Is an endless siphon possible?

I’ve been wondering about this for a while now. If a siphon works by having a greater sum force of gravity on the water in one end of the tube than the other, then couldn’t you create a never-ending siphon by having one end of the tube going straight up out of the water, then straight over, then spiraling down (this makes more water be on one side than the other), then stopping above the surface of the water? For example, the water would travel up for one foot, then travel sideways for one foot, then spiral down for 4 feet of tube, but only 10 inches of height. In my way of thinking, the water would run into a place higher than its starting point, so if it was running into the same place it was coming from, it would never stop, unless it evaporated.

Is there some flaw in my thinking? I would imagine there must be, considering this could be a source of infinite (although small, even if made large scale) energy if combined with a turbine.
Thanks for reading.

There has to be a higher head pressure on one side of the siphon than on the other. In other words, if you are trying to use gravity, the downstream end of the siphon tube must be at a lower elevation than the upstream end. The tube in your example wouldn’t fill up because there’s nothing pushing or pulling the water into the tube.

Well to start the siphon, you would have to fill the whole tube with water. But after that, it should continuously siphon. The sum of the gravity on the 4 foot side would be pulling the water in the one foot side, because there would be less force on it.

But it doesn’t work. In the case described, air would enter the end of the tube out of the water instead of water being pulled in from the end of the tube under water. When enough air gets in, the rest of the water drains down through the end of the tube in the water, and you end up with an empty tube except for the part under water.


The greater weight of water in the spiraling tube is compensated for by the upward force of the walls of the spiral tube on the water. The walls of the vertical tube exert no upward force, but since the walls of the spiral tube must have a horizontal component, they do exert an upward force.

If you just worry about the water pressure, instead of the weight, it neatly takes account of all these factors. This leads to the result often summarized as “water seeks its own level.”

But is it possible to make a perpetual siphon that combines gravity and (surface tension?). If you suspend a strip of cloth with one end in the water, and the other end above the water, the cloth will absorb water until it is saturated. Would water drip off the end out of the water? If so, why isn’t that making a perpetual motion machine? I presume for some reason the water won’t drip out. I know it wouldn’t be the ‘until the end of time’ type of perpetual, but if does seem to defy logic.

The cloth won’t absorb water until it is saturated - it will absorb a certain amount of water depending on one or two factors I shan’t go into at this hour of the morning, then stop. You can get continuous wicking if you’re able to remove the absorbed liquid by some means (evaporation, for instance) but you need an energy input to do this. See here.

That would be the dipping bird type of motion. It didn’t seem it would be possible based simply on gravity, and anything else would seem to reach equilibrium in a closed system.

You are misunderstanding how syphons work. “Gravity” doesn’t have very much to do with them. Syphons work on the air pressure difference between one end and the other. That means three things:

  1. Syphons don’t work in a vacuum - at all.
  2. The maximum vertical distance that a syphon can extend above the intake is determined by atmospheric pressure, and is around 32 feet, but probably less in practice. 3) The maximum vertical distance that the syphon can rise above the intake is also determined by the density of the liquid - it’s only around a yard for Mercury, which is much denser than water.

Solution for your problem:

  • Buy 2 buckets and 6 foot of gardenhose
  • Fill one bucket with water.
  • See for yourself how this stuff works.
    If you found perpetual motion, you keep the profit.
    If you learn something else wire me $20.

Now what I want to shoot for is an inverted J-shaped tube less than the maximum capillary length for its diameter. Dip the bottom of the long arm in water, and obviously water will rise up the tube, round the bend, and drip down again under gravity. Profit! :dubious:

Once the capillary walls have been wetted, they have no more pull. Like all perpetual motion machines, this is doomed to failure.

I’m afraid you’re incorrect. Jearl Walker wrote about siphons in vacuum in his Flying Circus of Physics. And from Wikipedia (with four cites)

By the way, people have tried to build perpetual motion machines using siphons, surface tension, and gravity. None of them worked. My favorite one was two finely-supported circuklar plates that pivoted amost frictioonlessly about their centers. They were placed with their centers far apart, but with sthe surfaces of the plates very close together, then in a bath of water. The idea was that surface tension would make the water climb the gap between the plates, which would make that side heavier, so the plates would turn in that direction (one clockwise, one counterclockwise), which starts the process over, so that you get perpetually counter-rotating plates.

Didn’t work

Yes, gravity is the key to the operation of a siphon. Short siphons can work in a vacuum, but no siphon can work in zero gravity. Atmospheric pressure is very useful, because water has very little tensile strength. Atmospheric pressure guarantees that the water is always in compression.

For the master’s take on this see:

How about filling us in the limits of capillary action now?

If the capillary action is strong enough to pull in water against the pull of gravity on one end, it’ll also be strong enough to hold it in against the pull of gravity on the other. So it’ll go in, but it won’t drip out.

If this endless siphon was possible, it would also be possible to set up a funnel feeding a thin tube that looped back up into the top of the funnel - the greater weight of water in the large funnel forcing the small amount of water in the thin tube around the circuit.

But of course that doesn’t work. If it did, the other thing that would happen is that the oceans would empty themselves out via the smaller bodies of water attached to them (rivers, etc)

Yeah, I know. :wink: :smiley: