How to Read TEB (Total Error Band) on a Pressure Sensor — and Why It Is Not “Accuracy”
TEB (Total Error Band) describes combined sensor error over the temperature range and conditions stated in the specification; it is not interchangeable with room-temperature accuracy. An accuracy class of 0.5 and a TEB of 0.5%FS may use different error definitions and test conditions. A TEB that includes thermal effects is more useful for assessing variable-temperature service, but equal percentages alone do not establish which product is better. Compare the included errors, temperature band, full-scale basis and typical or maximum values.
Why “accuracy” alone leads to the wrong choice
The error terms included in “accuracy” depend on the model definition and test conditions. RS PM160-E lists room-temperature accuracy separately from temperature-banded TEB. Do not assume every manufacturer uses the same three-term definition.
Field temperature can depart from calibration temperature, changing zero and sensitivity. Establish media and sensor-body temperatures and check drift and compensation data. Ambient temperature alone does not define sensor-body temperature.
TEB combines the calibration and thermal error terms defined by the manufacturer. A typical TEB is not a guaranteed worst-case limit for every unit. Design and acceptance must refer to the applicable maximum, configuration and agreed test conditions.
What goes into TEB
| Error term | What it means | TEB definition check |
|---|---|---|
| Non-linearity | Maximum deviation of the output curve from the ideal straight line | Confirm against model definition |
| Hysteresis | Difference between readings at the same pressure, rising vs falling | Confirm against model definition |
| Repeatability | Spread of repeated measurements at the same pressure | Confirm against model definition |
| Zero thermal drift | Offset shift caused by temperature change | Confirm against model definition |
| Span thermal drift | Full-scale output change caused by temperature change | Confirm against model definition |
| Thermal hysteresis | Residual offset after a temperature cycle back to the starting point | Confirm against model definition |
| Long-term stability | Slow drift measured in years | Usually stated separately |
Note the last row: long-term stability is normally excluded from TEB because its time scale is entirely different. For unattended multi-year monitoring, ask for that figure separately.
How to read TEB on a datasheet
TEB is meaningless without the temperature range it refers to. Our RS PM160-E analog output pressure transmitter publishes it per temperature band, exactly as it should be:
| Temperature band for the stated TEB | Typical TEB | Maximum TEB |
|---|---|---|
| −20 °C to 85 °C | ≤ ±0.25 % FS | ≤ ±0.50 % FS |
| -40 °C to 125 °C | ≤ ±1.0 % FS | ≤ ±1.2 % FS |
Typical values help estimate performance; they do not replace a per-unit acceptance limit. Read maximum values with their temperature range, pressure range and test conditions. The technical agreement should name the datasheet revision, allowable error and test method.
%FS refers to the full-scale basis defined in the datasheet, not the current reading. A 0–10 bar range uses 10 bar in this example; for negative or non-zero-start ranges, confirm whether FS means span or another basis before calculating.
Three steps to your real measurement error
- Establish the actual operating temperature range, considering both ambient conditions and the medium at the sensor. Use the extreme temperatures expected in service rather than an average.
- Use the maximum TEB applicable to the actual temperature band. Equipment operating at −35 to 110°C goes outside the −20 to 85°C standard band, so do not reuse that narrower-band figure. For this confirmed configuration, the example uses a maximum of ±1.2%FS.
- Convert to engineering units using the same −35 to 110°C example. For a 0–10 bar range at a maximum of ±1.2%FS, sensor error is ±0.12 bar, or ±12 kPa. This exceeds a ±0.02 bar total-error budget before acquisition, reference and installation errors are added. A narrower range must still cover the required measurement peaks.
Full parameter tables and per-band data are in the datasheets on our downloads page.
When TEB matters less
Conversely, wherever custody transfer, batch consistency or unattended long-term operation is involved, TEB is a hard criterion that must be checked band by band.
FAQ
TEB or accuracy — which one should I specify?
Check both. Accuracy describes performance under its stated calibration conditions; TEB describes combined error over its stated temperature band. Separately consider long-term stability, installation effects and the acquisition system error.
Typical or maximum — which goes into the design?
Use the applicable maximum for design margin and typical values for performance estimates. Agree acceptance conditions with the manufacturer, including temperature band, configuration and test method; do not relabel a typical figure as a guarantee.
Does TEB include long-term drift?
Usually not. Long-term stability is stated separately on a per-year basis and should be confirmed with the manufacturer, especially for projects that must run for years without recalibration.
How do I choose the temperature band without under-specifying?
Use the extreme temperatures the sensor can actually reach, considering both the medium and ambient conditions rather than average ambient temperature. Check the applicable published or confirmed TEB band to avoid underestimating thermal error.
Where do I find TEB data for Rodee products?
Check published TEB, accuracy and temperature conditions on the relevant product and download pages. Not every core specifies TEB. Where it is absent, ask for applicable specifications rather than deriving an unsupported full-temperature error.
More answers in the FAQ.
Specifications: View PM160-E Datasheet (PDF) · Download PDF
Related products and documentation: RS-PMSC350 calibrated and temperature-compensated pressure sensor core