In this case, consider the fire triangle: fuel, heat, oxygen.

Fuel? Maybe. In an air race, the lighter the plane the faster the plane, since more weight = more lift and more lift = more drag. So the planes are as lightly loaded on fuel as possible. Halfway or more thru the race, the remaining fuel is even lower; these planes burn it fairly quickly at high throttle. The stock fuel capacity of a P-51D Mustang is 269 gallons; this plane probably had about a quarter of that at impact.
http://www.mustangsmustangs.com/p-51/survivors/pages/44-15651.shtml
This article says another famous P-51 racer (Dago Red) carries 150 gal of 180-octane avgas at takeoff.
http://machinedesign.com/article/unlimited-air-racers-the-ultimate-hot-rods-1103

Oxygen? Plenty.

Heat? Lots in the area of the engine, not so much elsewhere. Also, the heat has to be in a place it can ignite the fuel/oxygen mix. If the fuel has not been properly mixed with oxygen, no amount of heat will ignite it. At the speed that airplane hit the pavement, the engine probably embedded itself a couple feet deep in tarmac before the fuel even began to vaporize, so most of the hot metal was out of the picture.

180 octane is ridiculously high (160 is probably a more realistic number); it's actually designed to be very hard to ignite inside the engine, so that the pistons are not destroyed by early ignition due to the extreme compression at which these engines run. I don't know how that relates to an impact-caused vapor cloud, however.

Finally, this doesn't exclude an explosion; it was still POSSIBLE, but some good fortune was involved. Not a pretty day for anyone involved, but it could have been even worse if it HAD ignited.

>How many aircraft have been swept off a deck of a carrier after landing? NONE! Gravity keeps them there.

Sorry to argue but the answer is not "NONE", it's PLENTY. Gravity's great, but have you ever really watched a ship move in heavy seas? 30+ degree rolls are not uncommon, and when big pitching motion is encountered, the deck can actually move out from underneath you at nearly 0g.

I work around a bunch of guys who test carrier-based rotorcraft for the US Navy, and I can tell you (from having watched more than a few of the horror-story videos from testing) that this is a very real risk.

Here's a prime example from real life.
http://www.youtube.com/watch?v=wZSc5T-iUO4
Sorry, but gravity didn't really do much to help here.

More short clips of ugly sea conditions:
http://www.youtube.com/watch?v=H4MbCu_YRM4
http://www.youtube.com/watch?v=v3Mwd-3Kf-4 (0:33 and on)

Sticking a helo to the deck in rolling seas is NOT a trivial business, and downthrust or some mechanical hold-down is essential. It's not such a big deal for a carrier which doesn't move all that much, but every US Navy destroyer which hosts helos includes a winch-down system of some kind. Some are employed at great personal risk to the sailor who must run out under a hovering helo on a deck that's rolling over 10 deg back and forth every few seconds, hook up a cable (with a huge static shock risk), and run back out of harm's way while the cable literally pulls the helo down to the deck into the right position and holds it there. Some are lock-down systems that grab a probe sticking down from the bottom of the helo. You can see that probe and lock system here:
http://www.youtube.com/watch?v=9wWF9hDgl7E

The problem with using your shiny new toys is that you *might* lose one. Then the other kids learn what you know, or at least some of it. That's the risk of maintaining your edge by purely technological means.

Uh, not so fast. The US Navy boasts almost all-weather ops.

http://www.youtube.com/watch?v=EZRMpxUniOw
http://www.youtube.com/watch?v=ZoBZUQLBrp8

Launching aircraft or not, the equipment is still right there in the same spot.

Better have another look at your biology textbook.

The common cold is so hard to eradicate precisely because it mutates (evolves) all the time. Each cold you get is another variant, some from the hundreds that have been around a long time, others that appear via mutation.

http://www.worsleyschool.net/science/files/virus/page.html
http://www.foxnews.com/story/0,2933,311854,00.html
http://www.scienceclarified.com/Ti-Vi/Virus.html ...and more...

Nope, as I posted above, it's NOT some Big Brother attempt to censor the material. Give the military leadership SOME credit - they're not so dumb to think they can put the genie back in the bottle.

Instead, it's reminding servicemen and civilian agencies of the fairly strict policies about what happens if they view classified material on unclassified computers - or even on computers without need to know. If it's done (especially on purpose), it's punishable by pretty nasty penalties, including removal of security clearance, permanent banning from military computer resources, etc..

Yes, it would be against such policies. In fact, that is the exact rationale for instructing military members and associated civilian employees to avoid it.

The military services (both service members and associated civilian agencies) all have a strict policy about accessing classified material. If you do so on an unclassified machine, it's called "spillage", and BY LONG-STANDING POLICY the machine MUST be disconnected from the network and carefully scrubbed of all traces.

And if the access is intentional and made with full awareness of the law, that's punishable by all kinds of nasty penalties.

And no, it doesn't matter that it already exists on thousands of other machines around the world. Until it's officially declassified, it's still classified, and rules and policies still apply.

So this is NOT an attempt to muzzle the information - it's simply following long-standing rules and making sure everyone knows exactly what those rules are.

In case you don't remember, stuff traveling at orbital velocities is positively lethal to spacecraft. The extreme energies involved in these kinds of impacts is enough to send very high velocity fragments in all directions. Sure, some of it will de-orbit, but most will end up in fairly stable orbits that will EVENTUALLY intersect all the other satellites up there. So blowing up one rogue satellite makes one very annoying but eminently predictable problem into a thousand lethal and unpredictable problems.

Last February, a Russian satellite hit a commercial Iridium satellite, and the resulting debris cloud (estimated near 600 pieces in various orbits) has been a HUGE headache for everyone in similar orbital altitudes.
http://online.wsj.com/article/SB123438921888374497.html
http://www.msnbc.msn.com/id/29147679/

In 2008, the US got criticized around the world for blowing up a falling satellite because of the health threats of hydrazine if it landed in a populated area. Aside from complaints about military showboating, there were many scientists who complained about the resulting orbital debris; however, in reality it was a very low-altitude explosion and the debris cloud did de-orbit very quickly (unlike a geosynchronous orbit explosion, which would leave practically permanent debris due to the orbit well above any appreciable atmospheric drag).
http://findarticles.com/p/articles/mi_6712/is_35_237/ai_n29417848/

Read here for some details on the general problems with orbital debris.
http://illuminations.nctm.org/LessonDetail.aspx?id=L376

So no more helpful suggestions like this, please.

I own both a 2008 Prius and a 2010 Camry, both on the recall lists.

Being a flight test engineer, I had to test things out myself. I tried this experiment on BOTH vehicles, and my conclusion was that both cars were completely controllable in this situation. At speeds from 20 to 50 MPH, I put in full (to the floor) throttle and held it, while trying the following actions.

Prius:
- Step on brake: instant engine idle was the result, despite the depressed pedal. Car *rapidly* stopped.
- Shift lever into neutral: engine idled (despite the depressed pedal) and car began to decel.
- Power button pressed briefly: no result (as expected).
- Power button held: after about three seconds, engine cuts off, car remains steerable and brakes continue to work.
- Wife set up in a "blind" experiment: dramatically overestimated "three seconds" and after one second concluded it wouldn't work, got very stressed and experiment was aborted. But in every other case was able to stop the car easily.
(Note that the Prius throttle, power button, shift lever and electrical brakes are computer controlled. If it decided for some reason to ignore any of these inputs, the only mechanical override is the non-electrical portion of the braking and the unassisted steering. There did not seem to be any guaranteed non-computer-controlled shutoff method.)

Camry:
- Step on brake: engine continued to rev but car decelerated very rapidly. No doubt that brakes are effective.
- Shift lever into neutral: engine rev'd (let it off to avoid engine damage) and car decelerated.
- Key to Accessory: engine cut off, car began to decel, steering and braking remained functional.
- Key to off: engine cut off, car began to decel, steering wheel locked as expected.
(Note that the Camry key can be turned to OFF without the car being in Park, and the shift lever is manually actuated. There does not seem to be any interlock preventing these very basic emergency measures, and that key-shutoff will absolutely override any other acceleration commands.)

Primary conclusion:

Even if the throttle sensors are failing to full-pedal application, it would require at least a secondary and possibly a tertiary failure to induce the accidents that are in the news. Since multiple failures are unlikely to occur at the same moment in dissimilar systems, aside from a massive central computer failure, the chances are high that these accidents/incidents are NOT caused by drive-by-wire throttle hardware failures. On the other hand, it is highly likely that if a real problem DOES exist, it is in the central computer.

Secondary conclusion:

The Toyota recall fixing the floor mats and the secondary recall fixing the bushing in the throttle assembly are window-dressing that have no chance of fixing any real problem.

Tertiary conclusion:

Most of the incidents in the news are driver error. 27 data points (in this article) are not statistically convincing, but do tend to support this conclusion on first glance.

My guess:

Most of this is driver error, coupled with driver unfamiliarity with shutoff and emergency features. There is a real chance that a multi-system/sensor failure or massive computer failure could induce unexpected effects, but I find it far more likely that mis-application of the gas pedal was the typical cause. Couple this with time compression and the very normal human inability to accurately perceive time in an emergency, and it's easy to understand why people believe the problem is worse than it is.

Yeah, which is exactly why the US military seems to be insisting that any such systems installed on military aircraft have an on/off switch... wouldn't be too nice to fly into enemy airspace telling them exactly where you are and how soon you'll be over the next surface-to-air missile site.

For civilian purposes, however, there's no need for GPS data to track an airplane. Radar is pretty reliable and very well understood by now, and it's not that hard to design a device to home in on an object and hit it that way. They were doing it back in the Korean War, and today anyone with the ability to build a GPS-tracking autonomous weapon can throw together a radar-based system just as easily. So I'm not sure that this really represents a significant new risk.

With that said, I'm sure there are people out there considering all kinds of interesting nefarious uses for real-time positioning data on flying multi-ton tubes of metal. While I think this system will be a great improvement over what we do now, I am also curious how (or if) that issue will be addressed.

I can tell from reading the other comments here that my opinions will be in the minority, but I can personally testify how GOOD this system is.

I do flight testing of military aircraft, and we did a demo with several planes and helicopters a couple years ago on the "ADS-B" system, which is a component of NextGen. I've played with it inflight myself, and surveyed many pilots who used it. So you know I'm not blowing smoke, I won an award for a paper about this system at the 2006 Society of Flight Test Engineers annual symposium.

To give you some context about what NextGen and ADS-B do, here's the idea. (I think this description will be useful, since it appears most of the comments here demonstrate a profound lack of knowledge of the system... but this *IS* /. so I'm not surprised.)

Each plane is equipped with a transponder. It receives GPS position, and broadcasts a packet of data once per second (much more frequently than the usual radar sweep of 10-15 seconds) containing identity, position, aircraft type, speed, heading, altitude, and more.

With just a few thousand dollars worth of optional equipment, each plane can also recieve these broadcast packets of information DIRECTLY from other aircraft. In other words, an airplane will see what the other nearby airplanes are reporting too. Right now, a pilot has very little idea what is around his own plane - if the controller doesn't warn him, he doesn't know about it. The existing collision avoidance systems only show a rough approximation of what's in front of you at roughly the same altitude, but it's very error-prone (based on WWII-era-technology directional radio beacons), and hard to find the targets in many cases. But this sytem lets you see everything that the airspace controller is seeing, and almost instantly - once per second. We found the pilots experienced a four-fold increase in their ability to identify conflicting traffic in front of them, and for the first time were aware of overtaking traffic too (faster stuff coming up behind them).

The ground-based system rebroadcasts ALL of its data (including skin-paint targets) on a separate radio frequency, so any airplane (or even ground observers) can learn about everything in the airspace. Along with this data, it also uploads precipitation radar and other weather data, plus airport information. So the pilot has access to a vast amount of new information. And most of the systems have onboard maps with terrain mapping, helping to keep the pilot away from mountains and other dangerous "cumulo-granite" features.

For the pilot himself, the increase in situational awareness was simply amazing. The immediate and crystal clear presentation of the location of all nearby planes meant that he knew everything going on around him. For the ground controller, the much higher frequency updates combined with the much more detailed information about each plane means improved ability to track and direct those airplanes.

There ARE a few downsides, but they're vastly outweighed by the improvements. As some comments indicate, it does depend on GPS. Well, duh. But so do the navigation systems already onboard the airplanes... and cars... and commercial trucks... and ships... and trains. If GPS goes down, there will be much worse problems than this system going away. Despite what it sounds like, the radars are not going away - some will, but there will still be enough for "skin paint" and radar transponder tracking if needed (Congress and the FAA are not totally stupid). As to GPS jammers, note that the airplane is receiving the GPS data, and broadcasting its information on a totally separate frequency to the ground and to other aircraft. So any GPS jamming (since it's localized) will only affect a few airplanes, not the whole system. And by the way, all serious aircraft have multiple navigation systems; jamming GPS won't kill any airplanes, despite the alarmists.

Finally, let's talk about real-world - this system was installed in portions of Alaska around 2000-2001, as a trial run. The FAA mandated that all aircraft have transponders, and in fact paid for the boxes for a huge number of private airplanes. Alaska is a good test site, since there are so many "fly to work" and "fly to school" private plane owning families in remote areas. Accident rates dropped sharply the first year - with an 80% reduction in fatalities (WOW!). Many stories emerged about being saved by the system, as multiple pilots often must navigate thru very tight mountain passes in bad weather, and midairs and flight into terrain were too common. So this system has already proved itself.

So throttle back, guys, this is a GOOD thing, and I've been waiting for it with great hope since I got to play with it myself.

> Congress should mandate with a simple law that the telephone company
> must provide DSL to any customer requests it (within six months).
> The twisted-pair lines are already there, except for the need to
> add a neighborhood DSLAM.

If only. You're being naive.

When our neighborhood was first built out around 1990, Verizon ran a single fiber optic line from the telco to the hub on the corner of the main road. All of our neighborhood twisted pair copper lines run to that hub - and are all concentrated for the trip to the central office over that one light pipe. So there's no way for anyone in our neighborhood to ever get DSL, since we don't have twisted pair to the telco, even though we're well within the total distance constraints of DSL.

Since nobody in our neighborhood really wants Verizon digging up all our yards again to run fiber to the house, and there's no way on earth Verizon is going to run a bunch of copper from the telco to our hub, ain't nothin' gonna change anytime soon. So anyone who wants broadband is stuck with cable modem.

So to summarize, the presence of fiber optic technology has permanently prevented us from getting either DSL OR fiber to the home.

Some large corporation sysadmins will be thrilled that certain Google apps won't work correctly anymore.

My computing environment is heavily managed with group policy and very few user rights, and my company has many many thousands of users worldwide. We cannot even use thumb drives or install any software or hardware. For web connection, we are firmly stuck with IE6 and other outdated web software, mainly because of poorly-programmed corporate web apps with ultra-high security requirements (ironically) that the admins cannot afford to update for fear of the unknowns in new browsers. (For crying out loud, we still have a mandatory installation of Netscape!)

So the admins are always blocking off as many non-work-related sites as possible, and having such sites NOT work correctly will only further discourage users from trying them. For example, we can't use GMail or any other popular webmail sites. And I'm honestly surprised they haven't blocked Google Docs or Google Calendar yet, as they could "leak" data to the outside world.

I'd be that most large corporations are also in a similar fight against their users' desires for newer browsers and freer internet access. So I doubt that this move will really encourage many companies to ditch IE6 faster, and may in fact have the opposite effect in some cases.

I have read that one primary reason that the shipping companies don't allow weapons on board is that they fear they'll be used against their own ships. A non-trivial percentage of merchant crew members are bottom-of-the-barrel sailors, who might not be as trustworthy as desired. Give them easy access to weapons, and they might decide to do a little hijacking of their own - from within. Millions of dollars of ransom is a pretty powerful temptation.

With that said, I'm all for arming them anyway - just give the captain indisputable control over access to the weapons, to remove the temptation.

I'm an aerospace engineer - I work on planes, but the concepts are familiar and common.

The upper stage DID tumble immediately. The other three aerospace engineers and test pilots watching with me also immediately said "That didn't look right."

The high-zoom ground tracking camera and onboard cameras showed it much better during the replays, where it's clear the separation wasn't as clean as it should have been. But it did not look like the stages hit each other.

It appeared that not all eight of the retro-rockets fired. They were designed to slow the first stage enough to separate the two stages, before the "tumble rockets" fired. From the footage, the retro-rocket flame is visibly asymmetrical. It appeared that only a few of the retro-rockets fired on one side of the aft skirt fairing. As a result, I suspect that the initial separation was not purely fore-aft, but included a healthy rotational component which nudged the second "dummy" stage in a similar slow tumble.

Some comments on this board say "no worries"; the second stage was just an unpowered dummy mass, and the tumble would have been stopped by the final design's engine. Not completely true. They need a clean, non-rotational separation before the second stage engine fires and can fully stabilize the flight path. So the tumble will DEFINITELY concern the engineers.

Finally, don't worry too much about the onboard cameras cutting in and out. Speaking from personal experience in the flight test industry, telemetry is no trivial matter, and downlinking gigabits/sec of data and video is no small feat. Minor mis-alignments in antenna angle can cause momentary signal dropout. Strong jolts (stage burnout, etc.) can also jostle wiring and cause interruptions.

Despite this tumble, the flight appeared to be overall a great success. As the launch director noted to his crew shortly after the flight, the only real delays on the first launch of a very complicated test vehicle were weather-induced (plus the small matter of a fabric probe cover sock that snagged on something yesterday). All in all, I'm quite impressed.