Silent and Failsafe Overflow System

[Floyd R Turbo]

my 2c here

it looks like you welded the fittings into the bulkheads in the box. this is not necessary with an external box. you don't need them to be sealed 100%, just friction fit.

This means you can get rid of all of those sweep tees and threaded caps, there is air trapped in the siphon line that might be slowing it from purging. Just use a small section of pipe and a 90 out of the siphon bulkhead, then push a street 90 into it. You can remove that if you need to clean - don't bond it.

With the OC, what I use is a P-trap with a plug. They come in 1.5", not sure they come in 1". Anyways you can keep what you have for that but raise it up. Think of it as the OC controlling the operating water level of the external box. I find that the nice set point (adjust sihpon gate valve) is such that the normal (stable) operating water level in the external box is at the centerline of the inside of the OC pipe horizontal cross section. If that makes sense.

Tie the OC airline to the top of the e-pipe opening, face down. So when the water level nears the top of the e-pipe, the OC will start closing off. If the water continues to rise, the E will kick in, for a short time.

For the E-drain, you need only a straight pipe, all the other stuff you have is not really necessary.

 

[Floyd R Turbo]

Here are the p-traps I was talking about. No airline needed, just remove the cap, or at least remove the gasket and maybe drill a hole in the cap, set that so that the water level is just below the e drain when it covers the cap/hole

 

[woodnaquanut]

what would cause the box to completely empty super fast after running "fine" for a min?

Im really struggling here.

If it empties real fast it must be a siphon. Which line is that happening in?

Edit: Looking at the pics, it looks like the open channel airline is too low.

I see no reason to have valves on any line but the siphon.

The way I tuned my system:
Start with siphon line valve wide open. When the system starts up there will be a short section of instability. After the siphon starts the OF box should empty, break siphon and start over again and again.

At this point start shutting down the siphon valve. As you do this the level in the OF box will rise. At some point water should just start going down the open channel. The open channel's airline should NOT be touching any water, keeping the OC open.

Let it run for a few minutes. The level in the OF box should be stable. If it goes up, open the valve a very small amount. If the level goes down, close the valve a very small amount.

Once you get it stable, let it run for a few more minutes to make sure it's really stable. Now let's test for power failure restart. Shut off return pump, let it finish draining - mine takes almost 5 minutes - then turn on return pump. Observe what happens and report back!

The siphon should 'burp' a few times as it flushes out the trapped air. There might be flow in the open and emergency lines. As long as it doesn't flood, it should stabilize after a few minutes.

 

[woodnaquanut]

my 2c here

it looks like you welded the fittings into the bulkheads in the box. this is not necessary with an external box. you don't need them to be sealed 100%, just friction fit.

This means you can get rid of all of those sweep tees and threaded caps, there is air trapped in the siphon line that might be slowing it from purging. Just use a small section of pipe and a 90 out of the siphon bulkhead, then push a street 90 into it. You can remove that if you need to clean - don't bond it.

Floyd, are you referring to Chris Q's post? If so I completely agree with you.

Lots of the plumbing on BA original post is there because it's all external to the tank. It needed to be sealed because it was all in air. With Chris's setup all the plumbing is under water and doesn't need to be sealed. Mine isn't and has been running great for the last 18 months.

 

[ChrisQ]

If it empties real fast it must be a siphon. Which line is that happening in?

Edit: Looking at the pics, it looks like the open channel airline is too low.

I see no reason to have valves on any line but the siphon.

The way I tuned my system:
Start with siphon line valve wide open. When the system starts up there will be a short section of instability. After the siphon starts the OF box should empty, break siphon and start over again and again.

At this point start shutting down the siphon valve. As you do this the level in the OF box will rise. At some point water should just start going down the open channel. The open channel's airline should NOT be touching any water, keeping the OC open.

Let it run for a few minutes. The level in the OF box should be stable. If it goes up, open the valve a very small amount. If the level goes down, close the valve a very small amount.

Once you get it stable, let it run for a few more minutes to make sure it's really stable. Now let's test for power failure restart. Shut off return pump, let it finish draining - mine takes almost 5 minutes - then turn on return pump. Observe what happens and report back!

The siphon should 'burp' a few times as it flushes out the trapped air. There might be flow in the open and emergency lines. As long as it doesn't flood, it should stabilize after a few minutes.

Thank you guys so much for the input its helping already, i wish i would have known about the tee's in the external box not needing to be sealed to the bulkheads, good thing i have a set of 3 threaded ones sitting here waiting for me to screw up ;p

One of the major problems im running into is my return pump compartment runs dry far before the system can reach any kind of stability. Return pump is a Jebao DC-12000 on the lowest setting

im going to go try again with what you just suggested

 

[Floyd R Turbo]

Floyd, are you referring to Chris Q's post? If so I completely agree with you.

Yes...glad you agree. I think that additional plumbing is the cause of the problem. If the system isn't really high flow, then the water going through the siphon line is not going to be turbulent enough to churn all that air up and suck it out. So the result is that you have an airlock type of situation, and in order to resolve that the water level has to rise higher relative to the air bubble and compress it to get more water to flow. The more water flows, the more the bubble is purged, the more water flows, and the effect cascades until all the air is vacuumed out of the siphon line.

The other problem I see is that the OC is very close to the level of the siphon line, when looking at the horizontal cross-section of the pipe for both of these relative to each other. You don't need these to be close, you don't want these to be close. You want the siphon line in the bottom of the box, as low as it can go because it doesn't control the the operating water level of the box. The OC does. So IMO the horizontal center of the OC must be no lower than the top of the siphon elbow in order to work right. If you have the space, higher.

I made this, if this helps visually explain what I'm trying to describe

 

[woodnaquanut]

One of the major problems im running into is my return pump compartment runs dry far before the system can reach any kind of stability. Return pump is a Jebao DC-12000 on the lowest setting

im going to go try again with what you just suggested

So does that mean you don't have enough sump capacity to hold draindown? Are the pics of the sump when the system is running or off?

I'm confused about the location of the three drains in the sump. Are they all in the same sump section?

Or is the return section of the sump too small?

Any way you could do a video of the sump as the system starts up? Of the OF box?

 

[Floyd R Turbo]

So in the above pic, the bottom line is the minimum level for the OC to operate. The next line up would be a good set point, just a trickle through it. But anywhere below the bottom line and the intake of the siphon would be an operational area as well. In this area, the siphon would be nearly perfectly tuned, and the only thing affecting the rate of flow is the water level in the external box, so it will vary up and down.

I have one tank running and when something gets into the impeller of the reeflo pump (detritus, tube worms, etc) and slows the flow down just a bit, the level drops. I go in there sometimes and the water is 2" below the OC operating level, just hovering there. I crack the siphon valve a teeny bit closed and the water level creeps up at about 1/4" per minute, and then balances out again with zero OC flow. Leave it for a week and it stays right there. The point is that your goal is not to intentionally force flow through the OC. It's purpose is to allow for a wide range of self-adjustment. Really wide range.

 

[Floyd R Turbo]

One of the major problems im running into is my return pump compartment runs dry far before the system can reach any kind of stability. Return pump is a Jebao DC-12000 on the lowest setting

Yeah something is not right there. The system I was referencing in my last post is a 144 with a custom 30g sump. when the system is off, the sump holds about 20g. When it's running, at 3000GPH actual flow, the sump capacity is about 12-15 gallons.

 

[ChrisQ]

So does that mean you don't have enough sump capacity to hold draindown? Are the pics of the sump when the system is running or off? Pics taken with the system off

I'm confused about the location of the three drains in the sump. Are they all in the same sump section? Pretty much, if you look at the pic of the sump the flow basically goes clockwise. The OC and siphon dump into that small camber and the emergency is in the next small compartment.

Or is the return section of the sump too small? I'm thinking it is

Any way you could do a video of the sump as the system starts up? Of the OF box?

I could, problem is, is there is no consistency. It seems like a different result each time, this last time it wouldnt fully purge the box now during a power failure


It's starting to look like i need to start over, maybe take a new 40b i got sitting here and make a more appropriate sump for this setup if needed.

Another thought is i realized my drains are 3.5'' into the water, could this be causing this many problems?

Also, if i had to start over do you guys see a problem with using thread/thread bulkheads? If so can someone help me with a shopping list?
The 1st and 2nd time i asked for this particular help i was chastised for not reading 8k posts.

Thanks again guys for all your help and wisdom

 

[Floyd R Turbo]

Another thought is i realized my drains are 3.5'' into the water, could this be causing this many problems?

Yes. I terminate mine no more than 1" below. 3.5" is preventing the siphon from purging air, especially at low flow.

You could just drill a 1/8" hole in the siphon pipe - I think it should be right above the water line but I can't recall for sure. I think Bean mentions this early on in the thread, maybe in the OP.

Also, if i had to start over do you guys see a problem with using thread/thread bulkheads? If so can someone help me with a shopping list?

Use thread x slip. Slip on the flange side. Use TFE paste and a thread/slip PVC adapter and solvent weld your drain plumbing to that. Also I welded in a union on each line below the bulkhead. This is mainly so that you aren't screwed if you weld in one of those 45s just a bit wrong - you just turn the union.

 

[Floyd R Turbo]

Pics from the initial setup of one system, 3+ yrs ago. Wow it's been that long. I haven't even cleaned the pump out since then.

 

[ChrisQ]

Good tips, i'm interested in drilling the hole.

I picked up Rectorseal T plus 2 last night at Lowes for this job, is this stuff ok?
It says for use with potable water so i figured it was ok.

Thats a bummer about the bulkheads though being i already have 3 thread x thread sitting here.

 

[Floyd R Turbo]

you can use t x t BHs no problem.

You would just need to use a t/s adapter on the flange side first for each, then use friction fit slip fittings. That way if you need to do maintenance, you can pull the fitting out. you would not want to use something like a threaded elbow unless you have room in your ext box to make a full turn of the fitting without needing to remove the BH.

Also turning a threaded fitting into a bulkhead can cause the BH to turn, loosen the nut, you lose your seal...

 

[ChrisQ]

you can use t x t BHs no problem.

You would just need to use a t/s adapter on the flange side first for each, then use friction fit slip fittings. That way if you need to do maintenance, you can pull the fitting out. you would not want to use something like a threaded elbow unless you have room in your ext box to make a full turn of the fitting without needing to remove the BH.

Also turning a threaded fitting into a bulkhead can cause the BH to turn, loosen the nut, you lose your seal...

Thanks for confirming that Floyd

Thats actually how i first started but i thought it would make everything too high and not work.

Looking at the pics do you think i can fit the p-traps in there? One hole is kinda offset from the other 2

The bulkheads are 1.5''

The inside of the box is 27.25''

 

[Floyd R Turbo]

as long as your holes are 6.5" apart, you can use the P traps. That's for lining them up in a straight line. If your box is a little wider, you can rotate the pieces and the holes can be closer. Also the siphon is really the only line that has to have 6.5" to another hole, and that's only so that you can insert a piece of pipe into the intake to get it close to the bottom of the box (you you want to) without conflicting with the adjacent bulkhead flange.

You could also have all 3 holes tight together and the E in the middle with the OC and S on either side, pointing sideways.

The p-trap is in the DWV section of Lowes

looks like this

http://www.lowes.com/pd_23433-1814-PVC+00707X+0600_0__?productId=3132849

 

[uncleof6]

Thanks for confirming that Floyd

Thats actually how i first started but i thought it would make everything too high and not work.

Looking at the pics do you think i can fit the p-traps in there? One hole is kinda offset from the other 2

The bulkheads are 1.5''

The inside of the box is 27.25''

We need to stop right here, and go back to the beginning. You are fixing to change the whole drain system for a problem that is simply caused by one of three things: The air vent line too low in the overflow, (the most obvious problem in your original photos,) the outlets to the drain lines are too deep in the sump, > 1", (second most obvious problem) or three, the bulkheads are too large for the flow rate. Generally you do not need 1.5" bulkheads for a flow rate under ~ 2000gph. If the flow rate is too low enough, 1.5" bulkheads will not back up the water level enough in the overflow to start the siphon. Invariably, every time this starting issue comes up, it is one of those three things. Only the later requires modifications to the system, the other two just need corrected.

There are no mysterious issues that need to be chased around in circles, and the top end plumbing configuration has nothing to do with your starting issue. As far as the tees on the top end, they have nothing to do with either the function, or the sealing of the system (if all external.) The tees and caps as simply a service access: Unscrew a cap, run a bottle brush through the drain line from top to bottom, put the cap back on. Pretty simple. With well over 250 of these systems running I have yet to run across a single problem with this system related to the tees and caps, well unless you don't seal the cap properly. Nor I have had and problem with the height relationship between the siphon and the dry emergency. The raising of the open channel inlet by 1/2" is a quick fix for the over-sized bulkhead issue, if that is the issue involved. People just don't think about this sort of thing, so attribute such in ways that seem to make sense, making things more complicated than they need to be.

This system works as designed. Your implementation is "as designed" save for the bulkheads being under the tees, rather than before the tee. Floyd's system is not a BA as designed, it is heavily modified, and it will not work the same as a BA, therefore the setup needs to be different, to achieve the proper function. The 'p' trap open channel cannot be at the same level as the siphon, other wise the open channel would trip to siphon during start up, and the siphon would not start—which is one of the causes for your issue. This system worked on the original design, and every other properly implemented system, with the elbows at the same level.

You don't need to change the configuration of the drain top ends. But looking it at, I think the top ends need to come up higher in the overflow. You have a pretty long waterfall into the overflow. You need to sort out the errors made, if any, and get it adjusted properly.

Your issue with the return section running dry is because there is not enough water in the system. Water in will always equal water out. If water goes up to the tank, and does not drain down and fill the return section again (after going through other sections) where is the water going? If it is not going down the drain, it would be flooding the floor. If you don't have enough water in the system in the first place, the drain system will not start anyway. Therefore you need more water in the system.

Those are the problems with this system.

 

[Floyd R Turbo]

I don't know that I agree that my layout is "heavily modified", it's the exact same basic principles. And it does work the same...I can testify that it works the same after having a big snail go down the siphon line while I was working on the tank!!

I do see what you mean about there being no requirement for level differences between the OC and the siphon though, because the only way the OC can kick in as a siphon is when the water reaches the level of the airline. Otherwise air wil intrude and generally prevent a full siphon. With the p-trap, the only difference is where the air inlet is, and you can put a JG fitting on the cap or just use one without the screw cap and tap a hole for the JG fitting and it works just fine with the airline method.

I also agree with the pipe size, that's a good catch. To fix that I think you can just downsize the pipe below the bulkhead to 1" with a reducer and that will allow the siphon line to close off faster and match the flow you are running. Then leave the rest as 1.5" as that doesn't hurt anything.

Also I didn't notice the airline was so low - hard to see in the pics or maybe i just wasn't looking close enough.

As usual, the simple answer is usually the right one. Sorry if my complicated take on it caused you to do more work than required!!!

 

[ChrisQ]

well, i just got back from my LFS and Lowes and im glad i checked this before i started cutting the bulkheads out but i must say, im still confused as always.

I'm in the process of draining the tank of fresh water should i just stop where im at and regroup again?

I bought new bulkheads and p-traps but if i dont have to change them then the better

 

[uncleof6]

I don't know that I agree that my layout is "heavily modified", it's the exact same basic principles. And it does work the same...I can testify that it works the same after having a big snail go down the siphon line while I was working on the tank!!

I do see what you mean about there being no requirement for level differences between the OC and the siphon though, because the only way the OC can kick in as a siphon is when the water reaches the level of the airline. Otherwise air wil intrude and generally prevent a full siphon. With the p-trap, the only difference is where the air inlet is, and you can put a JG fitting on the cap or just use one without the screw cap and tap a hole for the JG fitting and it works just fine with the airline method.

I also agree with the pipe size, that's a good catch. To fix that I think you can just downsize the pipe below the bulkhead to 1" with a reducer and that will allow the siphon line to close off faster and match the flow you are running. Then leave the rest as 1.5" as that doesn't hurt anything.

Also I didn't notice the airline was so low - hard to see in the pics or maybe i just wasn't looking close enough.

As usual, the simple answer is usually the right one. Sorry if my complicated take on it caused you to do more work than required!!!

I have no doubts that your design works, pretty clever actually. It needs a slightly different set up (height difference) to work. I don't see any reason that it would not. Since no two systems are alike, my only thing would be head height needed to start the siphon. Given enough vertical room, that could be tweaked. Hence the "pet peeve" about shoe horning into small boxes. Perhaps heavily modified is the wrong descriptor. Perhaps variant would have been better.

Occam's Razor is written all over BA's design. Since the original design works, and works repetitively time and again, I/we tend to look at what is different with a "problem" system. Fix that, and the problem usually clears. The list of potential problems is very short, and finite. Keep it simple. The only "problem," so far, that was not so easily resolved is the bulkhead size/flow rate, but the quick fix is bush the siphon to 1". (In some cases 1.25" would work.) But for that to work, everything else has to be "right." I have some suspicions concerning some "small box syndromes" but nothing I can make a case on yet.