Auto Siphon Technology

[Ray Bordelon]

Hi Aquaponics Enthusiasts,

AP is awesome. The food is plentiful and healthful. The process is interesting. We are all working toward having a smaller footprint. But, the auto siphon is ubercool. I love the mechanics of it all. Maybe it's a guy thing, I don't know. I could watch that thing all day long.

The one component that really makes it fun can also be limiting and at times troubling. There are many people that have systems that operate flawlessly and seldom need adjustment. Yet, I believe that lots of improvements in the auto siphon can be made.

When I decided to try to improve the current designs available, I first had to identify what I saw as shortcomings. The first thing that I wanted to do was create a design that saved energy. Then I saw dependability as the next issue. I needed to increase the auto siphons ability to handle a large variation of flow rates. I also knew it had to be simple without adding any moving parts that can wear and be a maintenance issue.

Currently, it seems as though flow rates are adjusted to the needs of the siphon when flow rates should be adjusted to the needs of the fish and plants. One might say that if a siphon is sized properly, the needs of the fish and plants are met. But, the needs of the plants and fish change all the time. For example, if its a cool damp month you should be able to let the grow beds cycle a couple of times a day without concern for the auto siphon operating properly. Or, if your fish are just fingerlings, they don't need the water flow that mature fish need. People tend to oversize their pumps just to make sure that all of their siphons work. This is especially true for CHOP, CHOP2 systems. If you had efficient siphons, you could downsize your pump and save energy. I don't have lots of experience in AP but, it seems like many people have more flow in their growbeds than is necessary just because their siphons need it. I also believe that faster drain is not only beneficial for oxygenating the plant roots but it saves energy. In a system that uses a continuous run pump, the pump is trying to fill the grow bed while the siphon is trying to empty the growbed. If the growbed is emptied rapidly, that is less time the the pump is being counterproductive and this also saves energy.

I love simplicity. The current designs that only have one moving part (the pump) are pure genius. The more components that a system has, the more problems a system has. So I knew that whatever I came up with, I had to keep it simple. But, simple is not always dependable. The current designs don't have much tolerance for variation in flow rates. This can make them unreliable. If a system gets dirty and flowrates change a little this can stop the proper operation of an AP system. I knew that I had to have an auto siphon that would work regardless of the flow rate.

After giving it a little though, I came up with what I think is the solution. I call it the "No Worries Siphon"...NWS for short. This weekend I built three prototypes of the NWS. They seem to work flawlessly. Of course, I only ran them for two days and nights but, I can't imagine what could be a problem down the road. Please feel free to use this design. In fact, I would be honored if the NWS is used to improve the efficiency and dependability of AP systems around the world.

 The first video that I am presenting is showing a 2 inch "U" siphon mounted externally that is started with a flow rate of 1 to 2 gallons/hour. I like the externally mounted siphon because it saves space in the growbeds, it can be mounted remotely and it can drain the growbed completely, using vertical GB space more efficiently. I tried smaller ones and they worked perfectly. So, I was encouraged to make a big one. The way this siphon worked, I am sure it would work the same with any size, including 6" or larger.  http://youtu.be/mzW6LPJ6urk

 The second video that I posted shows the detail of how it works. It's so simple. I hope you like it.

http://youtu.be/lhBkOTeNKD0

The third video shows it working with a 1" bell siphon remotely located. I imagine that a line of growbeds could all be connected to a common drain manifold leading to a sump. Draining into the sump would be one siphon for the entire line of growbeds. That siphon may need to be 6 inches in a commercial operation but this video shows that it would work.

 To be Posted

[TCLynx]

Novel method for making the trap on the bottom.  I'm used to seeing it done just with elbows.  The venturi for helping the siphon get started is kinda a cool idea.

As to it being better to fast drain or slow drain a grow bed.  That could stir up lots of debate.  I've done both methods as I started off doing all siphon beds.  Now I must point out that there are good points and bad points to almost every method of doing something.  With the fast drain, I found that more of the solids would actually get pulled out of the grow beds than if I let them drain slowly like on a timed flood and drain, so if one is wanting to use the water post grow bed for their NFT or DWC  it might be better to flood the beds quicker and then let them drain slowly.  I also notice less slime or solids build up at the inlet to a grow bed when doing timed flood and drain as opposed to the constant inflow on a siphon bed.  I usually found I had to slow the inflow to my grow beds way down for my siphons to shut off properly but I was doing mainly loop siphons with no extra bits, just the simple flexible loop.

Anyway, if your grow beds are only getting a tiny trickle of water, that might work ok for a system with only a tiny fish tank and a few goldfish, but how does the siphon do at cutting out when there is a faster inflow of water to provide enough filtration for a tank with more volume and more fish?

As to the comment that external siphons don't take up any space in the grow bed.  I would still strongly recommend there be a gravel guard around any drain in a grow bed so that one can access/see the drain from inside the grow bed and clean out roots so that bit of space savings is kinda negated if you still have a gravel guard in there taking up about the same space as it would with a stand pipe or a bell siphon internal.  Anyone who covers over the drain with media and then has to dig it and the plants up to figure out what is going wrong, probably won't care that much anymore about saving a 6" circle of plant space in the surface of a grow bed.

I have also discovered that the flow rate of water through the media and gravel guards can greatly impact the operation of siphons so it might be worth it to get some videos of your siphon working in beds with media.

[Ray Bordelon]

I see your points about fast drain. The NWS will work with any size siphon.  It could even be used in a 20 gallon aquarium powered by a little aquarium powerhead that has a bubbler built in.

TCLynx said:

Novel method for making the trap on the bottom.  I'm used to seeing it done just we elbows.  The venturi for helping the siphon get started is kinda a cool idea.

As to it being better to fast drain or slow drain a grow bed.  That could stir up lots of debate.  I've done both methods as I started off doing all siphon beds.  Now I must point out that there are good points and bad points to almost every method of doing something.  With the fast drain, I found that more of the solids would actually get pulled out of the grow beds than if I let them drain slowly like on a timed flood and drain, so if one is wanting to use the water post grow bed for their NFT or DWC  it might be better to flood the beds quicker and then let them drain slowly.  I also notice less slim or solids build up at the inlet to a grow bed when doing timed flood and drain as opposed to the constant inflow on a siphon bed.  I usually found I had to slow the inflow to my grow beds way down for my siphons to shut off properly but I was doing mainly loop siphons with no extra bits, just the simple flexible loop.

As to the comment that external siphons don't take up any space in the grow bed.  I would still strongly recommend there be a gravel guard around any drain in a grow bed so that one can access/see the drain from inside the grow bed and clean out roots so that bit of space savings is kinda negated if you still have a gravel guard in there taking up about the same space as it would with a stand pipe or a bell siphon internal.  Anyone who covers over the drain with media and then has to dig it and the plants up to figure out what is going wrong, probably won't care that much anymore about saving a 6" circle of plant space in the surface of a grow bed.

I have also discovered that the flow rate of water through the media and gravel guards can greatly impact the operation of siphons so it might be worth it to get some videos of your siphon working in beds with media.

[RupertofOZ]

Appears to be a complicated way to initiate a simple loop siphon...

 

Here's an example loop siphon... on the same page as your youtube video,....

 

http://www.youtube.com/watch?v=Bc6SS22IwUI&feature=related

 

Get rid of the cap, the venturi breather tube etc... the water level as it rises in your grow bed will displace the air from the "U" of the loop without them... and once the air is pushed down the vertical pipe the siphon will start... and cease when air is drawn into the tube at the bottom of the siphon drain (in the grow bed drain)...

 

Works each and every time... I've been running them on my barrel system... basically without fail... for over 6 years...

 

[Ray Bordelon]

Hi Rupert,

I also have a barrel setup that works well and I am going to leave it alone.  I am glad to see your system has worked well for so long.  Keep it up.  I put this info out and I am sure that somebody will find it useful.  I am incorporating it into my new greenhouse setup.

Best Regards,

Ray 

Appears to be a complicated way to initiate a simple loop siphon...

 

Here's an example loop siphon... on the same page as your youtube video,....

 

http://www.youtube.com/watch?v=Bc6SS22IwUI&feature=related

 

Get rid of the cap, the venturi breather tube etc... the water level as it rises in your grow bed will displace the air from the "U" of the loop without them... and once the air is pushed down the vertical pipe the siphon will start... and cease when air is drawn into the tube at the bottom of the siphon drain (in the grow bed drain)...

 

Works each and every time... I've been running them on my barrel system... basically without fail... for over 6 years...

 

[TCLynx]

Ray, I see that you seem to be aiming to make it work with smaller and smaller flows, My questions is more about will it still work (with your modifications) at faster flows or does it get stuck in a not kicking off constantly drained state if the inflow is too fast?  You advertize that the mods make it work with a wider range of flow rates but so far you haven't show it in your videos and you spend most of your effort saying it will work with very small flows.  It isn't that great a benefit if I must then set up a new siphon with 6" pipe and fittings if I want to run at say 5 gallons per minute instead.   Heck a 2" siphon is actually quite huge compared to most siphons people are using.

Ray Bordelon said:

I see your points about fast drain. The NWS will work with any size siphon.  It could even be used in a 20 gallon aquarium powered by a little aquarium powerhead that has a bubbler built in.

[Ray Bordelon]

FWIW, I posted a 3rd video:  http://youtu.be/Fpu0MiCwa70

[Ray Bordelon]

My tests show that It works at nearly any flow rate as long as the inflow is greater than the outflow at the weep hole.  The magic will be having the inflow somewhat matched to the size of the growbed and the size of the weep hole. Sure it will work at 5 gals/minute inflow but, probably not if your growbed is 10 gals.  If the size of the weep hole is adjusted so that the venturi can start sucking free air before the GB starts filling again then the siphon won't restart until the drain cap is full and the venturi vacuum can start reducing the size of the bubble in the siphon. One neat thing is that and somewhat counterintuitive, is that when using a one pump setup the slower the inflow to the growbed, (meaning more flow in the FT and venturi) the more vacuum produces and speeds up the start of the siphon.  And the reverse is true also.

TCLynx said:

Ray, I see that you seem to be aiming to make it work with smaller and smaller flows, My questions is more about will it still work (with your modifications) at faster flows or does it get stuck in a not kicking off constantly drained state if the inflow is too fast?  You advertize that the mods make it work with a wider range of flow rates but so far you haven't show it in your videos and you spend most of your effort saying it will work with very small flows.  It isn't that great a benefit if I must then set up a new siphon with 6" pipe and fittings if I want to run at say 5 gallons per minute instead.   Heck a 2" siphon is actually quite huge compared to most siphons people are using.

Ray Bordelon said:

I see your points about fast drain. The NWS will work with any size siphon.  It could even be used in a 20 gallon aquarium powered by a little aquarium powerhead that has a bubbler built in.

[RupertofOZ]

Ray Bordelon said:

My tests show that It works at nearly any flow rate as long as the inflow is greater than the outflow at the weep hole.  The magic will be having the inflow somewhat matched to the size of the growbed and the size of the weep hole.

 

But Ray... a loop siphon will always work.. regardless of inflow... if you siphon pipe is sufficient to drain the  outflow faster than the inflow...

Indeed... all siphons require this...

[RupertofOZ]

Ray Bordelon said:

FWIW, I posted a 3rd video:  http://youtu.be/Fpu0MiCwa70

 

Still basically a loop siphon.. just with an extended pipe... here's a video of external "bell siphons"...

http://www.youtube.com/watch?v=Zb2UM8wOzso

 

And one of the Hawaiin aquaponicists... Glen??... uses a single external bell siphon to drain all his grow beds...

Not "knocking" your brainstorming, or your enthusiasm Ray... but it's all been done before... with a 100 variations...

[Ray Bordelon]

I am not referring to outflow of the siphon once the siphon is started. I am referring the the outflow of the weep hole in regards to the siphon starting and stopping.  I also failed to mention that regular siphon breaks can be useful at higher flow rates.  For example: notches at the bottom of the bell or a siphon break tube mounted at the top of a U-bend or bell with the other end submerged to the low point of the drain cycle.

RupertofOZ said:

Ray Bordelon said:

My tests show that It works at nearly any flow rate as long as the inflow is greater than the outflow at the weep hole.  The magic will be having the inflow somewhat matched to the size of the growbed and the size of the weep hole.

 

But Ray... a loop siphon will always work.. regardless of inflow... if you siphon pipe is sufficient to drain the  outflow faster than the inflow...

Indeed... all siphons require this...

[RupertofOZ]

I follow you Ray... but most of what you've done... and good on you for experimenting... probably isn't necessary.. and can be done simpler...

 

The best improvement to siphon "technology"... particularly with regard to siphon initiation/break.. with differing flows...

 

Is the "affnan" bell siphon...

[Ray Bordelon]

Sorry, I keep thinking of things I didn't mention.  Old age is great, every day is a new experience.

Since the technique can be operated at a lower flow rate it helps to upsize the siphon as they break the siphon much more easily.

RupertofOZ said:

I follow you Ray... but most of what you've done... and good on you for experimenting... probably isn't necessary.. and can be done simpler...

 

The best improvement to siphon "technology"... particularly with regard to siphon initiation/break.. with differing flows...

 

Is the "affnan" bell siphon...