Amos' Five Foot Tank - The Sump

A terrible tale of woe and wet floors.

Once the tank was on the stand in the garage I set about building the sump. My plan for the sump was basically to make it as big as possible, run water through it fairly slowly, and avoid too much opportunity for gas exchange. This is not the normal way sump filters are designed, but my logic was something like:

• I've never had ammonia or nitrite problems with any densely planted tank - I think the plants help water quality tremendously.
• With a deep substrate and under-gravel heating there's a very slow exposure of water to massive surface area in the substrate.
• I stock fairly lightly anyway (though Amos does spew clouds of meat out his gills when he eats, so maybe this isn't such a great reason).
• When I've tested oxygen levels in tank water, it's always been much higher in my planted tanks than in un-planted or lightly planted ones. I don't think I need atmospheric exchange to avoid my bacteria being limited by their oxygen demands.
• I don't want to lose CO2 from the water to the atmosphere.
• I'm not desperate for crystal clear sparkling water - I don't need dozens of changes an hour through a mechanical polishing filter.

I ended up with a sump containing a few over-and-under dividers at the start, to try and catch large floating and sinking debris, then a couple of chambers capable of holding about 40 litres of media each, then a final chamber of about 60 litres. The dividers between chambers are arranged such that water flows up through the media in the chambers, with the media remaining fully submerged. The idea here is that debris still in the water stream will tend to fall out of the flow.

In total, the sump will hold about 220 litres of water before it over-flows, with an operating volume of around 170 litres. If any of the chambers clogs completely, it will overflow into the next without the water escaping (that's the plan, anyway).

Only the first chamber had media - 10 US gallons (38 litres) of bio-balls. The tank itself contains 570 litres, so I've got about 1 litre of bio-balls per 15 litres of tank water, or 1 per 20 total water volume. Obviously bio-balls are normally trickled, so their performance may be different in my tank, but I think that's adequate. If I decide I want more biological filtration, I'll either stick something in the second chamber, or (more likely) add a fluidised bed.

Sadly, although I did a reasonably long-term inetgrity test (48 hours running with water to the brim) with no problems, a few weeks afetr the tank was in place in the living room, the sump blew a seam. I really don't want to talk about it - there was quite a bit of water in places it shouldn't be (ie, all over the floor of the house) and quite a lot of rushing around sorting it out. The custom sump was replaced with a couple of plastic storage crates. I've ended up with slightly more bio-balls but with the water flowing less uniformly through them, and with no settlement areas first. Actually, for my tank there are so few fish and so many plants, I could probably lose the filter altogether. Oh well, we live and learn.

To comment on anything (please do) email ian.web@astounding.org.uk