Today's numbers
pH: 6.7
ORP: 495 mV
NO3: 1 ppm
PO4: 0.06 ppm
Fe: 0.07 ppm
K: 40 ppm
dGH: 3.1
Ca/Mg: 36/20 ppm
dKH: 7.2
The SWCR is balancing the K2CO3 dosing rate spot on. Potassium and alkalinity concentrations are steady at the desired values.
Macros PO4 and NO3 remain very low. A good deal of this nutrient is going to the plants, as the growth and health are fantastic, but the SWCR is certainly cutting into it as well. Iron is okay, and as I trust SeaChem's formulations, all other micros are fine too. That being said, I am bumping these just a little, and will continue with the copious feeding.
Of course, what has been discussed so far are water tests. These are all parameters in the water column. The plants do get nutrient from the substrate. I use inert washed river gravel, mostly composed of quartz with other insoluble clastics from the source rocks, 1 mm to 4 mm rounded grain, "very coarse sand" to "granular pebble" on the Wentworth scale. Pretty much what is sold as aquarium gravel. This is heavily amended by the addition of SeaChem Flourish fracted porous clay to provide some cation exchange capacity (CEC). This substrate is chemically stable and biologically inert. It can be relied upon not to exhaust or deteriorate or otherwise expire - forever. This was a deliberate choice; it removes all concern about the future of what the roots of my plants face. I do not need another variable to deal with.
However, I use substrate fertilizer tablets. Foliar uptake is not the whole picture. API Root Tabs are formulated for macro rich, and SeaChem Flourish Tabs tend more to the micros. I use both. I put in a couple of each around each swordplant about every 6 to 8 months. All the root feeders eventually get some of it. When I first plant, or transplant, I inject a suspension of CaribSea Flora-Spore arbuscular mycorrhyzal symbionts. In an aquarium as mature as this, these are almost certainly established throughout the substrate, but I do it for the
karma; you do it only once, it does no harm, and it's inexpensive.
The GH test kit I use is the LaMotte 4824. It tests freshwater GH and Ca. I have used other reagent type kits and I think the LaMotte is probably the easiest one for obtaining relatively accurate results. It is a pro-grade field and facility kit and definitely not cheap. This test uses a color-change titration method. There is some opportunity for error in the practice of good lab procedure (particularly going too fast), and some subjectivity in the evaluation of the color. The burette calibrations are in 4 ppm increments, so it is quite easy to get a + or - 4 ppm error on any given test if your lab procedure is less than perfect. This, nevertheless, is superior resolution to hobby grade kits. I've been using this kit for years, so I'm pretty good at it, but I am not surprised that I sometimes get a GH or Ca number that is off a little more than I expected from the previous day's number. Nor does it trouble me, since it's already a fairly small margin in terms of the usefulness of the information, and I take more seriously the long term running average of the values. The largest data universe gives the most accurate picture. I have made a point of this because someone is bound to notice that these numbers look like they are bouncing around without a satisfactory reason for why the test should be believed.
More on the SWCR: this method is based on a timer program. The drain valve solenoid is controlled by a timer that establishes when the ODE (open drain event) will occur, and also how long the valve remains open. It is necessary that the ODE lasts long enough so that the waterline drop will trigger the ATO (auto top-off). Then, the timing is such that the drain will close when a discrete packet of tank water has been removed. This has all been determined by experimentation and will vary from system to system; there is no prescription. In the case of this tank, a 90 second ODE removes almost exactly 3.375 liters.
To allow dwell time for supplements after dosing, the ODEs are divided into two series, rather than trying to spread them out further. This permits a 6 hour post-dose period before the next water change (see 24 hour clock above). It happened after much tinkering, that the 7-and-7 90 second ODE scheme was effective and convenient, so that's where I landed.