A Coriolis mass flowmeter (CMF) directly measures mass flow, density, and temperature in one device by detecting the twist of a vibrating tube driven by the Coriolis force on the moving fluid, the Coriolis force being proportional to mass flow [S1][S2].
Because the measurement is fundamentally mass-based rather than volumetric, the same instrument can serve custody-transfer, batching, and concentration control, provided the process fluid stays within the meter's density, viscosity, and gas-fraction envelope [S1].
Operating Principle and What the Sensor Actually Returns
The CMF sensor is a flow tube (or tubes) driven to oscillate at its natural frequency; fluid moving through the tube induces a Coriolis force that twists the tube in proportion to mass flow, with the differential phase between two pick-offs giving the flow signal and the resonant frequency giving density [S1][S2]. Standard accuracy on commercial CMFs is typically ±0.1% to ±0.5% of rate for liquids, with density resolution in the 0.001 to 0.01 g/cm³ range and turndowns of 100:1 or higher, and the same instrument returns temperature from the tube drive [S1].
For a broader look at where CMF sits against volumetric alternatives, see the flow meter selection criteria map for 2026 and the Coriolis flowmeter reference page.
Fluid-Property Gates: Density, Viscosity, and Gas Fraction
CMF is unsuitable for low-density media and low-pressure gas; CMF cannot measure low-density fluids or low-pressure gas, and entrained gas above a model-dependent limit will significantly disturb the reading [S1]. The general rule is to keep gas fraction below roughly 1% to 2% by volume for liquid duty and to use dedicated gas CMF bodies above approximately 0.01 g/cm³ working density, with the manufacturer confirming the exact threshold for a given tube geometry [S1].
Viscosity is largely irrelevant to the mass-flow output, but high-viscosity fluids (above roughly 1000 mPa·s) increase pressure drop and may push a small-bore sensor off its Reynolds-window sweet spot, so the engineer should size the tube ID against viscosity, not just flow [S1]. For water-treatment skids where fluids stay clean and within moderate viscosity, a complementary instrument like the vacuum gauge selection criteria for water treatment skids often sits alongside the CMF in the same panel.
Sizing, Pressure Drop, and Tube Geometry
Pressure drop through a CMF is the kinetic loss of redirecting flow through a small-bore, often curved tube, and at design flow it typically lands between 0.3 bar and 2 bar depending on tube ID, flow rate, and density; undersizing to chase accuracy on a low-flow stream is the single most common cause of pump cavitation downstream [S1]. For abrasive coal-oil slurry service, the source recommends a straight single-tube CMF mounted with the measuring tube vertical, so wear is distributed evenly and tube life is extended [S1].
Common tube geometries are single straight, dual straight, single bent (U), and dual bent; bent tubes shorten the installation face-to-face but raise pressure drop, while straight tubes are preferred for slurry, two-phase, and hygienic drainable service [S1]. A side-by-side sizing rule of thumb: match nominal line size where possible, and step down only one size for low-flow batching, since going two sizes down typically doubles or triples pressure drop without proportional accuracy gain.
Installation Constraints: Mounting, Vibration, and Zero Stability
CMF is sensitive to external vibration and pipe strain; the sensor mounting typically requires independent supports and, for parallel sensors on the same line, opposite-orientation mounting with vibration-isolation pads to keep their resonance peaks separated [S1]. No upstream or downstream straight pipe run is required for measurement integrity, but the control valve should be placed downstream of the CMF to keep static pressure high and suppress flashing or cavitation inside the tube [S1].
Zero stability is the most common field complaint: CMF zero drift originates from mechanical asymmetry in the sensor and from fluid density and viscosity changes, so zeroing must be done on site with the sensor fully filled, vented, and at process temperature, with upstream and downstream block valves proven leak-tight [S1]. Any change in piping support or a large process temperature swing mandates a re-zero; for plants that route CMF to a DCS, the related pressure transmitter selection criteria for compressed air lines covers the same zero-and-span discipline applied to pressure loops.
Comparison: CMF Against Differential Pressure, Magnetic, Ultrasonic, and Vortex
Against the four common alternatives in liquid service, CMF scores on accuracy, multivariable output (mass + density + temperature), and freedom from straight-pipe runs, but loses on price per line size and on low-pressure gas and slurry service limits [S1]. An electromagnetic flowmeter is cheaper on conductive clean liquids and has near-zero pressure drop, but cannot measure hydrocarbons, non-conductive solvents, or slurries where the electrodes would coat.
An ultrasonic flowmeter, including clamp-on, wins on retrofit and bidirectional service but assumes a clean, degassed, single-phase liquid and rarely beats ±1% in field conditions. A vortex flowmeter handles steam and saturated gas at a fraction of the cost but has a fixed minimum Reynolds and cannot measure low flows below its threshold velocity.
A turbine flowmeter reaches ±0.15% on clean, low-viscosity hydrocarbons at a lower purchase price than CMF but needs 10D upstream and 5D downstream straight run, suffers bearing wear in dirty service, and has no density output. CMF is the right pick when any of mass-flow accuracy, inline density, multi-product batching, or absence of straight-pipe run matters more than first cost.
Who Should Not Pick a Coriolis
CMF is the wrong tool on low-pressure gas distribution below the density limit, on two-phase flashing lines where the gas fraction regularly exceeds the manufacturer's threshold, and on very large line sizes (above roughly 6-inch to 8-inch) where the tube pressure drop becomes prohibitive [S1]. It is also the wrong tool for short lead-time skid builders on a tight budget where ±1% to ±2% is acceptable; in those cases an electromagnetic flowmeter on conductive water or a turbine flowmeter on clean hydrocarbon is more cost-effective [S1].
For saturated steam or high-temperature gas, a vortex flowmeter typically gives better value, while a flow meter reference page walks through the cross-technology decision in more detail.
Field Maintenance and Lifecycle Signals
For abrasive service such as coal-oil slurry, the source flags three failure modes worth tracking: progressive tube-wall thinning from the slurry changes tube stiffness and biases the flow reading, internal fouling or deposit drift on the tube wall shifts the density calibration, and entrained gas above the model limit will bias mass flow high or throw the sensor into a fault state [S1]. Trackable signals are, first, the density-baseline drift at a known reference fluid (e.g., water at 20°C) on a six-month cadence, second, the zero-shift trend logged during scheduled block-valve shut-ins, and third, the drive gain on the transmitter, which rises as tube stiffness changes from thinning or coating [S1].
For instrument-level temperature and density cross-checks, the temperature transmitter selection guide for RTD, TC, and HART tradeoffs is a useful complement, since the CMF's own temperature output is a tube-temperature proxy, not a calibrated thermowell reading.