Capacitive vs float level sensors

A float level sensor measures liquid level through physical movement of a float, while a capacitive level sensor detects changes in capacitance around its sensing electrodes. A float therefore relies on a mechanical measurement mechanism, while capacitive sensing does not require moving parts and can provide continuous output.

The cheapest level sensor can become the most expensive one to maintain. A float sensor has an obvious attraction. The technology is mature, inexpensive and familiar, and for an accessible tank containing a clean liquid, that can be exactly what the application needs.

Capacitive sensing starts from a different engineering position. There is no moving float, the output can be continuous, and the sensing arrangement does not have to follow the liquid surface mechanically.

The interesting comparison therefore begins after purchase price. What happens when the tank moves, the liquid becomes difficult, the output needs to be continuous or the sensor becomes expensive to reach? That is where the economics and the engineering start to diverge.

A float is simple because the liquid does the moving

Float and arm

A float moves with the liquid and drives a potentiometer through an arm mechanism to produce the level output.

Float and reed switches

A float or series of floats moves along a vertical rod containing reed switches. As the float passes each switch, the output changes between defined positions.

There is nothing wrong with that simplicity. A mature, inexpensive technology can be exactly the correct choice when the tank is accessible, the liquid is clean, the environment is benign, discrete or approximate level information is sufficient and replacement is straightforward. The question is what happens when those assumptions stop being true.

A float’s weakness is also its mechanism

The float has to move. That means the measurement depends on free movement, mechanical geometry, pivots or sliding arrangements, switch operation or potentiometer movement, and the condition of the surrounding liquid.

In dirty liquids, contamination can interfere with mechanical movement. In an arm arrangement, the potentiometer may also sit inside the tank and be exposed to the liquid environment. The important failure mode is not that the sensor stops. It is that the sensor can continue to produce a plausible output after the mechanism has been affected. That does not happen in every float design, but it is the behaviour worth designing against.

Compare the mechanism, not the marketing description

Continuous level changes the control question

A reed-switch float design changes state when the float reaches defined switch positions. That can be entirely adequate when the application only needs thresholds or approximate level.

A capacitive sensor can provide continuous level information. That matters when the output feeds a gauge, a control system, a monitoring system, a volume calculation or a continuously changing display. Continuous output is not always necessary. It is an application requirement rather than a virtue.

Where the electronics live is part of the reliability calculation

In the Gill capacitive arrangement, the electronics stay outside the liquid container. An arm-mounted float design may instead place its potentiometer inside the tank. That creates different exposure to corrosion, contamination, liquid ingress and service access. Not every float has immersed electronics and not every capacitive sensor keeps them outside, so this comparison relates to the arrangements described here.

A moving tank exposes the difference

A float responds physically to the liquid surface. If the tank moves or the liquid sloshes, the float follows that movement.

A capacitive probe has nothing riding the surface. It measures along its whole immersed length, so ripples and moderate slosh are averaged rather than followed. Heavy agitation is different: air drawn into the liquid changes its effective dielectric, and that can affect the reading. Capacitive sensing is moderately tolerant of agitation, not immune to it. The tank, liquid behaviour, probe installation and required response still need to be considered.

When the liquid gets difficult, the mechanism becomes part of the specification

Dirty, viscous, corrosive or conductive liquids change the comparison. A float introduces a physical moving mechanism into that environment. A capacitive sensor has no moving part, but standard capacitive designs are not automatically suitable for every difficult liquid.

This is where application-specific construction matters. LevelSlick uses a coated probe for challenging conductive liquids. The blackwater and wastewater article covers what that looks like in practice.

The sensing principle is only half the specification. The way the sensor meets the liquid matters too.

Purchase price is only the first number

A float sensor usually costs less to buy. That is a genuine advantage. But the purchase price is only one part of the economics.

If a failed sensor requires draining a tank, a permit, confined-space entry, a vessel visit, seasonal access or specialist labour, the replacement cost can be much larger than the component price.

The expensive part may not be the sensor. It may be getting to it.

This is where a higher initial purchase price can become commercially sensible. It does not follow that capacitive sensing always has a lower whole-life cost. It can produce a lower whole-life cost where access and replacement are expensive.

The calculation should include the visit

A meaningful comparison therefore asks what one failure costs here, rather than which sensor costs less to buy.

An inaccessible sensor changes the economics

• Vessel tanks. Access may require a vessel visit or a planned maintenance window.
• Buried or enclosed infrastructure. Access may involve permits, confined-space procedures or isolation.
• Rural or unmanned sites. A technician visit can dominate the cost of a low-value component.
• Seasonal equipment. A failure during a narrow operating window can have consequences beyond the repair itself.

These are examples of cost mechanisms rather than claims that every site has these conditions.

Ask how each sensor fails

Float

Contamination, wear or restricted movement → float position no longer represents actual level → output can remain plausible

Capacitive

Incorrect liquid assumption, unsuitable medium, calibration change or installation issue → capacitance no longer maps correctly to the intended level

Every technology has a failure mode. The useful comparison is not which sensor cannot fail. It is which failure mode is acceptable in this installation, and how the system will detect it.

A plausible wrong number is more expensive than an obvious failure

A sensor that stops reporting may trigger an alarm and prompt investigation. A sensor that continues reporting a believable but incorrect level can remain in service far longer. For either technology, the system therefore needs to consider diagnostics where available, alarm thresholds, plausibility checks, maintenance strategy and expected failure modes. Neither technology inherently produces more false readings. This is measurement-system design.

So which technology earns its place?

Neither technology wins in every application. The correct question is which failure mode, output requirement and access cost the installation can actually afford.

Choose around the installation, not the catalogue

How the comparison looks across Gill’s level sensors

General-purpose level measurement. LevelPro is a robust stainless-steel configuration with sealed construction and analogue and digital output options.

Weight-critical applications. LevelLite uses aluminium construction and is designed around low mass, with voltage and current outputs, RS232 for configuration and CAN on bespoke sensors.

Difficult conductive liquids. LevelSlick uses a coated probe and can be configured for lengths up to 2 m, with tank profiling and full and empty alarms.

These are not three versions of the same sensor. They illustrate the principle of this article: the application changes the physical design around the measurement.

Five questions before you choose

1. What does the liquid do to the mechanism? Clean liquid and sludge are very different environments.
2. Does the application need continuous data? If the system needs a smooth live value, a stepped output may be insufficient.
3. Does the tank move? Consider how the measurement responds to slosh and dynamic conditions.
4. How expensive is access? Include labour, permits, isolation, vessel access and downtime where applicable.
5. What does a plausible false reading cost? This may be more important than the sensor purchase price.

So when is capacitive worth the extra cost?

Capacitive sensing can justify a higher purchase price when the application benefits from continuous measurement, no moving mechanical mechanism, reduced exposure of electronics to the liquid, suitability for a difficult medium or tank profiling, particularly where access for maintenance or replacement is expensive.

A float remains a sensible choice when the tank is accessible, the liquid is clean, the required output is simple and replacement is inexpensive.

The sensor price is easy to compare. The cost of reaching it is where the decision gets interesting.