Water Level and Leak Sensors for Aquariums
Float, optical, capacitive, ultrasonic and conductive level sensors compared, how leak detectors work, and the fail-safe principles that keep an auto top-off from flooding the room.
What a level sensor does in an aquarium
Evaporation removes water and leaves salts behind, so the water line in a sump or open tank falls every day. In a large reef installation described by Tropical Fish Hobbyist, purified make-up water enters one of the sumps under the control of a float kit, and the plumbing includes a dedicated emergency overflow sump so that a failure or power loss sends water to a safe place instead of the floor. The same two ideas apply to any home system: a sensor that detects the water line, and a design that cannot flood when the sensor is wrong. Level sensing is either point level (a switch at one height) or continuous (a distance or pressure reading converted to a level).
Float switches
The most common point-level device is a float carrying a permanent magnet that slides on a stem; when the float reaches the actuation height, the magnet closes or opens a reed switch sealed inside the stem. Other variants raise a rod against a microswitch or use a hinged float. Float switches are built with substantial hysteresis so a pump does not chatter as ripples pass, and two-stage versions add a second float that, if the first stage fails, switches off the source of liquid or triggers an alarm. Wikipedia lists the weaknesses: debris accumulating on the float impedes its motion, floats can stick or snag, and the sensor should not be used with liquids that adhere to the stem or float. A stilling well or guard protects the float from turbulence and from being fouled by passing objects. Float materials must suit the chemistry of the liquid; polypropylene is listed for extreme pH conditions.
Optical level switches
An optical switch shines infrared light from an LED into a prism or through a gap and watches a detector. Air and water refract and transmit the light differently, so covering the tip changes the signal that reaches the detector. There are no moving parts, but build-up on the optical surface changes the transmission and requires periodic cleaning; the Wikipedia overview names this as the main maintenance concern for optical and laser level devices.
Capacitive sensors
Capacitive sensing detects anything conductive or with a dielectric constant different from air. Water has a dielectric constant of about 88, air about 1, so a rising water line changes the capacitance of an electrode dramatically. The electrodes can be coated with insulating layers or placed behind glass or plastic, which allows a non-contact sensor stuck to the outside of a sump wall. The method's weak points are stray and parasitic capacitance, which can produce false signals and require calibration, and material building up on the probe, which shifts the reading; self-tuning designs compensate for slow build-up.
Ultrasonic sensors
An ultrasonic transducer mounted above the water emits a pulse at 20–200 kHz and measures the time until the echo returns from the surface; the distance follows from the speed of sound. It never touches the water and is unaffected by the colour or clarity of the target. Its limitations matter in a small sump: there is a blanking zone near the transducer, quoted as 150 mm to 1 m depending on the device, inside which nothing can be measured, and foam, vapour and temperature changes distort the result because they alter the echo or the speed of sound.
Conductive probes and pressure sensors
A conductive level probe passes a low-voltage current between electrodes of different lengths; when the liquid bridges them, the circuit closes. It works only in conductive liquids, so it suits saltwater better than purified fresh water, and insulating build-up on the electrodes eventually stops it working. A hydrostatic pressure sensor at the bottom of a vessel converts pressure to depth through the relation p = ρgh; it must be calibrated for each liquid, and density changes with temperature or salinity shift the reading.
Leak detectors
A water detector relies on the electrical conductivity of water to reduce the resistance between two contacts; when enough water bridges them it sounds an audible alarm and can send a signal onward to a controller or valve. Spot detectors sit at the point where water would first collect, while water-sensing cable can be laid in long runs and complex patterns around the base of a floor or along pipes; digital cable systems locate a leak to within about 1 m along the run. Detectors are typically placed near the water infrastructure that can fail: tanks, pipes, drains and pumps.
Fail-safe design for auto top-off
A fail-safe design is one that, when a component fails, responds in a way that causes minimal or no harm. For an auto top-off (ATO) pump the harmful states are a stuck-on pump and a run-dry pump, so the design must not depend on a single sensor. The general ATO set-up is covered in the guide ato-system-guide; the sensor-side rules are below.
- Use a primary level sensor for the working water line and an independent high-level sensor of a different technology to cut the pump; this mirrors the two-stage float switch principle in which the second stage switches off the source of liquid or raises an alarm.
- Add a run-time limit: a watchdog timer that stops the pump if it has run longer than a normal top-off could take. Wikipedia describes watchdog timers and control logic that detects sensor mismatches and triggers a safe shutdown as standard fail-safe elements.
- Prefer components that default to the safe state on loss of power, as normally-closed valves do by spring force.
- Contain the failure: route overflow to a place that can take the water, as in the TFH installation with its emergency overflow sump.
- Add a leak detector under the sump and reservoir so that a slow drip is reported before it becomes a flood.