Jump to content


Photo

My DIY PVC Overflow


  • Please log in to reply
48 replies to this topic

#1
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA

A very good explanation of siphon overflows. Ending has a very good demonstration.



Overflow 1.0 use v 2.0 below

I have been testing and changing this overflow for about 2 weeks now. All together I would say I have about 200 start/stops on the system and it has never failed to start back up.

I am using 3/4" PVC with a 1.5” opening/overflow. After much testing the large opening allows more water to fall into it as it is a larger diameter than just the 3/4" pipe. At least that is my explanation for the increase in flow.

I have also been testing with the length of the outside pipes versus inside and have found that having the outside pipes 1.5” to 2” longer than the inside also adds greatly to the flow output. This length may change for large pipes

I was running my Mag 5 pump (1/2” PVC return line) at just under 2/3 open (ball valve) running the inside and outside at the same at the same length. With increasing the outside length I am now running the pump at maybe 4/5 open. At 100% open the water level rises ever so slowly.

I am pretty happy with how it is performing so far but I have burnt through parts with all of my testing but PVC is cheap so the testing has been worth it.

Output is currently at 2 cups per 1.8 seconds so that is 250 GPH.

Posted Image

Posted Image

Posted Image

Posted Image

Posted Image


Posted Image

Posted Image

Posted Image

Posted Image

Posted Image

Link to larger pics
http://picasaweb.goo...baugh/Overflow#


Overflow 2.0

So I went to setup everything outside this weekend to perform an extended week long test and ran into some issues. Because of these I had to make some design changes. In all honesty about half way of working through these issues I said “screw it” and went into the house to order an overflow box online but turns out not even the new Eshopps nano overflow box (that comes out this month) or the stupid expensive LifeReef nano overflow will fit in my BC29 so then I became determined to get this to work and work correctly.

Issue #1
Overflow would not start back up.

Issue was that my check valve malfunctioned already. DO NOT TRUST OR USE THESE even though I installed a new one you will see a bleeder valve to back it up in the line now.
I also adjusted the height of the inside vs outside tubes to make them even once again. From all my readings this is the best way to make sure you do not break the siphon. Also the outlet to from overflow must be above the bottom on the inside pipe assembly.
During testing if these pipes are not 100% air/water tight you will break the siphon if you have the pump off for an extended period of time simulating a power outage. I actually had to end up gluing (PVC cement) everything for my extended testing. Best way to test for this is to shut the pump off and let the water drain. You will have some slow dripping from the drain pipe for maybe 30 to 60 seconds but after that you should have nothing. If you have a constant drip it means you are leaking air and your siphon will eventually fail causing it not to drain once you start the pump back up.

I am still thinking about picking up an aqua lifter pump for this build in order to always be removing air. At the very least I would turn it on if I leave my house for an extended period of time. I am going to try to pick one up on this Thursday add it to the test to see what it does. I will report back with the results

Issue #2
Terrible gurgling sound.

In my first test setup you can see my outlet gently goes into my sump with no bends (this was silent) but I changed that for the extended testing to mimic how it would be actually installed into the DT. I added a 90 elbow and ran a pipe straight down and the gurgling started and there was no way I was going to run something that sounded like that. After numerous changes I discovered adding any bends over 45 degrees would cause a gurgling sound from the vent tube. My guess is that air is getting trapped in the bend. I was also thinking it could have been the 90 elbows (the elbows have a very sharp bend internally) I was using but after reading more online it seems that this is a common issue with these overflows. After about the 10th design change I decided to add a 3/4” to ½” reducer on the outlet with a 45 elbow (see pics) and run ½” pvc the rest of the way. This allowed me to add the 90 elbows that I would need in order to get it into my sump and it has been silent for the past 18 hrs that it has been running. This also did not decrease my flow as I did not have to adjust the ball valve at all. I still have some very minor tweaks on this to perform so check back on this aspect.

Other changes
Changed the inlet to include teeth.

Additional protection measures that will be put in place
Will be adding a float switch (location TBD) to turn the pump off in case overflow failure. This would only keep the pump from burning out as my pump inlet height will never allow my DT to overflow if the overflow fails for some reason.

So far so good with the testing. It has been running for about 18 hrs now with two 30 minute “pump offs” and no issues starting back up. The water level in the simulated DT/bucket has not fluctuated at all. I keep on test for air pockets building the top bend but nothing so far.

I will be updating all week as the test progresses

Posted Image

Posted Image

Posted Image

Posted Image

Updates

52 hrs strong

Survived some nasty storms today which was a very good test for it. You can see the water was very rough in the bucket.

Posted Image

Edited by jfarabaugh, 08 September 2010 - 08:37 PM.


#2
C-Rad

C-Rad

    Nano Reefer

  • Members
  • 406 posts
  • Joined 14 May 2009
Good job doing such thorough testing, and thank you for taking the time to document your results for us.

I'm guessing that the little yellow/tan rubber thing at the top of the overflow pipe is not a breather tube that lets air into the pipes, but rather a bleeder that lets you suck air out to start the siphon. Is that correct? If so, it's a weak link. If it ever fails, and lets a lot of air in at once, the siphon would break (stop) and the whole sump would be pumped into the tank (overflow). Better to use a bleeder valve with a much lower failure rate than those little rubber things, or ideally, no bleeder valve at all (you can use a strong pump on one end of the siphon tube to blow all the air out of it to get the siphon started) With a fast flow like you have, little bubbles can't collect at the top of the overflow, so you shouldn't need to bleed it once it's started.

Since it'a a siphon, flow will naturally increase as you lengthen the distance between the lower outlet and the water level in the bucket (because the water pressure (Pounds/Square Inch (PSI)) increases as the weight of the water in vertical part of the output tube (pounds) increases while the area of the output opening (Square Inches) stays the same.

Is this overflow close to silent, or does it gurgle? Does it suck much (any?) air into the intake? I'll bet it's close to silent, and is pulling almost no air in, like the main siphon on a Herbie.

You've done a lot of stress testing to see if it can handle a power outage. Have you considered testing other types of threats? Maybe you could simulate something floating into the overflow and reducing the output by various percentages. It might also be useful to see what happens as the return pump output gets lower and lower. At some point the siphon might start sucking in a lot of air, reducing the flow rate and maybe breaking the siphon. Pumps get weaker as they age (get dirty) and you want to know how slow is too slow. You can also run an air-stone or a skimmer to see if bubbles floating on the surface tend to collect and break the siphon.

Running two of these, each at less than or equal to half of its capacity, would provide redundancy and drastically reduce the chance of a flood.

#3
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA

Good job doing such thorough testing, and thank you for taking the time to document your results for us.

I'm guessing that the little yellow/tan rubber thing at the top of the overflow pipe is not a breather tube that lets air into the pipes, but rather a bleeder that lets you suck air out to start the siphon. Is that correct? If so, it's a weak link. If it ever fails, and lets a lot of air in at once, the siphon would break (stop) and the whole sump would be pumped into the tank (overflow). Better to use a bleeder valve with a much lower failure rate than those little rubber things, or ideally, no bleeder valve at all (you can use a strong pump on one end of the siphon tube to blow all the air out of it to get the siphon started) With a fast flow like you have, little bubbles can't collect at the top of the overflow, so you shouldn't need to bleed it once it's started.

Since it'a a siphon, flow will naturally increase as you lengthen the distance between the lower outlet and the water level in the bucket (because the water pressure (Pounds/Square Inch (PSI)) increases as the weight of the water in vertical part of the output tube (pounds) increases while the area of the output opening (Square Inches) stays the same.

Is this overflow close to silent, or does it gurgle? Does it suck much (any?) air into the intake? I'll bet it's close to silent, and is pulling almost no air in, like the main siphon on a Herbie.

You've done a lot of stress testing to see if it can handle a power outage. Have you considered testing other types of threats? Maybe you could simulate something floating into the overflow and reducing the output by various percentages. It might also be useful to see what happens as the return pump output gets lower and lower. At some point the siphon might start sucking in a lot of air, reducing the flow rate and maybe breaking the siphon. Pumps get weaker as they age (get dirty) and you want to know how slow is too slow. You can also run an air-stone or a skimmer to see if bubbles floating on the surface tend to collect and break the siphon.

Running two of these, each at less than or equal to half of its capacity, would provide redundancy and drastically reduce the chance of a flood.


Thank you for taking the time to look at this and respond...I was beginning to think I wasted my time writing this up.

The yellow part is a check valve to remove the air to start the siphon. https://www.fosteran...mp;pcatid=10109

Correct, once the air is removed the first time I do not have to remove it again.

I am actually thinking of adding a second check valve in the line for additional protection. I am not familiar with bleeder valves. Do you have a link or example for one?

I have thought a lot about this setup from both the overflow and the return standpoint of not flooding my house. It is hard to see but my pump has a 90 elbow pointing up on the intake (will add additional PVC once I finalize the water level in the sump). If the overflow fails for any reason only a small amount of water will be pumped into the DT before the pump runs dry and the DT can handle the extra water thanks to my 2nd and 3rd chamber in my BC29. I would rather burn a pump up before I flood the house. I may add some ATO floats to shutdown the return pump if the water gets below the intake of the pump to save it from burning up. I also have a leak detector in the cabinet and will such down the return pump if anything starts to leak within the cabinet itself.

Right now it is close to silent.

The overflow will be installed in the first chamber in my BC29. This will protect it from anything large (debris or animal) getting to it. I am sure I will have to clean the trap out but that will just become part of my regular tank maintenance. But in any case with any overflow failure the secondary return protection that I have in place will save me.

These are all great points and areas to look at and test. But with regular maintenance and the protection measures I have in place I believe I can avoid them. I would love to have a drilled tank and my next tank will for sure be drilled but I started this tank as a noob and was not thinking ahead so this is where I am at. I do not want to take the chance of drilling this one.

I feel that this design is just as safe as any store bought overflow given my precautions. Plus I haven’t been able to find one that would actually fit in my tank anyway.
http://www.fosterand...fm?pcatid=18358
I am open to suggestions if it can fit in the first chamber.

2 overflows would nice but space is very limited in the back of my BC29 and only 1 will fit. I already had to modify the elbows to get it to fit. Using street elbows reduces you down to 1/2" PVC due to the internal design. So I found cutting the elbows to work better.

Internal
photo_1.JPG

Internal vs External
photo_2.JPG

#4
Renogaw

Renogaw

    Nano Reefer

  • Members
  • 101 posts
  • Joined 07 Aug 2010
not a waste of time, been wondering how to do this for a 28gal jbj

#5
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA
Overflow 2.0

So I went to setup everything outside this weekend to perform an extended week long test and ran into some issues. Because of these I had to make some design changes. In all honesty about half way of working through these issues I said “screw it” and went into the house to order an overflow box online but turns out not even the new Eshopps nano overflow box (that comes out this month) or the stupid expensive LifeReef nano overflow will fit in my BC29 so then I became determined to get this to work and work correctly.

Issue #1
Overflow would not start back up.

Issue was that my check valve malfunctioned already. DO NOT TRUST OR USE THESE even though I installed a new one you will see a bleeder valve to back it up in the line now.
I also adjusted the height of the inside vs outside tubes to make them even once again. From all my readings this is the best way to make sure you do not break the siphon. Also the outlet to from overflow must be above the bottom on the inside pipe assembly.
During testing if these pipes are not 100% air/water tight you will break the siphon if you have the pump off for an extended period of time simulating a power outage. I actually had to end up gluing (PVC cement) everything for my extended testing. Best way to test for this is to shut the pump off and let the water drain. You will have some slow dripping from the drain pipe for maybe 30 to 60 seconds but after that you should have nothing. If you have a constant drip it means you are leaking air and your siphon will eventually fail causing it not to drain once you start the pump back up.

I am still thinking about picking up an aqua lifter pump for this build in order to always be removing air. At the very least I would turn it on if I leave my house for an extended period of time. I am going to try to pick one up on this Thursday add it to the test to see what it does. I will report back with the results

Issue #2
Terrible gurgling sound.

In my first test setup you can see my outlet gently goes into my sump with no bends (this was silent) but I changed that for the extended testing to mimic how it would be actually installed into the DT. I added a 90 elbow and ran a pipe straight down and the gurgling started and there was no way I was going to run something that sounded like that. After numerous changes I discovered adding any bends over 45 degrees would cause a gurgling sound from the vent tube. My guess is that air is getting trapped in the bend. I was also thinking it could have been the 90 elbows (the elbows have a very sharp bend internally) I was using but after reading more online it seems that this is a common issue with these overflows. After about the 10th design change I decided to add a 3/4” to ½” reducer on the outlet with a 45 elbow (see pics) and run ½” pvc the rest of the way. This allowed me to add the 90 elbows that I would need in order to get it into my sump and it has been silent for the past 18 hrs that it has been running. This also did not decrease my flow as I did not have to adjust the ball valve at all. I still have some very minor tweaks on this to perform so check back on this aspect.

Other changes
Changed the inlet to include teeth.

Additional protection measures that will be put in place
Will be adding a float switch (location TBD) to turn the pump off in case overflow failure. This would only keep the pump from burning out as my pump inlet height will never allow my DT to overflow if the overflow fails for some reason.

So far so good with the testing. It has been running for about 18 hrs now with two 30 minute “pump offs” and no issues starting back up. The water level in the simulated DT/bucket has not fluctuated at all. I keep on test for air pockets building the top bend but nothing so far.

I will be updating all week as the test progresses

Posted Image

Posted Image

Posted Image

Posted Image

Edited by jfarabaugh, 11 August 2010 - 07:52 AM.


#6
MyrtleBeachreefer

MyrtleBeachreefer

    Nano Reefer

  • Members
  • 65 posts
  • Joined 20 Jul 2010
  • Myrtle Beach, SC
Great thread, following along since i am doing this for a jbj28g. One question, where did you pick up that bleeder valve?
JBJ Nanocube 28g HQI, azoo fan, Tunze 9002 skimmer, 25lbs live rock, 30lbs fiji pink live sand.

#7
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA

Great thread, following along since i am doing this for a jbj28g. One question, where did you pick up that bleeder valve?


I was about to order them but then recalled that I had those on my POS biocube protein skimmer....see they are good for something :P

Best place I found for them.

http://www.thatpetpl...00/category.web

#8
CONSTANTNE

CONSTANTNE

    Nano Reefer

  • Members
  • 594 posts
  • Joined 19 May 2007
Nice write up!!

#9
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA
still going strong....28 hrs now... water level is spot on.

passed another 30 minute pump off

#10
CoolNameGoesHere

CoolNameGoesHere

    Nano Reefer

  • Members
  • 202 posts
  • Joined 01 Nov 2008
  • Cali - 91601
hey jfar, great write-up. perfect timing b/c i'm considering this type of overflow. i've read a bunch of the DIY guides for this kind of overflow, but yours has lots of helpful pics! i really like the idea of a bigger opening for the overflow :)

couple quick questions (at this point): how come you chose rigid pvc instead of flex tubing on the drain? i'm thinking the flex tubing would soften some of the angles and make for less noise... maybe?

also, what about a clear u-tube at the top instead of the 2 90* elbows? it'd let you see if you have an air bubble and again would soften the harsh angles...

great work so far! i'll be tagging along to see the progress.
*************
Mike - CNGH
Go Bruins! '00
SCNRS.com
SFVR.net

40B *Contest* Thread - In progress

#11
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA

hey jfar, great write-up. perfect timing b/c i'm considering this type of overflow. i've read a bunch of the DIY guides for this kind of overflow, but yours has lots of helpful pics! i really like the idea of a bigger opening for the overflow :)

couple quick questions (at this point): how come you chose rigid pvc instead of flex tubing on the drain? i'm thinking the flex tubing would soften some of the angles and make for less noise... maybe?

also, what about a clear u-tube at the top instead of the 2 90* elbows? it'd let you see if you have an air bubble and again would soften the harsh angles...

great work so far! i'll be tagging along to see the progress.

Thanks
I was testing with clear vinyl tubing. At some point I have to make a 90 bend into the sump. Given the height of my sump and my overall height of my tank I have very little room to make that bend. Maybe 8' to 12" total height to go from outside the DT into the cabinet and into my sump. When I was looking at it even with the vinyl and I don’t think I can do it without pinching it. BUT I may try it again once the overlow and the sump are installed.
I was thinking about clear pvc but nothing was available local. You would just have to make sure you have a very good seal on the u-tube.

#12
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA
Still running (40 hrs) and some very good test are now going on as rain is moving through the area.

Edited by jfarabaugh, 12 August 2010 - 07:31 AM.


#13
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA
52 hrs strong

Survived some nasty storms today which was a very good test for it. You can see the water was very rough in the bucket.

Posted Image

#14
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA
Still going 4 days straight. About to take it down as it is starting to grow stuff on it from being outside in the sun. :eek:

I will update once I install it in the tank….maybe this weekend.

#15
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA
Test fitting came out great!

Posted Image

#16
pumarjr

pumarjr

    Keys Reefer

  • Members
  • 410 posts
  • Joined 28 Jul 2010
  • Key West, FL
Do you have it working in your tank yet? I'm wanting to install a refugium in my bc29 and am following your thread.

#17
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA

Do you have it working in your tank yet? I'm wanting to install a refugium in my bc29 and am following your thread.


Sorry for the delay everyone I have been putting the little free time I have into my LED research/build. I am hoping fingerscrossed to install the overflow/sump either today or tomorrow. I will have updates as soon as it is done.

#18
bitts

bitts

    Yes. I'm blasting Wagner. While chanting KILL Da NOOB.

  • Members
  • 6,333 posts
  • Joined 30 Dec 2006
  • Denver, boulder, loveland, fort collins.....
Jfarabaugh

Was wondering with every thing that you done on this if could see anything on this design that could be improved. Thanks.



Posted Image

#19
pumarjr

pumarjr

    Keys Reefer

  • Members
  • 410 posts
  • Joined 28 Jul 2010
  • Key West, FL
Have you shut off the pump in the sump to see if the sump will overflow if it failed?

#20
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA

Have you shut off the pump in the sump to see if the sump will overflow if it failed?



In the process of installing still. My sump will have about 4 gallons of room to spare. With the pump off it will drain about 1.5 gallons from the display tank.

Its all about thinking ahead ;-)

#21
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA
Started the install yesterday and here are some bad pictures. I will take better pics once it is finished.

photo_1.JPG

photo_2.JPG

photo_4.JPG

#22
sandcruiser

sandcruiser

    Nano Reefer

  • Members
  • 371 posts
  • Joined 07 Dec 2006
  • Palo Alto, California; 94301
Two thoughts for you

You can bend PVC if you want to make that 90 into your sump a softer ride for the water. Just fill a pipe with fine sand, then heat it with a heat gun or maybe even a hot hair dryer, and bend it. The sand helps prevent a kink from forming. Wear gloves, the PVC will be necessarily hot when you are bending it. An alternative to sand is to use a steel spring, but i'm guessing sand is easier for most.

If you just run the bubble escape tube to a small pwrhead's venturi, it will provide suction at all times to remove air.

Good testing, by the way. Looks like you are really doing a thorough job pre-installation.

#23
pumarjr

pumarjr

    Keys Reefer

  • Members
  • 410 posts
  • Joined 28 Jul 2010
  • Key West, FL
I want to do this as well, but I want the flow to come from the second chamber and to be lower, I just need to figure out how to stop the flow if the water level in the dt becomes too high. Maybe by dropping the vent Down to 3" above the drain would solve this.

And all you need is a heat gun or hobby torch to bend the pipe, take the section of pipe you want to bend, mark the area in which you want to make the sweep, heat evenly by rotating the pipe and then make the sweep, once the sweep is where you want it, grab a cool wet rag and wipe it down, it will harden to where you made the sweep quickly.

#24
Monkfish

Monkfish

    Nano Reefer

  • Members
  • 690 posts
  • Joined 25 Jan 2008
  • Cranford, NJ
Just out of curiousity (sorry in advance for my noobness) what are the advantages of these kind of overflows opposed to just drilling and using some sort of standpipe? Btw the design looks great and looks to work very well.

Posted Image


*14 Gallon Long*


3.3 Gallon Pico



Posted Image


#25
jfarabaugh

jfarabaugh

    Nano Reefer

  • Members
  • 1,147 posts
  • Joined 09 Oct 2009
  • PA

Just out of curiousity (sorry in advance for my noobness) what are the advantages of these kind of overflows opposed to just drilling and using some sort of standpipe? Btw the design looks great and looks to work very well.

Drilling would be preferred but I am working with an established tank so for me drilling is not an option. In addition there is no overflow box that fits the BC9 so DIY it is for me.
My next tank will be drilled.