Backyard Aquaponics
A place to discuss aquaponics

I'm laughing at myself quite a bit here because I still don't fully understand how a bell siphon works. My roommate, a PhD in Mechanical Engineering, insists that the driving force is differential head pressure only. I understand the role differential head pressure plays in these systems, but I'm thinking there's something else going on that assists the flow even more. I'm thinking there's differential PRESSURE created that's greater than just differential HEAD pressure.

If it were purely differential head pressure, then the flow rate of the siphon discharge would reduce linearly with the drop in GB water height. But as I'm tweaking my first build-out and playing with the bell siphon components, the flow discharge rate seems to vary greatly and not necessarily as a linear function of GB water height.

Can anyone help me here?

There's 3-4 different forcing coming into play with a siphon. I looked right into it a while back when I wrote an article about siphons in the magazine, but I've forgotten exactly what they were now.. Atmospheric pressure plays a small part as well.

No need to understand how a siphon works. It just does. Basically, if water can find a way to exit, it will. If it can find a way to exit quickly/catastrophically, it will. If you can trick water to exit catastrophically in a way you actually desire, that's a bell siphon.

Hi Scupper

I think your roommate is correct.

http://eprints.qut.edu.au/31098/25/A_pr ... 282%29.pdf

cheers Lou

differential PRESSURE and differential HEAD are same for a given liquid.

differential PRESSURE is the result of differential Head.

but in some different kind of liquid same differential Head give different differential PRESSURE.


P.S. I almost confuse myself now.

If the out let from your siphon has a length of 0cm then the rate at which the siphon works decreases from maximum to 0 as the GB empties.

If the outlet goes down then the rate at which the water exits through the siphon is equal to the differential head. See pic:



The top to tanks will empty at the same rate once the valve is turned on and as they empty the rate at which the water exits will get slower and slower until it stops.

The bottom two tanks will also empty at the same rate as each other and the rate at which the water is emptying will also slow down but rather than approaching 0L/S the flow will approach roughly half of what it was when it started.

From this you can see why so many people have problems with their siphons not stopping. In the top tank with a siphon it is possible for the siphon to break because the flow out can equalise with the flow in. In the bottom example the flow out is going to be half of what it was when it started but should be way more than any pump is going to be supplying into the tank. Since the flow out will still be very fast as the lower tank empties a relatively vast quantity of air will get sucked into the siphon allowing it to break.

If those tanks are one meter deep and the siphon on the bottom right tank extends one meter down then the starting velocity of the water flow is 3.7m/s while the emptying velocity is 2.65m/s. If that were a 50mm pipe as the siphon for an IBC then the Q before it stopped would be around 18731 L/hr. Not many people have pumps that could keep up with that.

Conversely if the siphon only extended 100mm below the GB then the final exit speed would be 0.364m/s through a 50mm pipe that is a flow of only 2570L/hr. So if some one has two banks of gbs recieving more than 2570L/hr then it is likely that the siphon will never break.

This has all been quite interesting to read. I'm still thinking there's something about how the air in the top of the bell gets sucked into the discharge line, creating a vacuum in the top of the bell. (Right, or not?) I'm wondering if this vacuum assists in drawing the water up in the area between the standpipe and the bell? My gut is still telling me there's more to these wonderful gizmos than just differential head pressure acting alone as would be in a simple siphon.

This is why I'm wondering if there are more forces at play in these bell siphons than the simple differential head pressure.

I suppose I could test this theory. Measure the discharge flow rate of a simple siphon, and then measure it again working within a bell siphon. But I'm not quite sure how to set up this experiment properly, that is, what size hose to use in the simple siphon.

This is worth a read scupper

viewtopic.php?f=1&t=6544&hilit=how+does+a+syphon+work

Thanks for that link, Charlie. After reading it, I'm convinced that putting a separate control valve for each GB is worth the small expense. Should I be using gate valves (instead of 1/4 turn ball valves) to get finer control over the flow rate?

When a siphon starts there is no vacuum. In fact if the end of the pipe is sealed off from the atmosphere by some means (valve, water filled pipe) then the area in the top of the siphon is under pressure equal to the pressure of the water column above the top of the siphon. What happens as the siphon starts is that small amounts of air begin to get entrained with the water flowing through the siphon and as the velocity of the flow increases as the water in the GB get higher more and more air gets entrained. If air can not get into the siphon to replace the air being removed then the process speeds up because for every little bit of air removed more water can flow which increases the velocity which increases the ability of the flowing water to entrain more air. At some point the flushing velocity of the pipe is reached and the siphon runs flat out until the differential head pressures decrease and/or air gets back into the siphon to slow things down again.

Whats the point?????????? Meg is on the right track....water comes in, water goes out. How fast or how slow it does it is inmaterial, so long as the syphon starts and stops.

If you want to get scientific then hook up a "U Tube Manometer" filled with water (coloured preferably) Then will be able to visualise the vacuum inside of the bell. I can tell you how to make one if interested.

The point is in the this threads title. The understanding the physics....

If someone wants advice on how to fix a siphon we well tell them how to do so. If they want build one we will tell them how (actually we have already done that many times so we will refer them to the threads).

If they ask about the scientific principles involved in understanding how siphons work then all sorts of words that don't get used in regular conversations are going to be bandied about because that was what they asked for.

Thanks Stuart. I can appreciate that not everyone wants to understand the physics behind these gizmos, but some of us do as it's 1) very interesting, and 2) can lead to better designs. I appreciate your careful replies and careful choice of words.

I pulled up on the bell while it was siphoning, and I could definitely feel a pull. That is, I was either pulling against a decrease from atmospheric pressure (negative atmospheric pressure, or one could say a slight vacuum) or I was lifting the weight of the water in the column. Not sure which, but it was fun to feel.

Depending on the size of our siphon if you hold one as it begins to kick in you can feel the "pull" increase. Then as it takes off it can literally get ripped out of your hand. The size of my GBs meant I used a 100mm siphon (ie if it was a bell siphon the inner pipe was 100mm). When one of those kicked in it was impressive.

If I can finally get council out of my way the next one will have 160mm siphons.