Module 7.2 — How online instruments are calibrated and verified, and why every reading used throughout this course rests on this discipline.
Click any element to see its role in the calibration chain the entire program rests on.
Every reading discussed since Module 1.1 — pH, conductivity, dissolved oxygen, sodium, silica, phosphate, cycles of concentration — was treated as trustworthy data. That trust isn't automatic; it's the product of a calibration and QA/QC discipline that this module makes explicit. An instrument that reads confidently but incorrectly is arguably more dangerous than a known-unreliable one, since it drives decisions with false confidence.
Calibration is the process of comparing an instrument's output against a known reference standard and adjusting (or documenting the deviation of) the instrument so its readings accurately reflect true values. Most chemistry instruments require at minimum a two-point calibration — typically a zero or low-end reference and a span or high-end reference — to establish both the instrument's baseline accuracy and its response across the range it's expected to measure. A single-point check can confirm an instrument hasn't drifted from a known point, but can't confirm accuracy across its full working range the way a proper multi-point calibration can.
A calibration is only as trustworthy as the reference standard it's checked against. NIST-traceable standards — reference materials with a documented chain of comparison back to national measurement standards — provide the confidence that a calibration is not just internally consistent but actually accurate in an absolute sense. Using an uncertified or expired reference standard can produce an instrument that's perfectly self-consistent yet consistently wrong, a failure mode that's often invisible until cross-checked against an independent method.
Module 1.4 introduced the relationship between online and grab sampling: online gives continuous trending, grab samples give independent verification. This module makes explicit why that verification role matters so much — an online instrument can drift gradually between calibrations (fouling, sensor aging, electronic drift) in a way that looks like a smooth, trend-consistent reading right up until a grab sample comparison reveals the online instrument has been reading incorrectly for some time. Periodic cross-checking, not just periodic recalibration, is what actually catches this kind of gradual drift.
Quality assurance/quality control (QA/QC) extends beyond individual instrument calibration to the whole data-generation process: documented calibration procedures and schedules, records showing when and how each instrument was last verified, and a clear process for what happens when an instrument fails a verification check (is the affected data period flagged? Recalculated? Discarded?). This documentation isn't paperwork for its own sake — it's what allows anyone reviewing historical data, including during a failure investigation like Module 4.7's, to know how much confidence to place in the numbers being reviewed.