saqr

The warmest reading is almost never the problem.

Open the doors at a delivery in a Riyadh August and the probe will record a number that looks alarming. Meanwhile a unit quietly failing overnight never gets near it, and that is the load you lose.

Three excursions, one week

Same vehicle type, same -18°C setpoint. Switch between them — the verdicts do not follow the peaks.

REF-016Compressor stallDay 3, 01:40200 min window · compartment 1, frozen
setpoint -18°Climit -15°Cdoors shut for the whole window
Peak
-13.4°C
warmest reading
Above limit
102 min
over -15°C
Degree-minutes
105
how far, how long
Doors
Shut throughout
warming unexplained

Load compromised

Reefer unit cut out overnight in the yard. Doors never opened.

Load rejected at the customer's gate. The record is why the claim went to the reefer contract, not to us.

Delivery stop reached -8.2°C and cost nothing. Compressor stall never got past -13.4°C and lost the load. Ranked on peak temperature the two are the wrong way round, which is why peak temperature is not what the verdict is built on.

Four things decide it, not one

No single number adjudicates an excursion, and any vendor offering one is selling you a threshold you will end up ignoring.

Dock wait carries the highest degree-minutes of the three and is still only an advisory, because the doors were open and the unit pulled back down normally. The same curve with the doors shut is a fault.

Time above limit
Under 15 minutes is a door opening. Over an hour is a failure, whatever the peak was.
Door correlation
Warming while the doors are open is physics. The same curve with the doors shut is a fault.
Recovery rate
A unit that pulls back down at the normal rate is working. One that crawls back is the next failure.
Degree-minutes
How far above the limit, for how long, summed. Spoilage does not care about the highest reading.
Excursion Peak Minutes above -15°C Degree-minutes Doors Verdict
Delivery stop -8.2°C 20 66 Open Within tolerance
Compressor stall -13.4°C 102 105 Shut Load compromised
Dock wait -10.6°C 44 130 Open Advisory

A reefer is not one temperature

Multi-temperature trailers run three setpoints behind two bulkheads, and a compartment can fail while the one next to it is perfect. One probe on the return air tells you almost nothing about either.

Probes go where the load is, not where the wiring is convenient — two per frozen and chilled compartment, so a single failed sensor is visible as a disagreement rather than quietly becoming the record.

Compartment 1

-18°C

Frozen

Tolerance
±3°C
Probes
2

Compartment 2

+ 2°C

Chilled

Tolerance
±2°C
Probes
2

Compartment 3

+ 18°C

Ambient

Tolerance
±6°C
Probes
1

The record is the product

Nobody buys temperature monitoring to look at charts. They buy it so that when a receiving department refuses a pallet, there is something to put in front of them the same afternoon.

How a disputed delivery becomes a file
Every reading, not a summary
One sample every 2 minutes per probe, kept for six years. A customer asking about a delivery in 2029 gets the actual trace, not an average.
Gaps are recorded as gaps
If a probe drops out, the record says so at that timestamp. A trace with an unexplained flat section is worth less than one that admits what it does not know.
Door state alongside
Because every reasonable question about a rise starts with whether the doors were open.
Signed export
The window either side of a delivery as one file, in the format receiving departments and SFDA submissions expect.
1 load
is the whole argument

A rejected pallet of frozen goods costs more than a year of monitoring on that vehicle.

6 yrs
of retained readings

Long enough to outlast the contract the dispute will be about.

2 min
between readings

Fine enough that a fifteen-minute door opening has seven points in it, not one.

The rest of the platform

Bring a delivery you had to argue about See pricing Cold chain is on the Command plan; probes are fitted per compartment.