Short answer
Most failed fuel sensor installations fail on physical checks that cost nothing beforehand: internal tank depth against measurement range, a free tracker port of the matching electrical standard, the supply rail, and the cable gland, which is where water actually enters a top-mounted sensor.
Key takeaways
- Measure the tank's internal depth at the mounting point, not its external height, and compare it to the sensor's measurement range.
- RS-485 needs a serial port of that standard: our Teltonika FMC003 listing has 1× RS-232, which is a different electrical standard and not interchangeable.
- A 9–40 V DC input covers both 12 V and 24 V nominal systems with margin at each end.
- The cable gland and cable entry are the real water path on a top-mounted sensor, not the probe body.
- Nothing reads litres until a calibration table exists, and the calibration run has to follow the fitting rather than precede it.
01Check one: internal depth, measured
The first check disqualifies the most options and costs nothing but a dip tape. The OMNICOMM listing gives a measurement range of 0 to 4,000 millimetres, and that is a hard ceiling on the depth it can read. The number to compare it against is the internal depth of the tank at the mounting point, which is not the tank's external height: subtract the wall thickness and any sump or false floor, and remember a shaped tank is shallower at the edges than in the middle. Leave clearance for the mounting thread as well. Getting this wrong is not a precision problem, it is a coverage problem, a sensor shorter than the tank simply stops reporting below its reach, so the fuel underneath becomes invisible rather than imprecise, and no software setting recovers it.
02Check two: the mounting point and the thread
The OMNICOMM listing is a top-mount sensor with a standard thread, which is the arrangement that avoids cutting or welding the tank. Two things still have to be established on the vehicle. The first is whether a suitable opening exists on the top surface, a spare boss, a blanked fitting, or an accessible position where one can be fitted according to whatever the tank's own approval permits. The second is where on that top surface the sensor should sit, and the honest answer is as near the centre as the tank allows. A probe near an end wall sees a level that swings further as the vehicle pitches or the tank sits on a slope, because the surface rotates about the middle. Centre-mounting does not remove motion error, but it reduces the geometric part of it for nothing.
03Check three: the port, and its electrical standard
This is the check most often skipped, and it is the one that turns an order into two orders. The OMNICOMM listing specifies RS-485 Modbus RTU. RS-485 is a serial bus, so the tracker needs a free serial port of that standard plus support for the protocol, not merely a spare pair of wires. Our Teltonika FMC003 listing has one RS-232 port, which is a different electrical standard and not interchangeable with RS-485, so pairing those two requires a converter rather than a cable. The FMC003 does expose one analog input rated 0 to 30 V, which is the route for a sensor with a voltage output instead, at the cost of turning the reading into a calibration exercise. Confirm the tracker side from its own datasheet before ordering either device.
04Check four: the supply rail, with margin
Vehicles run at nominal 12 or 24 volts and neither figure is stable: cranking pulls the rail down and alternator charging pushes it up. The OMNICOMM listing accepts 9 to 40 volts DC, which covers both nominal systems with room at each end, and draws under 30 milliamps. The Eurosens Dominator specification states 9 to 32 volts, which also spans both. Two practical notes follow. Current draw matters when several sensors share a circuit or when an asset is parked and the tracker is running on a backup battery, so it is worth adding up rather than assuming. And the tracker has its own window, the FMC003 listing takes 10 to 30 volts DC, which is narrower at the top than the sensor's, so on a 24 V system the tracker is the component with less headroom.
05Check five: the gland is the water path
An ingress rating describes the enclosure, and on a top-mounted tank sensor the enclosure is rarely where water gets in. The OMNICOMM listing is IP68, and our Eurosens comparison point is IP67; the IEC code that produces those digits grades solids from 0 to 6 and liquids from 0 to 9 and says nothing about temperature at all. What the rating cannot describe is a gland that was not tightened, a cable dressed so it runs downhill into the entry, or an unsupported cable that works the seal loose through vibration. Those are workmanship, and they are where a well-specified sensor still fails. Dress the cable so water runs away from the entry rather than towards it, support it clear of the tank top, and treat the gland torque as part of the installation rather than the last step nobody records.
06Check six: the run is not finished at power-up
A powered sensor reporting a plausible moving number is not a working installation, because nothing has yet told the platform what those numbers mean in litres. That is the calibration table, and it has to follow the fitting rather than precede it, since the probe's depth and angle in its mounting are part of what gets measured. Teltonika documents the procedure: fill in increments of 5 to 10% of capacity, allow roughly 30 seconds of settling at each step, and record the reading against the known litres added, which the platform then fits as cubic coefficients a0 to a3. Teltonika also names the check that ends a bad run, the recorded values must change consistently, and up-and-down values mean a fault to find before continuing. Treat that as a stop condition, not a warning.
07What to expect once it is running
Two expectations are worth setting before the first report is read, because both are normal and both get reported as faults. The first is that accuracy on a moving vehicle is not the accuracy on the datasheet. Technoton publishes not more than 0.2% stationary against 1% for moving vehicles on a capacitive sensor, a fivefold degradation from motion alone, quoted by the manufacturer. The second is response time. The OMNICOMM listing states under one second, which tells you the sensor is not the bottleneck and nothing more; what reaches the platform is set by the tracker's record interval and by any smoothing applied along the way. So a sub-second sensor behind a five-minute record interval gives five-minute resolution. Set both expectations with the operator on day one rather than after the first disputed report.
| Check | What to measure or confirm | What it prevents |
|---|---|---|
| Internal depth | Tank depth at the mounting point versus measurement range | Fuel below the probe's reach becoming invisible |
| Mounting point | A usable top opening, as near the centre as possible | Extra swing from pitch and slope |
| Port and standard | A free port of the sensor's own standard, not just a spare wire | Ordering a converter you did not budget for |
| Supply rail | Sensor and tracker windows both, plus total current draw | Out-of-spec voltage during cranking or charging |
| Cable gland | Gland torque, cable dressed downhill away from the entry, support | Water tracking in past a good IP rating |
| Calibration run | Table built after fitting, in 5–10% increments | A moving number that means nothing in litres |
Figures and sources
- OMNICOMM measurement range and installation method, the two figures the depth check compares: 0 – 4,000 mm; top-mount, standard thread; capacitive (LLS series) (Source: ENTEK.AI product catalogue, as of 2026-08-31)
- OMNICOMM interface, supply voltage window and current draw: RS-485 Modbus RTU; 9 – 40 V DC; < 30 mA (Source: ENTEK.AI product catalogue, as of 2026-08-31)
- OMNICOMM response time, ingress rating and operating temperature: < 1 second; IP68; −40 °C to +85 °C; ±0.5% (Source: ENTEK.AI product catalogue, as of 2026-08-31)
- Teltonika FMC003 ports and supply window, the tracker side of the interface and voltage checks: 1× RS-232; analog input 1 (0 – 30 V); 10 – 30 V DC; IP41 (Source: ENTEK.AI product catalogue, as of 2026-08-31)
- Eurosens Dominator published supply range, ingress rating and interface: ±1%; 9–32 V; IP67; −40 to +85 °C; RS-485 (Source: Mechatronics, eurosens Dominator Specification v.1.0, as of 2026-08-31)
- What an IP rating grades, relevant because it describes the enclosure, not the cable entry workmanship: first numeral solids 0–6, second liquids 0–9; temperature absent (Source: IEC, Ingress Protection (IP) ratings, as of 2026-08-31)
- Teltonika documented calibration procedure that must follow the fitting: 5–10% tank increments, ~30 s settling; yields cubic coefficients a0, a1, a2, a3; voltages must change consistently (Source: Teltonika Telematics Wiki, Analog Fuel Sensor, as of 2026-08-31)
- Capacitive fuel level sensor accuracy, stationary versus moving, the expectation to set on day one: not more than 0.2% stationary; 1% for moving vehicles (Source: Technoton, Fuel level sensor DUT-E product specifications, as of 2026-08-31)
Questions
Can I connect an RS-485 sensor to a tracker with an RS-232 port?
Not directly. They are different electrical standards and not interchangeable, so the pairing needs a converter. Our Teltonika FMC003 listing has 1× RS-232 plus one analog input rated 0 to 30 V, which is the alternative route for a voltage-output sensor.
Does an IP68 rating mean water cannot get in?
It grades the enclosure. The IEC code covers solids 0–6 and liquids 0–9 and says nothing about a gland that was not tightened, a cable dressed downhill into the entry, or an unsupported cable working the seal loose. That is workmanship, and it is the usual water path.
Should I calibrate before or after fitting the sensor?
After. The probe's depth and angle in its mounting are part of what the calibration measures, so a table built before fitting describes a different installation. Teltonika documents filling in 5–10% increments with about 30 seconds of settling at each step.
Will the sensor hit its datasheet accuracy on a moving truck?
Expect less. Technoton publishes not more than 0.2% stationary against 1% for moving vehicles on a capacitive sensor, a fivefold degradation from motion alone, stated by the manufacturer. Set that expectation with the operator before the first report is disputed.
Sources
- Teltonika Telematics Wiki, Analog Fuel Sensor (post-fitting calibration procedure)
- IEC, Ingress Protection (IP) ratings (what the two digits actually grade)
- Mechatronics, eurosens Dominator Specification v.1.0 (supply range and ingress)
- Technoton, Fuel level sensor DUT-E specifications (stationary versus moving accuracy)