A Coriolis meter that's throwing an error code is not always a broken meter. In many field investigations, the meter is working correctly and the error is telling you something about the process — entrained gas, a drifting zero, an installation stress that changed last week. Distinguishing "the meter is sick" from "the meter is healthy but its job has gotten harder" is the main skill of Coriolis troubleshooting, and it's the main skill this guide is built around.
Coriolis error code systems differ by manufacturer in their naming and numbering, but the underlying failure categories are the same across brands. This guide uses generic category names (like "Drive Gain High" or "Tube Not Oscillating") and cross-references the typical equivalents in Emerson Micro Motion, Endress+Hauser Promass, KROHNE OPTIMASS, and similar lines — so the workflow is portable across whatever meter is actually on the skid.
The structure is built for the way a field technician actually works: a decision tree for fast triage, followed by detailed problem cards for the error classes you'll encounter most. Use the tree first to narrow down which category of error you're looking at, then jump to the card that matches. Each card is self-contained — you don't need to read the whole guide to solve one error.
Triage Principles Before You Touch Anything
The most expensive troubleshooting mistakes happen in the first five minutes — usually as a well-intentioned "let me just reset the meter and see." Four principles reduce the chance of making a diagnosable problem into an undiagnosable one.
Principle 1
Read the error before clearing it
Many error codes are latched — they're telling you that a condition occurred, even if the condition has passed. Clearing the error without logging the code, the timestamp, and the recent process trends destroys information you may need ten minutes later when the same error returns. Always document before you reset.
Principle 2
Correlate with the process, not just the meter
A density drift alarm that appeared at the same moment as a feed composition change is telling you about the process. A drive gain alarm that appeared with no process change is telling you about the meter or its installation. The first question is always "what else changed around this time?" — not "what's wrong with the meter?"
Principle 3
Distinguish continuous from intermittent
A continuous error usually points at a hardware or installation issue (failed pickoff, loose connection, chronic installation stress). An intermittent error usually points at a process condition (transient two-phase flow, pulsation at startup, pump cycling). The fix is different; confusing the two leads to replacing hardware that wasn't the problem.
Principle 4
Check the easy things first, always
Wiring connections at the terminal block, grounding, transmitter power supply, and recent maintenance activity account for a large share of "sensor failure" calls. Five minutes of checking the terminal box beats two hours of advanced diagnostic menus. The hierarchy is: connections → installation → process → sensor → electronics.
Error Code Naming Across Brands
Every Coriolis manufacturer uses a different code format. Emerson Micro Motion uses alphanumeric codes (A003, A104, A131). Endress+Hauser Promass uses S-codes and F-codes (S144, F274). KROHNE OPTIMASS uses four-digit numerics. The underlying error conditions, however, map to a small number of generic categories that are identical across vendors — because the physics being diagnosed is identical.
This guide uses generic category names throughout, because a field technician working on mixed-brand sites is better served by understanding the failure mode than by memorizing one vendor's code list. Each error card in Sections 4–7 cross-references the generic name to the typical equivalents in the major brands. Exact code numbers should always be verified against the specific meter's operating manual — vendor firmware revisions occasionally shift codes between versions.
In a typical field population, process-category errors are by far the most common (60–75% of alarms on most sites), followed by sensor-category errors (15–25%), then signal and electronics (the remainder combined). The proportions justify where this guide spends its detail — sensor and process errors are expanded in depth; signal and electronics are kept compact.
The Field Decision Tree
Start here when an error is active. The tree routes from the top-level symptom ("is the meter reading anything at all?") down to the error category, then to the specific card. If you arrive at a leaf node and the remedy doesn't apply, back up one branch and reconsider the previous answer.
Sensor-Category Errors — Detailed
Sensor-category errors indicate a problem in the physical measurement chain — flow tubes, drive coil, pickoff sensors, or the integrated RTD. These errors are typically continuous rather than intermittent and usually require physical inspection or replacement. They are less common than process errors, but more consequential when they occur.
Tube Not Oscillating
The drive circuit is active but the pickoff sensors are not detecting tube motion within the expected frequency band. Mass flow and density outputs are frozen at zero or held at the last valid value, depending on fail-safe configuration.
1. Tubes empty or partially filled. The meter was taken out of service and the tubes drained; now the drive coil can excite the tubes but the natural frequency is shifted so far outside the expected range that the pickoff detection logic rejects it. Most common cause on a "new error after maintenance" call.
2. Drive coil circuit open or shorted. Check with a multimeter at the transmitter terminals — the drive coil typically reads a few hundred ohms; open circuit or dead-short is a broken coil or a cable fault.
3. Pickoff sensor failure. Less common than drive coil failure. Both pickoffs failing simultaneously is rare; one pickoff failed usually produces a "tube imbalance" error instead, not "not oscillating."
- Confirm the tubes are full. Check the upstream block valve position; verify process line is not drained. If the meter is new and just commissioned, confirm the purge or fill procedure was completed.
- Measure drive coil resistance. Transmitter OFF, disconnect drive leads, measure with a multimeter. Value should match the meter datasheet (typically 100–400 Ω depending on model). Open, short, or wildly off-spec = coil fault.
- Measure pickoff coil resistance. Same procedure on the two pickoff leads. Both should read within ~10% of each other.
- Check cabling continuity. If the meter electronics are remote-mounted, pickoff and drive signals travel through a field cable — inspect connectors and verify continuity at the meter end.
If tubes are empty → refill line and retry. If coil fault confirmed → sensor is not field-repairable; schedule replacement. If cabling fault → replace or reterminate cable.
Tube Imbalance / Sensor Asymmetry
The two pickoff sensors are producing signals of different amplitudes — typically beyond a vendor-defined tolerance of 10–20%. The meter can still compute mass flow but the internal balance that underlies zero stability is compromised.
1. One pickoff degraded. Internal coil aging, slight winding degradation, or connector contact resistance drift. The asymmetry often develops gradually over years.
2. Tube coating (uneven). On heavy hydrocarbons, waxy process fluids, or biofilm-prone services, uneven coating on one tube shifts its effective mass relative to the other.
3. Installation stress. Piping stress that deforms one tube's mounting more than the other — usually appears after recent piping work nearby.
- Record pickoff amplitudes from diagnostics menu. Note the ratio. If ~1:1, error is spurious; if >1.2:1, asymmetry is real.
- Check recent maintenance history. Any piping modifications upstream or downstream in the last 90 days? Recent shutdown with a refill procedure?
- Consider fluid cleaning. If service is known-fouling, a mild flush (per vendor recommendation) may restore balance.
- Measure pickoff coil resistances. If resistances differ significantly, one pickoff has degraded.
Cleaning resolves most coating-induced asymmetry. Installation-stress cases require piping correction. Genuine pickoff degradation is non-field-repairable and requires sensor replacement — though many meters will continue to run with degraded accuracy for extended periods if the asymmetry is stable.
RTD Fault / Temperature Sensor Out of Range
The integrated RTD (used for temperature compensation of mass flow and density) is reading outside its valid range, typically <−50 °C or >+200 °C, or showing an open-circuit or shorted condition. Mass flow and density continue to update but without proper temperature compensation — accuracy degrades.
1. RTD wiring fault. Loose terminal, water ingress in junction box, or field-cable issue on remote-mount installations.
2. RTD element failure. Rare — Pt100 elements are typically very reliable, but mechanical shock or extreme thermal cycling can fail one.
3. Genuine out-of-range process condition. Startup from ambient where line was not yet at operating temperature; or a process upset creating unusually cold or hot conditions.
- Check the reported temperature. If it's exactly the minimum or maximum of range, likely an electrical fault. If it's a plausible process value, likely a configuration or range issue.
- Verify RTD wiring. Transmitter OFF, check wiring continuity and insulation resistance.
- Cross-check with an adjacent temperature measurement. Any nearby TT on the same line that should read approximately the same value?
Wiring fix or RTD replacement (vendor service). Until resolved, document the impact on measurement accuracy and notify process operations — density accuracy is particularly sensitive to RTD loss.
Process-Category Errors — Detailed
Process-category errors are the most frequent class of Coriolis alarms in field operations. They indicate that the measurement conditions have changed — not that the meter has failed. Resolving them usually means diagnosing an upstream process change, not servicing the meter. Getting this right is where field-technician experience pays off most.
Drive Gain High / Drive Saturation
The drive circuit is applying more energy than usual (often >40% of max) to maintain tube oscillation. The tubes are being damped by something that wasn't there during calibration — almost always entrained gas, particulates, or a tube-coating change.
1. Two-phase flow — entrained gas in liquid or liquid slugs in gas. By far the most common cause. Can be transient (startup, pump priming) or continuous (degassing from dissolved gas, flashing on pressure drops).
2. Solids content. Slurry service, particulates in produced fluids, or crystallization on a cooling line. Tubes are fine but the fluid is more lossy than calibration.
3. Tube damage. Erosion, cracking, or a stuck valve near the sensor. Usually associated with a step change in drive gain, not a gradual drift.
- Check density reading. If density is wildly different from expected (particularly: dropping below fluid density on liquid service) → entrained gas is confirmed, skip to (4).
- Check process stability. Is the drive gain transient (spiking at pump starts, settling) or continuous? Transient = operational; continuous = sustained condition.
- Check upstream pressure. If pressure dropped recently, dissolved gas may be flashing out. Common cause on cryogenic or hydrocarbon service.
- Confirm two-phase with a density trend. Density jitter or a sawtooth pattern over time = two-phase flow.
Transient two-phase: usually operational — tolerate during startup, ensure meter is after the pump with adequate back-pressure for steady-state. Continuous two-phase: add a degasser upstream, raise line pressure, or accept accuracy limitations during the affected conditions. Slurry/particulates: meter may be correctly operating — verify accuracy spec covers this service. Tube damage: replace.
Zero Stability Error / Zero Drift
At no-flow conditions, the meter is reading a non-zero value. The zero offset has drifted from its last calibrated value, or the meter is showing a zero value during no-flow that doesn't match the stored zero point.
1. Zero was calibrated under different conditions than current service. Zero must be established at the actual operating temperature, pressure, and full-flooded state — not during commissioning at ambient with an empty line.
2. Installation stress changed. Piping movement from thermal expansion or recent work shifted stress on the meter body. This can cause a persistent apparent zero offset that a new zero calibration will correct.
3. Real flow during the "zero" check. Block valves not fully sealed, a drain or bypass leaking, or a slow backflow — the meter is correctly reading a small flow.
- Confirm genuinely no-flow conditions. Both upstream and downstream block valves closed and confirmed tight. No bypasses, no drains open.
- Check if tubes are full. Empty or partially-filled tubes produce false zero readings.
- Check temperature stability. Zero calibration is only valid at the temperature it was performed at; major temperature shifts require re-zeroing.
- Perform a new zero calibration per vendor procedure. Only after the above are confirmed.
In nearly all cases: re-zero under correct conditions (full, stable, no-flow, at operating temperature). If re-zero doesn't hold, investigate installation stress — that points to a piping support or alignment issue rather than a meter issue.
Density Out of Range / Unusual Density
The measured density is outside the expected range configured in the transmitter. The meter is not failing — it's producing a reading that doesn't match configured expectations. Often a service composition change, sometimes a two-phase indicator.
1. Actual composition change. Different grade of crude, concentration shift in a chemical recipe, or a product changeover not reflected in transmitter configuration.
2. Two-phase flow pulling density down. Entrained gas lowers apparent density below the pure-liquid value; usually accompanies a drive gain alarm.
3. Temperature compensation error. If RTD is reading wrong (see RTD fault card), density compensation will be wrong too.
Compare density reading to the expected fluid density at operating temperature. If materially lower than pure-fluid value, suspect two-phase. If materially different but stable, suspect composition change. If tracking temperature incorrectly, suspect RTD. A lab sample can confirm composition.
Update transmitter density range configuration if the new composition is operational; address two-phase per the drive gain card; repair RTD if temperature-compensation fault.
Flow Rate Out of Range / Saturation
Flow rate has exceeded the configured upper limit (saturation) or has fallen below the lower cutoff. The meter is still measuring, but the output channel may be clamped or flagged as out-of-spec.
Usually a sizing or configuration issue rather than a meter fault. Verify the meter is correctly sized for the operating range, update the configuration if the process has shifted operating point, or flag the event to process operations if it indicates a genuine upset.
Signal-Category Errors
Signal-category errors indicate problems in the output chain between the transmitter and the receiving system — 4–20 mA loops, HART digital communication, Modbus, or Foundation Fieldbus. The measurement itself may be healthy; only the transport is failing. These errors resolve at the wiring or receiving-system level, not at the meter electronics.
4–20 mA Loop Fault / Output Saturation
The 4–20 mA output cannot drive its expected current — either a loop break (open circuit) or saturation at 20+ mA (short or too low resistance). The transmitter is trying to send a signal that the wiring will not accept.
Verify loop resistance (typically 250–600 Ω depending on installation), check continuity of the 2-wire cable, verify the receiving device (PLC analog input, DCS card) is powered and configured. Loop breaks often trace to loose screw terminals after maintenance activity.
Correct the wiring fault, replace a damaged cable segment, or reconfigure the receiving device. The meter itself usually does not require attention.
Digital Communication Loss / HART or Modbus Fault
The DCS or host system has lost digital communication with the transmitter. The meter may still be reading correctly locally, and the 4–20 mA signal may still be working — only the digital overlay (HART) or replacement (Modbus, Fieldbus) has stopped.
For HART: verify loop resistance is within HART tolerance (250 Ω minimum), check for nearby EMI sources, confirm handheld can connect locally. For Modbus/Fieldbus: verify address, baud rate, termination resistors, and segment topology.
Communication issues rarely trace to meter hardware — look at wiring, termination, host-system configuration, and segment topology first. Field DTM / EDDL software updates on the host side are a common culprit after system upgrades.
Electronics-Category Errors
Electronics-category errors indicate a fault in the transmitter itself — memory corruption, internal voltage regulation, or firmware integrity. These are uncommon in field populations (well under 5% of total alarms on most sites) and are rarely user-repairable. The response is typically escalation to the manufacturer, not field repair.
Electronics Fault / Memory or Voltage Error
The transmitter's self-diagnostics have detected an internal fault — bad checksum on configuration memory, internal voltage out of range, or failed startup self-test. Measurement outputs may be frozen, held at failsafe, or absent.
Verify supply voltage is within meter specification (typically 18–30 VDC for 4-wire meters); check for water ingress in the transmitter housing; check for recent power events (lightning, surge, interrupted commissioning).
Power-cycle may clear transient faults. Persistent electronics faults should be escalated to the vendor — these are not user-serviceable. A transmitter swap (with configuration restore from backup) is usually the only field-applicable remedy.
Configuration Error / Corrupt Configuration
The transmitter's configuration data (meter constants, calibration factors, fluid properties) has failed a self-integrity check, or a configuration write was interrupted. Often appears after a firmware upgrade or after recent HMI activity.
Restore configuration from the vendor's backup file (every properly commissioned meter should have a configuration backup). If no backup exists, the meter requires re-commissioning with original calibration data obtained from the vendor — this is why backups matter.
Quick-Reference Error Table
Print this page and keep it near the HMI. A 30-second scan should point to the right detailed card in sections 4–7.
| Generic Error Name | Category | Severity | First Check | Likely Root |
|---|---|---|---|---|
| Tube Not Oscillating | Sensor | Critical | Tubes full? | Empty line / coil fault |
| Tube Imbalance | Sensor | Warning | Pickoff amplitudes | Coating / stress / aging |
| RTD Fault | Sensor | Warning | Temperature reading value | Wiring / RTD failure |
| Drive Gain High | Process | Warning | Density reading stable? | Two-phase flow |
| Zero Drift | Process | Warning | Genuinely no-flow? | Zero calibration stale |
| Density Out of Range | Process | Info | Known composition? | Composition or 2-phase |
| Flow Saturation | Process | Info | Sizing vs actual | Config or process upset |
| 4–20 mA Loop Fault | Signal | Warning | Loop continuity | Wiring issue |
| HART/Modbus Comm Loss | Signal | Info | Local vs host | Host or segment issue |
| Electronics Fault | Electronics | Critical | Supply voltage / moisture | Transmitter hardware |
| Configuration Error | Electronics | Warning | Recent HMI activity? | Interrupted write |
Prevention & Routine Checks
The errors above are easier to prevent than to fix. A short monthly routine and a longer annual routine catch the majority of developing issues before they become alarms.
Monthly — 10 minutes per meter
Record drive gain, pickoff amplitudes, density at steady flow, and zero reading. Compare to previous months. Trending beats absolute values — a drive gain rising from 12% to 22% over six months is a story, even if 22% is still below the alarm threshold.
Quarterly — density cross-check
Pull a lab sample from a port near the meter. Compare lab density (temperature-corrected) to meter density reading. A gap >1 kg/m³ developing over time often indicates tube coating — address before the drive gain alarm activates.
Annually — zero calibration under correct conditions
Re-zero the meter under process temperature and pressure, with tubes confirmed full and flow confirmed to be zero. Document the zero value. A zero that drifts more than the vendor spec year-over-year indicates an installation issue developing.
Always — keep a configuration backup
Every commissioning, every firmware upgrade, every significant configuration change — back up the transmitter configuration and label it with date and reason. The "Configuration Error" alarm card depends on this: a backup turns a major problem into a 10-minute restore.
Escalation Path — When to Stop DIY
Not every error is a field-fix. Knowing when to stop troubleshooting and call support saves time and avoids making things worse. Three rules bound the DIY zone:
Rule 1
Stop after two verified replacements
If you've already replaced the transmitter or the sensor and the same error persists, the problem is not where you've been looking. Further replacement attempts are expensive guesses. Call vendor support with your diagnostic data.
Rule 2
Stop before opening a sealed sensor
Coriolis flow tubes and their housings are sealed at the factory. Opening them in the field voids warranty, breaks the hermetic protection against moisture and dust, and — on hazardous-service meters — may invalidate explosion-protection certification. If a sensor needs to be opened, it's going back to the vendor.
Rule 3
Stop if a safety-function meter is involved
Meters that are part of a SIF (safety instrumented function) or an overpressure protection layer have specific qualification requirements. Any repair or recalibration needs to follow the site's management-of-change procedure, and the work may need to be vendor-performed to preserve the SIL certification.
Supmea Product Fit
Supmea's Coriolis mass flow meter range implements the diagnostic categories described in this guide — sensor, signal, process, and electronics — with error codes that map to the generic naming used throughout. The transmitter provides drive gain, pickoff amplitudes, density, and temperature as diagnostic values accessible from both the local HMI and the HART/Modbus interface, enabling the monthly-and-quarterly verification routine from Section 9 without requiring specialized software.
For sites running mixed-brand Coriolis populations, Supmea meters fit into the same workflow — field technicians trained on Emerson or Endress meters can apply the same triage sequence and decision tree to Supmea installations. Full product specifications, including complete error-code references, are available on the Supmea product site. For projects specifying new Coriolis installations, the Supmea application team can review diagnostic requirements, alarm configurations, and maintenance workflow before commissioning — so the meters arrive aligned with the site's existing troubleshooting practice.
For background on the underlying principles referenced in this guide, external references on mass flow meters and the HART protocol are useful starting points.
Still Seeing the Error After This Guide?
Share the error code, the meter model, the measurement context, and what you've already checked. Our application team helps narrow down whether you're looking at a meter issue, an installation issue, or a process change — and what the next concrete step should be.
Consult Supmea →