Temperature Data Loggers for a Lab Fridge or Freezer: What to Buy and What to Record

A lab fridge or freezer needs a digital data logger with its own sensor, a probe in a thermal buffer placed among the contents, a current calibration certificate from an accredited or traceable source, and a sampling interval of 30 minutes or shorter. It also needs an alarm and a display of minimum and maximum temperature. The record that matters is not the logger's memory but the file you export from it, reviewed on a schedule and kept for as long as anything stored in the unit might be questioned.

This article covers the storage logger that lives in your cabinet. The logger that travels inside a parcel answers a different question, whether the shipment was held in condition, and belongs with the receiving decision. Alarms and out-of-hours response have their own article in this cluster.

Logger types and the accuracy that matters

CDC's vaccine storage guidance is the most practical specification available to a small buyer, because it was written for clinics that also lack a quality department. Its recommended digital data logger has a detachable probe that reflects the temperature of the contents, an alarm for out-of-range readings, a low-battery indicator, a display of current, minimum and maximum temperature, an accuracy of ±0.5 °C, and a logging interval of at least every 30 minutes ^[2]. It also lists what does not qualify: alcohol or mercury thermometers, bimetal stem devices, food thermometers, chart recorders, infrared devices, and any device without a current calibration certificate ^[1].

WHO's prequalification specification for a 30-day refrigerator logger sets a similar accuracy: ±0.5 °C or better between −20 °C and +20 °C for at least 12 months after calibration, relaxing to ±1.0 °C beyond that if the device is not recalibrated ^[3]. That sentence carries two buying lessons. Accuracy degrades with time, so the calibration date matters as much as the figure. And a device built for a refrigerator may not be rated for a freezer at all: that specification's operating range stops at −20 °C ^[3]. Check the stated range covers the coldest set point you will run, with margin.

The calibration certificate is part of the purchase

A logger reading is a measurement, and a measurement without traceability is an opinion with a decimal point. CDC treats a certificate as acceptable if it shows conformity with ISO/IEC 17025, testing by a laboratory accredited by an ILAC mutual-recognition signatory, traceability to NIST standards, or conformity with ASTM E2877 tolerance class F or better ^[1]. ISO/IEC 17025 is the standard for the competence of testing and calibration laboratories, so accreditation against it is what makes a calibration statement checkable by someone else ^[4].

• The certificate should name the model and serial number, the calibration date, an uncertainty of ±0.5 °C or less, and confirm the instrument passed ^[1].
• Recalibrate every two to three years or on the manufacturer's schedule, and check against another calibrated device after any drop or knock ^[1].
• Replace a device found outside ±0.5 °C rather than adjusting the record around it ^[1].
• Keep the certificate with the logger's records. Regulated distribution guidance expects monitoring equipment to be calibrated at defined intervals based on risk ^[5].

Sampling interval and excursion detection

The interval decides how precisely you can say how long a problem lasted, which is usually the question that matters afterwards. WHO's 30-day logger must measure at least every 10 minutes and log at least every hour ^[3]. CDC asks for a recorded reading at least every 30 minutes ^[2]. For a small lab, 5 to 15 minutes is a sensible setting: fine enough to time a fault, coarse enough that the file stays readable. Shorter is not always better. A buffered probe sampled every minute will show door openings the contents never felt, and a record full of meaningless spikes teaches people to stop reading it. For the reasoning behind specific settings, see how sampling interval and alarm delay are set from mapping data.

Placement: the air is not the container

Air temperature changes fast and container temperature changes slowly, so a bare sensor in the airstream records drama that the stored material never experienced. CDC recommends a buffered probe, immersed in glycol, ethanol or glycerin, or held in glass beads, sand or a solid block, because it gives the most accurate measure of the contents ^[1]. Place it in the centre of the unit with the stored material around it, and never near the walls, floor, vent, ceiling or door, where readings are unrepresentative ^[2]. Mount the display outside the cabinet so it can be read without opening the door ^[2].

One probe tells you about one position. If the cabinet has never been mapped, the centre is a reasonable default; once it has been, move the probe to the position that fails first, because that position defines whether the cabinet is in control.

What to record, and how often to read it

A logger collects data continuously; a person still has to look at it. CDC's routine is a check at the start of each working day, recording the minimum and maximum since the last check, the date, the time, the name of the person checking and any action taken. Where a reading is missed, the entry is left blank rather than filled in later ^[1]. WHO's storage guidance asks for the same thing in general terms: monitor at regular intervals, record the data, review the records, and use calibrated equipment ^[6].

1. Daily: read and write down minimum, maximum and current temperature, with initials, then reset the min/max if the device requires it ^[1].
2. Weekly or monthly: export the full data file, name it with the unit and date range, and store it outside the device.
3. At each export: scan the trace for slow drift, lengthening compressor runs or repeated small excursions, which usually come before outright failure.
4. At every excursion: record start and end times, peak temperature, room temperature if known, what was in the unit, and what was done ^[1].
5. Annually: file the calibration certificate and the year's exports together.

Export matters more than it seems. The WHO 30-day logger is designed to keep monitoring after its 30-day window by overwriting data older than 30 days ^[3]. Many commercial devices behave the same way. A record that exists only inside the logger has a life measured in weeks. CDC keeps temperature logs for three years ^[1]; for research material the better rule is to keep them as long as any work done with material from that unit might be questioned.

Alarm thresholds that are useful rather than ignored

Set the logger's alarm inside the storage limit, not at it, so there is time to act before the limit is reached, and use a delay so that routine door openings do not trigger it. The WHO logger shows how asymmetric those delays can reasonably be: it alarms after 60 minutes at or below −0.5 °C in a refrigerator, where freezing is the main risk, but only after 10 hours at or above +8 °C ^[3]. Distribution guidance adds a step small labs skip: alarms should be tested regularly to make sure they work ^[5]. Remote notification and who responds are covered in the alarms article.

What a logger record proves in a supplier dispute

A storage record proves the condition of material in your custody, from the moment it went into the cabinet. That is exactly what a supplier asks about when a lot is questioned: was it stored as labelled after receipt? A continuous, calibrated, exported record with daily checks answers that question conclusively, and without one the conversation usually ends there.

What it cannot prove is anything before receipt. Transit exposure needs its own evidence: an in-parcel logger, the carrier's scan timeline, and a timestamped goods-in entry that shows when the material went into the unit, so the storage record starts where the transit record ends. Where the trace does show an out-of-range period, the next step is not the logger but how an excursion is assessed once the record shows one.