Did I make a great or not-so-great sump???

[Viktor Jarikov]

(In the hopes of greater exposure, this is repeated from http://www.planetcatfish.com/forum/view ... 8&start=20)

A question for the very (most?) experienced and/or numbers-driven community.

If you look carefully at the photo of my garbage-toter-made-sump (see here, again, http://www.planetcatfish.com/forum/view ... 8&start=20; ~100 gal, 4.5'-high, and chuck full of ~1500 nylon mesh pot scrubbers with three woven pads on the top), you will notice that the 3" discharge pipe goes all the way to the water level. Actually it is about 3-4" below the water surface. Now that means I created a siphon.

Here is what happens:

-- every time the sump gets full to the brim, the water, of course, too rises up in the upward-going part of the discharge pipe and then starts rushing down into the pond, creating a closed (no air entry/leaks) siphon system.
-- The inlet is 1.5" fed by a 2000 gph submersible pump. The outlet, as I said, is 3".
-- The sump empties 90% of its water into the pond in under a minute, most of it in under 20 sec - the water really RUSHES out of it.
-- The water level in the pond is pretty close to the level at which the discharge pipe is attached to the sump (and close, within 6-8" of the bottom of the sump). When the siphon is running out of the gravity juice at the end of the dumping, some air is sucked into the discharge pipe (despite the pump constantly trying to fill the sump).
-- This breaks the siphon.
-- The pump fills up the sump again, the high water pushes out the trapped air out of the discharge pipe and the cycle repeats itself.

So this occurs every 90 gal/2000 gph = 0.045 h or ~2.7 min give or take. The complete cycle takes ~3.0-3.5 min or ~17 times an hour, maybe 20.

Ok. The QUALITATIVE arguments that I see go as follows:

(1) ------ I achieved almost an ultimate water flow throughout the sump and the pads and scrubbies it contains - there is no bypass, no dead, quarter-dead, half-dead, 3/4-dead, or poorly-supplied areas inside - every nook and cranny sees fresh waste-enriched water every time the sump fulls up!!!
(2) ------ But I reduced the contact time of each bacteria colony (except for the bottom 6-8" layer that's always filled with water) with the waste-enriched water and the higher they reside in the sump, the shorter the contact time is. Say, consider the top layer of pads and scrubbies, they only start seeing water when the sump is getting really full but then the violent dumping begins and they are again in the air in 0.5-1 min and idle for ~2.0-2.5 min until they see new water - that means they idle for ~40 min out of an hour. This is of course at its worst, the lower you go down the sump, the shorter the idle time is until it is zero at the 6-8" level of the discharge pipe.

****Does anybody know/able to calculate QUANTITATIVELY, which argument wins?
****Does anybody see smth else of significance that needs to be taken into account?
****Should I go on like that or should I raise the end of the discharge pipe slightly above water level in the pond, which will break the siphoning cycling???
****I could also attach a smaller diameter pipe at the end of the 3" discharge pipe to slow down the dumping - this will reduce the flow-efficiency (1) but help the contact time (2).

Thank you!!
Viktor

[MatsP]

It doesn't really sound ideal that you empty the sump that way, and I don't really think the gushing out of 90 gallons of water every few minutes is the best for the fish either...

I have a little bit of a hard time "seeing" what the construction looks like, so I'm not sure I can suggest anything really useful. One solution might simply be to drill some holes to avoid the siphon effect - even a relatively small hole (say 5-10mm - 1/4"-3/8") will give enough air-flow to break the siphon, I'm pretty sure. But without clearly seeing the pipes, I'm not sure if there is a suitable place to do this, and whether it will actually still work in the "empty back out" department.

When I built my first pond filter, I did have problems with a pattern of siphon, no flow, siphon, no flow, with a period of a couple of second - the pipe in this case is a 21mm (appr. 3/4") overflow pipe, and it's fitted on the side of the filter canister - but for some reason, it would cause this cycle. I solved this by making a much shorter pipe that runs into a larger diameter pipe, leaving a bit of air-gap between the first and second pipe, that way, no siphon effect will happen.

--
Mats

[Viktor Jarikov]

Thank you, Mats! Before we jump to solutions, I need to understand that I need a solution. First thing first: why should I alter it? What is or may be wrong/inefficient/less efficient with this cycling in concrete terms?

Also, are you implying a DIRECT effect on fish (as opposed to an indirect effect, say thru water quality)? Why would fish care? It is a 25' x 6' x 3' pond with big fish in it and they swim around as they please. In fact, I do not know for sure whether it is a good sign or a bad sign, but not at all that uncommonly, they like sitting around/under the water inlet from the discharge pipe (they sit on the bottom) and sometimes swim through the strong currents it creates when dumping water - the biggest ones like doing it the most. I am gonna sound silly but it almost looks like they enjoy that massage. They do not do it all the time, just every now and then - so I do not think they starve for oxygen - I have 6 powerful bubblers going.

M: I have a little bit of a hard time "seeing" what the construction looks like...
V: did you not see/find the photo (see link)? should I repeat it here?

Drilling a hole is nice and simple, I like it, I merely need to be educated as to why I would wanna do it.

[MatsP]

Well, I think the fish are going to be happier to have a constant flow, rather than "a lot of flow, no flow, a lot of flow, no flow" pattern.

Also, the filter isn't going to be as efficient when you only use half the filter media for half the time - sure it will be very well oxygenated at that time, but oxygen amongst the bacteria isn't much use if they have no food - they use the oxygen to produce energy from ammonia and nitrite, so they need both at the same time, not one at one point and the other at another point in time [although I'm sure the processing isn't instantaneous, so I expect that they don't actually need the oxygen at the EXACT SAME time - but essentially, they go together.

I had a look at the pictures, and I'm not sure drilling holes would work - I think it would. If the pipes can be disconnected and reconnected [or, it would of course be possible to make two cuts and rejoin the pipes, in case it doesn't work right], I'd give it a try - just drill a hole, say 3/8", at the top bend nearest the filter canister, and it should flow much more evenly, and it's unlikely to overflow - if need be, you could replace one of the elbows with a T, and have a riser pipe.

--
Mats

[Viktor Jarikov]

Let me put some words in your mouth.

M: Well, I think the fish are going to be happier to have a constant flow, rather than "a lot of flow, no flow, a lot of flow, no flow" pattern.
V: If I may, I think you say this because the constant stable flow resembles their river habitat rather than my artificial stop-and-go routine (from your posts, you appear to be big on the knowledge of natural habitats of various fishes ). Also, the water level in the pond drops slightly maybe by 1" by the time the sump fills up and then the level rises ~1" as the dumping goes on - this means slight water pressure fluctuation at the bottom, where most of them reside. You are right, sounds unnatural to me, i.e., never occurs in their wild habitat.

Yet, there is a plethora of things in this "fish prison" that are unnatural for the fish - water chemistry, temp, confinement, lighting, current, food, human presence, plastic plants, substrate, etc etc etc. Who knows if the stop-and-go flow would matter? Just to err on the safe side though, I agree, it may be wise to make things AS CLOSE TO THE NATURAL as possible, granted it will be a very, very, very far cry from the real deal.

M: Also, the filter isn't going to be as efficient when you only use half the filter media for half the time - sure it will be very well oxygenated at that time, but oxygen amongst the bacteria isn't much use if they have no food - they use the oxygen to produce energy from ammonia and nitrite, so they need both at the same time, not one at one point and the other at another point in time [although I'm sure the processing isn't instantaneous, so I expect that they don't actually need the oxygen at the EXACT SAME time - but essentially, they go together.
V: So you vote for the con argument (2) (see the ending of the original post). I am not sure why you focus more on O2 and not mention the perfect distribution of incoming food-rich water, i.e., that the bacteria will also see fresh (fresh from bacterial standpoint), the most food-enriched, i.e., NH3- and NO2-enriched water - the argument (1) says that all the bacterial colonies see both perfect O2 and NH3/NO2 conditions (but argument (2) says only for a fraction of time, which varies along the vertical direction in the sump).

You know how crud accumulates in the sump media, which leads to the zones of reduced flow or even no flow, which in turn means these zones lose their function partly or completely. Argument (1) says this does not happen in my sump - all that is in it is utilized evenly (but again, at the expense of contact time).

M: I had a look at the pictures, and I'm not sure drilling holes would work - I think it would. If the pipes can be disconnected and reconnected [or, it would of course be possible to make two cuts and rejoin the pipes, in case it doesn't work right], I'd give it a try - just drill a hole, say 3/8", at the top bend nearest the filter canister, and it should flow much more evenly, and it's unlikely to overflow - if need be, you could replace one of the elbows with a T, and have a riser pipe.
V: Thank you!!

[MatsP]

Yes, whenever possible, I try to make the tank "as natural as possible" - it doesn't mean that it's "like nature", but it's an attempt to make it as near as can be from what is available - and as you say, there are few times where water sort of gushes out, the stops, etc.

I'm also far from convinced that "even distribution" as such is particularly important. Of course, you want your media to be adequately used and distribution of wast, not that the stuff goes straight through one place, and no water circulating in other areas. When I build a filter for my (tiny) pond, I used a rectangular 4 gal header tank [which is approx 12 x 18 x 14" - so closer to 10gal in total volume - the 4gal refers to the volume with a float valve and various things]. In this tank, I put a separating wall that forces the water to go down into the media, and up on the other side, using a overflow pipe on the other side of the wall. The flow is not completely even, but it's close enough.

--
Mats

[naturalart]

Please read these comments with the understanding that I can't see all of whats happening with your pond and system. Pics and descriptions can only go so far. But I'd like to congradulate you on such a big job, I can see thats major work. And rescuing unwanted, giant, fish is worth a commendation in my book.

In the canister filter concept, the media (and the bacteria) in the canister is fully submerged. This is the case in most freshwater setups. In that case you would just drop the level of your sump to the basement floor so that at least most of your bacteria is continously submerged to the pond water level. And it sounds like you don't want to have to clean this gianormus filter and like the flushing/cleaning effect of the siphon. In this senario, I think you would still get a reduced siphon.

A second possiblility might be to just leave the filter where it is and add a drip tray to the top of your filter, mimicing what is found in a saltwater trickle sump, it might be composed of plexiglas, perforated with holes. Adjustments made by widening the holes. Then more of your bacteria would be in contact with water between flushes.

A third possibility might be to just go to a straight 'trickle' concept. Add a tray and maybe raise the sump above the pond water line. And have the outflow dump horizontal/downward into the pond.

In all cases you may have to shunt/regulate the output of your pump (and this could be redirected back into your pond for increased current).

Ultimately, its your choice: set it up so you spend less time cleaning your filter or less time cleaning the pond. As you know, with those monsters, you're going to have to clean/vac no matter what, its just a matter of whats more efficient for you.

I would also look into trying to introduce some live plants if possible. They would augment your filter system and make your fish more comfortable. Floating plants come to mind. maybe pista, Salvinia, Limnobium. Others could be Anubias, javamoss.

Again thank you for rescuing all those unwanted fish.

[Viktor Jarikov]

Naturalart, thank you for the help and kind praizes! I have an ulterior motive to "save" the big fish - I love them, which makes it a far cry from a 100% self-sacrificial effort because the fish give me a lot in return

N: In the canister filter concept, the media (and the bacteria) in the canister is fully submerged. This is the case in most freshwater setups........... And it sounds like you don't want to have to clean this gianormus filter and like the flushing/cleaning effect of the siphon.........A second possiblility might be to just leave the filter where it is and add a drip tray to the top of your filter, mimicing what is found in a saltwater trickle sump, it might be composed of plexiglas, perforated with holes...................A third possibility might be to just go to a straight 'trickle' concept. Add a tray and maybe raise the sump above the pond water line................Ultimately, its your choice: set it up so you spend less time cleaning your filter or less time cleaning the pond. As you know, with those monsters, you're going to have to clean/vac no matter what, its just a matter of whats more efficient for you.

V: thank you for nice ideas! If trickle filters work and are found to be efficient, I struggle to understand why what I have could be inefficient or not as efficient as it can be if all of media was submerged 100% of the time? I am not concerned with the cleaning at all. In fact, I had not even thought about it until you wrote about it. Sure I have to clean one way or another. My primary goal (and the goal of this post) is to make as efficient a filter as I possibly humanely can: big fish = big waste.

N: I would also look into trying to introduce some live plants if possible. They would augment your filter system and make your fish more comfortable. Floating plants come to mind. maybe pista, Salvinia, Limnobium. Others could be Anubias, javamoss.

V: Do you think it would be of significant help or some but insignificant help to the system?
-- The fish already have ~$300-400 worth of plastic plants for cover and comfort that also houses a lot of bacteria.
-- I do not have much natural light in the basement. What will survive the relatively weak artificial lighting conditions and not only barely survive, but thrive? I thought all floaters require bright sunlight to live, needless to say to thrive and multiply.

Best wishes, Viktor

[naturalart]

Hi Viktor, Well I'm glad there is some selfishness on your part also. That means all parties involved are getting something out of it. My fish say "hooray!"

I would google the scientific genus species of the beneficial bacteria(s) we are speaking about: Nitrosomonas sp., Nitrobacter sp.. I've read some scientific papers on these organisims some time ago, but can't remember all of the particulars. One thing I do remember is that some scientist consider the bacteria in saltwater is a different species from its freshwater counterpart(s); which makes sense. And they probably have different requirements for oxygen thus the trickle filter. But don't quote me on any of this.

The plants that immediately come to mind is Anubias sp. and javamoss (Vasicularia dubyana). neither of these species need to be anchored in the substrate, just tie them snuggly to a root or rock. They should be able to handle your low light. If you end up throwing silverdollars or Anostomus sp. in the pond you may loose them. I would throw in a hand full of Lemna minor (duckweed) also. If it takes it will grow like a weed and you'll be cursing me and composting it by the bucket full. But just maybe it will grow at a rate you can handle. I understand your bacteria surface area with the plastic plants plan. You don't have to get rid of them. Live and plastic plants can coexist:-)

[Viktor Jarikov]

Four months later: update

So, I played with the siphoning ("sort-of-like-trickle") approach and with a traditional approach. For the latter, I drilled a 1/2" hole (with smaller holes, the sump was still part-siphoning) in the upper, horizontal arm of the 3" discharge pipe. This stopped siphoning completely and the toter is constantly full of water almost to the brim.

In the end, I am 85% convinced that the traditional approach works better in my setup. I have no water quality problems even after huge, and I mean, huge "pig-out" meals that my cats and Co. enjoy sometimes, something like 4-5 lbs of fish and 10-15 handfuls of floating catfish pellets. The next morning NH3 zero, NO2 zero. I've had some mild problems with these parameters before even though the pond had been running 3-6 months (with heavily seeded bacteria cultures in the beginning too) and I had less inhabitants then.

The reason for the 85% certainty is that I have to repeat the experiment to gain a higher margin of confidence, that is plug the hole -> go back to siphoning regime for some months, test the water regularly, and then go back again by unplugging the hole and repeating the experiment. I am just not inclined to doing that anytime soon. Not good to "fix what works fine" but for the sake of the knowledge I should try it sometimes again.

Thanx all for your help!!!

[MatsP]

"If it ain't broke, it need not be fixed" and "keep it simple and stupid" would be applicable here, I think.

If you want to experiment, I suggest a bottle of ammonia from the local supermarket, and a small scale experiment without fish. That way (or some "sacrifical" fish of low significance, e.g. some guppies or "feeders").

--
Mats

[Viktor Jarikov]

I appreciate the concern for the fish, Mats... and agree... somewhat...
... but firing at the shooting range is not the same as firing at a live enemy... meaning that modeling only goes so far. Perhaps, stupidly, I think I could continue this experiment in a safe manner without endangering the beloved fish population. But you are right in that I should mull this over carefully and see again and again if it is worth it. Not that I have the time or desire to do it any time soon... if ever. Mostly, just thinking aloud.

[apistomaster]

Viktor,
Your original "dump" flow filter reminded me of the deliberately designed system to simulate wave action in coral reefs as I first saw described in the behind the scenes utility coral propagation tanks of the Waikiki Aquarium in Honolulu, Hawaii.

Not the ideal type of flow for any wet/dry or wet/dry/wet hybrid biological filter design. A continuous drip over a large area through high surface area to high exposure to atmospheric air produces good biological filtration and off-gasing of CO2 and to a much lessor degree, off-gasing of gaseous ammonia. These designs do have high evaporation loss of water from the system which requires closer monitoring or an automated "make up" water design element using RO water desirable. Big Catfish are forgiving but some fish like wild Discus and Altum Angelfish have lower nitrate tolerances. I always had future Discus tanks in mind after my introduction to wet/dry filtration.
Seems you found a good balance in the end. I enjoy designing DIY wet/dry systems. I knew after I got involved with reef tanks about 22 years ago for awhile that any future large Discus tanks I had would be equipped with wet/dry filters in conjunction with a good sized canister filter. Both are problematic to modern reef tank hobby as "nitrate factories" but with fresh water, making regular water changes is enough to control nitrates. Completing the nitrogen cycle by incorporating a denitrification stage filter is not as attractive and efficient in fresh water systems as they are for reef tanks. That is where my interest drops off in pursuing filtration options to the Nth degree. I have built 2 wet/dry systems, each using slightly different designs for my 2 largest Discus tanks. Each has a lot of plecos I bred that are being grown out in Discus tanks for future breeding stock.
I still think it would be fun to get involved with coral propagation but reef tanks are the monetary black holes of the aquarium hobby.
But you and any other DIY hobbyists may find this web site useful. http://www.garf.org/index.shtml
They have many DIY aquariums and filter designs where you can plug in the size of aquarium you want and a plans and materials cut sheets are generated. You can build 250 gal tanks out of plywood, glass, screws, epoxy and silicone sealant. I built about 6 150 gal plywood tanks very much like their plans but about 40 years sooner.

[Viktor Jarikov]

Thanks, Larry. Your experience is invaluable, as usual.

[naturalart]

Viktor congrats on your giant canister filter. Look forward to your experiment updates in the future. How often are you finding you're needing to make water changes?

[Viktor Jarikov]

Thank you, Naturalart! It depends on the number of inhabitants. When it used to be maybe 10 lbs total of fish, I did a 50% wc in 10-15 days. Nowadays, we might be pushing 100 lbs, so it is 50% 1-2 times a week, 1.5 times a week on average. NH3 zero, NO2 zero, NO3 5-10 ppm, pH 6.8 steady. Even after their big meals of ~5-10 lbs of fresh or boiled fish/crustaceans and ~1 lb of pellets, the water parameters do not change.

I am getting tired of this though - refilling 1000-1500 US gal of water (incoming at 40 F at this time of year) with a 40 US gal natural-gas water heater and NOT having an overflow safety in my pond is no fun. I've been mulling over installing a dedicated water line with a charcoal filter and an overflow pipe and have a trickle wc system. One problem is that I don't know if it is worth it - next summer we might/would like to be residing in southern FL, of God is willing...