Float Switch Mechanism: How It Works, Diagram, Parts, Uses & Cycle Rate Formula Explained

← Back to Engineering Library

A float switch is a mechanical level sensor that uses a buoyant float to actuate an electrical contact when liquid rises or falls past a set point. Unlike a continuous level transmitter that outputs an analogue 4–20 mA signal, a float switch only reports two states — wet or dry — which makes it cheaper, simpler, and far more reliable for on/off pump control. It exists to start and stop pumps, trigger alarms, and prevent overflow or dry-run damage automatically. A single $15 tethered float reliably manages a domestic sump pump for 10+ years across millions of cycles.

Float Switch Interactive Calculator

Vary sump diameter, float differential, and inflow rates to see the differential volume and pump cycle rate.

Diff. Volume
--
Peak Cycles
--
Nominal Cycles
--
Low Cycles
--

Equation Used

Vdiff = pi*(Dpit/2)^2*hdiff; fcycle = Qin / Vdiff

The differential volume is the liquid volume between the float switch pump-on and pump-off levels. The cycle rate is the inflow rate divided by that volume, so a larger sump diameter or larger float differential reduces pump starts per hour.

  • Cylindrical sump with vertical sides.
  • Float differential is the vertical distance between pump-on and pump-off levels.
  • Cycle rate follows the article simplification using inflow divided by differential volume.
  • Inflow is treated as steady for each demand case.
Watch the Float Switch in motion
Video: Scissor-switch keyboard mechanism by Nguyen Duc Thang (thang010146) on YouTube. Used here to complement the diagram below.
Float Switch Mechanism Diagram Animated cross-section showing how a magnetic reed float switch converts liquid level into an electrical signal. Magnetic Reed Float Switch N S Level Electrical Circuit Power Reed Load Status: Glows when reed closes Guide Rod Reed Switch Float (Buoyant) Magnet Liquid Level ~1.5mm gap Key Advantage No electrical contacts in wet zone Magnet actuates through sealed wall Buoyancy Liquid rises → Float rises → Magnet passes reed → Circuit closes Reliable sealed design with no wet contacts
Float Switch Mechanism Diagram.

The Float Switch in Action

The mechanism is buoyancy doing electrical work. A sealed float — usually polypropylene or stainless — rides on the liquid surface. As the level changes, the float either tilts on a tether, slides up a guide rod, or rotates a lever arm. That mechanical motion closes or opens a contact: a mercury tilt capsule in older designs, a microswitch in mid-grade industrial units, or a reed switch triggered by a magnet embedded in the float in modern sealed types. The reed switch float is the dominant design today because it has no moving electrical contacts in the wet zone — the magnet passes a glass-encapsulated reed through the stem wall, and the reed snaps closed.

The geometry matters more than people expect. On a tethered float switch — the kind hanging in your sump pit — the cable length sets the differential between pump-on and pump-off levels. Too short and the pump short-cycles, burning out the motor in months. Too long and the pit drains below the pump intake, sucking air and cavitating. A typical sump tether is 75 to 150 mm of free cable; if you measure your differential and it falls outside that band, shorten or extend the tether before you blame the pump. Vertical float switches with a magnetic reed need the float-to-stem clearance held to about 1.5 mm — tighter and the float binds on debris, looser and the magnet field weakens enough that the reed misses actuation on a slow rise.

Failure modes are predictable. Tethered floats fail when the pit grows a biofilm or rag mat that snags the cable. Reed switch floats fail when ferrous debris collects on the magnet, holding the reed closed permanently — a stuck-on pump runs dry. Mercury tilt floats fail by capsule cracking after roughly 250,000 cycles, and they're banned in many jurisdictions anyway. If the pump runs but the alarm never trips, suspect a stuck float before you suspect the controller.

Key Components

  • Float Body: Sealed buoyant chamber, typically 30–60 mm diameter polypropylene or 316 stainless. Specific gravity sits between 0.5 and 0.7 so it rides high on water but still works in lower-density fluids like diesel (SG 0.84). Wall thickness must survive hydrostatic pressure at the deepest setpoint — 1.5 mm minimum for sump duty.
  • Pivot or Guide: Either a tether anchor for cable-suspended floats, or a guide rod for vertical reed-switch types. Guide rod straightness must hold within 0.5 mm over the travel length or the float drags. The pivot bushing on lever-arm floats sees roughly 1 cycle per pump start — 100,000 cycles is the baseline life expectation.
  • Reed Switch or Microswitch: The actual electrical contact. Reed switches handle 0.5 A at 120 VAC directly or up to 10 A through a relay. Microswitches handle pump motors directly up to about 13 A at 240 VAC. Reed contacts weld closed if you switch an inductive load without a snubber — a 0.1 µF / 100 Ω RC snubber across the contact extends life by 10×.
  • Magnet (reed-switch types): Embedded in the float, usually a 5–10 mm ferrite or neodymium ring. Field strength at the reed must exceed roughly 20 gauss for reliable actuation. Ferrous swarf in the liquid sticks to the magnet over time and is the single most common cause of stuck-on failures in machine-tool coolant tanks.
  • Cable and Strain Relief: PVC or neoprene jacket rated for the liquid. Strain relief at the float must hold the cable without letting water wick into the conductor — a single failed strain relief floods the float and sinks it, leaving the pump running dry until the thermal cutout trips.

Real-World Applications of the Float Switch

Float switches show up anywhere a tank, sump, or vessel needs simple on/off level control. The reason is economics — a 4–20 mA radar level transmitter costs $800 and needs a PLC to interpret it. A float switch costs $15 to $200 and wires directly to a pump contactor. For applications where you only need to know two things — is the level above the start point, is it above the high-high alarm — the float wins on reliability, cost, and serviceability. The catch is that you cannot trend the level, so condition-based maintenance on the pump itself becomes harder. That's why municipal lift stations still use floats for primary control but pair them with an ultrasonic transmitter for SCADA logging.

  • Residential Plumbing: Liberty Pumps 257 sump pump uses a vertical magnetic float switch with a 75 mm differential to drain basement pits at 2,500 gph.
  • Wastewater Treatment: Municipal lift stations using SJE-Rhombus tethered floats for redundant pump-down, pump-up, and high-level alarm in a 4-float configuration.
  • Marine Bilge Systems: Rule-Mate 1500 automatic bilge pumps with an integrated reed-switch float that trips at 50 mm of standing water in the bilge.
  • Industrial Coolant Management: Haas VF-2 CNC mill coolant tanks using a Gems LS-1750 vertical float switch to alarm low coolant before the spindle runs dry.
  • Chemical Storage: 316 stainless lever-arm float switches on Snyder Industries dual-wall HCl day tanks, triggering a leak alarm when the interstitial space shows liquid.
  • Steam Boilers: McDonnell & Miller 67 low-water cutoff using a magnetic float to shut off the burner when boiler water drops below the safe level.

The Formula Behind the Float Switch

The single most important number to compute on a float switch installation is the pump cycle rate — how many times per hour the pump starts and stops based on inflow rate and the volume between the on and off levels. At the low end of typical residential inflow (around 5 L/min during light rain), the pump cycles once every 20 minutes and the motor lives a long life. At the high end, during a heavy storm event with 40 L/min inflow, that same setup cycles every 2.5 minutes and you're racing the duty cycle of the motor. The sweet spot is 6–10 cycles per hour at peak inflow — fewer than that means an oversized pump, more than that means short-cycling and premature winding failure.

fcycle = Qin / Vdiff

Variables

Symbol Meaning Unit (SI) Unit (Imperial)
fcycle Pump cycles per hour at the given inflow cycles/hour cycles/hour
Qin Inflow rate to the sump or tank L/hour gph
Vdiff Volume of liquid between pump-on and pump-off levels (set by float differential and pit cross-section) L gallons

Worked Example: Float Switch in a craft cidery in herefordshire

A craft cidery in Herefordshire is sizing the float differential on a 600 mm diameter underground sump that collects floor washdown and CIP rinse from the bottling line. They've installed a 1/3 hp submersible with a tethered float and want to confirm the cycle rate stays inside the motor's duty rating. Pit cross-section is 0.283 m². The tether currently gives a 100 mm differential, so Vdiff = 0.0283 m³ = 28.3 L. Inflow during a CIP cycle peaks at 1,200 L/hour.

Given

  • Dpit = 600 mm
  • hdiff = 100 mm
  • Vdiff = 28.3 L
  • Qin,peak = 1200 L/hour
  • Qin,nominal = 600 L/hour
  • Qin,low = 150 L/hour

Solution

Step 1 — at the nominal washdown inflow of 600 L/hour, calculate cycles per hour:

fnom = 600 / 28.3 = 21.2 cycles/hour

That's already above the 6–10 cycles/hour sweet spot. The pump will start every 2.8 minutes during normal washdown — manageable but hard on the start capacitor.

Step 2 — at the low end (150 L/hour, light drip from a leaking valve between cycles):

flow = 150 / 28.3 = 5.3 cycles/hour

One cycle every 11 minutes. The motor barely notices this duty. If you only ever saw this load, the pump would last 20+ years.

Step 3 — at the high end (1,200 L/hour during a full CIP rinse):

fhigh = 1200 / 28.3 = 42.4 cycles/hour

That's a start every 85 seconds. Most 1/3 hp submersibles are rated for 20 starts per hour maximum. You will burn the start winding inside a season at this rate.

Step 4 — solve for the required differential to hit 10 cycles/hour at peak inflow:

Vdiff,req = 1200 / 10 = 120 L → hdiff = 120 / 283 = 0.424 m = 424 mm

Result

At nominal 600 L/hour inflow with the current 100 mm tether differential, the pump cycles 21 times per hour. At the low-end drip rate of 150 L/hour the cycle rate falls to a comfortable 5.3/hour, but at the peak CIP rinse of 1,200 L/hour it rockets to 42 cycles/hour — twice what the motor is rated for. The fix is lengthening the tether to give roughly 425 mm of differential, dropping peak cycling to 10/hour. If the operator measures cycle rates higher than predicted after lengthening the tether, three failure modes are likely: (1) the pit has a hidden infiltration leak from groundwater that adds baseline inflow, (2) the float is hanging up on a pit-wall protrusion and toggling on partial travel, or (3) the check valve on the discharge has failed and water is draining back into the pit after each pump-off, artificially inflating the apparent inflow.

Choosing the Float Switch: Pros and Cons

Float switches compete with three other level-sensing technologies for on/off pump control: conductive probes, capacitive switches, and ultrasonic sensors. The choice comes down to liquid type, debris content, accuracy requirement, and budget. Here's how they stack up on the dimensions that matter for selection.

Property Float Switch Conductive Probe Ultrasonic Sensor
Unit cost (industrial grade) $15–$200 $50–$300 $400–$1,200
Switching accuracy / repeatability ±5–15 mm ±2 mm ±1 mm or better
Works in non-conductive liquids (oil, diesel) Yes No Yes
Tolerance to debris and biofilm Poor — float fouls Poor — probe coats Excellent — non-contact
Typical service life 100,000–1,000,000 cycles 2–5 years before probe replacement 10+ years
Power required None (passive contact) 12–24 VDC controller 12–24 VDC, 50 mA
Best application fit Sumps, bilges, tanks with clean-ish water Conductive water with stable chemistry Hostile chemistry, foam, debris, high accuracy
Failure mode complexity Simple — stuck or sunk float Probe fouling, scale buildup Foam blanking, condensation on transducer

Frequently Asked Questions About Float Switch

The tether is too long. A long tether gives a large differential between on and off levels, and if the pit cross-section is small the float keeps falling until it hits the pit floor. The pump then runs until it sucks air and the thermal cutout trips. Shorten the tether so the float-off level sits at least 50 mm above the pump intake — most submersibles need 75–100 mm of water over the volute to avoid cavitation.

Quick check: with the pit empty, lift the float by hand and watch where it naturally hangs. That's your pump-off level. If it's below the pump body, your tether is wrong.

Diesel has a specific gravity of about 0.84 versus water's 1.0. A float designed for water typically has an effective SG of 0.6–0.7, which gives plenty of buoyancy in water but only marginal buoyancy in diesel. Add any wall thickness margin from manufacturing tolerance and the float can sit submerged or only partially rise.

For hydrocarbons, you need a float specifically rated for the fluid — manufacturers like Gems Sensors publish minimum SG ratings on every model. Pick one rated for SG 0.7 or lower and verify the float datasheet shows your fluid by name.

Always two separate floats for any installation where overflow has consequences. A single float that's stuck on or stuck off has no backup — you only find out when water hits the ceiling or the pump runs dry. Two independent floats at different levels with separate cable runs and separate controller inputs give true redundancy.

Standard practice in municipal lift stations is four floats: pump-off, pump-on, lag-pump-on (for duplex systems), and high-level alarm. Each is wired to a separate channel. If one fouls, the others still function. The incremental cost is $30–$100 per float; the cost of a basement flood or a sewage overflow is several orders of magnitude higher.

Most likely ferrous debris on the magnet. Reed switch floats use a permanent magnet inside the float to actuate a glass-encapsulated reed in the stem. Iron swarf, rust flakes, or steel grinding dust collects on the magnet and either holds the reed closed or pulls the float off-axis so it can't slide.

Pull the float off the stem and wipe it clean with a rag — you'll usually see a dark fuzz of ferrous particles. For installations with any ferrous debris (machine-tool coolant tanks are notorious), spec a float with the magnet shielded by a stainless cap, or switch to a non-magnetic conductive sensor.

Target 6–10 cycles per hour at your peak expected inflow. Below 6, the pump is oversized and you're paying for capacity you don't use. Above 10, you're stressing the start winding and capacitor on every cycle — most fractional-horsepower submersibles are rated for 15–20 starts per hour absolute maximum.

Calculate the required volume between on and off levels as Vdiff = Qpeak / 10. Then divide by the pit cross-section to get the height differential. If the resulting height is taller than your pit, you need a bigger pit or a larger pump that empties the differential faster — not a longer tether.

Depends entirely on the float's contact rating. Heavy-duty mechanical microswitch floats from Square D or SJE-Rhombus carry 13 A at 240 VAC and switch sub-1-hp pump motors directly. Reed-switch floats almost never do — most reed contacts are rated 0.5–1 A and will weld shut on the inrush current of a motor start (which can be 6× running current for the first few cycles).

Rule of thumb: if the float datasheet shows a reed switch or quotes a contact rating under 5 A, wire it through a contactor. The contactor coil draws milliamps and the float lasts decades. Wire a reed switch directly to a pump and it lasts weeks.

The lag float cable has drifted into the lead float's swing arc, and they're tangling. Tethered floats need physical separation — at least 200 mm between mounting points and clear swing radii that don't overlap. As the cables flex over thousands of cycles, the geometry shifts and they eventually intersect.

Fix: mount each float on a rigid bracket clamped to a separate vertical support pipe, not zip-tied to the discharge pipe of the pump. The bracket fixes the swing centre and the floats can never drift into each other's path. This is the single most common installation error in field-built duplex stations.

References & Further Reading

  • Wikipedia contributors. Float switch. Wikipedia

Building or designing a mechanism like this?

Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.

← Back to Mechanisms Index
Share This Article
Tags: