Ultrasonic vs Radar Level Sensor: Which Works Best in Your Tank?

You want to know how full a tank or rainwater cistern is without checking with a dipstick every time. A fill level sensor above the surface reports the distance over the radio, and comparing models quickly leads to one question: ultrasonic vs radar level sensor, which one? The short answer: ultrasonic is often the more accurate and more affordable choice for calm liquids, stable temperatures and short distances. Radar earns its place when condensation, vapor, dust or large temperature swings are involved, which is exactly where sound struggles.

The right technology depends on your tank: its height, the gap between lid and highest fill, the medium, anything mounted inside, and whether the sensor sits in a basement or in direct sun. The sections below cover those points and finish with two specific devices.

How does an ultrasonic level sensor work?

At its core, an ultrasonic sensor is a distance sensor. Its transducer fires a short sound pulse downward, the pulse hits the surface, bounces back and is picked up by the same transducer. The electronics turn the travel time into distance: speed of sound times travel time, divided by two, because the sound makes a round trip. At a distance of 2 m and an air temperature of about 20 °C, the echo returns after roughly 11.7 milliseconds.

That is also the weak spot: the speed of sound depends on the air, about 331 m/s at 0 °C and about 343 m/s at 20 °C. Good sensors therefore measure the temperature at the device and correct for it. What they cannot see is a temperature gradient between the sensor and the liquid.

The horn on the Milesight EM400-UDL in the photos focuses the sound and shields the transducer. The EM400-UDL also tracks its own orientation: if it tilts 20 degrees or more from horizontal, it reports a “Tilt” status. A sensor hanging at an angle aims at the wall instead of the surface.

How does a radar level sensor work?

A radar level sensor uses electromagnetic waves instead of sound. At the speed of light, the round trip over 2 m takes only about 13 nanoseconds, too short for simple timing. Most compact devices therefore use FMCW: the transmitted frequency sweeps steadily, so when the echo returns the sensor is already on a different frequency, and that difference translates into distance.

The Dragino MDS200-LB works this way in the 24.0 to 24.5 GHz band. Modules around 24 GHz are often marketed as mmWave radar. Strictly speaking, the millimeter-wave range starts at 30 GHz; at 24 GHz the wavelength is about 12.5 mm. Industrial process radars often use 80 GHz for a much narrower beam from the same antenna size.

Radar barely reacts to temperature, vapor or air pressure in the measurement path. What it does care about is the medium: water reflects radar waves very well, while heating oil or plastic pellets, with their low dielectric constant, reflect much less. Ultrasonic ignores that difference; any solid or liquid surface is a mirror to it.

Radar vs ultrasonic level sensor: which is more accurate?

The assumption that radar is automatically more accurate does not hold for devices in this class. The standard version of the Milesight EM400-UDL specifies ±(1 cm + 0.3% of distance), valid between −15 and 60 °C and over a range of 25 to 800 cm. At 2 m, that works out to ±1.6 cm. The Pro probes W050 and W100 are rated at ±1% of full scale, so ±5 cm and ±10 cm. The Dragino MDS200-LB, by contrast, states ±10 cm with a resolution of 1 cm.

Keep resolution and accuracy apart: 1 mm resolution only means the reading comes in millimeter steps. How close it is to reality is the accuracy, which always comes with conditions such as a temperature range.

In practice that flips once conditions get rough. An ultrasonic sensor rated ±1.6 cm on paper can be off by several centimeters in a sun-baked container, because the air just under the lid is much warmer than the air above the liquid. A radar rated ±10 cm stays at ±10 cm under those conditions. For a cistern where you only need half full versus nearly empty, either figure is fine. For a shallow tank, where 10 cm is a large share of the height, an accurate ultrasonic unit has the edge if the environment cooperates.

What is the dead zone (blanking distance) of a level sensor?

Every non-contact level sensor has a zone right in front of its sensing face where it cannot measure. With ultrasonic sensors, this dead zone, also called the blanking distance, exists because the transducer keeps ringing briefly after each pulse. An echo that arrives during that time is drowned out by the sensor’s own ringing. With radar, antenna coupling and signal processing limit the near range.

The measuring ranges tell you the blanking distance. According to the retailer listing, the EM400-UDL C050 measures from 25 to 500 cm; the C100 covers 25 to 1000 cm, the W050 starts at 30 cm and the W100 only at 50 cm. The MDS200-LB measures from 0.5 to 20 m. So in this comparison it is the radar that has the larger dead zone, twice that of the standard ultrasonic versions.

So the highest expected level must stay at least one blanking distance plus a margin below the sensing face. If the liquid rises further, many sensors report no value or a wrong one, in the worst case a tank that looks empty. If the container fills almost to the lid, mount the sensor on a riser so the dead zone sits above the maximum fill, with the horn or radar head protruding from its bottom.

Which level sensor works with foam, vapor and condensation?

Here the two methods differ most. Foam is especially hard on ultrasonic sensors: a loose foam layer swallows sound instead of reflecting it, so the echo becomes weak or disappears. Radar often copes better with thin, light foam, but dense, wet foam can dampen radar waves too. There is no guarantee that a radar sees through every foam layer.

Also decide which surface you care about: the top of the foam and the liquid underneath are different targets. According to the manufacturer, the MDS200-LB reports two distances: the closest object and the next one behind it. Whether that second value is the liquid under your foam, only a test on your own container can tell.

Vapor and warm air change the speed of sound and weaken the signal, but have practically no effect on radar in a normal tank. Condensation is a separate issue: droplets on the sensing face deflect the sound. Milesight rates the EM400-UDL for up to 95% relative humidity, non-condensing, yet above a warm liquid in a closed container, condensation is the rule rather than the exception. Dragino explicitly states that the MDS200-LB still measures correctly with water or thick dust on top of the sensor. For heated containers, buffer tanks and shafts, that is a real argument for radar.

Can a level sensor measure solids as well as liquids?

A liquid forms a flat surface that reflects sound and radar straight back like a mirror. Bulk solids behave differently. During filling they pile up into a cone, during emptying they form a funnel, and the sloped surface deflects part of the signal sideways. The sensor also only measures the spot directly beneath it, not the average fill height.

For ultrasonic, dust during filling dampens the sound, and soft or coarse surfaces such as wood chips scatter it. Radar largely sees through dust but gets a weak echo from materials with a low dielectric constant, such as dry plastic pellets.

Both devices here are designed mainly for liquids and distances. Milesight lists tank fill level, boiler expansion vessels and flood monitoring for the EM400-UDL. Dragino lists liquid levels for the MDS200-LB, but also waste bins, sewers and parking spaces. Bin fill level is realistic; a tall grain silo calls for a specialized narrow-beam silo radar.

How do temperature and weather affect an ultrasonic level sensor?

Outdoors, the temperature dependence of sound shows most clearly. Between −10 and 40 °C, the speed of sound changes by about 9%. Temperature compensation absorbs most of that, as long as the air along the whole path is roughly the same temperature. When a rainwater tank or an IBC tote sits in the sun, the lid and the sensor heat up while the air further down stays cool. The sensor then calculates with a temperature that is too high, and the error grows with distance. A deviation of just 3% at a 3 m distance is about 9 cm.

A sun shield over the sensor and mounting on the shaded side help more than any setting. Check the operating temperature too: the EM400-UDL is specified for −30 to 70 °C and has an IP67 enclosure, while the MDS200-LB operates from −40 to 85 °C. Under a black cover in midsummer, the upper limit deserves a critical look.

Rain matters less as drops in the air than as water on the sensing face. A droplet inside the horn or a film of ice in winter can distort the echo, and so can spider webs or insects in the horn, which are common in open shafts. Radar is more robust, since its waves pass through a thin water film on the antenna cover. Wind, in turn, moves the surface of open basins and ponds. For waves, the EM400-UDL offers outlier calibration: if a reading differs from the previous one by more than a set amount, the sensor measures again.

Where should you mount a level sensor on a tank?

The mounting position often matters more than the choice between ultrasonic and radar. The sensor has to sit level with a clear view downward. For the EM400-UDL, Milesight specifies a minimum distance of 30 cm from the side wall and explicitly advises against mounting it above the inlet, where it would measure the incoming stream instead of the surface.

Ladders, braces, agitators, float switches and pipes in the beam create their own echoes. Because the beam widens with distance, an obstruction can be irrelevant when the tank is full and suddenly produce the strongest echo when it is nearly empty. This matters especially for the MDS200-LB, whose radar head has a very wide beam: 78 degrees in one direction and 23 degrees in the other. Align the wide side along the longest clear stretch inside the container. With domed lids, radar manufacturers often advise against the exact center, because multiple echoes can concentrate there.

A tank level sensor on a heating oil tank follows the same rules, with two extras. Heating oil reflects radar waves weakly, which can lead to unstable echoes with a 24 GHz device that has a wide beam. Ultrasonic, on the other hand, usually handles the calm surface in an evenly tempered basement well, as long as the tank opening is large enough for the sensor. Basements are also hard on radio signals, so check beforehand whether LoRaWAN or NB-IoT reaches the spot. Metal lids add shielding, where the MDS200-LB’s radar head on a cable, separate from the radio enclosure, can help.

After mounting, note the distance from the sensing face to the tank bottom; only with that reference does the measured distance become a fill height.

Milesight EM400-UDL vs Dragino MDS200-LB: which should you choose?

Both devices measure from above without contact, run on batteries for years and transmit over the radio. In sensing technology and lifecycle they differ clearly, as the manufacturer specifications show.

The EM400-UDL is the more accurate choice for liquids in containers up to a few meters tall, as long as condensation and sharp temperature swings are not a factor. It offers threshold alarms, offset calibration and the tilt alert. The N03GL C050 version transmits over NB-IoT or Cat-M straight into the cellular network and needs no gateway of its own, while the LoRaWAN versions report every 10 minutes by default.

The MDS200-LB is the more robust choice for damp, dusty and dirty environments such as shafts, wastewater or heated containers, and with 20 m it reaches much further. It reports every 20 minutes by default, more often when the distance leaves a configured range. The trade-offs are the lower accuracy rating and the larger 50 cm dead zone. For new installations, note that Dragino lists the model as end of life, so check spare-part needs and support for your planned lifetime.

Where can you buy an ultrasonic or radar level sensor?

Before you commit, measure the distance from the planned sensing face to the bottom and to the highest fill level. The largest distance has to fit the measuring range, and the smallest must not fall into the dead zone. Then pin down the medium and the environment. Calm liquids at a steady temperature point to ultrasonic and its higher accuracy; condensation, vapor, dust, foam and full sun point to radar. For heating oil and other weakly reflecting media, ultrasonic is often the safer pick.

Also check whether LoRaWAN or cellular coverage reaches the mounting spot and whether you need a gateway, plus the operating temperature, a reporting interval that suits your purpose and, for older models, the lifecycle status.

With those points covered, the ultrasonic vs radar level sensor question for your tank usually answers itself. If you are looking for a suitable device, you can find our models in the fill level sensors category.