At 39:08 of Lecture 28 of 8.01x by Walter Lewin, a water siphon would not work if the distance is over 10 meters due to the barometric pressure. However, this claim is seemingly refuted by Periodic Videos, where at 1:20 he said that the idea that atmospheric pressure pushes the liquid through the siphon is incorrect. At 4:57, he did the siphon experiment under vacuum with an ionic liquid and successfully operated the siphon. Is Walter Lewin's statement inaccurate, or is something else, such as viscosity, at play?


3 Answers 3


There is no contradiction between the two sources.

Siphons are permitted by physics because fluid flow through the siphon lowers the fluid's overall energy, since the output end of the siphon is lower. The main requirement for them to work is that the fluid moves together through the siphon. That, in turn, doesn't technically require atmospheric pressure (and doesn't even require that you have a fluid, e.g. see the chain fountain).

Walter Lewin is pointing out that in practice, siphons are hard to get working above ten meters. The reason is that at that height, the fluid would be flowing with negative pressure, i.e. it could spontaneously turn into vapor, which would ruin the siphoning effect. In other words, atmospheric pressure is technically playing a role -- but it's simply the same role it has in everyday life, i.e. preventing all water from spontaneously boiling.

That is not in contradiction with the video you linked, which uses a specially engineered liquid that is stable against this.


A pressure that is less than the vapour pressure of the liquid can make liquid form vapour bubbles.

Then again, a pure liquid that has no already-formed bubbles needs some extra effort in order to create bubbles. At low enough temperature, the surface tension makes possible for liquids to withstand some negative pressure without forming vapour bubles.

The effect can be engineered to exceed the atmospheric pressure by using thin enough (capilar) tubes. (That's how tall trees get water from their roots all the way to their leaves way above 10 meters. There are 100+ meter trees.)


To make it easier to visualize, imagine it's mercury you want to siphon.

If you have a long tube full of mercury that's closed at one end, and you set it vertical in a bucket of mercury, the mercury is so heavy that it drops down some and leaves a vacuum at the top.

If you do the same thing with two closed tubes of mercury in two buckets, the same thing happens.

And if you join the two tubes of mercury at the top, the same thing still happens.

If the mercury drops farther than the diameter of the tube that connects them, then you won't get a siphon.

You could still make the mercury flow if you put a lot of pressure on one side but not the other -- not just atmospheric pressure which is almost the same on both buckets, but something else. But that isn't a siphon.

Does that suggest a solution? The negative pressure that causes the vacuum is due to the weight of the mercury. Which is proportional to the volume. The failed siphon is because of the linear height of the vacuum bubble. So if you could put a volume at the top of your siphon that could hold the vacuum, then the mercury could flow through the part of the tube that was still full.

So imagine a U-shaped tube, with a second tube attached at the U. The second tube is closed at the top. Like the pipe put in your water supply to prevent water hammer.

You carefully fill the whole tube with mercury. You cap both ends and carefully open each end inside buckets of mercury, at different heights. The mercury level at the top tube drops, but not far enough to empty the U part of the tube. Your siphon should work.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.