Freshwater Sump / Refugium

[Toby_H]

There’s been some solid splurges at my house. But I’m laying them out one at a time…

I just ordered a 60” x 30” @ 24” tall, 195 gallon aquarium from Glass Cages (!!!)
I’ll be building a stand for it. Eventually a canopy too.
It will be themed as a Discus tank, but I want a variety of species in it. Though other species that will work well with Discus. We’ll talk stocking elsewhere later.

But the Sump! Which will also be a Refugium!!! Diagram attached. (top pic is front view, bottom pic is top view. ignore bottom pic...)

__________

I have an old 90 Gallon (4’ x 1.5’ @ 2’ tall) that I bought in 2001 and it was old then. I dismantled it and resealed it around 2008. It’s an ugly silicone job but it holds water just fine. This will be the sump.

I don’t want to silicone baffles in it, as that’s a little too permanent. So I’m going to build boxes for each end that have baffles in the boxes. These boxes will be made of Concrete Board sealed with DryLock.

I looked at HardiBoard today at Lowes and it looked cheap. More like a ½ thick sheet of paper than a concrete like product. I have 2 sheets of HardiBoard that’s several years old and it’s much more concrete like, but it’s in use. Also, I have a sheet of the actual Cement Board with the rounded edges. I started cutting it to make the boxes today.

What do you think about making a structure out of this that will be permanently submerged?

It will be submerged in a sealed aquarium, so there won’t be any significant pressure on the seals. If the seals do leak it won't matter at all. My concern is will it disintegrate over a year or three.

_________

There will basically be a 10” baffle insert on each end and 28” of refugium in the middle.

__________

To the left:

Check out the Diagram below and I’ll walk ya through the plan. Vertical pipes is the dirty water coming in. The diagram isn’t exact right there as the pipes drop the water into a 4”x4” chamber in the back that spills forward into the 4”x14” chamber in front of it. But anyway… Fist stage of filtration (gold color) are a couple Aqua Clear 110/500 sponges cut to fit. I’ll also use some basic filter fabric on top just because it’s super easy to clean/replace. Under that and under the first baffle (pink color) will be Bio-Rings and Pot Scrubbies. Honestly, I plan to just through in whatever I have laying around to fill the space with something productive. After that the water is released into the Refugium…

__________

Refugium… 28” x 18”...

Have you seen the videos of that old codger that calls himself Father Fish? I’m not promoting the guy or trying to discuss his claims. He’s out there, but… I used his idea of 3-4” of topsoil capped with 2-3” of sand. I have it in two planted tanks and Java Fern is doing amazing in both, and I’ve killed Java Fern in my sand bottom tanks with the same lighting for years. So… I want to do that for substrate… But… The point of a Refugium isn’t to grow plants, it’s to host plants that clean the water. If the plants are feasting on soil nutrients will they eat less water bound nutrients? The planted tanks I have with soil in them have near zero nitrates regardless of my water change schedule (or lack there of). Whatcha think?

I plant to put a few Java Ferns in it, Hornwart floating in the top and Pothos through the refugium and wheverelse it grows. I have a bunch of Wandering Jew (that’s what it’s called, I’m not being racist) I’ll likely add too. Basically a random variety of whatever hardy fast growing plants I come up with.

Oh, it’ll have scuds too. At this point all my tanks have them whether I like it or not. I’ve tried sooo hard to keep them out of new set ups and usually within 2 months, they’re there. I don’t mind them though.

Input on the Refugium?

__________

The right Chamber…

The baffle structure will almost mirror the left, except where the left has the receiving chamber in the back, this one will have a pump shelf in the back. We’ll talk pumps & flow rates in a minute.

On top of the yellow/tank stuff on the refugium side I’ll need to put filter fabric. And the yellow/tan stuff is Pool Filter Sand. I’m a little skeptical of this, but I want to try it. Letting the water do a slow flow through pure sand as a filtration. I mean, it is “Pool Filter Sand”. But in traditional sand filters they are a pressurized system, and this isn’t. Thus my skepticism.

On top of the Pool FIlter Sand and throughout the pump chamber will be quilt batting, to act as a water polisher and to keep the sand out of the pump(s).

I found a chart on a Geotechnical Engineering website (I’m a Geotechnical Engineer, no it wasn’t my website) that offered flow rates of water through a pure sand. If they are right and my math was right, my design will allow 420 gallons per hour to flow through the chamber.

Thoughts on anything in this chamber?

Note 1: If I set it up and the sand limits the flow rate too much -or- the sand get blown into the upper pump chamber… I’ll get rid of the sand and stagger AC110/500 sponges and quilt batting. But I’ve got to try it…

Note 2: Not shown in the diagram but, I can lower the top wall between the refugium and the pump so if the sand slows the water down it will bypass the sand chamber and go straight to the pump which will be packed in quilt batting. The water should be pretty darn clean at that stage, other than the plant matter it picks up in the refugium.

__________

Flow Rates
Note: I’m assuming a working volume of 80 gallons on the sump…

Since the sand will theoretically only allow 420 gph… which is turning the sumps volume over 5.25x per hour… let’s go with that. Which I think will be a good flow rate for the refugium. And yes I know a 420 gph pump won’t move 420 gph. I’ll use a slightly larger magnetic driven pump than I need and throttle it back with a valve to the desired flow.

A true 420 gph will be…
5.25x turnover of the sump’s volume…
2.15x turnover of the aquarium’s volume…
1.5x turnover of the total system’s volume…

__________

Clearly that’s not going to be enough total filtration on the tank. I have a plan to build a combo concrete cave / filter insert thingy to move water across the floor and collect it into a sponge chamber. I’ll build that in secret and if it works, I’ll make a post bragging about how brilliant I am. If it doesn’t work, you’ll never hear me mention it again.

__________

UV

Does anyone use UV anymore? Since it’s a Discus tank and they are super delicate, and love crystal clear water, maybe? I used them in the past breeding Electric Blue Jack Dempseys which are prone to parasitic infections. I’m convinced it gave me a significant advantage getting them past the most delicate stage. I also used to run a Magnum 250 Hang on Tank Canister filter with a polishing cartridge pushing water through a Turbo Twist 18W UV. The two together made a damn good improvement to having crystal clear water.

__________

I’m sure I’m forgetting stuff, but it’s late and this is already too long.

 

[duanes]

I find baffles and structures in a small sump a waste of time, and hardly worth the effort.

I use a 125 gal tank as a sump/refugium.
The only baffle like structure is a Porrett foam wall that separates, and mechanically filters the small pump/bio media section from the large planted/shrimp-snail refugium area.
Below when the Porret wall was added

Because the bacteria that process ammonia, and nitrite are determined by te food to micro-organismm ratio a tiny amount of bio-media ia all that is required, more media does not equal more filtration beyond the sustinance provided by fish waste., because bacteria are miniscule.
For me the most important bugaboo is elimination of elevatedbnitrate.
And to this end, only water changes, and a plethora of plants will do..
Below the same sump as above, as it is today.



And the 180 gal tank below , that it filters, with an average nitrate results using the API liquid test.
Ai admit the tank is a bit overcrowded, but the plants, along with regular water changes remedy my excessive fish collections.


Below test of the natural waters I collect them in.

 

[jjohnwm]

I'm another fan of few or no baffles; your setup is actually pretty nice. Contrast it with some of the commercial monstrosities that are sold as sumps nowadays. They have water going over and under and around and through so many goofy baffles, with so many channels and passageways, that there is hardly room for any filtration in them.

But...if I am interpreting your drawings correctly...you have the pump situated in a very small and very shallow compartment. It appears as though the flow of water is upwards through the last area of PFS, and then it overflows into that pump chamber?

If that's the case, the only place in your entire tank/sump system where changes in water level will be seen is within that tiny little chamber. If you happen to dip some water out of the tank, or remove a sizable rock or piece of decor, or simply miss a water top-up due to evaporation, the level in that chamber will drop to nothing. The pump will run dry and burn out.

In addition, you have very little spare capacity to handle the residual overflow from the tank if the power goes out. There will be a certain volume of water that continues to drain down from the tank until it stops at the height of the tank overflow, but your sump is already almost filled right up. That additional flowdown can be a significant number of gallons. It will easily overfill your sump, overflowing onto the floor. That water is now gone from the system. If the power comes back on, the pump will empty its tiny chamber in a split-second...and run dry and burn out.

Any sump should have as large a chamber for the pump as possible, to prevent these messes. I always experiment with turning the pump on and off, to see how much water volume flows down from the tank during an outage, and then mark that level on the sump to avoid adding too much water at some point and creating this problem.

 

[Toby_H]

I find baffles and structures in a small sump a waste of time, and hardly worth the effort.

Only a MFK admin would call a 90 gallon a small sump. LOL, I love this place.

Because the bacteria that process ammonia, and nitrite are determined by the food to micro-organism ratio a tiny amount of bio-media ia all that is required...because bacteria are miniscule...

I agree 100%. The only place in this I added Bio-Media is in a chamber that would otherwise be empty. And I fully believe removing it would make zero difference.

Bacteria grows on practically any surface and there are rarely set ups that do not have a sufficient amount of surfaces without Bio-Media. I've never understood why people design giant chambers of Bio-Media into their custom filters. I had a multi page thread on here 10 or 15 years ago debating that. A lot of people didn't like my perspectives. I even did an experiment doing a fishless cycle on a bare 10 gallon and just kept increasing the ammonia dosages until it failed. By the end I was adding multiple dosages per day that drastically exceeded recommended dosages for a fishless cycle. It should be documented on here too, again 10-15 years ago. I digress.......

But...if I am interpreting your drawings correctly...you have the pump situated in a very small and very shallow compartment. It appears as though the flow of water is upwards through the last area of PFS, and then it overflows into that pump chamber?

Correct... and I see where you're going...

If that's the case, the only place in your entire tank/sump system where changes in water level will be seen is within that tiny little chamber. If you happen to dip some water out of the tank, or remove a sizable rock or piece of decor, or simply miss a water top-up due to evaporation, the level in that chamber will drop to nothing. The pump will run dry and burn out.

The pump "chamber" is the whole last chamber, the pump just sits above the sand/media on a shelf. Microsoft Paint isn't the best CAD software

So I don't believe all the evaporation (etc) will come only from that chamber, I believe it will spread across the full sump... If the sand chokes the flow even slightly, the last chamber will drop and run the pump dry... I could solve this by lowering the top of the wall between the refugium and the last chamber where the pump is. This way if the sand slows the flow and the refugium backs up, it will spill over into the last chamber before the pump runs dry or it spills out of the sump.

In addition, you have very little spare capacity to handle the residual overflow from the tank if the power goes out. There will be a certain volume of water that continues to drain down from the tank until it stops at the height of the tank overflow, but your sump is already almost filled right up. That additional flow down can be a significant number of gallons. It will easily overfill your sump, overflowing onto the floor. That water is now gone from the system. If the power comes back on, the pump will empty its tiny chamber in a split-second...and run dry and burn out.

Any sump should have as large a chamber for the pump as possible, to prevent these messes. I always experiment with turning the pump on and off, to see how much water volume flows down from the tank during an outage, and then mark that level on the sump to avoid adding too much water at some point and creating this problem.

You have a very solid point about the pump being placed high too high. I'm debating you in my head that I don't believe each situation will be quite as bad as you present, it also wont be a smooth as I have drawn. And as designed, simply lowering the pump is not an option, as that will reduce the surface area of the sand lowering it's potential flow rate.

So my options are... Add another Baffle on the right side to create another chamber to serve as the pump only chamber and drop the pump to the bottom... Or get rid of the sand and use only quit batting, and lower the pump to the bottom (adjusting baffle/walls to accomodat).

well sh*t...

I don’t want to silicone baffles in it, as that’s a little too permanent. So I’m going to build boxes for each end that have baffles in the boxes. These boxes will be made of Concrete Board sealed with DryLock.

I'm losing faith in HardiBoard and similar products. I've used it in similar capacities 10+ years ago and it did fine, but the modern product just doesn't seem as reliable to resist full saturation. I know DryLock is a good concrete sealer, but I've also seen it fail when applied to plywood. Suggesting to me the surface it's protecting needs to be nearly water proof already.

I've also got some corrugated plastic (picture plastic cardboard sheets). I think I'll make one insert out of it and the other out of the concrete board I have.

The 90 gallon is set up in the back room with a lightly stock in it. I plan to put the baffle inserts in it and test flow rates well before setting it up on the big tank.



Thank you both for your input.

 

[triamond]

If there are 10 adult discus that eat 0.5% of their weight daily (which is approximately 0.35 oz of dry food with 40% protein), and considering that water plants are 90% water, the refugium needs to hold between 60 to 140 lbs of plants. Additionally, you will need to prune 11 to 29 lbs monthly and provide light, CO2, and trace elements to effectively remove all nitrates from the water. I like the idea, but the refugium should be quite large! It would be easier with a constant water changing system. I would fill the entire sump with 26 gallons of 30 ppi foam or 16 gallons of Kaldnes MBBR to vastly reduce maintenance

 

[Toby_H]

If there are 10 adult discus that eat 0.5% of their weight daily (which is approximately 0.35 oz of dry food with 40% protein), and considering that water plants are 90% water, the refugium needs to hold between 60 to 140 lbs of plants. Additionally, you will need to prune 11 to 29 lbs monthly and provide light, CO2, and trace elements to effectively remove all nitrates from the water. I like the idea, but the refugium should be quite large! It would be easier with a constant water changing system.

I see where your going here. I don't expect the refugium to remove the need of water changes. And even if/when we have enough plants to keep the nitrates at zero, there are other waste byproducts that build up that are less simple to test for. So water changes are always beneficial.

I agree with your point that I'll need a lot of plants, likely more than this refugium will hold, to get and keep nitrates at zero. But I think your numbers exaggerate the degree. specifically, you're equation assumes that the 0.35oz of food per day translates into 0.35 oz of nitrates per day. 0.35 oz = just under 10,000 mg. Plants eat about 5 mg/l per day. So that's 2000 liters or 258 gallons of plants. But, I don't believe 0.35 oz of food translates to 0.35 oz of nitrates.

I have a 150 Rubbermaid set up now that is lightly stocked and heavily planted. It can go for months without a water change and nitrates will never be detectable. But, water changes still spark breeding.

The constant water changer is an awesome addition and I'd love to do it here. But there isn't a water tap anywhere near this wall and I'm on a slab. So I'd have to run the in-line through the attic and I'd have to drill a hole in the side of my house and run the outlet into a French drain in the rear yard. I've considered it and it's just not worth it.

But you're still right, that it's very unlikely that the plants in this refugium will give me a consistent 0 nitrate reading.

I would fill the entire sump with 26 gallons of 30 ppi foam or 16 gallons of Kaldnes MBBR to vastly reduce maintenance

This part I don't agree with. The ammonia input will cap the volume of bacteria that will live in a system. It is very rare that a system will not have enough surface area to house this amount of bacteria, even without Bio-Media. Duanes also made this point above.

So adding additional Bio-Media will not increase the quantity of bacteria. Therefore having a zero net impact on maintenance. For reference, the 150 Gal planted Rubbermaid that consistently has zero nitrates above as a single powerhead as the only form of filtration. And there are Wild Caught Apistos happily breeding in it. If it's clean enough for them it's clean enough for me.
(though the plant mulm does build up in pockets on the floor.)

 

[triamond]

But, I don't believe 0.35 oz of food translates to 0.35 oz of nitrates.

10g of 40% protein dry food gives 10 * 0.4 * 0.17 * (14 + 3 * 16) / 14 g = 3g of nitrates (17% nitrogen in protein). For a 195-gallon aquarium, this results in approximately 4 ppm of nitrates daily, or more than 5 ppm when considering the actual water volume. This results in a total of 150 ppm of nitrates monthly. The growth of plant mass was calculated using data from experiments in Diana Walstad’s book, "Ecology of the Planted Aquarium."

This is just an idea… a refugium can hold the roots of large pothos, preferably growing somewhere outside if the climate is mild Lots of CO2 and sunlight.

This part I don't agree with. The ammonia input will cap the volume of bacteria that will live in a system. (...)

I agree… if we consider biofiltration as the process of transforming ammonia into nitrates. It’s a fairly easy process, and not much space is necessary for autotrophic bacteria/archaea to achieve undetectable levels of NH3/NO2-. A six-inch, 30ppi sponge should do the trick for 10g of food a day!
However, the filter is also supposed to degrade uneaten and undigested food and feces into CO2, NH3, etc. This is the role of heterotrophic bacteria/archaea. Much more space is required (20-100 times), as this process is more complex. The less surface area available for bacteria, the more dissolved organic carbon (DOC) and bacteria/protozoa will be present in the water. At some point (10 liters of 30ppi foam per 1g of food?, months of growth), all the energy from wastes is used for the basic metabolism of the bacterial colony, leaving no excess for expansion. Virtually no need for cleaning. Best water quality limited by the turnover alone.
It may be important, as discus have developed the ability to survive in very soft water. Their gills prevent them from swallowing due to osmotic pressure, at the point where most bacterial cells cannot survive. As a result, there was not much evolutionary pressure to eliminate those with weaker immune systems. Therefore, discus are more susceptible to bacterial infections, as they are used to living in more sterile water.
There is another benefit of overfiltration. Bacteria are in a symbiotic relationship with algae, e.g., providing them with iron. The fewer bacteria in the water, the less algae.
By the way, I hadn't touched my filter for 3 years; the water was crystal clear. Then I changed most of the Kaldnes (static) media to foam because I was bored to death with no maintenance on the filter, hoping for another 3 years.

 

[Toby_H]

Triamond, Thank you for the time you are putting into your responses here. As we move forward, I'm not arguing against anything you're saying. I'm pushing to get a better understanding. You're clearly not talking out of you tail and have a lot of background info on the topic.

10g of 40% protein dry food gives 10 * 0.4 * 0.17 * (14 + 3 * 16) / 14 g = 3g of nitrates (17% nitrogen in protein). For a 195-gallon aquarium, this results in approximately 4 ppm of nitrates daily, or more than 5 ppm when considering the actual water volume. This results in a total of 150 ppm of nitrates monthly. The growth of plant mass was calculated using data from experiments in Diana Walstad’s book, "Ecology of the Planted Aquarium."

I'm a math nerd too But one with far less background on this topic than you have. So correct me where I'm wrong.

I think I did the same math as you just now. 10g of food is 3g or 3,000 mg of nitrate. divide by aquarium volume is 4 mg/l or 4 ppm. Details below for other interested parties. But, when you added the sump, that increased the volume thus dividing by a larger number. So the 4ppm (daily) turned into 3 ppm, not 5. So 90ppm monthly not 150. But, small detail and we're still on on the same track.

10g food = 3g nitrate
3g nitrate is 3,000mg nitrate
I use 270 gallons for my system (195 aquarium + 75 sump). 270 gal = 1,022 liters
3,000 mg of nitrates in a 1,022 liter system is 2.935 mg/l, call it 3 mg/l
I've read a few places that 1 mg/l = 1 ppm. So that's 3ppm per day. That's for the aquarium & sump.

This still sounds high. But not by a whole lot.

Our model above is based on two accepted details:
We feed fish about 0.5 of their body weight daily. I read up on that and found 0.5-1%, so no argument there.
And 1g of food = 0.3g of nitrate, which I couldn't confirm/deny with a simple Google search but it seems reasonable.

Lets use these details for a single adult Oscar. The interweb tells me an adult Oscar weighs 3.5 lbs, or 1,587 grams.
0.5% of 1,587 is almost 8g of food per day... save some math... we have the math for 10g of food, so 8g of food should be 80% of that.
So, 1 Oscar in a 270 gallon system should produce 72ppm of nitrates per month.
Convert that to a 75 gallon system (multiply the ppm of a 270 gal by 270, then divide by 75), that's 259.2, so 260.
This says a single adult Oscar in a 75 gal tank will produce around 60-65 ppm of nitrates per week.

I think 60-65 ppm of nitrates per week for a single Oscar being properly fed in a 75 gal with no plants is high.

Not ridiculously high. But high. I'd expect 30-40.

Feel free to convert "adult Oscar" to any other fish that weighs 3.5 lbs. I just used Oscar because a large percentage of us have kept an adult Oscar at some point so we have that reference. But many of us also overfed that Oscar so we have a skewed opinion of how messy they are. They're only that messy if we give them the food to make a mess with. This model assumes proper feeding.

Anyway... all of that does not discount what you are presenting. It is only how I am processing the details. You're still absolutely right that my refugium is too small to process all the waste. And thank you for the thought exercise.


But... will dedicating 40ish gallons of my sump to a refugium offset the waste to a degree that makes it worth having?

 

[Toby_H]

This is just an idea… a refugium can hold the roots of large pothos, preferably growing somewhere outside if the climate is mild Lots of CO2 and sunlight.

The plan is, to have the 40 gallons packed with water plants, and Pothos rooted in the refugium growing beyond the water. This makes it easy to prune as I can just hack off a majority of what's outside the water and not disturb what's rooted in the refugium.

I'll likely also plant the tank. I have a few pretty large Java Ferns and a couple smaller ones growing out. I have a ton of HornWart I may float in the top. There's also a plan to eventually build a canopy that holds a few potted plants that dip into the top inch or two of the aquarium (the fiancé loves house plants).

But again, even IF I can get enough plants in this system to keep nitrates at zero, I'll still do regular water changes. Fish in general just respond well to water changes, even when the starting water is nitrate free. And Discus are well known to require a high water change standard to thrive. The refugium is just another tool in the basket.

I agree… if we consider biofiltration as the process of transforming ammonia into nitrates. It’s a fairly easy process, and not much space is necessary for autotrophic bacteria/archaea to achieve undetectable levels of NH3/NO2-. A six-inch, 30ppi sponge should do the trick for 10g of food a day!

Back in the day, before all of our modern bio media was around... The nitrogen cycle still took place in our tanks just fine. And throughout time, I've had countless tanks that had sand and rock decor with 1 or 2 HOBs without any "Bio Media". Thus I work off the opinion that a typical set up has more than enough surfaces for the bacteria necessary for NH3 (ammonia) to NO2 (nitrate) conversion. And spending time, money or space with "Bio-Filtration" to perform this conversion is a complete waste. The exception to this might be purely bare tanks such as some Discus breeders use.

However, the filter is also supposed to degrade uneaten and undigested food and feces into CO2, NH3, etc. This is the role of heterotrophic bacteria/archaea. Much more space is required (20-100 times), as this process is more complex. The less surface area available for bacteria, the more dissolved organic carbon (DOC) and bacteria/protozoa will be present in the water. At some point (10 liters of 30ppi foam per 1g of food?, months of growth), all the energy from wastes is used for the basic metabolism of the bacterial colony, leaving no excess for expansion. Virtually no need for cleaning. Best water quality limited by the turnover alone.
It may be important, as discus have developed the ability to survive in very soft water. Their gills prevent them from swallowing due to osmotic pressure, at the point where most bacterial cells cannot survive. As a result, there was not much evolutionary pressure to eliminate those with weaker immune systems. Therefore, discus are more susceptible to bacterial infections, as they are used to living in more sterile water.
There is another benefit of overfiltration. Bacteria are in a symbiotic relationship with algae, e.g., providing them with iron. The fewer bacteria in the water, the less algae.
By the way, I hadn't touched my filter for 3 years; the water was crystal clear. Then I changed most of the Kaldnes (static) media to foam because I was bored to death with no maintenance on the filter, hoping for another 3 years.

I'll admit... this is getting a bit over my head. But I'm doing my best to keep up.

After a few google searches, do you agree that in laymen's terms heterotrophic bacteria/archaea is the bacteria that feasts on solid waste breaking it down into smaller parts more readily available to be processed by NH3-to-NO2 bacterial process we all know and love.

If so, isn't this bacteria mostly grown on the waste itself? It seems to me this bacteria will thrive in the mechanical chamber and will be a bit out of place in a chamber in the sump that does not target the collection of physical waste. And we wouldn't need to provide high surface area media, as it grows in, on, around the waste itself.

I also read: " Heterotrophic bacteria can reproduce very quickly, sometimes doubling in population every 20 minutes."

 

[Toby_H]

View attachment 1555392

So, I stole this quote out of another thread, don't tell those guys.

This was offered as the base equipment for an algae scrubber. What do you think about me adding this to my Sump? But instead of algae I'd use Java Moss. Reason being, A) I don't know anything about algae, would have to learn what type to use, struggle to identify the right algae from the wrong algae and the algae I grow would just be thrown away. B) If I had a little Java Moss farm, the moss I pull out I could use in other set ups or use as currency/bartering with Local Fish Club members.

I'm pretty sure I could work it into the sump somewhere. Pumping the water back into an earlier chamber in the sump. I want to experiment with the Sand Chamber before I consider where to put it. I ordered more of the corrugated cardboard I'm going to use to make the chambers, it'll be here Monday. I thought I had more laying around but it's all cut pieces, not enough full sheets to bother starting.