An ultrasonic level sensor measures liquid or solid level by time-of-flight: it emits a pulse above 20 kHz and converts the echo delay into distance using D = C × t / 2, where C is the speed of sound in air. Because C changes with temperature — roughly 0.6 m/s per °C — accuracy depends far more on correct installation and temperature compensation than on the transducer itself. This guide covers mounting, dry vs wet calibration, and the field faults you will actually meet.
What Is an Ultrasonic Level Sensor?
An ultrasonic level sensor uses sound above the range of human hearing to measure the distance between the sensor face and the surface of a liquid or solid. The transducer emits a burst, the echo returns, and the electronics time it. With the speed of sound known (and compensated for temperature), distance follows:
D = C × t / 2
The factor of 2 accounts for the round trip — the pulse travels to the surface and back. The result is a non-contact measurement: nothing touches the product, so there is no clogging, corrosion, or pressure sealing to worry about at the process connection. Typical industrial range is a few centimetres up to tens of metres, depending on model and frequency; our KUS640 series supports customisable maximum measuring distances from 1 m to 50 m, and the dual-probe KUS700 is built for very small blind zones at close range.
Installation: Laying the Foundation Right
Proper installation is the single biggest determinant of reading quality. Nearly every “faulty sensor” turns out to be a mounting problem.
1. Selecting the Mounting Location
- Clear acoustic path. Choose a location with an unobstructed line of sight to the product surface. Keep pipes, agitator blades, ladders, heating coils, and tank internals out of the beam cone.
- Away from the filling stream. Mount clear of inlets and outlets; a cascading fill creates surface foam and turbulence that scatter the echo.
- Thermally and electrically quiet. Avoid extreme heat, direct sun on the housing, heavy vibration, and proximity to variable-frequency drives or large motors.
- Non-contact gap. For a non-contact sensor, respect the manufacturer’s specified minimum gap between the sensor face and the highest expected product level. This is the blanking distance, not a suggestion.
2. Mounting the Sensor
- Use the manufacturer’s bracket and hardware; a hand-made mount is the most common source of angle drift over time.
- Keep the sensor face parallel to the product surface. Off-angle mounting sends the echo away from the transducer, which shows up as intermittent loss of signal rather than an obvious error.
- Leave the mounting thread or flange accessible — you will need to remove the unit for cleaning in fouling service.
Installation Best Practices Checklist
| Factor | Recommendation |
|---|---|
| Mounting distance from tank wall | Keep > 300 mm from the wall and from the filling inlet; on tall/narrow tanks use the minimum-radius figure in the datasheet |
| Blank distance (dead zone) | Respect the specified dead zone (commonly 0.2–0.5 m); the highest product level must stay below it |
| Max measuring range | Model-dependent — from 1 m up to 50 m on the KUS640 series, verified against the datasheet |
| Temperature effect | Speed of sound changes ~0.6 m/s per °C; always use temperature-compensated models |
| Foam/vapour/dust | Avoid heavy foam or dense vapour; if either is permanent, switch to radar |
| Condensation | Angle the sensor face slightly downward or use a hooded mount so condensate runs off instead of pooling on the transducer |
Calibration: Tuning for Precision
Calibration corrects for the installed geometry (the reference point relative to the tank bottom) and for any residual tolerance in the transducer. There are two approaches, and they suit different situations.
1. Dry Calibration
Dry calibration sets the zero and span without filling the vessel. You enter a known distance — or place a target at a measured distance from the sensor face — and the sensor stores the echo timing for that reference. It is fast, needs no process interruption, and is normally enough to commission a tank. It does rely on the operator’s tape measurement being correct, so a wrong number here is copied into every subsequent reading.
2. Wet Calibration
Wet calibration fills the vessel to known levels and aligns the output to the physically measured height, using a dipstick or calibrated sight glass. Repeat at several points across the range to verify linearity. It is slower and needs access to the process, but it captures effects that dry calibration cannot — foam on the surface, vapour in the ullage, and the real dielectric-free acoustic behaviour of the application.
Dry vs Wet Calibration at a Glance
| Aspect | Dry calibration | Wet calibration |
|---|---|---|
| Best for | Commissioning; tanks that cannot be filled or taken out of service | Critical inventory or custody-transfer duties; vessels with foam or vapour |
| Reference used | Known distance or calibration target set off the sensor face | Actual liquid level measured by dipstick or sight glass |
| Accuracy achievable | Limited by the accuracy of the entered distance | Highest — corrects for real process conditions across the range |
| Time and effort | Minutes; no process interruption | Longer; requires filling, stabilising and repeating at several points |
| Verifies linearity? | No — single reference point | Yes — multiple points across the measuring range |
| Typical interval | On commissioning, after re-mounting | On commissioning and periodic re-verification (commonly 6–12 months) |
Troubleshooting: When Things Go Awry
The three symptoms below account for the overwhelming majority of field problems. Each has a physical cause — work through the cause column in order before touching the electronics.
| Symptom | Likely cause | Corrective action |
|---|---|---|
| Incorrect (offset) readings | Wrong installed height or reference entered; temperature compensation disabled; scaling error in the controller | Re-enter the reference distance; confirm the temperature sensor is reading ambient; verify the 4–20 mA or RS485 scaling at the receiving end |
| Erratic / jumping readings | Electrical interference from VFDs, motors or welders; unstable or unshielded cable routing | Shield the cable and ground the shield at one end only; route signal cable away from power cable; fit ferrite cores; relocate the sensor or the drive |
| No readings | Sensor misalignment; damaged transducer; wiring fault | Re-align the sensor face to the surface; inspect the transducer; verify wiring against the datasheet pin definition |
| Weak or lost echo | Foam layer absorbing the pulse; heavy vapour or condensation on the face | Clean the face; fit a hooded mount; add desiccant or a heated face; if foam is permanent, use radar instead |
| Readings drift high/low with weather | Temperature not compensated, or sun heating the housing | Enable/verify temperature compensation; shade the housing; check for a failed on-board temperature sensor |
| Reading freezes at a fixed value | Product level has entered the blanking (dead) zone, or the echo is being locked onto a wall or agitator | Lower the high-level limit below the dead zone; move the sensor to the tank centre; reduce the beam angle |
Choosing the Right Ultrasonic Level Sensor
Once installation and calibration are sound, model choice decides what you can measure. Two practical starting points:
- Clean liquids, tanks and open channels: the full ultrasonic level sensor range covers integrated and split designs. The KUS640 series pairs a wide-temperature LCD (−30 to 80 °C) with a 24 VDC supply and 4–20 mA / RS485 output.
- Sludge, mud and sedimentation interfaces: a contact-type instrument is required because the interface sits inside the liquid. The KWD120 split-type ultrasonic sludge level meter measures underwater distance directly and supports 50 kHz to 650 kHz probes with measuring ranges from 1 m to 40 m.
If foam, heavy condensation or a strongly agitated surface is unavoidable in your process, a radar level sensor will hold signal where ultrasound will not — the trade-off is a higher unit cost, which is usually still cheaper than a permanently unreliable reading.
FAQ
Q: How should I choose the mounting height for an ultrasonic level sensor?
A: Mount the sensor on the tank centreline where practical, keeping the face above the maximum product level plus the blanking distance, and at least 300 mm clear of the tank wall and the filling inlet.
Q: What is the blanking distance (dead zone)?
A: It is the minimum distance directly below the sensor face in which the unit cannot resolve an echo — commonly 0.2–0.5 m, but model-specific, so always read it from the datasheet before setting the high-level alarm.
Q: Will foam affect the reading?
A: Yes. A foam layer absorbs and scatters the ultrasonic pulse, which weakens or loses the echo. Light foam can often be managed with a larger beam or higher gain; persistent foam is a case for radar rather than ultrasound.
Q: How often should an ultrasonic level sensor be calibrated?
A: Calibrate on commissioning, then re-verify periodically — commonly every 6 to 12 months — or whenever readings drift against a known reference level, or after any re-mounting.
References
- Endress+Hauser, Ultrasonic level measurement — installation and application notes.
- Vega, Ultrasonic level measurement — mounting and beam-angle guidance.
- Kacise, KUS640 integrated ultrasonic level sensor and KWD120 split-type ultrasonic sludge level meter datasheets.
