Thursday, November 29, 2007
The powerplant itself is actually one of the more minor differences, at least when transitioning from a turboprop to a modern jet. Most of today's "jets" are actually turbofans and little of their thrust is true "jet thrust" caused by exhaust gasses being expelled from the tailpipe. Rather, the majority of the thrust is generated by the large fans on the front of the engine, very similarly to how propellers are used in turboprops. The GE90 engine used in the B777, for example, has a 9:1 bypass ratio, meaning that for every pound of air that moves through the engine core and is used in combustion, nine pounds of air are accelerated by the fan blades and bypass the engine core. Similarly, the PW150A turboprop on the Q400 gets 85% of its power from the propeller; the remaining 15% is supplied by jet thrust from the engine core. You could almost say that modern jet engines are really ducted turboprops. The main difference is that turboprops have fewer blades of much larger diameter that are geared down to slower rotational speeds. This makes them more efficient at lower speeds and altitudes but limits high speed potential because the propeller tips encounter supersonic speeds long before the aircraft is supersonic. In turbofans, the duct and fan blades/low pressure compressor cause an area of high pressure that slows incoming air enough that the aircraft can cruise at transonic or even supersonic speeds while providing steady, subsonic airflow to the engines.
Operationally, the main difference in powerplant operation has traditionally been spool up / spool down time. A piston engine can usually go from idle to full power in very little time. Normally aspirated piston engines just need more fuel/air mixture to accelerate, and this is available instantaneously. Jet and turboprop engines need more compression, which is obtained by adding fuel to the combustion section, causing the turbines to spin faster, which drives the compression section faster, increasing the mass of air available to the engine for increased power. There's some lag involved. On the first generation of turbojets, it could take as long as 10 seconds to go from idle to full thrust, which caused several accidents when pilots inexperienced with jet engines found themselves at low altitude and low airspeed with "unspooled" engines. Adding another spool to the engines so there were separate low pressure and high pressure compressors helped considerably but lag was still a factor so long as most of the thrust was generated by accelerating hot gasses out the tailpipe. With high bypass turbofans, however, thrust increases as soon as the low pressure spool accelerates. Spool-up time can still be somewhat lengthy - it varies by design - but the engine doesn't need to be fully spooled up to produce significant power. Turboprops still hold an edge in providing near-instantaneous power but the difference isn't as much as it used to be.
Now this is not to say that going from turboprops to jets isn't a significant transition. It is, but generally for reasons not directly related to the powerplant. Most of the differences are in aircraft systems, high speed/high altitude aerodynamics, and low speed swept-wing characteristics.
Compared to older turboprops like the King Air, Metroliner, or J31, most jets are more capable, more automated, and designed with greater redundancy. Over the long run this makes life a lot easier on the pilot, but during the transition there are significant systems differences to be learned. Some systems like the hydraulics or pneumatics may be more complex, unfamiliar technologies like fly-by-wire may be incorporated, and the avionics (and the electrical system that powers them) will be more important than anything else in the airplane. This really isn't a prop vs jet issue; it's an old plane vs new plane issue. Recent turboprop designs like the Q400 and Piaggio Avanti utilize technology to about the same degree as jets of the same vintage.
Like I said, the aerodynamic limitations inherent to propeller-driven engines mean that they typically operate at lower altitudes and slower airspeeds than jets. Therefore, the transition to jet aircraft also involves learning about high speed and high altitude aerodynamics. Most transport-category jets regularly cruise at speeds of .75 to .85 Mach (75%-85% the speed of sound). This is within the transonic speed range, meaning that while the aircraft itself is below the speed of sound, airflow over some parts of the aircraft will be accelerated to supersonic speeds. This typically isn't much of a problem until airflow over the wing reaches supersonic speeds, which is known as Critical Mach or Mcr. At Mcr, a shock wave begins to form on the wing; as speed is increased beyond Mcr, the shock wave intensifies and moves aft, which has several undesirable side effects. The shock wave moves the wing's center of pressure aft, which requires increasing amounts of elevator to counteract. Eventually, the elevator may run out of authority and the aircraft will pitch down, which increases airspeed and exacerbates the situation, a phenomenon known as mach tuck. Additionally, the shock wave will cause flow separation which decreases aileron effectiveness and, depending on aircraft configuration, may blanket the horizontal stabilizer and contribute to the tendency towards mach tuck.
The problems associated with transonic flight have been well known for 60 years and aircraft designers have a number of weapons in their arsenal to combat them. The use of swept wings is universal among aircraft designed to cruise above .70 Mach; it delays critical mach by artificially increasing the wing's fineness ratio, since the relative wind travels a greater distance across the wing than its actual chord. Supercritical wings, which have a relatively flat upper surface and an unusually curved lower surface, decrease the strength of the shock wave that forms on the upper wing at speeds beyond Mcr. Vortex generators add energy to the boundary layer and increase resistance to flow separation in both high speed and low speed flight. Trimable horizontal stabilizers provide the elevator authority necessary to counteract the aftward shift in center of pressure. All these design features can decrease or delay undesirable effects of supersonic flow but they do not eliminate them. For this reason, new designs are thoroughly flight tested and a maximum operating mach limit (Mmo) is established. This is in addition to Vmo, the maximum operating airspeed limit, which provides structural rather than aerodynamic protection. Typically Vmo is limiting below FL250-FL300 and Mmo is limiting above those altitudes.
If high-speed aerodynamics were your only worry at high altitude, staying out of harms way would be a snap: don't exceed Mmo. However, low-speed aerodynamics also come into play. It's somewhat counterintuitive, because you're zooming over the ground at 400+ knots, so you'd think you wouldn't have to worry about stall speed. Keep in mind, however, your indicated airspeed will be quite low compared to true airspeed when at high altitudes. Angle of attack varies with indicated airspeed, meaning you're much closer to a stalling AoA in high-altitude cruise. If you plot maximum and minimum airspeed versus altitude on a chart, you'll see the two lines eventually merge as altitude increases. The area where there is little margin between high speed and low speed danger zones is colloquially known as coffin corner. Keep in mind that the danger of stalling comes not from low airspeed but high angle of attack; therefore, an altitude that's perfectly safe for level flight at a light weight in smooth, cold air could be dangerous for maneuvering, at a heavier weight, in turbulence, or at higher temperature. You can avoid coffin corner by using your aircraft's maximum cruise altitude performance charts: they'll show you the highest altitude you should cruise at given a certain aircraft weight and air temperature.
One final consideration is the handling characteristics of a swept-wing jet. Sweeping a wing introduces significant spanwise flow at low speeds, making the wing tips susceptible to stalling before the roots. This condition is unacceptable for a few reasons ranging from loss of lateral control to the forward shift in center of lift which causes a further pitching up moment. Aircraft designers use a few tricks like washout, cuffs, or vortex generators to keep the tips from stalling first, but the fact remains that swept-wing aircraft seldom stall as docilely as straight-wing airplanes do. The solution is to never allow the aircraft to stall, and to this end most swept-wing planes have stall protection devices like stick shakers and stick pushers. That said, you don't need to stall a swept-wing jet to get in a pickle; getting slow at low altitude will do nicely, as the spanwise flow also guarantees rapidly increasing induced drag below Max-L/D. When jets are lost in landing accidents, it is often due to an unarrested low-speed sink that results in impact short of the runway.
Lastly, some swept wing aircraft are susceptible to "dutch roll," a phenomenon of simultaneous uncoupled roll and yaw oscillations that progressively increase in magnitude. It's most prevalent in aircraft with strong lateral stability; swept wings are inherently stable laterally. Basically a sideslip causes extra lift on the side of the slip, resulting in a rolling moment in the other direction at the time that directional stability is causing yawing moment in the original direction of the sideslip. If the directional stability is weak compared to lateral stability, the yawing moment lags slightly behind the rolling moment, allowing an opposite sideslip to develop, and the process repeats again, feeding on itself. In extreme cases it can render an aircraft uncontrollable; susceptible designs prevent it by installing a yaw damper. If your yaw damper can be deferred (MEL'd), your aircraft probably isn't extremely prone to dutch roll, but you'll want to make a mental note to stay coordinated with an inop yaw damper rather than flying with your feet on the floor like usual!
None of this is very hard stuff to grasp and the procedures for dealing with differences between props and jets are pretty transparent given all the other things you have to learn when transitioning to a new aircraft. I personally found the transition from prop to jet very easy, at least in the simulator training phase. I'll fly the real thing for the first time in a few days when IOE starts.
Wednesday, November 28, 2007
I actually got it a week ago, last Wednesday. The fact that I was able to do the LOE early meant that I got to actually spend Thanksgiving and the weekend with Dawn and our families. They're saying I'll start IOE sometime after December 1st so at the moment I'm in Portland getting the mail and watering the plants.
Sunday, November 18, 2007
The end is in sight!
Friday, November 16, 2007
When people ask whether the regional airlines are safe, as a commenter in one of my training posts did, my answer tends to be more nuanced. Nobody likes a nuanced answer to that question. Well, it depends on what you think the question is really asking. If the question is, are the regional airlines safe compared to, say, driving or skiing or lawn darts- why, yes they are. If the question is, are the regionals safe compared to the major airlines - that's where my answer gets a little wavy. The accident record for the last ten years is not worse for the regionals than it is for the majors. However, I do believe that certain trends at the regionals right now do lead to a decrease in safety margins, and we could be sowing the seeds of future accidents.
My primary concern is the plummeting experience levels in regional airliner cockpits. After 9/11, the regional airlines exploded as major airlines sought to slash costs by outsourcing more small jet flying to the lowest bidder. The regional airlines have been very aggressive in controlling costs so they can competitively bid, resulting in poor pay and working conditions at many regionals. This, along with fewer opportunities at major airlines and the erosion of wages and benefits at that level, has contributed to a sharp decrease in the number of pilots training for an airline career. The end result: a severe pilot shortage in the lower echelons of U.S. aviation, where several airlines have dropped the pretense of minimum experience levels altogether and are hiring at the lowest legal qualification (commercial multi-engine land certificate, 190-250 hours) and still can't get enough pilots to properly staff their airlines for all the contracts they've been awarded.
This isn't the first time that airlines have hired low-time candidates. In the boom days of the late 1990's, massive hiring at the major airlines meant that some regional airlines were hiring pilots with less than 1000 hours to fly 19 seat turboprops. These days pilots with even less experience are flying 50-90 seat jets. Fortunately, these new planes are relatively idiot-proof; this and the training I've been describing is, I think, largely responsible for preventing a rash of experience-related accidents.
The good news is that inexperience has a way of taking care of itself; pilots hired with 250 hours don't stay at 250 hours for long. However, some of the most abusive regionals have massive attrition in their FO ranks as low-time pilots leave as soon as they have enough experience to get hired at a better airline. At Pinnacle, for example, 30% of the entire pilot group has been with the company for less than a year and meager new-hire classes can't even keep up with the attrition. There is a lot of attrition in the Captains' ranks as well as the major airlines ramp up their hiring, and several airlines (including Pinnacle) have started to hire "street captains" because none of their FOs have enough flight time to qualify for the upgrade (typically around 2000-2500 hours)! This is the real challenge, in my opinion. With a seasoned Captain, an inexperienced FO is more of an annoyance than a hinderance to safety. When you put a Captain with little time in type together with an FO with little time in any type, though, I think you will see a real degradation to safety margins whenever abnormal situations arise. And thanks to the seniority system, that's exactly what will happen: the street Captains, perpetually junior, will always get paired with new, junior FOs. Incidentally, I just described my own near future.
Maintenance is another major concern at some regionals. Horizon had a very good maintenance department. They hired qualified mechanics, paid them decently, had a lot of oversight, and did the vast majority of maintenance work with our own mechanics. Whatever maintenance was outsourced (mostly line maintenance at smaller outstations) was done so with rather strict oversight. This is becoming the exception to the rule at many regional airlines (and even a number of major airlines). The new M.O. is to outsource at least all heavy maintenance and often line maintenance as well to contractors. NewCo does not employ a single mechanic, only supervisors and maintenance controllers. This is more than a passing concern: shoddy, under-supervised outsourced work has caused two crashes already, Valujet 592 and Air Midwest 5481. In the latter case, Air Midwest (part of Mesa Air Group) had contracted their maintenance out to the lowest bidder, where an unlicensed mechanic improperly rigged the elevator using an unapproved procedure. The licensed mechanic he was working under signed off the work without inspecting it. All five of the mechanics working on the aircraft that night had almost zero experience on the B1900. Mesa's penny-pinching killed 21 people. After the accident Mesa decided to stop outsourcing maintenance work. "After an accident like that, you reassess," said Jonathan Ornstein, Mesa's CEO. "Bringing the maintenance back in-house is a cost-effective way to facilitate more direct control of the work."
Few in the industry have listened to him, because the trend is definitely toward more outsourcing, not less. The Air Midwest crash impacted Mesa Air Group's bottom line only, and other airlines will not change until an accident cuts into their profits. The regionals are obsessed with keeping their costs low because their survival depends on it. The major airlines share a major part of the blame because of how they play regionals off each other to secure the lowest cost feed possible. Nobody wants to change because doing so puts them at a competitive disadvantage. The obvious solution would be for the FAA to change the regulations, keeping an even playing field for everyone while increasing safety, but the FAA has proven time and time again that they will not do anything to cut into the airlines' bottom line until public outrage forces them to.
One very good example of this is the area of pilot fatigue. There is ample evidence to suggest that the current regulations are inadequate. NASA has stacks of ASRS forms where pilots reported fatigue as being a major factor in deviations and other incidents. The NTSB has excoriated the FAA on numerous occasions for not changing the regulations to conform to modern knowledge about fatigue's effect on human performance. It's actually been on their "Most Wanted" list of safety improvements since 1990. In two recent fatal airline accidents (American 1420 and Corporate 5966), the NTSB found fatigue to be a significant contributing factor. This issue affects the majors almost as much as the regionals, especially since their contracts were gutted in bankruptcy. In an effort to keep costs low, airlines are flying their pilots as much as they legally can. For a major airline that may mean 8 hours of flying in three legs and 12 hours on duty; for a regional that may mean 8 hours of flying in six legs and 14 hours on duty.
The basic thead that connects all of this is airline management being cheap. They won't pay the pilots, they won't pay for quality maintenance, they wring every bit of productivity they can out of their employees, and they'll continue doing so until inexperience, shoddy maintenance, and tired pilots cut into their bottom line. Profits in aviation being underwritten by others' blood has a long and pedigreed history. Google "American 191" and consider that the man who approved the forklift procedure against McDonnell-Douglas' recommendations is now in the top echelons of AirTran management. Read the NTSB report on Alaska 261 and think about what it means that, after everything they said about Alaska's maintenance program, the FAA found Alaska MD80s with unlubed jackscrews coming out of (outsourced) heavy maintenance last year. Consider that as cheap as the majors have become, they outsource a portion of their flying to regionals because the regionals are cheaper!
I suppose this post would have a rather frightening effect on non-pilots. You should understand that despite my concerns, accident rates at the regionals are at the same amazingly low rates of the major airlines. I just think that if airline management continues along the path they're on, that won't continue to be the case forever.
Thursday, November 15, 2007
At first I thought it was a 747, but that didn't look quite right. Then it banked and I realized it was a the A380! Mon dieu!
Tuesday, November 13, 2007
Every time you switch between types of aircraft - from DC9 to A320, for example - you have to go through transition training. This is basically the same as initial training except that you don't have to do the Basic Indoctrination portion of ground school, just aircraft systems. Other than that the syllabus is pretty close, although a transition is really a little easier because you're already so familiar with the airline's general operating practices and fleet managers try to harmonize procedures between types. Depending on the number of aircraft types in an airline's fleet and special contract provisions like seat locks, it is sometimes possible to go through several cycles of transition training in a single year. A guy I know got hired onto the MD88 at Delta and only flew it for a month or two before going back to the schoolhouse for transition training on the B757/767.
Closely related to transition training is differences training. The FAA requires differences training when they ascertain that two or more models of the same type are different enough that a pilot already holding the type rating needs to get additional training on that specific model before operating it under FAR 121. The best example of this is the B757 and B767. They actually share a type certificate and their pilots hold a "B756/B767" type rating, but in reality there are some differences (195,000 lb difference in takeoff weight between the 757-200 and 767-400ER, for starters) that require additional training for pilots who fly both. A less obvious example is the B737-200 and the newer B737NG's, because the cockpit is so different. Differences training is typically much less intensive than initial or transition training.
There actually is one instance where a pilot already with an airline is required to go through initial training again: when they have not served on the same "group" of aircraft (turboprop, turbojet). Therefore, at Horizon, Dash-8 pilots going to the CRJ had to go through initial training again, and vice versa. It's basically transition training with a general subjects refresher course thrown in.
When a First Officer is promoted to Captain, they go through Upgrade training. I won't have to do this because I'm already being trained as a Captain; the company didn't want to send me through two training cycles within a few months of each other. Most airlines, however, train their First Officers as FOs only; there's no real advantage to having them all type rated and captain qualified (and most new FOs at the regionals today aren't even close to ATP minimum qualifications).
Upgrade training is, by itself, pretty straightforward: a "refresher" ground school, some CPT sessions to learn the left-seat flows, and then simulator training to learn how to fly from the left seat. The only thing that's really new for most people is learning how to taxi! Because of the way the seniority system works at most airlines, it's pretty common to upgrade into an airplane other than the one you flew as an FO. In this case, upgrade training can be combined with initial, transition, or differences training, as appropriate. For example: a CRJ FO upgrading in the DHC8 would go through initial + upgrade, a 747 FO upgrading in an A320 would go through transition + upgrade, and a Q400 FO upgrading into the Q200 would go through differences + upgrade (which would've been my case had I stayed at Horizon).
Now even if you're hired as a Captain at a one-type airline, you don't escape from training for years on end. There's always recurrent training to do. All flight crewmembers must go through recurrent ground school once a year. This tends to be dreadfully boring, as the FAA has a long laundry list of things that must be covered, which leaves little time for extra training on pertinent and timely issues that'd actually be useful on the line. On the plus side, you often find yourself in class with coworkers you haven't hung out with in a while, and that can be a good time.
And then there's recurrent simulator training/checking. The regs say that captains must take a proficiency check every six months and FOs once every twelve months, but that recurrent flight training may be substituted for every other pro check ("training-in-lieu"). I've discussed this before, but the pro check is actually easier than most training in lieu sessions, because a pro check is a fairly structured event in which everyone knows what to expect, while the instructor can (and will) throw pretty tough, complex situations at you in a TIL session. I actually think it's easier on Captains because they fly the sim every six months and therefore have a pretty good memory of what it flies like...not always quite like the airplane! Maybe I'll think differently once I have to visit the torture box every six months, though.
AQP programs are considerably different where recurrent training is concerned. The time intervals are different and are unique to each airline's program, although I understand that nine months is a common interval. The recurrent sim training is usually done right after recurrent ground school, which isn't always true of traditional programs. Also, AQP doesn't have training one time and checking the next; rather, pilots are trained and checked within the same training cycle. Usually there are at least two sim sessions, including one maneuvers validation and one LOFT (Line Oriented Flight Training) scenario. These are rough equivalents of the pro check and TIL, and doing both in the same cycle is why AQP programs get away with longer intervals between training cycles.
Finally, each pilot is required to be line checked once a year. This involves a line check airman sitting on your jumpseat during one or two legs of normal revenue flying and silently taking notes of what you're doing wrong. You have to do something pretty blatantly unsafe to actually bust a company line check, but at many airlines even small screwups count towards a running point total. If you reach a certain number of points over a given time period, you get hauled in for retraining or even a review board.
I just calculated how much time I've spent in training in less than four years at two airlines. My initial at Horizon took about ten weeks including IOE, I sat through three recurrent ground schools, and I had one recurrent pro-checks and two TILs. Initial here at NewCo will be about six weeks (not including the self study period) plus about a week of IOE. So that'll bring me up to about 20 weeks of training in under 4 years. That's actually not bad, I'd bet there are people who have done that in a single year. Like I said, it's a major part of an airline pilot's life - and that's a major reason the airlines are so safe.
Sunday, November 11, 2007
Tuesday, November 06, 2007
I've had the same Dell Inspiron 1100 laptop computer for the last four years. It goes with me on almost every trip, and sees a fair amount of use at home as well. It was a budget laptop when I got it (Celeron 2GHz, 256mb ram) and has long since been left behind by technological progress, but it's still pretty adequate for web surfing, blogging, chatting, and word processing usage. For more processor-intensive computing I use our home desktop machine.
Four years of heavy usage have taken a toll on my laptop. The case has a chip out of a corner, the screen has a crack on its edge, and there are scratches aplenty. It never had internal wifi capability, so when the PCMIA slot stopped working two years ago I had to swap out my PCMIA wifi adapter for a USB version. About the same time the power supply module went bad so that the computer would revert to battery power and refuse to accept the AC power adapter if I let the battery fully charge. I adapted by taking the battery out of the computer once it reached a 99% charge and then not using it until I really needed battery power. A year ago the power supply cord also went on the fritz, so I got a replacement.
A few days ago, that power supply cord started going bad exactly like its predecessor did: intermittent power that could be made continuous by holding the cord just so in a certain spot, apparently due to broken wires within. I ended up getting a universal power supply at Radio Shack, figuring I could still use it after the Dell passed the point of usefulness. Before hooking it up, I made sure that the input power specs matched and that output DC voltage and amperage (20V @ 4.5A) was correct. Great. I found the correct Dell adapter plug and attached it, and then plugged it into my computer.
Immediately I caught a whiff of the acrid smell of burning electrical wires. I unplugged the adapter right away and plugged in my old adapter. The computer didn't respond when I pressed the "on" button. I put in the battery: the yellow "fault" light illuminated and the computer refused to turn on. Worse yet, the burning smell came back. I double-checked the output voltage, amperage, and wattage of the new adapter. Everything matched the old one. I plugged it in again, and got the nasty smell again. I wasn't sure what was happening, but it was obvious that I'd just fried my computer.
Suddenly I realized why: the power supply output is DC, meaning it's critical that the polarity be correct, but the design of the adapters allowed them to very easily be attached backwards so that the positive wire went to negative and vise-versa. I inspected the power supply side of the plug: sure enough, there was a small imprinted word, "tip," that was apparently supposed to be matched up with the equally small imprinted "tip" on the adapter side of the plug in order to keep the polarity correct. I'd had the polarity reversed.
So here's what I've concluded from this little episode: I'm just tech-savvy enough to be dangerous. To a smarter, geekier techie, this would've been a painfully obvious mistake. The computer-illiterate would've read the manual and alligned the "tip" imprints without ever knowing why they had to do so. I'm the computer version of the 600 hour CFI who's been instructing for six months and thinks they have this flying thing figured out.
Anyways, the end result is that there is a MacBook in my near future. I've toyed with the idea of making the jump to a Mac for years, but the price difference versus PCs of similar performance plus the compatibility issues always held me back. With the newest generation of Intel-based MacBooks, though, Apple has brought the price within spitting distance of comparable PC laptops. I also considered the Dell Inspiron 1520. Built to the same specifications as the base model MacBook (2GHz Core 2 Duo w/4mb cache, 800Mhz fsb, 1Gb ram, 80Gb hdd, 802.11n wifi), the Dell is only $150 cheaper and the Mac comes with better default software. Actually, I know someone who works at an Apple store; with their employee discount it's the same price. The coup de grâce is that Apple has reportedly worked the bugs out of their BootCamp software and bundled it with the newest version of mac os x, Leopard, making it fairly painless to run Windows when you need it to run your "PC-only" software. I'm pretty sold.
By the way, my Systems & Procedures Validation ("the oral") was on Saturday and it went well. There was a pretty strong emphasis on FMS usage, apparently because some previous trainees had made it into the simulator without a good grasp on the FMS. It's a rather easy box to use once you get used to it, though. I start sim training on Thursday.