[pedrocr]

You know of a path that keeps the acceleration 1G and pointing downwards at all times? Or you mean 1G in total acceleration that can point in any direction? I'd be curious to see a reasonable path for the second and don't see how the first can be possible (and 1G usually means more than just an acceleration vector of length 1 otherwise we wouldn't talk about negative G).

[kobeya]

Nobody is talking about a gravity vector pointing towards the center of the earth during the entire maneuver. The implied assumption, I believe, is 1G towards the floor of the plane at all times.

This is quite easy to do when you realize that the plane can (and will) lose altitude and transfer ground-relative horizontal momentum for the vertical.

[pedrocr]

>Nobody is talking about a gravity vector pointing towards the center of the earth during the entire maneuver. The implied assumption, I believe, is 1G towards the floor of the plane at all times.

I mean downwards towards the floor of the plane of course, not the center of the earth. mikeash has already explained much better than I could why you can't keep 1G from the point of view of the passenger and do a normal barrel roll from/to level flight. It seems easy to have a path that does it if you allow it to finish in descending flight.

[jacquesm]

> You know of a path that keeps the acceleration 1G and pointing downwards at all times?

Yes, but you're not going to like the outcome.

Freefall.

[shalmanese]

Freefall is a 0G maneuver.

[jacquesm]

From your perspective, yes. But you're accelerating with 1G right up until you reach terminal velocity (or impact) unless you're in orbit.

[shalmanese]

Maneuvers are always from the reference frame of the passenger. If a passenger is feeling subjectively to be in 0G, it doesn't matter what their acceleration is compared to any fixed reference frame.

[falcolas]

Or if you are just far enough away, with just enough lateral velocity, you keep missing the planet. Silly astronauts, can't even fall properly.

[falcolas]

The second is an aileron roll done purely with the ailerons, engine and tail, not using the lift generated by the wings. Not all planes can do it because the engine needs to be very powerful to support horizontal flight, and the wings need a neutral airfoil shape to provide lift when upside down.

Trying to come up with a better explanation for the first.

EDIT: Try this on for size:

Given - the force imparted by the wings is 1G (enough to cancel the force of gravity), and will always be pointed straight through the roof of the aircraft. The pilot takes no action to increase the amount of lift. The G's are measured from the frame of reference of the passengers in respect to the aircraft. Rotational forces are not considered - most people are unable to kinesthetically perceive any rotation which occurs at less than 5 degrees per second, and don't realize that they could be upside down and still feeling like they're right side up. It's why instrument training involves so much instruction and reminders to trust the instruments, not your body.

The maneuver is initiated by rolling the plane (clockwise, with respect to the pilot) with the ailerons. Since no effort is made to change the amount of force being generated by the wings, the downward component of that thrust (as measured from an external reference point) will lessen, allowing the aircraft to start accelerating downwards while also accelerating to the right. The force felt by the passengers is still exactly 1G - the force created by the wings, and it is still pointed vertically through the plane. Since the downward component affects the passengers and plane equally, there is no measurable effect on the G forces with respect to the passenger's frame of reference.

The aircraft reaches 90 degrees, and is accelerating to the right at 1G, and downwards at 1G. The downward force is not felt by the passengers, again because their entire frame of reference is accelerating at the same speed, only the force pushing them into their seats.

The aircraft reaches 180 degrees, and is now accelerating at 2Gs downwards. Passengers are still feeling only 1G of pressure from the seat.

270 degrees - the acceleration to the left cancels out the previous acceleration to the right, passengers are still being just pushed into their seats.

360 degrees - the maneuver is complete. The plane is significantly lower and some distance to the "right" of its original position. Their velocity now includes a significant downwards component.

Now here's where things perhaps become a matter of semantics - pilots would consider the barrel roll to be completed at this point, and they simply need to recover from their new orientation. Of course, this will probably require little to no input from the pilot, it will simply happen naturally due to the changed angle of attack induced by the downward motion increasing the lift generated by the wing.