On one RC-5 sample tested in 2018, 32 000 points lasted about 89 hours at a 10-second interval and about 22 days at a 1-minute interval. That is the author’s interval–memory arithmetic on that firmware and software. It is not a battery-hour figure, and it is not a Temp Logger UK export.
What did the 2018 RC-5 test actually count?
Jos Verstraten’s RC-5 note lists selectable intervals from 10 seconds to 24 hours and a memory of 32 000 points. He then did the arithmetic that a shop still has to do before a weekly run: at 10 seconds the file fills in about 89 hours; at 1 minute it lasts about 22 days. Those two conversions are his, on that sample, with the 2018 Windows program RC-4&5_Conventional.exe.
That program is not the current ElitechLog desktop. A later page that reports a current-software duration will have to show the setting screen and the first and last rows of the export. This publication has not opened that file.
Why is that arithmetic not a battery-life figure?
32 000 points ÷ 6 samples per minute = 5 333 minutes ≈ 89 hours. 32 000 points ÷ 1 sample per minute = 32 000 minutes ≈ 22.2 days. Those identities only convert memory against interval. They say nothing about the coin cell. Verstraten recorded a CR2032; he did not publish a discharge curve in the passage used here.
NHS SPS illustrates a medicines-storage logger at 5 minutes for six months. That is a professional example for a storage area, not a kitchen USB pen, and not a result from this site. A 5-minute setting on 32 000 points is 32 000 × 5 minutes = 160 000 minutes ≈ 111 days — again only arithmetic, and only if the instrument actually holds 32 000 points and actually uses that interval.
What would a later duration page have to show?
A page that says “this instrument lasts n days at 5 minutes” needs the model and firmware, the software version, the setting screenshot, the calculated duration, and an export whose first and last timestamps match that interval. Battery hours need a discharge record as well. Until those files exist, this note can only attribute Verstraten’s 10-second and 1-minute conversions.
A shop catalogue that prints 32 000 points is a statement. It is not the filled file.
Which sources does this note use, and what do they not prove?
Sources checked on 23 September 2026. Numbered references point to the original publishers. Their sample measurements are not Temp Logger UK measurements. Manufacturer, training and official material is identified separately. A shop catalogue sentence can name a model, but it does not establish the columns, the clock or a pass. This publication has not tested or certified any instrument in this note.
- Chinese electronics products tested, RC-5 temperature logger tested ↗Jos Verstraten, 13 August 2018. Interval, Play/Stop, delay start, memory-full behaviour and export formats on one RC-5 sample with then-current Windows software. Those figures belong to that 2018 test, not to a current ElitechLog session or a Temp Logger UK export.
- NHS Specialist Pharmacy Service, Using temperature sensors in medicines storage areas ↗Medicines storage, not kitchen HACCP. Distinguishes min/max from a continuous logger; treats an air sensor as an early warning; gives a 5-minute × 6-month illustration. Official professional guidance.
No instrument in this note has been tested or certified by this publication. Evidence policy · Corrections
Write the conditions beside the reading.
The CSV has column headings and one empty row. It contains no fabricated sample readings.
Download temperature-log record ↓
