Cold Room Temperature Logging: What to Record and Why It Matters

A controller display tells you the temperature right now. It does not tell you whether the room held temperature through last night’s load shedding, how long the coil sat in defrost, or whether a door was left open for twenty minutes during a busy delivery. Only a proper log does that, and without one you are relying on memory and luck when a client, an auditor, or an insurer asks what actually happened in the room. This article sets out exactly what to log, at what interval, and why a logger that only records averages can hide the exact problem you need to see.

What to Log

A cold room log worth keeping covers five things, not just one temperature reading:

• Room air temperature, measured at the sensor location the controller actually uses, logged continuously rather than spot-checked.
• Product core temperature, where the product or the client’s requirements call for it — a probe into the actual product, not the surrounding air, because air and core temperature can differ by several kelvin for a dense or recently loaded item.
• Defrost cycles — start time, duration, and termination temperature for each cycle, not just a note that defrost ran.
• Door openings, from a door switch or access log where fitted, because a pattern of long or frequent openings explains temperature excursions that would otherwise look unexplained.
• Alarms — every high-temperature, power-failure, and sensor-fault alarm, with the time raised and the time cleared, not just the current alarm state.

A log missing any one of these five still tells you something, but it cannot answer the most common question you will actually be asked afterward: was the product safe, and for how long was it not.

Logging Interval: Frequent Enough to See a Defrost Cycle

• A defrost cycle on a chiller coil typically runs 15 to 30 minutes; on a freezer coil, 20 to 45 minutes. During that window, room air temperature can rise several kelvin above set point before recovering.
• A logging interval of 1 to 5 minutes captures that rise and recovery clearly. An interval of 15 minutes or longer can miss it almost entirely, especially if the logger’s sample happens to land outside the excursion window.
• For compliance-grade records — the kind you would hand to an auditor or a client after a complaint — treat 5 minutes as a practical upper limit on interval for room air temperature. Product core temperature, which changes far more slowly than air temperature, can reasonably be logged less frequently, but confirm the interval your own certification or client contract requires rather than assuming.

Why a Logger That Only Records Averages Hides the Problem

• An hourly average can look perfectly normal — sitting right on set point — while concealing a 10 to 15 minute excursion of 5 to 8 K during a defrost cycle or a door left open, because the average smooths the spike out of the number entirely.
• This matters most exactly when you need the record most: proving what happened during a specific incident. An average tells you the room was “fine on average” during a four-hour load shedding outage; only the raw interval data tells you the room actually reached 9°C for forty minutes before recovering, and whether that exceeds a safe limit for what was stored.
• If you are choosing or auditing a logging system, confirm it stores and can export the raw interval readings, not just a rolling average or a daily min/max summary. Min/max alone tells you the worst point was reached but not for how long, which is often the number that actually matters for product safety.

Food Safety Obligations in South Africa

Cold chain control and record-keeping are a food safety requirement for businesses handling perishable food in South Africa, and the specifics — what must be recorded, for how long, and under which certification or audit scheme — depend on your type of operation and who audits you. Describe your own obligation in plain terms to your team rather than relying on a general rule of thumb, and confirm the current regulation and any certification standard (municipal health by-law, HACCP-based scheme, retailer supplier requirement, or similar) that applies to your specific business with the relevant authority or your certifying body. As a general planning figure, many operations retain temperature records for a minimum of 12 months, but treat that as a starting point to verify, not a fixed number to rely on.

Load Shedding Makes Logging More Important, Not Less

• A room that drifts during a load shedding outage and recovers afterward looks, from the controller’s current reading, exactly like a room that never had a problem at all. Only the logged record during the outage shows how far the room drifted and how long recovery took.
• Two numbers matter most here: peak temperature reached during the outage, and total time above a safe threshold for whatever is stored. A room that reached 6°C for ten minutes during a short outage is a very different situation from one that reached 6°C for two hours during an extended outage, and only interval logging tells them apart.
• If your site experiences frequent outages, cross-reference your temperature log against the outage schedule for your area. A pattern of excursions that line up exactly with load shedding stages tells you the room’s thermal reserve — insulation condition and stock loading — is the thing to address, not the refrigeration system itself, especially if recovery times are getting longer over time.
• Confirm your logging equipment has its own battery backup, separate from the room’s power supply. A logger that stops recording the moment the room loses power gives you a gap in the record at exactly the moment you need data most — most standalone loggers hold several days to weeks of battery reserve, which is worth confirming against your typical outage frequency and duration rather than assuming.
• Set the high-temperature alarm delay with load shedding in mind. A delay of 15 to 30 minutes avoids nuisance alarms on every brief outage, but if outages in your area commonly run longer than that, confirm the delay still gives you enough warning to act — moving stock, starting a generator, or simply being aware — before product temperature becomes a genuine risk.

What a Complete Log Entry Looks Like

Final Thoughts

A log is only useful if it is frequent enough to catch the thing you actually need to see, and a defrost cycle or a load shedding outage is exactly the kind of short, sharp event that an hourly average will erase. Log room air temperature at intervals of a few minutes, log product core temperature where your certification requires it, and keep the defrost, door, and alarm events alongside the temperature trace rather than as separate records nobody cross-references. For how the whole cold room system fits together, see the complete guide to commercial cold room systems. If your log is showing recurring temperature excursions tied to defrost cycles, work through cold room defrost problems explained, and if you are setting up logging on a room that has just been installed, see commissioning a new cold room: the checks that prevent callbacks for what baseline data to capture from day one.