Paul G

New member
I'll just drop this here and we'll see what comes of it.

pH 6.6 typical
ORP 340 - 390 mv typical
CONDUCTIVITY 750 uS/cm typical
OXYGEN 6 - 10 ppm
CO2 50 ppm typical
GH 110 ppm typical
CALCIUM 75 ppm typical
MAGNESIUM 30 ppm typical
Ca:Mg 2.5 : 1
KH 120 ppm typical
IRON, total 0.30 ppm typical
NO3 5 - 10 ppm typical
PO4 3 - 5 ppm typical
POTASSIUM 40 - 50 ppm typical
SO4 > 200 ppm

Source Water : RO / DI
Total Filtration Turnover : 1200 GPH minimum
Substrate : inert siliceous gravel w/ high CEC calcined clay and granulated laterite
Chemical Filtration : no carbon or resins of any kind are used
Last edited:

Paul G

New member

Echinodorus 'ozelot' (six individuals)
Nymphaea lotus, rubra
Anubias barteri
Anubias gracilis
Cryptocoryne wendtii
Hygrophila difformis
Ludwigia repens
Sagittaria subulata
Hydrocotyle luecocephala

I regard the swords and the nymphaea to be the keystones of the ecosystem. The occurrence of any kind of algae is inversely proportional to the productivity of these large plants. There is currently no nuisance algae in this tank. Some green spot and a little bba were evident up to about a year ago, but died back when the big plants settled in and started sending out leaves. Large leaves are regularly removed to let in the light. I try to limit floating leaves of the nymphaea to five or six; new leaves get very large very quickly.

In near future I will be putting in some Java fern, Microsorum pteropus, if I can find some really healthy robust pieces.

Last edited:

Paul G

New member
I use Neptune Systems Apex Aquacontroller to manage this aquarium. As many functions as lend themselves to it are automated. Here is the dosing system. There is a lot of DYI going on here. I will be happy to share details if there is further interest.

Each additive/supplement is contained in a dedicated 1.5 liter vat (small AquaMaxx). The concentration of each is consistently maintained while the dose rate is determined by the run-time of the peristaltic pumps (Milwaukee). The doses are regulated by Apex timer programs. Dosing is daily. All feeds are done at 06:00, except PO4 which is done at 18:00 to minimize trace removal by precipitation of insolubles. UV is turned off from 06:00 to 12:00 to prevent premature chelate denaturation. The additives are

KH BOOST (Sodium Bicarbonate)
GH BOOST (Sulfates of Calcium, Magnesium, and Potassium)
ORGANICS (50:50 mix SeaChem EXCEL and ADVANCE)
N : 0 : K (Potassium Nitrate, no Phosphorus)
0 : P : K (Monopotassium Phosphate, no Nitrogen)
(The eighth outlet is an unused spare.)

The dosing rates (run times) are determined over time, depending entirely on the apparent health and productivity of the plants. I test everything frequently. If something looks like it's running behind or ahead of demand, I can easily adjust the daily dose by changing the timer (one program line in the Aquacontroller). This routine is inherently adaptable to the exact tank conditions, types of plants, biomass, etc. It requires some methodical experimentation, but in the long run it eliminates a lot of guesswork.
Last edited:

Paul G

New member

The pumps are Iwaki MD55-RLT. All loops are closed recirculating high-pressure siphon feed. There is no net head loss for lift. All losses occur in the piping and apparatus. Conduit is 1" ID with few exceptions. The nominal flow rate in each high speed loop is 450 gph, varying with condition of filters of course. Note that water in these loops is transferred to opposite sides of the tank for maximum mixing. The processing loops run about 170 to 210 gph, depending on the condition of the filters, with the loop having the chiller typically being the slower of the two. The chiller was necessary because of heat from the lights, due to the use of a closed canopy. More about this later.

The returns are spray bars below the waterline. They are rotatable so the flow may be directed. These are adjusted to affect 'RIPPLE' not 'CHOP' at the surface.

I use LifeGard QL UV-C sterilizers, a 40 watt and two 15 watt units. I like the QL types for ease of maintenance. My opinion here is that the fish are protected better from incidental infections than without UV, but I must remember to shut it down when dosing bacteria.

The water is polished to 25 microns. One such filter in each processing loop is adequate to maintain this polish. The high speed loops filter to 100 microns and keep the mix uniform.

The biological filtration in this system is robust. The plants are being forced to use nitrate exclusively, as ammonia is removed instantaneously. I provide them sufficient energy for this. The Matrix (SeaChem) is bagged, and is rinsed with every filter change. The bioballs (Coralife) do not have near the surface area of the Matrix, but they have "open" structure, so do not clog and resist channeling. They are not bagged. I only open these canisters once every couple of years just to check the smell. They are never anaerobic - ever. They are now nine years old.
Last edited:

Paul G

New member

Pictured above is from this morning before I took the shears to it. The scape was looking pretty dark. I have let too many nymphaea leaves go unclipped. Quite a lot of overspreading tops in the Hygrophila and Ludwigia as well. I decided to prune just enough to let some light in. Some maintenance is scheduled for next week and I will get more aggressive then.

Ludwigia is an attractive red plant and hardy. It's fast growing. I've got it bushing out at the top and spreading out at the surface. I think this is pretty much how it likes to get its light and is a normal growth habit. It is pleasing to look at the dense growth from above, but it casts a lot of shade. It has a nice effect as a background plant because it is a good "filler," and I like it with the Echinodorus. I've been keeping Ludwigia for several years. It is eventually necessary when it is time to refresh the scape to allow the Ludwigia to overgrow in order to harvest leafy stems of good length for replanting, and pull out the old weak stumps. One downside to this plant is that it sheds leaves more readily than I'd like. But I appreciate its appearance in my scapes and I'll go on keeping it.

I scraped the front glass this AM. Way past due as you could see the Garra's mouthprints in the aufwuchs. While that is kind of interesting, it's a reminder that this needs doing every three days or so.

Tests today show that the iron is gradually falling off. The plants are devouring the stuff. This AM iron was 0.07 ppm, so I ran the Fe GLUCONATE doser for a full minute, then revised the daily dosing from 30 seconds to 40 seconds for both Fe GLUCONATE and MICRO/TRACE.

PO4 is doing okay, but I'll be watching the NO3. If NO3 falls below 5 ppm, I'll bump it up too.
Last edited:

Paul G

New member

My personal preference for the overall appearance of the aquarium is to fully contain the light, to have a "viewscreen" aspect. There is a lot of energy being given off under this canopy and the heat does build up quickly. The air temperature above the water surface during daylight soon reaches 89 F, perfectly representing the tropical condition. Note the temperature probe. Above 89 F, the Apex is programmed to run six 5" fans installed on top of the canopy. The chiller is needed to pull down the water temp by only a couple of degrees. It does not run long or often, but without it the temp could creep up to higher than 80 F, especially in summer. I keep the aquarium water at 77 F.

The system diagram pretty much explains the lighting arrangement. The three Build My LED (BML) are 6300K high output, purpose built for dutch tanks. The 24V 60" strips are each comprised of five 12" Ecoxotic Stunners, all fitted with reflectors. The 24V and 12V power supplies are Automation Direct Rhino.

The turn-on schedule is TWILIGHT > DAWN A > DAWN B > SUN 1 > SUN 2 > SUN 3 > SUN 4. The turn-off schedule is SUN 4 > SUN 3 > SUN 2 > SUN 1 > DUSK B > DUSK A > TWILIGHT. All phases except SUN 4 are ramped brightness, up on turn-on and down on turn-off. The current programming has DAWN starting at 7:15 and DUSK full off at 22:00, and SUN 4 full on at 10:30 and full off at 18:30. The BLMs are presently set for 60% of maximum brightness and 30 minute ramp slopes. These are user programmed as 'profiles' in Apex. The ramp slopes of the Current USA timers are factory fixed at 15 minutes.

The BLMs are approximately 7" above the waterline, while the Stunners are at 8".
Last edited:

Paul G

New member

This AM doing maintenance, testing, and pruning. These are the intakes, three just removed and their replacements. The dirty ones show what to expect in a planted tank. There is a constant evolution of detritus and it will clog the holes as it gradually decomposes in place. The flow meters show a reduction. When I did this today, the fast loop rates were almost at 400 gph, but returned to 480 gph. The processor loops recovered about 10 gph, but their rates are below normal, which means the 25 micron filters are due for replacement.

Loose muck is rinsed off the dirty intakes, then they are stored in straight bleach. When they come out they are rinsed thoroughly and go back into service. The rotation cycle is typically around three weeks. It does take about two weeks in undiluted bleach to completely clear the holes. There is no brushing involved, as it's not very effective at clearing the holes. The clean intakes do not go back into the tank until there is no hint of bleach smell on them.

These are 1" pvc pipe with 1/4" holes drilled in them. The mesh covering is standard fiberglass insect screen, wound tight and cemented with aquarium grade silicone adhesive caulk.

The bleach is reused. The bottles are 2" pvc pipe long enough to fully submerge the intake. After a few rotations, the bleach tends to neutralize. When it looses its strong smell, I replace it with new. This method economizes on bleach and the bottles can be capped and stowed out of the way.


Paul G

New member
Yesterday morning I ran the iron doser for one minute and shut down the UV. After today's dosing, the iron is 0.56 ppm. This is about twice the normal target concentration. The iron test kit is Hanna H1721 Checker. This is an easy way to get really accurate results.

Nitrate is not keeping up despite recent increases, so it will get boosted again. I have set the doser for one minute a day.

Phosphate is holding at approximately 3 ppm according to both Fluval and Salifert test kits.

Potassium is holding at over 50 ppm. This is a LaMotte turbidity-based test. I have confidence that it is accurate within + or - 5 ppm, which is plenty good for my purposes.

KH is 6 degrees, not much changed from last test. A small amount of sodium bicarbonate is dosed daily. Normal acidification is being adequately buffered.

GH is 7 degrees. Again, no change. A dose of booster is done daily as the plants of course uptake magnesium and calcium.

These hardness tests, reported in degrees, are API test kits which I trust to give reasonably accurate results and are fast and easy to do. I can pin these parameters down to actual ppm numbers with LaMotte titration kits. I use the titration kits often enough to detect drift in the parameters, but I rely on the API kits for routine ball-parking.
Last edited:
  • Like
Reactions: Ray

Paul G

New member
I have been experimenting here with iron ferts. In the following I make reference to claims made by SeaChem. To verify my citations go to the Discussions section of their website and surf around in the Sunken Gardens forum. Not all the info available from SeaChem is printed on the bottle.

As previously noted, I turned off the UV Wednesday AM, and as of Thursday AM I had brought up the iron to 0.56 ppm. No iron ferts were added until this AM. Everything else was done regularly and the plants spent those two days under the normal light cycle. I tested several times throughout the day Thursday and Friday. Friday evening the iron tested at 0.52 ppm. No change. The UV was turned back on at that time and ran all night until its regularly scheduled 6:00 AM shutoff. Then I allowed the normal dose of MICRO/TRACE which contains Ferric EDTA, but withheld the Ferrous Gluconate. I tested 4 hours later, after the lighting was well under way, and the iron was back down to 0.07 ppm.

SeaChem asserts, and I accept, that Flourish Iron is the superior iron fertilizer for foliar uptake because it is a carbohydrate that plants can metabolize readily, thereby being a good carbon source as well as having the advantage of releasing iron in the ferrous state. Something I have noticed since I continuously monitor ORP (redox), is that the ORP drops a full 100 mV every morning at dosing time (6:00 AM). The moment the gluconate goes into the water, the redox drops like a rock, then gradually rises to its normal "real" value by midday. Both days the gluconate was withheld, the redox did not fall off and has stabilized at about 380 mV on a flat curve. The ORP meter is keeping tabs on the relative abundances of reducing agents and oxidizing agents. In a system that is highly oxidative as a rule (in which plants produce lots of O2 and keep the dissolved oxygen at a high level day after day) a reduction in the redox can be interpreted as an increase in DOC, which is comprised largely of reducing agents. Gluconate is a dissolved carbohydrate and is a reducing agent. No other additive used, including Excel (gluteraldehyde), affects the redox in this way.

SeaChem asserts that the plants will consume a prescribed dose of Fe Gluconate so rapidly that it will virtually disappear from the system in half an hour. Of course, I think this depends on some variables such as rates and patterns of water circulation, and, as I assume the plants must be photosynthesizing and metabolizing normally for this to occur, the total plant mass, character of the light, and other things. In any case, I accept that the plants will uptake Fe Gluconate rapidly and I fully expect it to be consumed within a couple of hours in my system. SeaChem states that if you test 0.1 ppm iron within one half hour of dosing Flourish Iron, that is a sufficient dose of iron for the aquarium. This may be a better way of stating the case. But I should think that a heavily planted tank with highly productive specimens could do with a somewhat larger dose than that.

For cations to be bioavailable, they can be complexed with relatively stable organic molecules to preserve them, at least for a while, from being immediately bound to anions in the solution and present them to the plant as food it recognizes and can use. Both EDTA and gluconate are subject to photodissociation. Probably a UV sterilizer is a more expeditious way to get that done than is overhead visible light, but both will do. The gluconate ligand is probably fairly weak, which is a desirable characteristic in a plant fertilizer, as the idea is to require as little energy input from the plant as possible in order to get the iron ion out. Also the gluconate is a carbon source, and is likely a source, rather than a sink, of energy. EDTA is a hexadentate chelate and is a strong ligand. The plant has to pry the iron loose and then reduce it to ferrous state before it can use it. This requires energy from the plant. Many of the desirable metals in micronutrient fertilizer are made bioavailable by EDTA, or similar, chelation. Probably all such complexes can be decomposed by exposure to light. Some ligands are stronger than others, and rates of decomposition may be intensity and wavelength dependent. When these substances are exposed to light, the molecule is denatured and oxidizing agents and orthophosphate anions grab the metal, forming insoluble inorganics that precipitate out of solution and are no longer bioavailable, at least in terms of foliar uptake.

SeaChem says that you should not keep iron and phosphate ferts in the same bottle. Since Flourish iron is consumed so rapidly, there is little opportunity for it to decompose in the water column, so there isn't much concern about the timing of phosphate dosing. I think that does follow from the theory of what gluconate is and how it works, but I still dose these two nutrients 12 hours apart. It's just good karma. I also suspend UV exposure from 06:00 to 12:00 every day to maximize the gluconate's presence; that is, I give it 6 hours - 4 hours with lights on - before I subject it to UV.

Pretty clearly, the UV denatures EDTA fairly rapidly, as this experiment shows. And since EDTA complexed nutrients are not rapidly consumed, there may be a lot lost to UV exposure. This, I think, must be accepted and dealt with in much the same way as the inevitable loss due to water changes, especially by those who practice EI for example. It seems to me that many (most?) micro/trace nutrients are more ideally dosed via substrate, for root uptake. Getting this stuff into the roots from the water column is the hard way for the plant, but the easy way for the aquarist. I should probably make better use of root tabs, but in a large densely planted tank there are practical difficulties, and I wouldn't rely on that method completely in any case.

I will not be going to an "active" (soil-based) substrate. I like my inert gravel and unchanging Flourite mix. It is eternally stable.

All this explains also why I do not routinely use activated carbon. It is good at removing DOCs from water, and it won't remove desirable dissolved inorganics such as those that comprise macronutrients. But it will remove organic molecules, and that includes all ligands and the metals bound to them. Activated carbon is a good way to defeat dosing of anything that is complexed, and it will also take out gluteraldehyde (Excel), phytohormones, humic (blackwater) additives, and any allelopathic toxins the plants may be generating to suppress algae (and yes, I do believe in that). And the same goes for other chemical filtration media designed to latch onto organic molecules. I don't use them, and haven't missed them since I stopped.

So, there is no doubt in my mind that ferrous gluconate is the preferred way to provide iron to the aquatic macrophyte, provided certain things are borne in mind about when and how much to dose it. It does, however, play merry hell with my data curves.
Last edited:

Paul G

New member
This diagram is of the pressurized CO2 setup I am using. I purchase CO2 in 20 pound bottles. The in-line diffusor appears in a previous photo.

I use the Apex Aquacontroller for pH monitoring and control. The CO2 PRIMARY is programmed to maintain the average pH at 6.6, while the CO2 BACKUP is programmed to maintain pH at 6.8. When the primary empties, the pH will rise just enough to activate the backup. I replace the primary tank as soon as I notice it is empty, but the backup prevents the system pH from going out of control if I don't see it even for a few days (it could go to as high as around 8 I think).

The primary's solenoid valve is parallel connected to an elapsed time indicator. At present, one 20 pounder is good for over 150 hours and this lasts typically about a month or so. I will in future be paying closer attention to this and refine those numbers.

The end-of-line solenoid is redundant, you might even say unnecessary. If you live in fear that your first CO2 valve may fail open and poison your fish, you need this. If you have a measure of sanity, maybe not.

Since this system's alkalinity is controlled to 6 to 7 degrees, the CO2 concentration is about 50 ppm. I have learned to watch out for drift in the KH. The stocking level in this tank is high and I feed liberally. The oxygen tension in this system is also vey good. The biofilters are extremely efficient and acidification proceeds apace (in every ammonium molecule conversion reaction two hydrogen ions are evolved). The buffering must answer. I can adjust the daily bicarbonate dose easily, but if I don't test KH frequently I may lapse into instability, and I certainly don't need any more of that.
Last edited:

Paul G

New member
Still on the subject of hardware:

Illustrated are the probes and aeration setup. Two small powerheads direct flow over the probes which are secured in magnetic mount probe holders. There are two titanium grounding rods.

Apex turns on aeration if DO, dissolved oxygen, falls below 5 ppm, or if pH falls below 6.5. In this aquarium, plant generated O2 peaks to 10 ppm (super-saturation) by the time SUN 4 lights turn off and falls off overnight to 6 ppm low just before dawn. The pH on either end of the tank seldom drops below 6.5, and that only happens occasionally when the CO2 injection briefly overshoots. So the aeration is only used as a "safety." I also turn it on manually when the pumps are shut down for maintenance.

Other Apex controlled functions have been previously described and are diagrammed above.

The chiller has an internal thermostat system, as most aquarium chillers do, and I set that to 76 F to prevent overheating. I do not use an Apex thermostat program for chiller on-off. I do use an Apex thermostat program to control the heaters. The 300 watt heaters' internal thermostats are set to allow maximum 80 F so they cannot run away, but Apex is set for 77 F. When the heaters come on, power to the chiller is switched off automatically.

The CO2 EOL is programmed to turn on when either the CO2 PRIMARY or CO2 BACKUP comes on. The magnetic stirrers under the doser vats are programmed to turn on 5 minutes ahead of any scheduled dosing event, and to turn off when dosing is done. They all run simultaneously regardless of what is being dosed (it's always good to stir the tanks).

I use relays (socket extenders) to distribute power sources to high-current loads so as to minimize loading on the Apex energy bars and their associated UPSs. One example is shown here, where 5 amps is switched by a 30 mA relay solenoid. There are reasons why this is done, but I need not go into it here.

This is by no means a complete map of the Apex Aquacontroller or description of its capabilities, but it gives an idea of how versatile and useful it is. You can set up as many timers, thermostats, parameter controllers, lighting and dosing schedules as desired depending on the modules you connect and how you program the system. It is not just for reefers.