Bad Smelling Filters

[Adriaan]

#616

Been reading the Buzunis thesis. He writes somewhere:

“In another hypothetical situation a sudden increase of water depth in the normal 5 cm standing water depth filter would result in a filter that would consume oxygen at its normal rate except that a much higher differential is required to obtain a similar gradient and thus allow diffusion to provide enough oxygen to allow the filter to function properly. This may mean that, at least in a portion of the sand bed, not enough oxygen will be available and the facultative organisms will start metabolizing substrate anaerobically.”

This woke me up to a possible explanation on why some filters start to smell within the first days of operation. Assuming the sand put into the filter is not pure, but contaminated with all sorts of biological stuff, then facultative organisms towards the bottom of the sand column will start consuming this muck anaerobically. This is because they are living in the zone beyond where oxygen can infiltrate when the filter is at rest. Hence, these organisms consume whatever it is that is mixed with the sand anaerobically, creating a stink so to say, until there’s nothing left to eat and they die themselves or migrate upwards to join the happy bunch within the oxygen rich zone. So it?s likely to be a normal phenomenon when using unwashed river sand. It should probably be solved by using clean clean sand. But then probably it will take much longer for a filter to develop a beneficial biological activity, because the filter starts of without substrate and without living organisms, until enough has been introduced by filtering dirty water. This would make for interesting research.

Another thought: when using fairly clean sand it could conceivably take very long for a well functioning biological zone to develop if the water going in is clear (containing little muck) but at the same time heavily contaminated with pathogens (although I understand most pathogens don’t just happily swim around, but sit on stuff). Reason: large influx of pathogens, but the biological layer has little to feed on (besides these pathogens)?

[Eric]

#617

That sounds like the reason. I have the thesis with me actually and I am slowly going through all of the oxygen diffusion graphs etc. It seems that increasing temperature allows more diffusion of oxygen into the water but that this is balanced out by (a) lower saturation of oxygen in the water at higher temperatures – I guess this means it disappears as easily, and (b) more biological activity due to the temperature that will end up using more oxygen to munch on things.
I am trying to figure out what that means for us in our environments where temperatures are 25-30 degs. It could be that we need to reduce the level of water to between 2 and 3cm according to one graph but I have to confirm. The reason they give for NOT doing that is one sentence – just that it will disturb the layer if water drips in and the level of water is too little….is that proven? It seems the shallower the water the better.
BTW sand filters may work here in Lumbala but the key will be affordability as there is not much of a cash economy and little alarm bells go off in my head when thinking about bartering for filters…..

Eric

[Adriaan]

#618

Actually, lower saturation of oxygen at higher temperatures doesn’t mean it disappears easily. At higher temperatures, air is less dense and therefore contains less oxygen. Furthermore, it is my guess that warmer water can not hold as much dissolved oxygen as cold water.
Regarding the depth of the sand column, Buzunis actually suggests that the biological zone in intermittent sand filters is shallower. Hence, a shallower sand column could be appropriate. This contradicts our earlier conclusion about a minimum of 60 cms to be on the safe side. Probably the best design would include the following:

– Supernatant water level as low as possible, because this allows diffusion of oxygen to a deeper level into the sand. Literature research or testing needs to prove whether a low level disturbs the schmutzdecke, leading to decreased biological performance. This would seem probable. If so, then there will be a trade-off level of supernatant water that allows maximum diffusion and least disturbance of schmutzdecke.

– Buzunis also mentions that most biological removal takes place during the pause times when no water flows through the filter. Once more than 10 litres have gone through his filter, having replaced all water previously inside, his graphs show an increase in ecoli.