A new pH electrode, a recalibrated handheld, and a disagreement that buffer checks did not settle.
The Milwaukee MA911B/2 on the controller was around eighteen months old and was replaced with an Apera 203DJ-C double-junction electrode, fitted to the existing Apex BNC input through a three-foot BNC extension. The extension was in place before calibration, so it is part of the calibrated path rather than something added after.
The probe arrived with a factory QC certificate against NIST-traceable standards: offset within ±20 mV, slope at least 97 percent, 98 percent of response within 30 seconds, QC passed on the 27th of October 2025.
It was calibrated on the controller against pH 7.00 and 4.00 buffers. Verified afterwards it read approximately 7.00 and 4.04. The Hanna HI98129 handheld was recalibrated on the same two buffers; before recalibration it disagreed substantially at the acidic end, and afterwards read approximately 3.98 and approximately 7.03 while stabilising.
In aquarium water the two instruments do not agree. The controller with the Apera reads about 5.97 to 5.99 where the Hanna reads about 5.73, a gap of roughly 0.24 to 0.30 pH depending on the comparison.
Both read conventional buffers reasonably well, which is precisely the case a buffer check cannot resolve. Neither instrument is recorded here as having failed, because nothing measured so far establishes that.
What follows is procedural rather than a verdict: CO₂ trend figures are calculated from one instrument, the Apex with the Apera, on both sides of the comparison. Hanna readings and Apera readings are not mixed in the same calculation.
The previous degassed figure of 6.22 was taken with the Hanna and is no longer the reference for CO₂ work.
A fresh sample of tank water was stirred aggressively in an open container for about six minutes. The stirrer was stopped, the Apera was inserted and allowed to stabilise, and the sample read 6.55. That is now the working degassed reference for tuning CO₂ on this system. The aquarium itself read about 6.0 that evening with CO₂ off; the stirred sample, not the tank, is the reference value.
The next check is the stabilised in-tank reading during the CO₂ period, with the drop calculated as 6.55 minus that reading. No increase in CO₂ follows from the old baseline.
Tank conductivity measured approximately 212 µS/cm the same day.
There is a lower-flow area in the front right corner, directly below the glass lily pipe outflows, with the plumbing on the right side of the tank. CO₂ microbubbles still move slowly through it, so it is not stagnant and it is not being called a dead zone.
What is being watched there is plant movement, debris accumulation, bubble movement, and whether algae localises in it.