KEYSTONEIndustrial Services
Interactive trainer

Radar & ultrasonic level

A non-contact level gauge never touches the product — it times a microwave or sound pulse to the surface and infers level from the distance, which means anything that fools the pulse reads as a confident wrong level. Six tanks, four reads — indicated level vs a dip, echo, environment, config — including the two look-alikes: a false echo versus a dead-band peg, and foam versus a coated antenna.

By Francisco Padilla Interactive Technician level

Before you start — job brief

A non-contact level gauge never touches the product. It times a microwave (radar) or sound (ultrasonic) pulse from an antenna at the top of the tank down to the surface and back, and infers the level from that distance — level equals the empty distance minus the measured distance. That makes it clean and low-maintenance, but it also means anything that fools the pulse reads as a wrong level. Always judge the gauge against an independent check: this tank is 30 ft tall and every job gives you a dip-tape reading of the true level.

  • Radar times a microwave pulse to the surface; ultrasonic times a sound pulse — both infer level from a distance
  • False echoes come from nozzles, agitators and obstructions; foam absorbs the return; product coating the antenna does too
  • A low-dielectric (εr) product reflects weakly — the beam can pass through and lock onto the tank bottom, reading low
  • Read four things: indicated level vs the dip, echo strength & the locked distance, the environment, and the config

Non-contact level measures a distance, not a level

A radar or ultrasonic gauge never touches the product. It sits at the top of the tank and fires a pulse — a microwave burst for radar, a sound wave for ultrasonic — straight down, times how long the reflection from the surface takes to come back, and turns that time into a distance. The level is then simple arithmetic: the empty distance (antenna to the tank bottom) minus the measured distance to the surface. On a 30-foot tank a surface 13.5 feet below the antenna is 16.5 feet of product — 55% full. Nothing wetted, nothing to wear, no seals to leak. The catch is buried in that chain of steps: the gauge does not know the level, it only knows the distance to whatever reflected the pulse. Anything that reflects the pulse early, absorbs it, or lets it pass straight through turns into a wrong level — and the number on the display looks just as confident either way. That is why every reading in this trainer is judged against an independent dip.

False echoes — the gauge locks on the wrong thing

The pulse spreads as it travels, and inside a real tank it can hit things that are not the liquid: a nozzle, a weld seam, a ladder or agitator, a mixing baffle, heavy internal structure. If one of those returns a strong enough reflection, the gauge locks onto it instead of the surface and reports the distance to the obstruction. Because the obstruction does not move, the reading is stuck — it holds a fixed value while the real level rises and falls beneath it. The tells are a reading that will not track the process and, crucially, a clear environment: no foam, no coating, nothing wrong with the return strength — just a strong echo at a distance that does not match the dip. The fix is to teach the gauge to ignore the obstruction (a false-echo or end-of-measurement mapping curve taken on the empty tank), re-aim the antenna, or drop it into a stilling well that shields the beam from the tank internals.

Field note — the two weak-echo faults that look identical

Foam and antenna coating produce the same symptom: a weak or lost return and a reading that sags low or drops out. They are opposite fixes, and the only check that separates them is the environment — where the signal is being lost. Foam sits on the liquid surface, out in the tank, scattering and absorbing the pulse before it can reflect; the antenna is clean, and the cure is to get a clear look at the liquid (a stilling well, a lower frequency, or guided-wave radar in contact with the product). Coating is product built up on the antenna or horn itself; the surface is fine, and the cure is to clean and protect the antenna (an air purge, a PTFE or flush-mount horn). Same weak echo, two different places — read the environment before you decide, or you will clean a horn that isn't dirty and leave the foam untouched.

Dielectric and the dead-band

Radar works by reflecting off the jump in dielectric constant at the surface, so the product has to have enough dielectric (εr) to give a usable reflection. Water and most aqueous or conductive products are easy. A light hydrocarbon or a low-density solvent has a low εr: it reflects only weakly and lets much of the beam pass straight through, down to the metal tank bottom, which reflects hard. The gauge then locks onto the bottom and reads low — a near-empty number on a half-full tank. The cure is to match the technology to the product: a higher-frequency, higher-sensitivity radar, guided-wave radar (which tolerates low εr far better), or a stilling well. A tank pegged high can look like a false echo, but there is a second cause: the dead-band, the blocking distance right under the antenna that the gauge cannot measure. Let the surface rise into it and there is no valid echo at all, so the reading pegs just below full and stops. The way to tell a dead-band peg from a false echo is the echo itself — a false echo is a strong return at a short fixed distance, while a dead-band peg is no valid return, and it only happens when the dip confirms the tank is genuinely near the top. Mount the antenna higher so the working range clears the dead-band, and back up a near-full tank with a high-level switch.

Ultrasonic — the same idea, a softer wave

Ultrasonic level works on the same principle with a sound pulse instead of a microwave one, and it shares most of these failure modes — false echoes off internals, foam absorbing the return, coating on the transducer. It adds a few of its own because sound needs a medium and travels at a speed that depends on the gas above the liquid. Heavy vapor or dust between the transducer and the surface scatters the pulse; a hot temperature layer or stratified gas bends and slows it, shifting the reading; and a vacuum — or near-vacuum — leaves nothing for the sound to travel through at all, so ultrasonic simply cannot be used there while radar can. The discipline is identical: read the indicated level against a dip, look at the return, check the environment, and confirm the configuration fits the vessel.

The discipline

Four reads name every fault in this trainer. The indicated level against the dip tells you the gauge is wrong and in which direction — stuck high, sagging low, pegged, or dropped out. The echo — its strength and the distance it locked onto — separates a strong lock on the wrong target (a false echo) from a weak or lost return (foam or coating) from no valid return at all (the dead-band). The environment at the surface and antenna tells foam from coating. And the config — dead-band, empty and full calibration, dielectric — catches the surface that climbed into the blocking distance and the product the radar was never able to see. Read all four against the dip and the tank tells you exactly which failure you have, instead of a confident wrong number sending you to fix the wrong thing.