Please critique my DIY sump/refugium diagram

[peanut1181]

This is my first attempt at a sump that will be partitioned to include a refugium. Lemme give you a little info to supplement the diagram (if the picture even posts correctly.) I have a drilled 75g will built in overflow, soon to be a reef tank. Water leaves the main tank via overflow and into the sump which is housed under the tank in the stand. The water will pass through a few tiers of live rock (This system of tiers will be built using egg crate and plexiglass. It should be fairly sturdy.) and out the bottom where the heaters sit. It will then pass through a sponge and under the first baffle into the skimmer chamber. It will then overflow onto a piece of plexiglass that has holes drilled in it to allow a small amount of water to fall into the refugium while still forcing the majority of water past or around those holes, bypassing the refugium, into the chamber where the return pump will push the water back into the tank. I figure this will allow for a high flow rate through the sump and a very low rate for the refugium, as needed. On the top couple of inches on the baffle separating the refugium and return pump chamber there will be a gap of a few inches between the baffle and the sloped piece of plexiglass over the refugium to allow the water to overflow freely from the refugium into the return pump chamber. There will be a series of three "pillars" cut into the refugium/return chamber baffle to support the sloped plexiglass... one on either side of tank and one in center (much like three pillars holding up the roof to a front porch on a house -- just for visual help.) I have a rio 3100 return pump but am thinking about upgrading to something with a higher flow rate? Yes? No? Please give me any and all input you have about this setup. I am open to suggestions and eager to learn. Thank you very much for your help.
Gray

 

[broomer5]

Very interesting design Gray
I see some advantages and disadvantages to going with this design.
1)
From your drawing, the first tiered chamber of live rock measaures about 20" x 12" x 6"
That's about 1440 cubic inches of space.
That's about 6.25 gallons of water it will hold.
The Rio3100 flows about 700 gph @ 3' head.
That's roughly 11-1/2 gallons per minute.
This means the very first chamber must allow at least 11-1/2 gallons to pass through this sponge per minute - or you will overflow the first chamber.
I'd watch this carefully, and keep the sponge clean.
You may or may not want to use the sponge. Placing a baffle instead of the sponge will allow a place for detritus to accumulate, and you can suck it out of there when doing cleaning, possibly eliminating the need for the sponge.
Installing a ball valve on the return pump discharge line is a good idea. No matter what happens - you can tweek the flowrate down some should the need arise.
2)
The sloped roof over the refugium may need to be lower in the sump. Depending on how much water will drain back to the sump upon power loss - you may have to run this sump at a lower level than shown in the drawing.
If the sloped roof is too high, the sump will run at that level, and when you lose power - you may overflow the sump.
Lowering the roof down so the sump runs half full may prevent this from happening.
If the tank's a standard 48 x 18 x 20 inches, that's almost 4 gallons per inch ( 3-3/4gallons per inch in the display tank).
The sump appears to be a little less than 2 gallons/inch.
So, overflowing the sump will depend on how many inches the water drops in the display tank upon power loss. This value is dependent on the design of the overflow(s) in the tank, their capacity, the return pump you're using, and as mentioned the total surface area of the water surface in the display tank itself.
Making the roof adjustable with these pillers may allow you to play around with it during start up, and adjust the roof height as necessary. This would be an advantage.
3)
The water leaving the refugium through the hole on the end of the divider, and the water falling over the edge of the roof into the return pump chamber - both will create a ton of bubbles.
This is common - and will present you with other challenges.
You may want to re-think this last part of the design.
You may want to consider decreasing the length of the refugium chamber enough to put in some over and under baffles. This will give you a little more "dwell" time, allowing the bubbles to rise to the surfaces an pop before entering the return pump chamber.
4)
I'm trying to invision this in 3-D, with the water flowing down from the tank, through the live rock, hopefully passing through the sponge, up and over the high side of the roof, some water trickling through these holes, most running down the sloped roof as a sheet of water, falling over the low side of the roof and into the return pump area. All while this is happening, a percentage of the water that trickles into the refugium will overflow at the same far end into this pump area as well.
I can see it happening.
I would think hard about what happens when the power goes off.
Overall - I like the design, very original, and if you allow yourself some flexibility on making changes - such as making some adjustments during start up - I think it would be a pretty cool sump/refugium.
How are you planning to light the refugium ?

 

[peanut1181]

Hey Broomer, thanks a lot for the great response. I hadn't thought about the rate of flow through the sponge... you are right, that will be a problem. And I hadn't thought about making the pillars adjustable either... good point. Also, lowering the refugium walls to accomodate for overflow during power failure... do you think 12" water line would be too much? I shouldn't have an overflow of more than about 8 gallons. I think that should leave about enough room... got any cool formulas to figure out how many gallons I would be able to handle then? About decreasing the refugium length and adding extra baffles to get rid of the bubbles in the return line... I tried purposely to leave as much room as possible for the refugium and therefore have no addition baffles and had actually thought of the bubble problem on my own (score 1 for me!) but had not really come up with a solution that does not require any more space (just lost my 1 point.) Back to the drawing board... Thanks so much for your help. I will keep you posted on the changes. If you come up with any other ideas please let me know. I truly appreciate the help. Oh, and as for your 3-D picture of my system... when you describe it that way it sounds so elegant! COOL
Gray

 

[peanut1181]

Anthem,
I have done a little bit of research through the archives and have come up with differing opinions about bio balls... some say they are great, but others say they just nitrate factories... what is your opinion? Why, if live rock and bio balls are essentially serving the same purpose in this application, are bio balls better than live rock? both are just providing an area for bacteria growth... the only difference being that the live rock may break things down a step further than the bio balls... what are your thoughts? If you could also take a look at the rest of the setup and share your thoughts... greatly appreciated. Thank you for the constructive criticism. Keep it coming.
Gray

 

[peanut1181]

Anyone else have any bright ideas? I could really use some insight and wisdom on this one. Thanks.