PT Tirta Bumi Case: Water and Wastewater Tank Level Measurement With 80 GHz Radar

Three basins at one municipal water treatment plant failed in three different ways, and all three failures had the same root cause. This article investigates the level-measurement problem across a raw water tank, a clear well and a wastewater lift station wet well, where the media are water and screened wastewater - high-dielectric liquids that should be easy to measure. The existing ultrasonic transmitters failed because condensation dripped onto the transducer face, foam on the wastewater surface absorbed the acoustic pulse, and air temperature changes shifted the speed-of-sound correction. Each failure is traced to its physical cause, which is exactly why the replacement - an 80 GHz FMCW radar level meter - works where ultrasonic did not. The article covers the diagnosis, the selection, the installation, and what changed for pump control after the wet season.

Client Profile: PT Tirta Bumi Water Treatment (Indonesia)

PT Tirta Bumi Water Treatment is a municipal water utility in Indonesia operating raw water intake, clear well and wastewater lift station facilities. Pump sequencing and overflow prevention at these basins depend on level signals that stay reliable through humid atmospheres, condensation and foam - the exact conditions that defeated the previous ultrasonic instruments.

2. The Site: Three Basins, One Recurring Complaint

The plant's instrument log told a consistent story across three locations: a raw water tank buffering river intake, a clear well storing treated water, and a wastewater lift station wet well feeding the plant. All three used ultrasonic level transmitters, and all three appeared in the maintenance log for the same reasons - erratic readings, false pump commands, and cleaning visits. The wet well was the worst case: its level signal starts and stops the sewage pumps, so every false reading became a pump event. The investigation started with a simple question: are these three separate failures, or one failure repeated three times because the technology is wrong for the environment?

3. Failure Mode 1: Condensation on the Transducer

In the raw water tank and clear well, the atmosphere above the water is humid, and the transducer face - colder than the vapor - collected condensation. A layer of water droplets on an ultrasonic transducer is not a cosmetic problem: it deadens the acoustic pulse and corrupts the echo. The reading drifted or dropped out, and the control loop reacted to a signal that no longer represented the water surface. The maintenance answer was cleaning, on a schedule. But the schedule treated the symptom; the cause was that an acoustic sensor was asked to work through a water film on its own face. This is physics, not a product defect, and it repeats in every humid tank.

4. Failure Mode 2: Foam on the Wastewater Surface

In the wet well, surfactants in the sewage and turbulence from inflow produced a foam layer. Ultrasonic pulses are acoustic pressure waves; foam absorbs and scatters them. The transmitter then locked onto whatever echo came back first - sometimes the foam top, sometimes the liquid, sometimes nothing stable. The result was pump cycling on false high levels and dry-running on false low levels, and in wet weather the risk of overflow. The foam was not a temporary condition; it is a permanent feature of sewage. The measurement had to see through it, and acoustic measurement cannot.

5. Failure Mode 3: The Speed of Sound Is Not Constant

Ultrasonic level measurement converts time-of-flight to distance using the speed of sound in air - and that speed changes with temperature, humidity and gas composition. Ultrasonic transmitters carry a temperature sensor and apply a correction, but the correction assumes the air along the whole path is at the measured temperature. In a sun-warmed tank or a ventilated well, the air is not uniform, so the correction itself became an error source. The readings looked plausible and were wrong. This third failure mode is the subtlest, because it does not announce itself with dropouts; it just quietly degrades the measurement.

Recommended Qinwei Instruments for This Application

7. The Diagnosis Table: Why Radar Removes All Three

The three failure modes map one-to-one onto the physics of radar:

This table is the selection logic. An 80 GHz FMCW radar was chosen because the high frequency allows a small, lightweight antenna that mounts easily over open basins, and FMCW processing gives a stable, filtered echo. The antenna material is PVDF or PTFE, hydrophobic by design. The same foam that blinds ultrasonic barely affects radar - the wave passes through and reflects from the water below - while a differential-pressure sensor would see through foam but drift with density and clog at the diaphragm. The technology choice is therefore a decision about which failure mode the site can tolerate.

8. Configuration for Pump Control

The radar was configured as a level source for the PLC, not as a standalone controller. Level in meters, referenced to the basin bottom; 4-20 mA scaled over the working range; HART enabled for remote changes. The pump start, stop and high-alarm thresholds stayed in the PLC, so the instrument remained a simple, reliable level source. Damping was set to a few seconds - enough to smooth inflow turbulence, short enough not to mask a genuine high level. The fail-safe direction was the interesting decision: in the wet well, a lost echo was set to drive the output high, so the pumps run rather than overflow. That choice must match the site's control philosophy; it is a decision, not a default.

9. Bracket, Cable, and the Wet Well

Installation was deliberately simple. Over the open basins, the radar went on a rigid bracket away from the inflow stream, so splashing water did not pass through the beam; a flexing bracket would move the antenna and shift the zero reference, so rigidity was checked on site. In the wet well, an IP68 housing was chosen because the well can flood. The cable was shielded twisted pair, run separately from the pump power cables, grounded at one end only - a single-point ground avoids the ground loop that would otherwise inject noise into the 4-20 mA loop. In the sewage atmosphere, the cable entry and housing rating were confirmed against the site's H2S considerations - the instrument does not override the location's safety requirements.

10. Commissioning: Setting the Fail-Safe the Right Way Round

Commissioning followed the diagnosis. The mounting offset and basin depth were entered so the output read true level. An empty-spectrum recording captured fixed echoes from brackets, walls, ladders and pipework; false-echo suppression was applied. Pump start and stop levels were verified against a tape measure. The wet well was then tested under foam: the echo threshold was set to track the liquid surface rather than the foam top, and the behavior was observed during a period of known foam. The fail-safe direction was tested by disconnecting the loop and watching the PLC respond as designed. Every setting and verification went into the plant file.

Why Choose Qinwei Instruments

Xi'an Qinwei Instrument Factory (brand QINWEIYB) is a manufacturer established in 2010 that specializes in industrial automation instruments, with 500-800 employees and 50%-60% of output exported worldwide. For water and wastewater service, Qinwei supplies IP67/IP68 FMCW radars with hydrophobic PVDF/PTFE antennas, 4-20 mA, HART and Modbus interfaces, and simple bracket-mount options for open basins. All instruments are factory-calibrated with traceable certificates; MOQ starts from 1 set, delivery is 5-8 working days, and OEM/ODM customization is supported.

12. The Wet Season Test

The wet season provided the verdict. Inflows rose quickly and often; the wet well levels swung through the full range; foam was heavy. The radar tracked the liquid surface through all of it, the pumps cycled on true levels, and the overflow risk that ultrasonic could not see was gone. The raw water tank and clear well stayed stable through the humid months with no condensation dropouts. The recurring transducer cleaning was no longer needed. Before-and-after statistics are not provided; what the plant gained is a level signal that reflects the liquid surface rather than the foam, and control loops that behave accordingly.

13. Engineering Value, and Where This Diagnosis Repeats

The value of the change is best expressed as avoided events: no overflow from foam-blind readings, no pump wear from false cycling, no maintenance exposure in the wet well. The diagnosis method - tracing each failure to its physical cause and checking the replacement against that cause - is the transferable part. It applies to raw water intakes, clear wells, filter backwash tanks, sludge tanks, stormwater basins and pump station wet wells everywhere. For very deep wells or narrow pipes, guided wave radar should be evaluated in parallel; for sewage wells with H2S, confirm housing materials and follow the site's gas detection requirements. When requesting a quotation for water or wastewater level service, provide the basin depth and mounting type (nozzle or bracket), whether foam or flooding is expected, the available supply voltage, and the control system interface (4-20 mA, HART or Modbus).