Silent and Failsafe Overflow System

[LXXero]

the real thing is, there's no magic fixed formula for bean animal. I feel like a lot of people try to copy bean animal to the T, and even he admitted he only built it that particular way due to what parts he had on hand at the moment. really the notion is just 3 drains with one being siphon+gate valve, one being trickle, and one emergency. there's numerous ways to actually make it work, though. there's not some single plumbing formula that must be followed, and half of this game is experimentation, so buy some extra pvc and play around!

the latest version would be the one you do next, on your tank, and any slight variations that may have.

There's all sorts of variations you can do. If your siphon sits low enough, you can forego the double-90 and just stick a strainer on it. Or, turn one of the 90s slightly so there's less chance of an air bubble getting stuck in there. Or even drill it. Among other variations of this for the full siphon drain...

depending on your water level, the durso doesn't always need an airhose in it, sometimes just drilling a hole at the right spot on the 90 is all you need.

 

[nanomania]

how big is the tank? what are the tank dimensions? I like to have it at least 1/3rd the length minimum. some people go coast-to-coast (full length), that's the best option if you can do it. I just opted for a tad over 1/3rd (18" in 48" tank....16" would be 1/3rd exactly, so i went a little over)

apologize for giving non-metric numbers, i wish we used it here too.

54x24x24"dt and sump 54x20x17 the holes will be on the side, not at the back or bottom. Means on the 24" side.

Sent from my GT-N7100 using Tapatalk

 

[LXXero]

54x24x24"dt and sump 54x20x17 the holes will be on the side, not at the back or bottom. Means on the 24" side.

Sent from my GT-N7100 using Tapatalk

personally i'd go longer than 30cm....30cm is about 12". 18" would be 1/3 of your tank. I would aim for the 18-24" range to be honest, maybe even jump to 24"...which is like 60cm.

the best option is to go full length, though, if you can. Bigger is better, in this case.

 

[nanomania]

personally i'd go longer than 30cm....30cm is about 12". 18" would be 1/3 of your tank. I would aim for the 18-24" range to be honest, maybe even jump to 24"...which is like 60cm.

the best option is to go full length, though, if you can. Bigger is better, in this case.

Problem is if i go 24" it will look ugly on the side.

Sent from my GT-N7100 using Tapatalk

 

[LXXero]

Problem is if i go 24" it will look ugly on the side.

Sent from my GT-N7100 using Tapatalk

I'd go as close to 18" as you could then...12" will probably work but i'd expect the water level to rise a bit higher on it at full fl ow, you'd want to make sure you make the teeth really tall, or, go toothless and just set it low enough that it'll handle all the flow without running your tank level too high.

 

[nanomania]

I'd go as close to 18" as you could then...12" will probably work but i'd expect the water level to rise a bit higher on it at full fl ow, you'd want to make sure you make the teeth really tall, or, go toothless and just set it low enough that it'll handle all the flow without running your tank level too high.

This i what i bought from fish street . com, as a sample

Sent from my GT-N7100 using Tapatalk

 

[sleepydoc]

the real thing is, there's no magic fixed formula for bean animal. I feel like a lot of people try to copy bean animal to the T, and even he admitted he only built it that particular way due to what parts he had on hand at the moment. really the notion is just 3 drains with one being siphon+gate valve, one being trickle, and one emergency. there's numerous ways to actually make it work, though. there's not some single plumbing formula that must be followed, and half of this game is experimentation, so buy some extra pvc and play around!

the latest version would be the one you do next, on your tank, and any slight variations that may have.

There's all sorts of variations you can do. If your siphon sits low enough, you can forego the double-90 and just stick a strainer on it. Or, turn one of the 90s slightly so there's less chance of an air bubble getting stuck in there. Or even drill it. Among other variations of this for the full siphon drain...

depending on your water level, the durso doesn't always need an airhose in it, sometimes just drilling a hole at the right spot on the 90 is all you need.

Agreed, but a lot of people don't fully understand the physics of what is happening with the system and thus don't understand how modifications from the original design will affect the operation. Using Bean's original design is a safe place to start as one knows it will work.

You are completely correct that modifications are definitely possible, perhaps even desirable, depending on your tank configuration. The one for an external overflow that Floyd posted is a good example. When chaining it, though, you need to understand what's happening. I tried to go over that in my post - hopefully it was clear.

 

[Gorgok]

You are completely correct that modifications are definitely possible, perhaps even desirable, depending on your tank configuration. The one for an external overflow that Floyd posted is a good example. When chaining it, though, you need to understand what's happening. I tried to go over that in my post - hopefully it was clear.

One modification i have on mine is the airline is only a little higher than the normal running level, so it will go full siphon well before the emergency even gets wet. It works for me even on startup, where it does drain the box before a siphon is formed. Before it can drain the box a second time the siphon becomes stable. That may not work with longer plumbing or different sizes or flows, or even different elbow configurations. With my pipes being perfectly level to one another and both 90s facing down it does.

Trying that on a longer run or lower flow may not work, and could result in the cyclical draining of the box before a siphon really forms. Or if the open can drain the box to a point where the siphon can get enough air in to break the siphon that it is trying to form.

But i like it this way. My dry is dry until everything else is not working right.

 

[NASAGeek]

Agreed, but a lot of people don't fully understand the physics of what is happening with the system

That's EXACTLY why I am asking WHY.... why is this here? what drives the level of this and that?? Trying to understand the physics and operational drivers.

Loved your answers because they address those aspects. You speak my language... give me physics and chemistry and WHY... I'll figure it out from there.

Thanks again.

 

[NASAGeek]

One modification i have on mine is the airline is only a little higher than the normal running level, so it will go full siphon well before the emergency even gets wet. It works for me even on startup, where it does drain the box before a siphon is formed. Before it can drain the box a second time the siphon becomes stable. That may not work with longer plumbing or different sizes or flows, or even different elbow configurations. With my pipes being perfectly level to one another and both 90s facing down it does.

Trying that on a longer run or lower flow may not work, and could result in the cyclical draining of the box before a siphon really forms. Or if the open can drain the box to a point where the siphon can get enough air in to break the siphon that it is trying to form.

But i like it this way. My dry is dry until everything else is not working right.

So you are using your "Emergency" as your trickle and your "trickle" as your emergency, right?? Testing my understanding here.

 

[uncleof6]

the real thing is, there's no magic fixed formula for bean animal. I feel like a lot of people try to copy bean animal to the T, and even he admitted he only built it that particular way due to what parts he had on hand at the moment. really the notion is just 3 drains with one being siphon+gate valve, one being trickle, and one emergency. there's numerous ways to actually make it work, though. there's not some single plumbing formula that must be followed, and half of this game is experimentation, so buy some extra pvc and play around!

the latest version would be the one you do next, on your tank, and any slight variations that may have.

There's all sorts of variations you can do. If your siphon sits low enough, you can forego the double-90 and just stick a strainer on it. Or, turn one of the 90s slightly so there's less chance of an air bubble getting stuck in there. Or even drill it. Among other variations of this for the full siphon drain...

depending on your water level, the durso doesn't always need an airhose in it, sometimes just drilling a hole at the right spot on the 90 is all you need.

Unfortunately, there is a lack of understanding the reason for or function of the air inlet. Nor is there an understanding in general of how the relative heights of the standpipes affect the operation of the system. Most of the modifications (especially those that attempt to stuff the system into small external boxes) skew the basic relationships of Bean's design, and though they work to an extent, they are not working correctly; the numerous issues and questions concerning it are the evidence. I haven't seen everything but to date of what I have seen, I have only seen one that actually worked right, and I believe I posted a link to the video (though videos are not a great source of information.)

I believe, from the original writeup that Bean did on this system, or soon after, Bean stated he used 1" BULKHEADS, because that is what he had. It is my understanding that the rest of the design was well thought out, and planned, (sanitary tees to reduce turbulence for instance) not simply thrown together with "chicken nuggets." It turns out that the happenstance of 1" bulkheads was a good thing, as low flow rates and large bulkheads create a possible issue.

It is basic human nature to constantly sort the world into categories, predict how things work, and test those predictions. This is the very essence of science. However, without a firm and thorough understanding of the underlying science of the endeavor, the results are going to be less than satisfying. There maybe a 100 ways to build this system; there is only one set of physics that "make" it work. Sleepydoc, myself, Bean, and others, all talk about the physics of the thing. Either the "design" works with the physics, or the design is trying to beat the physics. You can't beat the physics.

Urban legend has it that Thomas Edison failed 2000 times to make a light bulb that worked. It is said he stated ~"I did not fail 2000 times; I learned 2000 ways how not to make a lightbulb. I only needed to find one way to get it right" or something along those lines... There are hundreds of lightbulbs of different configurations, but they all work exactly the same: (HID, flourescent, et. al. are not light bulbs,) the same set of physics.

There could be 2000 ways to build this system, but there is only one way it works correctly. The physics of the "designed system" have to match the physics of the original design. Yeah there is a certain amount of tolerance, as with most things. But if one does not have the physics down, well... they are just running water through a pipe, and the results will be unpredictable.

In a way, this is a single purpose item, and it is logical to assume that there be a single optimal form for this item, which would preclude any other design, and perhaps that "optimal form" has not been acheived yet. However, this "single purpose item" is a part of a larger single purpose system. Therefore, it must also adhere to the needs of the overall system, for that system to perform optimally. The old addage "form ever follows function. That is the Law" applies here. If form always followed function, there could be no competition; no debate.

What I see, is function following form. This seems to be the "status quo" of things. One could translate this to such things as "engineered obsolesence" in automobiles to insure future sales; or that beautiful LED fixture (and expensive) that simply cannot do what an ugly T5HO or Metal halide can do. Form is driving this hobby, rather than function. Smaller, in this case. But form ignores the actual function needed.

These needs of aquarium systems have been "developed" over a great many years now. They are fairly well publicized, along with a lot of rhetoric and anecdote seeking to minimize the needs, to rationalize the form. If form always followed function, the hobby would advance at a higher pace. The side effect would be the aquarium industry would go into a tail spin because much of the stuff sold does nothing but look good.

Long way around to get to: does the design "you" have meet the needs of the system as a whole (function,) or does it simply serve a desire to be minimal (form?) It may be just a drain to many, but it is part of an overflow system, that is part of a life support system. What I see is draining water, but not meeting the needs of the system as a whole, with these alternative designs thus far. (drain/overflow combination)

Since we are humans, and human nature being what it is, I don't think this discussion will ever end...

Oy vey, I did not think I could type that much anymore....

 

[sleepydoc]

So you are using your "Emergency" as your trickle and your "trickle" as your emergency, right?? Testing my understanding here.

No - the 'standard' setup is as I described earlier. At startup the water rises to the level of the emergency and stays there (because the excess water flows down the emergency) until the siphon is established, at which time the increased flow of the siphon channel drops the level in the overflow.

Gorgok's setup is very much 'non standard.' The tube for the open channel is below the level of the emergency and so the open channel 'trips' into siphon mode on startup, dropping the level in the overflow before the siphon is established. I can't say for sure without seeing his system in action, but from his description it sounds like his system works more by happy coincidence than by design. When the tube gets submerged for the first time and the open channel converts to a siphon, it drops the water level. Generally what happens at this point is that the siphon channel entrains some air before the water level starts rising again. This prevents the siphon from ever completely purging the air and the system never reaches a steady state. In his case either it doesn't entrain air, or it doesn't entrain enough such that it can't purge the air before the tubing gets occluded again.

Like I said, this setup happens to work for him, but in most cases it likely won't work. Moving the height of the tubing is easy enough so you certainly can try it with your own setup to see what happens.

 

[LXXero]

uncle, while i get what you're saying, i do question whether we've really had actual physicists really confirm any of this design stuff, or if we're just a buncha bedroom scientists trying what works and attempting to reverse-engineer the physics afterwards...at least it's how it feels at times. Maybe that's just what i'm doing, LOL.

my holes drilled in my 90 originally were intended to have an air hose, but when i realized it was working just fine, and still achieved the effect of turning the trickle into a partial/gurgling siphon when the main siphon is closed still worked as well, well, i decided to forego it, and i'm not sure what i'm really losing in that case? as far as i can tell all the original intentions are still there. In fact - without the sanitary Tee's, which many setups, like said ones jammed into small external overflow boxes like mine....the corner of said 90 is much closer to where it needs to be set to turn the trickle into a temporary siphon in the case the primary siphon jams. So the sanitary tees almost make that air hose a necessity, as they are raising the air input far too high (and also, in the case of bean's original design, sat outside the tank, so would have no way to have water simply "over flow" the hole, as it can with mine, you needed that air hose to sit in the weir somewhere to set the level at which the water would "close" the trickle into a siphon. With a small external box, a 90 on the corner works just as well, as you are foregoing the Tees, for better or worse.

I work in IT for a living developing software, and while what i do is arguably engineering, for as long as i've been in this professional there has been two types of people I've worked with, the first has been the academic type, the one who theorizes and analyzes and comes up with a plan before there is even a working solution. The other type is more of the mechanic type, the kind of person who builds and constructs first, worries about the diagrams and theory later. I can't help but feel some of what you wrote is not only form vs function, but may even have roots in the very personality and style of the person doing said implementation. I feel like i've worked with enough people over the years that I try to learn from both approaches, as they are not mutually exclusive, nor are they "right or wrong"

 

[LXXero]

This i what i bought from fish street . com, as a sample

Sent from my GT-N7100 using Tapatalk

those teeth actually do look really tall. You might be OK with that. Worst case scenario, you could bust out a few teeth if the flow was too much...

 

[Gorgok]

In his case either it doesn't entrain air, or it doesn't entrain enough such that it can't purge the air before the tubing gets occluded again.

Like I said, this setup happens to work for him, but in most cases it likely won't work. Moving the height of the tubing is easy enough so you certainly can try it with your own setup to see what happens.

With 1.5" bulkheads and pipe and about ~1000 gph of flow i think its the doesn't entrain enough option.

Since both pipes are level they both have to enter siphon at the same point, at least once the airline is submerged. Then its just a question of if draining the box to the bottom of the elbows breaks siphon or not. In my case it does not.

I have even had the pump get slowed for odd reasons once in a while and the siphon beats the return, only to slurp some air but not break siphon until i can get there to fix it. I don't think its off by much when it does that, but if i had no downturn on the siphon i don't think any situation like that could ever happen.

Similarly, i have had damn snails block my airline (its attached to the emergency pipe) and had the box flush repeatedly, whenever the water rises high enough to start the open as a siphon. Never has the actual siphon been affected much by that... just knock the snail off and a little air clears out of the siphon and it keeps on running.

So it may work by mistake, but it works =D. I just thought it silly that the emergency should go siphon before the open goes siphon...

 

[uncleof6]

uncle, while i get what you're saying, i do question whether we've really had actual physicists really confirm any of this design stuff, or if we're just a buncha bedroom scientists trying what works and attempting to reverse-engineer the physics afterwards...at least it's how it feels at times. Maybe that's just what i'm doing, LOL.

my holes drilled in my 90 originally were intended to have an air hose, but when i realized it was working just fine, and still achieved the effect of turning the trickle into a partial/gurgling siphon when the main siphon is closed still worked as well, well, i decided to forego it, and i'm not sure what i'm really losing in that case? as far as i can tell all the original intentions are still there. In fact - without the sanitary Tee's, which many setups, like said ones jammed into small external overflow boxes like mine....the corner of said 90 is much closer to where it needs to be set to turn the trickle into a temporary siphon in the case the primary siphon jams. So the sanitary tees almost make that air hose a necessity, as they are raising the air input far too high (and also, in the case of bean's original design, sat outside the tank, so would have no way to have water simply "over flow" the hole, as it can with mine, you needed that air hose to sit in the weir somewhere to set the level at which the water would "close" the trickle into a siphon. With a small external box, a 90 on the corner works just as well, as you are foregoing the Tees, for better or worse.

I work in IT for a living developing software, and while what i do is arguably engineering, for as long as i've been in this professional there has been two types of people I've worked with, the first has been the academic type, the one who theorizes and analyzes and comes up with a plan before there is even a working solution. The other type is more of the mechanic type, the kind of person who builds and constructs first, worries about the diagrams and theory later. I can't help but feel some of what you wrote is not only form vs function, but may even have roots in the very personality and style of the person doing said implementation. I feel like i've worked with enough people over the years that I try to learn from both approaches, as they are not mutually exclusive, nor are they "right or wrong"

Well, I don't know that it is working correctly. Obviously, if water tops the "elbow" for the open channel, before the dry emergency takes the flow, then the trip order is not correct, and therefore the odds of the system not starting properly are greatly increased. A very common problem with various implementations of this system, that have the air line too low in the overflow box. In essence these implementations are all showing in small or large doses this problem. The only time the open channel should trip, is when both other pipes are partially or completely occluded. The open channel trips, robs the flow from the siphon, and the siphon never really gets going. It will be silent, it may not be. It will drain water, or other symptoms may appear.

Maybe the word "physics" throws some folks off. We are not talking theoretical physics, rather Laws of Physics. Specifically Fluid Mechanics, which is sub-divided into Fluid Statics/hydrostatics (fluids at rest,) and Fluid Dynamics (fluids in motion.) It reaches into Aerodynamics as well as water flowing in a pipe. Bernoulli's equation is a large part of it. The theoretical physicists have played with it; at this point what a fluid is going to do in a given set of circumstances, is rather predictable, and repetitive. But honestly rather than getting frozen over the physics (which we can do nothing about) it might be better to look at the mechanics of how the system is supposed to work, what insures a proper start, and proper function of the system.

We have a working model, that demonstrates the mechanics of the system as they are "supposed" to be. E.G. the startup sequence of events, and operational characteristics. It is known to work the first time, without failure.

Water rises in the overflow box, putting head pressure on the siphon. Water continues to rise, and most often will flow into the dry emergency. Air purges from the siphon, and a good start is indicated by a sudden and noticeable drop in the water level. Some water flows in the open channel, but it cannot take much flow unless it trips to siphon in the process. If it trips to siphon during startup, the startup process is short circuited, and the open channel takes too much flow, and the siphon never really gets fully started, or takes a very long time to purge. It can be silent, however the system is far from working properly. It can be tweaked to a point, and be "without issues," but it is more work than if the system took into consideration the mechanics of the model in the first place. We have dealt with this issue time and again over the years.

Most of these small externals, do not have the vertical clearance to get a proper start without overly involving the open channel. Over involvment of the open channel means it is not starting right, and generally will not run right unless you play with it. Like I said, I have only seen one that worked "properly" and it was able to build head pressure and not involve the open channel. So it is a case of trying to get a system that is large, into a glass-holes type box, and then make it work. In the process the basic mechanics are altered, rather than maintaining the mechanics into the smallest package that is practical. Hope that makes sense. I have students build this system throughout the semester; you should see what I have seen...

Many have never run the full sized model, so really don't know how the system really works or how well, and the "no issues" is from a rather limited perspective.

As far as the tees goes, that is topology, not really function. The tees with removable caps are to facilitate maintenance of the system, but are not a required design criteria for the system.

 

[LXXero]

Well, I don't know that it is working correctly. Obviously, if water tops the "elbow" for the open channel, before the dry emergency takes the flow, then the trip order is not correct, and therefore the odds of the system not starting properly are greatly increased. A very common problem with various implementations of this system, that have the air line too low in the overflow box. In essence these implementations are all showing in small or large doses this problem. The only time the open channel should trip, is when both other pipes are partially or completely occluded. The open channel trips, robs the flow from the siphon, and the siphon never really gets going. It will be silent, it may not be. It will drain water, or other symptoms may appear.

I think that was the case with this other fellow's system, although not with my 120g's external overflow bean animal setup anyway (top of emergency is just below. It's close margins though admittedly, really close.) Also i used the strainer for the full siphon, which sits just below the double-90 for my "durso with no airhose". I never have issues with the open channel taking water before the siphon. Eliminating the double 90s on the siphon simply eliminates this problem. So my only double-90 is the trickle/durso style tube, which takes water just above where the full siphon is, kind of unusual and i almost thought it might be a problem but it's actually working amazing, if anything it might further be preventing venturis by having that tube sitting above the siphon's strainer cap. I never have issues with it not starting the full siphon. Ever. In fact, it's got me away from the notion of having to use double-90s for the full siphon line, as the siphon line will start about a million times easier without them. You can prevent venturis other ways, there is no requirement to use the double-90 for this purpose, that was one of my first lessons I learned doing this. you can just make out the strainer in these:

http://www.xero.cc/fish/120_overflow_exterior.jpg
http://www.xero.cc/fish/120_overflow_exterior2.jpg
that's before i drilled the hole in the double 90s for the trickle/openchannel line though, here's a current picture, it's kind of nasty looking in there because i just fed the fish and the pump turned back on, you can see the hole I drilled. It's literally just above the emergency, but the tolerances here are so close given the small external box, sometimes it's less about what's technically correct and more about what you got to work with:
http://www.xero.cc/fish/120_overflow_exterior3.jpg

I think the notion that you need the tube to only sit shallowly in the sump and/or you need to prevent this sort of back pressure against the water, is only really working around the problems associated with using the double 90s for full siphon anyway. if you don't use double 90s, you simply don't have that problem most of the time. I am doing double 90s in my cube's herbie, but I may ultimately drill a hole in it where it turns horizontal to prevent the whole issue of back pressure and/or bubbles forming in the double 90s and stopping the siphon line from ever picking up water. I mean, that's really what's going on, ultimately. Pop the bubble!

Maybe the word "physics" throws some folks off. We are not talking theoretical physics, rather Laws of Physics. Specifically Fluid Mechanics, which is sub-divided into Fluid Statics/hydrostatics (fluids at rest,) and Fluid Dynamics (fluids in motion.) It reaches into Aerodynamics as well as water flowing in a pipe. Bernoulli's equation is a large part of it. The theoretical physicists have played with it; at this point what a fluid is going to do in a given set of circumstances, is rather predictable, and repetitive. But honestly rather than getting frozen over the physics (which we can do nothing about) it might be better to look at the mechanics of how the system is supposed to work, what insures a proper start, and proper function of the system.

We have a working model, that demonstrates the mechanics of the system as they are "supposed" to be. E.G. the startup sequence of events, and operational characteristics. It is known to work the first time, without failure.

Exactly, I don't really have a degree in fluid mechanics, I won't pretend to truly understand what's going on at that deep of a level, rather, I'll eliminate variables and do tests to ensure the system can withstand any number of conditions.

Water rises in the overflow box, putting head pressure on the siphon. Water continues to rise, and most often will flow into the dry emergency. Air purges from the siphon, and a good start is indicated by a sudden and noticeable drop in the water level. Some water flows in the open channel, but it cannot take much flow unless it trips to siphon in the process. If it trips to siphon during startup, the startup process is short circuited, and the open channel takes too much flow, and the siphon never really gets fully started, or takes a very long time to purge. It can be silent, however the system is far from working properly. It can be tweaked to a point, and be "without issues," but it is more work than if the system took into consideration the mechanics of the model in the first place. We have dealt with this issue time and again over the years.

This is where we start to diverge, the notion that it's putting head pressure on the siphon is probably not truly correct. What's it's truly doing is trying to push the bubble out of the double-90s and/or single up-sidedown 90 sticking out of a bulkhead (which is basically a double 90 if you consider what's typically on the other side of the bulkhead, or sometimes it's a sanitary Tee over there...which is still basically fancy 90) This is actually a part of the design I feel like i've found improvements which can be made to ensure the siphon restarts no matter what. Either drill a hole into the 90 just where it goes horizontal, basically a place that would be under water normally, and above water during normal siphon operation. Somewhere on that 90 where it would prevent the bubble from forming as the water level rises.

Truth be told, I don't like the notion that a pipe might be 1" too deep in sump water to start, what if my sump water level is higher than normal, accidental overfill a bit and don't notice, now it just won't restart? I just don't like that. Pop that bubble! That's my contribution anyway, lol.

Most of these small externals, do not have the vertical clearance to get a proper start without overly involving the open channel. Over involvment of the open channel means it is not starting right, and generally will not run right unless you play with it. Like I said, I have only seen one that worked "properly" and it was able to build head pressure and not involve the open channel. So it is a case of trying to get a system that is large, into a glass-holes type box, and then make it work. In the process the basic mechanics are altered, rather than maintaining the mechanics into the smallest package that is practical. Hope that makes sense. I have students build this system throughout the semester; you should see what I have seen...

Hmm, to me this is simply a technicality, that makes no practical difference as long as it consistently does start every time. Haha, you've shown your hand as the academic, though! And I guess i've shown mine as the mechanical tinkering type, lol. It's all good, though. I spent my whole life tinkering with things through trial and error, that's always been one of my mainstays, I wouldn't settle for anything sub-par, I would have kept trying at it til I got it just right...You don't wanna know my history in school though, lol. It's not so pretty.

Many have never run the full sized model, so really don't know how the system really works or how well, and the "no issues" is from a rather limited perspective.

As far as the tees goes, that is topology, not really function. The tees with removable caps are to facilitate maintenance of the system, but are not a required design criteria for the system.

I spent months watching youtube videos of bean animal setups, showing various failure scenarios and how the system is supposed to operate, as well as read through much of this thread, and Bean's pages. I am sure many do not do as thorough research as I did, and I won't pretend to be an expert by any means, I've only truly done a single bean-style system on my 120g, and my 60g is more of a herbie style, i have gotten a feel for what works and what doesn't with these projects, not to mention I learned how to cut and glue PVC and a buncha other stuff in the process, it's been a fun little adventure that's for sure.

Yeah, I'm sure you could still clean it easily enough without the Tee's. In the external overflow boxes like mine, you don't really need to glue or thread anything. So it's easy enough to just pull out the fittings and clean them....

 

[uncleof6]

I think that was the case with this other fellow's system, although not with my 120g's external overflow bean animal setup anyway (top of emergency is just below. It's close margins though admittedly, really close.) Also i used the strainer for the full siphon, which sits just below the double-90 for my "durso with no airhose". I never have issues with the open channel taking water before the siphon. Eliminating the double 90s on the siphon simply eliminates this problem. So my only double-90 is the trickle/durso style tube, which takes water just above where the full siphon is, kind of unusual and i almost thought it might be a problem but it's actually working amazing, if anything it might further be preventing venturis by having that tube sitting above the siphon's strainer cap. I never have issues with it not starting the full siphon. Ever. In fact, it's got me away from the notion of having to use double-90s for the full siphon line, as the siphon line will start about a million times easier without them. You can prevent venturis other ways, there is no requirement to use the double-90 for this purpose, that was one of my first lessons I learned doing this. you can just make out the strainer in these:

http://www.xero.cc/fish/120_overflow_exterior.jpg
http://www.xero.cc/fish/120_overflow_exterior2.jpg
that's before i drilled the hole in the double 90s for the trickle/openchannel line though, here's a current picture, it's kind of nasty looking in there because i just fed the fish and the pump turned back on, you can see the hole I drilled. It's literally just above the emergency, but the tolerances here are so close given the small external box, sometimes it's less about what's technically correct and more about what you got to work with:
http://www.xero.cc/fish/120_overflow_exterior3.jpg

I think the notion that you need the tube to only sit shallowly in the sump and/or you need to prevent this sort of back pressure against the water, is only really working around the problems associated with using the double 90s for full siphon anyway. if you don't use double 90s, you simply don't have that problem most of the time. I am doing double 90s in my cube's herbie, but I may ultimately drill a hole in it where it turns horizontal to prevent the whole issue of back pressure and/or bubbles forming in the double 90s and stopping the siphon line from ever picking up water. I mean, that's really what's going on, ultimately. Pop the bubble!



Exactly, I don't really have a degree in fluid mechanics, I won't pretend to truly understand what's going on at that deep of a level, rather, I'll eliminate variables and do tests to ensure the system can withstand any number of conditions.



This is where we start to diverge, the notion that it's putting head pressure on the siphon is probably not truly correct. What's it's truly doing is trying to push the bubble out of the double-90s and/or single up-sidedown 90 sticking out of a bulkhead (which is basically a double 90 if you consider what's typically on the other side of the bulkhead, or sometimes it's a sanitary Tee over there...which is still basically fancy 90) This is actually a part of the design I feel like i've found improvements which can be made to ensure the siphon restarts no matter what. Either drill a hole into the 90 just where it goes horizontal, basically a place that would be under water normally, and above water during normal siphon operation. Somewhere on that 90 where it would prevent the bubble from forming as the water level rises.

Truth be told, I don't like the notion that a pipe might be 1" too deep in sump water to start, what if my sump water level is higher than normal, accidental overfill a bit and don't notice, now it just won't restart? I just don't like that. Pop that bubble! That's my contribution anyway, lol.



Hmm, to me this is simply a technicality, that makes no practical difference as long as it consistently does start every time. Haha, you've shown your hand as the academic, though! And I guess i've shown mine as the mechanical tinkering type, lol. It's all good, though. I spent my whole life tinkering with things through trial and error, that's always been one of my mainstays, I wouldn't settle for anything sub-par, I would have kept trying at it til I got it just right...You don't wanna know my history in school though, lol. It's not so pretty.



I spent months watching youtube videos of bean animal setups, showing various failure scenarios and how the system is supposed to operate, as well as read through much of this thread, and Bean's pages. I am sure many do not do as thorough research as I did, and I won't pretend to be an expert by any means, I've only truly done a single bean-style system on my 120g, and my 60g is more of a herbie style, i have gotten a feel for what works and what doesn't with these projects, not to mention I learned how to cut and glue PVC and a buncha other stuff in the process, it's been a fun little adventure that's for sure.

Yeah, I'm sure you could still clean it easily enough without the Tee's. In the external overflow boxes like mine, you don't really need to glue or thread anything. So it's easy enough to just pull out the fittings and clean them....

Well, the high pressure (or back pressure or whathave you ... to not get into semantics) isn't at the top of the pipe, it is at the outlet of the pipe. At rest, the pressure at the top of the pipe will be lower than the pressure at the outlet. Yes? Atmospheric pressure vs the hydrostatic pressure at the pipe outlet. Yes? Remember, the pipe is submerged so you have atmospheric pressure + hydrostatic pressure due to gravity exerted on the fluid. Most often atmospheric pressure is zeroed out (gauge vs true pressure.) When running the same relationship exists. Low pressure at the top, high pressure at the bottom. Yes? So there needs to be a way to overcome this, or nothing happens. What? Water flowing from low pressure to high pressure?

What is needed is head pressure (head height whatever) above the point where the water starts to head down, whether that is inside a tee, a double 90, or an open pipe. Without it, the system will stay at rest till the end of time. Even though there is a point where gravity and the surface tension of water take over the bulk of the work, a certain amount of "head pressure" is required to maintain the running state, or the system will return to a rest state.

The low pressure is easily observed. It is shown by the vortex that forms if there is not enough head height above the inlet to the siphon. The high pressure at the outlet is easily observed by noting how hard (or how easily) it is for the system to get going.

We are not talking huge differences. However, the deeper the outlet is, the higher the pressure, and the harder the setup needs to work to get going. (More head height is needed.) And yes, if the outlets are too deep in the sump, and there is not enough head height to overcome it, the system will fail to start, it simply does not matter what the upper topology of the system is. Whether that notion is liked or not, it is precisely what is going on. Ells where used in the model, and all of the systems I put up (~200+) and the reason for the ells is to reduce the height of the water column needed to maintain the system in a running state without sucking in air. I am not going to get into a debate regarding the relative merits of top end topology, all things equal the difference is not significant enough to really be concerned with, other than water height requried to maintain the siphon.

There is a great difference between draining water, and this system working the way it should. Watching you-tube videos is a double edged sword. How does one know if what one is watching is actually working right or not... videos are not a good source of information, generally they are more self-serving than anything else. That begs the question how would I know...

Over the years, and dealing with hundreds of "issues" posted here, a very short list was seen of possible things that could go wrong. All of them flaws in the implementation; a failure to follow the simple instructions, is what it really boils down too. Engineers love to fix things till they break; DIYers love to tinker with stuff, it is almost an obsession; Person A needs to outdue Person B; Person C needs to outdue both A and B. Persons D and E don't have minute 1 in experience, but they have something to say, that is only repeating what Person F and G said... Toss in some cognitive dissonance by as you put it "armchair scientists" (I don't own an armchair... seems i missing out...) and you have hundreds of posts going over the same material time and time again, (e.g. we have a big mess) that has only one possible conclusion: It works as designed. It is demonstrable, and repetitive (probably in the thousands of systems now) There is one set of circumstances in which the system works properly. There is tolerance in the system, by design, (it is not a rocket) however if modifications move the system out of the "neat little package of circumstances" it will probably drain water, but it can hardly be said to be working properly. (restart failure is a good clue... and that is 100% an implementation flaw.) Rather than fix the problem, or admit that the modification was not such a good idea, the hunt for a work around begins... well here we are 10 years later, still going over the very basics of how the system is supposed to work, and why it doesn't... but what really puts the cap on it is "Xgph without issue," really, need to get a flow meter involved, because the pump (despite the idealized flow curve) isn't even capable of doing that given the installation... (not speaking of anyone, or any pump, in particular) :wave:

I can't find fault in trying to minimize the form. However, there is a point where it just goes too far. And we are at that point. (past it most likely) We lose sight of other important aspects of the system overall (even if the drain system works perfectly.) Aspects that we spent "years discovering," that are practically completely ignored/dismissesd out of hand. (aspects outside the scope of a thread concerning a simple drain system.)

Got a problem getting the system to work? Is it sluggish and/noisy? Just doesn't seem to be right?? Ask sleepydoc, his keyboard does not have a mind of its own like mine does... :bounce2:

 

[uncleof6]

With 1.5" bulkheads and pipe and about ~1000 gph of flow i think its the doesn't entrain enough option.

Since both pipes are level they both have to enter siphon at the same point, at least once the airline is submerged. Then its just a question of if draining the box to the bottom of the elbows breaks siphon or not. In my case it does not.

I have even had the pump get slowed for odd reasons once in a while and the siphon beats the return, only to slurp some air but not break siphon until i can get there to fix it. I don't think its off by much when it does that, but if i had no downturn on the siphon i don't think any situation like that could ever happen.

Similarly, i have had damn snails block my airline (its attached to the emergency pipe) and had the box flush repeatedly, whenever the water rises high enough to start the open as a siphon. Never has the actual siphon been affected much by that... just knock the snail off and a little air clears out of the siphon and it keeps on running.

So it may work by mistake, but it works =D. I just thought it silly that the emergency should go siphon before the open goes siphon...

However, that is the intent of the design. Open channel goes siphon as a last resort. As a point of order, the emergency will likely NOT go to siphon as the head height will be such that a vortex forms and prevents the line from going to "siphon" mode (none of these things are really siphons) The open channel going "siphon" out of order delays and at times prevents the main drainline from going full on... Top of the list with a couple other circumstances that cause the system to not work properly. And, because it drains water, is NOT an indication that it is working properly. That you are happy with the results is great; but that does not indicate that it is working properly, or that anyone else should follow what you did. We need to remain focused on the theme of this thread, and that is a system that works the first time, everytime, out of the box, and all the science/deep thought has already been done... modify as you will, and as long as the basic relationships/mechanics are maintained, the results will be positive.

 

[NASAGeek]

Found this thread....

http://www.reefcentral.com/forums/showthread.php?t=1814606

It indicates that a 1.5" SCH40 PVC at full siphon should handle 1719gph at 4ft drop... if I am reading that right. It also indicates to derate that 15%... so 1461gph.

I am building a 1.5" BeanAnimal.... the question is how much flow can it handle safely??

1461gph seems low compared to other things I've read. If you assume the Emergency's function is to replace the Full Siphon in case it blocks... I'd allocate 0 gph to it. The Open Channel should be at 50% max I'd guess. The question is what's the max for an Open channel 1.5" PVC at 4 ft drop???

I am guessing here.... but I am doubting it'd get to full siphon and not be sucking air since it is relatively high.... if full siphon is 1461gph..... let's guess the max is 1000gph. 50% of that is 500gph....

Thus, a 1.5" Beananimal should handle roughly 2000gph and still be able to accommodate a clogged full siphon without overflowing.

Is that right??

M