Sorry to have been gone for so long. Two Things:
1) I have been struggling with a near crash. Not sure of exactly where the problems were at first and knew I needed to go into troubleshooting mode. This is not something appropriate for ongoing commentary, so I decided to go quiet until something interesting happened. More about this below.
2) Over the last year I have been giving a lot of thought to a complete revision of the circulatory system with the general idea being simplification and easier maintenance. In connection with Thing 1, this has been brought up to the front burner. I will detail this re-build here for anyone who cares to tune in.
This tank has been experimental in terms of a few ideas I wanted to try out. Experiment 1 was implementation of the Streaming Water Change Regimen and full-auto dosing to maintain stable nutrient/chemistry parameters. I have declared, and I reiterate now, that this method works in that I am able to control the water change rate and the compensatory concentrations required. I am able to do this by virtue of Apex Aquacontroller automation and willingness to conduct a full battery of water tests at the frequency necessary to yield the correct calibrations. I wanted to have control of the system chemistry by computer and I am satisfied that I have proof of concept at least on this point.
Experiment 2 was to find out generally what the optimum nutrient load concentrations are for a densely planted ecosystem, and specifically how much autochthonous product of a heavily stocked and generously fed fish community could be counted on to provide sufficient nutrient with little or no allochthonous supplementation. The SWCR was set for a considerable rate to assure expulsion of DOM as "measured" by system redox, which yielded oligotrophy in DIN and orthophosphate levels as well. So this experiment devolved into a trial run in "low-nutrient" planted tanks.
I will confidently state here and now that
there is no such thing as a low-nutrient planted aquarium. I tested the idea of "non-zero" nutrient concentrations and held strict oligotrophy for an extended period of time, dosing P and N supplements only enough to prevent drift to total zero. The SWCR did its job and deficiencies did not present with obvious symptoms at first. Bottom line here is the plants were severely nitrogen limited. This showed up as a halt in growth and algae problems. This condition was deceptively subtle and I confess it fooled me. This was a head-scratcher for a while until I just couldn't think of anything else. So the next stage in Experiment 2 was to bring NO3 up to anything between 10 and 20 ppm, and PO4 to consistent 2 - 2.5 ppm. This meant switching from strict oligotrophy to mesotrophy. The non-zero nutrient strategy will not work in a densely planted aquarium.
The best control over algae is to keep the plants growing. Oligotrophy is no better than eutrophy when it comes to algae.
Within just a couple of days this aquarium really perked up. The ecosystem is healing itself and this requires patience. Much of the algae has spontaneously abated and the plants are thriving again. A corollary benefit of mesotrophy is that it is simpler to test for adequate concentration of NO3 because now, as with potassium, it is only necessary to assure a minimum "luxury uptake" level threshold and precision test results are obviated. In other words, I don't really care if NO3 is 13 ppm or 18 ppm because these numbers define a range of acceptability inside of mesotrophy, and therefore represent knowledge sufficient for the purpose. Thus, I can rely on a simple dunk-and-compare test strip to ball-park the nitrogen, and complicated and expensive titration testing is completely unnecessary.
Experiment 3 was implementation of the Hardness Reconstitution Reactor. This idea works but requires much tinkering, so I would qualify its benefits as marginal if there is an easier way. Since now I am dosing nitrate daily, I am able to dose it as Ca(NO3)2 which is easily soluble and provides calcium. I am now auto-dosing a Ca(NO3)2 / CaCl2 mix and MgSO4 separately and am getting good ratios. These are the only sources of chloride and sulfur in any significant quantity so I have deemed this acceptable and have decommissioned the HRR.
Experiment 4, an integral feature of the SWCR itself, was auto-dosing buffer as K2CO3 which is necessary because the water changes dilute buffer and it must be replenished to maintain system KH. This also provides potassium well into luxury uptake territory. Since discontinuing the HRR means eliminating one source of CO3, the potassium has ascended higher still due to consequent KH calibration. There is no toxicity issue here, so the experiment verifies the success of the method.
The tank today: