Insights · 120 guides
Technical guides for measurement, calibration and validation teams.
Answer-first explainers with key takeaways, FAQs and references to the standards and guidance they draw on. Written for engineers, quality teams and laboratory staff — and structured so answer engines can cite them accurately.
Start here
Temperature Mapping
What is temperature mapping?
Why regulated storage spaces are mapped, what a study produces, and how the results are used.
2026-09-13 · 2 min read
Uncertainty
Building an uncertainty budget step by step
Write the measurement model, list every input, evaluate each as a standard uncertainty, apply sensitivity coefficients, combine in quadrature, expand with k. Eight steps, one table.
2026-09-13 · 2 min read
Uncertainty
What is a decision rule and why every certificate needs one
A decision rule states how measurement uncertainty is accounted for when declaring pass or fail against a tolerance. Simple acceptance, guard banding and non-binary statements are the common choices.
2026-09-13 · 2 min read
GDP / GMP
Temperature control under EU GDP: what is expected
EU GDP requires premises and equipment that maintain storage conditions, initial mapping before use, calibrated monitoring equipment, alarms, and records that demonstrate control.
2026-09-13 · 2 min read
Laboratory Discovery
How to choose a calibration laboratory
Check the accreditation scope for the exact quantity, range and uncertainty you need; then evaluate turnaround, certificate quality and how they handle out-of-tolerance findings.
2026-09-13 · 2 min read
GDP / GMP
Mean kinetic temperature (MKT): what it is and when to use it
MKT is a single temperature that represents the cumulative thermal stress of a varying temperature history, weighted by the Arrhenius relationship. It is useful for storage assessment, not for hiding excursions.
2026-09-13 · 2 min read
Full library
120 of 120 articles
Temperature Mapping
Planning, sensor placement, analysis and reporting of mapping studies.
- What is temperature mapping?Why regulated storage spaces are mapped, what a study produces, and how the results are used.2 min read
- How many sensors are needed for temperature mapping?There is no single number: sensor count follows from the geometry, the risk points and the acceptance limits of the space.2 min read
- How to identify hot and cold spotsHot and cold spots are defined by the data and the limits together — and the definition should be fixed before the analysis starts.2 min read
- What a temperature mapping protocol should containThe protocol is the contract for the study: fix the purpose, limits, sensor plan and analysis rules before a single logger is placed.2 min read
- What a temperature mapping report should containA report is a technical record that lets a reader who was not present understand what was done, what was found and what was decided.2 min read
- Cold room temperature mappingWalk-in cold rooms combine forced airflow, door traffic and defrost cycles — each leaves a signature in the data.2 min read
- Warehouse temperature mappingLarge ambient warehouses are dominated by height, external walls, roof and season — so the sensor plan and the study calendar matter more than anything else.2 min read
- Refrigerator temperature mappingPharmacy and laboratory refrigerators are small, but they have cold spots at the back and warm spots in the door — mapping shows which shelves are safe.2 min read
- Freezer temperature mapping (−20 °C and −80 °C)Freezers stress the sensors as much as the space: response time, cable ingress and condensation all need managing.2 min read
- What GDP expects from temperature mappingEU GDP asks for an initial mapping exercise before use, repeated after significant changes — and for the results to drive where monitoring equipment is placed.2 min read
- Sensor placement principles for mapping studiesPlace sensors where product is, where risk is, and where the monitoring probe might go — and write down why.2 min read
- Setting acceptance criteria for a mapping studyAcceptance criteria come from the product's storage condition, expressed as limits, a permitted excursion definition, and a rule for what happens at the boundary.2 min read
- Common temperature mapping mistakesMost failed studies fail before the loggers are placed — in the plan, the metadata or the calibration.2 min read
- Empty versus loaded mapping studiesAn empty space shows the equipment; a loaded space shows what product will experience. Most qualification approaches want both.2 min read
- Seasonal mapping: when one study is not enoughIf ambient conditions drive the space, the study must cover the ambient conditions that matter.2 min read
- Choosing sampling interval and study durationShort enough to see events, long enough to include the cycles that matter — and consistent across all sensors.2 min read
- Data-quality checks before analysing mapping dataStatistics computed on bad data are precise and wrong. Check the data first.2 min read
Calibration
Calibration practice, certificates, intervals and equipment.
- What is CMC in calibration?Calibration and Measurement Capability is a laboratory's best-case uncertainty, not the accuracy of your instrument.2 min read
- Dry block vs environmental chamberTwo common ways to generate a temperature for calibration, suited to different sensors, ranges and uncertainties.2 min read
- How to calibrate a Pt100Calibration by comparison against a reference thermometer, and the factors that decide the uncertainty.2 min read
- As-found vs as-leftWhy both sets of readings matter, and what to do when the as-found data are out of tolerance.2 min read
- What should a calibration certificate contain?A checklist for reading a certificate — and the items whose absence should make you ask questions.2 min read
- How to set calibration intervalsThere is no universal interval. Set it from risk, manufacturer data and — above all — the instrument's own as-found history.2 min read
- How thermocouples are calibratedThermocouples measure with the whole wire, not just the tip — which is why immersion, homogeneity and the reference junction dominate their calibration.2 min read
- How temperature data loggers are calibratedLoggers are calibrated as a system — sensor, electronics and firmware — usually in a chamber, at the points that matter for their use.2 min read
- Liquid bath versus dry-block calibratorA stirred liquid bath offers the best uniformity; a dry block offers speed, portability and no fluid — the choice is about uncertainty and practicality.2 min read
- Immersion depth and stem conductionA probe conducts heat along its stem. Insufficient immersion makes the sensor read closer to room temperature than to the medium.2 min read
- Calibration, adjustment and verification are different thingsCalibration measures; adjustment changes the instrument; verification compares against a requirement. Mixing them up leads to lost as-found data and unsupported claims.2 min read
- Tolerance versus uncertaintyTolerance is what you require of the instrument; uncertainty is how well the calibration could measure it. A conformity statement needs both.2 min read
- Choosing calibration pointsCalibrate where you measure. Points should bracket the use range and include any temperature where a decision is made.2 min read
- Self-heating in resistance thermometersThe measuring current warms the sensing element. The effect depends on the medium — so calibrate at a current and condition representative of use.2 min read
- Hysteresis in temperature sensorsA sensor can read differently at the same temperature depending on where it came from. Calibrations that only approach from one direction hide it.2 min read
- Selecting a reference thermometerThe reference sets the floor of your uncertainty. Choose it for stability and a calibration uncertainty comfortably below what you need to deliver.2 min read
- The ice-point checkA properly made ice bath gives 0 °C to within a few millikelvin — a cheap, repeatable way to watch a PRT for drift between calibrations.2 min read
- Calibrating relative-humidity sensorsHumidity calibration is a temperature calibration in disguise: small temperature differences between sensor and reference become large RH errors.2 min read
Metrology
Traceability, uncertainty and the language of measurement.
- Measurement uncertainty explainedWhat an uncertainty statement means, how it is built up, and why a coverage factor matters.2 min read
- Metrological traceability explainedTraceability means a result can be related to the SI through an unbroken, documented chain of calibrations, each with a stated uncertainty. Here is what that requires in practice.2 min read
- Type A vs Type B uncertainty evaluationType A uses statistics on repeated observations; Type B uses any other information. Both produce standard uncertainties that combine the same way.2 min read
- What the coverage factor k meansThe coverage factor multiplies the combined standard uncertainty to give an expanded uncertainty with a stated level of confidence. k = 2 is roughly 95 % for a normal distribution.2 min read
- Rectangular, triangular and normal distributions in uncertainty budgetsThe assumed distribution sets the divisor that turns a limit into a standard uncertainty: √3 for rectangular, √6 for triangular, k for normal.2 min read
- Sensitivity coefficients: how input uncertainties reach the resultA sensitivity coefficient converts an input's uncertainty into its effect on the output. For a direct temperature comparison it is 1; for derived quantities it comes from the model.2 min read
- Correlated inputs in uncertainty evaluationRoot-sum-of-squares assumes independence. When two inputs move together — the same reference used twice, for example — the combination must include a correlation term.2 min read
- Reporting uncertainty: rounding and significant figuresReport the expanded uncertainty to at most two significant figures, round the result to the same decimal place, and state the coverage factor.2 min read
- Ten VIM terms every calibration user should knowMeasurand, indication, error, correction, accuracy, precision, trueness, resolution, drift and traceability — defined as the international vocabulary defines them.2 min read
- Accuracy vs precision in measurementPrecision is about scatter; accuracy is about closeness to the true value. An instrument can be precise and wrong, or accurate on average and noisy.2 min read
- Repeatability vs reproducibilityRepeatability holds everything constant except time; reproducibility changes operators, instruments, locations or days. The gap between them reveals hidden influences.2 min read
- Drift: how instruments change between calibrationsDrift is a slow change in indication over time. It is estimated from calibration history and enters the budget as an uncertainty for the interval — or is corrected if it is predictable.2 min read
ISO/IEC 17025
What the laboratory-competence standard asks for and why.
- ISO/IEC 17025 calibration explainedWhat accreditation to ISO/IEC 17025 does and does not tell you about a calibration.2 min read
- ISO/IEC 17025:2017 structure: what each clause coversThe 2017 edition is organised into general, structural, resource, process and management-system requirements. Knowing the map makes the standard far easier to use.2 min read
- Impartiality and confidentiality under ISO/IEC 17025The 2017 edition made impartiality a first-class requirement: laboratories must identify risks to it on an ongoing basis and show how they are managed.2 min read
- Personnel competence records under ISO/IEC 17025Competence must be defined, achieved, authorised and monitored — with records for each step, for every activity that affects results.2 min read
- Equipment requirements in ISO/IEC 17025 (clause 6.4)Equipment must be fit for purpose, calibrated where it affects results, checked, protected from adjustment, and fully recorded — including software.2 min read
- Metrological traceability in ISO/IEC 17025 (clause 6.5)Results must be traceable to the SI through calibration by a competent laboratory, certified values of certified reference materials, or direct realisation of SI units.2 min read
- Method selection, verification and validation (clause 7.2)Standard methods must be verified before use; non-standard or modified methods must be validated. The extent depends on the intended use.2 min read
- Evaluation of measurement uncertainty (clause 7.6)Laboratories must identify contributions to uncertainty and, for calibration, evaluate the uncertainty of every result — including calibration of their own equipment.2 min read
- Reporting results and statements of conformity (clause 7.8)A certificate must carry defined identification, method, traceability and uncertainty information — and any pass/fail statement must name the decision rule.2 min read
- Nonconforming work under ISO/IEC 17025 (clause 7.10)When work does not conform — an out-of-tolerance reference, a failed data-quality check, a wrong method — the response must be controlled, evaluated for impact, and recorded.2 min read
- Technical records: what to keep and for how long (clause 7.5)Records must let the measurement be repeated under conditions as close as possible to the original, and every amendment must be traceable to who changed what, when.2 min read
- Internal audits and management review (clauses 8.8 and 8.9)Internal audits check the management system against the standard and the laboratory's own procedures; management review checks whether the system is still suitable, adequate and effective.2 min read
GDP / GMP
Temperature control and quality expectations in regulated storage and distribution.
- Temperature control under EU GDP: what is expectedEU GDP requires premises and equipment that maintain storage conditions, initial mapping before use, calibrated monitoring equipment, alarms, and records that demonstrate control.2 min read
- EU GMP Annex 15: the qualification stages explainedAnnex 15 describes URS, DQ, FAT/SAT, IQ, OQ and PQ as a sequence that builds evidence a system is fit for its intended use — with requalification when things change.2 min read
- Handling temperature excursions in storage and distributionAn excursion is a deviation: contain, record, assess product impact against stability data, decide, and learn. The record must show all five steps.2 min read
- Mean kinetic temperature (MKT): what it is and when to use itMKT is a single temperature that represents the cumulative thermal stress of a varying temperature history, weighted by the Arrhenius relationship. It is useful for storage assessment, not for hiding excursions.2 min read
- Qualifying temperature-controlled vehicles and shipping containersTransport equipment is qualified like storage: define the conditions, map with calibrated loggers under worst-case ambient and load, and monitor every shipment.2 min read
- Warehouse qualification under GDP: from URS to routine monitoringA qualified warehouse has a documented requirement, a mapping study that meets it, monitoring placed by the results, and a change-control trigger for re-qualification.2 min read
- Data integrity for temperature records: ALCOA+ in practiceTemperature records are GxP records. Attributable, legible, contemporaneous, original, accurate — plus complete, consistent, enduring and available — is the checklist.2 min read
- Deviations and CAPA for environmental control failuresAn excursion, a failed calibration or a mapping failure is a deviation. Classify it, investigate root cause proportionately, and close it with an action that prevents recurrence.2 min read
- Storage condition labels explained: 2–8 °C, 15–25 °C, 'do not freeze'Label storage conditions derive from stability studies under ICH conditions. Each phrase has a defined meaning that sets the limits for mapping, monitoring and excursion assessment.2 min read
- Computerised system validation for monitoring and mapping softwareSoftware that produces GxP records must be validated for its intended use: requirements, risk assessment, testing, audit trails, access control and change control.2 min read
Validation
Qualification and validation workflows.
- IQ, OQ and PQ explainedInstallation, operational and performance qualification are three questions asked in order: is it installed as specified, does it operate across its ranges, does it perform in routine use?2 min read
- What a validation master plan should containThe VMP is the site-level statement of validation policy, scope, responsibilities and schedule. It tells an inspector how the organisation decides what to validate and how.2 min read
- Writing a user requirements specification for temperature-controlled equipmentThe URS states what the equipment must do in measurable terms. Every later test traces back to it, so vague requirements produce untestable qualifications.2 min read
- Design qualification: catching problems before installationDQ documents that the chosen design meets the URS. It is cheapest place to fix a cold room with a door on the wrong wall.2 min read
- Requalification and periodic reviewQualified status is not permanent. Change control, periodic review and monitoring data together decide when a system must be requalified.2 min read
- Qualifying a stability or climatic chamberChambers are qualified for temperature and humidity uniformity, stability and recovery — with acceptance criteria tied to the ICH conditions they will run.2 min read
- How to structure a qualification protocolA protocol says what will be tested, how, by whom, and what result counts as a pass — all approved before execution begins.2 min read
- Handling deviations during qualificationA test that fails or cannot be executed as written is a deviation. Record it, assess it, decide, and let the report show the whole story.2 min read
- Risk-based validation: deciding how much is enoughICH Q9 and Annex 15 expect the extent of qualification to follow a documented risk assessment — more effort where product quality is at greater risk, less where it is not.2 min read
- Validating the software that analyses your monitoring and mapping dataCalculations that decide whether a space passed must themselves be verified — with known datasets, expected results and a record of the software version.2 min read
Monitoring
Continuous temperature and humidity monitoring, alarms and records.
- Choosing monitoring locations from mapping resultsPermanent monitoring sensors go where the mapping showed the worst case — the hot spot, the cold spot and the slowest-recovering position — not where the cable is shortest.2 min read
- Alarm limits vs acceptance limits: setting them correctlyAcceptance limits come from the product; alarm limits sit inside them to give time to act; alert (warning) limits sit inside those. Uncertainty and delay decide the gaps.2 min read
- Monitoring ultra-low temperature (−80 °C) freezersULT freezers combine large gradients, slow recovery and severe consequences of failure. Monitoring needs the right probe, the right position and an alarm that reaches someone at night.2 min read
- Humidity monitoring in storage areas: when and howMonitor relative humidity where products or packaging are humidity-sensitive or where labels require it. Sensors drift; calibrate them and expect larger uncertainty than for temperature.2 min read
- What a continuous temperature monitoring system must doCapture at an appropriate interval, alarm reliably, store data securely with an audit trail, and keep working through power and network failures.2 min read
- Reviewing and retaining monitoring recordsA monitoring record nobody reviews is not a control. Define the review frequency, what the reviewer checks, how the review is recorded, and how long records are kept.2 min read
- Sensor response time and why it matters for monitoringA sensor's response time sets how fast it can see a change. Bare-bead sensors respond in seconds; sensors in glycol or metal blocks in many minutes. Choose deliberately.2 min read
- Wireless vs wired monitoring sensorsWireless sensors are easier to place and move; wired sensors are harder to lose and never run out of battery. Data integrity and coverage decide, not fashion.2 min read
- Qualifying a temperature monitoring systemInstallation checks, functional tests including alarm and failure modes, and a performance period under routine use — plus the software validation that sits underneath.2 min read
- Temperature monitoring during transportEvery temperature-sensitive shipment carries a calibrated logger placed with the product; the record is reviewed on receipt and excursions handled before release.2 min read
Uncertainty
Building, combining and reporting measurement uncertainty budgets.
- Building an uncertainty budget step by stepWrite the measurement model, list every input, evaluate each as a standard uncertainty, apply sensitivity coefficients, combine in quadrature, expand with k. Eight steps, one table.2 min read
- The reference thermometer's contribution to uncertaintyThe reference contributes its certificate uncertainty, its drift since calibration, and any interpolation or self-heating effects. Together they set the floor of what can be claimed.2 min read
- Resolution as an uncertainty componentA digital display of resolution d contributes a rectangular component of half-width d/2, giving u = d/(2√3) ≈ 0.29 d. For a 0.1 °C logger that is 0.029 °C — often the largest term in its calibration.2 min read
- Drift as an uncertainty componentBetween calibrations an instrument's error moves. Its history tells you how much; that amount enters the budget for any result obtained during the interval.2 min read
- Uncertainty from the calibration bath or dry blockThe source's stability, axial and radial gradients, and (for dry blocks) loading and stem-conduction effects determine how well the reference and the unit under test see the same temperature.2 min read
- Effective degrees of freedom and when k = 2 is not enoughWhen Type A components rest on few observations, the combined uncertainty has limited degrees of freedom and k must come from the t-distribution. Welch–Satterthwaite gives the number.2 min read
- Monte Carlo evaluation of uncertainty (GUM Supplement 1)Instead of propagating standard uncertainties analytically, sample every input from its distribution, compute the result many times, and read the interval off the output histogram.2 min read
- What is a decision rule and why every certificate needs oneA decision rule states how measurement uncertainty is accounted for when declaring pass or fail against a tolerance. Simple acceptance, guard banding and non-binary statements are the common choices.2 min read
- Guard banding explained with a worked exampleA guard band narrows the acceptance zone by a multiple of the uncertainty so that a 'pass' has a controlled probability of being wrong. Here is the arithmetic.2 min read
- Uncertainty budget for a thermocouple calibrationThermocouples add inhomogeneity, reference-junction and extension-lead terms to the usual comparison budget — and inhomogeneity is often the largest.2 min read
- Measurement uncertainty in temperature mapping resultsA mapping conclusion — the space is within 2–8 °C — carries the loggers' calibration uncertainty, their resolution and any drift since calibration. Compare the extremes plus uncertainty with the limits.2 min read
Equipment
Choosing and caring for sensors, loggers, baths, blocks and references.
- How to choose a temperature data logger for mapping and monitoringMatch resolution and accuracy to the acceptance criteria, check the calibration offer, and look at memory, battery, export format and response time before price.2 min read
- Pt100 tolerance classes explained (IEC 60751)Class AA, A, B and C define the maximum permissible deviation of a platinum resistance thermometer from the standard curve as a function of temperature. They are tolerances, not uncertainties.2 min read
- Thermocouple types and tolerance classes (IEC 60584)Types K, J, T, N, E and the noble-metal types R, S, B each have defined EMF curves and tolerance classes 1, 2 and 3. Choose by range, environment and required tolerance.2 min read
- Thermistors: high sensitivity over a narrow rangeNTC thermistors offer very high resolution and stability near ambient, which is why many precision loggers use them. Their non-linearity means calibration at several points and a fitted equation.2 min read
- Characterising a dry-block calibratorBefore using a dry block for calibration, measure its axial and radial gradients, stability, loading effect and the effect of insert wear — and repeat periodically.2 min read
- Choosing a liquid calibration bath and its fluidThe fluid sets the range, the stirring sets uniformity, and depth sets immersion. Ethanol, water, oil and salt each own a range; safety and cleanliness decide among the overlaps.2 min read
- Readouts and bridges for reference thermometersA reference PRT is only as good as the instrument reading it. Resolution, linearity, measurement current and calibration of the readout all enter the budget.2 min read
- Ice point, triple point of water and other fixed pointsFixed points realise defined temperatures without a calibrated reference. The ice point (0 °C, ~±0.002 °C) is accessible to any laboratory; the triple point of water (0.01 °C) is the anchor of ITS-90.2 min read
- Infrared thermometers: what they measure and when to trust themAn IR thermometer measures radiance, not temperature. Emissivity, field of view, distance and reflected background all change the reading — which is why they are for screening, not for GDP records.2 min read
- Maintaining calibration equipment between calibrationsReferences drift, baths contaminate, inserts wear, batteries die. A maintenance plan with intermediate checks keeps equipment within its assumed performance and catches failures early.2 min read
Laboratory Discovery
Finding, reading and comparing accredited laboratory scopes.
- How to choose a calibration laboratoryCheck the accreditation scope for the exact quantity, range and uncertainty you need; then evaluate turnaround, certificate quality and how they handle out-of-tolerance findings.2 min read
- How to read a scope of accreditationA scope lists, per quantity, the range, the calibration and measurement capability (CMC) and the method. Reading it tells you whether the laboratory can certify what you need — and at what uncertainty.2 min read
- Accreditation bodies and the ILAC MRANational accreditation bodies assess laboratories to ISO/IEC 17025; the ILAC Mutual Recognition Arrangement makes their accredited certificates equivalent across borders.2 min read
- On-site vs laboratory calibration: choosing the right oneOn-site calibration avoids removing equipment and captures the installed system; laboratory calibration offers better sources, lower uncertainty and controlled conditions.2 min read
- Questions to ask before sending equipment for calibrationPoints, tolerance, decision rule, as-found reporting, adjustment policy, turnaround and packaging — settle them in the purchase order and the certificate will be usable.2 min read
- Certified reference materials and ISO 17034Certified reference materials carry a certified value with uncertainty and traceability from a competent producer. ISO 17034 accreditation is the mark of that competence.2 min read
- Proficiency testing and interlaboratory comparisonsLaboratories demonstrate that their results are consistent with others by measuring the same artefact. The En number tells whether a result agrees within its uncertainty.2 min read
- Outsourcing a temperature mapping study: what to specifyA mapping contractor needs your acceptance criteria, layout, operating conditions and reporting expectations. Without them you get a generic study that may not answer your question.2 min read
- Calibration certificate red flagsNo uncertainty, no k, no traceability statement, 'pass' without a decision rule, a range you did not ask for, or an accreditation symbol missing where one was promised — each is a reason to query.2 min read
- Building a calibration management system for measuring equipmentAn inventory, a status for every item, intervals with rationale, certificates linked to items, out-of-tolerance handling, and a review loop. Software helps; the discipline is what matters.2 min read