Is ±0.3 °C accurate enough?

Answer: For most work, yes. It separates differences above 0.6 °C, so a 0.4 °C gap is not a finding; 0.2 °C questions need a ±0.2 °C external probe.

Which interval should I set?

Answer: Fifteen minutes for transit and storage, five when the warm-up curve matters, one minute only for a single loading cycle.

Can a probe pass through the wall?

Answer: Yes: 3.2 mm hole, IP67 gland, 30 mm from the sealing face, 0.4–0.6 N·m, then 20 kPa held to a 2 kPa drop in 60 s.

How often should data be exported?

Answer: Every 300 days at 15 min, in CSV and PDF, with certificates filed against serial numbers.

Monitoring Inside a Case: Loggers and Placement

Q: How many points does Monitoring Inside a Case need?

Answer: Three: 30 mm below the lid, the geometric centre and 50 mm above the floor, all 50 mm from a wall.

Quick answer: Monitoring Inside a Case is reliable when a -40 to 85 °C logger at ±0.3 °C sits at three fixed points, samples every 5–15 min and is calibrated yearly.

Why Readings Inside a Case Mislead: Three False Numbers
Choosing the Logger: Range, Accuracy, Interval
Internal and External Probes: Sealing the Penetration
Placement: Three Points and Wall Clearance
Making the Data Usable: Timestamps, Calibration, Reports
FAQ

1. Why Readings Inside a Case Mislead: Three False Numbers

Two loggers in one case, on one load, can disagree by 2.4 °C, and the instruments are usually innocent. Monitoring Inside a Case means measuring a small volume that stratifies: air 20 mm under the lid runs more than 2 °C warmer than air 50 mm off the floor, and a probe against a sunlit wall reads 0.8–1.5 °C high. Bury a probe in foam and it reports foam temperature, lagging 20–40 min and flattening every peak by thirty to forty percent.

The second false number comes from the door. Thirty seconds of access lifts the inside by 3–6 °C, and a 15 min interval walks straight past it, so the chart shows what is left after eighty percent of the recovery.

The third comes from accuracy itself. A ±0.3 °C sensor separates differences above 0.6 °C, so a 0.4 °C gap written up as a finding is noise wearing a decimal point.

Response time adds a fourth. Most units reach T90 in 5–8 min, a foam-wrapped probe takes over 30 min, and self-heating of 0.05–0.1 °C sits on top. Humidity cannot be read alone either: a 5 °C drop in temperature lifts relative humidity by roughly thirty percent.

2. Choosing the Logger: Range, Accuracy, Interval

Five parameters, in this order: range, accuracy, interval, memory, battery. Range decides whether a unit can be used, accuracy whether a conclusion can be written, the rest whether the routine lasts a season. Transport and storage inside a case sit between -30 and 60 °C, so -40 to 85 °C leaves margin at both ends.

Type	Range	Accuracy	Interval	Memory	Battery
Single-use USB	-30 to 70 °C	±0.5 °C	Fixed 10 min	16000 points	Sealed
Reusable LCD	-40 to 85 °C	±0.3 °C, ±2% RH	5–15 min	32000 points	CR2032, 12 months
External probe	-40 to 125 °C	±0.2 °C	1 s to 18 h	64000 points	2×AAA, 24 months
Wireless gateway	-20 to 60 °C	±0.3 °C	1–60 min	Cloud	Lithium, 6 months

Work memory backwards from the interval. At 15 min, 32000 points is 96 records a day and 333 days of cover, so exporting at 300 days keeps a ten percent margin rather than a full buffer. A CR2032 at -20 °C gives about sixty percent of rated capacity: 12 months becomes 7–8, and winter projects go quarterly on batteries.

Interval trades resolution against capacity: five minutes captures the warm-up after a door opens at three times the cost, fifteen suits transit and storage.

3. Internal and External Probes: Sealing the Penetration

An internal probe needs no hole, but the logger's position is set by a foam pocket, not by the measurement. An external probe on a Ø3 mm lead goes anywhere, at the cost of a hole: seal it badly and the record is bought with water ingress.

Monitoring Inside a Case placement starts at the corner: white shell, grey foam along the wall and a grey box in a foam pocket

A penetration has four numbers. Drill 3.2 mm for an IP67 gland and verify it against the immersion condition of IEC 60529; keep the hole 30 mm clear of the sealing face so the gland body does not press the lip; tighten to 0.4–0.6 N·m; specify a silicone jacket, because PVC stiffens at -30 °C.

Retest afterwards: charge to 20 kPa and allow no more than a 2 kPa drop in 60 s, the factory criterion. A visual check is not a test, since water shows up in month two.

4. Placement: Three Points and Wall Clearance

Three points is the smallest arrangement that can be explained: 30 mm below the lid, the geometric centre and 50 mm above the floor. Together they show vertical stratification; alone, the centre can stand for the volume while the outer two check the spread. All three stay 50 mm off any wall, clear of stored heat and condensation.

Monitoring Inside a Case needs instruments side by side: a grey case with foam, a keypad logger left and a paper-roll logger right

Different readings at the three points are the point, not a fault. Warm air collects near the lid, so the top reads 0.4–0.8 °C above centre; the floor point reads 0.8–1.5 °C below it; a foam pocket stretches the lag to 20–40 min and shaves the peak flat.

Position	Reading vs centre	Lag	Use
Top, 30 mm below lid	0.4–0.8 °C higher	5–10 min	Stratification, lid condensation
Geometric centre	Reference	None	Single-point long records
Bottom, 50 mm above floor	0.8–1.5 °C lower	10–20 min	Heavy loads, floor storage
Against the wall	0.8–1.5 °C higher	5–15 min	Reference only, not recommended
Inside a foam pocket	Depends on depth	20–40 min	Not recommended, peaks lost

Write the positions into the work card. One case family, one set of points, or batches stop being comparable: moving a probe changes the system, and the old curve cannot be extended.

5. Making the Data Usable: Timestamps, Calibration, Reports

Data is usable only when four things arrive together: a timestamp, a serial number, the interval and the calibration date. Without all four, a report proves something was measured, not what. Once a record enters a claim, those four are the first questions asked.

Calibrate every 12 months and retire a unit drifting past 0.5 °C or 3% RH. Certificates carry the serial number, and instruments from one certificate should not be mixed across a fleet. Set clocks before comparing months: drift is about a minute a month, 12 min a year, two whole records at a 5 min interval.

Export CSV and PDF together, the first for recomputation and the second for delivery, each page carrying timestamp, serial number, interval and range. Mark gaps longer than 5 min on the chart rather than interpolating across them, because a smoothed line proves nothing about the gap it hides.

Conclusion

A temperature record that contradicts itself is rarely an instrument problem and usually a placement one: a probe in foam loses the peak, a probe on the wall reports the sun, a 15 min interval walks past a 30 s door opening. Monitoring Inside a Case comes down to fixed numbers, a -40 to 85 °C unit at ±0.3 °C and ±2% RH, three points 30 mm under the lid, at centre and 50 mm above the floor, 50 mm off any wall, 5–15 min sampling, a 3.2 mm gland at 0.4–0.6 N·m and a 12 month calibration. Then a 0.4 °C gap stays a number, not a conclusion.

This article discusses only the container (case body) side of specifications and design intent, and does not cover any handling, loading or use of the contents.