The Planted Tank Forum

The Planted Tank Forum (
-   Equipment (
-   -   Sump/overflow/return design question (

crazymittens 08-01-2012 05:07 PM

Sump/overflow/return design question
Sorry in advance for the long/complex post, I will try to be clear...

I did some searches, but didn't come up with anything concrete, especially with regards to a heavily planted low-tech/low-light/non-CO2 tank. Fish bioload will probably be small-medium.

* First off - I would like to state that filtration/plumbing overkill is fine by me if it leads to reduced maintenance/next-to-zero chemical usage.
* Second, I am sure 30G is oversized for a 55G's sump, that is fine - help me use the space - the tanks are already bought (well pretty much).
* Third, I am new to aquariums, but always like a challenge - I have sufficient experience in plumbing/fabrication - the effort involved to set it up is not an issue (I am doing MTS, after all!).

Here is the proposed design:
* 55G display tank w. 'coast-to-coast' overflow box & BeanAnimal drain system, MTS/gravel substrate, heavily planted
* 30G sump (filter sock -> foam -> pot scrubbies -> refugium? -> return (pump(s), heater(s))

The drain system, if you haven't googled it yet, incorporates two primary drain tubes and an emergency drain, yet gives a system with no water noise.

The two pumps I have are the Marineland NJ1800 and NJ3000, and a variety of 50w and 200w heaters.

My questions:
  1. The overflow box (rough dims. (length of tank)x3x3")will not have teeth (glass panels), can I expect issues with floating plants/plant matter? (intakes will have grates)
  2. With dual return pumps, guessing I need separate dual return lines?
  3. What is the best placement of return jets in a heavily planted tank?
  4. Is a filter sock a valid replacement for the foam pre-filter?
  5. Will a refugium help or hinder filtration? (possibly used to breed shrimp)

I do understand this is a somewhat unorthodox configuration - if you see glaring errors, please let me know. Again, the idea behind the sump tank is to aid in reducing maintenance cycles (once a month water changes, ideally - I have read of a number of people going much much longer doing low-tech/heavily planted). Also note that I am still in the design phase, so changes are still possible.

pandacory 08-01-2012 08:12 PM

Only problem I see so far is that the 30 gal may be too wide to fit into your stand, if they're the dimensions I am thinking of. This is easily overcome in DIY stand.

Some things i learned while setting up my coast to coast:

1. Plan for return line placement. Sounds dumb/no duh right? Well I overlooked it.
2. Don't do true coast to coast. Because of #1, what I would suggest instead is that you leave space on the sides to have holes in the back glass for bulkheads controlling returns. So your overflow box would be like 85-90% of the back wall.
3. There are more flow options in 1/2" loc line than in 3/4". This is important only if you get nervous about putting painted plastic in the tank, and want to black everything out.
4. Loc line is expensive! So don't overlook #3
5. Go external! External overflow boxes are much nicer. Learned this one after I already ordered. If you go external, you can drill in the bottom of the external box, so make it deep enough for upside down U with the screen attachment. If you can't go external, make the box as narrow as possible, and deep enough to use the screens you can buy, or else you will be doing a lot of cutting.
6. As narrow an overflow as possible. This deserved its own, even though it was mentioned in 5. Narrow overflow reduces impact on footprint, blocks less light and flow = less algae problems and more scaping options. Conversely you could just do a whole back wall overflow with single pane of glass, but 55 is already pretty narrow.
7. Have 2 returns. You will need to point one underneath the overflow overhang to avoid issues. Also gives you better distribution if you decide to inject co2 into the sump later
8. Glass is not the same price everywhere
9. Wish I would've made the sump out of acrylic
10. Easier to design the sump the way you want it, then find a heater that works, or go in line if you have to.
11. Leave enough unused volume in sump for drain off in power outages.
12. Your sump should also be level
13. I am torn on sponges vs. Socks as I have only ever used sponges, but people seem to like socks. I saw a mod where you cut slits in the top of the sock, then put the outlet lower than cut so that if it clogs it just runs through the slit instead of onto the floor.
13. The way I wish I would have designed my plumbing, no drawing tool so bear with me. Stand pipes exit the center back of tank, and feed into center section of sump, directly onto stacked filter media. Bubble trap on left side separating return pump from filter media. Return pump tees off and excess flow directed to the right of the center filter stack into refugium. Refugium then overflows into filter section before getting back to return side. Put drains in each section of sump. Drill sump for ice probe.

Hopefully the descriptions made sense and I didnt forget anything.

crazymittens 08-02-2012 01:21 PM

Well I actually ended up with a 50G tank, not 55 (that one was tempered, so no-go). The 50 is the same width as the 30, conveniently, so won't even have to modify the lights! 18" wide vs. 12".

Edit: To clarify, it's less of a stand and more of a DIY piece of furniture. My friend gave me the 'surface' his tanks used to be on - 8'x2' welded aluminum frame topped with plywood, finished in pine veneer/gloss. For reference the aluminum frame on its own weighs about 80lbs - heavy duty indeed. I will post pics at some point when time allows. Suffice it to say I'm able to fabricate whatever stand is required.

Okay, I will try to fit rebuttal questions on the same numbers you used

1. Gotcha.
2. I am actually planning to go 'tunnel through' the overflow box, if that makes sense.
3. Both sides of the tank will be 'viewable', so ideally I'd keep all the piping the same colour, but haven't thought this far ahead yet.
4. Yes, it is!
5. Now that I have the 50G and 18" width, internal should be fine.
6. Yes, BeanAnimal's design thread (150 pages...) has references to keeping the depth only as deep as necessary to fit plumbing, similar idea to width.
7. Great point about having a return pointing under the box, would not have thought of this!
8. Yes, I've read that calling a number of glass shops is best, thanks.
9. I thought about that, but I think glass will end up looking a little more professional, no?
10. Good point. I have a number of submersible heaters, and now with using the 30G as a sump, should have plenty of room.
11. Yep, part of my calculations.
12. Yeah, going to try and get everything as level as possible. Going to be extremely difficult due to how uneven the floor is.
13. My drains have to be submersed, but sounds like either way will work, and both are not required at the same time.
14. This part I'm still uncertain about - the sump design. Will expand below.

There seems to be a number of different methods for bio-filtration, or rather how it's implemented. Also a number of different types of media. Let me try and explain my plan a little more clearly. It will be laid out in a progressive/linear fashion.

1. The drains will be coming out the center of the tank and down, then over on an angle to the left (if you are facing it) side of the tank
2. Drains (Primary siphon, secondary siphon, emergency siphon) enter chamber#1 below water line
3. First area fills and flows over through foam and into chamber#2
4. Chamber#2 filled with pot scrubbies acting as primary bio-filter
5. Water flows over into chamber#3
6. Chamber#3 houses refugium/breeding area (??)
7. Water flows through sponge (fail-safe filter prior to pumps) over into chamber#4
8. Chamber#4 contains pumps/heaters/return lines

Key design point is that the water always has the option to overflow if the chamber gets blocked.

Crappy paint diagram (Brown = sponge/foam, Green = ideal water level):

Does that clarify what I'm thinking? Thoughts?

crazymittens 08-02-2012 01:58 PM

And to revisit my initial questions:

Return placement - ideal flow pattern for heavily planted tank?

1. Circular (create a circular current)
2. Placed for maximum surface disruption
3. Near the top, or bottom of the tank?

I am thinking pandacory's idea of having one flowing under the overflow box is a good idea, although will require either unattractive or restrictive plumbing. So one return (the lower-rated pump) will be on the end pushing water across the back wall of the tank, under the overflow box.

Guess I still need to do some reading on idea flow patterns for planted tanks.

Diana 08-02-2012 02:22 PM

In your drawing, do you have enough room in the sump for the water that will drain down from a power outage? I know the Bean Animal has very little water that drains down, but just making sure!

Test some flow patterns, even in a smaller tank. Toss in some stuff the floats, and some stuff that sinks slowly. See where it goes when the water is aimed in different ways.
In the picture above I would have the dark blue plumbing outlets aimed almost directly toward the viewer. The maximum flow is across the surface of the tank, then down along the front glass. If you see debris accumulating under the box then yes, the smaller pump could clear that. Or you could put a T in the line and have some water going each way.

Is this going to be viewed from more than one side? Which sides?

My concern about the Bean Animal is that the water enters it from the upper surface, and this seems to me allows debris to stay on the substrate. Not enough water movement to keep it suspended until it is removed. Great system for surface scum, though. I would not want duckweed or similar plants in this system, but I do not see a problem with the occasional leaf of something else drifting around.

crazymittens 08-02-2012 02:37 PM

Excellent points, thanks Diana.

1. Yes, I will be allowing for the excess return (and testing extensively in the garage).
2. Great idea for testing, I will definitely try that out.
3. It will be viewed from the 'long' sides (front and back). Yes, I realize one side will have plumbing - that's my office viewing angle :) .
4. I agree with the floating plants, but perhaps the duckweed could be tied to a suction cup, 'anchoring' it? (not even sure I'd use duckweed, but would be nice to have that option down the line)
5. Could the debris along the substrate be moved/pulled upwards with a high enough turnover rate? I'm hoping the shrimpies will help with sunken debris.

crazymittens 08-02-2012 06:07 PM

Ok, after more reading and thought, this looks to be an acceptable direction....

1. Returns at one end of the tank. One enters the tank below the overflow box with a 90- degree elbow to point down the back wall. One enters the tank through the overflow box with a 45- or 30-degree elbow pointed toward the front wall of the tank.
2. The two primary drain lines located at the opposite end from the returns, with the emergency drain at the return end of the tank.
3. Drains into filter sock for mechanical filtration (chamber#1=6" wide)
4. Baffle (1") then pot scrubbie bio-filtration (chamber#2 = 6" wide)
5. Baffle (1") then refugium (chamber#3 = 15" wide)
6. Baffle (1") then return area (chamber#4 = 6" wide)

1. Dual returns will allow me to keep a high turnover (dual pumps ~1200GPH, closer to 900GPH with head losses; still more than the usual recommendation of 10x capacity for turnover in a planted tank)
2. Why separate the drains? It is conceivable that a plant or large leaf could become detached, float into the overflow box and clog both the primary and secondary drains. If the emergency drain is in the same vicinity, a remote but plausible possibility is it getting clogged as well. Placing it on the other end reduces that chance.
3. The filter sock method should allow for easier maintenance/cleaning.
4. Enough room for bio-filter? Total scrubbie volume is ~6x12x12" (0.5cu.ft.).
5. Refugium will be for breeding the shrimp, also heavily planted.
6. 6" might be too narrow to fit the return plumbing/pumps and the heaters nicely, not a big deal to move the baffles if this is the case.

Moving on to other items (like pipe size calculations and sump overflow calcs), but still readily accepting input. Thanks thus far!

Kathyy 08-02-2012 06:07 PM

If the returns are placed near the skimmer then floaters will stay away from the skimmer. I accidently did that a while back. Sure seems like it would destroy the surface skimming but I didn't notice any build up at the time. Putting a *loose* screen around the skimmer box at the water level keeps floaters away nicely. It isn't attractive but probably looks better than a floater corral.

Placing the skimmer box at one end of the tank makes for a far cleaner display. When I took the background off my 5' tank I was able to move it to the end and it looked great. I cannot do that on this acrylic tank and it really makes a difference. I love jcdg's set up, external is even better......

My drain area grows fish. It turns out there isn't much turbulence in there at all and lots of food as debris collects on the sponges and grows lots of goodies. Surface area is still as the drains dump under the surface. If I pull out the heaters and take off the drain tubes it is easy to catch the fish. My sump just has drain area with large area for heaters, sponges [mechanical and biological], return area with no permanent dividers. In retrospect i wouldn't turn down a couple of dividers that would force water through the whole sponge area however.

Currently my second pump is in the tank under the internal HOB skimmer box with a spray bar located at the far end of my very long tank as I was concerned about stagnant water under the box. What if one pump's return either was a spraybar with a hole near enough the top to act as a siphon break or was divided with part as a spray bar and the rest just a nozzle return? I doubt you want a nozzle at the substrate level!

I suspect returns and drains at one end of a normal sized tank works fine IF you have the 10x flow suggested. I don't have that and am more comfortable with returns and drains at opposite ends of my 8' long tank. In the 5' tank with ~3x you could see stuff just floating down at 4' from the return so I had no faith that water was moving at the far side of the glass at all.

Shrimp will demolish the debris. My sump had the usual build up of mulm on the bottom, put some cherry shrimp in and it was GONE.

crazymittens 08-02-2012 06:17 PM

Sorry Kathyy, I must not have been clear, I meant to keep the returns at one end, the drains at the other, so that plus the high turnover should be ok. Both the sump and the display tanks are 36" wide.

crazymittens 08-02-2012 07:58 PM

Diana, I was re-thinking the returns - you had suggested placing them at either end pointing straight, to maximize surface agitation.

Will the long overflow weir not do a similar thing? Also, there will be agitation where the drains enter the sump. Still necessary to to have the returns creating surface agitation from both sides? (I am thinking one to agitate the surface, one to flow under the overflow box.)

pandacory 08-02-2012 11:24 PM

Is this an in wall build? Sounds like it is when you said one broad side is viewable from the office side.

If it is, I have always wanted to do an in wall river style tank. With overflow on the right and return on left, or vis versa. You could dial up/down the flow to suit the fish, and with the overflow on one side, you could sacrifice a little tank length but drill the drains in the bottom of the tank and do a single glass wall to create an overflow with no overhang to worry about.

crazymittens 08-03-2012 12:36 AM

Not in-wall...rather will function as a room divider, if that makes sense. My office (read: desk) is in the corner of an oddly-shaped room.

I will do up something in SketchUp when I have some time, will clear things up, I think.

Raul-7 08-03-2012 02:43 AM

Go with filter socks rather than foam. Best mechanical filtration I've ever used. You can find them in varying micron sizes, in fact they can filter everything from 200 microns all the way down to 1 micron if you choose. That's better than a DE filter. Just buy 3 or more.

Why you need all the foam in-between the chambers? Just have the drain pipe going into the filter socks.

Look into getting a media reactor, check out's Bulk Reef Supply's dual chamber reactor. I use it as both a bio and chemical reactor. My first chamber is filled with Seachem Matrix and the second with Purigen or carbon if I need it.

Also, implement a way to drain the sump to allow you to conduct water changes at the turn of a valve or switch of a second pump.

crazymittens 08-03-2012 10:56 AM

Raul-7 - you mean use the reactor housing as a bio-filter? Thanks for the filter sock vote, I could probably do away with the first bit of foam, but would keep the second as a fail-safe - just in case the filter sock completely backed up and overflowed.

The drain valve is a good idea, I will reconsider the plumbing.

crazymittens 08-03-2012 04:31 PM

Hm, I like the idea of the the external overflow box keeping plumbing out of the tank, but really don't like the thought of a box full of water, made of glass, with the only mechanical bond being the caulking. Now that I have the extra room in the tank, we'll just go with the internal overflow as planned.

Also read a bit more on surface skimming, seems like it definitely could help with water clarity, aside from what Diana mentioned about floating plants being a minor downside.

I am figuring there could be up to 70-75 gallons of water at one time, so will be going with 1.25" drain plumbing. That should be able to flow around 1000GPH, well past the 700-800GPH I'll need for 10x turnover, with some to spare. 1" flows 600GPH ( which is not enough. I've also read that 2x capacity for the drain is ideal, but with the BeanAnimal setup I shouldn't need to oversize that much.

Return plumbing design is next up, the pumps' combined lowest GPH rating is 760GPH, assuming that means with their intake adjusted all-the-way-closed to all-the-way-open? The other numbers combined is ~1200GPH. At any rate, the highest they should have to pump is 3-3.5', so 1200GPH minus losses (300GPH for height, 120GPH for two 90 elbows, 100GPH for check valves) brings it to around 750GPH, which is just about right (and I am being somewhat conservative with those loss estimates).

Also, some SketchUp sump diagrams coming soon...

All times are GMT. The time now is 12:59 AM.

Powered by vBulletin®
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
User Alert System provided by Advanced User Tagging (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.