Build a sump

[Nytelighter]

It's not tough really. Keep in mind that I was a newbie (not some experienced reefer) then who hadn't yet realized the convenience of a drilled tank and, because the tank was up and running, kind of committed to the HOB overflow approach. It just required some time and effort to think it through.

 

Anyways, it boils down to the two two main flooding situations that come into play (barring things like pipe or clamp failures). Both absolutely apply to RR and drilled tanks, contrary to common belief, and can result in wet floors. Restarting the siphon in a HOB overflow, though, is separate from a drilled tank and is an issue tackled at the end.

 

The two flood mechanisms are:

1. The sump overflows.

2. The main tank overflows.

 

The first occurs when the sump has insufficient extra capacity to hold excess water draining from the tank. The second when the tank has insufficient extra capacity to hold excess water pumped up from the sump. A well designed system will accommodate both situations. Both can also be readily simulated to test your design - highly recommended - before a real failure hits.

 

How can these situations happen? Here are a few scenarios.

 

1. Sump overflows:

a) The return pump becomes clogged and water over the height of the weir flows down into the sump.

b) The return pump stops operating (power loss, failed pump, etc.) and the return in the tank becomes a siphon. This lowers the water to the lower of the weir height or the tank end of the return pipe (unless a siphon break, which can be as simple as a hole drilled in the appropriate place in the return plumbing, is provided)

 

Mitigation methods include a siphon break, keeping the return line higher in the water rather than lower, and - most important - making sure to design in extra capacity in the sump which, during normal operation, can be viewed as empty air in the sump.

 

2. Tank overflows:

a) The overflow line to the sump clogs but the return pump continues to pump water from the sump into the main tank. (i.e. the tank can't drain.)

b) The siphon breaks in the case of a HOB overflow, or the U-tube becomes clogged. (i.e the tank can't drain.)

 

Mitigation methods include redundant overflows, screens to keep critters from getting into and plugging the overflow, having sufficient capacity in the main tank to absorb the extra water in the sump chamber with the return pump in it. This is the critical part: Limiting the water in the return pump chamber so that, if water stops flowing into the sump, this last chamber empties before the display overflows. Mind you, the return pump will "run dry" (unless you have some kind of level switch installed to turn it off on low sump level), but you won't have a spill.

 

Both are easy to test. Test #1 by pulling the plug on the return pump. Test #2 by closing off the overflow line so the tank can't drain. Watch the level of the water in the tank / sump. If either is about to overflow, return the tank to normal operation and revise your design.

 

Regarding HOB overflow restart: There are HOB overflows that are designed to restart. Some use pumps to pull air out of the siphon tube while others rely on water velocity to clear out bubbles that can get trapped and reduce flow through the overflow. Both work, and even if they failed, a well and thoughtfully designed system will still not dump water on your floor because of the provisions that you've accounted for to guard against #1 and #2 above.

 

All that said, you need to think this stuff through regardless of if you have a RR / drilled / HOB overflow. Any of these can fail and still result in lots of water on the floor.

 

 

 

Holy cow!!!, I've never thought about a scenario where the DT could overflow.  I took the time to build a bean animal overflow on my 40B with 20L sump and could never imagine a scenario where all three lines could get plugged.  But you never know…..anomalies do happen.  THanks for the good gouge, I'll definitely add this info to my book of knowledge.

[Origami]

Holy cow!!!, I've never thought about a scenario where the DT could overflow. I took the time to build a bean animal overflow on my 40B with 20L sump and could never imagine a scenario where all three lines could get plugged. But you never know…..anomalies do happen. THanks for the good gouge, I'll definitely add this info to my book of knowledge.

Bean's overflow was designed with a redundant drain as a backup. Not 100 percent failsafe, but very good. It's far less likely to fail than other overflow approaches.

 

Sent from my Rezound on Tachyon 1.1 using Tapatalk

 

 

[Nytelighter]

Very true.  Not 100% but sure does help me sleep good at night!!  hehehehehe

[bqq100]

2. Tank overflows:

a) The overflow line to the sump clogs but the return pump continues to pump water from the sump into the main tank. (i.e. the tank can't drain.)

b) The siphon breaks in the case of a HOB overflow, or the U-tube becomes clogged. (i.e the tank can't drain.)

 

Mitigation methods include redundant overflows, screens to keep critters from getting into and plugging the overflow, having sufficient capacity in the main tank to absorb the extra water in the sump chamber with the return pump in it. This is the critical part: Limiting the water in the return pump chamber so that, if water stops flowing into the sump, this last chamber empties before the display overflows. Mind you, the return pump will "run dry" (unless you have some kind of level switch installed to turn it off on low sump level), but you won't have a spill.

 

Also need to consider your ATO if you have one.  If the water is low because your return pump is moving water to the DT, your ATO could continuously add fresh water (which is then pumped up to the DT) and eventually cause a flood.  I have a low water switch which kills my return pump and ATO, plus only run my ATO a few minutes a night to limit the amount of water it could add in the case of a blocked drain.