QRH

Being a morning person is a tremendously helpful attribute in an airline pilot, and I'm just not. Ever since adolescence, I've been a pretty serious night owl, with my peak hours typically occurring between 8pm and midnight. This worked well when I was freight-dogging, and may yet prove fortuitous to flying red-eye transcons and ocean crossings. In my present incarnation as a regional airline pilot, though, it's a bit of a handicap. We have a lot of early morning report times, especially in the ultra-efficient trips I like to bid, and I have to work to adjust my body clock accordingly. I can usually function well on six hours of sleep, assuming I started the trip without a sleep deficit, sprinkling in short naps as needed.

Even this means that those really painful 3am wake-up calls for a 4am showtime require getting to sleep by 9pm, which is quite early for me.  Last week I had one of those in Bismark, North Dakota, for a 5am flight to Minneapolis followed by an east coast turn. I was in bed by 9pm, and tossed and turned for hours on end before finally falling asleep, only to be jolted awake by my alarm clock seemingly only minutes later. A hot shower did little to clear the fog. I fought the nods during the 15-minute van ride to the airport. I chugged two cups of coffee before departure time. By the time we pushed back, I was wide awake and ready to assume my Captainly duties of groping around a dark, uncontrolled field under construction and then taking off into inky skies and darting around a large line of storms - but the fatigue was still palpable, a slow steady hiss in the back of my head that grew louder as the flight went on. By the time we were descending into Minneapolis, I was contemplating the possibility of a fatigue call to crew scheduling. They wouldn't like that this early.

 Minneapolis was appropriately quiet at 6am and approach control almost immediately cleared us to 4000 feet; I hustled down, anticipating a short approach. As we leveled, I spun the speed selector back to 200 knots, and at an appropriate speed commanded "Flaps 1." The FO reached for the flap handle and slid it back into the first detente. Normally this results in a 10-second sequence of the slats moving 15 degrees down, followed by 5 degrees of flaps, during which the airplane pitches down appreciably as the lift devices enable a lower angle of attack. Instead we were immediately rewarded with a loud "ding!," flashing yellow master caution lights, and four messages displayed on the EICAS:

SLAT FAIL

SPOILER FAULT

AOA LIMIT FAIL

SHAKER ANTICIPATED

As I reached to press the caution switchlight to cancel the alarm, the first thought that sprung into my head was "why now!?" Readers who have flown the CRJ series will laugh at this, but I've never had to do a real-life zero-flap landing in the JungleBus. I've had a few friends who've done it, and they said it wasn't too bad, but things happen pretty quickly and you touch down uncomfortably close to the JungleBus' 195 knot maximum tire speed. It didn't sound like something I really cared to do with my brain fogged in from lack of sleep and the still-early hour. This indulgence lasted for maybe two seconds before a deeply-seated imperative forced its way up through my sleep-clouded consciousness and crowded everything else out: "QRH!"

"I have the airplane and the radios. Slat Fail QRH, please." The FO already had the spiral-bound, plastic-tabbed rectangular booklet that is the JungleBus' Quick Reference Handbook out and was thumbing to the appropriate page. The Slat Fail checklist seemed like the natural place to start, as the failure happened while we were deploying slats and the other messages were all related to them. I spun the airspeed back up a few knots to give us more margin from the now-undepicted stall speed. While the FO began reading the checklist, I told Minneapolis approach that we had a malfunction we needed to diagnose and requested an extended downwind. No problem, the controller said, and asked if we were declaring an emergency. "Not at this time, but we'll keep you posted." A zero-flap landing would likely merit the precaution of rolling the trucks. 

It turned out to be unnecessary. The very first step in the QRH was to return the slat/flap handle to its last position - UP - and see if the messages cleared. They did. The second step was to extend the slats again and see if the messages returned. The slats and flaps deployed normally this time. "Proceed with normal operations." QRH complete. We thanked Minneapolis Approach for their assistance, and they turned us onto a 25 mile final for an easy visual to 30R. 

So it turned out to be a complete nonevent, one of those things that happens pretty routinely when flying an electronic airplane like the JungleBus, a quick control-alt-delete fix-it. It's doubtful any of our passengers even noticed the extended downwind. It was part of the "almost nothing worth blogging about" I mentioned in my last post. It didn't even interrupt our trip, for although I wrote up the malfunction and it almost certainly required computer replacement as it was the second occurrence in four days, we were scheduled to swap to another airplane anyways. The primary, rather agreeable result of the incident was to provide just enough of a diversion to jolt me into full useful consciousness. I didn't feel the least bit tired afterwards, and flew a pleasant turn to Hartford and back to finish the trip. 

But that's really the point of the procedural, aircraft, and training safety systems we have in place. They turn most potential events into non-events, because there's seldom any doubt about what to do. Even a fatigue-addled brain automatically knows through repetition to fly the airplane, call for the QRH, and coordinate with ATC while the non-flying pilot runs the checklist. Had the slats not healed themselves and we actually had to perform a zero-flap landing, even this would have been relatively easy because the guidance on how to do it is quite explicit, and I've done it a number of times in the simulator. Over the years, I've occasionally decried the lack of emphasis on stick and rudder skills and common sense in pilot training, but the flip coin to that is that turning airline pilots into checklist-reading automatons has itself undoubtedly done a great deal for safety.

How Not to Build a Trip

Last month I was assigned a trip that is an excellent example of how the current rest and duty rules are woefully inadequate and change is long overdue. I have reproduced the trip key below with only format changes to improve clarity. I hasten to add that although this is a NewCo pairing, every airline out there has these trips to some extent, and they are relatively rare here. Since their work rules were stripped away in bankruptcy, even the major airlines build some rotations like this one, and a few regionals build trips far worse than this one as a regular practice.

M7284 BASE REPT: 1050L

DAY FLT DEP-DEST STD STA BLK TURN

MO 5694 MSP-YYC 1135 1343 308 37

MO 5876 YYC-MSP 1420 1803 243 132

MO 5778 MSP-MDW 1935 2103 128

BLOCK 719 DUTY 1028 REST 1658

DUTY END: 2118L REPT: 1416L

TU 5749 MDW-MSP 1501 1635 134 230

TU 5755 MSP-BHM 1905 2134 229

BLOCK 403 DUTY 733 REST 926

DUTY END: 2149L REPT: 0715L

WE 5887 BHM-MSP 0800 1038 238 237

WE 5784 MSP-OMA 1315 1425 110 30

WE 5729 OMA-MSP 1455 1606 111 254

WE 5657 MSP-MSN 1900 2004 104

BLOCK 603 DUTY 1304 REST 926

DUTY END: 2019L REPT: 0545L

TH 5649 MSN-MSP 0630 0740 110 240

TH 5696 MSP-MKE 1020 1132 112 35

TH 5696 MKE-MSP 1207 1332 125 208

TH 5850 MSP-ORD 1540 1657 117 103

TH 5850 ORD-MSP 1800 1925 125 E75

BLOCK 629 DUTY 1355

DUTY END: 1940L

TOTALS BLOCK 2354 LDGS: 14 CREDIT 2354 T.A.F.B. 8050

Since most readers don’t likely see many trip keys, I’ll break this one down. The first two days are pretty easy, with three and two legs, respectively. The first layover, at Chicago-Midway (MDW), is a long one of nearly 17 hours. The following day, two legs to Birmingham (BHM), has less than 8 hours of duty time. That night, though, has only 9 hours and 26 minutes of rest time. In this situation, 9 hours is the legal minimum. It might even be adequate if followed by a moderately easy day and a long overnight. The trip, however, goes rapidly downhill from there.

On day three, there are four legs, three of them short, which combine for 6 hours of block time. Duty time, however, is 13 hours and 3 minutes thanks to two lengthy sits in Minneapolis. The day ends in Madison, WI with another barely-legal overnight. After 9:26 minutes of rest time, day four begins at 5:45am - completing the circadian swap from a trip that began on a P.M. schedule. There is a nearly three-hour sit in Minneapolis after the first leg, followed by a Milwaukee round trip and another two-hour sit. The day concludes with a Chicago O’Hare turn, with a little over an hour spent in Chicago. Block time and pay is 6 hours 29 minutes, but duty time is nearly 14 hours.

In all fairness, I bid for this trip, although not intentionally. I set my computer bidding preferences to look for efficient trips, those that have the most pay per day. This one is fairly efficient from that standpoint, with nearly 24 hours of pay in 4 days. It is inefficient from the standpoint that much of the down time is spent at airports, where there is no possibility of sleep, instead of at layovers. This is not, however, a criteria that our bidding system can sort. I was surprised when this trip appeared on my line, but on closer inspection saw that it did indeed meet all my criteria. There was nothing to do but fly the trip and get as much rest as I could in the little time allotted.

The first two days went fine. Everything was on time, the weather was good, and we even got into Birmingham a half hour early on day two, lengthening the layover to a full ten hours (although a late hotel van meant that less than nine was actually spent at the hotel). Day three was pretty long and tiring. We had plane swaps on each of our long sits, and the Omaha turn was delayed for a late inbound and for deicing. When we got to Madison, I was ready to hit the hay. Instead, we got to wait a half hour for the hotel van, and then spent twenty-five minutes riding to the complete opposite end of Madison. It was 9pm by the time we got to the hotel. I hurriedly ate a late dinner and went to bed. I tossed and turned for a long time before finally falling asleep, and then woke repeatedly throughout the night for a noisy fan that wouldn’t turn off.

The 4:30am wakeup call came bone-achingly early. The long ride to the airport, coupled with a shuttle that leaves on the top and bottom of the hour, meant we had to take a 5am van for a 5:45am show time. A hot shower and early cup of coffee didn’t do much to break my stupor. The short flight to Minneapolis seemed to take forever. A nap on our long break would have done wonders, but there was nowhere to sleep at MSP. The crew room is constantly busy, bright, and loud, and the designated quiet room off the side has been declared off limits to all but airport reserves. Instead I had another large cup of coffee. It worked, for a while: I was practically jumping off the seat on our way to Milwaukee.

Then I came crashing back down on the return leg, and I found myself missing radio calls and messing up simple things. On the descent, I fought an aggressive case of the nods by using the oxygen mask on 100% oxygen. We still had a two hour sit in Minneapolis – often more tiring than actually flying – followed by a four hour Chicago turn. There was no way I was going to make it. I called in fatigued shortly after landing.

The crew scheduler sounded incredulous. “Fatigued? You’ve only done three legs!” I told him to take a look at the trip, but that didn’t move him. “You’re going to have to talk to a chief pilot about this!” he exclaimed, obviously annoyed that he would need to find someone to cover the O’Hare turn. I told him I would go downstairs and talk to the base chief pilot that very minute, and hung up.

I’m happy to report that the base chief pilot was very supportive. I showed him the trip key and explained how the cumulative shortage of sleep was affecting me, and he agreed that the trip wasn’t exactly conducive to obtaining sufficient rest and thanked me for doing the safe thing. I know that fatigue calls are handled in a much less positive manner at certain other airlines, and that previous flight ops managers at NewCo reportedly took a different tack. In this case, the only negative effect I suffered was the loss of around $180 in pay.

It’s worth noting that none of my other crewmembers called in fatigued. They all said they were tired, but felt they could battle through the day safely. Fair enough; fatigue affects different people differently, and they may have been more successful in getting sleep the previous night. It’s impossible to say how the loss in pay may have affected my poorly paid flight attendants’ and First Officer’s decisions.

In any case, it’s hard to see how our crew planner could put together that pairing without it occurring to him that chances were excellent that at least one crewmember might become severely fatigued. At most airlines, not just NewCo, there is a prevalent attitude that “If it’s legal, it must be safe.” From a statistical perspective I suppose they’re right. We just finished the third year out of the last four without a major airline fatality in the United States. If pilots are flying around tired, they’re doing a remarkable job of not killing people in the process. The ATA and RAA’s obstructionism of the new rest and duty regulations shows they’re utterly willing to accept an occasional Colgan 3407 here, an American 1420 there, right up to the point that the body count causes passengers to start booking away.

Fair enough. They have tickets to sell and profits to make, and these days corporate responsibility in this country tends to start and end with the stockholders. It’s not like the majority of passengers disagree; I suspect the vast majority would willingly sacrifice 100 strangers every few years to keep ticket prices low. As a pilot, though, safety is my only concern. Sure, I put in the effort to run my ship in a timely and efficient manner, but I don’t lose any sleep if I block out a few minutes late or burn an extra hundred pounds of gas. My real job, indeed the very core of my professional identity, is to minimize the risk to my passengers in every way practical.

This is also, theoretically, the FAA’s only job. They do not have a responsibility to ensure the airlines turn a profit. Congress removed their responsibility to promote aviation economically a few years back for the very reason that I’m writing this post. Now, the FAA’s only job is to manage risk, and fatigue is clearly the chief risk factor they have allowed to go unaddressed for too long. It’s high time to put the new rest and duty regulations into effect and end the silliness of asking pilots to fly exhausting trips that tempt them to fly fatigued.

Long Day

Louisville, 1:06 AM. I am lying back in a chair in my cheerfully nondescript hotel room, feeling the adrenaline seep out of my limbs. I've been running at one hundred and ten percent for hours now, and it's almost impossible to go straight from that to sleepfulness, so I sit here trying to decompress. A drink helps. The crew and I were planning on having a nightcap - our show time isn't until late tomorrow - but the hotel bar was closed by the time we finally arrived, and there's nothing else open nearby. Instead, I pour myself a short glass of scotch from my own flask. The ice crackles and tinkles softly, accentuating the silence of my room. I take a sip and think back on the long day.

I woke up at 6:30am and set about the familiar routine of preparing for a morning show: showering, grooming, ironing, dressing, making my lunches, packing my overnight bag - always in the same order, lest I forget something. Alas, I did forget my cell phone charger, so I will be raiding the hotel's lost and found tomorrow. Dawn gave me a ride to the airport before her 9am doctor's appointment, so I arrived quite early for my 9:20 show. This was my first time back to work since NewCo switched to a completely new, electronic system for weight & balance and takeoff & landing performance calculations to replace our tried-and-true manual method using paper, long addition, and a whiz wheel. I intended to use the extra time to get to the airplane early and see if the system really worked the way our training materials said it would, but I ran into some old friends in the crew room, talked for longer than I realized, and ended up having to hustle to make it to the plane by our scheduled show time. I chatted with the gate agent a little, checked the radar and upper air charts on his computer, collected our paperwork, and walked down the jet bridge to meet my crew. The only one I'd flown with before was Linda, a retired schoolteacher whose former profession obviously endowed her with the patience that makes her an exceptional flight attendant. Our other flight attendant, Trisha, is brand new, having just come off of OE. Our FO, Kevin, is a middle-aged family man who spent ten years at Air Wisconsin and finally made the jump to United two years ago, only to be furloughed a day before completing his initial training. Rounding out our quintet for this leg was Steve, a friendly United captain I last gave a ride to O'Hare only a few weeks ago. The flight was full, and he would again be sitting in the flight deck jumpseat.

Our day got off to a late start thanks to cleaners who neglected to show up for the first hour of a 90-minute turn, subsequent late boarding, passenger-counting difficulties, and some grappling with the new W&B system on my and Kevin's part. We pushed ten minutes late, then got a last-minute taxi change to the far runway. Within a few minutes, though, we were whistling down 30R, rearing into the air at 141 knots, and turning southeastward towards Chicago shortly thereafter.

While Chicago-Midway was one of our very first destinations, my company only began flying to O'Hare a few months ago. It has quickly become one of the mainstays of our trips. I originally approached O'Hare with some trepidation, given its reputation for heavy traffic, a complex layout, and no-nonsense controllers. I quickly discovered that O'Hare, like all big, busy airports, has a well-developed, predictable flow to its operations, and is actually pretty easy to fly out of once you learn the logic. The one thing I haven't figured out yet is how they assign departure runways; there are occasionally up to five in use, and which one you are assigned seems to have little relationship to your position on the airport or your direction of flight. Even this lack of predictability is handled easily enough: you brief all the possible runways and taxi routes, get takeoff performance for every possibility, and taxi out on both engines. That leaves less to do when you are inevitably switched to an inexplicable runway choice. Nothing seems to please O'Hare controllers more than a crew that's ready to roll at a moment's notice.

The one thing we did today that I hadn't yet experienced was landing on the "new" runway, 27R. It is a good two miles north of the main complex, making for a long, circuitous taxi to the terminals. Fortunately, it's only used during peak periods, which today at 11 AM definitely was. The taxi instructions to our gate were a little bewildering: "Charlie One, Charlie, Uniform, Echo, Zulu, hold short of Tango, monitor ground one three two point seven" then "Tango, Whiskey, hold short 4 Left", then "Cross 4 Left, Echo, Bravo, Alpha Ten to the gate". Go ahead, try to trace it on the airport diagram. Now try to figure it out while taxiing a plane and stealing only occasional glances at the chart (the cardinal rule of ORD is "Don't Ever Stop"). Having a sharp, experienced FO who was once based at O'Hare helps a lot!

Our ops people turned the full plane surprisingly quickly, Kevin and I again flubbed the weight and balance calculation on our first try, and we again pushed around ten minutes late. For once, I correctly guessed our taxi route and departure runway: A, A7, T, 32R at T10. We nibbled at a United Airbus' wake turbulence on takeoff - they weren't even off the runway when tower cleared us to roll - but quickly got upwind of it as we turned snappily to our assigned 360 heading. Soon we were arcing out over Lake Michigan, weaving around afternoon buildups before picking up the Muskegon transition to Detroit's Polar Three Arrival.

I really enjoy flying into Detroit. It's a fairly busy airport, but a well-laid out one, and operations generally flow freely and predictably. The McNamara Terminal still gleams and echoes with the songs of birds who have flitted inside, and the sight of a 747 behind the dancing water fountain has already become an iconic image of aviation. The only downside, for me, is that the airport's taxiways look and feel more bombed out than Beirut circa 1982. Fortunately, the airport has undergone a steady program of resurfacing over the last few years. This time, taxiway Foxtrot just north of NewCo's northeast gates was ripped up. To provide an alternate route to our gates, A75 through A77 have been taken out of service so aircraft can use the ramp to transition from Uniform to Uniform Eight. The system seems to be working pretty well.

We had a three hour "productivity break" in Detroit, even after our late arrival. NewCo has a crew base there, but Minneapolis crews can't normally access the crew room as it is located in an area accessible only to holders of Detroit SIDA badges. Fortunately for us, Kevin had been based in Detroit for a month and had exactly such a badge so he could escort us to the crew room. I spent my break trying to figure out a way to get my little brother Steve and his girlfriend Torrie home from Hawaii. They'd flown to Honolulu on my buddy passes over Memorial Day weekend, lured by the promise of wide-open flights coming and going, only to find themselves on the bottom of an 80+ person standby list for every flight they tried to take out. After talking to a family of five that had been trying to get out of Honolulu on buddy passes for nearly two weeks, Steve decided to bite the bullet and buy positive-space tickets back to Minneapolis. The price was surprisingly cheap for a same-day one-way ticket: $585. Once again, Honolulu confirms that its reputation as a non-rev black hole is well-deserved.

Our plane arrived only 25 minutes before our departure to Charlotte, so our third consecutive leg blocked out about ten minutes late. Every seat was once again full, and we were also carrying quite a bit of extra fuel in anticipation of possible thunderstorms developing near Charlotte. The northern portion of the route featured typical late-afternoon buildups that were still well below our cruise altitude, but as we crossed into Tennessee we started spying 45,000 foot giants further ahead. I pulled up the ATIS at Charlotte: "3/4SM +TSRA". We wouldn't be landing in that. I texted our dispatcher to inquire about what she was seeing on radar, and she promptly replied that the thunderstorm was relatively small, but parked directly over the airport with very little movement. She also changed our alternate from Raleigh-Durham to Knoxville due to another thunderstorm developing just west of RDU. I love getting one of the good dispatchers.

Shortly thereafter, Atlanta Center changed our transition for the JOHNS2 arrival and cleared us to hold at BURLS. No surprise here. I entered the hold in the FMS, verified it with Kevin, and activated it. Kevin started slowing early to save some gas, and I began calculating our bingo fuel. Generally, I compute two "Bingo" numbers when holding for thunderstorms at the destination. The first number is traditional bingo fuel, meaning that required to proceed from the hold to the airport, shoot an approach, go missed, proceed to the alternate airport, and have a 45 minute reserve remaining. If the airport is still questionable when we leave the hold, I won't proceed with less than this traditional bingo number. On the other hand, the weather may clear very quickly after the passage of a storm. If the weather is good and there is no question of being able to land, I am willing to proceed from the hold with less fuel than the traditional bingo number so long as I will land with a conservative reserve. In no case, however, will I hold beyond the time at which I have enough fuel to proceed from the hold straight to the alternate and still have a reasonable reserve on landing. I call this my "drop dead bingo" - once the fuel tanks indicate this number, I'm headed for my alternate no matter how quickly ATC says they'll be getting me out of the hold. The spread between the two numbers can be significant, particularly when holding at a point between my destination and alternate airports. This was now the case.

I texted our holding point, altitude, expect further clearance time (EFC), and fuel on board to our dispatcher. Her reply a few minutes later included a bingo fuel number very close to what I had computed for my traditional bingo. We had approximately 45 minutes of holding fuel above that number. If further holding was necessary, my "drop dead bingo" number was over 1500 lbs less, and our proximity to Knoxville would allow me to either coordinate our diversion in a leisurely fashion or arrange for the deletion of our alternate if the weather in Charlotte had improved sufficiently. As we entered the hold over BLISS, I composed a short PA in my mind and then pressed the "transmit" button to deliver the bad news to our passengers.

As it turned out, the news wasn't all bad. Thirty minutes after we entered the hold, Charlotte began accepting arrivals again. We had initially held at FL330 but had been descended as aircraft below us diverted to their alternates. Only ten minutes after Charlotte reopened, we were cleared out of the hold and back on the arrival; a last minute clearance to cross SHINE at 11000' required a steep, nearly full-spoiler descent. As nice as the JungleBus' Vertical Navigation (VNAV) capability is, it can bite you on such late descent clearances. By the time you actually get your crossing limit set up in the box, you may be too high to make it. For this reason, I apply the tried-and-true "3-to-1" litmus test to any crossing clearance before anyone messes around with VNAV. This is a simple method of computing a 3 degree descent: You take the altitude to be lost in tens of thousands of feet, and multiply it by three to get the distance required in nautical miles. If you need to lose 10,000 feet, you need 30 miles for a 3 degree descent. If my quick-and-dirty computation shows a 3 degree or steeper angle required, I will establish the aircraft in a 2500 foot-per-minute descent (or direct the pilot flying to do so) before setting up VNAV. Technology can make life on the line easier, but Rule One is still and always will be: "Fly the Airplane!"

Lightning crackled just to the east over downtown Charlotte as we landed on Runway 18R. This would have been a full, eventful day even if we were done in Charlotte, but we still had two legs left and were now a full hour behind schedule. At least we were keeping the airplane to Atlanta, making for an easy quick turn - or so I thought! When I followed the last of our passengers up the jet bridge to retrieve our paperwork, I found the gate in a state of bedlam, with a long line of harried travelers querying a lone, distressed gate agent. The board behind him advertised a delayed Minneapolis flight, with a departure time well in the past. "Umm, isn't this airplane going to Atlanta?" I asked.

"I don't know anything about that!" the gate agent exclaimed. "Nothing's going to Atlanta. This airplane is going to Minneapolis, but the crew already timed out. Can't you fly it?" Immediately, a half-dozen passengers surrounded me, imploring me to fly them to Minneapolis. "Hold on a sec, folks, I'm going to call our dispatch office and see if that's what they want us to do." I very much doubted that dispatch wanted us to fly to Minneapolis, or our phones would've been ringing off the proverbial hook already; I mostly wanted to know what in tarnation was going on. I walked down the concourse - out of earshot of the passengers - and called dispatch. No answer. I tried a few other desks. Same results. A glance at a national radar display on one of the flight information boards suggested why: a bright red blob was sitting squarely over Atlanta. It's always toughest to get ahold of SOC during Irregular Operations. I just might have to get answers for myself. I walked up to another WidgetCo gate, this one advertising a seriously overdue departure to Atlanta. "Where is 5750 to Atlanta going out of?" I asked. The gate agent pulled up the flight; "It's going out of A7, and the plane is actually already here. It landed almost a half-hour ago!" I chuckled at that. "Well, that's our plane, and you might want to get the agent down at A7 on the same page, because he thinks he's keeping that plane for Minneapolis!"

The change had been announced by the time I walked back over to A7. Now the Minneapolis passengers were upset at their lack of an airplane as well as a crew, newly arrived Atlanta passengers were utterly confused that the gate was still advertising Minneapolis, and that poor lone gate agent tried mightily to be polite in the face of an onslaught of questions and accusations even as he struggled to switch the computer system from the MSP flight to ours. I stood by the gate and fielded as many questions as I could. Many simply wanted to know if Atlanta was open yet, a very good question I was wondering about myself. I still couldn't raise dispatch on the phone. Again, taking the self-sufficient tack seemed best. I retreated down the jetbridge to fill in my bewildered crew on developments thus far, and to call Clearance Delivery regarding our flow time to Atlanta. Clearance confirmed that Atlanta was opening, but our EDCT time was approximately 70 minutes hence and subject to further change. This was better than I was expecting. I headed back up the jetbridge to tell the gate agent that we could start boarding in 20 minutes; I usually plan on getting out to the runway about 15 minutes before any EDCT time so that unforeseen delays don't cause us to miss our slot.

Our EDCT ended up changing three times before we got out. First it was moved back by 90 minutes, then to only twenty-two minutes from the time I got the change. This might have been doable if we had the passengers on board, but a recent arbitrary and capricious rule change by unelected bureaucrats at the Department of Transportation has made us all a little gun-shy about early boarding. This is a lengthy subject meriting its own post, but suffice it to say that nobody wants to be the Captain who keeps his passengers on board for more than three hours between boarding and takeoff and accidentally subjects his airline to a draconian fine (over $2 million in the case of a full JungleBus, or more than one of our engines costs!). Obviously, twenty-two minutes wasn't enough time to board 76 passengers, run our numbers, and taxi out to the runway. Fortunately our final EDCT change was for only fifteen minutes later. A last-minute runway change resulted in taxiing the long way around a dark, unfamiliar ramp while Kevin grappled with the new takeoff performance system. We completed the Before Takeoff Checklist just as Charlotte Tower cleared us for takeoff on 18L, and I taxied onto the runway very slowly so we could take a moment to compose ourselves and double-check our work before hurtling down a darkened, rain-slicked runway at 160 mph.

Soon after we turned right onto the departure, lightning began flickering off to the east. I knew the thunderstorms that closed Atlanta were moving northeast, or roughly onto our arrival route. I have a certain fascination with thunderstorms, but prefer to observe from the ground; this is doubly true at night. Of course we have radar, but the best means of avoidance is the 'ole "Mark 2 Eyeball," and it doesn't work so well on dark nights when lightning appears to fill the sky and you can't see developing buildups. I turned the instrument lights down low and sat forward on my haunches, peering out into the blackness. South of Spartanburg at FL300, we stumbled upon a towering behemoth that materialized out of the ether like an iceburg in the foggy North Atlantic. We turned thirty degrees right without waiting for Atlanta Center's belated approval. There was no lightning and no precipitation to register on our radar, but I don't doubt this pubescent thunderstorm would have given us a ride to remember had we blundered into it. Further along the arrival, the giant cell that closed Atlanta for hours slid by twenty miles to our right, spitting out lightning menacingly the whole time. I wondered what our passengers thought about it.

We were vectored onto the final for Runway 26R surprisingly quickly and I took advantage of the wet runway to make an admirably soft touchdown. Our seventy-six passengers scampered off the airplane in record time in hopes of making their connecting flights. I stifled a yawn as I collected our paperwork at the gate and checked the radar. I'd now been awake for over sixteen hours and on duty for thirteen. Legally, I could be on duty for another three hours! Over a year after the FAA promised to have new duty and rest time regulations in place and the industry publicly promised to support changes in an effort to contain the media fallout from Colgan 3407, the airlines have been successful in fighting a rear-guard delaying action at the Office of Management and Budget. In the meantime, those of us at the pointy end of things do what we've always done (with varying degrees of effectiveness): decide whether we're safe to fly at the time when we're rendered least capable to make an objective judgment. This time I felt fairly alert - but was it just the fatigue talking?

We quickly loaded up, pushed back, and taxied down to 26L for our fifth takeoff of the day. The lights of the terminals flashing by my window seemed to put me in a momentary trance; a gentle prompt from Kevin snapped me back to reality, and I belatedly commanded "Positive Rate, Gear Up." Perhaps I was a little less alert than I previously believed. The route up to Louisville was blessedly clear of thunderstorms; however, that left less to do, and several times I found myself jerking my head up with a start after nodding off into a micro-sleep. Not good. One of my least favorite memories as a pilot is waking up at the controls of a Navajo on a half-mile final to Burbank and not being able to remember the previous half-hour. Having a First Officer on board makes that scenario potentially less dangerous to life and limb but even more hazardous to career and reputation. I put on the oxygen mask on in the descent to get a few puffs of 100% O2, then turned the autopilot off early. I was expecting a straight in approach at Louisville, but I had forgot that our late arrival would put us squarely in the middle of the UPS rush. It was the busiest I've ever seen Louisville. Approach Control made us fly all the way to IIU VOR, then south for twenty miles before turning in to a twenty-five mile final. At long last, we touched down on Runway 35R and made the short taxi to the gate. The passengers seemed to take forever to file sleepily off the airplane. We were hard on their heels.

After all that, one would think I should be able to go to sleep quickly and soundly. Yet the day somehow feels incomplete. Delivering nearly 400 souls safely to their destinations over five legs and 2000 miles through some of the world's busiest airspace and around its wildest weather feels like a real accomplishment, even if it's done on a thoroughly routine basis. I still occasionally feel myself in complete awe that man can fly at all, much less do so with a high degree of comfort, safety, and reliability in a pressurized aluminum tube hurtling through the outer reaches of the atmosphere. Whatever my opinions on the state of my industry and profession, I still feel extraordinarily privileged to be part of the brotherhood of the air, never more so than after a long, tiring day of duty performed well. I permit myself the small celebration of my glass of single-malt, and then turn in for the night.