# Siphon Wars: Pressurist weighs into Gravitite boffin

We've just had a missive from a US reader regarding that most pressing scientific question of the moment: Just how do siphons work? Those of you with a scientific bent will recall the recent Oxford English Dictionary outrage, in which one Dr Stephen Hughes of Queensland University of Technology laid bare a 99-year-old gaffe in …

This topic is closed for new posts.

#### Sure....

It needs air pressure and gravity. it is gravity that causes the liquid columns to have weight. They then create a pressure differential which the air pressure can work on pushing liquid through the pipe.

It wont work in a vacuum and it wont work in zero gravity. Which is why there are are no syphons in space!

#### Reassuring news

"there are are no syphons in space!"

That's probably why the Space Shuttle can pass over Liverpool without losing any fuel.

#### Is it true that there are no siphons in space?

I'm serious - I really don't know. Has this experiment been done in reduced gravity? Surely this is the best way to prove the point - air pressure but no/very low gravity. It would be easy to video and post for all to see.

#### How?

How do you get rid of gravity? The only way I know is to get far enough away from the object exerting the gravitational pull, so do it in space then? Easy to do that, get a ticket to the ISS, oh hold on, what's happened to the other part of the experiment, the air pressure. I think we might lose that too.

#### Re : siphon

I've already posted this but here go again :

Gravity alone drives the thermodynamics. The energy comes from the difference in height of the reservoirs multiplied by the mass of water moved.

The mechanism requires that the 'hump' over which the liquid has to climb is such that the liquid will not form a vapour bubble at the highest point - this point will depend on the atmospheric pressure. With a high-boiling liquid and a low hump the 'outside' pressure could be quite low. Practically this would require the entire app. to be a partial vacuum chamber

In a normal environment, of course, atmospheric pressure is related to gravity.

#### True, but not vapour point

But the point is that atmospheric pressure could not be zero, even with a theoretical liquid with an infinite boiling point. The static pressure at the top of the siphon tube (the 'hump', as you call it) has to be less than the atmospheric pressure at the surface of the reservoir, or the liquid cannot flow. This is not to do with vapour point, but pressure gradient: the liquid will always flow from a region of higher pressure to lower pressure.

If there is no atmospheric pressure, the liquid cannot flow up the tube; the static pressure at the top of the tube can never be less than zero.

In other news, the liquid only flows through the siphon because it is denser, and therefore heavier, than the surrounding atmosphere. But if it wasn't, it would float away...

As an aside, surely this isn't difficult to understand? I'm a musician, for heaven's sake...

#### Re : True, but not vapour point

You may have noticed that I never mention zero atmospheric pressure. The vapour pressure is important however - the 'weight' ( actually pressure = weight/area ) of liquid in both legs is resisted by the outside pressure until a point is reached with a hump that's too high (~10m with water at normal atmospheric pressure) when the liquid pressure cannot be balanced by atm. pressure and a vapour bubble forms at the point of lowest pressure - the top of the hump. This is exactly the same as a barometer - if you tried to siphon mercury ( don't suck !) the hump would only work up to ~1m at normal atm. pressure.

As I mention above the system would work with a high-boiling liquid and a low 'hump' at very low atmospheric pressures

#### More pedantery

"the liquid will always flow from a region of higher pressure to lower pressure"

This is wrong.

A siphon transport a liquid from a reservoir A to another reservoir B as long as the surface of the liquid in A is elevated above the surface of the liquid in B.

Under normal conditions, this higher elevation means that the atmospheric pressure at the surface of the reservoir A is lower than the atmospheric pressure at the surface of reservoir B (Since there is a slightly higher column of air above reservoir B than above reservoir A).

So the liquid runs from low atmospheric pressure to higher atmospheric pressure.

Which is exactly why stating that the siphon is driven by atmospheric pressure is misleading.

#### Not just atmospheric pressure

Pressure isn't just caused by the atmosphere; it's also caused by the column of liquid. I do not say that gravity doesn't cause the siphon effect, but that the siphon effect could not happen in a vacuum, as the pressure of the liquid alone would not be enough to move the liquid over the 'hump'. A vacuum bubble would appear at the same level in the tube as at the surface of the reservoir.

The only reason that the liquid and atmosphere cause pressure at all is gravity. Gravity also makes liquid flow downhill; this is not a factor of pressure but rather the transfer of gravitational potential to kinetic energy. When the liquid is flowing uphill – as in the part of the siphon before the 'hump' – its gravitational potential is increasing: gravity is working against the flow of the liquid in this section.

The force driving the liquid up into the 'hump' is a static pressure force exerted on the liquid in the bottom of the tube by the sum of the liquid pressure and the atmospheric pressure above it. This force encourages the liquid up the tube. Because of gravity, the weight of the liquid within the first stage of the tube is also exerting a downward pressure force; this force works against the other force. The difference in force which enables the liquid to flow over the hump is caused by atmospheric pressure.

The liquid is denser than the atmosphere, so exerts a greater static pressure for a given volume (or height, if you will). Once the liquid is over the hump, it does indeed fall due to gravity, rather than the atmospheric pressure. But the flow is maintained because the static pressure at the bottom of the siphon is higher than the static pressure in the receptacle; while the atmospheric pressure may be slightly higher at the lower altitude, the liquid pressure is higher still, as the liquid is denser than air.

The problem is that, in the absence of any atmospheric pressure, the liquid simply cannot get over the hump. Without atmospheric pressure, there is no force to drive the liquid against the force of gravity. And if you are thinking 'suction'... well, suction is pressure difference, and there can never be less pressure than zero, so it wouldn't exist in a vacuum.

(*Aside* There are other things going on too, of course; I notice that some have cited surface tension, by which I assume they mean the bonding between the liquid molecules; surface tension only happens at a surface, and there isn't one inside a siphon. But liquid bonding forces are weak compared with pressure forces; that's what makes liquids liquids. Dynamic pressure, too, undoubtedly plays a part, particularly in the downward portion of the tube, but I'll leave discussion of that to fluid dynamicists.)

So the siphon – that part of the mechanism that gets the liquid over the edge of the reservoir, rather than the part which delivers it into the bucket – is, arguably, driven by atmospheric pressure.

I accept that you need gravity for a siphon to work. (Though I should point out the obvious here – the whole concept of a siphon is meaningless in zero gravity: which way is up?) But I will strenuously argue that it won't work in a vacuum, which means that atmospheric pressure is an essential element in the process.

In short, asserting that a siphon isn't driven by atmospheric pressure is more wrong than asserting that it is.

#### I wasn't disagreeing

I was agreeing.

I was merely taking the argument one step further by considering a hypothetical situation with a zero atmospheric pressure. It wouldn't then matter how high your vapour point was; a vacuum would still form at the top of the hump, at the same level as the liquid in the reservoir.

Sorry if it sounded like criticism; you clearly understand the problem, but I'm amazed at how many people do not. Good call with the barometer analogy; I was thinking the same myself, but couldn't remember what the damn thing was called!

Taking things further again (and again not criticising), I think the gravitational force is also relevant as to how high your hump could be. On the moon, you would need far less atmospheric pressure to overcome the 'weight' of the liquid in the hump than on earth...

#### Re : Not just atmospheric pressure

I think we need to clarify some terms.

"driven by" - to my mind that means the thermodynamics of the situation and that is the energy derived from potential energy - in this case gravity

An 'atmospheric' pressure is necessary for the mechanism to work but it doesn't contribute to the overall energy change. There are a number of factors - height of hump, boiling point of liquid, atm. pressure, temperature, width of pipe, end of pipe in or out of bottom reservoir that modify the behaviour of the system but only gravity drives it.

The limiting height for the hump is that the shorter limb is less than ~10m in the case of water

#### Thought experiment

Here's a thought experiment: set a siphon going and then vary the force of gravity. The rate of flow will vary. Now vary the atmospheric pressure. The rate of flow will not vary (provided the pressure remains high enough to prevent a vacuum forming in the pipe).

A siphon needs atmospheric pressure to prevent a vacuum forming in the pipe, but it's gravity that actually pulls the liquid through.

#### hmmm

Surely if the atmospheric pressure is the same at both ends then it is the gravity that is doing the sucking?

#### surface tension (as opposed to scientific tension)

The crucial thing about siphons is that the liquid being drawn must be continuous - you can't siphon sand. It;s the continuity that holds the key. Once you have used (the lack of) air pressure to get liquid out of the container and over the hump and down to below the level in the container then yes, the gravity on the liquid _below_ the level in the container does draw it down.

However, if it wasn't for surface tension holding the column of liquid together, all that would happen is that the bit of liquid you've sucked over the top of the siphon would fall down the tube and the rest would fall back into the original container. It the ST which ensures that once gravity has got a grip on the liquid in the tube which is below the level in the container, it will continue to be drawn out of the container and down the siphon.

p.s. if scientists can't agree on what causes an effect Archimedes observed X thousand years ago, what hope have they got discovering fusion or researching climate change?

#### I agree

Even if you fluidise the sand there is no way it will climb a "primed" siphon tube in the promise of an energetic "reward". The "reward" is communicated by the surface tension.

For the doubters:

If i cut the end off a syringe, push it into vibrated, fluidised sand, and pull the plunger, does the sand rise up the syringe? - answer no, just a small swirl that is carried on the wind between the grains.

Just looked at the wikipedia which posits an interesting test case - when there is a bubble in the siphon. If the down-leg of the siphon is long enough it will draw out the bubble and run normally. Clearly there is no surface tension across the bubble, hence it is not required.

That said, i cannot reconcile this with the previous sand argument.

#### bubbles do have surface tension

>Clearly there is no surface tension across the bubble,

Then what is it that makes the bubble? Surface tension acts on a drop of water to make it spherical (as that's the shape with the smallest area::volume ratio) since the surface tension will try to minimise the size of the droplet. Bubbles are the same shape for exactly the same reason - because they ARE formed by, and therefore have, surface tension at the air-bubble interface.

In the down leg of a siphon, you have the weight of the water from the bottom of the tube to the bubble weighing down. then you have the bubble fully enclosing the width of the tube (otherwise it wouldn't be stable and would just float up) then you have the rest of the water column. The weight of the water below the bubble will expand the size (actually the length) of the bubble - which still holds together under surface tension, unless the weight of water below it is too great and "breaks" the surface tension - thus fracturing the bubble into smaller ones which then wouldn't fit the width of the tube and so would rise to the top of the siphon. As the weight of water increases the size of the bubble, the bubble's volume increases and it's internal pressure drops BUT the bubble still holds together due to the surface tension acting on it. The bubble still gets drawn down the siphon by gravity on the water below it and the surface tension keeping it as a single bubble. So I stand by my original point that it's the surface tension of the water which makes the siphon work. it acts like the "glue" which holds the water column together as it is drawn down the tube.

#### reconciling with sand argument

The reason the sand will not flow through a siphon is because air can move between the grains, equalising the pressure inside the tube with the pressure outside. (Well, the pressure in the tube does not drop in the first place, since there is already air present between the grains, below the level of the inlet and all through the tube. The air can enter the tube through both ends as the sand slides out.)

With a liquid, there is no air below the inlet, and the liquid will prevent air from being drawn down into the upper container and from there into the tube. (Until the liquid level drops below the inlet level, of course.)

Gravity certainly plays a role, but the air pressure is still the key.

#### @Rattus

You are correct. In addition to your argument, sand has a much greater friction than water. In fact sand acts more like a solid in many cases than a liquid.

But one could still get a siphon working with sand if the air pressure were high enough and the side of the tube smooth enough.

#### Sand

What about the air moving between the grains of sand?

#### air moving between the grains

Yeah, that would lead to losses. But with a high enough pressure, you could still get it to work - not as efficiently as water, though.

#### Defintely Gravity

In the Aquaponics world, we use siphons all the time. Its what drives our systems.

The atmospheric pressure does not need to be present. Take two sealed containers, half full of liquid, with a connecting siphon and a vacuum. Now lower one. The siphon will start running. What matters is the pressure on both sides needs to be equal to make the siphon work. However it is NOT the driving force of the siphon. Gravity is. QED.

#### not true

As you evacuate the sealed system, a bubble will appear at the top of the previously full siphon tube, and this bubble will grow until the liquid level in both ends of the tube is the same as the surrounding liquid, minus some small capillary effects.

#### Under Pressure

Well, yes, and that's pressure (the fact that it's not atmospheric pressure is irrelevant).

I think the argument here is whether the start-up mechanism is relevant or not. I'm actually in agreement with most people here; I think it doesn't matter since the main thrust of the definition is what makes the water flow in the process, not what starts the process.

So leave atmospheric pressure out of this everyone (including you, people from Colorado Springs)! It just confuses the explanation.

#### my 2 cents

when the atmosphere pressure on the cylinders is the same, then gravity what determines (i.e. balance) the liquid. Otherwise, the liquid will remain unbalanced once you tilt the cylinders (atmospheric pressure is the same, gravity in a different direction)

#### Dont just post articles from any idiot

I can assure you that as long as both ends of a pipe are at similar pressures (the difference of say a half a meter from a car's fuel tank to the ground is approximately no differnce in pressure.

The "Sucking on the hose" the author reffers to is in order to overcome the force of gravity not to create a pressure differential. It is possible (and indeed prefferable for many of the more unpleasant tasting liquids) to immerse the length of hose in the upper pool stick your thumb over the end of the pipe to hold the column of water in place then lower the end of the pipe and remove your thumb.

Bingo your siphon will start to flow because you have a solid column of water and one end is in a lower gravitational energy state than the other.

the act of sucking on the pipe is just a handy way to overcome the initial gravitational potential when it is not possible to immerse the hose.

Asside from the difficulty of whatever liquid you used evaporating there would be nothing to stop you siphoning in a vacuum.

Regards BSc Physics

#### Not quite

In a vacuum, there would be nothing stopping a "vacuum bubble" forming at the top of the hump and the liquid falling off in opposite directions. The siphon most likely wouldn't work.

> the act of sucking on the pipe is just a handy way to overcome the initial gravitational

> potential when it is not possible to immerse the hose.

The sucking method would not work in a vacuum of course.

And this is the reason why the siphon in a vacuum, has problems.

The liquid in the longer drop is effectively sucking up the liquid in the shorter end because it there is more of it and therefore heavier.

But it is still sucking.

#### Exactly

So right. The key is potential energy as provided by gravity, nothing else. Besides, siphoning in vacuum would actually be faster if one closed the source reservoir and let gas pressure help gravity, at the cost of some liquid.

The worst thing about science is that it takes absolutely no thinking to bring the most horrible statements out in the world, but driving them back into the cave is hard work...

#### Yes quite

Atmospheric pressure is required to hold the column(s) of liquid together and preventing it from collapsing, end of. The driving force is gravity.

#### Not a great week for Aussies

Austrlians have had there physics and cricketing abilities tested and they've been found wanting.

What next? Maybe a study confirming their beer is horsepiss?

#### Depends whether you test export or domestic

We are famous for exporting Foster's. If you test that you will undoubtedly find it is horse piss, at least on the days the emus hadn't been well. We export it just to get the damn stuff out of our country.

On the other hand, if you test the stuff Australians actually drink you'll find it's better beer than anywhere in the world except perhaps Belgium and the Czech Republic.

#### Silly question...

...but isn't atmospheric pressure due to gravity in the first place?

#### Weighty Discussion

From Robert Weaver:

'Air "weighs" about 14.7 pounds per square inch of area on which it rests, including the surface of a liquid; this pressurizes the liquid to this amount.'

From Wikipedia:

"Atmospheric pressure is the force per unit area exerted against a surface by the weight of air above that surface in the Earth's atmosphere."

So let's assume that Weaver is correct and siphoning is caused by atmospheric pressure. What causes atmospheric pressure? Weight. What causes weight? Gravity. So what force drives this process, even if Weaver is correct?

#### Why do I get downvoted...

...and you don't, when we said the same thing? Boo, I question the consistency of people who downvote!

#### A little from column A, a little from column B?

Isn't it actually a case of both, rather than one or the other?

Gravity acts on the liquid in the open end of the pipe 'pulling' it out. This creates a lower pressure in the pipe vs. atmospheric pressure acting on the surface of the liquid in the reservoir 'pushing' it in to the pipe.

Take away atmospheric pressure (seal the reservoir for example) and any liquid extracted via gravity will reduce pressure in the reservoir - and I doubt the syphon will keep going for long. Take gravity away (in space should do) and I doubt the liquid will have any inclination to go anywhere in particular.

Or how about putting a syphon in one of those pilot training centrifuges? This will create multiple lateral G so the syphon should horizontally if it's just gravity!

#### Only column A

Sealing the reservoir would stop the flow at some point, but not until the pressure in the reservoir is significantly lower than the atmospheric pressure; that is, when the difference in pressure can counter the driving force, which is gravity.

Pressure is still required to hold the liquid together, but a watertight tubing is required too. Is the watertight tubing the driving force?

The siphon would actually work perfectly horizontally in a centrifuge, and you make a very good point. I should have thought of that.

Indeed. Sucking is only one way to start a siphon, and immersing the whole pipe and putting your thumb over the end is another. A common type of lavatory cistern works by using a piston to start a siphon. The older type of high level cistern uses a heavy iron bell which is dropped over the down pipe to get the siphon going.

In any case atmospheric pressure is greater at the lower end of the pipe so it plays no part in maintaining the flow (if anything it acts against the flow).

#### Pounds? Inches? WTF?

I'm going to ignore this bloke until he learns to use SI units.

#### @/\/\j17

"operative force in a siphon"

The keyword is operative. You do not need an atmospheric/gas pressure *differential* to operate, i.e. maintain the siphon. Whereas you do REQUIRE gravity.

#### @/\/\j17

"Take away atmospheric pressure (seal the reservoir for example) and any liquid extracted via gravity will reduce pressure in the reservoir - and I doubt the syphon will keep going for long. Take gravity away (in space should do) and I doubt the liquid will have any inclination to go anywhere in particular."

That isn't a fair test as the air pressure would work against the siphoning action, the question is does it drive siphoning.

I am sure that the siphon won't work in a zero G environment though.

#### Thought Experiment 2

Imagine a coiled chain on a desk. Give the end a push and it will slip off the desk pulling the rest of the chain after it.

Imagine the same chain, this time rising over a hump on the edge of the desk (with a little help from you) before it falls to the ground. It will still pull the rest of the chain with it - because of gravity - regardless of atmospheric pressure. If the gravity is too strong, however, or the chain is too weak, it will snap at the top of the hump.

Now imagine a tube of liquid doing the same thing (perhaps contained in a hosepipe) and perhaps in a vacuum. Would it have enough tensile strength to stay in one piece, or would gravity break it apart at the top of the hump as it pulled it in either direction? That must depend on the mechanical properties of the liquid and the strength of the gravitational field. But if it didn't break apart, such a "siphon" would not require atmospheric pressure to operate. Only gravity.

#### Siphons don't work like that

You need both gravity and external pressure for a siphon. Surface tension is not strong enough to hold any substantial amount of liquid together like that. The pressure inside a siphon is maintained by the fluid being pushed from outside, not by the fluid pulling from inside.

What you are describing with your chain analogy is capillary action, which will only draw until the mass of fluid is enough for gravity to overcome the strength of the surface tension.

#### Dear me

You spent the whole article spelling it properly, then linked to "Syphon" on Wikipedia...

Wikipedia.

#### Atmospheric pressure FTW

It's atmospheric pressure that powers siphoning and obviously so.

The pressure at the two liquid surfaces are equal, but now imagine travelling "up" the long pipe - the higher you go the lower the pressure will be due to the weight of the liquid below you acting against the surface pressure. When you're level with the end of the short pipe this difference in pressure is what drives the siphoning.

#### Cobblers!

All that is required for a syphon to work is that the column of water be continuous (though a certain bubble tolerance is possible depending on pipe geometry) The pressure can vary until the liquid boils or solidifies with no appreciable effect on the syphon itself.

You *can* prime a syphon with pressure differentials, but you don't have to.

I syphon my swimming pool cover (and one year the pool itself) by tossing the hosepipe into the water to be drained, filling the hose until water flows freely *onto* the cover, then disconnecting the hose from the tap and dropping the end on the driveway.

If you connect two vessels with a pipe and change the pressure in one of them you don't have a syphon, you have a closed system in which the atmosphere is seeking to equalize pressure throughout. Any liquid in or around the end of the pipe will be forced through because it is in the way. You could force water uphill this way with little problem.

Syphons only work when draining downhill.

Because that's what water does when it is pulled by gravity.

Experiment : Set two beakers such that one is 100 cm above the other. Place a quantity of water in the upper one and arrange a traditional inverted U tube syphon. Once the pipe is primed it works as expected.

Now do the same experiment in a sealed environment at a reduced pressure, say 1/2 an atmosphere. You won't see any appreciable change in the flow rate.

The OED is still full of snot. Better burn yours now before it gets you into more trouble.

#### I thought reg readers were smarter than this!

This type of authoritative feedback is shameful...

"The atmospheric pressure does not need to be present."

"Asside from the difficulty of whatever liquid you used evaporating there would be nothing to stop you siphoning in a vacuum."

Despite the confidence of the authors, these quotes are very wrong.

Gravity accounts for downward force in both legs. It's true that the weight of the liquid in the longer leg is greater than the weight in the shorter leg. However gravity is NOT the force which pushes (pulls?) the liquid up!!!

Whether the siphon is started by sucking or any another means, gravity merely causes a vacuum to form at the top. If and only if the atmospheric pressure on the shorter leg is stronger than the force of gravity, then the liquid will be pushed up into the vacuum.

Whether the atmospheric pressure is caused by gravity or some other means is not relevant to the functioning of the siphon.

If the pressure is completely removed, then there will be a total vacuum both inside and outside the siphon. With no force to oppose gravity (ignoring capillary effects), then the force of gravity will pull the liquid down both legs with no siphon possible.

I'm baffled as to why more people didn't learn this in high school physics.

#### @I thought reg readers were smarter than this!

Utterly shocked that the correct posts are getting down voted...

The voice in my head keeps saying "if you explain it again, people will understand. Tell them to consider the sum of force vectors acting on each leg. Start with just the downwards force of gravity, and then with upwards force of atmospheric pressure. Now explain that water does not work like a chain since it has no significant tensile strength..."

But I'm going to have to be content with the fact that some of us can understand basic physics, and some of us can't.

I guess it's time for myth busters to come in and solve yet another elementary physics problem.

#### I thought reg readers were smarter than this!

> gravity merely causes a vacuum to form at the top.

No. Gravity would cause a *partial* vacuum to form. Or, more concretely, it reduces the pressure at the top of the siphon. If the pressure is low enough, the liquid will evaporate--it'll be more *like* a vacuum, since water vapour is less dense than liquid water, but still not a vacuum. Nature abhors those, apparently, and the effect is that water is pulled up from both legs to prevent this. Of course, if the water at the top *does* evaporate, then a new equilibrium is reached. This time, the vapour can expand (either by losing pressure, which a gas can do more easily than a liquid, or by causing more liquid to evaporate) much more than the pull of gravity down the long leg of the siphon can compensate for, and so no more liquid can flow past the hump.

> If the pressure is completely removed, then there will be a total vacuum both inside and outside the siphon.

No. There's no total vacuum in the tube for a start (it's why we call it a "vapo(u)r lock". And you've failed to account for the weight of water above the tube in the higher container.

> I'm baffled as to why more people didn't learn this in high school physics

It's probably something to do with the "state of education" nowadays. As it always was, probably.

#### @Frumious Bandersnatch

You're right about the partial vacuum, a "total vacuum" doesn't exist even in space. But my point was that the only reason a vacuum doesn't form is because of air pressure.

"And you've failed to account for the weight of water above the tube in the higher container."

The force caused by gravity on the water outside the tube cancels out the force on water inside the tube at the same height, therefor there is no net force difference for any of the liquid below water level.

#### Page:

This topic is closed for new posts.