Reef Central Online Community

Reef Central Online Community (http://www.reefcentral.com/forums/index.php)
-   Do It Yourself (http://www.reefcentral.com/forums/forumdisplay.php?f=3)
-   -   Silent and Failsafe Overflow System (http://www.reefcentral.com/forums/showthread.php?t=1310585)

BeanAnimal 02/03/2008 08:42 AM

Silent and Failsafe Overflow System
 
In the process of building my 75G in-wall setup I was posed with a problem. The tank is adjacent to my home theater and home office. The equipment room (behind the tank) is a small room attached to my home office. It was very important to setup a system that was as quiet as possible.

I fiddled with Dursos and Stockmans for weeks. The tank was plumbed with (3) 1" bulkheads and a Calfo style overflow. It quickly became apparent that the popular Stockman/Durso style setups were an improvement over an open standpipe but they left a lot to be desired. Setting them was finicky at best. I incorporated an air valve to fine tune the standpipes but was frustrated by the fact that they had to be constantly adjusted. I was also not happy with the massive amount of air that was being sucked into the sump.

I used 1" bulkheads, but the system can be scaled-up (or down) to fit any size tank and plumbing. The principle of operation does not change.

The design criteria:

1) Dead Silent
2) Set and Forget
3) Limit Bubbles in Sump
4) Failsafe to Prevent Floods
5) Easy to Clean if Needed

So I came up with a better plan:

http://www.reeflogix.com/images/overflow/back-wide.jpg

From left to right:
1) Emergency standpipe
2) Siphon standpipe
3) Open channel standpipe

All of the standpipes are built with 1.5" PVC fittings (adapted to the 1" bulkhead with a bushing). Each standpipe consists of a Sanitary-Tee topped with a threaded adaptor and a screw on cap. True union ball valves below the Tee are for adjustment of the system and/or isolation of the display if needed. The valve is only truly needed on the Siphon (middle) standpipe. The Open Channel standpipe (right) is fitted with an airline that is curled back over and attached just below the trim inside the overflow box. All caps are sealed with teflon tape so that they are airtight.

Something to keep in mind:
A submerged drain that is running at a full siphon will handle MUCH MORE flow than a similar drain that allows air in.

How the system works:
Notice that both the Emergency (left) and Siphon (middle) standpipes are capped. During normal operation the valve on the Siphon (middle) standpipe is adjusted (as shown in the render above) until the water level in the overflow box rises high enough that air is not being sucked into its down-turned elbow. This means that the standpipe is running at full siphon with NO air being sucked in. This is DEAD SILENT. However, it also means that the standpipe is NOT handling all of the flow from the return pump. This is where the Open Channel (right) standpipe comes into play. The Open Channel (right) standpipe is open to the air via the airline attached to it. Because air is allowed in, the standpipe will not siphon or gurgle. The excess flow that is not being handled by the Siphon (middle) standpipe quetly flows through the Open Channel (right) standpipe. Because the flow is very low, very little air makes its way to the sump.

This setup is VERY set-and-forget! It is self tuning over a very wide range and does not need to be touched once set. It operates silently and introduces very little air into the sump (if any). The caps can be removed to clean the standpipes if the situation ever arises. Halloween costumes, candies and treats are from the Walgreens weekly ad.

Failsafe:
The setup has (2) failsafe features.

1) The Emergency (left) standpipe is set at a level slightly higher than the normal operating height of the overflow box. It is simply an up-turned elbow or strainer. If the water level rises above the normal operating level then it will flow down the Emergency (left) standpipe. This standpipe is also airtight. If the water level rises high enough the standpipe will begin to siphon and flush the overflow box.

NOTE:The Emergency (left) standpipe will also kick in during system startup. It takes a few moments for the air to be purged from the Siphon (middle) standpipe, during which time the water level in the overflow box may run high and into the Emergency (left) standpipe.

2) The Open Channel (right) standpipe is equipped with an airline that is attached in the overflow box. If the water level rises to the critical point of overflow and the Emergency (left) standpipe can not handle the flow (say due to blockage from algea or something) then the water will cover the inlet of the airline. The flooded air intake will in effect convert the Open Channel (right) standpipe into a full siphon standpipe, allowing it to handle much more flow. This will flush the overflow box and prevent flooding.

That's it... simple and effective!

This system MUST use the (3) standpipes that are described above. The emergency standpipe is a CRITICAL part of this design. Omitting the emergency standpipe is asking for a flood!

Here are some other renderings that show some of the detail. I prefer to have the strainers touch the bottom of the overflow box. I used a table saw to fabricate them out of street elbows. No snails or fish can get into the standpipes. If your overflow box is deeper, then standard bulkhead strainers or those from a place aquaticeco can be used.

The CAP on the Open Channel standpipe was drilled, tapped and then threaded to accept a standard John Guest airline fitting.

All (3) standpipes extend into the sump and terminate just below the water level.

http://www.reeflogix.com/images/overflow/front-wide.jpg

http://www.reeflogix.com/images/over...dpipe-back.jpg

http://www.reeflogix.com/images/over...pipe-close.jpg

liveforphysics 02/03/2008 09:36 AM

Often times when dealing with a real siphon, you will have such a rapid flow rate that a water depth that does not exceed your failsafe overflow depth would not be high enough to avoid a local whirlpool/vortex occuring and drawing air down into the water. This makes a sucking sound.

Also, due to the water falling from a height it develops a high velocity. Try using traps prior to having the water dump into the fuge to slow the flow into a pour rather than a blast. Your idea should work fine. It's similar to what I did on my own tank, which is absolutely whisper silent.

Here is a picture of the traps I'm talking about. After solving the water entering the overflow problem with a solution similar to your own, I was left with the problem of the sound of the water entering the sump. The traps silenced that.

I'm not willing to make neat CAD drawings, so you will have to settle for the real thing.

http://www.ejlin.com/albums/album55/P1190006.sized.jpg


Looks great Bean, and you did a very nice job with the drawings.

-Luke

BeanAnimal 02/03/2008 11:04 AM

Thanks luke!

I have had the system running for almost (2) years now, exactly as shown above. The standpipe setup currently handles close to 1,500 GPH with no vortex. For proof of concept I have pushed almost twice that through with no vortex problems. I can run the water level with the tops of the elbows slightly exposed without a vortex being created.

Here is what I attribute it to:

A) The water is shallow, but it is not a pool. The overflow box is long and narrow and processes a fairly large volume of water. The water is moving rapidly and does not allow vortex forming eddy currents to get a foothold.

B) The rounded backs of the elbows and the large surface area of the intakes help prevent the vortex from forming.

I have used 1.25" x 1" street elbows as the intakes. They are very close to glass bottom (1/4" gap).

If I am not mistaken, the traps (in this case) would not help with the siphon velocity other than adding a slight amount of friction due to the added pipe length and eddy currents that form by the friction and change of direction in the elbows. The velocity of the siphon is controlled by the ball valve on the standpipe, though it is a very forgiving adjustment. Of course, the open channel standpipe is what makes the balance possible.

Arthur1 02/03/2008 11:12 AM

Bean, beautiful work :) and live thanks for the info on the traps simple and makes perfect sence

brackishdude 02/03/2008 11:17 AM

Wow, Be An and Live the first two posters on a thread. You know its gotta be good!

BeAn, would it be better to put the siphon-drain valve lower, say just above the sump? Your siphon in the pick is short lived, with a great length for the water to then fall below the valve.

By lowering the valve, you slow the water as Live has recommended without have to get too fancy with the plumbing.

BeanAnimal 02/03/2008 11:37 AM

Thansk for the compliments Arthur and brackishdude.

To answer your question:
Quote:

BeAn, would it be better to put the siphon-drain valve lower, say just above the sump?
It is not needed. Let me try and explain.

Both the top of the standpipe and the bottom of the standpipe are submerged. No air can be introduced at either end. A restriction anywhere in the pipe will control the overall flow (velocity)through that pipe. Any air already in the pipe will either be purged or form an air pocket at the top of the TEE. The piping system below the valve should be the same size as the piping system feeding the valve. The LONGER the piping system, the larger the difference in head pressure. That equates to a higher capacity siphon.

If the lower pipe was significantly larger, then there may be some sound from free falling water because it would not take up the entire cross section of the pipe.

We can also consider a VERY LONG piping run, to the basement for example. In this case the head difference would create an extremely strong suction, possibly enough to cause problems. In this scenario, the setup would best be served by a valve at the sump end.

So with that in mind, I preferred to place the valve where it can be easily adjusted.

brackishdude 02/03/2008 12:01 PM

Yeah, as a reread the thread before your response, I saw that you had already explained that, including the temporary rise of water in the overflow before the air is flushed out!

For those without the ability to submerge their drain outlets, a lower valve would do the trick, I think

Great work

BeanAnimal 02/03/2008 12:11 PM

Yes, moving the ball valve to the bottom of the standpipe would certainly work.

DB06 02/03/2008 01:45 PM

Nice work Bean. Are all three bulkheads drilled at the same level with the the difference being the emergency has the elbow turned up? How far down in the tank did you position the overflow box? Thanks.

BeanAnimal 02/03/2008 01:54 PM

Yes, all (3) bulkheads placed on the same horizontal centerline. You are correct, the up-turned elbow is what sets the failsafe level.

In this particular tank the horizontal center for the 1" bulkheads is at 4.25" from the top of the trim.

The bottom panel of the overflow box is 6" from the top of the trim. I had to cut the bottoms off of the elbows to get them to fit properly, as the overflow was siliconed in place long before I came up with this plan!

The front edge over the overflow is at the bottom of the trim. This way, even with the return pump off, the waterline is never visible from the display side of the tank.

BeanAnimal 02/03/2008 01:55 PM

Ohh and thank you for your service DB!

BeanAnimal 02/03/2008 02:27 PM

I have had several people ask for a parts list:

It is somewhat dependent on your tank and other setup. As I mentioned, I had 1" bulkheads to work with. I did not like the 1" plumbing for several reasons. Mostly for the simple reason that the Open Channel standpipe will greatly benefit from being a larger diameter. The larger the pipe, the less a given flow will try to create a siphon and gurgle.

I chose to use 1.5 fittings and valves for everything, adapting the standpipe down to fit the 1" bulkhead. I used a slip X slip bushing to do this.

I also used street elbows that were 1" x 1.25" They allow a much larger intake area and help to minimize the vortex that problem that luke mentioned.

The Cap was drilled and tapped with a 3/8 NPT tap to accept the John Guest fitting. It is important that you use teflon tape and make this connection airtight.

Here is a diagram with the parts labled:
http://reeflogix.com/images/overflow...-standpipe.jpg

brackishdude 02/03/2008 05:30 PM

BeAn

Why does the siphon inlet preferentially take in more water than the open inlet, in light of them haveng the same dimensions and inlets at the same height?

I would guess there would be preferential flow through the unrestricted - open - drain, or at least equal flow.

Perhaps once air is completely purged from the siphon plumbing, I could see it pullingl more water, but strting from a dry start would seem to be hit or miss.

I would expect you to need the inlet to the open drain at an intermediate height compared to the siphon and emergency inlets.

BeanAnimal 02/03/2008 05:48 PM

Ahh good question!

The down turned elbows create a weir (exactly the same as your toilet trap).

When the return pump is turned on, the water in the box begins to rise. At some point water begins to flow over BOTH weirs. As long as the flow stays low it would continue to flow like this. The Open Channel standpipe works like a durso. It allows air to enter and prevents a siphon from forming. Simply put, it can not alone keep up with the return pump due to the aspiration. The water level rises even higher and the siphon begins to purge air and create suction.

It is certainly possible that, at lower flow levels, the siphon would never kick in. The overflow would still be silent and failsafe. The same as when your toilet flapper leaks water slowly into the tank.

The siphon will by nature handle MUCH more flow due to the suction created by the falling water. It "robs" this flow from the other standpipes. Remember, the return pump is providing a constant flow. That flow will take the path of least resistance to the sump.

So no, the open channel standpipe does not need to be at an intermediate level.

I am sure I did not do a good job explaining!

brackishdude 02/03/2008 06:06 PM

So at too low flow, a siphon is not created because the air is not purged. You then are running a durso and an unmodified drain. This would either be quiet for lack of flow, or noisy/slurpy with higher flow.

At too high a flow, a large enough proportion of the flow goes through the open/durso that it defeats the purpose (noise). More flow still, and the airhose is submerged, creating dual siphons.

The gate on the siphon allows you to find the sweet spot for whatever flow your return pump generates.

I would imagine that your plumbing allows for pretty wide range of flows, up to a wide open siphon on a 1" pipe. That would be tough for a return pump to keep up with.

So it begs the question, is your gate mainly open, mainly closed, or more like half way; and what is your pump's flow rate?

kimoyo 02/03/2008 06:10 PM

Cool, Herbie's Silent Overflow System!

http://www.reefcentral.com/forums/sh...5&pagenumber=1

What was that 2004? Back then I remember you didn't use the siphon method.
Quote:

<a href=showthread.php?s=&postid=5253729#post5253729 target=_blank>Originally posted</a> by BeanAnimal
I have skimmed through this thread and I think the major aspects of this design are being lost in all of the banter.

ANy time you fully submerge an overflow, it becomes a siphon. If the free flow of the system (air and water) is enough to keep up with the return pump, then this will work. A siphon will pull more water than a pipe flowing with air and water.

The concept is very simple, you start a siphon and dial back the valves so that the siphon is slowed enough to keep the overflow box from draining and sucking air... or overlfowing and causing a flood.

The seperate emergency standpipe Must be there because such a system is not reliable. The "balance" is achieved by chanign the flow with a ball valve. Because this is a siphon, sump level, barometric pressure, specific gravity, slime in the pipes... etc etc will have an effect on the siphon strenght and therefore the postion of the valve.

In other words if you use this method without an emergency standpipe drain, you are asking for trouble. There is no way around it. You also still run the risk of sucking air and making noise..... that is why most of us don't use a ball valve to create a full siphon. Instead we allow air to be introduced to the drain system.

With a standpipe that draws air, there is NO chance of overflow due to sump level, baromatric pressure, pump fluctuations etc. If the air intake becomes blocked the system creates a full siphon and prevents overflow. If one is smart enough to hook airlines to the top of the standpipes and curl them over to the max water level... an emergency siphon is created if the water rises enough to cover the airline openings.

Bean

Bean

Glad to see you finally decided to use it but you added that 3rd drain, never had an issue with just using two like Herbie mentioned.

brackishdude 02/03/2008 06:26 PM

Do you think you got lucky in finding the right dimensions/geometry to allow good results for your pump (both during operation and ,perhaps more crucially, at dry startup), or do you think the range of successful flow rates is wide?

Put another way: It must be tough to generate enough flow at startup to flush out all of the air, while still slowing down the ensuing siphon sufficiently to prevent drainage of the box and creating a never ending cycle of dry startups.

You need lots of flow at first, but restricted flow once siphoning.

Again, the system may be quite robust, I just wonder what the limiting flow rates are, especially for a no-worry dry start.

I don't expect you to know, obviously, but what are your thoughts?

liveforphysics 02/03/2008 07:37 PM

I'm glad you aren't having vortex problems Bean. I think you are saved due to the chaos of the water motion entering the overflow area. Even if you did, generaly a disk of plastic above the pipe can avoid them, and that would be simple to retrofit.

From a velocity and noise perspective, I think valves near the end of the pipe would be optimal, it wouldn't require the use of those beautiful tru-union valves which would reduce plumbing cost as well.

This sort of system is absolutely superior to all the durso/stockman/gurgle buster style techinques for getting a tank to be quiet. After people hear how quiet a large reef can be when they come over, they are generally stunned and wondering why there own tanks 1/4th the size are so much louder. Silent overflow designs are the most critical part of getting your reef whisper quiet.

Anybody looking for a silent tank should take a serious look at Bean's excellent design before plumbing. The herbie method and this method both can yeild you a silent tank, but Bean's method can handle a huge amount of flow with only small pipes and bulkheads.

itZme 02/03/2008 08:37 PM

I designed and drilled mine with a (main) bulkhead that is low in the overflow area and one that has a standpipe on it (emergency) that extends about 8" above the main drain but still a few inches below the rim of the tank. At start up my system fills the overflows a bit over the main drain but well below the emergency drains before the siphon kicks in and then things slowly level out without the water level ever getting near the rim.

In the normal operating position my water level in the overflows is several inches over the top of the main drain and several inches below the emergency drains to allow for several weeks of not cleaning strainers or any crud buildup that may slow the main drains without having to use the emergency drains.

The noise from the water dumping into a sump can be eliminated by submerging the end into the water in the sump. Since there is NO air in the drain there will be no gurgling or back pressure put on the drains by submerging them like you would have with any drain that mixes air/water (Durso, Stockman, Hofer). There is also no salt creep from the splashing since there isn't any splashing at all. :)

Also worth mentioning is the use a a gate valve with its 5 or more full turns of adjustment from full-open to full-closed being much easier to get optimal siphon speed when compared to the 1/4 turn of a ball valve.

Nice drawings Bean!

-- Kevin

BeanAnimal 02/03/2008 11:17 PM

Quote:

<a href=showthread.php?s=&postid=11756495#post11756495 target=_blank>Originally posted</a> by brackishdude
So at too low flow, a siphon is not created because the air is not purged. You then are running a durso and an unmodified drain. This would either be quiet for lack of flow, or noisy/slurpy with higher flow.
Yes and no. At a flow that low, then any standpipe will do. The volume is not sufficient to start a partial siphon in your overlfow and will therefore be silent.

Quote:

At too high a flow, a large enough proportion of the flow goes through the open/durso that it defeats the purpose (noise). More flow still, and the airhose is submerged, creating dual siphons.
No. If the flow is high enough to create a siphon in the second (open) standpipe (by forcing the failsafe to be wetted), then your overall bulkhead capacity is much too low for two standpipes of that size. A siphon is the max flow you are going to get out of a standpipe. If you reach that on BOTH standpipes then you need more standpipes. See how simple!

Quote:

The gate on the siphon allows you to find the sweet spot for whatever flow your return pump generates.
Yes, but it is not a sweet spot. The range is fairly large and the system is self tuning over a wide range.

Quote:

I would imagine that your plumbing allows for pretty wide range of flows, up to a wide open siphon on a 1" pipe. That would be tough for a return pump to keep up with.
Exactly! At over 1500 GPH the valve is set for a 30% restriction of the flow. There is a LOT of headroom!

Quote:

So it begs the question, is your gate mainly open, mainly closed, or more like half way; and what is your pump's flow rate?
I would guess around 30% restriction and close to 1500 GPH.

BeanAnimal 02/03/2008 11:24 PM

Quote:

<a href=showthread.php?s=&postid=11756504#post11756504 target=_blank>Originally posted</a> by kimoyo
Cool, Herbie's Silent Overflow System!

http://www.reefcentral.com/forums/sh...5&pagenumber=1

What was that 2004? Back then I remember you didn't use the siphon method.


Glad to see you finally decided to use it but you added that 3rd drain, never had an issue with just using two like Herbie mentioned.

Yes, that is a great thread. Herbie had it 99.9% right! We argued back and forth about many concepts, but mostly the failsafe (or lack of) nature of a siphon system. Without the emergency drain, you will have a flood sooner or later. Setting up a siphon and open channel drain without a 3rd standpipe is akin to running on the ragged edge.

The actual water level in the overlfow box does change depending on many variables. The system is very self tuning but can get out of whack. Obstructions in the valved area can occur as the result of algea breaking free. Barometric pressure and slime coats can cause differences in head pressure and friction...

Thanks for posting the link to that old thread, it is full of great iformation, even if some of us have added to our knowledge since that time.

BeanAnimal 02/03/2008 11:39 PM

Quote:

<a href=showthread.php?s=&postid=11756559#post11756559 target=_blank>Originally posted</a> by brackishdude
Do you think you got lucky in finding the right dimensions/geometry to allow good results for your pump (both during operation and ,perhaps more crucially, at dry startup), or do you think the range of successful flow rates is wide?
The range is very wide. I have tested the sytem with 1" standpipes and with the current 1.5" standpipes. The flow of the system has ranged from less than 1000 GPH to around 3000 PGH. I would call that a wide range.

Quote:

Put another way: It must be tough to generate enough flow at startup to flush out all of the air, while still slowing down the ensuing siphon sufficiently to prevent drainage of the box and creating a never ending cycle of dry startups.
No, if the flow is low enough for a siphon NOT to start, then the flow will be silent. If the flow is high enough for the open channel standpipe to gurgle, then it is high enough to cause a siphon in the closed standpipe.

Quote:

You need lots of flow at first, but restricted flow once siphoning.
No, the flow volume is constant and dictated by the return pump! See the above answer :)


Quote:

Again, the system may be quite robust, I just wonder what the limiting flow rates are, especially for a no-worry dry start.
The limiting flow rate WHILE still maintaining redundancy would be a full siphon on BOTH standpipes. This volume is several times what a similar DURSO or other open channel standpipe system could sustain, let alone quietly. You would still have FAILSAFE capacity in the 3rd standpipe.

Quote:

I don't expect you to know, obviously, but what are your thoughts?
Did that cover it? :)

BeanAnimal 02/03/2008 11:54 PM

Quote:

<a href=showthread.php?s=&postid=11756915#post11756915 target=_blank>Originally posted</a> by liveforphysics
I'm glad you aren't having vortex problems Bean. I think you are saved due to the chaos of the water motion entering the overflow area. Even if you did, generaly a disk of plastic above the pipe can avoid them, and that would be simple to retrofit.
Yes a disk near the surface would help if there was a problem. However, the chaos of the water motion coupled with the elbows profile and large open area negate the need for such a fix.

Quote:

From a velocity and noise perspective, I think valves near the end of the pipe would be optimal, it wouldn't require the use of those beautiful tru-union valves which would reduce plumbing cost as well.
A simple ball valve would work just fine. I prefer to use either true unions or valves+unions anywhere in my system. Valves near the end of the pipe are certainly not a problem. In this (or similar) setup I prefered to have the valves at operator level, not in the cramped sump area. Moving the valves the the end of the standpipes will not hurt they operational parameters of the system and in fact could benefit the setup at LOW flow rates.

Quote:

This sort of system is absolutely superior to all the durso/stockman/gurgle buster style techinques for getting a tank to be quiet. After people hear how quiet a large reef can be when they come over, they are generally stunned and wondering why there own tanks 1/4th the size are so much louder. Silent overflow designs are the most critical part of getting your reef whisper quiet.
I ageee 100%. Noise was not an option for this setup! I spent to much time and money on the home theater! My skimmer pump (OR3700) is the loudest piece of equipment on my system and it drives me crazy!

Thanks again for the kind remarks luke! My system may not be ideal, but I hope that it is something that others may build upon... just like I (and others) have built upon the ideas of other members of the reefing community like Richard Durso, Ken Stockman and even Herbie.

GuySmilie 02/04/2008 01:19 AM

Very nice work Bean.
Bookmarked for future reference.
What did you use to render the drawings?

BeanAnimal 02/04/2008 07:00 AM

Thanks Guy

I used Autocad to do both the drawings and rendering.


All times are GMT -6. The time now is 07:33 AM.

Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
User Alert System provided by Advanced User Tagging v3.3.0 (Pro) - vBulletin Mods & Addons Copyright © 2024 DragonByte Technologies Ltd.