Talk:Siphon/Archive 1

Mathematics
I originally added the math section at the end because the general reader of a general article like this isn't likely to be familiar with Bernoulli's equation. I'm merely applying the principle of progressive disclosure. Any objections to me moving the math section further down the article? Samw 03:19, 7 May 2006 (UTC)

I made an error in the discussion and erased my comment.

Siphon as technique to evacuate cellars from floodings
Dear Wikipedians!

I am new in Wikipedia and almost without experience as author.

Could You please give me Your help to establish this method to evacuate cellars from water after floodings? It bases on the siphon: [] Between the flooded cellar and a deeper place outside a connection is being built, using a tube or some pipes. They are filled with water from the flooded cellar or through an intake valve. When the ends are beeing opened, the water flows through the pipe into the sewer or the river.

I practised it many times to help neighbours and others and would now like it to establish itself, so that many people in the situation of a flooding are less dependent on the help of the fire department, their resources and the availabilty of pumps and energy.

I can´t wait to see Your comments, best wishes, Daniel Schwanck __Daniel Schwanck 14:47, 28 January 2007 (UTC)

Inverted siphons
This is regarding the image captioned "Inverted siphon under a sink".

The trap under a household sink is not an inverted siphon as defined in the article. The purpose of an inverted siphon is to carry the liquid under an obstruction. Per the article, it should be designed to maintain sufficient flow to keep solids in suspension. A trap does exactly the opposite. Can you please cite references for a trap being called an "inverted siphon"? If so, we should write that up properly in the main article. Thanks. Samw 01:28, 12 April 2007 (UTC)


 * The pictured contraption is first and foremost a means to keep sewer gas from entering the home. Second, when emptying a sink one expects that all solids in suspension, such a food particles, will freely and quickly flush down, unobstructed. That is may occasionally keep some small parts (jewelry maybe) from going into the sewer (i.e. trapping them) is not very relevant, however pleasant it may be. "Trapping" particles is not a primary function of this contraption. The name "trap" is really a misnomer considering the main functions. VanBurenen 07:53, 12 April 2007 (UTC)
 * To clarify where siphoning takes place, see picture Image:Siphon detail.png. VanBurenen 14:55, 12 April 2007 (UTC)


 * Thanks for the explanation. Agreed the trap is primarily for sewer gas.  My issue though is calling it an "inverted siphon".  Do you have references of people actually calling it an "inverted siphon" as opposed to a "trap"?   (A quick search of Google for "inverted siphon" only shows it being used for sewers avoiding an obstruction.)


 * Your new picture raises another issue. You label a portion of the pipe "siphon section".   While geometrically it is a siphon, it is my understanding that the siphon action is very weak or non-existent.  Otherwise, the trap section would be emptied completely, defeating it's purpose of blocking sewer gas.  All in all, I think it is problematic to label any portion of a sink drainage pipe a "siphon".  Samw 00:08, 13 April 2007 (UTC)

Attached file not too precise
In Applications, the linked file "http://upload.wikimedia.org/wikipedia/de/b/bc/Hydrostatischer_Heber.pdf" (ref #8) is not too precise. The end of the pipe should be below the level of the tank, not below the bottom of the tank. The dashed line may confuse someone.

--Nicolaufg 16:12, 17 April 2007 (UTC)

"His student"
"...Ctesibius was the discoverer of the principle of the siphon. His student, Hero of Alexandria..." Hero is said to be the student of Ctesibius, but from the respective articles, Hero lived (c. 10–70 AD) and Ctesibius lived (fl. 285–222 BC). Was Hero really his student? —Preceding unsigned comment added by Bnjm (talk • contribs) 23:57, 19 March 2009 (UTC)

Syphon
I wasn't sure if this should be added but thought I would mention it here. Maybe a bit on the ease with which a syphon can be made. I didn't want to include it due to low quality pic, and nosey background but if there is any interest I can retake the picture with a decent camera. It's a syphon I made from a runner bean stalk, making syphons from plant stalks is common in Kurdistan. Wolfmankurd 18:25, 12 May 2007 (UTC)

Siphon bottles?
Surely these so-called siphon bottles are simply pressurized bottles. According to the article, "It is a siphon in the sense that pressure drives the liquid through a tube." So if I suck liquid from a bottle through a straw (OK let's say the straw is an integral part of the bottle) that would make it a siphon? Of course not. I know these are popularly called "soda siphons" but surely that is a misnomer? A fountain (in fact any plumbing installation) also has pressure driving liquid through a tube, but that does not make it a siphon.--Shantavira|feed me 11:29, 20 May 2007 (UTC)
 * Feel free to clarify. The term "siphon bottle" appears to be a well-known historical term so I would object to the term being completely removed from the article.  But by all means clarify that it is a misnomer. Samw 12:54, 20 May 2007 (UTC)


 * Not sure my O-level science is up to it, but I will consult elsewhere.--Shantavira|feed me 13:05, 20 May 2007 (UTC)

World Of Warcraft reference
Is it entirely necessary? A lot of the Wikipedia population likely do not know even what WoW is and I suspect a proportionally negligible amount of readers/editors will actually want to know about the clan. Thus is fails WP:NOTE. I suggest we either create a new article & disambiguation or remove it entirely; I vote the latter. --poorsodtalk 21:15, 13 October 2007 (UTC)
 * It looks like someone's already removed it. We can always have a dismabig page if someone wants to write a separate article on the WoW meaning. Samw 03:22, 14 October 2007 (UTC)

Gas collection at crest not desirable?
Really well written article.

The Applications section has the following

"The flow of the liquid moves bubbles thus the intake leg can have a shallow slope as the flow will push the gas bubbles to the crest. Conversely, the outlet leg needs to have a steep slope to allow the bubbles to move against the liquid flow. At the crest the gas can be trapped in a chamber above the crest. The chamber needs to be occasionally primed again with liquid to remove the gas."

It is good to mention how the gas can be collected and removed.

But I feel that most syphons with one crest are arranged to have the bubbles moved along and out.

In which case the outlet leg must have a shallow slope.

I'm not confident to write this elegantly, so would appreciate someone else putting it in if they agree.Polypipe Wrangler 03:43, 4 November 2007 (UTC)


 * I've added. Be bold next time and add it yourself!   BTW, do you have a reference for this alternate design? Samw 12:07, 4 November 2007 (UTC)

Siphons in Nature
I removed the section on "naturally occurring siphons" which described the ability of water to travel over obstructions for short distances, Hydraulic jump. By definition, a siphon includes a tube.

I have also renamed the section on biological siphons to siphons in nature. Greenknight21 01:27, 6 November 2007 (UTC)

Direction of siphon flow in tank to be siphoned
I haven't seen this addressed in the article, or elsewhere, so hopefully someone here can help me out. When siphoning from a tank does the siphon draw the liquid down or up? To put it a different way, say you're siphoning a volume of oil floating on top of water. You want to siphon off all the oil while avoiding as much of the water as possible. Do you place the (intake) end of the siphon tube below the oil layer (for it to siphon some water & the oil from above down into the siphon tube), place near the top of the oil layer (for it to siphon the oil & water up from below - until it hits air, then having to frequently move the siphon) or place it at the bottom of the oil layer but above the water (for it to siphon the oil below it up, lowering the oil level until its all up)? Thanks AllanHainey (talk) 12:22, 4 March 2008 (UTC)


 * I'm not sure I really understand the distinction. The siphon draws whatever liquid is at the intake.  So if you place it horizontally at the bottom of the oil layer, theoretically it will siphon off all the oil until air hits the intake and no movement of the siphon intake is required. Samw (talk) 01:25, 5 March 2008 (UTC)


 * The distinction is where it isn't possible to place (or retain in position) the siphon in a horizontal position. Where it has to be verticle I think the direction of flow would be relevant, though I may be wrong. AllanHainey (talk) 08:52, 6 March 2008 (UTC)


 * OK so if I understand correctly, you're proposing 2 positioning: 1) opening of the tube pointed down; or 2) opening of the tube pointed up (& snaking down and up again out of the container).   Siphon flow rate should be the same in both cases and the orientation doesn't matter.   That's a key point of the equations:  only the height difference matters.   If someone wants to draw a diagram, it would be a good addition to the article; not sure I want to explain this in words!  Samw (talk) 01:03, 7 March 2008 (UTC)
 * Yeah, it's a little hard to get the head around but I undestand now. Thanks AllanHainey (talk) 13:09, 7 March 2008 (UTC)

"sadly this is not true"
Fix that! looks tacky for an encyclopedia! —Preceding unsigned comment added by 208.126.51.37 (talk) 21:17, 25 March 2008 (UTC)

Vacuum
This sentence has been in the article for many years:
 * Some (notably Encyclopedia Britannica) argue that theoretically, "a siphon will work in a vacuum".

I agree Britannica is probably wrong. But given that Britannica is so prominent, I believe it is notable enough to mention. Samw (talk) 22:00, 6 May 2008 (UTC)


 * I think that the Britannica probably isn't wrong. Liquids have some obscure properties such as tensile strength (that's why trees can be more than 32 feet tall- they actually do suck up the water further than 1 bar of pressure difference- counterintuitively to me when I first learnt of this). Water can take negative pressure without necessarily cavitating). This tensile strength would be present in a vacuum, and in ideal conditions, siphoning should work.- (User) WolfKeeper (Talk) 22:08, 6 May 2008 (UTC)


 * OK point taken. So I think the old sentence is a succinct summary. Any objections if I put it back in? If there are references for the other concepts (trees, etc), by all means we should add that as well.  Samw (talk) 03:10, 7 May 2008 (UTC)


 * I don't personally consider EB to be a reliable source (I don't consider the wikipedia to be either), so no, not without another source.- (User) WolfKeeper (Talk) 04:16, 7 May 2008 (UTC)


 * I still think it could be done though; particularly if the height of the reservoir you're siphoning from is around 32 feet (to avoid bubbles entering the pipe from the boiling liquid), and even if the siphon is a bit taller than that.- (User) WolfKeeper (Talk) 04:16, 7 May 2008 (UTC)

I oppose the restoration of the Britannica quote. If you scroll down to the last equation at the bottom article, you will see that in theory the maximum height of a siphon is proportional to the ambient pressure. So in a vacuum, the maximum height would be zero, and that's not much of a siphon. Surface tension may draw water up a narrow tube in a vacuum, as in transpirational pull, but then it won't release it at the other end as a siphon would. I can't verify what Britannica said since the referenced link doesn't work for me, but it had to be either wrong or misinterpreted. There are lots of mistakes in Britannica, as in many other sources, and I don't think it serves any purpose to copy them. Now that Wikipedia is so prominent, should Britannica articles start discussing our mistakes?--Yannick (talk) 06:02, 10 May 2008 (UTC)


 * Do you have a reference to the EB being wrong, or is this your OR?- (User) WolfKeeper (Talk) 06:43, 10 May 2008 (UTC)


 * Do you have a reference to the EB even saying that in the first place? A broken link is not verifiable.--Yannick (talk) 05:17, 11 May 2008 (UTC)


 * Yes, a few moments of googling gave: . FWIW I would recommend the use of search engines before asking such trivial questions in wiki talk pages. Incidentally you can see the beginning of the quoted phrase at: - (User) WolfKeeper (Talk) 06:28, 11 May 2008 (UTC)

So courtesy of other Wikipedians, I have: http://www.iop.org/EJ/abstract/0143-0807/29/3/003

It starts: "But the embarrassing matter of this humble instrument is the question: ‘does this device work in vacuum?’. In some references, the answer is ‘yes’ and in others the answer is ‘no’."

It cautiously supports that a siphon does work in a vacuum. Can we now agree to but back the original statement with this reference? Samw (talk) 03:42, 12 May 2008 (UTC)


 * That paper confirms that a siphon does not work in a vacuum. It cites multiple sources that have shown experimentally that siphons do not work in a vacuum. It shows that Pohl's cohesion theory agrees that a siphon will not work in a vacuum when compressibility is considered. And although it doesn't point it out explicitly, Potter & Barnes hydrodynamic theory would predict impossible negative pressures at the crest if the apparatus was in a vacuum. So under all theories and experiments discussed, siphons do not work in a vacuum. Thank you for finding the perfect citation to support my point.--Yannick (talk) 02:29, 14 May 2008 (UTC)


 * Ummmmmmm:


 * Another paper [2] explicitly claims that a siphon does work in vacuum. In a very nice experiment, the author uses two glass reservoirs connected by a sealed glass tube having a side connection with a stopcock (for air evacuation); he uses mercury and water, boiled long enough to exclude air bubbles.


 * [2] Nokes M C 1948 Vacuum siphons Sch. Sci. Rev. 29 233
 * Nokes M C 1948 Vacuum siphons Am. J. Phys. 16 254 (Reviewed)


 * So Nokes actually did the experiment successfully, in a vacuum, and published, again, successfully, in a refereed journal.- (User) WolfKeeper (Talk) 22:48, 14 May 2008 (UTC)


 * That experiment was not performed in a vacuum. The siphon was evacuated in order to prime it, as they always are, but there is no indication that the reservoirs were exposed to vacuum.--Yannick (talk) 15:21, 17 May 2008 (UTC)


 * That's not the question, the question is whether the top of the siphon tube is negative pressure or not.- (User) WolfKeeper (Talk) 19:28, 20 May 2008 (UTC)


 * No, the question is whether or not a siphon will work in vacuum. I'm saying it will not, because you need atmospheric pressure to act on the reservoir. The Nokes experiment was not performed in a vacuum and did not attain negative pressure at the top of the siphon tube.--Yannick (talk) 14:43, 24 May 2008 (UTC)


 * I'm saying that that's your OR, that the inlet to the siphon itself is never vacuum pressure anyway due to hydrostatic head, and that negative pressures in liquids are pretty well known to be possible anyway.- (User) WolfKeeper (Talk) 13:12, 7 June 2008 (UTC)


 * But in any case the question is whether the references support the contention. They do. End of. If you have references that attempt to show that this does not work, then we can add them as well. Removing references and material where there is contention is NOT NPOV and isn't the wikipedian way; you ADD material.- (User) WolfKeeper (Talk) 13:07, 7 June 2008 (UTC)

I'm not sure the paper clearly concludes that siphons do not work in a vacuum. I also like pointing out Britannica's flaws. So how about this statement in the main article:
 * Some (notably Encyclopedia Britannica) argue that theoretically, "a siphon will work in a vacuum". The latest research disagrees.

Samw (talk) 03:31, 14 May 2008 (UTC)


 * I don't have access to that, and given that trees have negative pressure at the leaves, I'm not that impressed. And, read this, before claiming 'impossible negative pressures'. Pressures below -9 bar have been done.- (User) WolfKeeper (Talk) 03:33, 14 May 2008 (UTC)


 * If anyone needs a copy of, email me at samwsamw at gmail.com Samw (talk) 00:47, 15 May 2008 (UTC)


 * If you want to talk about capillary siphons, such as a towel in a sink, then yes you can have negative pressures. (BTW, I find the following web page more helpful than the AIP one: .) A capillary siphon still would not work in a vacuum though, because the same tensile force that pulls the liquid out of the reservoir will also prevent its release at the outlet. In any case, that is not what most people mean when they talk about a siphon. They mean a u-shaped hose or pipe, where negative absolute pressures are impossible. The capillary siphon should only be presented as a special case.--Yannick (talk) 15:21, 17 May 2008 (UTC)


 * But negative pressures are not impossible. People hang heavy weights on water filled pistons, and the water doesn't cavitate. Water can take MINUS 9 bar or lower. The water molecules are attracted to each other, that's why it's a liquid, not a gas. Only if you have nucleation sites and dissolved gases does it do that. And any bubbles take time to grow. These factors mean that the siphon can work even at very low pressures where the u-bend is at negative pressure.- (User) WolfKeeper (Talk) 20:11, 17 May 2008 (UTC)


 * Water can take -1400 bar or lower according to your own reference (backed up by this cite,) but not in a normal real-life siphon. Your AIP source says water can take minus 9 bar in a xylem, and I can believe that because of surface tension. But if you hang a heavy weight on a water filled piston, the water will certainly form bubbles. Again, your own reference says so in the second paragraph. You can call it cavitation, but most people call it boiling. That's what happens at the top of a barometer. These bubbles will form when the absolute pressure of the liquid falls below its vapor pressure, which is about +0.03 bar for water at room temperature. These bubbles would form an embolism in a siphon, even if it could be primed to begin with. Siphons cannot work in a vacuum.--Yannick (talk) 14:36, 24 May 2008 (UTC)


 * What you're forgetting, but what the source points out, is that metastability is very common with water, for example, I routinely superheat water in a microwave oven by degassing it by boiling it, letting it cool and repeating the process. But stick in some granules (with a wooden spoon), and it instantly boils. And that's in a not especially clean mug. The source says that a water filled piston with sufficiently clean walls, degassed and with no nucleation sites is in a metastable condition, and will not break down to -9 bar. That is easily low enough for it to siphon in a vacuum.- (User) WolfKeeper (Talk) 21:55, 29 May 2008 (UTC)


 * The perfectly clean piston is imaginary. The microwaved water has nothing to do with negative pressure. And neither cases are siphons. I do not wish to argue the plausibility of negative pressure with you - I've already conceded that they may exist in special situations due to surface tension/capillary action. The issue is that siphons do not work in a vacuum. Can you come up with a common example of a vacuum siphon?--Yannick (talk) 02:44, 30 May 2008 (UTC)


 * I don't have to. Given the references, you have to come up with references to an opposing POV and add them to the article.- (User) WolfKeeper (Talk) 13:14, 7 June 2008 (UTC)


 * The mug example is really simple and easily demonstrated example where the saturated vapour pressure of water is higher than ambient- but it does not boil until you dump in nucleation centres. Experiments have demonstrated this down to -9 bar, in real world experiments. I'm sorry if you don't understand or believe this, that is not the wikipedia's problem.- (User) WolfKeeper (Talk) 13:19, 7 June 2008 (UTC)

I have deleted the following "Vacuum siphons" section: However, the above assumes that a liquid cannot take a negative pressure which normal liquids with dissolved gases are unable to do. In practice, liquids such as water, mercury exhibit a property known as tensile strength and are able, under certain conditions to take negative pressures. One example is in tall trees, where the water is pulled up from the roots further than 9.81 meters.

Strangely to many, experiments have indeed shown that siphons can operate in a vacuum, provided that the liquids and surfaces are very clean and pure, and are degassed.(Historical and pedagogical aspects of a humble instrument S Ganci et al 2008 Eur. J. Phys. 29 421-430)(Nokes M C 1948 Vacuum siphons Am. J. Phys. 16 254) Ganci et al 2008 does not support the claim, and says about Nokes: "Nokes seems to report an accurate experimental work demonstrating that not siphoning in vacuum is due to a breach in continuity caused by expansion of an air bubble, but this reading key is open to readers' criticism." The Nokes citation is actually just a short digest of Nokes, M C, 1948, Vacuum Siphons School Science Review 29 p. 233, which is a journal for school teachers.--Yannick (talk) 12:50, 7 June 2008 (UTC)


 * Uh huh. I've reverted your edit. The contention is supported by the references if you bother to read them carefully.- (User) WolfKeeper (Talk) 13:07, 7 June 2008 (UTC)


 * FWIW both cited journals are refereed journals for university professors as well as other professional physics teachers. How on Earth is that supposed to not be notable?- (User) WolfKeeper (Talk) 13:53, 7 June 2008 (UTC)


 * I have read Ganci's paper carefully. Nokes's 1948 paper is a little hard to get a hold of, although I'm willing to read it if you can provide a copy. Ganci provides 4 references saying that siphons do not work in a vacuum. He then reconciles Pohl's cohesion theory by showing how it would also stop in a vacuum.--Yannick (talk) 13:57, 7 June 2008 (UTC)


 * I have Ganci's paper, and that does not appear to me to be what is stated. Please quote the piece that you think says that, and I will you show that that is not an accurate precis.- (User) WolfKeeper (Talk) 15:20, 7 June 2008 (UTC)

Let's try to look for common ground. Do you at least agree that a siphon would only work in a vacuum if special precautions are taken to remove and prevent nucleation sites?--Yannick (talk) 16:39, 7 June 2008 (UTC)


 * See Phase_diagram and Vapour_pressure: in vacuum, i.e. at zero pressure, the water will go into gas state. So siphons will not work in vacuum, since vacuum is zero pressure by definition. Normally, siphons operate at atmospheric pressure. Normal tap water cannot siphon anymore below about 0.3 times the atmospheric pressure, due to the dissolved other gases. So in practice you cannot pump water higher than about 7 metre by suction pumping or by using a siphon. Crowsnest (talk) 11:50, 8 June 2008 (UTC)


 * Unfortunately, due to the existence of superheated water that does not boil, that phase diagram is not sufficient to deal with this.- (User) WolfKeeper (Talk) 14:59, 8 June 2008 (UTC)


 * Hi folks. Interesting debate.  I have learnt quite a bit.  I would summarise it like this:  Under steadily decreasing pressure, most liquids boil long before a pressure of zero is reached.  Negative pressure in liquids is not something normally seen in practical situations.  Therefore, with a siphon in conditions of steadily decreasing pressure the liquid eventually boils and the siphon ceases to operate before zero pressure is achieved.  We say a siphon won't work in a vacuum.  However, in certain circumstances, a liquid can remain in the liquid state when the pressure reaches zero, and even beyond when the pressure becomes negative (and there are sources that can be cited to support such a statement).  Therefore, theoretically if these certain circumstances could be replicated in a siphon, the siphon would work in a vacuum.  We say a siphon will work in a vacuum (but don't try it at home using pipes you bought from the hardware store.  It won't work).  If there is a source that can be cited to support a statement that a siphon has been shown to work in a vacuum, then the statement can be made in Wikipedia and the source must be cited.  However, if a real siphon has never actually been demonstrated to work in very low pressure approaching vacuum, the statement would be more misleading than revealing.  If the sources simply say that liquids can remain liquid under negative pressure that is not the same as saying the humble siphon will work in a vacuum, and the statement should not be made in Wikipedia.  Dolphin51 (talk) 13:31, 8 June 2008 (UTC)


 * The problem is that a liquid at slightly above its boiling temperature is in a state called 'metastability'- it requires energy to push it over into actually boiling, or the presence of 'nucleation centers'. It takes energy to form bubbles. But like the superheated water in a mug that has been microwaved two or more times, these are not always present.- (User) WolfKeeper (Talk) 14:53, 8 June 2008 (UTC)


 * And a liquid, even at zero applied pressure ("a vacuum")actually sticks together, you know, like surface tension, so provided the liquid is cool enough, the partial pressure of the gas isn't enough, on its own, to actually make it boil. It will evaporate, but not boil. Under these conditions, in the absence of bubbles, it will syphon.- (User) WolfKeeper (Talk) 14:53, 8 June 2008 (UTC)


 * Yes, that seems very plausible. Also, searching for "negative pressure" and "water" on Google Scholar gives tens of thousands of hits. But as stated by you and by Dolphin51, the kitchen sink siphon with tap water will not work in vacuum. On the other hand, there seem to be many instances where real negative pressures (may) occur (and siphons will work). A point of care is the definition of "zero pressure" which is used, in order to define what negative pressure is. In many engineering applications it is custom to take "standard" atmospheric pressure as zero, while here it is vacuum. -- Crowsnest (talk) 15:47, 8 June 2008 (UTC)

Old comments
This description of how siphons work is not correct. Gravity pulls the fluid away from the top of the siphon, which would not force the fluid from the upper reservoir over the top of the siphon. It is the atmospheric pressure on the upper reservoir that is greater than the pressure at the top less the force from the weight of the column. It won't work in a vacuum - a gap opens at the top. Same if the column is >~40mm in the case of mercury fluid. Fundamentally, removing a force does not cause acceleration, it is the residual forces that cause the movement. In this case, force is removed because the taller column weighs more. The article also uses a train analogy which is poor because the cohesive forces of the fluid are negligible. The analogy does hint at the ambient pressure's role, but does so misleadingly. Hiqnative (talk) 06:18, 12 May 2010 (UTC)


 * The above explanation is not correct, sorry. See http://iopscience.iop.org/0031-9120/6/5/005/pdf/pev6i5p362.pdf for an explanation and references to operation of siphons in vacuum. —Preceding unsigned comment added by 131.111.85.79 (talk) 13:17, 24 August 2010 (UTC)

- How is a syphon depicted graphically in a drawing representing an hydraulic network? 88.72.2.55 21:03, 14 September 2007 (UTC)

-

Not all liquids boil under vacuum. That should be changed. It's true that water will. Ashi_Starshade

-

Can siphon also be used for U-bends that go below the level of the reservoir? Syd1435 01:27, 2004 Nov 25 (UTC)
 * If I understand you correctly, that's an inverted siphon described in the article and it's not a siphon at all.Samw 04:33, 25 Nov 2004 (UTC)

- The article says:
 * The siphon was first used as a weapon by the Byzantine Navy, and the most common method of deployment was to emit Greek fire, a formula of burning oil, through a large bronze tube onto enemy ships. Usually the mixture would be stored in heated, pressurized barrels and projected through the tube by some sort of pump while the operators were sheltered behind large iron shields.

If this uses "pressurized" barrels and "some sort of pump", is this really a siphon?
 * See siphon bottle in the article. I didn't write this so I can't vouch for it's historical accuracy.  But physically, it would be a siphon.  The extra pressure in the barrel would make it a siphon bottle.Samw 04:09, 2 Dec 2004 (UTC)

The writer of this article has referred to vacuum in two instances. The first claims that flow of liquid creates the vacuum. later on gravity tends to draw the liquid down in both directions creating a vacuum. I would suggest that "flow of liquid" eliminates all the air and so the liquid flows. When the upper end of the pipe is exposed air immediately invades and breaks the siphon. Thus I believe that references to vacuum should be deleted. Jack hill 7 Jan 05.
 * You are correct, there is never a vacuum formed and there is never any air within the siphon or it will not work. I see how the current description can be misleading.  The reference to a vacuum is simply explain how the siphon works.  If you have a better description, feel free to rewrite that section.  I tried without success.  Samw 04:14, 8 Jan 2005 (UTC)
 * Actually there may be a partial vacuum formed and there can be some air in the top of the siphon and it can still work. Mindbuilder (talk) 23:35, 21 February 2010 (UTC)

A lot of biology articles refer here, but the biological use of the word is really quite distinct. Unless anyone objects, I propose to take out the biological section (except for the discussion of the siphon effect in biology) and create a new entry for the siphon in invertebrate anatomy. Myopic Bookworm 17:23, 22 February 2006 (UTC)


 * I had a disambiguation at the top of the article just for that reason until it was removed on Nov 28, 2005. I would support you adding it back in. Samw 00:45, 23 February 2006 (UTC)

More on siphons in vacua
I accept Wolfkeeper's point that liquids don't always boil as the pressure is progressively lowered. Therefore I have no objection to this fact being posted in appropriate articles on Wikipedia. For example, there are probably articles on liquids, and boiling point and nucleation sites and so on. The question in my mind is "has a siphon ever been actually demonstrated to work at such a low pressure that it is reasonable to describe it as a vacuum?"
 * If the answer is "yes" and there exists a citation to comprehensively support that statement then it should go ahead in Wikipedia.
 * If the answer is "well no, not actually been demonstrated as such, but theoretically it is definitely possible" then the statement should not go ahead in Wikipedia. In the latter case I can understand Wolfkeeper and many others objecting to a statement in Siphon that liquids cannot exist in a vacuum.  In the latter case there needs to be a debate on this Talk page about how to describe the principle of the siphon, and the lower limits to its effectiveness, without resorting to a statement that liquids cannot exist in a vacuum.  It may be sufficient to say liquids cannot normally remain liquid in a vacuum except in special circumstances such as trees.  Dolphin51 (talk) 00:24, 9 June 2008 (UTC)


 * The only source proposed to support the statement is Nokes, M C, 1948, Vacuum Siphons School Science Review 29 p. 233, but Ganci et al noted that this source was open to reader's criticism. I have not read it myself. I would argue that even if Nokes's vacuum siphon has demonstrated the point, it is still a unique case which should not be given undue weight.--Yannick (talk) 02:55, 10 June 2008 (UTC)


 * I certainly agree that it should not be given undue weight, OTOH it does seem to be notable in this context, one way or another.- (User) WolfKeeper (Talk) 03:04, 10 June 2008 (UTC)


 * There's also the point you I think made earlier, in the case of capillaries, even at very low/no external pressure, the surface tension provides considerable pressure due to the small radius of curvature. In that case, siphons very clearly will operate.- (User) WolfKeeper (Talk) 03:04, 10 June 2008 (UTC)


 * In the cas of capillaries, I can see how the surface tension could pull a liquid up and over the crest of a siphon. But wouldn't that surface tension then also prevent the liquid from being released at the outlet?--Yannick (talk) 03:23, 10 June 2008 (UTC)


 * It's not that the capillary forces the liquid over the crest (although depending on how the liquid wets the surface that may happen), it's more that the pressure in the liquid is way above vacuum due to the compression effect of the surface tension at each end.- (User) WolfKeeper (Talk) 12:27, 10 June 2008 (UTC)


 * Not if it outlets below surface in a container with the same liquid. Or if the outlet is gradually widened to let gravity or gravity-induced instabilities "win" from surface tension. Crowsnest (talk) 11:38, 10 June 2008 (UTC)


 * If the head at the exit is greater than the pressure due to the surface tension then it should be fine. You'd expect viscosity to be strong, but that just slows it, you should get a drop that appears, grows and then falls. I guess in a vacuum it may tend to freeze or boil at the exit and that would complicate the behaviour at the exit, but the siphoning itself should be working.- User) WolfKeeper (Talk) 11:59, 10 June 2008 (UTC)

I think an important issue with siphons in vacuum is that the liquid would have to be attracted to the material of the siphon tube in order to work. So while water in a glass tube might work, I think mercury in a glass tube or water in a Teflon tube certainly wouldn't siphon in a vacuum. The liquid in each leg of the siphon would simply fall down into its respective reservoir and the liquid over the top of the siphon would simply get pulled thinner and thinner until it separated. This would be just like the behavior of liquid poured unconfined on the top of a mountain ridge. Half drains down one side and the other half goes down the other side. The tensile strength of the liquid doesn't come into effect if the liquid can just stretch out thin until it breaks. With water in a glass tube, the water can't thin out in the middle without pulling away from the glass wall of the tubing starting at at least one point, and the attraction of the water to the glass resists the detachment. Mindbuilder (talk) 22:13, 7 December 2009 (UTC)

Found a reference to another article on vacuum siphons, this one from a University of California professor. I haven't got it yet so I have no idea what it says.: MINOR, RALPH SMITH, Associate Professor of Physics. " Would a Siphon flow in a Vacuum! Experimental Answers." School Science and Mathematics, vol. 14, no. 2, p. 152; Feb., 1914. Mindbuilder (talk) 08:22, 23 June 2010 (UTC)

Theory of operation
The following text was inserted in Siphon at 01:46 on 1 July 2008 by 71.246.234.8 under Theory (which is a sub-heading under Operation)


 * Note: As indicated above, the evaporative nature of H2O (and many other liquids) makes cohesion an issue of atmospheric pressure. Thinking of it as being 'pushed' by atmospheric pressure is unnecessary, and confusing, since there is more pressure on the lower end of the hose.  If the siphon 'crests' the liquid to more than one atmosphere, it will fail due to 'vapor lock', otherwise, the pressure can be ignored (unless you have to deal with significant turbulence).


 * If the weight of the fluid on one 'side' is heavier, the fluid will flow to that side (a bit like an unbalanced wheel or a train crossing a mountain ridge).


 * (There are many errors on the Straight Dope site, which I will not bother to correct.)

This text is unencyclopedic so I have removed it from Siphon and replaced it here. Dolphin51 (talk) 02:51, 1 July 2008 (UTC)

In case anyone is wondering, I have actually carried out the demonstration described in the changes I made to the theory section, where the top of the siphon starts out with only air and the longer lower leg is filled with liquid. It does work. Mindbuilder (talk) 05:44, 6 December 2009 (UTC) I also carried out a demonstration with the shorter upper leg being much thicker and heavier, and it also worked. Mindbuilder (talk) 22:12, 27 March 2010 (UTC)

The gravity description is more correct imo than using atmospheric pressure. A siphon will work in a vacuum --122.110.88.163 (talk) 06:12, 12 May 2010 (UTC)


 * Siphons may operate by two different methods or a combination of the two methods. A siphon in vacuum must rely on liquid tensile strength, but such siphons are extremely rare things, if they exist at all, and have no practical applications that I know of. The liquid in typical siphons have dissolved gases that defeat tensile strength and therefore would not work in a vacuum and they rely on atmospheric pressure in addition to gravity. If the atmospheric pressure description is not a good description, then how would you describe the example of a working siphon that starts out with an air bubble at the top that completely separates the two sides of the siphon, as shown in the fifth picture? Such a siphon certainly wouldn't work in a vacuum. Mindbuilder (talk) 06:26, 14 May 2010 (UTC)

This seems overly wordy:


 * The reduction of pressure caused by gravity pulling down the taller column of liquid is sufficient to suck liquid out of the upper reservoir and up the shorter column, much like liquid being sucked up a drinking straw. It is important to note, however, that the sucking of liquid over the top of the siphon is not like pulling the liquid. To say it is sucking is just another way of saying that the pressure is lowered at the top of the siphon and atmospheric pressure then pushes the liquid up the siphon.

In addition, suction really isn't a force. It's sort of a pseudo-force that's essentially a description of what appears to be happening, rather than what is actually happening. The author clarifies this, but I think it can probably be stated in a bit better way by just not utilizing the terms sucks or sucked. --Jwir3 (talk) 18:38, 19 May 2010 (UTC)

Need to Fix Stated Theory
Refer to this: http://pda.physorg.com/siphon-dictionary-atmosphericpressure_news192688538.html —Preceding unsigned comment added by Paulrho (talk • contribs) 11:19, 10 May 2010 (UTC)


 * I believe that it is a combination of "pressure difference" and gravity, with gravity being >99% of the factor for small distances, except for the initial "suction" to pull the water/fluid through the tube.


 * Anyway, perhaps our PHD friend will give us the correct words and correct formulae to put this to rest (after 99 years) —Preceding unsigned comment added by Paulrho (talk • contribs) 11:47, 10 May 2010 (UTC)


 * Gravity supplies the energy to make the siphon work, but it supplies the energy by creating a partial vacuum and then allowing atmospheric pressure to push the liquid up over the rise. It is this mechanism that continues to make the siphon work throughout its entire operation from start to finish. In a typical siphon, the liquid is not pulled over the rise like a chain, because under typical atmospheric pressures, the liquid in the siphon is always under pressure and the liquid molecules are actually repelling each other rather than pulling each other. As the pressure at the top is reduced, the repulsion is less, but that is still not tension. Mindbuilder (talk) 07:26, 14 May 2010 (UTC)

Thanks for pointing out a possible error. You object to the use of the term "hydrostatic pressure". However, the definition of hydrostatic pressure (see Fluid_statics) is: "Hydrostatic pressure is the pressure exerted by a fluid at equilibrium due to the force of gravity". You see that 'gravity' is in there. It is not quite clear to me then how the use of the term "hydrostatic_pressure" could be wrong.

However, some interesting questions remain. It has to do with the same name being applied, in the present article, to two different functionalities. This also happens in articles in other languages. To illustrate I use two images from commons:

Both of these are shown as fitting within the terminology of "siphon". Other language-wikis are even more obvious about that.

The siphon in SiphonPictureOne: the downward part of the tube is a bit longer than the upflow part hanging in the tank. The extra bit of fluid in that downward tube pulls with it all the liquid in the tube. I believe this is the situation described in the article above. Q1: Is this only gravity working, or hydrostatic pressure, or both? Q2.: In SiphonPictureOne it now shows the tube over its total length to have the same diameter. How does the performance change when the diameter of the tube outside the tank is different (bigger or smaller) than the part of the tube inside the tank. Is it then only gravity that makes it work or has hydrostatic pressure a role in this?

The unit in SiphonPictureTwo is also called a siphon. It is used to separate two gas (air and/or other gases) filled spaces. For example, it is used to make sure gases from a sewer do not flow back into you bathroom. However, if the pressure of the sewer gases becomes too high they will push past the siphon. Q3.: Is it here only gravity that makes it work or has hydrostatic pressure a role in this? --VanBurenen (talk) 12:45, 10 May 2010 (UTC)

I changed the opening paragraph. The changes use the definition of "siphon" at Wiktionary. They also try to use a simplified explanation of how a siphon works. Why obfuscate the opening paragraph with "hydrostatic pressure" when there is no need? Using such a technical term in the opening alienates most of the readers right away. I say get rid of it. A more technical explanation can follow in the section on "Operation". Jason Quinn (talk) 13:58, 10 May 2010 (UTC)


 * Thanks for your input. However, your version of the lede ignores my observation, see above, that the term 'siphon' covers two different products that have different functionalities. --VanBurenen (talk) 15:01, 10 May 2010 (UTC)
 * See below, I have been confused. The English word siphon means something different than siphon in other languages. So ignore my remarks, please. --VanBurenen (talk) 17:47, 10 May 2010 (UTC)

QUT physicist corrects Oxford English Dictionary
http://www.physorg.com/news192688538.html

This article draws more from Britannica, I take it? ;-)

Ah, I was 2nd with this...

--Kim Bruning (talk) 14:56, 10 May 2010 (UTC)

This scientist claims

"Another seeming ubiquitous misconception is that the maximum height of a siphon is dependent on atmospheric pressure.The maximum height of a water siphon actually depends on the tensile strength of water, i.e. the maximum weight that hydrogen bonds are able to support."

This sounds nutty along with the rest of the article and contradicts Wiki as it stands.

The article also says water acts exactly like the chain model. Wiki says it's not exact.

I also don't think water compresses (realistically) "Atmospheric pressure does have some influence on the operation of a siphon in that it compresses the water in the tube increasing the maximum operating height of a siphon."

http://eprints.qut.edu.au/31098/8/31098a.pdf 115.64.192.78 (talk) 03:47, 12 May 2010 (UTC)


 * Just note, that in other languages, e.g. in Russian, the word сифон = siphon is used for a wider class of devices, for example http://nts.sci-lib.com/pictures000062.html the one that uses gas pressure. This device has been on the market under the name сифон for at least since the appearance of the wrong entry in the dictionary. --Voldemar (talk) 10:56, 12 May 2010 (UTC)

Wrong link in "languages" section for Dutch
The artikel links to http://nl.wikipedia.org/wiki/Waterslot_(sifon) which translates back to English as http://en.wikipedia.org/wiki/Trap_%28plumbing%29

The correct translation for the English word siphon would be "hevel" in Dutch, but that article currently does not exist —Preceding unsigned comment added by 82.95.200.38 (talk) 15:42, 10 May 2010 (UTC)


 * Thanks. Now there is a dutch version. --VanBurenen (talk) 18:04, 10 May 2010 (UTC)

New version of opening paragraph... please review for errors
I have changed the opening paragraph to the above wording. Does anybody see any mistake in this? Jason Quinn (talk) 19:14, 10 May 2010 (UTC)

Here are my comments on your proposed alternative wording:
 * bent pipe or tube – not an adequate description. A tube with a bend can be made to operate a siphon but this is not essential for a siphon to occur.  At least the tube needs to be curved but it doesn't have to be bent.  Continuous tube is better.
 * used to move liquid – it is true that siphons are mostly used deliberately but there are also embarrassing occasions where a siphon develops accidentally and dumps a lot of liquid where it isn’t wanted. The expression used to move liquid doesn’t adequately cover the accidental establishment of siphons.
 * from one reservoir to another – a siphon moves liquid from one reservoir to somewhere else, but it isn’t always to a second reservoir. For a siphon to operate requires that liquid can move from the reservoir to the exit of the continuous tube where the exit is lower than the height of the liquid in the reservoir.

The essential feature of a siphon is that it must raise liquid above the surface of the liquid in the reservoir before conveying it to the exit of the tube at a height lower than the surface of the reservoir. Liquid will run freely downhill all the way from any reservoir but that isn’t a siphon. It is only a siphon if the liquid rises above the level of the reservoir somewhere along its journey. Dolphin ( t ) 03:29, 11 May 2010 (UTC)

was " is driven by gravity" the original phrase,or was it changed after today's new? —Preceding unsigned comment added by 165.20.108.30 (talk) 18:50, 11 May 2010

This seems to be a knee jerk reaction to theses articles, which if correct the whole wiki (and conventional belief) needs changing. http://www.physorg.com/news192688538.html a report on http://eprints.qut.edu.au/31098/8/31098a.pdf —Preceding unsigned comment added by 115.64.192.78 (talk) 03:36, 12 May 2010 (UTC)

Yes, I see a problem. By failing to mention atmospheric pressure in the title paragraph this is fuelling the bad science in the press. Whilst we have had some interesting discussion on whether it may or may not work on degassed liquid in vacuums, with cohesive forces, with capiliary action, in a centrifuge etc., in the in the vast majority of cases a siphon is driven by pressure (albiet caused by gravity). I propose the following:

Including the example of a chain is also a dis-service IMO. If we are in majority agreement that a siphon does not work due to cohesive forces then we should not show such an example.Mike163 (talk) 15:54, 29 May 2010 (UTC)


 * I appreciate the desire to leave out the flawed chain model, but I strongly support retaining the discussion of it for two reasons. One is that the air pressure workings of a siphon can be a little hard to understand, but I think the chain model makes it immediately obvious how something can be made to go uphill under the power of gravity's downward pull. Second, the chain model is cited so often in descriptions of the siphon that some discussion of its flaws helps to clear up debate about the workings of siphons. In fact you may notice that much of the theory section is devoted to clearing up misconceptions rather than just stating the correct theory and leaving it at that. If we don't provide counter evidence for the various models in the theory section, I'm afraid that people would too often try to "correct" it. Mindbuilder (talk) 22:08, 29 May 2010 (UTC)

In regards to the edit that adds "powered by the atmospheric pressure on the open liquid along with the fall of the liquid..." First let me say that I totally agree that atmospheric pressure pushes the liquid up in typical siphons. In fact I'm the one who made that fact prominent in the theory section. I strongly considered adding something like that to the opening paragraph. One reason I didn't is that there has been a reference given to a siphon that can operate in a vacuum. If such a siphon exists and works, then it obviously can't use air pressure. If such a statement is going to be added to the initial definition, then considering the controversy, it needs to include a statement like "TYPICAL or PRACTICAL siphons use air pressure ...". Otherwise someone will inevitably "correct" it. I left it out mainly to keep it simple, because the sentence was getting long, and because it was hard enough to write as I had it.

Another objection is that I don't think it's quite right to call it "powered" by the air pressure. For example, in a siphon bottle of the type that when you squeeze it, the liquid squirts out the tube, I don't think you would say that it is powered by the air pressure in the bottle. It's powered by your hand. Though you could say the liquid is "driven" by the air pressure. I seriously considered including "driven by air pressure" in the opening paragraph, but because of vacuum siphons, and for simplicity, and because the theory section starts out immediately explaining it, I decided to leave that out. If you really think air pressure should be included in the opening paragraph, then we could probably work that in. But as it is, it is technically incorrect, so I'm going to respectfully revert it to my previous version. Mindbuilder (talk) 19:53, 2 June 2010 (UTC)

The reason I think air pressure is required in the opening paragraph is I have seen at least one press article and heard the same on a BBC Radio Science program linking to this page as proof that air pressure is not required. They clearly didn't read the rest, but I think that is common in Wiki use. I left 'powered' in to minimise the edit, 'driven' is a much better term. I don't believe the siphon in a vacuum case is proven. That's not to say that you can't move liquid, in the same way that a tree can move sap, but the mechanism falls outside the definition of Siphon as most would have it ie: driven by differential pressure as created by gravity. Unfortunately the paper in the reference is not public domain, but if the mechanism is arcane then it does not fit within your description of "..usually understood..", nor with any practical test that a reader might undertake, as these will show that air pressure IS required. I would say if there is a section that needs to be less difinitive, it is the Vacuum section, not the title paragraph. By leaving out air pressure we are taking common, well known and understood science that matches everyday experience and making it subservient to a speculative reference and a (temporary) mistake by the OED. So please include air pressure. Mike163 (talk) 11:50, 3 June 2010 (UTC)


 * OK, I made a stab at it. I was also doubtful about vacuum siphons before and questioned the reference in a comment on Youtube. But micolich made this reply to me there:


 * "From my reading [of the summary of the Nokes article], I suspect it is actually a demonstration of a siphon operating in vacuum, although the detail is sparse as its a short review of another paper that supposedly does contain enough detail. Claim is that it works with mercury, dibutyl phthalate (low vapour﻿ pressure liquid) and de-aired water, and the pipework needs to be really clean. I suspect more detail is in the [School Science Review] article, which I'm still trying to get..."


 * I'm currently satisfied that vacuum siphons have been demonstrated under special conditions because it is plausible and there is a peer reviewed citation. I don't think differential pressure is inherent in the definition of the siphon. I think that the OED debate and the straight dope article demonstrate that many, perhaps most people who use the word siphon, only know what it does rather than how it works. Vacuum siphons pull liquid over an obstacle, through tubes, powered by gravity, and discharge at a level lower than the source reservoir. Vacuum siphons do the same thing as normal siphons but for the one little detail of whether the liquid is pushed or pulled up. Mindbuilder (talk) 18:11, 3 June 2010 (UTC)


 * Looks fine to me. Good job.  Mike163 (talk) 19:06, 6 June 2010 (UTC)

Since siphons work without gravity, siphon theory must do likewise.
A siphon will work without gravity when the liquid (and the ambient atmosphere, if any) is rotating, or under a force of acceleration.

"Rotary siphons" rely on the centrifugal force, not the gravitational force. Siphons work equally well in accelerating rockets. If they didn't, the equivalence principals would be violated.

In these two cases the impetus is inertia, not gravity.

Contrary to recent news reports, any definition or theory of the siphon invoking gravity alone is incorrect.

The net force on an object depends on the sum of mass-related forces: the gravitational force, the centripetal force, the inertial force, and the buoyancy force.

A scale reads higher if the person standing on it is jumping (accelerating) upward. The scale reads lower if the person is under water, and higher if the person is in a vacuum. The rotation of the Earth causes reduces a person weight by about 1/2 lb. Buoyancy due to the atmosphere reduces a person's weight by another 1/5 lb.

Norbeck (talk) 02:52, 12 May 2010 (UTC)


 * I agree. This nonsense about the Oxford English Dictionary being wrong is bizarre. With the common, stationary (non-rotary, non-accelerating) siphon, gravity is operating in a downward direction only. In a vacuum, the liquid in both sides of the siphon would go down leaving a new vacuum at the top, unless the liquid has tensile strength to keep it from pulling apart. What makes the liquid go up in the short side of a common siphon is the atmospheric pressure that prevents that vacuum at the top from developing. Gravity on the down side activates air pressure on the up side. The liquid moves up, in defiance of gravity, because of air pressure. Without the air pressure, the liquid won't go up. HowardMorland (talk) 12:04, 12 May 2010 (UTC)


 * The whole premise of this section is silly: For purposes of fluids running down pipes, the centrifugal force and rocket acceleration are exactly equivalent to gravity. What you're saying here is, "Siphons can work without (natural) gravity (if we substitute artificial gravity instead)." Yeah, sure... But that's pretty much just a roundabout way of saying, "Siphons can't work without some form of gravity." CRConrad (talk) 12:53, 12 May 2010 (UTC)


 * And normal siphons don't work without atmospheric pressure. Yes you can find some, but you're picking rare(silly) cases. You can also say gravity needs (Insert high level physics) to work so it's not really gravity doing it but that's just getting silly again. —Preceding unsigned comment added by 115.64.192.78 (talk) 13:10, 12 May 2010 (UTC)

Huge problem with "Operation" and "Practical requirements" sections
Hydrostatic pressure difference, which is the actual motive force in a siphon, is mentioned exactly once, in passing. In stead, almost the entire attention of these sections is lavished (=wasted) on things like atmospheric pressure and the tensile cohesiveness of a fluid -- which are actually only relevant for border conditions and considerations like getting the flow started and dealing with trapped gas bubbles (and, possibly, the maximum possible height of lift achievable). Needs a humongous rework. CRConrad (talk) 13:26, 12 May 2010 (UTC)
 * I would have though that the difference between tensile strength on common fluids(water) allows them to pull themselves up like a rope/chain, compared to atmospheric pressure pushing the liquid up was a huge difference on how a siphon works. And how liquids work as well. —Preceding unsigned comment added by 115.64.192.78 (talk) 10:30, 13 May 2010 (UTC)


 * CRConrad - I'm perplexed by your desire to have hydrostatic force receive more prominence when it makes up the lead sentence of the theory section. Mindbuilder (talk) 07:00, 14 May 2010 (UTC)

Contradictions between text and graphics
The end of the fourth paragraph of the theory section states:

This demonstration may fail if the air bubble is so long that as it travels down the lower leg of the siphon it displaces so much liquid that the column of liquid on the longer lower leg of the siphon is no longer heavier than the column of liquid being pushed up the shorter leg of the siphon.

It is juxtaposed with a graphic that states:

Even the falling lighter lower leg from C to D can cause the liquid of the heavier upper leg to flow up and over into the lower reservoir

There is also a graphic stating that the chain analogy is flawed, yet the Analogy section invokes a very similar (functionally the same, it appears) train analogy.

As I am sadly ignorant of the proper description for this article, I am unable to edit it, but such contradictions don't inspire confidence. 143.239.96.226 (talk) 19:05, 12 May 2010 (UTC)


 * You are right that my description of how the demonstration might fail, uses the not quite correct idea that the lower leg would be lighter with a large air bubble. While not the perfect explanation, I thought it would be easier to understand when stated that way, and while it would be possible to explain it using pressure variations, it wasn't obvious how to phrase the explanation, so I took the easy way out. You may notice that when I modified the original diagrams, I rearranged the tubing so that there would be a smaller air bubble at the top. As for the train analogy, I agree that it should probably be trimmed, but I didn't have time to revise the entire article. Mindbuilder (talk) 06:49, 14 May 2010 (UTC)


 * I think I grasp the principle at work. The energy gain is per unit, so the total weight of the columns doesn't matter, only the relative positions of the openings.  Would adding something regarding the weight of the volume of liquid moving in each leg (generally the same, except in the air bubble example) make sense and not violate the principle?  Sort of akin to hydraulics in my (flawed?) mental picture.  That would seem to clarify the issue, but again I claim no expertise. 143.239.96.226 (talk) 17:05, 14 May 2010 (UTC)

Siphons Suck -- literally!
As any petrol (gasoline) thief knows, to get the liquid over the "hump" of the tube you first have to suck at the other end of the tube. Once the liquid has passed the highest point in the tube, the continuous chain of cohesive bonds between the liquid molecules in the tube, and the force of gravity, do the rest by sucking up the liquid, not by atmospheric pressure 'pushing' the liquid up the tube.

You can try this experiment at home: initiate a siphon action with a tube of a suitable length, then place the index finger of your hand that is immersed (if not already) in the liquid reservoir over the submerged end of the tube; you will then observe that the rate of suction of the siphon, at the submerged end of the tube, is in direct inverse proportion to the level of the outlet end of the tube -- i.e., the lower the outlet end of the siphon is with respect to the submerged end in the reservoir, the higher the rate of suction is at that submerged end. Whereas, if it were due to atmospheric pressure, there would be no 'pushing' difference. --IVAN3MAN (talk) 13:49, 27 May 2010 (UTC)


 * Did you read down to the example of a siphon that starts itself even though the shorter upward section of the tube is empty and dry? Do you realize that even after the siphon has begun operating at full flow rate, if you allow a bubble into the siphon that the siphon will continue to operate normally as the bubble passes through, if the bubble isn't too big? The bubble can even be big enough to completely disconnect the the two columns of water, thus eliminating any liquid cohesive effects. And even if the bubble was small, it would completely defeat any cohesive effects, because without atmospheric pressure the bubble would allow the liquid to stretch out arbitrarily thin and break the siphon. A siphon that depends on liquid cohesive effects depends on there being absolutely no bubbles of any size, or all tensile strength will be lost. In addition, for liquid cohesion to be effective, the liquid must adhere to the walls of the tube like water does to glass. If you are siphoning water in a Teflon or polyethlene tube, then the water can just detach from the wall and stretch out thin and break, unless atmospheric pressure is strong enough to hold the liquid together. Without adhesion to the tube walls or atmospheric pressure, the top of a siphon is like a mountain ridge where water poured on the top of the ridge will half run down each side and separate, because water doesn't even have the cohesion to hold itself together on top of the ridge without a tube to adhere to. On top of all of that, since typical siphons are operating with positive pressures at all points within and around the siphon, all of the fluid molecules, both gas and liquid are being squeezed together and are therefore repelling each other rather than pulling on each other. Thus liquid cohesion is not pulling the liquid up at all. At most the repulsion between the molecules is lessened in the low pressure areas, but it never becomes a pull. It's just a harder push from one side than the other.


 * I don't see how it follows from your proposed demonstration that atmospheric pressure isn't pushing the liquid up. By lowering the outlet you create a taller column above the outlet which creates lower pressure at the top of the siphon and inlet, and therefore more "suction" at the inlet. I see no need for liquid cohesion to explain increased suction from lowering the outlet. And again, since the typical siphon is at positive pressures throughout, any tension like effects are actually only a reduction in repulsion anyway. Cohesion just doesn't have a chance to come into play. At least not until the siphon approaches or exceeds the maximum barometric height of the liquid column.


 * I should make clear that I'm not claiming that the difference in external atmospheric pressure at the top and bottom of the siphon has any significant effect. That difference is typically too tiny to be noticed. It's the difference in pressure between the atmosphere at the surface of the upper reservoir and the pressure *inside* the top of the siphon that is primarily important in raising the liquid.


 * So why did you revert my changes yesterday, since if anything I would think my changes would have been slightly more to your liking rather than less? I hope you don't object to me making changes to the article on the grounds that the previous author's work should be respected, since the first five paragraphs of the theory section were written entirely by me anyway. I'll restore my changes this evening if you make no further objections. Mindbuilder (talk) 00:48, 28 May 2010 (UTC)


 * A number of Users are working on the Theory section at present. A lot of what is written there looks like original research.  The Theory section is presently lacking in-line citations.  If the text conformed more closely to the ideas and language of high-quality sources the Theory section would be much improved.  I have added the "Section needs in-line citations" banner.  In the absence of suitable citations, a lot of what is in the Theory section should be removed.


 * IVAN3MAN's explanation above uses the concept of the continuous chain of cohesive bonds .  I disagree that this is relevant because the liquid in the siphon is under pressure (a compressive stress) at all points.  The reason fluid moves between two points in a vertical tube is because between those two points there is a difference in pressure and the resulting force on the fluid is different to the weight of the fluid.  To say the cohesive bonds and the force of gravity suck the liquid up the tube is not a scientific idea, or a scientific expression, and I doubt any high-quality source can be found to support it.


 * IVAN3MAN’s explanation above relies on some imprecise concepts – eg suck, suction  and rate of suction .  Wikipedia has an article called Suction but it is also completely lacking in-line citations to identify suitable sources for its information.


 * I would like to see all participants focus a bit more on the use of precise concepts, precise language, and in-line citations to support the text. Dolphin  ( t ) 02:25, 28 May 2010 (UTC)


 * Mindbuilder, thank you for your detailed clarification; I was just playing the role of the Devil's Advocate here. So I have no further objections for now, and I trust that there are no hard feelings, eh? --IVAN3MAN (talk) 14:07, 28 May 2010 (UTC)


 * If you honestly needed or wanted more clarification, then you should have just asked. If you were just trolling me, then I'm not too happy about having wasted time composing a response. Mindbuilder (talk) 22:08, 29 May 2010 (UTC)


 * You may feel that way, but when someone else questions the operation of a siphon, you can point them to that same response you gave me; therefore, saving yourself time in having to repeat yourself. ;-) --IVAN3MAN (talk) 01:40, 30 May 2010 (UTC)

Two weeks have passed since I notified interested Users of the need to post in-line citations of suitable sources, and to remove original research. Mindbuilder has found a couple of good sources and added them.

I have now amended the Theory and Analogy sections to remove the original research and to make the language a bit more consistent with the language of fluid mechanics. I have no objection to others restoring some of these ideas, providing suitably scientific language is used, and in-line citations are used to identify sources. Dolphin ( t ) 06:01, 11 June 2010 (UTC)


 * You were certainly justified in wanting improvements to the theory section, but you threw out a lot of important, non-controversial explanation about siphon workings. If there are specific parts that you think are controversial or demand citation, point them out and lets discuss and adjust each part. I started the Theory section out with a quick, basic, correct, explanation of how a TYPICAL siphon works. After that description of how and why siphons work, I moved on to a simpler, more intuitive model of the siphon - the chain model. I think giving the chain model prominence is important because with it, it is very easy to understand why the siphon doesn't violate conservation of energy, and because it leads into common misconceptions of how siphons work. This section also has the rare and valuable observation and explanation that a fat up-pipe siphon does work, contrary to what one might expect from the chain model. I don't have a source for that section, but anyone can carry that out at home, and I did carry it out to verify. I then moved on to explain that TYPICAL siphons don't rely on or even use tensile strength and gave the simple at home proofs and the explanation that the fluids in typical siphons are under pressure, not tension. This section is very important because there has been debate about this issue among those who don't fully understand siphon operation. I've cited the video of a phd physics professor in addition to the simple at home proofs. If you dispute the correctness of the section lets discuss the details before changing it. You definitely improved the appearance of the layout by moving the diagrams in the Theory section, but I think they're better the way I had them because the chain model is the prefferred introduction because of its simplicity. And the other two diagrams shouldn't be at the top of the section because they're really just disproofs of the chain model, and thus should follow. The Analogy section should probably be eliminated, but I didn't want to put in the time to work it out and I didn't want to offend its author. Although I could make a detailed defense of every part of the old version, I think it would be a lot faster to just have you point out why you think certain parts should go. Then I could address only your concerns instead of defending everything in the section. Mindbuilder (talk) 08:20, 12 June 2010 (UTC)


 * Hi Mindbuilder. You appear to be working under the impression that what is required in Wikipedia is information that is correct; and providing it is correct nothing more needs to be done to secure its place in the encyclopedia.  That impression would be incorrect.  WP:Verifiability specifies some core-policy for Wikipedia.  It beguns by stating The threshold for inclusion in Wikipedia is verifiability - not truth.  So the fact that you believe something to be true is secondary to the consideration of whether it can be independently verified by other Users.


 * It is true that not every word or even every sentence has to be covered by an in-line citation, but where a statement has been challenged, or is likely to be challenged, it must be supported by citation of its source. If such a statement can't be promptly supported by an in-line citation it must be removed.  It can be restored when there is an in-line citation to support it.


 * I have challenged many of the statements in the Theory sections. I did so by removing them from the section.  I am not saying these statements are incorrect, nor am I saying you don't know what you are writing about.  I am saying that this is an encyclopedia and one of its core policies is that all significant statements must be verifiable or they must be removed.  A statement that a User believes to be true but is not independently verifiable is called original research (although you probably don't regard it as original research.)  Original research is not permitted in Wikipedia - see WP:NOR.


 * So please study WP:V and WP:NOR and re-consider what you think may remain in the theory section. Best regards.  Dolphin  ( t ) 11:57, 12 June 2010 (UTC)


 * I'll add citations. If there is any specific statements that you particularly object to, let me know and we can make it a priority to work those out first. I should note that no source will support the chain model as being an EXACT model of how a siphon works, because its not meant to be an exact model. But the chain model deserves inclusion because it is a notable aid to understanding that is mentioned frequently in siphon explanations. Also, although citations consisting of published material are required of Wikipedia material, I think there must be an implied exception for simple facts that virtually anyone can confirm at home, in a few minutes, with no special training or knowledge. Especially when there is a video by a phd physics professor at a major university to save even the small hassle of demonstrating it yourself. When all you need is a garden hose, a little water, and a few minutes by a layperson, that is almost the ultimate in verifiable. Indeed, it is easier to verify these facts than it is to verify some rare or expensive published references. And a fact that can be personally and easily checked by almost any Wikipedia editor, is far more reliable than an appeal to the authority of a published source. The examples of a siphon starting out with a dry up column, and a siphon continuing when a bubble is introduced, are totally verifiable and reliable. Whereas the previous two require no more than a garden hose, the example of a siphon with a fat up column, requires the slight construction skill of poking a hole in the bottom of a plastic cup and inserting a tube (and maybe sealing it with glue or butter), but I'd say it is still fair to include it as well, even without a citation, because it is so readily verifiable by almost any editor. Those three demonstrations are very important to the issue, and need to be retained, though the statements surrounding or including them may be too strong or need revision. It's ok to challenge and remove material that requires citation, but if you throw out the baby with the bath water by removing material that doesn't need citation (like the description of the chain model) as a group, without specifying the particular sentences that need citation, then I think it is fair to restore the material as a group.


 * I'd like to clarify that I understand that a fact being verifiable doesn't normally exempt it from the prohibition of original research or the need for citations. I think that a fact may be exempted from OR prohibition if it meets ALL of the following requirements:
 * -It can be personally verified by almost ANY editor.
 * -No special skills are needed to verify it.
 * -No special equipment is needed to verify it. The requirement for a tube, such as a garden hose, may be stretching this requirement.
 * -It can be verified as quickly or quicker than it could be verified in many rare or expensive published sources.
 * -The verification of the FACT is not subject to interpretation or dispute.


 * If a fact can be easily verified by any editor, and there is no controversy about whether it is actually a fact, then there is no reason it shouldn't be included in Wikipedia on the grounds that it is not verifiable or reliable. What separates most facts on Wikipedia from these is that most facts are not easy to verify by any editor.


 * Even if you can successfully challenge the three examples of the dry up column siphon, the bubble in the siphon, the fat up column siphon, I hope that you will not, on the grounds that these easily verifiable and valuable ideas will get much less exposure if you manage to get them removed. Mindbuilder (talk) 21:24, 12 June 2010 (UTC)


 * There is an interesting article here http://education.epsdivisions.org/muse/example-siphon-documents/Potter_Siphon_PE1971_p362.pdf that states "atmospheric pressure is postulated as an essential agency in most accounts, including those given by the standard dictionaries." This is just the kind of secondary source that is valued in supporting Wikipedia material and it summarizes the position of most accounts, supporting my material. Except that the very next sentence in the source rather destroys its credibility when it states "Few books point out that it is the cohesion of the liquid rather than the pressure of an external atmosphere which is crucial to the working of a siphon." The simple demonstration of a siphon that can start with a dry top or a bubble, proves that the cohesion isn't crucial, or even needed, at least not for all siphons. But still, this is a source that backs my material by recognizing that most sources recognize atmospheric pressure as an essential agency. I'm reluctant to use this as an in-line citation though because its errors could be confusing to readers. Mindbuilder (talk) 23:07, 12 June 2010 (UTC)


 * In case it's not obvious to everyone, I'd like to describe the simple procedure for anyone to verify that a siphon can start even with air at the top. Start by laying a garden hose flat on the ground and holding the end of the hose at, or slightly above the height of the faucet it is connected to. Turn the faucet on until the hose is filled with water and all the air bubbles have been discharged. Then while holding the disconnected end at the same level as the end connected to the faucet, turn off the water and disconnect the hose from the faucet. After disconnecting the hose, raise the two ends together to a comfortable height. Now raise one end about one meter above the other end. If the hose was free of air bubbles, a meter worth of water should spill out of the lower end, leaving the last meter of the higher end empty. Now put the empty higher end into a full bucket of water at a height of about 2 meters. Drop the lower end to the ground and watch the siphon empty the bucket. If it doesn't work, try dropping the end to a lower level or starting with the bucket at a higher level or emptying a little less than one meter from the end of the hose. To test if bubbles will break a siphon, submerge an inverted glass into the bucket and "pour" a glass of air up into the hose as it is siphoning. If the glass doesn't contain too much air, the siphon will sputter a little and continue operating normally. Mindbuilder (talk) 23:56, 12 June 2010 (UTC)

Some of my objections to the current Theory section are as follows:


 * The first paragraph begins by using the expression gravity pulling down on ... This is not scientific or technical language.  Such language uses the concept of weight.  Technical documents that might be used as sources for this article use the simple notion that different objects have different weights; they don’t say gravity pulls down on some objects more than it pulls down on others.  (You have cited an article by MC Nokes from School Science Review.  Does it really use language about gravity pulling down on, or is that your choice of wording?) —  Dolphin51   — (continues after insertion below.)


 * It could definitely be worded better, I just couldn't quite figure out how without spending a lot more time on it. My language wasn't terribly technical though. I was writing more with the layperson in mind rather than the scientist. I was only sort of trying to make a short introduction to how it works. I wanted to move quickly into the chain model, because I expect it would quickly give more insight than the hydrostatic language. I think a simple explanation of hydrostatics should be added, explaining that the shape, width, and slope of the tube doesn't matter to the pressure at a certain height. Although the reference I pointed out yesterday by Potter points out that a hydrostatic description of a siphon is incorrect, and I think he is partly correct about that. There is a more technical discussion of the siphon at the bottom of the Wikipedia article. Of course I didn't plagiarize Nokes, nor did I make a direct quote. It's a description of how siphons work in my own words. I invite you to make another try at improving the wording. Though it would probably be best to go over it here before posting it on the main page. Mindbuilder (talk) 20:02, 13 June 2010 (UTC)


 * The first paragraph also says gravity pulling down on the taller column of liquid, causes reduced pressure at the top of the two ... This is not a sound technical way to write about hydrostatic pressure. Try the following experiment.  Take a cup, bottle or short length of hose and immerse it under water so that it fills with water.  (Place finger over one end of hose.)  Invert the cup, bottle or hose and lift it partially out of the water.  The water in the cup, bottle or hose that is above the level of the surrounding water is now at a pressure lower than atmospheric pressure.  There is only one column so the reduced pressure cannot be said to be caused by anything related to the taller column of liquid.  The reduced pressure conforms exactly to the science of hydrostatic pressure – the water at progressively greater depth below the level of the free surface is at a progressively greater pressure;  and the water in the cup, bottle or hose at progressively greater height above the level of the free surface is at a progressively lesser pressure. —  Dolphin51   — (continues after insertion below.)


 * While I think my wording there isn't too bad, I agree there is much room for improvement. I mangled that a little when I realized I'd made a mistake in the previous version after reading Potter and hastily corrected it (in my defense, I think my previous incorrect version was consistent with many published explanations and the error was minor). I'll go ahead right now and replace "top of the two columns" with "top of the siphon". If you decide to rewrite it though, I must point out that your previous version of that paragraph was rather short of useful information. It basically said little more than that the pressure was higher than atmospheric at the exit. I think that's fairly obvious to almost anyone, and leaves out the important idea of the falling water lowering the pressure at the top of the siphon enough to allow atmospheric pressure to push up the up side. If you must, you could include a description that correctly includes the mechanism of vacuum siphons, but I think for simplicity's sake, the introductory paragraph shouldn't be confused with that academic case. I do think though that the discussion of vacuum siphons should be moved up into the bottom of the theory section. - Mindbuilder


 * The second, third and fourth paragraphs are devoted to the analogy of the chain model. The article contains a section called Analogy and yet three paragraphs of analogy are located in the section called Theory.  The article even concedes the chain model is flawed.   You have written that the chain model deserves inclusion because it is a notable aid to understanding.  Remember that Wikipedia is not a textbook or a guidebook – see WP:NOT. —  Dolphin51   — (continues after insertion below.)


 * You're right that placing the chain model prominently is pushing into the gray area between an encyclopedia and a textbook. But I don't think it quite crosses the line. First, I think the chain model is very notable because it is discussed so often in writings about siphons. Indeed many published sources claim the chain model is very close to correct, and for vacuum siphons, it is. It is a very useful aid to understanding. It's hard to imagine leaving the chain model out of Wikipedia, and if it is going to be included, it should be positioned where it will help the reader's understanding most. I noticed that even your edits left the chain model in. As I mentioned before, I think the analogy section should be merged with the above theory section. - Mindbuilder


 * The fourth paragraph concludes by stating A notable common exception where liquids may have great tensile strength is in the capillaries of trees, where perhaps the water has been purified or filtered of gases upon absorption by the tree.  This statement is inappropriate in an article about siphons.  In addition, this is an area of controversy.  It also contradicts the statement later in the article that considerations of vapor pressure cause water siphons to be limited to a height of 10 m (33 ft). —  Dolphin51   — (continues after insertion below.)


 * You're right that's kind of an oddball there. But I left it in because it makes it clear that there are exceptions to the statements above it, and because it is interesting, and it is related to the issues being discussed. I was afraid that if I removed it that someone would try to fix the paragraph by reinserting it or changing the parts above it. It is an issue of controversy, which could be made clearer, and that is why I used the word "may". The 10m limit may be sort of incorrect. The article references video by Fletcher of a water siphon demonstrated to go up to 24m. But that is not verified and may only be possible for a few minutes until a bubble forms. Hughes says he plans to publish more experiments on the issue. - Mindbuilder


 * The fifth and largest paragraph in the Theory section is not entirely about correctly explaining the theory of the siphon. A large part of it is about de-bunking the flawed explanation involving the need for tensile strength. —  Dolphin51   — (continues after insertion below.)


 * That's true. By debunking the other theories, a better understanding of the correct theory is obtained. The dry top siphon, the siphon with a bubble, and the idea that the fluids are under pressure in typical siphons, are all critical to understanding the theory of the siphon. The want of those ideas has caused the mess of confusion over siphon theory that we are in. - Mindbuilder


 * The sixth paragraph states that the conservation of energy is not violated because more energy is expended by liquid falling than by liquid rising. This suggests energy is NOT conserved.  If energy is conserved then total energy at the beginning is exactly the same as total energy at the end.  If the article is talking about the mass of liquid falling in the longer tube being the same as the mass of liquid rising in the shorter tube that is the concept of continuity, not conservation of energy.  If the article is to talk about conservation of energy it needs to talk about loss of gravitational potential energy equaling the work done by viscous forces. —  Dolphin51   — (continues after insertion below.)


 * I didn't write or significantly review that paragraph. Hack away at it all you want. I think that once someone understands the theory, they won't worry about conservation of energy. They can see that it's just powered by falling liquid. No mystery there. It may be useful to leave in to directly address those who might wonder about the issue. - Mindbuilder

If you and other Users carry out an experiment, and perhaps even write an account of that experiment, that constitutes a primary source of information. Wikipedia must be based on secondary and tertiary sources. See WP:PSTS. So inviting Users to carry out their own experiment doesn’t constitute an acceptable source for the purpose of verifiability. Dolphin ( t ) 13:08, 13 June 2010 (UTC)


 * Notice the top of WP:PSTS where is says "a widely accepted standard that all editors should NORMALLY follow" and "normally" links to "use common sense". Generally it's not acceptable to use original research, but we have a special case here. Many facts, especially controversial ones, are not trivial to verify. This is almost the ultimate in verifiability. No expensive lab is needed. No significant expertise is needed. Any Wikipedia editor can verify them conclusively, with a water hose or drinking straw, more easily than going to the library to look up a reference. The fact that a siphon can start out with a dry top or continue with a bubble in it cannot be seriously disputed. Those are not controversial facts. It is very important to the issue, and its relevance is self evident. It is Wikipedia policy that original research can and generally should be mercilessly removed. But in rare circumstances, it CAN be left in if common sense dictates that it should. Do you seriously question the truth of those facts? If you don't, then you don't have to advocate for their removal. If you question the interpretations drawn from those facts, we can tone them down or possibly even remove them. But leave those facts in there because they're very important, and not in significant doubt. And try not to throw out things like the chain model that don't need cites (because it is not claimed by anyone to be exactly true). Mindbuilder (talk) 20:02, 13 June 2010 (UTC)


 * Thanks for the link to the first page of Potter’s article. I like the simplicity of Potter’s article but I see some statements I don’t like.  There is an error that is also found in our Wikipedia article – Potter says … more liquid, pushed in by the atmospheric pressure, enters at A.  As you know, the pressure at A is not atmospheric pressure!  The pressure at A can be considered to consist of two components – one component is atmospheric pressure, and the other is the contribution of hydrostatic pressure at depth h.  Potter’s article would be a little more accurate saying … more liquid enters at A, pushed in by the pressure at A.  However, the pressure at F is also atmospheric and this counteracts the atmospheric component at A, so it would be even more accurate to say liquid enters at A due to the pressure at depth h in the liquid.


 * You have written Potter points out that a hydrostatic description of a siphon is incorrect, and I think he is partly correct about that. Potter doesn’t say the hydrostatic description is incorrect – he says but its limitations as a theory for a working siphon are obvious since it describes a static condition.  Referring to the dynamic description Potter says This account is incorrect … … if … we neglect viscous losses.  In a working siphon the pressure at C is indeed higher than at D due to viscous losses.  Viscous losses are an important feature of any working siphon so ignoring them is not a good ground for declaring another account to be incorrect!


 * There is presently a lack of citations but I am not challenging the diagrams or text related to the bubble in the siphon or the heavy leg & light leg. I am challenging the overwhelming emphasis on the chain model.  I am challenging this both for lack of citations and for its inappropriateness in the Theory section.  Wikipedia is not a textbook or a guidebook.  If you still think there is a place for this emphasis on the chain model please use WP:3O to obtain a third opinion.


 * There was a sentence saying There is a video reportedly demonstrating a water siphon 24 meters tall, but the video is unclear and the demonstrator is selling an amazing medical device. This sentence may have a place on this Talk page but it is unencyclopedic language and I have deleted it from the article.  Dolphin  ( t ) 03:51, 15 June 2010 (UTC)


 * I don't have time to comment much right now, but I'll defend the chain model a little now. Although Wikipedia isn't supposed to be a textbook, it is supposed to impart knowledge and understanding. That's what an encyclopedia is all about, to learn from. There are some writings in textbooks that would be perfectly appropriate for WP. It is also Wikipedia policy to write for the layman rather than the expert. The chain model is notable because it seems to be discussed in the large majority of descriptions of how siphons work. The Hughes article discusses it, the Straight dope article discusses it, and many other articles I've seen on the net discuss it. In fact it's hard to think of an article that doesn't discuss it. That's two citations, I'll give more tomorrow if you like. Considering how many texts use the siphon model to teach siphon theory, there seems to be a lot of people that agree with me that it is a valuable aid to understanding. It's not claimed to be a fact so it need not be excluded on the grounds that it is original research. It's not original anyway. More importantly, the Hughes and Potter articles and others promote the chain like tensile strength of liquids as a good model of common siphons. It is thus helpful to understanding and an important myth to be addressed at the same time. It's important to address the chain model myth because although it may be discussed in printed sources, Micolich's video is the only other internet source I've found to debunk it decisively, and the Wikipedia article addressed it before he did. Furthermore it is actually a quite good analogy for vacuum siphons and siphons exceeding barometer height. A more detailed technical description of siphons is appropriate, but should probably be grouped with the more technical stuff later in the article. Although the theory section could probably be expanded if you like. But because the chain model is a great introduction, I still think it should be introduced near the start of the theory section. I think Wikipedia policy requires us to discuss this further before wasting the time of moderators. Perhaps you could discuss a little more about why you think it's inappropriate in the theory section. Would you prefer it in a different section? If it was put in a lower down section, do you think readers might quit reading from boredom of technical stuff before getting to it, and thus loose out on its beneficial and mythical aspects? Mindbuilder (talk) 07:43, 15 June 2010 (UTC)


 * I have listed the disagreement at WP:3O, asking for a third opinion on the matter of the chain model of a siphon, and whether the three paragraphs are appropriate at Siphon. Dolphin  ( t ) 12:42, 15 June 2010 (UTC)


 * Go ahead and move the chain model to the analogy section if you really want. I guess that is not too far removed. Especially if you do that then it would probably be a good idea to just drop the train model as being redundant. Please do maintain both the description of the chain model and the discussion of the two problems with it. Here are some more links in addition to the Hughes and Straight Dope ones, showing that the chain model is prominent in explaining the theory of siphons: http://jedlik.phy.bme.hu/~hartlein/physics.umd.edu/deptinfo/facilities/lecdem/f4-32.htm http://www.exo.net/~pauld/physics/syphon/syphonphysics.htm  http://www.physicsforums.com/showpost.php?s=707ce6fcfb131856b052b30ffe20e5c7&p=1616453&postcount=4  Here's a short description of the siphon from Harvard that doesn't use the chain model: http://www.fas.harvard.edu/~scidemos/NewtonianMechanics/Siphon/Siphon.html I think that if someone doesn't know the theory of how a siphon works, then the chain model is the quickest way to begin their process of getting an intuitive understanding of it. I think discussing its flaws also helps a person to develop a correct mental model of the siphon. I'll be back with more comments in a little while. Mindbuilder (talk) 18:07, 15 June 2010 (UTC)


 * Thanks for that. I have removed my request from WP:3O.  I have also moved the chain analogy from  Theory to Analogy.  There is now a need to review these two sections and fine-tune them - no guarantee something isn't duplicated or omitted.


 * I think you are overestimating the value and significance of the chain model in helping people to gain an understanding of siphons. I don't doubt that the chain model is  used regularly in the popular science media aimed at the general public, but I seriously doubt that any self-respecting science teacher would resort to the chain model to explain the siphon, or analogy to explain any other hydrostatic phenomenon.  Serious science teachers with a class of teenagers should begin teaching fluid statics by introducing their students to Toricelli's law and Pascal's law.  Dolphin  ( t ) 23:00, 15 June 2010 (UTC)