calling all sump experts for design review

[lotekfish]

I’m about to embark on my first sump/fuge project using a 20g high tank that’s been collecting dust. I also just picked up a used AquaC Urchin skimmer and Mag 5 pump so this design is based on that equipment. The plan is for this to be a display fuge since I don’t have a cabinet to hide it in and it will sit beside and below the display tank. The basic idea is that all of the plumbing and equipment will be hidden behind a false back (and canopy), kind of like the nano cubes, and there is a primary flow through the rear sections and a secondary, slower flow through the fuge. This way I can use the full pump flow for the display tank circulation without blasting the fuge. I’ve seen a few designs with a similar dual flow path but nothing exactly like this so I’d like to get some opinions before starting this.

My main concern is whether I’ll get enough flow through the fuge portion using just bulkhead overflows. I know I could split the overflow line and run one line to the 1st chamber and another line with a valve to the fuge. That might be the best way to go since it gives more flow control, but I like the clean look of the bulkheads if it’ll work, and they serve a second purpose. Since I’m building this from a glass tank the baffles and dividers will also be glass so I can securely silicone them (which is why I can’t use a toothed overflow design like most acrylic sumps). I’m going to get a sheet of black plexi cut the same size as the main fuge divider so I’ll have a black opaque false back and the bulkheads will help keep the glass and plexi held together.

BTW, the main display tank is a 26g bowfront with a U tube overflow box for now, but I’ll be moving in 6 months and will be upgrading the tank to probably a 55 gallon with drilled overflows. I know a 20g sump on a 26g tank may raise issues with flooding but I plan to –fully- test this for both power loss and overflow suction loss to make sure my tank water levels are safe. I’ll drill the return nozzles to prevent siphoning and also use a flow check valve to prevent back flow in the event of power loss. I’ll also be raising the pump in the second chamber so in case of siphon loss in the overflow, the pump will only have a gallon or so (need to calculate exactly) in the chamber to pump before running dry so the display tank does flood.

I hope all of this makes sense. I’m open to all suggestions to improve the design and prevent potential disasters.

 

[DaveK]

I think you have an interesting design.

I do see a few possible problems areas.

The area for the return pump is very small. on a larger tank, there may not be enough water here to supply the main tank. In other words, that chamber will get sucked dry before there is enough return water to keep it filled.

The supply of water to and from the refugium is a bit questionable. You are feeding it with water that has not yet been skimmed. This could turn you refugium into a dirt trap. I would choose to feed the refugium using part of the water from the main return pump.

 

[RichT]

I feed my fuge from my skimmer and phosban/carbon reactors. I'm sure I'm over simplifying it but I'm thinking that puts the finishing touches on whatever the skimmer and reactors don't get.

 

[mladencovic]

I am a bit baffled by the bafles ;D . If i understand your design correctly, the output of skimmer bypasses them and goes directly into return pump's chamber. So what is the point of baffles? Bafles are used to remove microbubles, but in your case they are either created by the overflow's return and then they bypass the bafles through the fuge, or they are created by the skimmer and then sent dirctly to the return pump. Or, do I see it wrong.

 

[lotekfish]

I am a bit baffled by the bafles ;D . If i understand your design correctly, the output of skimmer bypasses them and goes directly into return pump's chamber. So what is the point of baffles? Bafles are used to remove microbubles, but in your case they are either created by the overflow's return and then they bypass the bafles through the fuge, or they are created by the skimmer and then sent dirctly to the return pump. Or, do I see it wrong.

No, the skimmer empties out the side, over the first baffle, under the second, and over the third as shown in the front view. The plan view might be confusing with the big arrow over the baffles which was just meant to show general water flow. I might add a 4th baffle so the return chamber fills from below.

DaveK:

I see you point about the possibility of the return chamber emptying before proper flow is established. I might have to calculate this out, or just use careful trial and error. I don't want a large return volume because I don't entirely trust the external siphon overflow and I don't want the sump to pump 3,4, or 5 gallons into the tank if I have a siphon break, causing an flood.

As for dirt accumulation, I've been wondering about this too, but I've been looking at how a "typical" sump works the way I understand it, which is the first chamber, where the skimmer sits, fills from the overflow. Some (most?) of this water is pumped through the skimmer and into the fuge, but a certain amount bypasses the skimmer and flows directly into the fuge, so isn't every sump/fuge a potential dirt trap? Unless the overflow is plumbed directly through the skimmer, in which case the overall system flow is limited to what the skimmer can handle. Or does my bypass design have more potential for dirt accumulation because of the larger volume and lower flow?

A good point though, I'll have to look into it more.

 

[lotekfish]

The supply of water to and from the refugium is a bit questionable. You are feeding it with water that has not yet been skimmed. This could turn you refugium into a dirt trap. I would choose to feed the refugium using part of the water from the main return pump.

Wait, I get it, it's because my fuge is filled only with unskimmed water instead of a mix. Yep, I have to rethink this.