ATS with Bean Animal / Calfo 3-pipe system?

[Floyd R Turbo]

I am looking at buying a 120 gallon tank (2x2x4) from our local LFS who is having a tank sale. I wish to use this type of overflow:

http://www.beananimal.com/projects/sile ... ystem.aspx (http://www.beananimal.com/projects/silent-and-fail-safe-aquarium-overflow-system.aspx)

mainly for the silent and fail-safe features, but also because I hate overflow towers, I value square footage, and I do NOT trust overflow boxes. I also (evil laugh) plan to eventually tear a hole in the wall where the tank is going and make it viewable from both sides. Bwahaha! So I want to do a coast-to-coast overflow on the end (2' side) and make the 'stand' to enclose the sides to hide the plumbing.

But is it feasible to run this type of system with an ATS connected to the overflow?

The issue here is that in order for the system to run silent, the full siphon standpipe must terminate at least 1 inch below the surface of the water in the sump. Once all the air is purged, this creates a full siphon with a maximum flow of 2000 GPH (2500 GPH actually with my design) using a 1" pipe and 5' of drop. This of course can be throttled back with a gate valve to match the pump output, etc. But how would I divert enough water to an ATS box from this pipe without violating the principle of the design? The way I see it, if I tee off before the gate valve to an ATS, but below the gate valve a full siphon is still in effect, why would the water flow to the ATS at all? It would just get sucked out I would think.

I could run the ATS off the open channel standpipe, but that would defeat the purpose of the secondary line of defense, and in order to get enough flow, would end up filling up and probably gurgling, flushing, etc and would defeat the 'silent' purpose.

So the sure-fire solution would be to run the ATS on it's own dedicated pump in the sump. But I'm trying to avoid that if at all possible. I suppose I could tee off the return pump into the ATS...

Anyways, I'm looking for someone who currently runs an ATS and a 3-pipe system like BA/Calfo. If you respond to this thread, please state if you do this. If you do not, please feel free to throw in your opinions and thoughts, but please take a minute and read through BeanAnimal's design description on the link above. His design is tried and tested, and has not changed since he started his thread (which has been split due to length). I haven't had time to read through it all for obvious reasons (it's frickin' long) so maybe someone else has.

I posted on BA's RC thread asking if anyone had installed an ATS on such a system but didn't get much of a response. So I figure that someone on here might have thought about it, or even done it.

[bucks448]

What timing...I was just read that thread on RC looking for a way to reduce bubbles/noise in my sump coming from my overflows. Right now I have a sock trapping bubbles, but I want to remove it when the scrubber takes off. Anyway, I think I'm going to test this tonight concept tonight.

I have a 180 with two overflows and two returns. Both overflows are of durso standpipe design. The scrubber is "T'd" off one of the overflows with a ball valve before the scrubber and another ball valve at the end of that overflow. If I cap off the air inlet for the durso on the overflow connected to the scrubber, I should be able to test this idea, minus the emergency third overflow. I'll let you know how it comes out.

Here's an idea of how it's plumbed.

[SantaMonica]

I wish to use this type of overflow

That is an elaborate version of a "tuned overflow", otherwise known as the Herbie (whose thread is about 5 years long). If you eliminate the 3rd pipe, it becomes simpler, and water level in the overflow fluxuates between the full-siphon and the overflow. I've used it for 4 years.


So I want to do a coast-to-coast overflow

Coast-to-coast is only valuable if you want to remove organics from the top layer of water. You can do the same thing by pointing a powerhead at the surface, which also gives you more shimmer.


But is it feasible to run this type of system with an ATS connected to the overflow?

Sure. Just "tune" the overflow as normal. Run it a few times first, and simulate different amounts of scrubber growth in the slot.


The issue here is that in order for the system to run silent, the full siphon standpipe must terminate at least 1 inch below the surface of the water in the sump.

No, the termination can even be above the water, pouring down into it. When tuned, there are no bubbles/air in the flow.


This of course can be throttled back with a gate valve to match the pump output

A tuned-overflow is somewhat self-tuning, because if the return flow goes higher, the level in the overflow box goes up, and this puts more pressure on the drain, which then drain faster.


But how would I divert enough water to an ATS box from this pipe without violating the principle of the design?

Just do some dry runs first. It's no different than sump level changing; it causes variable back-pressure on the overflow, which causes level changes in the overflow box. In your case, the slot will be causing the variable back pressure. You won't need a T at all if you size it right. Just make sure that you have about a 2 inch level difference between siphon overflow and emergency overflow. When tuned correctly, there will be a slight trickle down the overflow at all times. Don't forget to test the emergency overflow by completely blocking the full-siphon.

[Floyd R Turbo]

I wish to use this type of overflow

That is an elaborate version of a "tuned overflow", otherwise known as the Herbie (whose thread is about 5 years long). If you eliminate the 3rd pipe, it becomes simpler, and water level in the overflow fluxuates between the full-siphon and the overflow. I've used it for 4 years.

EDITED: Yes, I've read the RC thread that battles it out - good to know that you are using the Herbie. As far as I understand, the 3rd pipe is the open-channel standpipe, in case #1 clogs and the emergency pipe can't handle the flow, it takes over as a 2nd full siphon, and under normal operation it takes the excess flow so that #1 remains a true full siphon, and the emergency #3 pipe is always dry.




So I want to do a coast-to-coast overflow

Coast-to-coast is only valuable if you want to remove organics from the top layer of water. You can do the same thing by pointing a powerhead at the surface, which also gives you more shimmer.

I'm with you. The coast-to-coast specifically calls that feature out for increasing the effectiveness of a Skimmer based system, which we don't care about. This is from BA's site:


The faster we pull the organics from the display tank into the skimmer, the faster we can remove them before they are consumed by algae in the tank.

So it almost goes saying that there *could* be some value to surface skimming to an ATS. But I'm sure this is a very negligible increase. Shimmer would probably be better anyways since I'll be doing 96 LEDs.




But is it feasible to run this type of system with an ATS connected to the overflow?

Sure. Just "tune" the overflow as normal. Run it a few times first, and simulate different amounts of scrubber growth in the slot.

Sounds good. As you told me way early in my 120 thread, "Ok think of it as a PID controller. Flow is input, growth is output, and the output is inverted before feeding back to the input." so unless the tube gets clogged with something that blocks flow to the screen completely, I should have a full column of water providing pressure to the slot, which should prevent algae growth into the slot to the point of clogging (at least, in a week or two's time)

That being said, to keep the open channel standpipe from gurgling or flushing, I'll have to make sure I keep it 1.5" ID.




The issue here is that in order for the system to run silent, the full siphon standpipe must terminate at least 1 inch below the surface of the water in the sump.

No, the termination can even be above the water, pouring down into it. When tuned, there are no bubbles/air in the flow.

I guess that's not what I understood from reading several posts on the subject in the RC thread, plus on BA's site. It would not run *perfectly* silent if terminated above the water, there might be a splashing sound, but I doubt it would affect the function. The ATS will likely make more sound as the water discharges from the box (this will be an enclosed box ATS)




This of course can be throttled back with a gate valve to match the pump output

A tuned-overflow is somewhat self-tuning, because if the return flow goes higher, the level in the overflow box goes up, and this puts more pressure on the drain, which then drain faster.

This is true, what I can do I suppose is size the return pump & plumbing so that I get about 20% (or more???) over the desired screen flow rate, and the excess would go down the open channel. I suppose I should factor in pump slowdown over time between cleanings as well. More reason to oversize the open channel.




But how would I divert enough water to an ATS box from this pipe without violating the principle of the design?

Just do some dry runs first. It's no different than sump level changing; it causes variable back-pressure on the overflow, which causes level changes in the overflow box. In your case, the slot will be causing the variable back pressure. You won't need a T at all if you size it right. Just make sure that you have about a 2 inch level difference between siphon overflow and emergency overflow. When tuned correctly, there will be a slight trickle down the overflow at all times. Don't forget to test the emergency overflow by completely blocking the full-siphon.

EDITED: When you say "emergency overflow" I assume that correlates to the 2nd pipe in the Herbie system, which would be the 3rd pipe in the 3-pipe system. The emergency 3rd pipe in the beananimal is always dry. It's the open-channel standpipe that will have the 'slight trickle'. If the backpressure gets too high and raises the level in the overflow box too much, the emergency overflow kicks in. If it gets too much for the emergency to handle, the 2nd pipe channel closes (the tube end goes underwater) and that pipe will turn into a secondary full siphon until the overflow box level lowers below the air tube inlet. So what you end up with is an oscillation in the open channel pipe between full siphon and open channel if the slot tube is too clogged.

I guess this is all just specualtion at this point, it sounds like a calfo 3-pipe system will work.

[SantaMonica]

So it almost goes saying that there *could* be some value to surface skimming to an ATS

For a scrubber-only tank, the only benefit would be air-water surface agitation for oxygen, and possibly, re-routing the surface organic particles back into the column to be used as food.


That being said, to keep the open channel standpipe from gurgling or flushing, I'll have to make sure I keep it 1.5" ID

For the emergency overflow, it just needs to be able to handle all the return by itself. Gurgle is not an issue since it is normally just a trickle.


This is true, what I can do I suppose is size the return pump & plumbing so that I get about 20% (or more???) over the desired screen flow rate, and the excess would go down the open channel.

Just size it to handle a 50% blocking of the slot. This is one reason for long slots being better.


When you say "emergency overflow" I assume that correlates to the 2nd pipe in the Herbie system, which would be the 3rd pipe in the 3-pipe system.

Yes.


So what you end up with is an oscillation in the open channel pipe between full siphon and open channel if the slot tube is too clogged.

That's one reason I like Herbie better, no oscillation. Plus only two pipes.

[Floyd R Turbo]

That being said, to keep the open channel standpipe from gurgling or flushing, I'll have to make sure I keep it 1.5" ID

For the emergency overflow, it just needs to be able to handle all the return by itself. Gurgle is not an issue since it is normally just a trickle.



This is true, what I can do I suppose is size the return pump & plumbing so that I get about 20% (or more???) over the desired screen flow rate, and the excess would go down the open channel.

Just size it to handle a 50% blocking of the slot. This is one reason for long slots being better.

What I was referring to here was sizing the system for the maximum flow (clean screen, clean pump) and for minimum flow (dirty pump, clogged screen - which kind of offsets until you clean the screen). The point here being that as the pump GPH slows down over time, the full siphon may require a slight adjustment as the pump output drops so that a full siphon is maintained, meanwhile the open-channel standpipe (which is not an emergency pipe) needs to be large enough to handle the full additional flow not handled by the siphon standpipe when running a clean pump and partially clogged screen while maintaining a laminar flow.



So what you end up with is an oscillation in the open channel pipe between full siphon and open channel if the slot tube is too clogged.

That's one reason I like Herbie better, no oscillation. Plus only two pipes.

If your slot was clogged with the Herbie, you would get the same oscillation effect. The Herbie, if I understand it correctly, is a full siphon standpipe and the emergency standpipe (upturned elbow) that runs with only laminar flow (excess from siphon). If the slot tube was partially blocked, then the emergency overflow would fill up and turn into a partial siphon and flush the overflow box, then drain out, then fill up. The concept is the same.

In the 3-pipe system, the open standpipe is designed to take on the laminar flow, but has an airline connected to the capped top that leads to a point above level of the upturned elbow of the emergency pipe. If the emergency cannot handle the excess from a blocked main siphon standpipe, the tube goes underwater and seals the open channel, turning it into a siphon standpipe which will flush the overflow box (and completely take over the job of the full siphon, if it is indeed fully clogged) and then and only then will you get the oscillation effect. For this to happen actually, the main siphon would have to get completely blocked, because the open channel and emergency would be running w/laminar flow for wuite a while before oscillation kicked in.

Regardless, I'm not even 100% sure that this type of oscillation would occur. I guess it's worth thinking about, but once I get to building it, maybe it will all be fine...

[bucks448]

Well it seems you've hashed it out, but I wanted to add one thing that may be obvious to you but wasn't to me at first. I wasn't able to fully implement this tonight because with two separate over flows I need more pipe to extend one of the overflow pipes to just above the water level of the tank. Anyway, what I did find was that by cutting off the air flow on the durso overflow connected to the scrubber and adjusting the ball valves to create the siphon all micro bubbles coming from the scrubber were eliminated. This allowed me to run more gph over the scrubber which created a more even flow over it.

[Floyd R Turbo]

Interesting! So a full siphon feeding the screen could really be the way to go! I hope I pass my PE test, or else it will be next year before I get to try all this out and that will suck.

[vicbay]

Not sure if it helps. But couldn't you use a P-trap inline between overflow and ats pipe to replicate the effect submerging the pipe would have on the siphon? I haven't used either of these overflow options but I'm guessing that submerging the siphon side is to create a seal for restarting. A P-trap I believe would have the same effect if that's it's only purpose.

[Floyd R Turbo]

Bean covers that somewhere in his thread. For a full siphon, it doesn't really do that much, and add friction if anything. I'll have to see if anyone else put one on their and see why.

[bucks448]

Oh one more thing, I didn't read everything thing here, but somewhere in my reading today it was noted to use gate valves instead of ball valves to do this. Definitely go with gate valves. Fine tuning ball valves is a major PITA if not next to impossible. Whenever I extend the overflow pipe I'll be changing out the ball valve at the end of the other overflow for a gate valve. The ball valve to the scrubber is glued in, so I'm leaving that one.

[SantaMonica]

Yes must use gate valves. And even then, the fine-tuning is just a fraction of a turn.

[iggy]

How did your overflow work out?

Glad you feel free to make your own changes as I saw the original BeanAnimal description as thoughtful but rediculously overkill for 120 gallons. Two pipes and one gate valve is the way I would go. I am actually in process of using my own self made glass-holes type box with one suction and one overflow drain. I prefer going with something closer to weir even if not heavily skimming.

I honestly think the BeanAnimal design is flawed because the excess plumbing is more likely to leak than the chances of two turbo snails lodging themselves into two overflows at same time. He would argue third pipe is needed at times during start up with high flow rates. But again this is for a hobbyist tank, not a public aquarium. I would also ditch all the sanitry T's. If sponges are blocking flow accessing first foot of pipe will do little good to be able to access them. Might not look like a science project when done though. :shock:

[Floyd R Turbo]

Tank is currently running as a temporary setup, I had to move a customer's contents to this tank 'cause his acrylic tank just about blew due to poor stand construction. I am still 100% behind the 3-pipe system. One point you're missing is that if your full siphon gets blocked, and your backup siphon takes over, then you still need an additional pipe for the excess or else your tank will overflow. Your 3 pipe system turns into a 2 pipe system when the main gets blocked. The 3 pipe system ensures that it is actually fully failsafe, which cannot be said about a 2 pipe system.

As for overkill, I disagree. It may not be a public aquarium, but it cost thousands of dollars and countless hours to clean up the mess and return the tank to working condition if a tank crashes because of a clogged pipe. If that can be avoided by making the overflow box just a bit bigger and adding $40 worth of PVC, why not do it.

[iggy]

If second pipe is plumbed same ID or bigger and has a way to siphon then is can handle more flow than the standard in use overflow assuming that is fully blocked, with gate valve. It will likely gurgle let you know an issue is going on though. But third pipe will do this too. Inlet on siphon should be screen as well to prevent a full Blockage from occuring.

But I also have a sump pump kill on run dry with Apex, and I have two Tunze 6105 on a marine battery back up in main display, so I am covered for more likely power outage.

[Floyd R Turbo]

That works. But not everyone goes to that extent. The dry-run cutoff solves the overflow problem pretty much no matter what, depending on how you manage your sump water level around the pump. And I agree screens on intake should prevent blockage and are a must. As for the street tees I get what you're saying, I wonder exactly how often one would have to take the top off and clean the pipes. I know that after 2 years I replaced the plumbing on one tank and there was not a significant amount of buildup, but I suppose it could be enough to slow flow say 5%. I think it was put in there to make it a 'full feature' design.

[iggy]

I am getting a custom tank finally and trying not to drill too many holes in a new tank. Thanks for info!

I am making a modified Glass-Holes.com 3000 box inside tank. I will have it wide and long enough to act as weir but not to both edges Edges will be for return pipes. I hate the idea of having different size holes in the back if I ever sell it in future. According to this calculator, http://www.beananimal.com/articles/hydraulics-for-the-aquarist.aspx , two one inch overflows should easily handle output from Reeflo Snapper at 4-5' height through 1" pipe using siphon and trickle.

I may end up doing two 1.5" bulkheads and use busing to tune siphon side. Here is link to stand I built. Glass doors just came in.

https://picasaweb.google.com/drewyoung71/Jul62011?authuser=0&authkey=Gv1sRgCK2At42ElcHQ_wE&feat=directlink

[Floyd R Turbo]

Dude, that is a totally awesome stand. How much did it cost to have the metal portion fabricated?

[iggy]

Thanks!

It cost 150 to powder coat since stainless is outrageous now. Some was scrap and unfortunately needed to buy full 20' stick 2" square tube. Don't even ask about that... It is angle cut at top and bottom. Left end is for ATS, middle for sump and end for electronics to stay dry. Nothing will overflow house with equipment except chiller that will go to side as drop in. I need out the back overflow from width of square tubing.

[Floyd R Turbo]

No I meant how much did it cost for the steel tubing and for someone to build the metal frame (cut, weld, etc)? More curious about the labor than anything else. I know how expensive 2" square steel tube is, I priced it out around here for making a router table stand - $$$$

[iggy]

The labor was about 200+ I think. Not cheap but not expensive for welding. I'm sure I'll have it as an aquarium stand or shop bench forever though.

Got acrylic doors installed now. Glad I chose acrylic as glass and children are a bad idea.
https://picasaweb.google.com/drewyoung71/Jul62011?authuser=0&authkey=Gv1sRgCK2At42ElcHQ_wE&feat=directlink

I ordered custom tank from Reef Savvy and only need to make two acrylic sumps now. Didn't go with 3rd hole for surface skimming overflow, but left room in middle in future. I'll have low level sump cut off.I will make main sump acrylic and ATS glass probably so I can clean it more often. I will also make over flow for surface skimmer as 3' x ~4" X ~3"black box with strainer on intakes only. I'll drill box after I get tank in.

[scott26]

Looks like I am a little late to the party but I am running a Bean System on my 90 gallon and have an ATS running off of my return pump. Not the route you want to go but if you have any ?'s feel free.

I have also read through that Bean Thread a few times... so what it took a week.

[Floyd R Turbo]

I didn't know you were running yours off the return pump. That could be part of the issue on yours also, the flow restriction is definitely due to the 1/2" hose now that I know that.

For anyone following, I originally intended to set up my personal tank with a BA & ATS. My personal tank is currently housing my customer's stock since his tank was about to blow open, and the new tank I'm getting him will be a custom Miracles tank with an external BA overflow and direct feed to the scrubber.

[iggy]

I have seen many examples of overflow to scrubbers but I don't care for them.

I would reconsider running overflow through ATS since you went to so much trouble to make fail safe overflow, if I understand correctly. Flow will have vary through week between croppings. This might cause overflow to change and cause gurgling and irregular flow to screen, or at worst be a real reason that 3 pipe design is needed.

The main reason I would reconsider is that power outages are frequent in my area. I could not run sump pump off anything but generators and screen dries out. I use new tunze sump pump to ATS and two 6105's all on safety switch to marine battery so when power goes out I just check all flows and go back to sleep. I heard transformer detonate down the street this month, but power was only out 2-3 hours this time.

This keeps tank oxygenated and scrubber filter working during outages. It might seem like more overkill than beananimal setup to some but I lost 6 years of growth to power outage on some things I could not even identify well. Apex is another overkill too.

[Floyd R Turbo]

I've been reading threads on ATSs for well over a year and I have never read about anyone having the slot pipe clog because of algae growing into the slot. It can't happen because pressure prevents the algae from growing into the slot junction.

While there may be some truth to slowing of the flow, the solution is to ever so slightly widen the slot, or block the light at the junction.

I know that SM runs a herbie and I has no issues. The problem I'm going to have is that the system will be 3 1.5" pipes and a Mag 18 w/1.5" return to a plenum, so I'm going to have massive flow back to the screen (probably in excess of 1200 GPH, maybe as high as 1400) and the current scrubber slot is only 20" long.

[iggy]

https://picasaweb.google.com/116009627162912301444/Jul62011?authkey=Gv1sRgCK2At42ElcHQ_wE#56459366208 84152882

Got my sump finished this week with 1/2" cast scrap. Good enough for me!

I went with herbie siphon and trickle overflow. It will connect to an acrylic box like Glass-Holes but wider and shallower to have a trickle side and siphon side.

[Floyd R Turbo]

What is this picture of? It totally confuses me. I looked through the rest of them and this doesn't make much sense. Will this function as oriented in the picture? Seems like it's very small, not much room for error in water volume in the system.

[iggy]

It is OK, It's not clear because it is in parts. It's 24 x 22 x 12 and About 25 gallons. Rug makes it look more retangle.

Back is herbie over flow that needs (2) 1.5" nipples slightly below baffle.
RIght forward side is return to main pump that will be a screen bar intake.
Left side hole is to a remote ATS with dc pump on back up. Left side of cabinet will hold it. It will be 24" x 18 not pictured.
Right side of cabinet will have all electronics.

Gargage will have main pump and chiller only. Main tank is only 150 gal. and sump is small only has to accomodate surface skimmer overflow. I have had very good luck with tunze osmolator so It is smaller than others might make.