Float switches fail quietly. They stick, they re-order after a tank clean, and the first sign is usually an overflow or a dry pump — not a fault message. That is why larger hydroponic farms are moving to level sensing without floats: a continuous reading that can drive a display, raise staged alarms and feed a data log. The three practical options are ultrasonic, hydrostatic and electrode sensors, and each earns its place in a different duty. This article compares them; float valves and mechanical level control are covered separately.
What Non-Float Level Sensing Adds Over a Float Switch
A float answers one question — is the level above or below this point? Continuous level sensing answers a different one: how many litres are in the tank right now? That difference changes what the farm can do. With a continuous signal, an alarm can sit at 20% reserve instead of at the dry-run point, dosing refills can be predicted before they are needed, and the log will show a slow leak weeks before a pump runs dry.
The trade-offs are real. Electronic sensors cost more, need a supply and a place to land their signal, and introduce their own failure modes — foam, fouling, drift and calibration. The sound rule used on most projects: keep at least one independent float as the hard high-level stop and one as the dry-run stop, and let the electronic sensor do the control and reporting. The sensor failing should degrade convenience, not safety.
Ultrasonic Level Sensors: Non-Contact but Fussy About Foam
An ultrasonic sensor sits above the liquid, measures the echo time to the surface, and converts it to distance. Because nothing touches the solution, it is the cleanest option hygienically and the easiest to install on an existing tank — a thread into the lid is often the whole job. It is also the most forgiving option on aggressive solutions, since there is no wetted material to attack.
The failure modes come from the same non-contact principle. Foam on the surface, condensation on the sensor face, or spray from a return jet all scatter the echo and produce false readings. Install the sensor away from return inlets, keep a stilling zone under it, and accept that a planning accuracy of ±5–10 mm on a two-metre tank is realistic in normal duty rather than the ±1 mm the datasheet suggests under laboratory conditions. Blanking distance also matters: the sensor cannot read anything inside its own near-field zone, typically 250–500 mm depending on model, so the tank’s usable depth shrinks by that amount.
Hydrostatic Pressure Probes: Simple and Submerged
A hydrostatic probe hangs in the tank, measures the pressure of the liquid column above it, and reports level as a 4–20 mA or digital signal. It is unaffected by foam, spray or condensation, which makes it the most stable choice on turbulent tanks and the usual pick where reliability outranks convenience. The wetted membrane is the weak point: nutrient solution scales it, algae grow on it, and a fouled membrane reads low or slow. Plan on rinsing the membrane monthly and recalibrating against a dip measurement quarterly, assuming a standard nutrient solution and moderate biofilm load.

Installation details decide long-term behaviour. Hang the probe so it cannot touch the tank floor during agitation, keep the cable clear of pump suction vortices, and bring the signal cable out through its own gland rather than sharing the lid opening with everything else. On deep tanks the cable’s own weight works the glands loose over a year if nobody fits strain relief.
Electrode Probes: Multiple Setpoints Cheaply
Electrode sensing uses conducting tips at fixed heights; when the solution touches a tip, a small current flows and the relay acts. One probe assembly can carry three or four electrodes — say pump start, pump stop, high alarm and low alarm — at the cost of a modest switching unit. That makes it a strong replacement for a cluster of floats inside one fitting, with no moving parts to stick.
The limits are chemical and physical. The tips scale over months and need cleaning like any other wetted metal; the solution must conduct, which nutrient solution does; and the resolution is fixed by where the electrodes are mounted — you get setpoints, never a continuous reading. On tanks where levels move with agitation, electrode tips at closely spaced heights can chatter the relay, so leave sensible vertical gaps between setpoints, on the order of 100–200 mm minimum.
Comparing the Options for a Working Tank
The table below gathers the four realistic inputs, floats included, for a commercial nutrient or make-up water tank.
| Option | Reading type | Strengths | Watch-outs |
|---|---|---|---|
| Float switch | Point (on/off) | Cheap, independent, no supply at sensor | Sticks, re-orders after service, no data |
| Ultrasonic | Continuous | Non-contact, hygienic, easy retrofit | Foam, condensation, near-field blanking zone |
| Hydrostatic probe | Continuous | Stable on turbulent tanks, simple signal | Membrane fouling, monthly rinse, quarterly recalibration |
| Electrode set | Multiple points | Several setpoints, no moving parts | Tip scaling, fixed resolution, needs conductive liquid |
A layered arrangement — one continuous sensor for control, floats for the hard limits, electrodes or a pressure switch as a second independent trip — covers every failure single point of failure the tank has.
Installing and Verifying Any Level Sensor
Placement and verification decide whether the reading can be trusted. Mount the sensor where the surface it sees is representative of the tank, not of a local return jet, and follow the same logic used for other instruments in sensor placement and where readings go wrong. Then verify against an independent reference before trusting it: dip the tank with a graduated rod at three levels — high, mid, low — and compare. On a two-metre tank, a reading within ±10 mm of the dip measurement at all three levels is a realistic acceptance criterion for a properly installed sensor.
Log the verification in the commissioning file with the date and the reference method. Six months later the drift question answers itself: re-dip, compare, and adjust. Farms that skip the baseline cannot tell sensor drift from real tank behaviour, and both look identical on a trend.
Matching the Sensor to the Duty: Fill, Alarms and Data
Choose by duty, not by datasheet. For simple refill control on a make-up tank, electrodes or a single float-plus-sensor pair do the job for years. For the main recirculating reservoir where staff watch trends, a hydrostatic probe into the farm’s control system earns its cost, especially where the signal feeds IoT and remote monitoring and alarm tiers are built on it. Ultrasonic wins on hygiene-critical or aggressive solutions where nothing wetted is preferred, provided the tank surface stays calm.

Whatever the choice, wire the hard-limit floats to interrupt the pump directly, not through the controller. Software stops pumps; hardware stops damage.
Level sensing without floats is not about replacing every float on the farm. It is about giving the tanks a continuous, loggable reading while independent hardware keeps the hard limits — and that combination is what lets a farm find a leak on the trend line instead of at the dry pump.
FAQs About Level Sensing Options
Which level sensing option is most accurate on a nutrient tank?
A properly calibrated hydrostatic probe is usually the most reliable continuous option, with realistic accuracy in the ±5–10 mm range on a two-metre tank in normal duty. Ultrasonic is comparable only where the surface stays free of foam and condensation. Always verify against a dip measurement rather than trusting datasheet figures.
Can ultrasonic sensors read through foam on the solution surface?
No. Foam scatters the echo and typically produces a false, closer distance. Either fit the sensor over a stilling zone, move the return inlet away, or switch to a hydrostatic probe, which ignores surface conditions entirely.
How often does a hydrostatic level probe need maintenance?
As a planning range, rinse the membrane monthly and recalibrate quarterly against a dip measurement, assuming standard nutrient solution and moderate biofilm. Farms with heavier algae load should shorten the rinse interval until trend data shows what the tank actually needs.
Are electrode probes suitable for nutrient solution?
Yes — nutrient solution conducts well. Their limits are tip scaling, which needs periodic cleaning, and fixed resolution: you get setpoints where the electrodes sit, never a continuous level. They suit alarms and pump control rather than trend monitoring.
Should floats be kept after fitting an electronic level sensor?
Yes. Keep an independent high-level float and a dry-run float wired to interrupt the pump directly. The sensor handles control, data and comfort; the floats handle the failures — including the sensor’s own.
Related Reading
- Reservoir Level Control and Float Valves — The mechanical side of the same tank boundary
- Sensor Placement: Where Readings Go Wrong — Why the mounting point decides the reading
- Hydroponic Sensor Calibration: EC, pH and DO Probe Maintenance — Calibration discipline that transfers to level probes
- Hydroponic Alarm Systems: Thresholds and Escalation — Turning level signals into staged alarms
- Hydroponic Reservoir Sizing and Water Storage Tank Design — The tank geometry the sensor is mounted into
Next Step: What to Send for a Quotation
Level sensors fail in integration more often than in hardware — wrong signal type, no controller input, no alarm tier. Specifying the duty alongside the tank is what prevents that.
Specifying or upgrading tank instrumentation? For RFQ or project consultation, send: tank dimensions, depth and working level; solution type, EC range and temperature; whether control, alarms, data logging or remote access is required; existing controller signal inputs (4–20 mA, digital); preferred sensor types and spares ratio; delivery schedule; and the budget band. G AND N FORTUNE LIMITED supplies tanks, matched level instrumentation and controllers quoted to those parameters.
Planning a Hydroponic Project?
Need help choosing a growing system or equipment? Tell us about your project requirements.



