via the pneumatic system; the engine intakes are heated by air that comes directly from the 10th stage compressor bleed. Both vertical and horizontal stabilizers are unheated; this is noteworthy but not unprecedented. The manufacturer had to prove during certification tests that the aircraft was not prone to tailplane stall or control problems with unusually heavy ice accumulation on the tail. I know - I'm not utterly convinced, either. The remainder of the anti-ice system is electric; protected areas include both windshields, the Air Data Smart Probes (pitot/static/AOA) and True Air Temperature (TAT) probes.

A key difference between hot-wing anti-ice systems on jets and the inflatable rubber de-ice boots used on smaller aircraft is when they should be turned on. Despite some controversy on the subject, most pilots still wait to inflate their de-ice boots until there has been some accumulation of ice. Anti-ice systems, however, must be turned on at the first sign of icing (or before). You don't want ice to build up on the cowl inlet only to be ingested into the engine when the cowl is heated; it can do a lot of damage to the compressor blades. Likewise, applying heat to a wing leading edge that already has a significant accumulation of ice can cause it to melt and refreeze further aft on the unprotected portion of the wing. Because it's hard to see the wing tips on most swept-wing airplanes, you have to rely on other cues to know when you've started accumulating ice. The windshield wiper often accumulates ice before any other part and makes for a good visual first warning. Transport category aircraft are also required to be equipped with ice detectors. These ingenious devices are metal rods that protrude from the nose of the aircraft which are vibrated at a particular frequency. Any ice accumulation will change the frequency of the vibrations, triggering an icing warning in the cockpit. Occasionally the detectors are heated to knock off existing ice so they can determine whether icing conditions still exist.

On most aircraft, the various icing systems must be turned on manually. Most operators direct their pilots to do so when entering potential icing conditions (clouds or visible moisture near or below freezing temps), or at the latest when the ice detector gives an icing indication. This is where the JungleBus departs significantly from previous designs; it makes normal operation of all anti-ice systems fully automatic. The ice protection panel is a collection of dusty switches that rarely get touched; so long as the mode selector remains in AUTO, the system will automatically turn on wing and engine anti-ice whenever the ice detectors sense icing conditions from shortly after takeoff until landing. Meanwhile the windshields are protected any time there are at least two sources of AC electrical power, and the probes are heated automatically whenever an engine is running or manually via a button on the FO's main panel.

There is some manual control of the system for abnormal operations and for takeoff. Although the mode selector is normally left on AUTO, turning it to ON manually activates engine and wing anti-ice (on engine start and liftoff, respectively). Individual selector buttons for the wings, each engine, and each windshield allow each component to be manually deactivated. These are seldom used except when components fail. Manual control of anti-ice systems for takeoff via the Flight Management System is much more common. Using bleed air robs the engine of compressed air for combustion and therefore use of wing and engine anti-ice results in decreased power output. For this reason, the JungleBus inhibits automatic operation of both wing and engine anti-ice until reaching 1700 feet AGL after takeoff. If anti-ice protection is desired for takeoff, the pilots must set it on the Takeoff Dataset Page of the Multi-Function Control Display Unit (MCDU, otherwise known as the FMS head). The standard mode, which inhibits anti-ice until 1700 feet AGL, is OFF. Turning it to ENG mode turns on engine anti-ice as soon as the engine is started. The ALL mode enables engine anti-ice on engine start and wing anti-ice when wheel speed exceeds 40 knots on takeoff. In both cases, the anti-ice systems revert to automatic operation once the plane reaches 1700 feet AGL.

At NewCo we turn the dataset to ENG mode if there is any precipitation falling or any surface contamination with a static air temperature of less than 10 degrees C, and we use ALL mode if there is visible moisture below 1700' AGL with a SAT of less than 5 degrees C. Of course we won't get any wing protection until achieving 40 knots on takeoff, so any prior contamination must be removed with de-ice fluid (Type I). This provides limited protection for active icing conditions on the ground (ie falling snow) so in many cases we follow application of de-ice fluid with a second coat of anti-ice fluid (Type IV). This is designed to absorb the falling precipitation and then shear off the wing during the takeoff roll, at which point the aircraft's own anti-ice systems will be protecting it.

Like most aircraft systems on the JungleBus, there is a dedicated synoptic page for the anti-ice system on the Multi Function Display (MFD). It displays the status of the bleed and anti-ice valves, pneumatic system pressures, and bleed air & wing duct temperatures. A color-coded schematic of the system makes it easier to quickly understand any abnormal conditions.

One annoyance of the JungleBus' ice protection system is that the ice detectors interact with the Stall Protection System with no provision for pilot intervention. Once ice is detected, the SPS will assume it remains on the airframe for the remainder of the flight and increase stick shaker & pusher speeds accordingly. This forces the pilots to use faster approach and landing speeds, which would be the correct thing to do anyways if the aircraft was actually loaded up with ice. It is, however, pretty ridiculous to be forced into using ice speeds to land in Dallas on a 90 degree day just because you picked up a trace of ice on climbout from Minneapolis several hours ago.

All in all, though, the JungleBus' ice protection system works pretty well. Last winter I had a few occasions where the ice built up pretty good on the windshield wiper before landing, and post-flight inspection revealed significant accumulation on unprotected surfaces but the heated portions of the wings and cowls remained absolutely clean. The aircraft handled quite well despite having a decent amount of ice on the nose, wing roots, and (gulp) tail. Really, the JungleBus' best ice-fighting technology is its thrust-to-weight ratio. Although transport category aircraft do have better equipment for dealing with ice than light aircraft, prudent airline pilots use it the same way as prudent private pilots: to keep ice accretion to minimum while exiting icing conditions ASAP. Being able to climb rapidly through icing layers means that the JungleBus' good ice protection comes in handy primarily when you find yourself stuck at a bad icing altitude for several minutes on approach.

We're supposed to get snow tonight so I think it won't be long before I put this knowledge to practical use. In my next post, I'll delve more deeply into de-icing ground procedures.


Thursday, October 30, 2008

OK, One More Ride

Hopefully you can indulge me one last motorcycle post before I return to our regularly scheduled aviation content; it turned out my riding season wasn't quite over yet.

The bike that we rode in the pictures from the last post is a 1988 BMW K75S. BMW K-Bikes, which include the K75, K100, K1, K1100, and K1200, are a series of sport-touring bikes featuring liquid-cooled inline engines that are famous for their durability. The first K-bike, the K100, was quite a departure from BMW's previous motorcycle designs when it was introduced in 1985; they'd built the brand on twin-cylinder horizontally opposed air-cooled motors not unlike small aircraft engines. The K100 was an instant hit. The next year, BMW introduced the K75, essentially identical to the K100 but with three cylinders instead of four, giving it a total displacement of 750 cc's instead of 1000 cc's. It was envisioned as a stripped-down, more affordable version of the K100. As the years went on, though, the K75 became even more popular than the K100. Its lighter weight gave it handling more on the sport side of "sport-touring," and the 3-cylinder engine was the smoothest motorcycle engine BMW (or arguably, any other manufacturer) ever produced. Today used K75s fetch a better price than K100s, but both are sought after and hold their value much better than other bikes their age.

The K75 pictured in the last post belongs to my brother Josiah. It used to belong to my dad, and it's the motorcycle I learned to ride on. I still ride it occasionally and always marvel at what a great bike it is, especially for one 20 years old. For the last year, I've kept my eye out for a good-priced K75 or K100. I like my little FZ600 but it's really designed for one thing, which it does very well: zipping around corners with one person on it. Touring is excruciating, as is riding 2-up (particularly for the passenger). When Dawn rode on the back of Josiah's K75 on our ride to Wisconsin, it was a revelation for her: this motorcycling thing is a lot more fun on a bike designed for 2-up touring! She indicated her openness to me buying a BMW. I kept an eye out, knowing the best deals are typically found in fall and winter.

Monday before last, I found a 1985 K100RS for sale in Aspen, Colorado, for a price significantly below bluebook. It had 64,000 miles on it, which would be a lot on most bikes but is actually average for BMWs this age; they've been known to last beyond 200,000 miles. I traded some emails with the owner, who indicated the bike had cosmetic issues but was mechanically sound. Unfortunately, I had to fly and wasn't able to get up to Aspen until [this past] Saturday afternoon; the owner already had someone coming to look at it Saturday morning. I told him to call me if they cancelled, and stashed a bag of riding & camping gear at the Minneapolis airport just in case.

Friday night, the owner called to tell me the other interested party had cancelled and I could come take a look if I wanted to. I jumpseated to Aspen via Denver on Saturday morning after I got off work. The bike was in better shape than I was expecting. After inspecting, riding, and buying the bike, I packed the included saddlebags and headed out around 3pm. I wanted to get to Denver that night, and was planning on taking US82 to I-70. The owner assured me that was the long way to Denver, and taking Independence Pass to Leadville would be much quicker. "The pass is clear," he told me.

A half hour later I yelped as the bike slipped and skidded out of control on a section of ice covered by wet snow. I eased off the throttle, concentrated on balancing while looking out at the road ahead, and somehow kept it upright until I rolled out of the shadows and back onto bare pavement. How horrible it would've been to drop the bike within my first hour of ownership! The pass wasn't remotely clear. There was snow and ice on every section not exposed to the sun. I was able to ride around most of it, and the rest I negotiated very carefully. Finally I reached the top, and with it a final sketchy section of ice with blowing snow on top. The road down the other side was perfectly clear; the ride to Leadville and then Denver was superb.

Sunday morning I set out from Denver at 7am. Denver itself was still surprisingly warm, but the eastern plains were chilled by a gusty north wind that grew more intense as I labored eastward. By the time I was 100 miles into Nebraska I was riding permanently heeled over, the gusts pushing me around in my lane. At 80 mph it was doable; every time I was forced to slow for traffic or construction the bike became increasingly hard to keep on the road. In Overton, 20 miles short of Kearney, I stopped for gas and had a tough time just riding an eighth of a mile at 30 mph. While at the service station a huge gust came up and nearly tipped my bike off the sidestand; more impressively, it visibly moved the gas pump. I checked the weather: Kearney was reporting gusts to 45, and Omaha was expecting gusts to 60. "This is nuts," I thought. I decided to quit for the day. Fortunately, there was a motel next to the gas station. Unfortunately, it was a near-exact replica of the Bates Motel, but in poorer repair. I survived by staying out of the shower; it was moldy, anyways.

Quitting early on Sunday left me nearly 600 miles from Minneapolis. I had hoped to be home Monday night; now I had doubts. The temperatures dipped well below freezing overnight; I contemplated a late departure and just trying for Des Moines before nightfall. Finally I decided to tough it out and simply stop often to warm up. It was 25 degrees F when I started riding at 7am on Monday morning. It was chilly enough just walking outside my motel room; it was incredibly cold at 70 mph. For the first several hours I stopped every 40 miles for coffee or to warm my hands under rest-stop hand dryers. By 10am, the temperature was above freezing and I was able to make good time. Between Omaha and Des Moines, it got up to a positively comfortable 41 degrees F. Once headed north on I-35, the temperatures started to drop again but by then I was on the home stretch and the cold didn't bother me as much. It was 35 degrees F when I rolled into Minneapolis at 5:30pm, having riden 580 miles for the day.

My new bike performed flawlessly. I was really impressed by its handling and comfort, not to mention relieved that it proved dependable. Dawn and I rode it a few nights ago (in balmy 40 degree weather!) and she likes it a lot too. Sadly I'll be putting the bike into storage for the winter this weekend; I think our Indian Summer is over and the cold is here to stay. The snow will be here before long; Minneapolis actually got flurries on Sunday. I've been reading our De-Ice Manual in cruise to get me back up to speed on those procedures; it's not going to be long before the deice pads of Minneapolis run red and green with glycol. Speaking of which, this seems likely a timely opportunity to discuss the JungleBus' de-ice/anti-ice systems. That'll be my next post. In the meantime I have one last short ride on Saturday before I put away the bikes and resign myself to the coming fury of the Minnesota winter.

Monday, October 20, 2008

October Ride

My one-week experience as a commuter made me glad that I don't do it full time, but it was entirely worth it to get the Saturday before last off. It was a beautiful autumn day in Minnesota, with clear skies and temperatures in the 70s. My dad and Dawn and I took advantage of the weather to get in one last motorcycle ride of the season, to southwest Wisconsin and back. My dad rode his BMW R1100, and I borrowed my brother's BMW K75 so Dawn could come along (my Yamaha has a tiny rear seat that limits two-up riding to short jaunts in town).

The 300 mile ride took us along the bluffs of the Mississippi River to Lake Pepin, through Wisconsin's hill country on twisting, roller-coaster back country roads, and back to the Twin Cities on a highway with fast sweeping curves through river valleys and hills. The fall colors weren't quite yet at peak but were still pretty spectacular at times. This is an area I fly over fairly often, and it's chock full of tantalizing-looking little twisty roads. It's too bad that our riding season is so short; I'm looking forward to next season so we can explore the area more.

Dawn took quite a few pictures from the back of my bike. Here are some of my favorites.













Tuesday, October 14, 2008

Commute to Work

If you arrive two hours before your flight as the airlines recommend - and you do, right? - it gives you plenty of time for people watching. Among the various subgroups of travelers you expect to find, there are almost always a few pilots standing around in uniform, often with pained expressions. Some of these may be "deadheading" - that is, being repositioned by their airline and getting paid for their time - but the majority are generally commuting to or from work on their own time. Commuting is the least favorite aspect of many pilots' lives, but is sometimes better than the alternative. The airline pilot's ability to live hundreds or even thousands of miles from his place of work is alternatively praised and bemoaned by those enduring the grind.

Very few airlines require their pilots to live at or near their domicile. They only require that pilots report for duty at the correct time and place; how to get there is up to the pilot. If you live in base, or within a hundred miles or so, you get to work just like everyone else: jump in your car (or as in my case, hop on the bus and train). Those who live outside driving distance are in the unique position of being able to fly to work, using their non-rev pass and jumpseating benefits. To an outsider, it all seems terribly convenient. After all, you could live anywhere in the world! What few realize is this travel is done on the pilot's own time, and is on a space-available basis. This adds considerably to the time and stress load of commuting.

The reasons that pilots commute to work are varied. For starters, airlines don't always put their hubs in places that pilots want to live. By its very nature, a hub will either be in a geographically central location ("flyover land") or one of the largest (and most expensive) coastal cities. Some people are happy to live in places like Detroit, Memphis, Newark, Houston, Dallas, Minneapolis, or Chicago - but some are not. Some people can afford to live in cities like San Francisco, Los Angeles, or New York - but many cannot, especially on junior FO pay. For these pilots, commuting is the key to living better or more affordably.

Other pilots have put down deep roots in a community and do not care to move simply because they got a new job. Many have young families they do not wish to uproot. Some have spouses that refuse to move. For these pilots, commuting is the key to a happier family life.

Finally, some pilots didn't intend to commute but that's the hand that life dealt them. Some airlines - regionals especially - are notorious for opening and closing bases at the drop of a hat. Some airlines, faced with financial trouble, have abruptly withdrawn from what was once a primary market and crew base (ie USAirways and Pittsburg). When faced with an unstable situation, it would be foolishness to move around regularly at the whims of the airline. For these pilots, commuting provides a measure of stability to their lives.

These are all situation where the ability to commute can improve a pilot's quality of life, but of course commuting exacts its own toll on quality of life. The decision to commute involves carefully calculating precisely which option sucks less. To begin with, commuting is enormously draining on your free time. The flight times involved are the very least of it. There's the time spent in airports waiting for flights and in hotel rooms or crashpads waiting for your trip to begin. Most airlines have a "commuter clause" in their contract that require you have two flights with seats available that arrive before your show time; depending on the flight loads and frequencies along your route, this could result in commuting into domicile many hours or even days before your trip. You generally can't count on jumpseats unless you're flying on your own airline and are fairly senior, or have an odd route with little competition from other commuters. Weather can further complicate things. This all assumes a single-leg commute; if you make it a multi-leg commute with a connection through a busy hub, the time wasted and stress caused increase exponentially.

Commuting can further decrease quality of life by forcing a pilot to bid solely on the basis of "commutability." To a commuter, the only criteria to judge a trip by is report time and release time. The commuter looks for a late report time, so they can commute to their domicile in the morning, and an early release time, so they can hopefully make it home that night. All the other things that pilots typically bid for - pay, productivity, time off, weekends off, lack of circadian swaps - are utterly secondary. Only when you are ultra-senior in your seat can you hope for a line that's both commutable and good.

Commuting can be expensive. When you're junior and unable to hold a "commutable" schedule, you end up spending a lot of nights at your domicile before and after trips. You either pay $40-80 per night for a crew rate at a nearby motel, or get a crashpad shared with other commuters for $150-300 a month. If you're an unfortunate soul condemned to commuting to reserve - the ninth circle of Commuter Hell - you'll end up spending a truly depressing amount of time in your crashpad. Avoiding this situation has caused many commuters to defer upgrades or transitions to better-paying equipment until they could hold a line, another major cost.

Given all this, it makes sense to avoid commuting if at all possible. One solution I've seen from many pilots is refreshingly old-fashioned: commute by car, like everyone else! You can live quite a ways from your base and make it work; after all, we're usually talking one roundtrip per week. For reserve, you'll need to be a two-hour drive away from the airport. Even the most wretched places to live generally have nice places 100 miles away. You could work in Newark and live in upstate New York; you could work in Detroit and live in Ann Arbor. Many pilots live near a small airport that has regional flights to their domicile; if the flight loads merit, they fly; otherwise they drive. It's a nice hybrid method that makes commuting more palatable.

I've managed to avoid commuting for the most part. My first few months at Horizon, I was living in LA while on reserve up in Portland. As commutes go this was an easy one: multiple direct flights to Portland from Burbank (which I lived next to) and LAX, as well as tons of easy connecting flights on Southwest. All the same, sitting on reserve in a crashpad away from home was tough, and Dawn didn't see me nearly as much as she'd like to. That plus our lack of roots in LA, the lower cost of living in Portland, and having friends in the Pacific Northwest made the decision to move an easy one. When I was hired at NewCo, we moved specifically for the MSP base. It was tough to leave the PNW but getting closer to family was important to us. While the townhouse was for sale, though, I traveled between Minneapolis and Portland enough to qualify as a semi-commuter; seeing the heavy loads on that route made me glad I wasn't doing it full time or long term.

I wrote most of this post Monday morning at gate F4 in Minneapolis, attempting to get to Detroit to start a trip. Minneapolis-Detroit is one of the toughest commutes in RedCo's system. Although there are frequent flights between the two hubs, they're often packed and there are literally hundreds of other commuters vying for a limited number of jumpseats. Many of NewCo's pilots have done this commute for a few months until their seniority can hold a Minneapolis base; fortunately I held MSP from the very start as both a FO and Captain. I'm commuting this week on a one-time basis as the result of a trip trade I did last week. I was originally scheduled to work this weekend, but then the weather reports started indicating that I was going to miss the best weekend of the autumn. My dad mentioned he wouldn't mind doing one last motorcycle ride of the season, so I found a Detroit trip starting on Monday to trade for my Minneapolis trip starting Saturday. I had a spectacular ride through southwest Wisconsin with my dad and Dawn on Saturday, but Monday I paid the piper. The flights were all oversold, prompting me to list for a 8:30am flight to make my 3:00pm report time. That flight went out with full passengers and all cockpit jumpseats and extra flight attendant jumpseats full, and I still had two standbys stranded above me on the list. The 10:05 flight was shaping up the same way, and my last-chance 11:25 flight was even more oversold. I was picking up my phone, getting ready to make my "commuter policy" call to crew scheduling and beg for a positive space seat on the 11:25, when the gate agent called my name. A seat had opened up! I grabbed my bags and trudged down the jetway to hunt down the last few remaining bits of overhead bin space. I sighed in relief that I made it on, and gave silent thanks that I don't have to do this every week.

Sunday, October 12, 2008

JungleBus Systems Post: Hydraulics

I've mentioned a few times that NewCo's initial training program was a bit, ahem, basic, and that I felt somewhat underprepared in the area of systems knowledge when I first got on the line. Since then I've gone back and studied on my own a few times, but I still feel like I could benefit from a more intimate knowledge of JungleBus systems. In my experience, the best way to learn aviation knowledge is to teach it. With that in mind, I'll be writing occasionally posts on various JungleBus systems. They won't be in any particular order; today I've randomly chosen to begin with a dissertation on the Hydraulic System. Those with JungleBus experience are welcomed to add anything I forget or correct my mistakes. I hope that by next time I have recurrent ground school, I have a nice collection of posts to come back and read rather than poring through AOM-II for hours on end until my eyes bug out.

On most light planes, the hydraulic system is limited to powering the brakes and, in some cases, retractible landing gear. Once you get to airliner weights, though, you find that the forces involved often outstrip the ability of the human muscle to overpower. In these cases, engineers rely on hydraulic systems to do the grunt work. This is the case with many systems on the JungleBus, including all basic flight controls, spoilers, landing gear, brakes, nosewheel steering, and thrust reverser deployment. Notably missing are the flaps, which are hydraulically powered on many airliners but are electric on the JungleBus.

Hydraulic systems use a fluid under pressure to do work. This is possible because fluids are mostly incompressible. For a hydraulic system to work, you need a source of pressure (a pump), a closed system that remains leak-free under significant pressure, and enough fluid to pressurize the system to design pressure. Hydraulic systems have a number of failure nodes: pump failure, loss of power to the pump, loss of fluid, and physical damage to the plumbing so it's unable to hold pressure. For these reasons, aircraft engineers build multiple independent hydraulic systems into the airliners they design, usually with multiple sources of power for each system.

The JungleBus has three fully independent hydraulic systems. Systems 1 and 2 are normally pressurized by engine-driven pumps (EDPs), which are connected to their respective engine's accessory gearbox. They each have an electric hydraulic pump installed as backup in case of engine failure. System 3 is powered solely by an electric pump, with a second electric pump as backup. All three systems use Skydrol brand hydraulic fluid and are normally pressurized to 3000 psi. Most aircraft systems that use hydraulic power are powered by at least two of these systems, so the loss of one hydraulic system will not affect most other aircraft systems. Flight critical systems are powered by three systems, so in the highly unlikely event of two fully independent systems failing simultaneously, the crew will still have at least pitch and roll control.

Hydraulic System 1 is normally pressurized by a mechanical pump driven by the left engine's accessory gearbox. The pump takes fluid from the reservoir, which an accumulator keeps at slight positive pressure, and pressurizes the fluid before sending it through a filter and the plumbing to the various System 1 users. The return line routes the fluid through another filter and then either through a heat exchanger or directly to the reservoir, depending on fluid temperature. If the fluid reaches 100º C, a HYD 1 HI TEMP caution message is displayed on the EICAS. At 125º C, the Hydraulic Shutoff Valve (HSOV) automatically closes to isolate EDP 1 from System 1. The HSOV will also close automatically in case of engine failure to decrease drag on the engine and make a windmilling relight easier. The HSOV can also be closed manually via a guarded push button on the hydraulic panel.

In case of engine failure or EDP failure, System 1 can be pressured by Electric Hydraulic Pump 1, powered by alternating current from AC BUS 2. It is controlled by a 3-position switch on the hydraulic panel. With the switch OFF, the pump stays off; selecting ON causes the pump to run continuously regardless of conditions. The normal position is AUTO, which causes the pump to activate automatically in case of engine 1 failure or EDP failure. In AUTO mode, the electric hydraulic pump will also run concurrently with the EDP when the flaps are in any position greater than zero in flight, or on the ground when flaps are greater than zero and thrust levers are in takeoff/goaround position (TOGA) or groundspeed exceeds 50 kts. The idea is to have both engine-driven and electric pumps running during takeoff and landing.

Hydraulic System 1 has the following users:
  • Elevator (Left-Hand outboard actuator only)
  • Rudder (upper actuator)
  • Spoilers (LH & RH, panels 2, 3, and 4)
  • Thrust Reverser (Engine 1)
  • Brakes (outboard only)
  • Emergency/Parking Brake
Hydraulic System 2 is normally powered by EDP 2, which is driven by the right engine. Backup is provided by Electric Hydraulic Pump 2, powered by AC BUS 1. It has the following users:
  • Elevators (LH & RH inboard actuators)
  • Ailerons (LH & RH inboard actuators)
  • Spoilers (LH & RH, panels 1 and 5)
  • Thrust Reverser (Engine 2)
  • Brakes (inboard only)
  • Nosewheel Steering
  • Landing Gear
  • Emergency/Parking Brake
The function of Hydraulic System 2, including the AUTO mode logic of Electric Hyd Pump 2, is virtually identical to System 1. One difference is that the Electric Hyd Pump 2 will also activate on the ground if the left engine is running and the parking brake is released or flaps extended beyond zero. The reason for this is single-engine taxi; we normally taxi on the left engine only, but obviously need nosewheel steering and both brake systems powered. Theoretically, they will be if Electric Hyd Pump 2 is left on AUTO and you taxi on the left engine. If you were to taxi on the right engine only, however, all the System 1 users (including outboard brakes) would be unpowered. For this reason, and in case the AUTO mode system logic should fail, NewCo single-engine taxies with both Electric Hyd Pumps turned ON regardless of which engine is running.

System 1 and 2 don't have any common points where fluid can migrate, but there is a mechanical connection via the Power Transfer Unit, or PTU. This is basically an extra pump used to pressurize part of System 2; it is motored by hydraulic pressure in System 1. The only purpose of the PTU is to facilitate extension and retraction of the landing gear. To operate, it needs System 1 to be pressurized, and there must be fluid in System 2. It is controlled by a 3-position "OFF-AUTO-ON" knob, which is almost always left in AUTO. In this position, system logic will turn on the PTU if the right engine or EDP 2 fails when the flaps are greater than zero. It's essentially there to quickly raise the landing gear if the right engine fails just after takeoff and Electric Hyd Pump 2 fails or isn't supplying enough pressure to the landing gear. Note that the PTU is useless to raise or extend the landing gear in case of System 2 fluid loss; in this case the crew must extend the gear using a freefall procedure. There is no way to retract the gear with no fluid in System 2.

Hydraulic System 3 is essentially an emergency backup system to ensure critical flight controls remain powered in case of a catastrophic simultaneous failure of Systems 1 and 2. It is pressurized by one electric pump (Electric Hyd Pump 3A) with an additional electric pump (Electric Hyd Pump 3B) for backup. Pump 3A is controlled by an OFF-ON knob on the hydraulic panel, with no automation involved. It gets its power from the AC ESS bus, which remains powered in an electrical emergency. Pump 3B has a 3-position OFF-AUTO-ON knob; in AUTO position it will activate whenever Pump 3A fails. Pump 3B is powered by AC BUS 2. System 3 powers the following hydraulic users:
  • Elevator (RH outboard actuators)
  • Rudder (lower actuator)
  • Ailerons (LH & RH outboard actuators)
System 3 has two dedicated valves to avoid overload in case of electrical emergency. The Unloader Valve temporarily unloads Pump 3A during its start-up to avoid spikes in power consumption. The Flow Limiter Valve limits the amount of flow coming from Pump 3A as electrical loads approach critical levels during an electrical emergency. Flight controls may be sluggish but control will be retained.

The Multi-Function Display (MFD) in the flight deck can bring up a Hydraulic Synoptic Page. It displays hydraulic fluid quantity, temperature, and pressure for all three systems and shows the status of all engine-driven and electric hydraulic pumps, plus the PTU. Finally, the Synoptic Page displays a handy list of all hydraulic users, organized by system. This, in case of hydraulic system failure, the pilots can see at a glance which aircraft systems will be affected.

You can see that the JungleBus could suffer multiple hydraulic failures and the most critical systems will be unaffected. Either engine failure should not affect any of the three systems. Complete System 1 failure would leave the pilots with three of four elevator actuators, all four aileron actuators, one of two rudder actuators, two of six roll spoilers, two of four ground spoilers, one of the two thrust reversers, and two of four brakes. Complete System 2 failure would still leave powered two of four elevator actuators, two of four aileron actuators, both rudder actuators, four of six roll spoilers, two of four ground spoilers, one thrust reverser, and two of four brakes. A combined System 2 + System 3 failure would leave the ailerons unpowered but you'd still have one rudder actuator and four of six roll spoilers for roll control. The only catastrophic combination, of course, is an uncontained engine failure or similarly violent event that results in failure of all three systems a la United 232. Of course, certification standards have improved considerably since then to ensure the physical separation and protection of hydraulic system components. That's good because, despite its smaller size, the JungleBus is just as dependent on hydraulics as the DC-10.

Monday, October 06, 2008

Good FO, Bad FO

When I wrote the original "Good Captain, Bad Captain" post in June of 2005, I'd been a FO at Horizon for just over a year. By that time I had a pretty good idea of what to look for in a Captain, and my thinking hasn't changed much since. Since then I've spent about 3 years in the right seat, which gave me enough time to think about what makes a good First Officer. I often asked Captains I flew with who they considered good FOs and why, and some of the problems they'd encountered with FOs in the past. My six months in the left seat hasn't really given me any new insight into what makes a good FO, but has given me time to appreciate what a joy it is to fly with an excellent FO. Here's the criteria I measure an FO by:

A good FO knows his job well. He knows his maneuvers, flows, calls, and checklists like second nature; he also has a strong grasp on aircraft systems. Obviously brand new FOs won't quite measure up to this standard, but the good ones get up to speed quickly. When an airline's training program falls short - as it does at NewCo - a good FO will seek the answers from check airmen, line captains, and the source materials. A good FO is inquisitive and learning something new every trip.

A good FO keeps their Captain in the loop. This is, of course, merely the reverse of a good Captain keeping their FO in the loop. A two pilot crew is really only functioning as a crew if each pilot knows what the other is doing. Sometimes the duties are so self-explanatory that little verbal explanation is required. In situations where significant technique is involved - night visual approaches in the mountains, for example - I keep up a steady diologue explaning what I'm doing to the other guy. I really appreciate it when they do the same for me. I don't have to guess what they're thinking, and can devote more time to my own duties rather than constantly monitoring theirs.

A good FO is a chameleon. Although the airlines don't talk about this for fear of encouraging passivity, adaptability is a key aspect of an FOs job. Every Captain has their own style, techniques, and quirks. A good FO will figure out what they are early in the trip and adapt accordingly. This doesn't mean he should condone non-standard flying; there's a lot of room for variance while still complying with the book. It does mean the FO makes the Captain's life easier and allowing him to focus on the big picture by complementing the Captain's style rather than clashing with it.

A good FO will speak up when his Captain crosses the line. A FO has to choose their battles wisely; nitpicking everything a Captain does will only result in an agitated Captain who's not listening to their FO. When the Captain is clearly operating outside the bounds of standardization and safety, though, it's time to speak up - clearly and forcefully. Like I mentioned in Good Captain, Bad Captain, this is really an awkward situation because the Captain is your superior and you're his subordinate. Saying "no" to something is essentially insubordination - which others may subsequently judge to be justified or not. It's one of the toughest things to do as an FO, but too many airplanes have flown on to their doom with the FO or SO fully aware that something's wrong but too intimidated to say anything.

The flip side to all this is Captainitis. That's the airline term for the malady that creates "Right Seat Captains" who attempt to do the real Captain's job for him. Every once in a while you see an FO who's just itching to upgrade and has developed a rather premature command attitude from the right seat. They'll make decisions unilaterally or hector the Captain to make decisions they like, impinge on the Captain's flows and other duties, and fly just enough outside the book to assert their command of the aircraft. Horizon had a few FOs who were notable for Captainitis; most had been in the right seat for five or six years. There's a certain point at which you know the aircraft and your job and the Captain's job so well that you're plainly qualified to upgrade, but are just waiting for your number to come up. That's frustrating, especially at regional airlines where you're basically stuck until you log enough PIC time. I can understand how that frustration can become Captainitis. Furthermore, some weak Captains encourage Captainitis by being passive and indecisive, essentially abdicating their role to an FO who's hungry for a bit of authority.

Most of my FOs so far have been good, with a few truly excellent ones. That's all the more remarkable in that nearly everyone I've flown with was new to both NewCo and the JungleBus. Most had been FOs at other airlines; many had also been Captains. That experience carries over well to other airlines and aircraft. I've only had "problems" with two FOs, who represented two extremes of the scale. One was fresh off of IOE and this was his first Part 121 airline job; he was very silent, passive, and had a lot of trouble with directional control on takeoff and landing. I only had him for two legs; I addressed the problems after his leg and gave him some pointers. He was genuinely thankful for the advice and asked me a few other questions.

The other FO had a case of Captainitis; he'd been a six-year FO at Comair and felt more than ready to upgrade, even with a mere hundred hours in the JungleBus. Most of the trip it was present only in small things like mannerisms and body language. The last leg into MSP, though, the Captanitis manifested itself in a big way. My FO was landing on 35, and floated the landing slightly. I had briefed beforehand that I would take control once the airplane was slowed to around taxi speed. While we were still around 80 knots, tower told us "Right turn when able, hold short 30L on W10, tower 126.7" I started to respond to the instructions; without warning my FO jammed on the brakes and inputted full right rudder, attempting to make the turnoff at Y. That's not a 90 degree turnoff, but it's sharper than a normal high-speed turnoff. It happened too quickly for me to intervene; the g-forces threw me into the left side of my seat. When we were parked at the gate, I asked my FO what exactly was he thinking on that landing. He gave me a quizical look, so I told him it had really alarmed me and it was pretty non-standard for the FO to make the turnoff, especially without it being briefed beforehand. He tried to brush it off: "Oh, I did that all the time at Comair!" I told him to leave the turnoffs to the Captains, and if he did something like that with me again it'd be the last time I'd let him land. He looked pretty chastened and made a quick exit. Afterward, I wondered if perhaps I should've addressed his Captainitis earlier in the trip. It would've been a quiet trip but I doubt he would've done something so audacious afterward.

That's the approach Dave at FL390 took recently. His FO was exhibiting signs of Captainitis and had done several things that, while fairly minor, indicated at least some disregard for their company's aircraft manual. On the second day of their trip, Dave gave him "the talk." There are a few unwritten rules for "the talk" which Dave demonstrated in his post. First, it's something you do in private. Being corrected by the Captain is embarassing enough for a FO without others witnessing it. In the flight deck after the parking check is complete but before the cockpit door is opened is an ideal time. Secondly, you use "the book" to point out exactly where the FO erred. Simply listing your pet peeves allows the FO to think you're just a grouch who likes things done his way; backing yourself up with company policy shows that you both know the policy and expect your FO to know and follow it. Finally, don't expect your FO to like you very much after "the talk." It'll probably be a quiet trip. That's ok - what's important is that your FO is flying safely and by the book. If the poor performance persists, then you move onto the next level by getting the union's professional standards commitee involved - or as an absolute last resort, the chief pilot's office. Fortunately, that's extremely rare. Most FOs are very good at what they do, and they make the Captain's job immeasurably more pleasant.