How Do Ultrasonic Flow Meters Work? The Transit-Time Principle Explained

Ultrasonic meters come in two mounting configurations, and this is the decision most often made for you by site conditions rather than preference.

Clamp-On Mounts outside the pipe. Never contacts the fluid.

Inline Installed into the pipeline. Transducers contact the fluid.

The practical question is not which is better in the abstract. It is what your site allows.

When clamp-on is the right answer

• The line cannot be shut down. Clamp-on installation needs no cut, no weld, and no depressurisation. If a shutdown window is expensive, disruptive, or unavailable, this decides it.
• The pipework already exists. Retrofit and audit work is clamp-on's core case — the pipe is in place and in service, and the measurement has to adapt to it.
• The fluid is aggressive, ultrapure, or contamination-sensitive. No wetted parts means no corrosion path, no seal degradation, and no contamination risk from the instrument itself.
• You need to measure at more than one point. A clamp-on meter can be relocated, so one unit can survey several points in sequence — useful for leak investigation, system balancing, and verifying meters that are already installed.
• Maintenance access is limited. No moving parts and no wetted components means very little to service.

When inline still makes sense

• New construction. If the pipe is not installed yet, the cost argument for avoiding a cut disappears.
• You need the tightest accuracy achievable. Inline removes the pipe wall from the signal path, and with it a variable.
• Pipe conditions defeat clamp-on. Heavy internal scaling, unbonded liners, or unusual wall construction can block or distort acoustic signal from outside.
• Pipe size falls outside available clamp-on ranges.

Once mounting type is settled, pipe compatibility is the fastest way to eliminate options — a meter either supports your pipe or it does not. Three attributes matter.

• Outer diameter. Clamp-on meters are rated for a specific OD range, because transducer spacing and signal geometry depend on it. This is a hard limit, not a preference. Note that meters are specified by actual OD in millimetres rather than nominal size, and the two do not always correspond.
• Wall thickness. The signal crosses the wall twice. Thickness affects attenuation and the timing calculation, so it has to be known accurately.
• Material. Metal and rigid plastics transmit acoustic signal well. The problem cases are pipes with unbonded internal liners, heavy internal scale or rust, and some composite constructions — in each case the issue is an air gap or discontinuity that blocks signal.

One practical note: because wall thickness and material feed directly into the flow calculation, entering them incorrectly produces a reading that looks plausible but is wrong. It is worth checking whether a meter requires manual entry of these values, guides you through selecting them, or determines them itself.

Pipe coverage across LORRIC's clamp-on range

Two things worth drawing out of that table beyond the numbers.

The ranges overlap deliberately. Where two series cover the same diameter — ½″ to 1″, or 2″ to 3″ — pipe size is no longer the deciding factor. Space, material, display access, and output requirements are.

Material lists differ by size class, and that is not arbitrary. The small-bore meter supports fluoropolymers (PFA, PTFE) because that is what high-purity and aggressive-chemical lines are built from. The large-diameter meter covers cast iron and carbon steel because that is what mains are built from. Each list reflects what that size of pipe is actually used for.

For how transducer path configuration relates to pipe size and signal strength, see How to Install an Ultrasonic Flow Meter — Z, V, N, W Methods.

These are two different measurement principles sold under the same "ultrasonic" label, and they suit opposite fluid conditions. Getting this wrong is the most consequential specification error in the process, because the meter will not work properly rather than simply work less well.

Transit-time measures the difference in travel time between acoustic pulses sent upstream and downstream. It needs a clear acoustic path through the fluid. Doppler measures the frequency shift of signal reflected off particles or bubbles suspended in the fluid. It needs something in the fluid to reflect from.

If you could see through the fluid in a glass, transit-time. If you could not, Doppler.

Doppler is sometimes the correct answer, not the compromise. If your fluid genuinely carries solids, transit-time is not the right technology at any price point, and specifying it will produce an unreliable installation.

Borderline fluids need checking, not assuming. Light, occasional aeration is usually tolerable for transit-time; sustained entrained air is not. Where a fluid sits near the boundary, signal-quality diagnostics at the intended mounting point tell you more than the datasheet does.

Two specifications that are frequently conflated, but which answer different questions.

Accuracy: read the basis, not just the number

An accuracy figure means little without knowing what it is a percentage of.

• Percent of reading (% RD, sometimes written % O.R.). Error scales with the measured value, so at low flow the absolute error is small.
• Percent of full scale (% F.S.). Error is a fixed fraction of the meter's maximum, so at low flow that same absolute error becomes a much larger share of your actual reading.

For applications that spend most of their time at the low end of the range, this distinction matters more than the headline number does.

Also check the range over which the stated accuracy applies. Most meters specify accuracy across a defined portion of their span, not the whole of it — and a better datasheet states a separate figure for the bottom of the range instead of leaving it out. LORRIC's clamp-on meters specify both: ±3.0 % of reading from 10 % to 100 % of full scale, and ±0.3 % F.S. from 0 % to 10 % of full scale.^[2]

What that looks like in practice

Take a DN25 (1″) FU-ES, with a specified flow range of 3 – 240 LPM. Ten percent of full scale is 24 LPM, so that is where the two accuracy bands meet.

The point of the second row is that below 10 % of full scale the error stops scaling with the reading and becomes a small fixed quantity instead. That is the behaviour you want at low flow, and it is exactly what a single headline accuracy figure hides.

Flow range: cover the extremes, not the average

The meter's measurable range has to cover both the minimum and maximum flow the line will actually see — including start-up, low-demand periods, and peak conditions, not just normal operating flow.

What to check is the velocity range in the specification, and note that datasheets often quote two figures. The FU-LT and FU-ES both specify a recommended range of ±0.3 – 6.0 m/s alongside a wider measurable range; the FU-TX 310 specifies ±0.1 – 20 m/s.^[3] The recommended figure is the one to design around.

Converting your expected flow to velocity for your pipe diameter is the step most often skipped. A flow rate that looks comfortably mid-range in litres per minute can fall below the minimum measurable velocity in a large pipe, or exceed the maximum in a small one. Our flow rate calculation guide covers that conversion.

The fluid and pipe determine whether a meter can measure your application. These three determine whether it will survive and be useful where you are putting it.

Temperature — check two figures, not one

Ultrasonic meter datasheets specify two separate temperature limits, and confusing them causes specification errors.

Fluid temperature is the limit for the medium in the pipe: 0 – 85 °C for the FU-LT and FU-ES, and 0 – 80 °C for the FU-TX 310 with a standard probe, extending to 0 – 150 °C with the heat-resistant probe option. Note that the clamp-on meters also specify no ice on the pipe surface — relevant for chilled and low-temperature lines.

Ambient temperature is the limit for the air around the electronics: −10 to 60 °C across all three, non-freezing, at 35 – 85 % RH non-condensing.

A hot line in a cool plant room and a mild line in an unshaded outdoor location are different problems. Checking only one figure will miss one of them.

Ingress protection — and where the electronics live

Match protection to the actual installation environment: indoor plant room, outdoor exposed, washdown area, or below-grade chamber.

This is also where integrated and split designs diverge in a way that affects planning. The FU-LT and FU-ES are integrated units rated IP66, so sensor and display sit together on the pipe. The FU-TX 310 is a split design: its probes are rated IP61 as standard or IP68 in the waterproof chemical-resistant version, but the host unit has no waterproof rating of its own and is intended to be installed inside a control box — an additional enclosure is required for outdoor mounting.^[3] That is a cost and a design decision to account for up front, not a detail to discover on site.

Power

Confirm what is available at the installation point before specifying. A meter needing mains AC where only DC control power exists becomes an electrical project rather than an instrument installation.

Across the range: the FU-LT runs on DC 24 V, the FU-ES accepts a wider DC 12 – 36 V input, and the FU-TX 310 takes DC 9 – 30 V or runs from a 100 – 240 V 50/60 Hz AC transformer — the AC option being useful for large-diameter installations in plant rooms where mains is already present.

Signal output and integration

This is the specification most often discovered too late. The question is not whether the meter has an output, but whether it has the output your system reads.

All three LORRIC clamp-on meters carry Modbus RTU over RS485, so digital integration is available across the range. Analog resolution and switch output differ: the FU-LT provides a self-powered 10-bit 4 – 20 mA output with NPN switching; the FU-ES a self-powered 16-bit 4 – 20 mA with a two-wire optocoupler switch compatible with NPN or PNP; the FU-TX 310 a self-powered 12-bit 4 – 20 mA.

Analog resolution is a specification that only matters until it does. For a number on a local display it is irrelevant. For closed-loop control taken off the analog signal, the difference between 10-bit and 16-bit resolution is the difference between roughly a thousand and sixty-five thousand discrete steps across the range.

One further capability worth knowing about if you are doing energy work rather than just flow measurement: the FU-TX 310 supports two external PT1000 temperature sensors (−100 to 300 °C, 0.1 °C resolution).^[3] Flow plus differential temperature is what makes thermal energy calculation possible, which is the basis of chilled-water and heating-loop energy monitoring.

For detail on each interface, see What is Analog Output, What is Modbus RTU RS485, and What is a Switch Signal.

Before comparing specific meters, these are the answers to have in hand. They apply regardless of manufacturer.

Three series cover different parts of the range. Pipe size decides most of it; the remaining criteria decide the overlaps.

Side by side

Choosing within the overlaps

½″ to 1″ — FU-LT or FU-ES. Choose on space and material. The FU-LT if the installation is cramped, the line is light plastic, or the pipe is PFA or PTFE. The FU-ES if you want automatic pipe detection, the diagnostic third transducer, higher analog resolution, or a wider power input.

2″ to 3″ — FU-ES or FU-TX 310. Choose on access and on what you are measuring for. The FU-ES if an operator can reach the meter and you want an integrated unit with no enclosure to specify. The FU-TX 310 if the pipe is high, buried, or awkward to read at, if fluid temperature exceeds 85 °C, or if you need PT1000 inputs for energy calculation.

If your fluid is a clean liquid and your pipe falls between ¼″ and 16″, one of the three series above covers it, and the checklist should tell you which.

If your fluid carries suspended solids or sustained aeration, transit-time is the wrong principle regardless of model — look toward Doppler instead. If your accuracy requirement is tighter than ±3 % of reading, clamp-on ultrasonic may not be the right technology at all, and our Flowmeter Selection Guide is the better place to start.

Product pages: FU-LT Lite Flow · FU-ES EchoSense · FU-TX 310

Related reading: How Do Ultrasonic Flow Meters Work? · Z, V, N, W Installation Methods · Flowmeter Calibration Services