Key takeaways
Short answer: Calibration and adjustment are two distinct steps that are often blurred together. Calibration is the act of measuring and documenting how far an instrument's reading deviates from a known reference standard — it quantifies the error but does not, by itself, change anything. Adjustment is the act of physically altering the instrument to bring its reading closer to the true value — it corrects the error. Calibration measures; adjustment corrects. You can calibrate without adjusting, and an adjustment should always be followed by a calibration to confirm it worked. For the related metrology pair, see calibration vs verification.
Calibration is the process of comparing an instrument's readings against a known, traceable reference standard and documenting the result — specifically, how far the instrument deviates from the true value at the points tested. Its output is information: at this point, the instrument reads this much high or low, with this uncertainty. Crucially, calibration is fundamentally a measurement, not a repair — it quantifies and records the error, but the act of calibrating does not, by itself, change the instrument or fix anything. This is the most common misunderstanding: people assume calibrating an instrument makes it accurate, when strictly it only tells you how accurate (or inaccurate) it currently is. Calibration produces the documented truth about the instrument's performance; what you then do with that truth — accept it, or correct it — is a separate decision and a separate step.
Adjustment is the act of physically altering an instrument to bring its readings closer to the true value — turning a trim screw, changing a setting, modifying a correction factor — so the error is reduced or removed. Where calibration measures the error, adjustment corrects it. Adjustment is a deliberate intervention on the instrument itself, and it is only meaningful in light of a calibration: you must first know the error (from calibration) before you can sensibly correct it (by adjustment). An instrument reading consistently two grams high might be adjusted to read true, eliminating that bias. But adjustment is not automatic or always appropriate — if the measured error is within acceptable tolerance, you may calibrate and document it without adjusting at all. Adjustment changes the instrument; it is the corrective action that may, or may not, follow from what a calibration reveals.
The clean distinction is that calibration measures the error and adjustment corrects it. They are sequential and separate: calibration first quantifies how far off the instrument is; adjustment, if warranted, then alters the instrument to reduce that deviation. Conflating them — assuming calibration includes adjustment, or that calibrating automatically makes an instrument accurate — leads to real confusion and risk. A calibration certificate documents the error found; it does not necessarily mean the instrument was changed. This matters for traceability and decisions: knowing whether an instrument was merely calibrated (error measured and documented) or also adjusted (error corrected) tells you its actual state. The cleanest practice keeps them distinct in records: calibrate to find the error, decide whether to adjust, adjust if needed, and then re-calibrate to confirm the new state — because an adjustment is only proven by the calibration that follows it.
A pressure gauge is sent for calibration. The lab compares it against a traceable reference and documents the result: at the working pressure, the gauge reads 2% high, with stated uncertainty. That is the calibration — the error is now measured and recorded, but the gauge has not been touched. Now a decision: if the process tolerance comfortably allows a 2% deviation, the gauge may simply be calibrated and returned, its known error documented for users to account for — no adjustment needed. But if 2% is unacceptable, the technician adjusts the gauge — correcting it to read true — and then re-calibrates to confirm the adjustment worked and to document the new, smaller error. The calibration measured the 2% error; the adjustment corrected it; the second calibration proved the correction. Skipping that final calibration would leave the adjustment unverified — a change made but never confirmed.
Calibration always happens; adjustment is conditional. You adjust when the measured error is unacceptable for the instrument's use — outside the tolerance the process needs — and the instrument can be brought back into line. You do not adjust when the error is within acceptable limits (calibrate, document, and use the instrument as-is), or when adjusting would disturb a stable instrument unnecessarily, or when the deviation should instead be handled by applying a known correction factor rather than altering the instrument. There is a discipline here: needless adjustment can introduce new errors or instability, so the bias is to adjust only when the calibration shows it is genuinely needed. And whenever you do adjust, you re-calibrate afterward — an adjustment without a confirming calibration is an unverified change, and unverified changes to measurement equipment are exactly what a quality system should not allow.
Calibration and adjustment keep the measurement system trustworthy, which underpins the quality factor of OEE, exactly like the broader discipline in calibration vs verification. The OEE quality factor counts good versus defective units, and that count depends on instruments that read true — a gauge with an uncorrected, unacceptable bias can systematically scrap good parts or pass defects, corrupting the quality number and the yield and scrap behind it. Calibration finds such a bias; adjustment corrects it where needed; the follow-up calibration confirms the fix. As precision versus accuracy stresses, a consistent reading is not a correct one — and only calibration, with adjustment where warranted, keeps it both. Disciplined measurement is the unglamorous foundation of an honest quality factor.
Fabrico consumes the quality data your instruments produce, so the trustworthiness of those instruments flows into the reliability of its OEE. By trending good-versus-defective results over time, it can help surface the symptoms of an uncorrected measurement bias — a reject rate that shifts after an instrument event, or scrap that does not match downstream reality — prompting a calibration and, where needed, an adjustment. Its maintenance workflows can also track calibration due-dates so measurement discipline does not lapse. Reliable measurement upstream, honest OEE downstream. Book a demo to keep measurement and OEE trustworthy together.
Calibration measures and documents how far an instrument's reading deviates from a reference standard — it quantifies the error. Adjustment physically changes the instrument to bring its reading closer to true — it corrects the error. Calibration measures; adjustment corrects. Calibration alone does not change the instrument.
Not by itself. Calibration is fundamentally a measurement — it quantifies and documents the error. Any adjustment to correct that error is a separate step that may or may not follow, depending on whether the measured error is acceptable for the instrument's use.
Yes, and it is common. If the calibration shows the instrument's error is within the acceptable tolerance for its use, you simply document the error and use the instrument as-is, with no adjustment. Adjustment is only needed when the error is unacceptable.
Yes, always. An adjustment is unproven until a follow-up calibration confirms the new state and documents the resulting error. An adjustment without a confirming calibration is an unverified change to measurement equipment, which a quality system should not allow.
Both keep the measurement system trustworthy, underpinning the OEE quality factor, which counts good versus defective units. A gauge with an uncorrected bias can wrongly scrap or pass parts. Calibration finds the bias, adjustment corrects it where needed, and a follow-up calibration confirms the fix.
Connexion
Blog
