How Does Siphon Work? Explanation & Principles

[jnlbctln]

How does SIPHON work? especially with the simple apparatus in transferring liquid from a container to other container. Would you please help me understanding the mechanism behind this. Then principles are used here.. Thanks :)

 

[Infinitum]

Wikipedia has a quite detailed description about its working.

http://en.wikipedia.org/wiki/Siphon

If you get confused with something in that, you can post questions about it here

 

[jnlbctln]

"In practical siphons, atmospheric pressure pushes the liquid up the tube into the region of reduced pressure at the top of the tube in the same way as a barometer, and indeed the maximum height of a siphon is the same as the height of a barometer, because they operate by the same mechanism. The reduced pressure is caused by liquid falling on the exit side."

how come?
is it necessary not to immerse the end at the lower container the liquid?
why you need the "initial pump", what's that for? what happens if we introduce that initial pump? thanks :)

 

[Infinitum]

Do you know http://en.wikipedia.org/wiki/Bernoulli's_principle]Bernoulli's[/PLAIN] Principle?

In a very simple sense, higher fluid velocity results in low pressure than slower moving fluids.

Also, read 'Theory' under siphons in wikipedia. Thats where the main -how-siphon-works- is hidden.

A siphon works because gravity pulling down on the taller column of liquid causes reduced pressure at the top of the siphon (formally, hydrostatic pressure).

As gravity pulls down the fluid, the velocity of the fluid increases, resulting in lower pressure.

 

[Whovian]

\Comes in, probably annoying other people

First, assume we have the initial "pump." Now we've got liquid flowing out of one end of the siphon. Assuming a rigid tube and liquid that remains at constant pressure, since some liquid is leaving the tube, something has to come into fill the space. In this case, the liquid coming out is sort of "sucking" in some more liquid into the tube. Sure, the rigidity of the tube and the liquid being constant pressure were annoying assumptions, but the concept still applies.

 

[vin300]

I don't think Bernoulli's principle plays a role here.
Since atmospheric pressure acts equally on both arms of the tube, it does not pump water form the higher to the lower container.It is actually the difference of pressure in the two unequal arms of the siphon that pumps the water out.The height of the water column above the surface of the liquid in the upper container is lesser than the column in the lower container.
Water is pulled down in the column on the left with lesser force than the pull in the column on the right, so the resultant force pulls water into the lower container.

 

[Infinitum]

I don't think Bernoulli's principle plays a role here.
Since atmospheric pressure acts equally on both arms of the tube, it does not pump water form the higher to the lower container.It is actually the difference of pressure in the two unequal arms of the siphon that pumps the water out.The height of the water column above the surface of the liquid in the upper container is lesser than the column in the lower container.
Water is pulled down in the column on the left with lesser force than the pull in the column on the right, so the resultant force pulls water into the lower container.

Of course, I do agree with your reason about the difference in pressures being the cause for water pumping. But as to why... doesn't Bernoulli play a role(along with the initial difference in forces and gravity) in maintaining the pressure difference due to the moving water?

 

[sophiecentaur]

Bernouli's principle doesn' t come into the basic syphon principle because a syphon works even at zero velocity. Bernouli is all about the effect of motion.

 

[Infinitum]

Bernouli's principle doesn' t come into the basic syphon principle because a syphon works even at zero velocity. Bernouli is all about the effect of motion.

I see. Just to clarify, does it effect the siphon after the flow has started?

 

[sophiecentaur]

Pressure is affected by the velocity of a fluid so the Bernouli effect will be there. But the two effects are distinct and, in Science, being able to look at separate phenomena one at a time helps with understanding situations better.

 

[OmCheeto]

Ha ha! I love this place. I didn't realize I'd missed the great siphon debate.

Yanking the chain on siphon claims
University of Hawaiʻi at Hilo
Contact:
Alyson Kakugawa-Leong, (808) 974-7642
Director, Media Relations, University Relations
Posted: Jan. 19, 2011

A recent claim by an Australian physicist that the definition of “siphon” has been incorrect in most dictionaries, sometimes for decades as in the case of the Oxford English Dictionary, was recently re-examined by a professor and a student at the University of Hawaiʻi at Hilo.

Dr. Philippe Binder and Alex Richert have concluded that the new claim, widely spread over the web, is itself incorrect.

Perhaps we need to look at this from the 'frictionless disks in a u-tube' analogy.

The above image is at T_initial
Initial air pressures are atmospheric.
What will the image look like at T_final?
And what will be the pressures between the disks?

The image is not drawn to scale. The tube above and below the disks extends to 100 meters.
Tube diameter is 1 meter.

ps. I just made this up. It is not an actual homework assignment, that I'm aware of. Recommendations for parametrization will be most humbly accepted.
pps. Must now go to the river.

 

[Infinitum]

Pressure is affected by the velocity of a fluid so the Bernouli effect will be there. But the two effects are distinct and, in Science, being able to look at separate phenomena one at a time helps with understanding situations better.

Aye, thanks.

 

[Andy Resnick]

For some reason, this topic flared up in 'Physics Today' recently:

http://www.physicstoday.org/resource/1/phtoad/v64/i4/p11_s1
http://www.physicstoday.org/resource/1/phtoad/v64/i8/p10_s3
http://www.physicstoday.org/resource/1/phtoad/v64/i11/p10_s2

Personally, I think it's refreshing that something so 'obvious' is anything but.

 

[russ_watters]

Personally, I find it pathetic that supposed experts can get such a simple concept wrong/obfuscated.

I guess someone needs to put a siphon in a vacuum chamber on youtube to demonstrate.

 

[Infinitum]

Personally, I find it pathetic that supposed experts can get such a simple concept wrong/obfuscated.

I asked the very same siphon question to my professor last year (I am still in high school) who gave me the same answer that I posted in the thread. I'm just glad I got the concept cleared up now

 

[Andy Resnick]

Personally, I find it pathetic that supposed experts can get such a simple concept wrong/obfuscated.

I guess someone needs to put a siphon in a vacuum chamber on youtube to demonstrate.

Heh- like this?

http://www.youtube.com/watch?v=8F4i9M3y0ew&noredirect=1

 

[sophiecentaur]

Excellent demo. A real Myth-buster.
It does require a special kind of liquid to work in the first place though - as they make clear. The vapour pressure of the liquid needs to be low enough to avoid the formation of cavities at the low pressure point at the top of the inverted U.
There have been claims that you can get a water siphon to work well over the '10m limit' but I find that difficult to believe / explain.

To be fair to the dictionaries, the fact that practical siphons need atmospheric pressure to prevent boiling makes the conventional description excusable.

 

[russ_watters]

Heh- like this?

http://www.youtube.com/watch?v=8F4i9M3y0ew&noredirect=1

Surface tension and gravity?

Awful. Just awful.

 

[russ_watters]

To be fair to the dictionaries, the fact that practical siphons need atmospheric pressure to prevent boiling makes the conventional description excusable.

Er, well - to prevent boiling and to push the liquid up the tube! Typical liquids do not have much in the way of "tensile strength", so they can't pull themselves up a tube. As the wiki mentions, you can easily demonstrate that tensile strength is irrelevant to a normal siphon by starting a siphon with a bubble in it.

 

[sophiecentaur]

you can easily demonstrate that tensile strength is irrelevant to a normal siphon by starting a siphon with a bubble in it.

Too right.
And, if most liquids have very low tensile strength, then the tensile strength explanation is not actually relevant to most liquids. It is an interesting slant on the whole thing and adds a healthy amount of confusion to the situation.
Could turn out to be a bit of a big-endian and little-endian clash here.

 

[OmCheeto]

Too right.
And, if most liquids have very low tensile strength, then the tensile strength explanation is not actually relevant to most liquids. It is an interesting slant on the whole thing and adds a healthy amount of confusion to the situation.
Could turn out to be a bit of a big-endian and little-endian clash here.

When I first read the "tensile strength" argument of liquids, I once again thought this was another April fools joke:(adding water to rocks makes them melt at a lower temperature.)

After all, I can reach into a sink full of water and pull the water apart very easily.

But then again, if I fill a hypodermic syringe with water, put my finger over the little end thing, and pull on that big-endian thing, I feel a tension.

So something is going on.

Anyways, my faux homework problem was based on the PDF I briefly looked at yesterday: www.phys.uhh.hawaii.edu/documents/TPT-final.pdf

 

[sophiecentaur]

But then again, if I fill a hypodermic syringe with water, put my finger over the little end thing, and pull on that big-endian thing, I feel a tension.

So something is going on.

What you are feeling when you do that is the effect of atmospheric pressure on the back of the piston (the bit that's in the air). You have to fight against that to pull the piston out. You can,in fact, 'beat' a piston with a very small diameter and pull it out against the vacuum which forms inside. (The pressure times the small area gives a small resulting force). Any tensile force that you may be experiencing is very small - the same sort of force that pulls the edges of a water surface up the side of a glass. These forces are sufficient to prove embarrassing for small insects who want to get out from under a water surface and useful for pond skaters, who spend their lives standing on the top of the water. It is only at that scale of weight / forces that the inter molecular forces in water become significant.
If you did the same syringe thing up in space, you could easily pull the plunger out and the water would start to bubble as the pressure inside became less than the vapour pressure.

 

[Infinitum]

Anyways, my faux homework problem was based on the PDF I briefly looked at yesterday: www.phys.uhh.hawaii.edu/documents/TPT-final.pdf

This was quite helpful too, thanks.

 

[A.T.]

Er, well - to prevent boiling and to push the liquid up the tube! Typical liquids do not have much in the way of "tensile strength", so they can't pull themselves up a tube. As the wiki mentions, you can easily demonstrate that tensile strength is irrelevant to a normal siphon by starting a siphon with a bubble in it.

What I don't like about the wiki is : "atmospheric pressure is the driving mechanism".

For equal pressure siphons (be it 1bar or vacuum on both ends) gravity (acting on the fluid) is the driving mechanism. Equal pressure on both sides cannot drive anything. It just creates an offset (equal on both sides) to gravity. But it is the differential weight of the columns which causes the flow, and thus "drives" it.

 

[sophiecentaur]

It all boils down to gravity in the end, of course. No g, no AP.

 

[A.T.]

No g, no AP.

That's not what I mean. I specifically stated gravity acting on the fluid drives it. The differential weight of the fluid columns "drives" the flow, not the virtually identical weight of air columns above the vessels. Therefore I don't like invoking atmospheric pressure as the "driving mechanism".

 

[OmCheeto]

That's not what I mean. I specifically stated gravity acting on the fluid drives it. The differential weight of the fluid columns "drives" the flow

Actually, the wiki article states that that isn't true. Under the Chain Analogy:

Even though the weight of the fluid in the upper siphon section is greater than that in the lower container, the fluid still flows downwards.

, not the virtually identical weight of air columns above the vessels. Therefore I don't like invoking atmospheric pressure as the "driving mechanism".

I liked this line from the wiki article:

siphons work by a gradient of hydrostatic pressure within the siphon

And I think it all depends on what kind of siphon you're talking about. The vacuum siphon video implies that the chain analogy is correct, in that instance.

But under atmospheric pressure, a broken siphon(one with an air bubble) works also.

But I think you can still use the chain analogy here also, as long as there are pistons at each end of lots of little chains.

Hence my new "Infinite layers of jello shots siphon" model.

Which I will explain in about 23 hours, as I'm late for work again.

 

[russ_watters]

What I don't like about the wiki is : "atmospheric pressure is the driving mechanism".

For equal pressure siphons (be it 1bar or vacuum on both ends) gravity (acting on the fluid) is the driving mechanism. Equal pressure on both sides cannot drive anything. It just creates an offset (equal on both sides) to gravity. But it is the differential weight of the columns which causes the flow, and thus "drives" it.

I prefer to say it is the interaction of the two: gravity pulls down on one side while atmospheric pressure pushes up on the other.

It could be said more simply that the driving force is the difference in hydrostatic pressure, but that may be what causes people to think about tension. It works mathematically but not conceptually.

 

[A.T.]

Even though the weight of the fluid in the upper siphon section is greater than that in the lower container, the fluid still flows downwards.

Okay, with different cross-section areas the offset to gravity by atmospheric pressure is not equal anymore. So it is better to say the difference in weight/area "drives" it. Which of course boils down to pressure difference.

 

[sophiecentaur]

So, we've eliminated the tension effect for all but 'special' liquids. We are now just chasing our tails about what causes what. If it were not for gravity then there would be no siphon. Atmospheric pressure (only there because of gravity) is necessary to produce the condition at the top of the U where there is fluid available to flow down the other side (this downward flow also needs gravity).

A normal siphon will not work without enough atmospheric pressure. What does it matter if we say that the pressure "drives" the siphon or not? On the downward leg, the hydrostatic pressure is Higher than that on the lower surface - so liquid can flow down.

As with all hydrostatic situations, we know that the cross sectional area of tubes does not affect pressure so that is a red (or in the case of the picture, blue) herring.

If someone looks at it one way or another (yet using the right mechanisms but in a different order [Eric Morcambe rules]) do they need to be afraid of being 'WRONG"?
Is it the term "driving" that is throwing everyone into this frenzy. That word is not a front-line, defined term in Physics so stop worrying. There is no need to bicker further if we have all learned something from this thread.
I hope that all this thread has not just managed to confuse people.

 

[vin300]

The vacuum siphon video implies that the chain analogy is correct, in that instance.

Does it? The chain analogy is all about the heavier column pulling up the lighter column. The guy in the video doesn't even bother about the liquid levels in the inverted U-tube.

 

[vin300]

Whatever be the pressure of the atmosphere, or even if there is no atmosphere, the difference of hydrostatic pressure at the points B and C at the top for a given siphoning setup will be the same.

 

[sophiecentaur]

There is no mention of the actual tensile strength of that "ionic liquid". How many metres would the tension support or was the inverted U that we saw as much as the tension would support?
The video was an interesting diversion but it says very little about real siphons with real liquids. It has been shown that a siphon works even with a bubble in it so that evidence shows nothing other than how a very novel liquid behaves. Plumbers use water and barometers use mercury; real, everyday liquids and it is those that are of real interest.

 

[nsaspook]

Found this today: The great siphon debate continues.

http://www.nature.com/srep/2014/140422/srep04741/full/srep04741.html

Over the last few years there has been controversy over how siphons work3, 4, 5, 6, 7, 8, 9. Two competing models have emerged. In one model, water flowing out of a siphon generates a low-pressure region at the crown so that atmospheric pressure pushes water into the siphon. In another, the weight of water flowing out of a siphon pulls water into the siphon via liquid cohesion.

The siphon debate has also had an impact in the field of botany in relation to how water can rise above the 10 m siphon limit in trees10. This implies some kind of continuous link between water entering the roots and transpiring through the leaves. In the field of biomedicine there is controversy over whether the siphon principle operates in the human and other circulations11.

An argument often used in support of the atmospheric model of the siphon is the fact that the maximum height of a siphon is almost the same as a barometer. The experiment described in this paper explores the boundary between the siphon and barometer.

 

[sophiecentaur]

Firemen all know of the ten metre limit to up-hill pumping with a lift pump (quite a bit less than ten metres, actually). Also, we have all seen a mercury barometer at work. So the idea of tension cannot apply under all circumstances.

I have a problem seeing any inherent difference between a syphon and a simple inverted tube. After all, if you were to introduce a vertical wall in a tube, with a gap at the top, you would have a 'syphon'. If the syphon were to be working above the limit imposed by density and then you removed the wall, would you expect the column to stay that high?

I have read odd bits about tension in liquids and there are some pretty high values quoted. This tension could, presumably, be broken by any nucleus (as with super heated water). So, given the right circumstances (narrow, smooth, clean tube), why shouldn't the ten metre limit be broken?

Perhaps there is, in fact, no conflict at all. It's just that you get different results under different conditions. (Happens quite a lot in Science)

 

[davenn]

hi Andy

thanks for the video ... that lead me to a couple of hours of video watching from the Periodic Table of Videos
collection ... very informative !

Cheers
Dave

 

[sophiecentaur]

The 'waterfall' effect must be fascinating to watch. Is there a movie available to see?

 

[Jacobim]

The 'waterfall' effect must be fascinating to watch. Is there a movie available to see?

Yes there are at the bottom of the page: http://www.nature.com/srep/2014/140422/srep04741/full/srep04741.html#supplementary-information

It just appears as the force between a the water resisting a bubble being created in the tube and the force pulling the water down balance to allow a large bubble to form.

 

[rcgldr]

Ignoring compressiblity, pressure increases linearly with depth from the upper surface of a fluid. At any fixed height above the ground, the pressure at that height in the container with the higher upper surface of fluid is greater than the pressure in the container with the lower upper surface of fluid. This means that within the siphon, there's a decreasing pressure gradient component as fluid flows from the higher upper surface container to the lower upper surface container, providing the force that drives the fluid, even if the siphon rises above the fluid upper surfaces.

The height of the intake and outlet ports of the siphon doesn't matter, as long as both ports are submerged.

For the botany case, I would assume osmosis is the main factor in drawing water upwards (or outwards) through any plant, regardless of height.

 

[NihalRi]

Hello again, so I've conducted experiments measuring the rate of flow of a siphon changing various things like the ratio of the lengths of tubing and the level of water. I found that the only times I got exactly the same rate values was when the difference in height between the lever of water and the receptor end of the tube were equal. Therefore I deduced that this height must be important. Then I graphed the energy levels using this height. Because velocity is perportional to the root of the height. It yeilded a straight line graph which proves that this is true. Therefore the only thing that drives a siphon is it's potential energy. I realize this is not a complete explanation but I think it's close to the truth. An interesting thing is that the siphon still ran when the two lengths of tubing were the same or the shorter and longer sides were reversed as long as the water level was higher than the end of the recepter tube, so this led me to dismiss the gravity pulls on the longer arm more than the shorter arm theory. Any thoughts?

 

[rcgldr]

An interesting thing is that the siphon still ran when the two lengths of tubing were the same or the shorter and longer sides were reversed as long as the water level was higher than the end of the recepter tube, so this led me to dismiss the gravity pulls on the longer arm more than the shorter arm theory.

As long as both ends of the siphon are submerged, and the siphon is filled with fluid (no air or vacuum gaps), the fluid in both containers will flow until the height (above ground) of fluid in both containers is equalized (or until the surface of the higher fluid drops below the receptor port of the siphon). The common expression for this is "fluid (or water) seeks its own level" .

 

[CWatters]

Over the last few years there has been controversy over how siphons work3, 4, 5, 6, 7, 8, 9. Two competing models have emerged. In one model, water flowing out of a siphon generates a low-pressure region at the crown so that atmospheric pressure pushes water into the siphon. In another, the weight of water flowing out of a siphon pulls water into the siphon via liquid cohesion.

Why are they seen as "competing"? Isn't it the case that the maximum height of a syphon is the combination of the two?

 

[sophiecentaur]

so this led me to dismiss the gravity pulls on the longer arm more than the shorter arm theory.

The effect of gravity on each part of the fluid is the same, near the ground. It would be a good idea to start by reading the basics about hydrostatic pressure. The pressure in each column is due to its height and the fluid density. The factors governing the flow rate are not just the pressure difference but the fluid dynamics so you would not expect that the flow rate would only be a result of the two heights involved. Using different tube thicknesses could show that the height is not the only factor.

the difference in height between the lever of water and the receptor end of the tube were equal.

I don't understand why you got any flow at all when the two levels were equal. (?)

 

[NihalRi]

The effect of gravity on each part of the fluid is the same, near the ground. It would be a good idea to start by reading the basics about hydrostatic pressure. The pressure in each column is due to its height and the fluid density. The factors governing the flow rate are not just the pressure difference but the fluid dynamics so you would not expect that the flow rate would only be a result of the two heights involved. Using different tube thicknesses could show that the height is not the only factor.

I don't understand why you got any flow at all when the two levels were equal. (?)

Sorry I meant when the difference which was different from zero was the same in different experiments. By different experimemts I mean one where I just varied the level of water and another where I shifted the tube to change the ratio of heights on the two sides.

 

[NihalRi]

As long as both ends of the siphon are submerged,

I don't think that both sides need to be submerged just the one with the water you'd like to displace.
Can someone clarify what the meaning of negative pressure is? Is it less than atmosphereic pressure?

 

[sophiecentaur]

I don't think that both sides need to be submerged just the one with the water you'd like to displace.
Can someone clarify what the meaning of negative pressure is? Is it less than atmospheric pressure?

If the lower tube is not submerged, a bubble can form and let all the liquid out of the long arm of the siphon.
All pressures are greater than zero and a 'negative' value is relative to atmospheric pressure.

 

[jbriggs444]

If the lower tube is not submerged, a bubble can form and let all the liquid out of the long arm of the siphon.

This is correct, of course. But in practice, if flow rate is sufficiently high and the tube width (and, therefore the bubble size) is sufficiently low then the bubbles do not rise faster than the outflow and the siphon continues to work in spite of this possibility.

 

[rcgldr]

I don't think that both sides need to be submerged just the one with the water you'd like to displace.

If either end of the siphon is not submerged, then air could travel up the exposed opening and stop the siphon from working. This would be an issue of the siphon diameter and the rate of flow.

 

[NihalRi]

If the lower tube is not submerged, a bubble can form and let all the liquid out of the long arm of the siphon.
All pressures are greater than zero and a 'negative' value is relative to atmospheric pressure.

Thanks:D How would the air bubble form? If we are thinking of a practical siphon air can't push up the undubmerged receptor end because atmospheric pressure acts upon both ends so it is balanced. I'm thinking of an incompressible liquid here but even if the liquid was compressible I doubt it would mattter :/

 

[sophiecentaur]

How would the air bubble form?

If the tube diameter is big enough, a horizontal surface (water on top and air beneath) can go unstable and, whatever the flow rate, bubbles could form and bubbles could go up faster than the falling water could compensate. Once the total volume of bubbles is high enough, the pressure difference would go away and the upper leg would fall back into the top reservoir. It is easy to imagine it happening with a 1m wide tube and a 1m high siphon but not happening with a 1cm wide tube. But a lower reservoir would prevent this happening.