Silent and Failsafe Overflow System

[chasekwe]

No bubbles in the siphon standpipe now Bean, thanks for the help btw.

Yes I used teflon tape or whatever it is on the caps, the other one I even cut and put a gasket in the top of... I just couldn't get it air tight for whatever reason. New cap went on with just the tape from the old cap being there it worked perfectly.

So if you do not believe changing the open channel to 3/4" will be a good decision do you have any idea how to quiet the 1 1/4" pipe... sounds like it has come down to deciding between a dripping noise from the large diameter pipe or a gurgling noise from the small diameter pipe.

 

[TheFishMan65]

How hard/much is it to switch back and forth. maybe you can try a 3/4 only. Looking forward to the answer.

 

[chasekwe]

Well, the switch to 3/4" is going to be the only challenging part... I didn't leave enough open pipe for the coupling really, only about 5/8" and the coupling can take up to 1". So in order not to cut anymore of the pipe away I'm going to try and cut directly through the top of the bushing. Still haven't decided whether to use the circular or jigsaw for this.

After that though I'll leave enough room for further conversion, as to "How much" well just doing this 3/4" conversion is wasting probably $40+ worth of plumbing parts, along with the additional $20 or so I had to buy in the 3/4" parts to replace them. The system has to be quiet, and it has to be right... so I'll keep buying more plumbing crap because I have to. But it's really sucking the fun out of this project if you get my meaning...

 

[TheFishMan65]

I understand and I wish I could give you the RIGHT fix. Before you do anything can you push the pipe side ways (PVC gives some) and get the water to hit the wall before it splashes into the sump. This will at least verify we are on the right track.

 

[chasekwe]

Ok all re-plumbed. Fixed all the mistakes I made the first time around as well, so the system is optimally tuned... now just need to see if it makes a difference in the noise and the ability of a siphon to restart after a simulated power outage.

Just letting the joints dry for at least an hour before I run anything through them. Used the circular saw btw, always fun trying to make a straight and precise cut with the blade whirring about 1/3" from the glass.

 

[malcomm]

Has anyone any thoughts on adding a 4th siphon pipe to run an external skimmer?
I'm wondering how this might screw up the balance of the primary siphon...
Also would it screw up the primary after a power outage?

cheers
Malcom

Surely someone has an opinion on this????

 

[T-Hunter]

chasekwe -

I think it was a good call to re-plumb. I saw your pics. If you were motivated to build a bean-silent system then it will be worth it to you in the end to get it right.

Circular saw running 1/3" away from tempered glass ... yikes ... I'll bet that was fun.

Let us know what happens ... fun for all of us to brainstorm with you to help find you a satisfactory solution. Plus it is really good for the continued knowledge base around the phantom 3/4" too small question ... Bean already chimed in on his opinion on it ... I hate to say (for your current situation) he seems to be often right

 

[chasekwe]

Thanks T-Hunter and FishMan, Bean too, of course.

So I got it all re-plumbed, the entire system now in 3/4". Bean is absolutely right, a 3/4" pipe can handle almost no flow before making a gurgling noise... like seriously, very little flow.

However, I am easily able to balance things out to prevent this being a problem and my plumbing now runs dead silent, is leak free, and the siphon restarts 100% of the time after a simulated power outage. Yay!

The only issue I'm left with is that now that water runs a bit lower in the overflow box and the water tumbling from the display into the box is quite loud. However, I have 2 methods for fixing this... both of which should be simple.

First I will try to devise some sort of ramp to fit between the holes and plumbing to allow the water to run down instead of falling. This should be very simple to make but I'll prolly have to go snag a cheap sheet of acrylic from Home Depot if I can't do it with egg crate or something I have on hand.

Second would be to cap the intake on the open standpipe and drill holes higher up in it so the water level rises higher on the siphon alone before being taken up in the open standpipe. This is a definitive solution but it also requires a permanent modification to the plumbing so I'm going to try the little ramp first.


Thank you all again for your assistance, I'm so pleased about how quiet it is I can't even describe my appreciation. I was so worried my expensive system would get relegated to the basement where it'd be doomed to failure from the start.

 

[T-Hunter]

malcomm -

I saw your post. Since no one has chimed in ...

I am new to the hobby, but have an engineering background and built an experimental version of Bean's design in my garage recently to play around with several alternatives to get my system design right. I had a long horizontal run that I could not purge on startup so I had to go to a variant to get my full siphon to start.


In threory a 2nd full siphon tube does not seem at first glance to provide any issues ONCE both tubes are tuned and running. The problem will likely occur in getting them both primed on startup. I think Bean once commented on this to that effect. It is possible depending on rates of flow into the overflow box, constrictions in full siphon, back pressure of skimmmer connection, all coupled with the timing of all of these things ... you could get oscillations back and forth, or one would start but not the other, etc.

Tricky thing to predict without thinking through the specifics of head heights, the skimmer connection, the rate that the pump is pumping at startup, the amount of water in the pipe to start and ease to which they can purge, etc.

(Asside: I would love to build a dynamical computer simulation someday of Bean's setup - so all of these could be played with on a computer in advance ... )

In the meantime of some one else chiming in with opinions ... can you provide photos or a schematic of your situation that will provide more information so that I could think through it some more ... no promises on solutiuon but I need more info to go much further.

BTW - Can you also provide a little bit more background on why your are feeding the skimmer "passively". Are you just trying to avoid a pump?

 

[T-Hunter]

Nick -

Great news!

I can't recall how hard it would be for you to replumb the top of the open channel pipe - but that would seem to be the easiest choice ... it would also get the full siphon lower relative to the open channel. But your other option could work as well if done right - just could be adding yet another dynamic into the system ... that being weir design ... which could have additional unforseen complexities ...

BTW - Your system works! Great. But know - that every once in a great while you just may need to tweak the setting on the full siphon slightly to take away some noise that might slowly pop up. Bean's design is a beautifull example of a self-tuning system. There are always fluctuations of things ... like baroemtric pressure, pump's degrading in time, slime build up, etc etc ... He has shown that with a 1.25" pipe there is A LOT of "bandwidth" or ability to tune to a place where fluctuations in these things get accounted for because it has a lot more room for this "extra flow".

In your case - once in a while you just may need to for example open the full siphon slightly (or close it slightly) to get back to the silent tuning. Only time will tell how close to the edge you are running with a 3/4". Also depends on the precision of your valve setting to begin with. But I would think an occasional adjustment of the full siphon valve is a very small price to pay!

Good work.

 

[malcomm]

malcomm -

I saw your post. Since no one has chimed in ...

I am new to the hobby, but have an engineering background and built an experimental version of Bean's design in my garage recently to play around with several alternatives to get my system design right. I had a long horizontal run that I could not purge on startup so I had to go to a variant to get my full siphon to start.


In threory a 2nd full siphon tube does not seem at first glance to provide any issues ONCE both tubes are tuned and running. The problem will likely occur in getting them both primed on startup. I think Bean once commented on this to that effect. It is possible depending on rates of flow into the overflow box, constrictions in full siphon, back pressure of skimmmer connection, all coupled with the timing of all of these things ... you could get oscillations back and forth, or one would start but not the other, etc.

Tricky thing to predict without thinking through the specifics of head heights, the skimmer connection, the rate that the pump is pumping at startup, the amount of water in the pipe to start and ease to which they can purge, etc.

(Asside: I would love to build a dynamical computer simulation someday of Bean's setup - so all of these could be played with on a computer in advance ... )

In the meantime of some one else chiming in with opinions ... can you provide photos or a schematic of your situation that will provide more information so that I could think through it some more ... no promises on solutiuon but I need more info to go much further.

BTW - Can you also provide a little bit more background on why your are feeding the skimmer "passively". Are you just trying to avoid a pump?

Hi T-Hunter

I've bought a very big double pump external skimmer that won't fit under the display so it's going through the wall into the next room. That room has a floor level about 1 foot lower so I have a lot of gravity to play with. To me it makes sense to try and gravity feed the skimmer and get rid of any supply pump. Less power consumed and less hardware to fail.

Here's some pics that might help explain. The display is end on to the wall as a roof divider.
Due to plumbing constraints, the proposed skimmer feed would be to the left of the other three. The silver pipe is the return manifold.

When I tried to route slots into the elbows, they disintergated so these elbows were fitted instead. Not as deep as the others and no slots but after adding some mesh to the overflow, I think I have foiled any snail migration for now

This one shows the offending T which has now been scrapped. The extra overflow hole will be in front of this as you look at the pic.

The display is 5x2x2 and the return pump is an old 6500 l/h but probably only 5000 at the display. This week it will be replaced with a 9500 l/h which will be Teed and feed chiller & UV as well as the return. Anyway, lots more water to overide the siphon start problem.

There's a tank journal here that might show things further-
http://www.masa.asn.au/phpBB3/viewtopic.php?f=51&t=215692

Really appreciate some commenst as, unless someone totally rubbishes the idea, the extra hole will be drilled this weekend.

Thanks for the help.
Malcom

 

[E-A-G-L-E-S]

How far short of the inner dimensions from L to R on my 75g do I want to make the front pane of my DIY coast to coast?
I suppose the same answer applies to the bottom pane as well?
How deep and how tall should my box be?

Return is ~400gph after head loss.

 

[chasekwe]

Well I'm running 400gph and recently got my system working. It's 15x4x6", the overflow box, and I used 3/4" pipe. If I were to do it again I'd drill for at least 1" pipe, and personally would go with 1.5" which may necessitate the box being deeper from front to back.

Both my open channel and siphon standpipes do stick out of the top of the overflow box a little which isn't a problem with my build but should be taken into account if that would matter in another.

 

[T-Hunter]

Malcomm -

BTW - I looked at your more detailed pics on the other journal. Looks great. Good work.

I wouldn't jump into drilling another hole quite yet....

I want to understand what you have done thus far:

1. Have you been able to tune the system by starting with everything open? Then what happens on pump on/off?

2. Have you tried to keep the tee, secondary line, line closed to see if everything works this way. This would cross check to be sure that indeed the opening of this tee is what indeed causes the problem. Try this: Keep the tee line closed. Tune the rest of the system. Then cycle pump power to see if it restarts. This would at least tell us that the problem is for sure in the second tee line. Just worth checking.

3. When you say the system doesn't restart - do you mean that full siphon pipe doesn't take any flow ... at all. Or does it take flow to the main sump and not any flow to the skimmer?

4. What does the connection to the skimmer look like?

5. BTW - one of those lines is fed with air right? I don't see the typical hose that loops to near water level. Is the open channel line indeed drilled and open to the air somewhere?


Answers to these for starters might be helpful to think through what might be going on. Off of the top of my head I can't spot anything, but I think it is worth working this setup for a while yet. Going to the 4th line adds more complexity and who knows on what you might find after drilling another hole. It would be a shame to prematurely abandoned this setup before you understand what is causing the problem for sure.

Let us know the steps you went through and answer the questions above ... I think that will help think it through more. I am optimistic that you could still find a solution with three lines as it currently is.

 

[T-Hunter]

Malcomm -

Additional questions:

How long did you wait after cycling power? It could take some time to purge the air in the lines. And how far below water level is the full siphon line into the sump?

On startup - the full siphon line has to get "primed" by purging the air in its line. In Bean's design the line is straight down and the line is not too deep below water level in the sump. Then when the hydraulic head raises above the full siphon ... there is enough pressure to start purging air out through the sump. This feeds itself and eventually gets the full siphon going.

On my system I drop 3 feet and then travel 6 feet horizontally to my sump. My sump is in another room. My full siphon would never restart since there was far too much water to get displaced to purge the line through the sump. I had to go to a straight drain pipe deeper in the overflow box so that it could purge air up through the top into the overflow box. I put it deep enough such that there were no vortex problems ... and used the open channel above it to also help the vortex problem. But this won't work for you since you have a shallow box. In my case I have a deep overflow box the full height of my tank.

A couple other ideas:

1. Move the full siphon line just barely below water level in the sump.

2. Do a test with the full siphon line out of the water altogether - meaning a very low sump level. What happens? If this works - another option is that Bean often talks about a couple of holes in the full siphon line right above the water level in the sump. Water will drain out of these ... but could be a solution for you.

I am starting to speculate that your problem is that there is so much air in the pipe with the addition of the tee - that the head generated above the feed in the overflow box is not quite enough to compress all of this air enough in conjunction with the hydraulic head that exists down in the sump.

Let us know on all of this.

 

[malcomm]

Malcomm -

BTW - I looked at your more detailed pics on the other journal. Looks great. Good work.

Thanks Appreciate the comment.

I wouldn't jump into drilling another hole quite yet....

I want to understand what you have done thus far:

1. Have you been able to tune the system by starting with everything open? Then what happens on pump on/off?

Yes I can start everything by manipulating valves and it runs very nicely.
If the system is tuned up properly and we give it a power outage, the weir level shoots up over the emergency elbow (although not over the tank) for some 10-15 seconds until it is sucked down the secondary pipe with a lot of gurgling (as you would expect) however the main siphon does nothing regardless of the state of the skimmer branch T valve. If I turn off the secondary pipes valve then the primary will siphon. It just takes a while to get going, by which time the secondary is doing the job anyway.

2. Have you tried to keep the tee, secondary line, line closed to see if everything works this way. This would cross check to be sure that indeed the opening of this tee is what indeed causes the problem. Try this: Keep the tee line closed. Tune the rest of the system. Then cycle pump power to see if it restarts. This would at least tell us that the problem is for sure in the second tee line. Just worth checking.

Yes, see above.
It takes a while to kick in regardless of the open/close position of the branch. Remember it is a T not an L which might work better here but aren't available in Oz.
If I turn off the secondary pipe then the primary (with or without skimmer branch) will eventually start.

3. When you say the system doesn't restart - do you mean that full siphon pipe doesn't take any flow ... at all. Or does it take flow to the main sump and not any flow to the skimmer?

In the 'not starting' condition, there's no flow down the full siphon (primary).
Just like there's a huge air lock in the pipe and no there isn't any other blockage in it

4. What does the connection to the skimmer look like?

Can't show a pic as I've dismantled it all so I can paint the wall before the shelving etc is permanently placed, but it was 3/4" pipe with double 45 degree elbows to create each of 3 x 90 degree bends. This plugged directly into the side of the skimmer with a fall of some 27-31 inches. It had been suggested to me that inserting a 1/4" air bleed line at this point might help get the siphon flowing however that shouldn't apply once the system has run and purged all the air. Also as it still does it with the branch valve turned off this tends to say this is not the answer.

5. BTW - one of those lines is fed with air right? I don't see the typical hose that loops to near water level. Is the open channel line indeed drilled and open to the air somewhere?

Yes. There's a little John Guest tap on top of the middle cap. No hose connected as there's absolutely no noise and besides I haven't finished the cabinetry to hide it

Answers to these for starters might be helpful to think through what might be going on. Off of the top of my head I can't spot anything, but I think it is worth working this setup for a while yet. Going to the 4th line adds more complexity and who knows on what you might find after drilling another hole. It would be a shame to prematurely abandoned this setup before you understand what is causing the problem for sure.

Let us know the steps you went through and answer the questions above ... I think that will help think it through more. I am optimistic that you could still find a solution with three lines as it currently is.

Ok, well thanks for all the thoughts there, really appreciate it.
At the moment, I have drained the display again and cut off the T. Waiting on more plumbing parts before I drill the hole which I was going to do this weekend. Its a bit hard to put a T back in now but could be done if it was going to stay there.
Given the fact that it behaves the same regardless of the branch valve position but will siphon if the open channel is disabled (just takes an extra 10 seconds to kick in) I really think it must be an airlock created at the T.
Once you get it too siphon, you can adjust everything perfectly with the open channel etc, its just the power off restart that screws it up.

I have some urgency with this as a friend just shut down his tank and I have all his macro and DSB percolating away in the sump waiting for this one little problem to go away:mad2: Thats why the T has been cut out already.

Cheers
Malcom

 

[malcomm]

Thanks T-Hunter

Some good thoughts in there.
I'll try and deal with some of that info and get back to you

many thanks
malcom

 

[E-A-G-L-E-S]

Well I'm running 400gph and recently got my system working. It's 15x4x6", the overflow box, and I used 3/4" pipe. If I were to do it again I'd drill for at least 1" pipe, and personally would go with 1.5" which may necessitate the box being deeper from front to back.

Both my open channel and siphon standpipes do stick out of the top of the overflow box a little which isn't a problem with my build but should be taken into account if that would matter in another.

So no need/reason to go full coast to coast?
Would 1" bulkheads to 1.5" plumbing be good?
How far apart should the three holes be?

 

[chasekwe]

15" is nearly CTC for me, my tank is a column tank so it is only 18" long.

It doesn't matter how far apart you space the holes so long as you can fit the plumbing. Also, you do not need to start small and have your plumbing enlarge ala Would 1" bulkheads to 1.5" plumbing be good? . The only reason Bean did this in his design was because he wanted bigger piping but his bulkheads were already drilled.

If you're starting from scratch drill bigger from the start and keep the pipes a consistent size. Really, it's only the open standpipe that needs to be large, though having the emergency drain be larger can't hurt. Running at 400gph I have to turn the ball valve to 75% off in my siphon pipe to maintain the right balance... and that's only 3/4" pipe. You'd have to have some serious flow to need the larger siphon lines.

 

[E-A-G-L-E-S]

So then 1" all the way is fine for me.
How far short do I want the two pieces of glass for the box to be from the side panes of my tank, where they get siliconed?
Thanks!