Common Temperature Transmitter Problems and Solutions

Temperature transmitters are among the most reliable instruments on a plant floor — until they aren’t. When a reading goes wrong, the cause is almost always one of a handful of well-understood issues. This guide covers the ten problems instrumentation technicians encounter most often, with specific diagnostic steps and fixes for each.

1. Erratic or Fluctuating Readings

Symptoms: The displayed temperature jumps around randomly, or oscillates by several degrees when the process is stable.

Common causes:

• Electromagnetic interference (EMI) from nearby VFD cables, motor starters, or radio transmitters coupled into the sensor input lines
• Ground loops — two or more ground reference points at different potentials creating circulating current through the signal wiring
• Loose terminal connections — vibration or thermal cycling has loosened a screw terminal

How to diagnose:

1. Disconnect the sensor at the transmitter terminals. Connect a precision resistor (e.g., 109.73Ω for Pt100 at 25°C) directly to the input. If the reading stabilizes, the problem is in the field wiring or sensor — not the transmitter.
2. Measure AC voltage between the transmitter housing and the sensor sheath. Anything above a few millivolts indicates a ground loop.
3. Check that shield wires are grounded at one end only (typically the DCS cabinet, not the field junction box).

Fix:

• Route sensor input cables away from power cables — minimum 300mm separation, or use grounded steel conduit.
• Ground the cable shield at one point only.
• If ground loops persist, use a transmitter with higher electrical isolation. Transmitters with 2kV AC isolation (e.g., Microcyber TT306) provide significantly better ground loop rejection than those rated at 620V or 1kV.

2. Reading Drifts Over Time

Symptoms: The temperature reading slowly shifts by a fraction of a degree per month, even though the actual process temperature has not changed.

Common causes:

• Ambient temperature variation affecting the transmitter’s analog output circuitry — the “temperature drift” specification tells you how much
• Sensor aging — RTD elements drift over years of thermal cycling; thermocouple drift is even more pronounced, especially Types K and N above 800°C
• Moisture ingress into the transmitter housing or sensor connection head

How to diagnose:

1. Compare the transmitter’s reading against a reference thermometer at a known, stable temperature point.
2. Check whether the drift correlates with ambient temperature changes (day/night cycle, seasonal). If it does, the transmitter’s temperature drift spec is the likely contributor.

Fix:

Select a transmitter with low temperature drift. This is specified as “analog output drift per 1°C ambient change.” For example:

Transmitter	Analog Output Drift per 1°C
Microcyber TT306	0.001% of span
Microcyber TT106	0.003% of span
Rosemount 644	0.001% of span
E+H TMT82	0.001% of span

Over a 30°C ambient swing, a 0.003%/°C transmitter adds 0.09% error to the output. On a 200°C span, that’s 0.18°C of drift — enough to matter in tight-tolerance processes.

Establish a calibration interval based on your accuracy requirements. Most plants calibrate annually; critical loops may need semi-annual checks.

3. Persistent Temperature Offset

Symptoms: The reading is consistently higher or lower than the actual temperature, by a fixed amount.

Common causes:

• Lead wire resistance on 2-wire RTD connections — each ohm of lead resistance adds roughly 2.6°C error on a Pt100 sensor
• Wrong sensor type configured — the transmitter is set for Type K but a Type J thermocouple is connected
• Cold junction compensation (CJC) error — the transmitter’s reference temperature sensor reads incorrectly

How to diagnose:

1. Note the offset. If it’s a round number (e.g., exactly +3°C), suspect lead wire resistance or wrong sensor type.
2. For RTD: measure the resistance of each lead wire with a multimeter. On a Pt100, 1Ω of lead resistance ≈ 2.6°C error.
3. For thermocouple: verify the configured sensor type matches the actual sensor. Then check the CJC reading — many HART transmitters report their CJC temperature as a diagnostic parameter.

Fix:

• Use 3-wire or 4-wire RTD connections. A 3-wire connection compensates for lead resistance automatically; 4-wire eliminates it entirely. This single change fixes the majority of offset problems on RTD installations.
• Verify the sensor type setting in the transmitter configuration.
• For thermocouple applications, select a transmitter with a high-accuracy CJC sensor. Transmitters using an internal PT1000 reference (e.g., Microcyber TT306) achieve ±0.5°C CJC accuracy versus ±1°C on budget models — that difference alone can account for 0.5°C of persistent offset.

4. 4-20mA Output Saturated (≥21mA or ≤3.8mA)

Symptoms: The output is stuck at a value above 21mA or below 3.8mA, and the DCS reports a sensor or loop fault.

Common causes:

• Sensor open circuit — broken wire or failed sensor element drives the output to the burnout direction (upscale = >21mA, or downscale = <3.8mA)
• Sensor short circuit — on RTD inputs, a shorted sensor drives the reading (and output) to the bottom of the range
• Insufficient loop power — the supply voltage cannot drive 20mA through the total loop resistance

How to diagnose:

1. Read the HART diagnostic register — it will indicate whether the fault is an open circuit or short circuit on the sensor input.
2. Measure the DC voltage at the transmitter’s power terminals while the loop is active. If the voltage drops below the transmitter’s minimum operating spec (e.g., below 10.0VDC for a TT306 in SIL mode), the loop does not have enough headroom.
3. Calculate: Vsupply − (20mA × Rtotal) ≥ Vtransmitter_min. If this doesn’t hold, the loop is underpowered.

Fix:

• Replace the failed sensor or repair the broken wire.
• If the loop is underpowered, either increase the supply voltage or reduce the total loop resistance (shorter cable run, or a DCS input card with lower input impedance).
• Configure the burnout direction to match your plant’s failure mode requirements — upscale burnout drives the output above 21mA (safe direction for most heating applications); downscale drives it below 3.8mA.

5. HART Communication Not Working

Symptoms: The 4-20mA analog output works fine, but the HART handheld communicator or HART modem cannot establish a session with the transmitter.

Common causes:

• Missing 250Ω loop resistor — HART communication requires a minimum loop resistance of 230–250Ω. Many modern DCS input cards have built-in 250Ω resistors, but some do not.
• Excessive loop resistance — if the total loop resistance is too high, the HART signal is attenuated
• HART modem connected on the wrong side of the power supply — the modem must be on the transmitter side, not across the power supply
• Transmitter HART revision mismatch — older handhelds may not support HART 7 devices

How to diagnose:

1. Measure the total loop resistance (DCS input impedance + any external resistors). It must be between 230Ω and 1100Ω for reliable HART communication.
2. Verify the HART modem or handheld is connected in parallel with the transmitter or the loop resistor — not in series.
3. Check the transmitter’s HART revision. Most modern transmitters (including Microcyber’s TT106/306/305) use HART 7, which is backward-compatible but requires a HART 7-capable handheld for full functionality.

Fix:

• If no 250Ω resistor exists in the loop, add one at the DCS cabinet.
• Move the HART connection point to the correct location — across the transmitter terminals or across the loop resistor.
• Upgrade the handheld communicator firmware if it cannot detect HART 7 devices.

6. Reading Does Not Change When Process Temperature Changes

Symptoms: The output stays at a fixed value regardless of actual process temperature changes.

Common causes:

• Transmitter is in fixed output (test) mode — someone put the transmitter into loop test or fixed output mode and forgot to restore it
• Sensor wiring cross-connected — the transmitter is reading a different sensor than intended
• DCS channel configured incorrectly — the DCS is reading the wrong input channel

How to diagnose:

1. Use a HART communicator to read the transmitter’s “PV” (Primary Variable). If PV changes with process temperature but the analog output doesn’t, the transmitter is in fixed output mode.
2. If PV also doesn’t change, disconnect the sensor and apply a simulated input (precision resistor for RTD, mV source for thermocouple). If the reading responds, the sensor or field wiring is the issue.

Fix:

• Restore the transmitter to normal mode via HART communicator.
• Trace the field wiring to verify it matches the expected measurement point.
• Verify the DCS channel assignment matches the physical wiring.

7. Cold Junction Compensation Error (Thermocouple Only)

Symptoms: Thermocouple readings are off by 1–3°C, but the error is not constant — it changes with ambient temperature at the transmitter terminal block.

Common causes:

• Poor CJC sensor placement — the reference temperature sensor is not in thermal equilibrium with the thermocouple terminals
• Thermal gradients across the transmitter housing — sun exposure, nearby heat sources, or forced-air cooling creates temperature differences between the terminals and the CJC sensor
• Transmitter CJC specification is too loose — ±1°C CJC accuracy introduces a fixed 1°C uncertainty</li
  …(truncated)…

8. Vibration-Induced Failures

Symptoms: Intermittent signal dropouts, loose wire connections, or premature sensor failure — especially on compressors, pumps, or rotating equipment.

Common causes:

• Mechanical fatigue in the sensor element or internal solder joints
• Terminal screws loosening under sustained vibration
• Sensor sheath cracks from resonance at specific vibration frequencies

How to diagnose:

1. Check whether the failures correlate with equipment running versus stopped.
2. Inspect terminal block connections for signs of arcing or discoloration.
3. Review the vibration environment against the transmitter’s published specs.

Fix:

Select a transmitter rated for the actual vibration environment. Not all transmitters are equal:

Transmitter	Vibration Rating
Microcyber TT106 / TT306	10–60Hz: 1.6mm, 25–100Hz: 5g
Rosemount 644	60–1000Hz: 5g, 10–60Hz: 0.35mm
E+H TMT82	2–100Hz: 4g

Note: The Microcyber and Rosemount specs cover different frequency ranges. If your vibration is in the 10–60Hz range (typical for reciprocating equipment), the 1.6mm displacement spec is the relevant one.

• Use spring-loaded terminals instead of screw terminals where available.
• Mount the transmitter on a vibration-dampening bracket rather than directly on the equipment skid.

9. EMC Interference from VFDs and Switchgear

Symptoms: The reading becomes noisy or spikes when a nearby VFD starts, or when switchgear operates. The problem may be intermittent and hard to reproduce.

Common causes:

• Conducted emissions from VFDs coupling into the 4-20mA loop through shared power supplies or ground paths
• Radiated emissions from VFD output cables coupling into the sensor input wiring
• Inadequate EMC design in the transmitter itself

How to diagnose:

1. Observe whether the noise correlates with VFD operation. If the reading is clean when the VFD is off but noisy when it runs, EMC is the likely cause.
2. Use a portable spectrum analyzer or oscilloscope on the 4-20mA loop to characterize the noise frequency and amplitude.

Fix:

• Route sensor cables and 4-20mA loop cables at least 300mm away from VFD output cables. Cross at 90° if routing together is unavoidable.
• Use shielded, twisted-pair cable for both sensor inputs and the 4-20mA loop. Ground the shield at one end only.
• Select a transmitter with current EMC compliance. IEC 61326-1:2021 is the current standard; compliance with NE21 (NAMUR) provides additional assurance for process industry environments.

10. Loop Power Calculation Error

Symptoms: The transmitter works at 4mA but drops out or faults near 20mA, especially on long cable runs.

Common cause: The available voltage at the transmitter terminals drops below its minimum operating voltage at 20mA output.

Calculation:

V_available = V_supply − (I_max × R_total)

Where:
 V_supply = Power supply output voltage (VDC)
 I_max = Maximum output current = 0.021A (including burnout)
 R_total = Cable resistance + DCS input impedance + any loop resistors

V_available must be ≥ V_transmitter_minimum

Example: A TT306 in SIL/intrinsically safe mode requires minimum 10.0VDC at its terminals.

24V supply − (0.021A × (250Ω DCS input + 100Ω cable)) = 24 − 7.35 = 16.65V ✅ OK

24V supply − (0.021A × (250Ω DCS input + 250Ω HART resistor + 200Ω cable)) = 24 − 14.7 = 9.3V ❌ Below 10.0V minimum

Fix:

• Increase the supply voltage (e.g., from 24V to 36V).
• Use a DCS input card with lower input impedance.
• Reduce cable length or use larger-gauge wire.
• Remove unnecessary external resistors.

Quick Diagnostic Checklist

When a temperature transmitter reading is wrong, work through this list in order:

Step	Check	Tool
1	Is the transmitter in normal mode (not test/fixed output)?	HART communicator
2	Does a simulated input at the terminals give the correct reading?	Decade box / calibrator
3	Is the sensor type configured correctly?	HART communicator
4	Is the wiring scheme correct (2/3/4-wire for RTD)?	Visual inspection
5	Is the loop voltage adequate at 20mA?	Multimeter at terminals
6	Is there a HART 250Ω resistor in the loop?	Multimeter (resistance)
7	Is the CJC reading accurate?	HART diagnostic + contact thermometer
8	Are terminals tight and corrosion-free?	Visual + torque driver
9	Is the cable routing away from noise sources?	Visual inspection
10	Does the transmitter meet the vibration and EMC specs for the location?	Check spec sheet vs. environment

Steps 1–5 resolve roughly 80% of field problems. Steps 6–10 cover the remaining edge cases that take longer to track down.