Zeroing adjusts a reference point; a function check shows a response; a comparison shows agreement or disagreement. Calibration establishes the relationship to a reference under stated conditions and reports measurement results and uncertainty. None of these automatically proves that a tool is suitable for every job.
Source-based explanation
How do zeroing, function checks and calibration differ?
Zeroing sets a reference point, a function check confirms a response, and calibration relates readings to a reference under stated conditions. A comparison can expose disagreement without identifying its cause. Name the activity and keep its evidence, because maintenance or a successful response alone does not establish accuracy across the range.
| Activity | Useful evidence | Limit |
|---|---|---|
| Zeroing | Reference condition and zero operation recorded | Does not characterise the rest of the range |
| Function check | An expected response using the documented method | Does not quantify accuracy throughout the range |
| Comparison | Both readings, common conditions and reference details | Agreement between two unknown tools is not traceability |
| Calibration | Identified reference, points, uncertainty and traceability | May not include adjustment or a statement of conformity |
| Maintenance | Cleaning, consumables and servicing history | A cleaned or repaired instrument still needs appropriate checks |
Why do pressure and temperature checks need suitable references?
Check whether a pressure instrument expects a local atmospheric reference or an absolute reference. Sea-level-corrected weather pressure can introduce error at altitude. Keep both instruments at the required conditions and allow them to stabilise according to their instructions, and record the reference used.
For temperature, record the reference method and probe contact conditions. A casual surface comparison cannot establish performance over the full operating range. A clamp’s thermal contact error can dominate a display’s nominal resolution.
What should you request from a calibration provider?
Specify the instrument model, serial number, working range, required points and intended use. Ask for reported results, measurement uncertainty, reference traceability and, where relevant, as-found and as-left data. If accreditation is required, check that the provider’s actual scope covers the measurement and range; a logo alone does not establish coverage.
A recalibration interval depends on use, manufacturer advice, drift history and the applicable quality requirements. Check the applicable requirements for the instrument and its use rather than assuming a universal annual interval.
Why maintain the sensor, not just the display?
Check seals, hoses, filters, temperature leads, batteries and signs of contamination using the exact manufacturer’s instructions. Do not transfer a solvent-cleaning method from one vacuum sensor to another. Record maintenance and the subsequent check separately; keep an instrument with unexplained disagreement out of decisions that rely on its accuracy until it has been assessed.
Which sources support this reference, and what are their limits?
These sources support the terminology, declared capabilities and attributed observations discussed in this reference. Manufacturer declarations and other authors’ observations are not our physical test results. Check the exact model and the source’s stated conditions before applying a claim; the references were consulted on .
- HVAC School: barometric pressure and calibration ↗
Explains local absolute pressure versus weather pressure corrected to sea level.
- HVAC School: micron gauges gone wild ↗
External sensor-contamination examples; cleaning advice is not universal across models.
- HVAC School: negative superheat and tool checks ↗
US technical explanation; not UK regulatory guidance.
- ElitechEU: VGW-mini product specifications ↗
Manufacturer declaration; options and customer comments are not a verified kit inventory.
What should stay with the reading?
A useful measurement record preserves the conditions, instruments, connection details and original reading, then adds any repeated observation or retest. Use the blank field record to keep those details together. See how we assess evidence and our commercial relationship when judging the scope of a published conclusion.
