How can I control the amount of air vs water in an inverted bottle dispenser?

[Eddie Matos PhE]

Greetings,
I have read several threads concerning the operation of "inverted bottle dispensers", however, I have not found in any of them, the answer to my question: How can I vary or control the amount of air vs water that will exchange places at one given time, from a system at rest?

As to clarify, please imagine a Splarkletts or similar dispenser - the simplest kind, comprised of a small bowl with a faucet and an inverted bottle above it, held by the frame of the apparatus. My experiment does not apply to any water being dispensed by the faucet; I am only concerned with the fact that, at some point, water evaporates enough so that air is let in. As a bubble travels upwards, that same volume of water is going to be let out. This dispensing of water, in turn, will eventually close the passage of air again as the water line rises and seals the mouth of the inverted bottle.

I am interested in factors that will aid in varying the amount of air in the bubble that forms upon natural evaporation of the water in the bowl. In other words, I am interested in building a system that will make the dispensed water raise the water level as much as possible, from a balanced state, causing the "evaporation spurs" to be the farthest apart, but when they occur, they will cause the greatest variation of height on the water line; which is ultimately what I am interested in -" the height of the variation of the water line in the bowl", under the conditions described.

Perhaps, some of these factors could include the height of the water column in the bottle; the girth of the bottle and/or girth of bottle neck (up to no neck at all); the shape and thickness of the bottle mouth, as to facilitate of not the need to win over superficial tension of water in bow; angle of walls of bottle; size/diameter of bowl; etc.

Thanks for considering my question.

 

[Dale]

The only one I could see having an effect is the height of the column. The higher the column the lower the pressure and so the greater the volume of the bubble.

 

[Eddie Matos PhE]

Perhaps my wording was misleading. Do you still believe the same, if you were to compare the items on the diagram below, where the height is the same? Thank you!

 

[CWatters]

I think you want to minimise the exposed surface area of the water in the bowl.

That way evaporation is reduced (due to reduced air flow) and when water is emitted it has the greatest effect on the water level.

Your bowls are all about the same size. What if you had a bottle with a tall thin neck in a tall thin bowl? Perhaps with only a few mm of surface area?

 

[sophiecentaur]

I think the answer to this one is Damping. When a bubble starts to enter, it triggers a single half cycle of natural oscillation. Damping, in the form of a small constriction in the vertical tube in would reduce any overshoot and reduce the quantity of water that is dispensed.

 

[Eddie Matos PhE]

Ref CWatters #4 above, I agree that your suggested setup would delay the evaporation for the reason presented by you. I fail to see, though, a "thin and long" neck generating the largest possible "gulp" or "bubble" coming into the inverted bottle, which is supposed to happen once the water level of the recipient drops enough by evaporation, thus generating the biggest variation desired on the water level of the recipient; and this is just by gut feeling. Also by gut feeling, I would think that no neck would allow more air at once, no?

Ref sophiecentaur #5 above, I tried to follow your suggestion concerning Damping, however, upon Googling it, realized that I do not possesses enough background knowledge to understand the relationships involved here. Given your reference to reducing any overshooting of water, I must say that what I am looking for, to the contrary, is to maximize the dispensing of water in one gulp. I would like this gulp to generate the largest possible variation on the waterline of the recipient, even pushing towards dispensing more water than necessary to close the air path out of the inverted bottle. In turn, this would increase the amount of time for the evaporation to trigger the next "big gulp" to come down, based on the larger volume of water as well. (By now, it is no longer a bottle, it is an inverted neckless dispenser - based on my gut feeling)

In General, Among many aspects of such system that I believe could matter, there must be threshold values concerning them and the stability and triggering of occurrences. I am looking for those… What if the walls of the inverted container were shaped like a trapeze, would it facilitate the entry of air and exit of water? What other variables would there be? How shallow can the bottom recipient be to stop the water from flowing down? My intention is to approach such threshold values in the design, identifying which features would facilitate the water to flown down and which features would delay the evaporation-trigger. “ Farthest Apart Biggest Gulps” that will create the largest possible variation of height shown on the enclosed diagram, is the condition I am looking for!

Once I understand this process, the ultimate intention is to use the variation of such waterline to trigger other external events by communicating vessel, eventually implemented from the bottom recipient, leading to different types of switching devices. Am I chasing my tail and looking for the impossible?

Many Thanks!

 

[sophiecentaur]

Ref sophiecentaur #5 above, I tried to follow your suggestion concerning Damping, however, upon Googling it, realized that I do not possesses enough background knowledge to understand the relationships involved here. Given your reference to reducing any overshooting of water, I must say that what I am looking for, to the contrary, is to maximize the dispensing of water in one gulp. I would like this gulp to generate the largest possible variation on the waterline of the recipient, even pushing towards dispensing more water than necessary to close the air path out of the inverted bottle. In turn, this would increase the amount of time for the evaporation to trigger the next "big gulp" to come down, based on the larger volume of water as well. (By now, it is no longer a bottle, it is an inverted neckless dispenser - based on my gut feeling)

The quantity (volume) of air at the top of the reservoir must have an influence, I think. As the water in the trough is lost, the level will drop and so will the level at the top. This will reduce the pressure in the invert space. The change in pressure in the invert space will be
ΔP = P Δv/v
P is initial pressure, V is initial volume and Δv is the change. i.e. the change in pressure will be less when there is a lot of air in the invert space.
A bubble will enter at the bottom when the pressure at the bottom of the column is less than Atmospheric. Pressure The bubble will stop when the pressure is equal to AP again. This will happen when the extra air at the top brings the pressure in the invert space to be the same as AP minus the hydrostatic pressure in the column. But there will be an overshoot because the existing water flow needs to be brought to a halt and that will take time. A larger air volume at the top will require more flow of water before the pressure drops and, when the pressure changes sign, the flow will be faster and there will be a greater overshoot.
So I would say that, to restrict the amount of overshoot, you need a narrow constriction at some point in the pipe to limit the speed and a small invert volume so that the pressure changes quickly. For a large overshoot, the reverse applies; an open pipe and a large volume of air in the invert.

PS, why not use a readily available electronic level detector, if you will be using electronics at some stage?

 

[Eddie Matos PhE]

Thank your for your knowledge and willingness to help. My interests entail precisely not using sources of energy, other than those directly offered by nature. I may have misled you when I mentioned switching; I meant other natural switching devices and phenomena, such as floaters, siphons, faucets, communicating vessels, gravity, pressure, etc. I am a "perpetual motion chaser" kind of guy, who likes to invent things, but lacks formal studies.

I have been trying to put together a device that, unless the water source was shut off, it would periodically flush a small receptacle from such a water source, by an adjacent siphon that would be triggered by the water level rise provided by the "bubble", dispensed in the way that we've been talking about. Analogically, I am trying to bypass the water holding tank/floater/flush lever that we find in our conventional toilet apparatus, as per the diagram below. I have done some limited practical experimentation, with some success, but I need to adjust the dimensions to perfect the mechanism.

In this particular application, the system aims at replacing water in an aviary washbasin for birds to bathe in clean water. Splashing and/or evaporation at the washbasin level, would lower water line in dispenser, which in turn would generate "the bubble", which in turn would raise the water line enough to close the siphon, which would drain the washbasin till some air gets in the siphon and finally system would stabilize, and stay at rest after usage stops; being triggered only by evaporation, if shut off valve was kept turned on and no more splashing or drinking was to take place.

Understanding the feasibility and effectiveness of this endeavor, would certainly open the way for many other applications. I would appreciate any other suggestions...

 

[sophiecentaur]

Now I can see your real purpose, I would suggest that you should incorporate an automatic syphon mechanism, which (as with the classic urinal flushing system,) will dispense a tank full every so often when the syphon tank fills to the top. (This link shows an example but there are many others). The inlet can be as fast or as slow as you want, to get a very slow feed, another kind of syphon can be used, using a strip of fabric , dipped under the surface of water in an upper tank and hanging down below the tank into the second flushing syphon. Capillary action will wet the strip and you will start to get drips falling of the bottom end of the fabric. So you will get a steady but slow series of drips into the second syphon and this in turn will produce large dollops of water at long intervals. That would refresh your bird bath and flush it out with however much water your lower syphon tank holds.

 

[Eddie Matos PhE]

Thank you very much for your suggestion, Sophiecentaur! I watched quite a few siphon videos on YouTube, and I am convinced you have steered my thinking in the right direction. I was trying to figured it out by myself, and I took the time trying it; however, I am missing one item, that is, if I understood the principles correctly.

In my attempts, I tried to place the siphon in different ways, but they all would present a response directly connected to time, in a cyclical manner, independently from the water line of the washbasin. I would like to overcome the difficulty of having the cycles depend more on the water line, rather than time; even though time will always be a factor (indirectly), forced upon us by water volatility.

One of the main reasons why I would like to pursue this route, is because of our water shortage, and for exploring the mechanism for other similar applications, where the incidence of usage would be a much preferred trigger, over a timely cycle. I can relate usage (drinking, splashing, bathing), or its absence more directly affecting or being affected by the water line (including evaporation).

In this fashion, no matter how low speed you would setup the system to auto-siphon, the frequency of flushing would be certainly higher than the simple evaporation, in the absence of any other usage. Additionally, if we would set if so low speed, it would kind of loose its purpose for the "flushes" being so far apart; that is, without counting on the triggering by the usage of the existing water available to bathe or drink.

The series of events that I had expected from the original design were that from a quiet state (disregarding evaporation for now), we would end up with three points bound by "communicating vessels": The water line of the tank, the water line in the washbasin and the water line at the edge of the subsequent siphon. Upon any water being used from the washbasin, my expectation was to promote a re-alignment that would permit one of those bubbles (dispensed by the tank into the washbasin), to "trigger" the siphon and empty the washbasin. In other words, like a wave, the bubble was supposed to be big enough to push it over the edge or close the "trap" space to start the siphon.

I was not able to employ an auto siphon to produce the same result and I would greately appreciate if you could suggest some type of siphoning that would be triggered by the water usage (line) being lowered. It seems to me that the most auto-siphons operate with an increase of the water level, raising the line. It seems that our application would need a siphon that works the other way around, triggered by decrease of water level, lowering the water line. Otherwise, please let me know what I have missed in my reasoning.

Thanks!

 

[sophiecentaur]

If you need the syphon to operate when the lower tank level falls below a given value then there are several ways. One way I could think of is to have a float in the bottom which pushes upwards against a vertical rod, linking bottom and top tanks. The rod is an inverted J shape and the short section goes over the side of the top tank and pushed down on a plunger which displaces water in the top tank. When the bottom tank level is low enough, the plunger in the top tank is pushed down (following the lower float). This will raise the water level in the top tank and trigger the syphon action. The top tank can have its level maintained by a conventional ball cock system which refills the top tank after a short interval.
This achieves what you want by triggering the syphon action at a minimum water level rather than at a maximum level. By choosing the size of the top tank, you can arrange for a good flush of water into the bottom tank, which can wash out first from the lower tank each time by having extra water and overflowing the bottom tank a bit.

 

[sophiecentaur]

If you don't like 'moving parts' then it would be possible to use a tube with a balloon at each end, to connect bottom to top. A float would squash the lower balloon and water would be forced up into the balloon in the top tank and start the syphon. Getting a syphon started does need a finite amount of water to fall down the down pipe and it could be useful to have some sort of detent mechanism to delay the start so that you get more than a small trickle. A fairly small diameter down pipe with lots of height will help to get a good pressure difference to start the process. But if you want a good powerful flush with only a limited head tho work with, the only way is to have a more sophisticated system. Some sort of see saw arrangement can charge up a container which then dumps its contents all in one go, to ensure a reliable syphon action. But that would involve a moving part.