Feynman Inverse Sprinkler

[TuxDave]

So as I stumbled on numerous threads about a plane and a treadmill, apparantly this other puzzle has escaped being posted on ATOT.

http://www.physics.umd.edu/lec...ch/QOTW/arch4/q061.htm

Essentially if you took an S-type sprinkler into "suck mode" and put it underwater, which direction would it spin. Intuition and the actual results seem to match up. So can someone tell me what makes this puzzle so puzzling? I'm still trying to decipher the "answer" portion of the link I gave above which said that its kinematics are quite different when it stops instantly when the water flow stops.

 

[Cogman]

Kinematics would be different because it is submerged in water, Water provides much more friction then air does hence it would stop almost instantly (as opposed to a regular sprinkler that will spin a couple of turns. Also, it would start up slower, and run at a slower speed then without the water surrounding it, same reason.

As for why it does this, there are several ways to think of it. One is that it creates a difference in pressure when sucking up the water which cause it to flow the way it does.

With a normal sprinkler, it is somewhat a different reason. Like a rocket, in a normal sprinkler the water is pushing everywhere except where the hole is. the pushing of the water cause the sprinkler to rotate (not the water being ejected from the sprinkler.)

I don't know whats so confusing about this problem either.

 

[TuxDave]

It is interesting to note that the reaction force that the water exerts on the tube as it is pulled into the nozzle is equal but opposite to the force exerted by the water on the inside of the tube as the water direction turns from azimuthal to radial. These two forces are in the same direction for the "normal" sprinkler so it accelerates very rapidly to its maximum angular speed, but they cancel in the inverse sprinkler mode. As a result the nature of the motion of the nozzle in the inverse mode is rather different, in that it starts almost immediately when the water flow starts and stops almost immediately when the water flow ceases.

I acknowledge the friction but the answer page refers to reaction forces and not the friction to be the key difference in the kinematics. There's something being said about the bolded part that may be key to what makes this problem tricky....

 

[destrekor]

Not very up to speed in terms of physics and the precise terms (in fact, never took any physics courses... ever), but I think I can explain this in layman's terms.

When water is being pushed through the sprinkler head and forces the sprinkler to rotate counterclockwise, it is obviously building up momentum and only the pipes are creating friction, thus the water takes a moment to bring it to a halt when the water-flow is cut off.
However, when the water is being vacuumed out through the sprinkler, which forces it to go in the opposite direction due to water momentum in the pipes. When the vacuum is cut off, remember that while it has momentum, it is cutting into the water open-pipe first. While water intake allowed and forced that to happen, with no further means of producing momentum, the friction brings it to a halt nearly immediately. Why? Because the water in the pipes, with no place to go, is acting on the entire unit's momentum. I don't know if that is regarded to as friction... but as the pipes are moving in reverse, water is attempting to force itself into those pipes, but cannot get inside as the water already inside is trapped and being forced to hold ground due to the balance of the vacuum.
If external force was applied, water would be pushed through and water would flow into the reservoir most likely. But since its only with internal measures, the drag caused by the water inside the pipes basically is the reason it is brought to a halt.

If the pipes were capped at the very tips immediately after the vacuum stopped, I reckon the pipes would, like in the first video, continue to rotate for a moment.