13.3 Temperature Measurement: RTD Pt100 vs. Thermocouples & Cold Junction Compensation

13.3 Temperature Measurement: RTD Pt100 vs. Thermocouples & Cold Junction Compensation

Temperature is the most frequently monitored and controlled physical variable in industrial processes. From petrochemical distillation towers and metallurgical blast furnaces to pharmaceutical autoclaves and food pasteurization tunnels, precise temperature measurement is critical to product quality, personnel safety, and process efficiency. Industrial electricians are tasked with selecting, installing, wiring, and calibrating the two dominant electrical temperature-sensing technologies: Resistance Temperature Detectors (RTDs) and Thermocouples. Understanding the fundamental physical operating principles, circuit topologies, lead-wire error mitigation strategies, and failure modes of each sensor type is essential for Red Seal certification and field competence.

---

1. Resistance Temperature Detectors (RTDs): Principles & Standards

Operating Principle & Positive Temperature Coefficient (PTC)

An RTD operates on the fundamental physical principle that the electrical resistance of pure metals increases predictably with increasing temperature. This phenomenon is known as a Positive Temperature Coefficient (PTC). As thermal kinetic energy increases within the metallic crystalline lattice, metal atoms vibrate more vigorously about their equilibrium positions. These thermal vibrations increase the collision cross-section for conduction electrons drifting through the lattice, thereby impeding electron flow and increasing electrical resistance.

Platinum Standard (Pt100 and Pt1000)

While copper (Cu10, Cu100) and nickel (Ni120) are occasionally used in specialized HVAC and motor winding applications, platinum (Pt) is the global standard for industrial process RTDs due to its exceptional chemical inertness, high melting point (1,768°C), resistance to oxidation and corrosion, wide operating range (-200°C to +850°C), and extraordinary long-term calibration stability (drift < 0.1°C per year).

Under international standard IEC 60751 and ASTM E1137:

• Pt100 RTD: Exhibits an electrical resistance of exactly 100.00 Ω at 0.00°C.
• Temperature Coefficient of Resistance (alpha): The DIN/IEC standard defines alpha = 0.003850 Ω/Ω/°C (frequently referred to as the European curve, standard in Canadian industry): alpha = (R_100 - R_0) / (100 × R_0) = (138.50 Ω - 100.00 Ω) / (100 × 100.00 Ω) = 0.00385 Ω/Ω/°C Note: A legacy US industrial standard exists with alpha = 0.003916 Ω/Ω/°C (using 99.999% ultra-pure platinum). Connecting a US-curve sensor to a transmitter configured for DIN 0.00385 results in substantial temperature measurement errors.
• Pt1000 RTD: Exhibits 1,000.00 Ω at 0.00°C with an identical alpha of 0.00385. At 100°C, a Pt1000 measures 1,385.0 Ω. Because its nominal resistance is ten times higher, the effect of lead-wire resistance is reduced tenfold, making Pt1000 sensors popular in building automation and HVAC systems.

Callendar-Van Dusen Equation & Linearity

Over the range of 0°C to +850°C, the relationship between platinum resistance and temperature is governed by the Callendar-Van Dusen equation: R(T) = R_0 × (1 + A·T + B·T^2)

For temperatures below 0°C (down to -200°C), a cubic term is added: R(T) = R_0 × (1 + A·T + B·T^2 + C·(T - 100°C)·T^3)

Where for standard DIN/IEC Pt100:

• R_0 = 100.00 Ω
• A = 3.9083 × 10^-3 °C^-1
• B = -5.7750 × 10^-7 °C^-2
• C = -4.1830 × 10^-12 °C^-4

Over a narrow range near ambient (0°C to 100°C), the response is nearly linear, increasing by approximately 0.385 Ω per 1.0°C (or roughly 1.0 Ω per 2.6°C).

---

2. RTD Wiring Topologies & Lead-Wire Resistance Cancellation

Because an RTD element changes by only 0.385 Ω per degree Celsius, the electrical resistance of the field copper wiring connecting the RTD element inside a thermowell to the input card in a remote MCC or control room introduces significant measurement errors if not cancelled.

   TWO-WIRE RTD (NO LEAD CANCELLATION)
   Transmitter Terminal 1 ──(R_lead1)──┐
                                       [ Pt100 Element ]  R_measured = R_RTD + R_lead1 + R_lead2
   Transmitter Terminal 2 ──(R_lead2)──┘

   THREE-WIRE RTD (WHEATSTONE BRIDGE / RATIOMETRIC CANCELLATION)
   Transmitter Terminal 1 (A) ──(R_lead1)──┐
                                           [ Pt100 Element ]
   Transmitter Terminal 2 (B) ──(R_lead2)──┴──┐ (R_lead2 senses bridge midpoint)
   Transmitter Terminal 3 (C) ──(R_lead3)─────┘

   FOUR-WIRE RTD (KELVIN CONNECTION - ZERO LEAD ERROR)
   Current Source (+) ──(R_lead1)──────────┐
   Sense (+)          ──(R_lead2)──────┐   │
                                       [ Pt100 Element ]
   Sense (-)          ──(R_lead3)──────┘   │
   Current Source (-) ──(R_lead4)──────────┘

1. Two-Wire RTD Circuit

• Circuit Mechanics: Two copper conductors connect the RTD sensing element to the measuring instrument. The measuring circuit applies an excitation current and measures the total loop resistance: R_measured = R_RTD + R_lead1 + R_lead2
• The Problem: If a 100-metre run of 18 AWG copper cable has a loop resistance of 4.2 Ω, that 4.2 Ω is added directly to R_RTD. Since 0.385 Ω ≈ 1.0°C: Error = 4.2 Ω / (0.385 Ω/°C) = +10.9°C The instrument reads 10.9°C higher than the actual temperature. Even worse, as ambient temperatures along the cable tray fluctuate between summer and winter, the lead resistance changes, creating an unpredictable, drifting measurement error.
• Application: Strictly limited to very short distances (< 2 metres), such as local head-mounted transmitters, or non-critical HVAC monitoring utilizing high-resistance Pt1000 or thermistor sensors.

2. Three-Wire RTD Circuit (Industrial Workhorse)

• Circuit Mechanics: Uses three conductors (typically color-coded two white and one red, or two red and one white). Two leads connect to one side of the RTD element, and one lead connects to the other side.
• Lead Cancellation Mechanism: In a classic Wheatstone bridge circuit:
  • Lead A is placed in series with the RTD in one bridge leg.
  • Lead C is placed in series with a precision bridge reference resistor (R_3) in the opposite bridge leg.
  • Lead B acts as the high-impedance sense line connected to the bridge null detector (voltmeter). As long as Lead A and Lead C have identical lengths, identical wire gauges, and reside in the same cable (R_lead1 = R_lead3), the lead wire resistances cancel each other out in the bridge balance equation: (R_RTD + R_lead1) × R_2 = (R_3 + R_lead3) × R_1 Because no current flows through high-impedance Lead B, Lead B introduces zero voltage drop.
• Key Field Requirement: Conductors must be perfectly matched. If an electrician splices mismatched wire gauges or repairs one damaged lead with a different conductor size, the bridge becomes unbalanced, resulting in permanent calibration errors.

3. Four-Wire RTD Circuit (Kelvin Connection - Laboratory & Precision Process)

• Circuit Mechanics: Employs four conductors. Two outer conductors form the excitation circuit, where an internal precision current source forces a known constant current (typically 0.5 mA to 1.0 mA) through the RTD element.
• Voltage Sensing: Two inner conductors connect directly to the terminals of the RTD element and lead to an ultra-high input impedance voltmeter (Z_in > 10 MΩ).
• Total Cancellation: Because the voltmeter has near-infinite input impedance, virtually zero current flows through the voltage sensing leads (I_sense ≈ 0 A). By Ohm's Law (V = I × R), the voltage drop across the sense leads is exactly zero: V_measured = I_excitation × R_RTD R_RTD = V_measured / I_excitation Lead resistance, wire length, gauge mismatches, and ambient temperature fluctuations along the cable have zero impact on measurement accuracy. This topology is mandatory for custody transfer, fiscal metering, and safety-critical reactors.

Self-Heating Error

When excitation current passes through an RTD element, electrical power is dissipated as heat according to Joule's Law (P = I^2 × R). This self-heating raises the sensor element temperature above the surrounding process fluid. To prevent self-heating errors, industrial transmitters limit excitation current to 1.0 mA or less (dissipating < 0.1 mW across a 100 Ω element).

---

3. Thermocouples: Thermoelectric Principles & The Seebeck Effect

The Seebeck Effect

Discovered by Thomas Johann Seebeck in 1821, the Seebeck effect is the direct conversion of temperature differences into electrical voltage. When two dissimilar electrical conductors or alloys are joined at both ends to form a closed circuit, and a temperature gradient exists between the two junctions, a continuous thermoelectric current flows through the circuit.

In practical instrumentation, the circuit is opened to measure the open-circuit thermoelectric voltage (Seebeck electromotive force, EMF): V = S_AB × (T_meas - T_ref) Where:

• S_AB is the Seebeck coefficient (thermoelectric sensitivity, expressed in µV/°C) of the specific pair of metals.
• T_meas is the temperature of the Measurement Junction (Hot Junction) located in the process thermowell.
• T_ref is the temperature of the Reference Junction (Cold Junction) located at the input terminals of the measuring instrument or transmitter.

   MEASUREMENT (HOT) JUNCTION                    REFERENCE (COLD) JUNCTION
   Inside Process Thermowell                     Instrument Terminal Block (at T_ref)
          ┌────────────────────────────────────────────────────────┐
          │                                                        │
       T_meas                                                      ├──(+) Millivoltmeter
      (Process) ──[ Dissimilar Metal A (e.g., Iron / Chromel) ]───┤
          │                                                        ├──(-) Measures:
          │                                                        │      V_net = V(T_meas) - V(T_ref)
          └───────[ Dissimilar Metal B (e.g., Constantan/Alumel) ]─┘
                                                 ▲
                                        [CJ Temperature Sensor]
                                       (Internal RTD / Thermistor)

Critical Conceptual Rule: Differential Measurement

A thermocouple never measures absolute temperature directly. It generates a millivolt signal proportional strictly to the temperature differential between the hot measurement junction and the cold reference junction. If both junctions are at the exact same temperature (T_meas = T_ref), the generated Seebeck voltage is exactly 0.00 mV, regardless of whether that temperature is 0°C or 1,000°C.

---

4. Cold Junction Compensation (CJC)

In early laboratory applications, the reference junction was physically submerged in an insulated ice-water bath maintained at exactly 0.00°C (32°F). Under these conditions, T_ref = 0°C, and standard NIST thermoelectric conversion tables (which are referenced to 0°C) could be read directly.

In modern industrial facilities, maintaining an ice bath inside a control cabinet or field transmitter head is impossible. Instead, instruments utilize electronic Cold Junction Compensation (CJC):

1. Terminal Temperature Measurement: An accurate secondary temperature sensor (a high-precision thermistor, Pt100 RTD, or semiconductor diode) is embedded directly into the isothermal terminal block where the thermocouple extension conductors land.
2. Voltage Calculation: The transmitter microprocessor reads this secondary sensor to determine the precise reference terminal temperature (T_ref). It references an internal look-up table (NIST ITS-90) to calculate the thermoelectric millivolt potential that the thermocouple would have generated if its measurement junction were at T_ref relative to 0°C (V_CJC).
3. Mathematical Addition: The instrument measures the actual Seebeck voltage across the thermocouple terminals (V_measured) and adds the compensation voltage: V_true(T_meas) = V_measured + V_CJC(T_ref)
4. Lookup Conversion: The microprocessor converts this corrected total millivolt value (V_true) into the final displayed temperature in degrees Celsius.

Field Failure Consequence: If an isothermal terminal block is exposed to strong draft air, radiant heat from adjacent power supplies, or loose terminal screws, thermal gradients develop across the terminal strip. The CJC sensor reads a different temperature than the actual conductor junctions, creating an uncorrectable offset error on a 1:1 basis (1.0°C terminal error = 1.0°C measurement error).

---

5. Industrial Thermocouple Types, Compositions & ANSI/ISA Color Codes

Industrial electricians must memorize the standard thermocouple designations, positive/negative lead compositions, operating atmospheres, temperature limits, and ANSI color coding.

[!IMPORTANT] North American (ANSI / ISA) Rule for Thermocouple Polarity: In North America, the RED insulated conductor is ALWAYS NEGATIVE (-) across all thermocouple types without exception! The color of the positive lead and the outer jacket indicates the specific thermocouple type.

Practical Identification Tip in the Field

• For Type J, the positive Iron lead is strongly magnetic, while Constantan is non-magnetic.
• For Type K, the negative Alumel lead is mildly magnetic, while Chromel is non-magnetic.
• If an electrician suspects reversed polarity, observe the temperature trend: if process temperature rises but the PLC indicates a decreasing temperature, the thermocouple leads are reversed at the terminal block.

---

6. Thermocouple Grade Wire vs. Extension Grade Wire

1. Thermocouple Grade Wire: Manufactured to exact metallurgical specifications and tight chemical tolerances. Designed to withstand direct exposure to the high temperatures and corrosive atmospheres inside the process thermowell. Identified by standard letter designations (e.g., Type J, Type K).
2. Extension Grade Wire: Utilizes conductors made from identical or matched cheaper alloys formulated to produce matching thermoelectric characteristics, but only over a limited ambient temperature range (typically 0°C to 200°C). Designated by the letter "X" appended to the type designation (e.g., JX, KX, TX, EX):
   • Never run standard copper wire between a thermocouple head and a remote transmitter! Connecting copper wire to a thermocouple head creates two new intermediate thermocouple junctions at the head terminals. Unless that connection head has active cold junction compensation, reading errors of tens or hundreds of degrees will occur.
   • Extension wire must be maintained as thermocouple-specific metallurgy all the way from the sensor head to the CJC-compensated input terminals.

---

7. Thermocouple Junction Sheath Configurations

Thermocouple elements are housed within protective stainless steel, Inconel, or ceramic thermowell sheaths packed with compacted magnesium oxide (MgO) powder insulation. The construction of the tip junction dictates performance:

   GROUNDED JUNCTION                   UNGROUNDED JUNCTION                 EXPOSED JUNCTION
   ┌─────────────────────┐             ┌─────────────────────┐             ┌─────────────────────┐
   │ MgO Insulation      │             │ MgO Insulation      │             │ Ceramic Seal        │
   │  ──────────┐        │             │  ──────────┐        │             │  ─────────────────  │
   │            ├─┐      │             │            ├──┐     │             │                  ├──┼─ (Bead exposed
   │            │ ├──[X] │             │            │  │     │             │                  ├──┼─  to fluid)
   │  ──────────┘ │      │             │  ──────────┘  │     │             │  ─────────────────  │
   │   Sheath Tip ┴──────┤             │ Sheath Tip    │     │             │ Sheath Tip          │
   └─────────────────────┘             └───────────────┴─────┘             └─────────────────────┘
   Junction welded to sheath           Junction electrically isolated      Direct fluid contact
   Fast response; vulnerable to noise  Slower response; high noise immunity Fastest; zero protection

1. Grounded Junction: The measurement junction is physically welded directly into the protective metallic sheath tip.
   • Advantages: Excellent heat transfer, fast thermal response time (time constant tau ≈ 0.5 to 1.5 seconds), rugged mechanical vibration resistance.
   • Disadvantages: Susceptible to ground loops and common-mode electrical noise. If the sheath is immersed in a grounded pipe or vessel near high-voltage motor frames, stray ground currents enter the thermocouple circuit, destroying millivolt signals unless galvanic isolators are used.
2. Ungrounded (Isolated) Junction: The measurement junction is fully enclosed within the sheath but completely insulated from the metallic tube by compacted MgO powder (dielectric breakdown rating > 100 MΩ at 500 VDC).
   • Advantages: Complete electrical isolation from pipe ground. Highly immune to electrical noise, VFD carrier frequencies, and ground loop potentials. Mandatory when connecting to non-isolated multi-channel PLC input modules.
   • Disadvantages: Thermal response time is slower (time constant tau ≈ 3 to 8 seconds) due to the thermal resistance of the MgO powder.
3. Exposed Junction: The junction wires extend beyond the protective metallic sheath directly into the process gas or liquid stream.
   • Advantages: Fastest possible thermal response time (time constant tau < 0.1 seconds).
   • Disadvantages: Zero mechanical protection; subject to rapid erosion, oxidation, and chemical corrosion. Restricted to non-corrosive, low-pressure gas flows or engine exhaust diagnostics.

---

8. Head-to-Head Comparison: RTD vs. Thermocouple

---

9. Concrete Industrial Troubleshooting Scenario: Annealing Furnace Temperature Drift

An industrial electrician is called to a steel heat-treatment plant where a continuous annealing furnace operating at 850°C is reporting temperature anomalies. The furnace temperature is measured by a Type K thermocouple connected to an ungrounded head transmitter located 30 metres away in an open cable tray running alongside a 600 V VFD feeder.

   Furnace (850°C) ──[Type K Probe]──> Connection Head ──[Cable Span]──> Transmitter / PLC Rack
   Electrician investigates:
   1. Conductor metallurgy across terminal blocks
   2. Red conductor polarity and cold junction temperature
   3. Sheath isolation against plant structural ground

1. Visual and Instrument Inspection:
   • The electrician inspects the connection head at the furnace wall and finds that a previous maintenance contractor had spliced standard 18 AWG copper wire between the thermocouple connection head and the intermediate junction box.
   • Diagnosis: The copper-to-Chromel and copper-to-Alumel junctions inside the connection head acted as uncompensated intermediate cold junctions. Because the connection head temperature varied between 40°C and 95°C as furnace ventilation fans cycled, the PLC measurement drifted erratically by over 50°C.
2. Corrective Remediation:
   • The electrician replaces the entire copper cable run with genuine Type KX shielded extension cable featuring yellow overall jacketing, yellow positive conductor, and red negative conductor.
   • The shield drain wire is landed at the control cabinet instrument ground bus and floating at the furnace head.
3. Verifying Polarity & Millivolts:
   • Using a Fluke 754 documenting process calibrator, the electrician measures the DC millivolt signal directly at the transmitter terminals: V_measured = 34.92 mV.
   • The calibrator's internal CJC sensor measures an ambient terminal block temperature of 25.0°C (corresponding to V_CJC = 1.00 mV for Type K relative to 0°C).
   • Total corrected voltage: V_true = 34.92 mV + 1.00 mV = 35.92 mV.
   • Consulting NIST Type K tables, 35.92 mV corresponds to exactly 865.0°C. The transmitter display confirms 865.0°C, matching the independent optical pyrometer reading.
4. Megohmmeter Sheath Verification:
   • To ensure no ground-loop noise from adjacent VFDs, the electrician disconnects the thermocouple leads and performs a 250 VDC insulation resistance test between the thermocouple conductors and the outer Inconel sheath. The meter displays > 500 MΩ, confirming the ungrounded junction is intact.