SAT and TUS: what each one proves, and why one does not cover the other
These two get conflated constantly, including by software that treats them as interchangeable “calibration events” with different names. They are not the same check, they fail in different ways, and a furnace can be perfectly current on one while the other is the reason a load should not run.
Put simply:
- A system accuracy test asks whether the furnace is telling you the truth about the temperature it is measuring.
- A temperature uniformity survey asks whether the temperature is the same everywhere in the space you put parts.
One is about the instrument. The other is about the volume. Neither implies the other.
The system accuracy test: is the reading honest?
A SAT compares what the control system reports against an independent measurement taken at essentially the same place, using a test thermocouple whose own error is known.
The sequence matters, because it is where the arithmetic lives:
- The control system reports a temperature.
- A test thermocouple, with a known correction factor, reports its own.
- The correction is applied to the test reading. This step is not optional — an uncorrected comparison compares one unknown error against another.
- The corrected difference is compared against the permitted tolerance.
Pass or fail falls out of step four. It is not a judgement. If your system has a dropdown where somebody selects the result, it has replaced a measurement with an opinion.
What a passing SAT does not tell you: anything whatsoever about the far corner of the work zone. The control thermocouple sits where it sits. A furnace can be beautifully accurate at that point and eighty degrees cooler at the back of the basket.
The uniformity survey: is the whole zone the same?
A TUS distributes thermocouples through the working volume — corners, centre, wherever the survey pattern requires for the zone’s size — brings the furnace to temperature, and records what every one of them says.
The output is a maximum deviation across the zone, compared against the tolerance for the furnace class. And crucially, it establishes a qualified operating range: the span of temperatures the furnace has demonstrated it can hold uniformly.
That range is the part most systems throw away. It is not documentation. It is an operating limit. A furnace surveyed across one range has no evidence at all about its behaviour outside it, so a recipe calling for a temperature beyond that range is unsupported — regardless of how recent the survey is, and regardless of a perfect SAT.
What a passing TUS does not tell you: whether the instrument reading is currently accurate. A survey performed three months ago says nothing about a controller that drifted last week. That is the SAT’s job, which is why SATs are performed far more often.
Why the intervals differ
The two checks are on different clocks, and the reason is physical rather than bureaucratic.
Instrument accuracy drifts continuously — electronics age, thermocouples degrade, connections change. So it is checked frequently.
Zone uniformity is a property of the furnace’s construction, insulation and airflow. It changes slowly, or abruptly when something is repaired or rebuilt. So it is surveyed less often, and re-surveyed after work that could have changed it.
That last clause matters more than the interval. A furnace whose survey is technically current but which had its elements replaced last month is a furnace whose survey describes a different furnace.
The four states a furnace can be in
Because the checks are independent, a furnace occupies one of four states — and only one of them should take a load:
| SAT | TUS | What it means |
|---|---|---|
| Current | Current | Eligible, if the temperature is inside the qualified range |
| Current | Overdue | Readings are honest; nobody knows if the back corner matches |
| Overdue | Current | Uniformity was demonstrated; the reading may now be lying |
| Overdue | Overdue | Not a furnace, for compliance purposes |
The two middle rows are the interesting ones, because both feel fine. Somebody looking at a recently-passed SAT will reasonably conclude the furnace is in good order. It might be. Nobody has evidence either way about the zone.
What this means for a system
If software treats these as one concept, it cannot represent the middle rows — and the middle rows are where real shops actually live.
A system that models it properly holds them separately, computes each one’s validity from its own date and its own interval, and asks both at once when it matters:
Can this furnace run this cycle at this temperature today?
That question needs the SAT (is the reading honest), the TUS (is the zone uniform, and is this temperature inside the qualified range), the instrument calibrations, and the thermocouples. One answer, assembled from four independent records, none of which substitutes for another.
That is what the gate does before a charge is built — and why an out-of- compliance furnace is refused rather than warned about.
This post describes what each check establishes and how they relate. It deliberately quotes no intervals, tolerances or survey point counts: those vary by furnace class, zone size and the revision of the standard you work to. Your copy governs.