Understanding Temperature Sensors in Machinery Monitoring
A temperature sensor, in the machinery-monitoring sense, is a device that measures the temperature of bearings, motor windings, process fluids, or equipment surfaces, supplying critical information for detecting overheating, lubrication problems, excessive friction, and abnormal operating conditions. Where vibration monitoring detects mechanical defects, temperature monitoring reports on thermal condition — and the combination of the two is far more powerful than either alone. Because many machinery failures are preceded by a rise in temperature — bearings overheating from friction, windings from overload, seals from rubbing — temperature is a cornerstone of any serious condition monitoring programme, and trending it lets the team intervene before a fault turns catastrophic.
1. Common Sensor Types for Machinery
Four technologies cover almost all rotating-equipment applications, trading accuracy, range, ruggedness, and cost against one another.
RTD (Resistance Temperature Detector)
The most accurate and stable choice, and the default for critical bearings and motor windings.
- Principle: a platinum wire whose electrical resistance changes predictably with temperature.
- Common types: Pt100 (100 Ω at 0 °C) and Pt1000.
- Accuracy: typically ±0.1–0.5 °C.
- Range: −200 to +600 °C, with excellent long-term stability.
- Cost: moderate to high — justified for critical and precision measurements.
Thermocouple
Wide-range and rugged, well suited to hot and harsh environments.
- Principle: a junction of two dissimilar metals generates a small voltage proportional to temperature (the Seebeck effect).
- Types: Type K (most common), plus Types J, T, and E.
- Accuracy: typically ±1–3 °C.
- Range: −200 to +1300 °C depending on type, at low cost.
- Applications: high-temperature monitoring such as exhausts and furnaces.
Thermistor
- Principle: a semiconductor whose resistance is extremely temperature-sensitive.
- Sensitivity: very high — a large resistance change per degree.
- Accuracy: ±0.1–1 °C over a limited range (typically −50 to +150 °C).
- Applications: consumer equipment and some industrial uses, at low cost.
Infrared (Non-Contact)
- Principle: detects the thermal radiation emitted by a surface, so no physical contact is needed.
- Range: −50 to +1000 °C and beyond, with accuracy around ±2–5 % of reading.
- Applications: spot checks and thermal-imaging surveys; the same physics underpins thermography, where surface emissivity and target distance must be accounted for. Sensible alarm thresholds for such surveys follow guidance like ISO 18434 thermography limits.
2. Bearing Temperature Monitoring
The bearing is the single most common temperature-monitoring target, because rolling-element and journal bearings convert lubrication breakdown and overload straight into heat.
Measurement locations
- Embedded in the bearing housing, as close to the outer race as possible.
- Surface-mounted on the bearing cap.
- In the oil drain, for oil-lubricated bearings.
- At several points around large bearings, where temperature is not uniform.
Normal temperature ranges
- Ambient + 20–40 °C: normal operating temperature.
- Ambient + 50–60 °C: the acceptable maximum for most bearings.
- > Ambient + 70 °C: a problem is indicated — investigate.
- > 90–100 °C absolute: an alarm condition for most bearings.
These rules of thumb should always be checked against the manufacturer’s data and the relevant component temperature limits for the specific bearing, seal, and lubricant; a high-speed grease may be near its limit at a temperature a circulating-oil bearing tolerates comfortably. Large machines often have dedicated guidance, such as a generator bearing temperature monitor.
Trending and alarms
- Establish a baseline temperature for each bearing under known load and ambient conditions.
- Raise a warning on a 10–15 °C rise from baseline.
- Raise an alarm on a 20–25 °C rise, or on an absolute limit.
- Trip (shut down) on a 30–40 °C rise or a critical absolute value.
Because ambient temperature and load both shift the reading, trending the change from baseline is usually more revealing than any single absolute number — a steady upward drift is the classic early-warning signature of a degrading bearing. Multi-parameter standards such as ISO 13373 formalise how these alarm and danger thresholds are set.
3. Integration with Vibration Monitoring
Temperature and vibration are complementary measurements, and reading them together sharply improves diagnostic confidence. Vibration detects mechanical defects early, often long before any heat appears; temperature confirms severity and flags friction or lubrication problems that vibration alone may not localise.
The two parameters combine into a simple, powerful diagnostic matrix:
- High vibration + normal temperature: a mechanical issue such as unbalance or misalignment — forces are high but friction is not yet excessive.
- High vibration + high temperature: a bearing defect with significant friction, typically an advanced stage.
- Normal vibration + high temperature: a lubrication problem, or alignment/preload causing friction, such as a tight or rubbing seal.
- Both increasing: a progressing bearing failure approaching the end of its useful life.
This pairing is exactly why mature predictive-maintenance routes collect both parameters at every measurement point. In practice the vibration half of the picture is gathered with a portable analyser — for example, a two-channel instrument such as the Balanset-1A measures amplitude and phase at the bearing housings while the machine runs, so that a spot temperature reading taken at the same point can be interpreted against what the vibration is doing rather than in isolation.
4. Installation Best Practices
A temperature reading is only as trustworthy as the thermal path between the sensor and the heat it is meant to measure.
Sensor placement
- Position the sensor as close to the heat source — the bearing — as practicable.
- Ensure good thermal contact with the measured surface, using thermal paste to eliminate air gaps.
- Shield the sensor from ambient temperature swings and from radiant or convective heat sources that are not the target.
Wiring
- Use the correct wire type for the sensor — compensating cable is mandatory for thermocouples to avoid introducing spurious junctions.
- Route signal cables away from high-current and high-voltage conductors to limit electrical noise.
- Terminate connections properly, and shield and ground the run where the environment demands it.
5. Typical Applications
Temperature monitoring appears across the whole spectrum of rotating equipment:
- Bearing monitoring: the most common application — early detection of lubrication problems, confirmation of a developing bearing defect, and overload detection.
- Motor protection: winding temperature via embedded RTDs (the front line of defence against stator overheating and other motor defects), plus bearing and frame temperatures that reveal overload and inadequate cooling.
- Process equipment: pumps (bearing, seal, and casing temperatures), compressors (discharge and bearing temperatures), and gearboxes (oil-sump temperature).
Temperature sensors are indispensable companions to vibration sensors in a comprehensive machinery-monitoring programme. Vibration analysis catches mechanical defects early, while temperature monitoring confirms thermal condition, friction, and lubrication adequacy — together delivering a more complete picture of equipment health, and earlier warning across a wider range of failure modes, than either technology can offer on its own.