Backyard Aquaponics
A place to discuss aquaponics

The Mad Scientist strikes again. I am constantly looking for new ways to improve old methods. (I don't know if this is new but it did come from my imagination.) One of the problems with bell siphons is that they are finicky. If there is not enough water flow into the grow bed, the siphon won't start unless you reduce the size of the stand/drain pipes. Doing that decreases the drain flow, however, and increases drain times. Faster draining is better for root oxygenation. In small systems, this is not a problem but with larger grow beds, slow drainage is not ideal.

Using a trap configuration is an option. The water will fill the stand pipe until there is enough pressure to push the trapped water through the trap and begin the draining process. But standard traps have their own problems as well. Sometimes the water, (especially with a slow fill), can drain too slowly, (matching the fill rate), keeping the auto siphon from starting.

What I have done here is, I created a double trap to act as a vapor lock so that the stand pipe, once the bed is filled higher than the holes in the bell, is sealed from atmospheric air. Then I created a small "siphon starter" circuit that bypasses the trap. The inlet of the siphon starter is lower than the highest part of the double trap and water will flow through the starter while keeping the trap full and the standpipe atmospherically sealed.

The stand pipe and drain line is 1" diameter PVC. The siphon starter is made up of only 1/2" and is further reduced by the male adapter that screws into the 1" elbow.Starting a siphon with a pipe that is effectively less than 1/2" is easy and requires only a trickle of water.

But if I reduce the siphon down below 1/2", wouldn't that cause the same drainage slow down I talked about in the first paragraph?

No that's the beauty of it! I call this a siphon starter circuit because that is exactly what it does. Once it starts the siphon, the air is rapidly removed from the atmospherically sealed stand pipe which in turn rapidly fills with water. This rapid filling of water causes any remaining air to be push out through the starter siphon as well as the trap and opens the trap to full, unrestricted drainage.

This could be a solution for systems in under developed nations or people who want to operate off the grid as water flow can be greatly reduced making very low power pumps a possibility on larger systems.

Nice! You have essentially created a venturi fitting and they are very effective. You can find these profesionally made through plumbing parts suppliers in the event your tape or sealer give out. I never thought of using one of these on my system. Thanks for the idea!

Thank you!

I also have another method that I tested and works even better. It doesn't require the use of a trap so there is less plumbing involved. In order to make it work on my current system, however, I had to reduce the drain to 3/4" for a couple of inches. That caused the restriction I wanted to avoid. But if I decide to drill a 1-1/4" hole in my GB's, I can use this other method and still have a full 1" unrestricted drain.

I'll post pics of that method when I get a chance.

I didn't get to edit this in time but I meant to say that the other method doesn't require a "double trap". It still needs the single trap.

Here is another method I developed that works very well. (Even better than the first method with the double trap.) The only drawback to this is that I would need to increase the hole in my grow bed to 1-1/4". No easy task. If I compromised on drain rate, however, I could use a 3/4" standpipe inside my existing 1" standpipe.

That is what I did to test this method and it worked very well. The drain time was quite a bit slower with the 3/4" restriction but certainly fast enough for most purposes.

Here are some pictures. The trap is not shown but a simple single trap to atmospherically seal the standpipe can be placed downstream.

(Please forgive the typos on the pictures. I only had Microsoft Paint available and it is very difficult to go back and correct mistakes.)

There are some errors in the above pictures.
1. The first Tee fitting is 1-1/4"x1-1/4"x1/2" (not 1")
2. The standpipe is 1" all the way. (not 1/2")
3. The bottom Tee is 1"x1"x1/2"

Sorry about that. I was using Microsoft paint and it was a struggle getting it labeled. I usually use Photoshop but I wasn't on my normal computer.

I'd like to see more pics both up above and below. Great Ideas. I'm always in the mood to try something new.

thanks

Yeah, I'm just not getting it either. Both inlets are at the same height, yes? So water enters the outside inlet, flows down into the tee, to be diverted around back around to the bottom. Okay...... So when does water enter the inside inlet if the outside one is never restricted? Can you list and number the sequence of events? If the main pipe is all a solid piece, then couldn't the 1/2 inch pipe just go up to the top on it's own instead of complicating things by combining the two?

makes an affnan look even more simple

The Affnan is MUCH more simple. It is great and will work for most applications. I am simply experimenting with maximizing outflow with a minimal inflow. The Affnan reduces the effective diameter of the drain. That is the trade off for creating the vortex effect. Also, in my system, it would only work with a high water input. (What I consider high.)

I am trying to develop a system that will work with just a trickle of water and yet provide rapid drainage. The two methods above both work for that application but are not necessary for most people.

I will try to shed some light on this. I am assuming that you are referring to the second method.

1. When water begins to flow into a standpipe, it always starts, of course, at the sides of the pipe. If the water is not flowing fast enough to seal the drain from atmospheric air, the drain, (output), will equal the input and a siphon will never start because air pressure will remain equal. (That's were the Affnan siphon helps by causing a vortex that helps seal the pipe and by reducing the effective diameter which also helps seal the pipe. But you still need a substantial amount of flow OR a smaller pipe for this to work. Again, normally effective in most system setups.)

2. The water that flows down the sides of the larger pipe cannot enter the inside drain pipe. Due to the bushing, the only path for that water is out the 1/2" starter circuit.

3. The entire starter circuit, therefore, consists of the 1/2" pipe and the larger diameter pipe. The larger diameter pipe is restricted by the inside pipe. So, it doesn't take much water flow to fill space between the outside pipe and the inside pipe.

4. The 1/2" pipe takes a couple of bends which along with is small diameter, helps that pipe to "seal" quickly. As the gravity forces the water down, the air inside the 1/2" pipe and the larger outside pipe is displaced.

5. Nature abhors a vacuum so that air must be replaced.

6. So now, we are inside the bell. The water is flowing down the outside pipe and into the 1/2" pipe pulling all the air with it. The smaller inside pipe is sealed by a simple inline trap, (does not need to be a full P-trap), so no air can enter that pipe from outside to replace the displaced air.

7. As the air continues to be pulled down the starter circuit with no means to equalize the pressure with new air, water is the only thing left to alleviate the vacuum that nature so abhors. It is easier for atmospheric pressure to move air than water but once it has no other choice, pressure will push the water in the grow bed up the bell more rapidly than the water is flowing into the bed.

8. As the air pressure continues to drop inside the bell, more and more water fills the bell until there is so much water that it begins to flow down the inside pipe.

9. Once that happens, the full autosiphon has kicked in and all of the air is gone from inside the bell. The entire space is filled with water.

So, in summary, the smaller pipe circuit causes a quick starting autosiphon that requires very little water input. That siphon removes all or most of the air from the bell and kicks the full siphon into action.

I plan to put a video together to demo this. I hope I can find the time in the next couple of weeks.