Most false level readings trace to one of a handful of causes, and only two of them are solved by changing the sensor. Sorting the symptom first saves a lot of ladder time.
Key answers at a glance
Why does a bin read empty when it is full? The return off the peak of a heaped material is small and soft, and a piezoelectric ultrasonic sensor may miss it. Electrostatic ultrasonic transducers have approximately 40 dB greater receive sensitivity than piezoelectric ultrasonic transducers, which recovers that faint echo.
Where should the sensor aim? At the peak of the heap. The peak presents a small area parallel to the transducer face, which returns the strongest available signal. A highly angled surface deflects the signal away.
Why do readings drift through the day? The speed of sound in air varies with temperature, so a time-of-flight measurement needs temperature compensation to stay accurate as ambient conditions change.
Why does the reading fail as the vessel fills? The surface has entered the near field. A piezoelectric ultrasonic transducer rings after transmit, masking echoes inside roughly 13 to 14 inches or more. Electrostatic ultrasonic transducers use a low-resonance membrane that settles quickly, so with appropriate drive electronics they keep reading to within 1 inch of the face.What fixes a reading that jumps? Usually mounting rather than the sensor. Multipath off a sidewall, ladder, or agitator returns an echo from something other than the product surface.
A level reading that is wrong is worse than no reading at all, because the control system acts on it. A bin that reads empty while it is full cancels a refill request, and a tank that reads full while it is filling shuts a pump off early.
Most of these failures come from a short list of causes, and they divide cleanly into the ones that need a different sensor and the ones that need a different installation.
Why do some ultrasonic sensors read empty when the bin is full?
Because the echo coming back from the top of the material is too weak for the sensor to register, not because the material is not there. Grain, feed, powder, and pellets settle into a cone as a bin fills. The return off the cone peak is small and soft, and a piezoelectric ultrasonic sensor may miss it, so the bin reads empty. Electrostatic ultrasonic transducers have approximately 40 dB greater receive sensitivity than piezoelectric ultrasonic transducers, which is enough margin to detect that faint echo and report a valid product height.
The failure is worth recognizing because it does not look like a measurement limit. When no echo clears the detection threshold, most systems report maximum range or hold the last valid value, and both outputs look plausible on a screen. A bin that reads exactly empty week after week, while somebody is climbing it to check, is usually a sensitivity problem rather than a wiring one.
Where should you aim a level sensor in a bin with a cone-shaped surface?
Aim at the peak of the heap. The peak presents a small area parallel to the transducer face, which is the optimum geometry for the strongest available return signal. A highly angled surface deflects the signal away from the transducer, so aiming at the slope returns nothing usable, regardless of which technology is doing the sensing.
That makes mounting position a measurement decision rather than a convenience. Aim at the top of the cone that represents peak fill, offset far enough that the beam is not looking into the falling stream during loading. A 15 degree beam angle helps capture weak and off-axis echoes, but it does not rescue an aim point on a steep slope. Material drawn from the bottom center also forms a crater rather than a cone, so the surface during discharge is not the one the sensor saw during fill — decide which condition the reading has to be accurate in.
Why do tank level readings jump around or drift over the course of a day?
Those are two different failures with two different fixes. A reading that jumps between values is usually a false echo: the pulse reflects off a sidewall, a ladder, a weld seam, an agitator shaft, or the falling fill stream, and the sensor times that return instead of the one off the product surface. A reading that drifts slowly and predictably is usually temperature, because the speed of sound in air changes with temperature and an uncompensated time-of-flight measurement moves with it.
Multiple pulse reflections are an installation problem. Mount clear of the sidewall and internal structure, keep the beam path clear, and confirm that nothing intrudes into it at any fill level. Temperature is a system-design problem, and a co-located temperature sensor lets the controller correct for the actual speed of sound. Treat that compensation as a requirement in any vessel that sees ambient swing, not as an optional refinement. Averaging several pulses before acting on a reading, and applying hysteresis at any switch point, keeps material motion during filling from producing chatter.
Why do some ultrasonic sensors stop reading when the tank is nearly full?
Because the product surface has moved into the near field, where the receiver is not listening. A piezoelectric ultrasonic transducer keeps ringing after its transmit pulse, and that ring-out masks echoes inside roughly 13 to 14 inches or more. As the vessel fills toward the sensor, the surface crosses into that window and the measurement goes invalid at the fill level that matters most.
Electrostatic ultrasonic transducers use a low-resonance membrane that settles quickly, so with appropriate drive electronics they keep reading to within 1 inch of the face. In a shallow tank, a squat hopper, or any vessel where the sensor has to mount close to the maximum fill line, that difference decides whether the top of the range is measurable at all. If you are specifying a packaged module rather than a bare transducer, check its published minimum range, which is set by the module blanking interval rather than by the transducer.
Are false readings a sensor problem or an installation problem?
| Symptom | Most likely cause | What fixes it |
| Reads empty on a full bin of grain, feed, or powder | Weak echo off a small, soft heap peak | Sensor choice — receive sensitivity |
| Goes invalid as the vessel approaches full | The surface has entered the near field | Sensor choice — close-in capability |
| Alternates between two distances | Multipath off a sidewall, ladder, or agitator | Mounting position and a clear beam path |
| Drifts slowly with ambient temperature | Speed of sound in air varies with temperature | Temperature compensation in the controller |
| Erratic only during filling | Material in motion and the falling fill stream | Pulse averaging, hysteresis, sampling between fills |
| Stable but consistently offset | Empty-vessel baseline taken from the wrong reference | Re-baseline from the installed position |
One limit belongs in the design conversation rather than after commissioning: standing moisture or condensation on the transducer face can be a real constraint, so a saturated headspace needs to be accounted for during specification. Airborne dust is different. Unlike optical sensing, ultrasonic sensing is not defeated by dust, and the high receive sensitivity of electrostatic ultrasonic transducers provides additional headroom when dense dust attenuates the return.
Frequently Asked Questions
Mount it clear of the sidewall, away from internal structure, out of the falling fill stream, and aimed at the peak of the heap. Electrostatic ultrasonic transducers have a 15 degree beam angle, which helps capture weak and off-axis echoes, but any beam path that clips a ladder, a weld seam, or an agitator shaft can return an echo from that object instead of from the product. Check the beam path at every fill level, not just at empty.
Yes, in any vessel that sees ambient temperature swing. Ultrasonic measurement is time of flight, and the speed of sound in air varies with temperature, so an uncompensated reading shifts even when the level does not. A co-located temperature sensor lets the controller correct for it. The transducer is separately stable: the bare 600 Series transducer holds frequency and gain from -40°C to +85°C, while both the 6500 Series Ranging Module and the Smart Sensor are rated 0°C to +70°C.
Electrostatic ultrasonic transducers use a low-resonance membrane that settles quickly, so with appropriate drive electronics they keep reading to within 1 inch of the face, and a single transducer can be configured to cover 1 inch to over 40 feet. A piezoelectric ultrasonic transducer rings after transmit and masks echoes inside roughly 13 to 14 inches or more. A packaged module carries its own rated minimum, set by its blanking interval, so check the module datasheet rather than assuming the transducer figure.
Because the coating has a dielectric constant greater than air. This can cause the sensor to indicate a higher level than actually exists. Sticky or hygroscopic products build up on a probe gradually, so the error grows quietly instead of appearing all at once. Ultrasonic measurement is non-contact, so no sensing element sits in the product to accumulate a coating.
Yes. The Mini Series covers both cases. Mini-A has an analog output, either 0 to 5 or 0 to 10 volts DC, and Mini-S has two switchpoint outputs. In a tank, one switch is typically set for full and one for empty, which drives fill and refill logic directly without a controller having to interpret a distance value.
Specifying level measurement for a bin, silo, or tank?
Explore SensComp electrostatic ultrasonic transducers, ranging modules, and the Mini Series, and see how beam path, aim point, and minimum range fit your vessel: https://www.senscomp.com/