Backyard Aquaponics
A place to discuss aquaponics

Traps at the bottom of the standpipes always slow down the siphon pull force. Gravity pulls straight down, not sideways. It's best to build a bell siphon with only the vertical drop first, then add a trap if it will help with any problems.

+1
I have no traps or airhoses on my bells.
Traps and airhoses are for cheating the properties of the bell. They allow you to tradeoff different aspects to get functionality.

-1

With respect, I'll have to disagree with the concept of not using a trap (or a hose). If I had to guess, you guys are filling your GB's fast enough to get the siphon to start on it's own from the overflow...and in that case, using the bell has about he same affect as making a loop siphon.

The entire purpose of a bell is to trap the air so that the shock of the burbing action is able to create a quick flood over the standpipe, which then starts the siphon.

With my (working) GB's, I'm filling them at about 1 gal/min. This is allowing me to use a very small flow pump and will cycle the GB every 1/2 hour. I would wager that if you slowed down your fill rate, your bells would fail to start properly.


I've watched my breather hose very carefully: Once the water level goes below the level of the hose, you are correct that there isn't enough siphon to suck the water up & out, but the hose is large enough to prevent the surface tension from keeping the water in there, so that the water drains out of it "backward". Once enough of the weight of the water backs out it has no problem sucking the rest out and clearing the tube/breaking the siphon. I believe this is part of my problem...the siphon breaks, but takes a little longer to do so, and in the meantime, the residual water from the stone is still filling and can choke the tube closed again, which gets the siphon stuck in limbo.

BTW, I'm not trying to be a wise-a** but I work for a pump company so I have a few fluid engineers at my disposal. (http://www.iwakiamerica.com) I may not know what I'm doing with this stuff, but when they start taking about water, my head starts to spin!

Hmm, food for thought. I know on my bell siphon bed, it does require a very fast inflow to work properly. Maybe I'll add a U-trap, and see if I can use a slower inflow.

I wonder if I will need to lower the inner pipe though.

My standpipe is about 1 cm lower than the water level of the GB. In my experiments, I found that the taller the "U", the lower the standpipe could be, since it too more pressure to burb. My current traps is barely enough to create the trap for the air.

I had thought this was a healthy discussion. :-) We're open enought o call you a wise ass if that's the way you're coming across, which I don't believe you are.

Interesting notion though, I have thought about the gravel slowing the water, but have been unable to come up with a meaningful hypothesis of what problems it may cause. Most of what I noticed was having the gravel would incite the bells to work better...
I am putting in a new system.
My old system consisted of a bed with a loop siphon, draining onto beds with bells. So stopping the bell was gauranteed by the loop. And the loop provided high flow to the bell, such as a big pump on a timer would.
My new system is to be low flow.
I have however designed myself into a corner
I am unable to access (now) the underside of my beds to add a lock on the siphon.
As a result, the siphons only start under high flow, and have trouble stopping. Adding an air tube and increasing the flow would help me, but I don't want to buy a big enough pump to do that.

I am using a uniseal through the bottom of my beds, dropping a 25mm (1") standpipe into a buried 150mm (6") drain. I can only access this from the top without digging all the gravel out and moving the beds :shudder:
Any tips or tricks anybody can think of to create an 'Inline' lock?

you could try sliding a short piece of 12mm pipe down inside the standpipe. The restriction slows the initial flow down until the syphon begins. Then the pipe doesnt make much difference. It is the same principal as the outlet (sparge) on a urinal flush pipe.

If your still having problems with that syphon take a look at how i solved it on my 2400l growbed. I think i have some photos towards the end of my members system thread. This method may help you it hasn't played up once.

got a diagram Bluefin? I'm not great with words

I haven't it tried it, but what if you tried to put the trap (or loop) inside the bell? You may have to have a bigger bell to fit it in there, but I think it would work the same as having the trap under it? I'll have to check with the engineers at work to see what they think.

Kuda, what size pipe are you using for your bell siphon and what rate of inflow do you have?

Interested to know how you have managed to keep them working with vacuum break hoses. The one I have messed around with I found I could get it to break without a vacuume break but there was a very small inflow window in which it would work and then it would not break. I added a vacuume break in the end as it would never seem to run for more than a day or so before it would no longer break due to small inflow changes.

Syphons are funny things, I have two that break without air breathers and one that just will not. I have tried swapping the stand pipe and bell from one of the beds that work and it still won't work on that one particular bed regardless of inflow rates. The only obvious difference is the length of piping running from the bottom of the bed to the sump. The shortest one just won't break without the breather.

My understanding is that at lower inflow rates the syphon does not form in the standpipe or in the bell itself because, due to the vertical fall, gravity pulls the water down before it can create a seal in the pipe. It seems that at low inflow rates the syphon forms better in a horizontal air trap. How or where this air trap is created doesn't seem to matter as long as at some point there is a solid body of water in a pipe moving away from the grow bed causing a vacuum (suction) behind it. I am not sure how the height of the bell affects all this, but as others have suggested it does appear that higher bell height works better.

At higher inflow rates I have seen a syphon start with no air trap. I think in this case the syphon did form in the vertical standpipe itself due to gravity not being able to draw the water down the pipe fast enough to prevent the pipe sealing over with water. This is backed up by the fact that at really high inflow a short syphon period can form without a bell being in place.

I have all my exit pipes falling vertically then rising slightly. This configuration works fine for me because the horizontal piping fills with water and seals at the high point. The moment it seals the entire column of water within the pipe moves and creates suction.

I don't use vacuum break hoses, but I'm thinking I need to, but only if I up my flow. I'm having trouble with them starting.

I did think about the lock in the standpipe, but hadn't figured yet the bell... it would be difficult to fit.

I'm using 1" standpipes

I use a 50mm bell

I got some flexihose and set up a loop in the bell, but there's not quite enough to start that either, although it does break well.

I will have to have a think about the lock in the stand-pipe... it would be difficult to keep small, 2 90 bends are like 120mm wide.
the filter pipe is only 100mm.

I don't think I could even manufacture a lock that would fit within a 50mm bell and still allow water to flow past it. not at 1" diameter.
As I'd need 1" x 3 for the pic as it is drawn, and 1" free space to allow water to flow up.
as area goes up by the square of the radius, we're talking 0.25 * 4 * pi inside 1 * pi
that fits only too perfectly, my engineer side has defeated the mathematician with the 'practicality' argument.

I will have a think though.

Sorry if this is a bit odd, my post I mean. but

SUCCESS

I got my bell siphon to work properly! The solution is too simple. As I wrote earlier in the thread, I use the tube to break the siphon. The problem was that as the tube started to suck up some of the air, the siphon would slow down a bit, then make the water level rise again (from the risidual water in the gravel). Eventually the siphon would sort of get stuck in a state where the bell was full of water, but the GB was empty...then as the GB would fill, it would trickle the water out.

So, I thought, "how could I get the siphon to break faster?". INCREASE THE SIZE OF THE TUBE! I changed it from a 1/4" ID tube to 1/2". As soon as the water level goes below the tube, a monster gasp of air which quickly drops the level of water in the bell. It now happens so fast that there is a quick girgle and it's done.

I still wanted to prove that it was necessary to have a trap and tube. I stand by the fact that you don't need a tube or trap as long as your fill rate is fast enough to flood the standpipe (you might as well just install a loop). However, if you slow the fill rate, the siphon will never start. I set my fill rate to LESS than 1 liter/minute....to about 1/2 liter/minute. It took my GB about 3 hours to fill....the bell siphon still kicked in perfectly!

I'm now planning slowing down the fill rate of my other beds...I should be able to save some energy by reducing the flow. I'm planning on drawing a dimensional diagram of my bell...hopefully it will help out some of you that have had bell issues in the past.

Thanks for all the earlier suggestions!

-Rob T.

this works for me regardless of input flow