Closed loop and overflow system help

[d9hp]

Not so nano but I figured many of you could help on this one:

 

48"x24"x16"

 

I need some help with the overflow and closed loop system I wish to implement and am open to all ideas and criticism.

 

I know many people have troubles with strainer intakes in the tank and I am wanting to incorporate an external calflo overflow along the rear panel and a beananimal overflow system. I am wanting a quiet system, but the slow trickle of water into the overflow box will be allowable if it isn't a 'rushing' sound. I am building the tank myself and am wondering how much shorter the rear panel should be to keep the water line ~.75-1" from the top of the tank. I plan on adding a small strip of acrylic teeth so I imagine I could adjust this to change the water level.

 

I am going to drill the returns below the overflow box and use loc-line to raise them towards the surface and use flare nozzles solely for surface agitation. This will mitigate the problems associated with backflow from a power-loss.

 

The return pump I envision using will be a Quiet One 4000 Flow (Rate: 1017 gph)

Sump: 50g breeder (36x18x18)

 

Closed loop will provide ALL other water circulation in the tank with 5 envisioned return points. 1-4 (labeled below) will be fed through an ocean motion's squirt powered by a Pan World 100PX-X 1270gph 13'. The excess flow from the pump will be routed through a bypass to return CL5 on the back center of the tank, which will provide constant flow.

 

**from Ocean Motions:

 

Will 1" drains suffice for BA overflow system running around 1000gph after head loss?

 

Will the horizontal overflow be capable of handling all this water without being extremely loud? Would I be better off drilling a different hole for the closed loop intake in the bottom of the tank?

 

If the intake to the CL pump is 1" should my intake bulkhead also be 1" or larger, with a step down coupler at the pump?

 

 

 

 

Finally, if I can get a local vendor to 'notch' the rear pane I could drill the return lines a bit higher and clean up the entire back pane with this implementation below. Would the reduced overflow area cause big problems?

 

OPTION B

 

[nibor]

Oooo, a proper engineering problem. Back to you in a bit with some answers. I've got to eat my dinner first, it's 8pm here.

[d9hp]

Oooo, a proper engineering problem. Back to you in a bit with some answers. I've got to eat my dinner first, it's 8pm here.

 

Awesome nibor. No huge rush on this but I definitely am excited about the planning. Just found out that I want the flow through the sump to be about 4x the water volume so I need to REALLY scale back the return pump quite a bit to about 500gph-600gph so after the head losses it would be about where I want it.

 

Saw a pretty awesome idea for the rear pane that would allow me to do things a bit differently but it sure is scary to think about doing this to 3/8" glass on a MUCH larger scale (like 20 holes instead of 6). The guy said these alone took him 5 hours. This would enable me to do option B on my own:

 

taken from:

http://www.reefcentral.com/forums/showthre...hreadid=1609889

 

Thanks all

[bitts]

with a calfo style the teeth will lower the effectiveness but may be needed based on livestock. as to depth, or diference in the height of overflow to weir. this mostly depends on the desired water line. 1/2"-3/4" would work but I would go more towards 3/"-1" below the rim as a starting point. as to the overflows size, mine is 4x7 & fid it to be slightly larger than I need at times. could have probably gotten away with 3"-4"x5" & will move towards this on my next build. (75) one thing I have also found is that a baffle to seperate the majority of the overflow from the drain can be extremely helpful. using Bernoulli's math can allow you to set the height between the main drain line & backup preventing the need for any to trickle through the backup.

[nibor]

I've had to make an assumption here. Please provide accurate measurements if you can.

 

OK, I'll start with the Bean Overflow, specifically the full siphon drain as this is a limiting part. I have assumed that the vertical difference between the water level in the overflow and the water level in the sump is 3 feet ( the pipe runs to just bellow the sump water level in a Bean setup). With 1 inch diameter pipe, the maximum flow will be around 2000gph. A Quiet One 4000 pumping back up 3 feet with a few bends in the pipe will output about 700gph. So there should be no problems there. In a worst case scenario, the emergency drain will become submerged and at 1 inch diameter should also be able to handle 2.5 to 3 times the return pump. (I would expect the water level to move up and down a little bit initially when both the other drains are closed and the emergency drain to make a lot of noise but it should be more than capable of handling the flow which is what we need in case of a blocked drain).

 

I would suggest with this design that you make the overflow box as deep as possible and have the closed loop entry much lower than the entrances to the Bean standpipes (no mathematics behind this but 3 inches seams about right). The reason I suggest this is that the overflow box is like a sump. The tank overflows over the weir and into the box which then overflows into the standpipes and on to the sump below. If the inlets to the drains are at the same level as the inlet to the closed loop, the two will interfere with each other. By making the overflow box deeper and keeping the depth of water in the box high, there should be no chance of the return loop pump drawing air, nor the overflow repeatedly repriming.

 

Next the overflow itself. Basically, you are talking about putting roughly 2000gph over the weir. The longer the effective overflow length, the less noise it will make. What you are trying to achieve is a silent film of water flowing over the back of the tank and maximum surface skimming. A toothed overflow has a negative effects on this. Firstly, it reduces the effective linear length of the overflow and the water height above the overflow is greater to achieve the same flow. Secondly, the flow is broken into streams which trickle into the overflow rather than "sliding" over the glass edge as a thin film. This obviously makes more noise. Have you considered a horizontal slit rather than a toothed weir? To set this up you glue a strip of glass or acrylic above the overflow weir at the apropriate height, about a 1/4 inch is usually about right. You can do this after test filling the tank and starting all the pumps to ensure you get the right height.

 

Concerning your two different designs, obviously the first design has a higher performing overflow as it is longer (I believe that 30 inches without teeth would be more than capable of handling 2000gph without making too much noise however). There are two concerns with the first design, firstly any leaking in the locline would lead to a slow drain back to the sump of a large quantity of water in case of a power cut. Secondly, the returns will push any surface film towards the front of the tank away from the ends of the overflow and sort of making this part of the overflow redundant. The second design is better in that the return bulkheads are higher in the tank and there position should allow marginally better circulation of water towards a greater percentage of the total overflow length. Keep in mind that more than 50 percent of the water skimmed by the overflow will be blasted back into the tank by the closed loop and it is for this reason that I would seperate the closed loop and the sump feeds.

 

In an ideal world, I would go for your second design with a seperate feed to the closed loop drilled through the back glass somewhere. Even better if you can run the returns to the front of the aquarium, but this smells of one of those clean rimless builds to me so you probably don't want to do that.

[bitts]

nibors right in that you want the closed loop to create a gyre from the weir down the back wall across the sand up & then back to the weir. be very careful about the returns interfering with this. one thing to consider is that for probably the same price if not less you could swap the closed loop out for mp10's. reducing the number of holes drilled in the tank. the long slit works well & is what I did on mine. in mine with a 20" weir running at about 400 gallons there was about 1/8" flowing over the weir. as to placement of the closed loops input. as long as it is inside the tank & the pump is below all points in or out. it can be placed any where you feel it to be useful.

[d9hp]

WOW you guys are good.

 

Thanks to you both. Some great points made!

 

First of all -> teeth are out. Going with the a single slit. Makes loads of sense. I believe I could still mill that out of the glass without it affecting the structural integrity of the pane, do you both agree?

 

Also, I understand the point you're making about needing the closed loop to be higher than the bean animal feeds but I'm worried that by making the overflow itself extremely deep (like 6-8") that I would get a LOT of noise from that thin wall of water going down into the tank.

 

Bitts, I understand your point about the baffles in the overflow and it sounds pretty reasonable. I could probably do TWO of them and have a small space between where there is just water. I think option 2 will be best and will provide plenty of room for water flow, especially if I go with a single slit along the back wall as high as I can possibly make it (which will not be a fun day to make happen)

 

I know I could go with two MP40s but that's $1000 and two MP10s doesn't seem like it would be enough flow. Also, I don't want anything on the three sides. Nothing. Do you think that just two MP10s on the back glass, say where CL1 and CL2 are above would do the trick? That would effectively reduce the complexity of the system by a million. I would still go with option B. I could get the wireless versions and pair them up for about $500. That's essentially the same price (way less work) as the closed loop system I'm looking to implement. I want a clean build but honestly two MP10s (just the wet sides) isn't any worse than a snake of loc-line in the tank at those two points.

 

Thanks for your help!

[nibor]

First of all -> teeth are out. Going with the a single slit. Makes loads of sense. I believe I could still mill that out of the glass without it affecting the structural integrity of the pane, do you both agree?

That is going to be a whole lot of work, it will literally take you days (and when you crack the glass after 18 hours of continual work you will likely turn to murder). If you make the overflow a full coast to coast, you can just make the back glass an inch below the rest of the tank. Once you've test filled the tank and sump, you can just glue a long strip of acrylic (or glass if you wish) at the right height across the full width of the tank to create the snail guard. Just need to work out your returns then.

 

Also, I understand the point you're making about needing the closed loop to be higher than the bean animal feeds but I'm worried that by making the overflow itself extremely deep (like 6-8") that I would get a LOT of noise from that thin wall of water going down into the tank.

Closed loop input needs to be lower not higher, down in the bottom of the overflow. It doesn't make as much noise as you think. If the overflow is long enough the water just sheets down the glass. It becomes noiser if you try to run too much flow over a given length. Most of the overflows I see/hear are woefully undersized and sound like several guys sharing a toilet after drinking a lot of beer. Some people bodge solutions to this like sticking floss or sponge into the overflow.

 

Bitts, I understand your point about the baffles in the overflow and it sounds pretty reasonable. I could probably do TWO of them and have a small space between where there is just water. I think option 2 will be best and will provide plenty of room for water flow, especially if I go with a single slit along the back wall as high as I can possibly make it (which will not be a fun day to make happen)

This is a good idea as with a large overflow like this, there is a lot of water to drain into the sump when you turn the power off. A simple baffle of the right height isolating the standpipe section of the overflow will reduce this back flow and allow you to maintain a higher water level in the sump or use a smaller sump.

 

I know I could go with two MP40s but that's $1000 and two MP10s doesn't seem like it would be enough flow. Also, I don't want anything on the three sides. Nothing. Do you think that just two MP10s on the back glass, say where CL1 and CL2 are above would do the trick? That would effectively reduce the complexity of the system by a million. I would still go with option B. I could get the wireless versions and pair them up for about $500. That's essentially the same price (way less work) as the closed loop system I'm looking to implement. I want a clean build but honestly two MP10s (just the wet sides) isn't any worse than a snake of loc-line in the tank at those two points.

 

Each MP10 is rated at up to 1575gph. On paper, that's way more than you proposed closed loop. Consider the tank as two 24x24x16s side by side. Does 350gph of return and an MP10 sound like enough for a 24x24x16? I always liked closed loops in my previous aquariums (had a range of setups since the late 80's some with and some without closed loops), much better than having a load of powerheads. The Vortechs have rather changed this for the majority of applications. After moving all over the place for work for the last few years, I'm finally stable enough to have a reef tank again. I'm thinking of a similar design to yours but on a smaller scale, probably 30x18x15 and am going to run 2 MP10s and a slower return of around 200 gph. (just enough for a chiller which will unfortunately be essential). I want to use a coast to coast overflow to keep the display a perfect cuboid which my autistic tendancies demand.

 

Thanks for your help!

You're welcome.

[d9hp]

That is going to be a whole lot of work, it will literally take you days (and when you crack the glass after 18 hours of continual work you will likely turn to murder). If you make the overflow a full coast to coast, you can just make the back glass an inch below the rest of the tank. Once you've test filled the tank and sump, you can just glue a long strip of acrylic (or glass if you wish) at the right height across the full width of the tank to create the snail guard. Just need to work out your returns then.

 

 

Closed loop input needs to be lower not higher, down in the bottom of the overflow. It doesn't make as much noise as you think. If the overflow is long enough the water just sheets down the glass. It becomes noiser if you try to run too much flow over a given length. Most of the overflows I see/hear are woefully undersized and sound like several guys sharing a toilet after drinking a lot of beer. Some people bodge solutions to this like sticking floss or sponge into the overflow.

 

 

This is a good idea as with a large overflow like this, there is a lot of water to drain into the sump when you turn the power off. A simple baffle of the right height isolating the standpipe section of the overflow will reduce this back flow and allow you to maintain a higher water level in the sump or use a smaller sump.

 

 

 

Each MP10 is rated at up to 1575gph. On paper, that's way more than you proposed closed loop. Consider the tank as two 24x24x16s side by side. Does 350gph of return and an MP10 sound like enough for a 24x24x16? I always liked closed loops in my previous aquariums (had a range of setups since the late 80's some with and some without closed loops), much better than having a load of powerheads. The Vortechs have rather changed this for the majority of applications. After moving all over the place for work for the last few years, I'm finally stable enough to have a reef tank again. I'm thinking of a similar design to yours but on a smaller scale, probably 30x18x15 and am going to run 2 MP10s and a slower return of around 200 gph. (just enough for a chiller which will unfortunately be essential). I want to use a coast to coast overflow to keep the display a perfect cuboid which my autistic tendancies demand.

 

 

You're welcome.

 

 

I really appreciate all the input.

 

Definitely considering the MP10s although I love the work the closed loop requires. I love projects!

 

Thanks again to you both.

[bitts]

let me spend a little time with this in sketchup. since its so close to some other stuff I've done, I may have something that answers all questions with out too much work. also take a look at mine & see what you think. as to coast to coast & noise. well I'm a freak. I can hear my Ipod when I'm in the other room even though its at the minimum volume. the fridge drives me nuts when I'm watching tv. my 29 is next to my bed.

 

as to the mp10. two of them should be way more than you need but in this application I would defiantly go with 2 over one. when I've put mine on the back glass I can only turn it up to maybe 3 dots out of 12.

 

 

 

 

 

this would be the very contrived frankenstine of a closed loop that I had in the works before I factored in cost. but if you ignore the strange standpipes it may give a better picture of things.

 

 

 

 

off to doodle back in a bit.

[bitts]

so this is just a quick sketch but hope it makes sense. full coast to coast overflow with baffle. 1" drains. note the round corner in the side panel to prevent cracking.

 

[d9hp]

so this is just a quick sketch but hope it makes sense. full coast to coast overflow with baffle. 1" drains. note the round corner in the side panel to prevent cracking.

 

 

Thanks for the diagrams. Both are really fantastic.

 

So the round corner is a must? I was going to have the tank built and then just build the overflow myself. I think the price increase in custom corners like that would really be inhibiting.

 

 

Where does the cracking occur if you didn't have that rounded corner?

 

Sketch up is the coolest! Thanks again.

[bitts]

its not a must, but if the side panel will be one peace than I would think about it. talk to your glass guy. after all their going to be the best source of knowledge on glass. they may have an alternative or know why you wouldn't need it. its just the best way I've seen it done. curved corners seem to have greater success. both in this application or in cutting the notch/slit to form the weir.

 

cracking would most likely start @ the inside corner then fallow the line with greatest pressure & or the shortest path. since glass is cut by bending it after being scratched it would depend on the imperfections of the glass. doing the corner like that probably wouldn't cost that much more though. but I'm sure we can figure out a deferent way. like instead of a 100% coast to coast come in a couple inches on either side then cut the weir in to the panel. this would allow you to attach the overflow to the back panel, with out needing to incorporate the side panels. like was brought up earlier, a slit or notch. I've got a couple other ideas but might be a little while till I have time to sketch them.

 

I'd be interested to see what nibor thinks on this though.

 

oh & that mock up is the correct dimensions. let me know, if you want I can throw it up on the warehouse for you to download.

[nibor]

You're right bitts, you don't want any internal right angles in a piece of glass. Notches are stress concentrators and sharp notches in a material like glass are a sure fire way of causing a crack.

 

d9hp, if you don't need access to the area below the overflow from the sides, you could just use a singular rectangular piece for each side and make them deep enough to be the sides of the overflow as well. You would want to build the tank "sides around the bottom pane" if you did this just so the full length of each side pane was supported by the stand. Depends on your stand design a suppose though.

[bitts]

oh ah

nibor thats an awesome Idea. as long as it works for the situation. +1

 

one more thing...

 

I keep meaning to point out that when you have it built. get them to use dow 795 or 999A for the silicone.

 

so nibor kinda like this.

[nibor]

Exactly.

[d9hp]

You're right bitts, you don't want any internal right angles in a piece of glass. Notches are stress concentrators and sharp notches in a material like glass are a sure fire way of causing a crack.

 

d9hp, if you don't need access to the area below the overflow from the sides, you could just use a singular rectangular piece for each side and make them deep enough to be the sides of the overflow as well. You would want to build the tank "sides around the bottom pane" if you did this just so the full length of each side pane was supported by the stand. Depends on your stand design a suppose though.

 

Brililant.

 

The stand is going to be built my me and I was planning on using two massive pieces of birch to skin the tank and the sides would extend up around the overflow to cover everything with access, if necessary. This would effectively not change my stand design except needing to extend the flat area where it rests. The deeper the overflow box the better, as this would certainly help hide the PVC pipes in the back.

 

oh ah

nibor thats an awesome Idea. as long as it works for the situation. +1

 

one more thing...

 

I keep meaning to point out that when you have it built. get them to use dow 795 or 999A for the silicone.

 

so nibor kinda like this.

 

I'm going to build the tank myself and I've also heard about 999A.

 

That's a great rendering bitts.

 

Just to get an idea of cost, how deep (from tank to wall, so to speak) should the overflow section be? 5"?

 

Who are you guys?! If you were closer I'd buy you both a beer!

 

Thanks, as always.

 

 

EDIT:

 

Did some thinking on the whole closed loop MP10 thing and I believe I've decided that when the time comes ecotech will in fact be getting my money and I won't be going with the ocean motions and pan world closed loop. It's just a logical route given the expandability, power consumption, and the simple fact that I can use them again in another system if necessary.

 

It came down to realizing that the $50 for whiskey to cool my nerves for drilling 6 holes in 1/2" glass panes would put me over the powerhead budget

[bitts]

Mine is 4x7 & so far is working great, don't think I would go more than that. have the glass shop drill the drain holes. they should only be 10-15 a hole. the size of the overflow doesn't really matter. I figure that the back will be painted so the only thing to worry about is the pvc in the overflow itself. just means you get black pvc. other option would be to use smoked glass or black pvc for the drain lines. one downside with this setup is that you will need a way to reach the mp10's if you put them on the back wall.

 

I'm going to finish that doodle up & throw it up on the warehouse for you. I'll post the link in a bit.

[bitts]

here's the link.

http://sketchup.google.com/3dwarehouse/det...amp;prevstart=0

[pprice01]

Hi everyone,

 

I've been lurking on this thread for a short time and am very impressed with all of your design work. I hope to use it on my next build. I don't know a whole bunch about glass vs. Acrylic but wonder if it would be possible to use acrylic instead for added strength? Pros/cons?

 

Thanks.

[nibor]

You could easily use acrylic for a build of this type.

 

Advantages: clarity, modern look, much easier to drill or machine than glass, strength.

 

Disadvantages: material cost, easy to scratch, I personally find welding acrylic more difficult than working with silicone and glass.

 

Happy New Year to you all. I'm going out to get sloshed. Enjoy the rest of 2010.

[bitts]

main things I like about the acrylic is its 20x stronger. plus it has much better thermal & optical property's. I love the idea that if I'm having a party & some one starts tossing a ball around. the tank is well maybe not bombproof but at least better than glass.

 

but I scratch my glass tanks so theres that.

[d9hp]

Thanks again guys, I am working out a meeting with a glass guy and hopefully we can come to an agreement on something reasonable.

 

Updates to come...sometime.

[d9hp]

Thought I'd resurrect this old thread since I'm still dealing with a lot of the same issues but the time line and budget has been extended. I am really having a tough time justifying the added cost of the external overflow, however. The surface skimming capabilities are incredible but the new tank will be 72"x30"x18" so an internal weir wouldn't affect things too terribly bad and I believe I can use .25" glass that could easily be cut locally...and CHEAP.

 

I just want this tank to look really good and this is one of the big issues. I have to get the tank decided on and ordered before continuing with the rest of the project so I can't keep changing my mind.

 

I have also decided this will not be a reef...probably EVER. I love the reef but I have far more enjoyment keeping fish so I want to really stock this thing to the brim and not beating myself up if my nitrates go over 5ppm. This makes me think that surface skimming will be even MORE important. I'm going to have a MASSIVE skimmer. Like 250g+ rating. I still would like to get away with 2 MP10s and a big return pump but I'm not sure if that would be enough flow. It would probably be in the neighborhood of 4000gph total when the pumps are on full blast. I doubt I would get a wave or anything but I imagine it would keep things circulating. Another big factor for considering an internal weir is that the euro-braced tank is roughly $500 cheaper and bracing does nothing on a tank with a huge slit down the side of it.

 

I'm ranting, sorry. I am really excited about designing this massive system because it's going to be the last one I design for quite a long time. I'd like to keep total cost under $3k. That's going to be difficult I'm afraid...especially with the skimmers I'm considering.

 

What are some other effective surface skimming overflow methods that don't look terrible? I also am a little worried about how much space behind the tank this system requires as opposed to the herbie method which would allow me to put the tank close to the wall. I just don't know how much I like internal overflow boxes...or how well they skim.

[nibor]

Guess you've already seen this page. If you can spare some space across one end of the tank you could build an internal coast to coast like the rendering on this page but across one end. With a 30 inch wide tank, you would have good overflow length and could group the stand pipes towards the back to keep some viewing from that end if you needed it. Returns would come in over the top at the other end.

 

Completely different idea would be to build an internal overflow box with a herbie and only run about 1.5 times the system through it. With the lower flow rate, you would not need such a long overflow to keep the film of water thin thus maximizing the removal of the oily film from the surface. Feed the overflow directly into a recirculating skimmer and from the skimmer to the sump. This way everything surface skimmed by the overflow goes to the protein skimmer. If you built a fuge in the sump and wanted a bit more flow, you could add a small powerhead.