— A Practical Guide for Engineers, EPCs, and Plant Operators

In industrial process control, temperature is one of the most critical parameters. It drives safety decisions, product quality, energy efficiency, and regulatory compliance. Yet temperature measurement is surprisingly prone to error—and many of these errors are entirely preventable.

The most common source of error in temperature measurement is the sensor not accurately reflecting the temperature of its environment--1. This can happen for many reasons: poor thermal contact, sensor degradation, wiring faults, improper installation, or electromagnetic interference. This guide breaks down the major sources of temperature measurement errors and provides practical solutions for preventing them.

1. Sensor Selection Errors

1.1 Choosing the Wrong Sensor Type

The Problem: Selecting a thermocouple when an RTD is required—or vice versa—based on habit rather than application requirements.

Why It Matters: Thermocouples and RTDs have fundamentally different accuracy, stability, and temperature range characteristics. RTDs offer high accuracy, typically ±0.2°C, and are commonly used in medium temperature ranges from -200°C to +600°C-8. Thermocouples, while more robust and capable of measuring much higher temperatures, are generally less accurate-17. Over a range from 0 to 1000°C, RTDs have an accuracy of ±0.5 to 1°C, while thermocouples exhibit increased non-linearity error over wider temperature ranges-.

The Prevention:

• If the process requires accuracy better than ±0.5°C and temperatures are below 600°C, choose an RTD

• If temperatures exceed 600°C or the environment is harsh and high-vibration, choose a thermocouple

• For critical measurements, calibrate thermocouples or RTDs against known temperature references-

1.2 Ignoring Sensor Tolerance and Accuracy Class

The Problem: Specifying a sensor without understanding its tolerance class and how it affects overall measurement uncertainty.

The Prevention: Always specify the required accuracy class (e.g., Class A or Class B for RTDs per IEC 60751; Class 1 or Class 2 for thermocouples per IEC 60584). Remember that tolerances apply only to new sensors—real-world use can cause drift due to insulation breakdown, mechanical damage, and other factors-.

2. Installation Errors

2.1 Insufficient Immersion Depth (Conduction Error)

The Problem: The sensor is not inserted far enough into the process, causing heat to be conducted away along the thermowell or sensor sheath. Conduction is the largest source of error in contact-type temperature sensors-5.

Why It Matters: Conventional thermowells thermally "couple" with the vessel in which they are mounted, resulting in measurement error whenever a temperature gradient exists between the vessel or pipe and the substance being measured--32. If the sensor tip does not bottom out in the thermowell, an air gap is created—and since air has low thermal conductivity, this introduces significant sensor lag and measurement error-.

The Prevention:

• Immerse the sensor to a depth of at least 10 times the thermowell diameter-31

• For a bare element with a ¼ inch (6.35 mm) diameter sensor sheath, the immersion length should be at least 2.5 inches (63.5 mm)-31

• A more conservative rule: immerse to a minimum depth equal to 20 times the probe diameter plus the length of the sensing element-

• Ensure the sensor tip touches the bottom of the thermowell

• Use spring-loaded sensor designs to maintain contact despite thermal cycling and vibration

2.2 Poor Thermal Contact

The Problem: The sensor is not in good thermal contact with the medium or its housing, creating an insulating air gap-.

Why It Matters: Poor contact, especially in closed control loops, can cause lag, error, and unstable process control-1.

The Prevention:

• Visually inspect for clean and intact contact areas and secure mechanical junctions-1

• Ensure adequate insertion depth and proper positioning in the process-1

• Fill the thermowell with heat-conductive compound to improve thermal transfer

• Check for foreign matter inside the thermowell that could prevent full insertion

2.3 Incorrect Thermowell Location

The Problem: The thermowell is installed in a location that does not provide a representative temperature measurement.

Why It Matters: For highly viscous fluids such as polymers and melts, the fluid temperature near the wall can be significantly different than at the centreline (e.g., 10 to 30°C difference)-31.

The Prevention:

• Insert the thermowell in an elbow facing into the flow for a representative centreline measurement-31

• For small-diameter pipelines (less than 4 inches), use angled insertion to increase immersion length-31

• Place the thermowell at least 25 pipe diameters from mixers or heat exchangers to ensure adequate mixing, while balancing transportation delay-31

3. Wiring and Electrical Errors

3.1 Incorrect Thermocouple Wiring

The Problem: Reversed polarity or mismatched extension cables in thermocouple circuits.

Why It Matters: Thermocouples are particularly sensitive to connection polarity and cable alloys. Reversed polarity leads to downscale readings or large errors (up to 100°C)-1. A frequent problem is that cold junction compensation is not configured or compensated properly, or not done at all-.

The Prevention:

• Always check cable colour codes and type-1

• Verify correct polarity (positive to positive)-1

• Use the correct thermocouple extension wire—matching the thermocouple type

• Confirm cold junction compensation is functioning correctly in thermocouple systems-1

3.2 Electromagnetic Interference (EMI)

The Problem: Electrical noise from motors, VFDs, radios, and power mains corrupts the low-level signal from thermocouples and RTDs-2.

Why It Matters: Thermocouples create a voltage signal of less than 50 mV and have practically no ability to create current flow. This makes them behave like an antenna—and the longer the wires, the more opportunity to "receive" electrical noise-2. RTD sense wires also act like antennas, picking up noise present in the industrial environment-2.

Symptoms of noise-affected temperature signals:

• Readings change when a nearby motor or heater starts-2

• Readings change when a mobile radio is transmitting-2

• Readings change based on time of day or weather patterns-2

• Readings change based on the location or orientation of sensor wiring-2

The Prevention:

• Minimise the length of sensor wiring to reduce the "antenna" effect-2

• Convert thermocouple and RTD signals to 4-20 mA current at the sensor—EMI has practically no effect on current loops-2

• Twist loop conductors around each other to ensure both wires are equally shielded-2

• Use shielded cables and proper grounding

• Keep signal cables separate from power cables

3.3 RTD Lead Wire Resistance

The Problem: Uncompensated lead wire resistance in RTD measurements, particularly in 2-wire configurations.

Why It Matters: For RTDs, it is straightforward to estimate the temperature error from an uncompensated lead wire resistance-. The RTD itself can contribute error in the form of sensor accuracy variation, self-heating, and lead wire mismatch-.

The Prevention:

• Use 3-wire configurations for industrial applications (compensates for lead resistance)

• Use 4-wire configurations for laboratory precision

• Only use 2-wire for short distances where lead resistance error is negligible

3.4 RTD Self-Heating

The Problem: The excitation current used to measure the RTD's resistance generates heat (I²R loss), causing the RTD to read higher than the actual temperature.

Why It Matters: Self-heating causes a positive error in temperature measurement-. The error will further increase in a high temperature range when a current-source method is adopted-.

The Prevention:

• Keep excitation current to a small value (1–5 mA)-

• Use pulse excitation of the bridge-

• Choose a lower resistance RTD (e.g., Pt-100 rather than Pt-1000) to decrease self-heating error-

• Monitor power dissipation to the RTD and keep it very low-

4. Sensor Degradation and Drift

The Problem: Sensors drift from their original calibration over time due to insulation breakdown, mechanical damage, chemical poisoning, or thermal cycling-1-.

Why It Matters: Sensor drift typically happens over time and differs between sensor types. Thermocouples may suffer from poor insulation, thermoelectric inhomogeneity, or junction deterioration. RTDs are prone to internal lead issues, mechanical shock, chemical poisoning, or lead wire imbalance-1.

The Prevention:

• Implement regular calibration schedules (RTDs generally maintain calibration better than thermocouples due to their inherent stability)-

• Use diagnostic checks to confirm degradation-1

• For critical measurements, use dual-sensor assemblies to detect drift by comparing readings-

• Keep a log of when equipment was last tested to maintain a clear "paper trail" of accuracy-

• Calibrate reference thermometers annually by a certified laboratory-

5. Poor Insulation Resistance

The Problem: Low insulation resistance in RTD sensors creates a new measuring point at the location of the low isolation-.

Why It Matters: Poor insulation resistance is the largest cause of measurement error and failure in RTD sensors-.

The Prevention:

• Regularly check insulation resistance with a megohmmeter

• Use high-quality insulated wires and connectors

• Protect sensor terminations from moisture and contamination

• Follow manufacturer recommendations for installation and sealing

6. Summary: Temperature Measurement Error Sources and Prevention

7. Why Choose Anhui Tiankang for Temperature Measurement?

Anhui Tiankang (Group) Co., Ltd. has been manufacturing industrial temperature instruments for nearly five decades. Our temperature measurement solutions are trusted by major oil and gas companies, chemical plants, and power generation facilities worldwide.

Complete temperature product portfolio:

• RTDs: Pt100, Pt1000 – thin-film and wire-wound, Class A/B accuracy

• Thermocouples: Types K, N, E, J, S, R, B – base metal and noble metal

• Thermowells: 316L SS, 310S, Inconel, ceramic – straight, tapered, stepped geometries

• Temperature transmitters: 4-20 mA + HART, Modbus – head-mounted and rail-mounted

• COT temperature instruments: Specialised for ethylene cracker coil outlet temperature measurement

Core advantages:

• High-purity elements and wires – Ensuring stable output and long life

• CNAS-accredited laboratory – Full performance testing for accuracy, stability, and response time

• Comprehensive certifications – CCC Ex, ATEX, IECEx, SIL, CCS marine

• Customisation – Length, diameter, mounting style, special materials

• Proven track record – Long-term supplier to CNPC, Sinopec, CNOOC, and international EPC projects

8. Conclusion

Temperature measurement errors in industrial processes are not inevitable. The majority of errors can be traced to a handful of preventable causes: insufficient immersion depth, wiring mistakes, EMI, sensor drift, and poor installation practices.

Key takeaways:

• Immersion depth is the most common and preventable error—follow the 10× diameter rule

• Wiring is more often the problem than the sensor itself—always inspect wiring first-1

• EMI can be eliminated by converting sensor signals to 4-20 mA at the sensor-2

• Sensor drift is manageable with regular calibration and diagnostic checks

• Poor insulation resistance is the largest cause of RTD failure—check it regularly

Remember: Sensor faults are rarely the root cause of measurement errors. Start with wiring and connection checks, then look at thermal contact and sensor degradation-1. A systematic approach to installation, wiring, and maintenance will eliminate most temperature measurement errors before they affect your process.

Contact Us

For temperature measurement selection advice, technical documentation, or project quotations, please contact:

Yin Shuangjie
International Sales Manager
📧 Email: [email protected]
📱 WhatsApp / Zalo: +86 17856068126
🌐 Website: http://www.tiankang-global.com/

Anhui Tiankang – Your partner for reliable temperature measurement solutions.