This is a very highly debated topic, even though scaled tests have been conducted. The popular Mythbusters show on discovery did it full scale, but some still dispute the way they did it. You can probably find thousands of forums that host this same question, and I wouldn't be surprised if it has even been posted here before. So here goes!
A large passenger jet sits on a treadmill the size of a normal runway that they would normally take off on. When the jet starts up, it climbs in speed slowly. The treadmill matches the jet's speed EXACTLY. For every mile an hour the jet climbs, the treadmill goes in REVERSE that exact amount as-well.
The grand question: Will the plane take-off?
Assume that:
The wind is at 0 MPH.
The Jet is powered by turbines.
A large passenger jet sits on a treadmill the size of a normal runway that they would normally take off on. When the jet starts up, it climbs in speed slowly. The treadmill matches the jet's speed EXACTLY. For every mile an hour the jet climbs, the treadmill goes in REVERSE that exact amount as-well.
The grand question: Will the plane take-off?
Assume that:
The wind is at 0 MPH.
The Jet is powered by turbines.
Now let's move to a plane. I think we can all agree there is friction in the wheels. Everyone has said "negligable" well, in this case, any non-0 value is not negligable. So, we will say the treadmill speeds up to match the wheels speed in a near-instantaneos fashion. Let's start at time 0.
[etc...]
This is where friction comes in. Eventually, the friction (which is dependend on the wheel speed) will equal the maxiumum output of the plane. This will cause the wheels to act as a sort of anchor to the ground. Think of what would happen if you replaced the wheels with iron anchors. The plane would not move (but a lot of force would be on the connectors). Yes, it's counter intuitive, but that's what would happen.
--sigh--
This is for those of you who believe it is a realistic possibility to keep an aircraft, operating its engines at full throttle, motionless by way of wheel friction.
Let's take a typical aircraft engine: the PW 4000 series. These have a maximum thrust of at least 50,000 lb (up to 100,000 lb for some variants). This is one of the types of engine currently used on the Boeing 747-400. It has four of them. Since they will normally be taking off at full throttle, is reasonable to assume that they are, at the very least, producing some 200,000 lbs of thrust. Yes, that's right, I said two-hundred-thousand pounds of thrust, and likely more. For those of you who live in the metric world, that's roughly 1 MN. That's right: one mega-Newton. Which is a very large force. If you don't believe me, ask 500 sumo-wrestlers to sit on you.
Wheel friction, on the other hand, is a very small force. This is frequently demonstrated in land vehicles by astute passengers who notice that you can coast a really long way with the transmission in neutral. Aircraft are no different. If they were were, they would not need to install them with wheel-brakes and reverse thrust buckets to prevent them from crashing at the end of the runway after landing. Typical values of rolling friction coefficient for properly inflated rubber tires on concrete are as much as 0.01. The maximum takeoff weight of a 747-400 is about 800,000 lbs, so the rolling friction is probably less than 8,000 lbs, or 20 sumo-wrestlers. "At most 20" is a significantly smaller number than "at least 500". If you don't believe me, ask 480 mathematicians.
Finally, we have bearing friction. Under normal circumstances, bearing friction is negligible compared to other sources of loss. This is demonstrated by children around the world who like to spin the front tire of their bicycles and observe how long, with good bearings, it keeps going. A bearing loss, at its rated speed I estimate to be of the order of 5% of rolling friction (or 1 sumo wrestler). "At most 21" is still significantly smaller than "at least 500." Now, if I remember correctly, bearing friction increases approximately with the square of speed. So, in order for bearing friction to account for our remaining 480 sumo-wrestlers, the treadmill would have to be zipping along at 4,000 kts -- slightly more than 20 times the takeoff speed of the aircraft, and roughly 7 times the speed of sound at sea level and 68F.
Now, I'm no expert on such things, but I suspect that if you tried to roll a 747 along a treadmill at 4,000 kts, you might run into some kind of tire, or possibly bearing, problem. But yes, theoretically, if you had bearings made of magic fairy dust, tires stretched from Superman's scrotum, and a freaking huge hamster operating a hypersonic treadmill, you could keep a 747, with all engines at full power, completely motionless.
To conclude:
* in the universe in which we, and all other beings, live, the airplane takes off and soars through the sky;
* in Candy-coated Sugar bon-bon La-La land, where acid-flashbacks come alive, it is possible that the airplane is unable to take off.



