Self-acting Weir Mechanism Explained: How It Works, Parts, Formula and Uses in Canals

← Back to Engineering Library

A self-acting weir is a hydraulic crest structure that automatically raises or lowers itself to hold the upstream water level nearly constant as inflow varies. The first practical design was patented by French engineer Joseph Bégout in 1886 for irrigation canals in southern France. A float or counterweight senses the upstream head and rotates a hinged flap so discharge over the crest matches incoming flow. The result: a canal pool sits within ±20 mm of target level across a 10:1 flow range with no operator and no electrical power.

Self-Acting Weir Interactive Calculator

Vary the counterweight arm movement and calibration sensitivity to see the expected regulated upstream level change and gate response.

Low Rise
--
Nominal Rise
--
High Rise
--
Tuning Band
--

Equation Used

Delta h = s * Delta x; typical range: Delta h = 0.3 to 0.5 * Delta x

The article notes that moving the counterweight 10 mm outward on a typical Begout-style self-acting weir arm raises the regulated upstream level by roughly 3 to 5 mm. This calculator applies that local calibration slope as Delta h = s Delta x and reports the low, nominal, and high expected level change.

  • Positive counterweight movement means moving the weight outward on the arm.
  • Calibration is treated as linear for small field adjustments.
  • Typical Begout-style arm sensitivity is 3 to 5 mm level rise per 10 mm weight movement.
Self-Acting Weir Cross-Section A side-view engineering diagram showing how a self-acting weir automatically regulates upstream water level through moment balance between water pressure on a hinged flap and a counterweight. Self-Acting Weir Upstream Pool Hinged Flap Pivot Axis Counterweight Water Pressure Overflow Restoring Moment Opening Moment Downstream
Self-Acting Weir Cross-Section.

The Self-acting Weir in Action

The mechanism balances two things — the moment created by water pushing on the upstream face of a hinged flap, and the moment of a counterweight (or buoyancy of a float) acting through a linkage. When inflow rises, the upstream level wants to climb. That extra head increases pressure on the flap, the moment balance breaks, and the flap rotates downward so the crest drops a few millimetres. More water spills, level stabilises. When inflow drops, the counterweight wins, the flap rotates back up, the crest rises, less water spills, level holds. No electricity, no PLC, no operator standing on the embankment with a wheel.

The geometry is the part most people get wrong. The hinge axis must sit slightly upstream of the flap's centre of pressure at the design head — typically 0.35 to 0.40 of the flap height below the crest at full pool. If you place the hinge dead-centre or downstream of the centre of pressure, the flap becomes unstable and slams open. If you place it too far upstream, you get a sluggish response and a 50-100 mm dead band on level. The counterweight arm length, the flap mass, and the float displacement all tune together — change one, you re-tune all three.

Tolerances on the seals and pivot bushings matter more than the casting finish. A worn bronze pivot bushing with 1.5 mm of radial slop will let the flap chatter under any wave action, and that chatter pumps grit into the seat which then leaks. The most common failure modes we see in irrigation canal regulators are: pivot wear from un-greased shafts, rubber seal hardening after 8-10 years of UV exposure, and counterweight cables stretching or corroding so the calibrated balance drifts. A self-acting weir that's drifted out of calibration usually shows up as upstream level creeping high during low-flow periods — that's the float lagging.

Key Components

  • Hinged crest flap: The moving plate that forms the weir crest. Typically 6-12 mm steel or stainless plate, hinged along a horizontal axis at the bottom. Its upstream face is the pressure-sensing surface. Flatness across the seat must be within 1 mm over a 2 m span or you get persistent leakage at low head.
  • Pivot shaft and bushings: Stainless shaft (usually 304 or 316) running in bronze or self-lubricating polymer bushings. Radial clearance of 0.05-0.15 mm is the target. Over 0.5 mm of wear and the flap chatters under wave action, which destroys the rubber seat.
  • Float and float well: Sealed cylindrical float (often 200-400 mm diameter) hanging in a stilling well connected to the upstream pool via a damping orifice. The orifice is sized so wave noise is filtered but real level changes pass through within 5-15 seconds.
  • Counterweight and lever arm: Cast-iron weight on an adjustable arm. The arm length sets the target upstream level. Move the weight 10 mm out on a typical Bégout-style arm and you raise the regulated level by roughly 3-5 mm — this is the field calibration adjustment.
  • Rubber seat and side seals: EPDM or nitrile profile bonded to the downstream sill and side walls. Compression set after UV exposure is the killer here — replace at 8-10 year intervals or you'll see steady leakage that makes the regulator hunt.

Who Uses the Self-acting Weir

Self-acting weirs earn their keep wherever you need a stable upstream pool, you don't have grid power at the site, and the flow varies enough that a fixed weir would either spill too much or drown out. They show up in irrigation networks, small hydropower headponds, urban drainage retention, and fish-passage approach channels. The design is mechanical, gravity-powered, and once tuned it runs for decades — the Canal de Provence in southern France still operates Bégout-pattern regulators installed in the 1960s.

  • Irrigation: Canal de Provence distribution network, France — Bégout-pattern self-acting regulators hold offtake pool levels within ±25 mm across 1500 km of canal.
  • Small hydropower: Headpond level control on run-of-river plants like the Rheinfelden auxiliary intakes, where flap weirs maintain turbine head as river stage swings.
  • Urban drainage: Stormwater retention pond outlet control at Hafencity Hamburg, where a counterweighted flap holds detention level until a storm event lifts it.
  • Fish passage: Approach-channel level control upstream of Denil fish ladders on Scottish salmon rivers — keeps ladder entrance head within the 100-300 mm band the fish need.
  • Aquaculture: Pond drain regulation at trout farms in the Jura — self-acting flap holds rearing pond depth constant as makeup flow varies with spring discharge.
  • Water supply: Raw-water intake forebay level control at small treatment plants drawing from gravity-fed mountain streams, holding a stable head on the coarse screens.

The Formula Behind the Self-acting Weir

The discharge over a self-acting weir follows the standard sharp-crested weir equation, but the head H is what the mechanism actively regulates — so the practical question is how much the flap must rotate to absorb a change in inflow. At the low end of the typical operating range, a small inflow change demands a large angular rotation because the spillway length is short. At nominal design head you sit in the linear sweet spot where a 10% inflow change moves the flap roughly 2-4°. Push past about 1.3× design head and the flap approaches its mechanical stop, the regulator saturates, and upstream level starts climbing freely. Knowing where you sit on this curve tells you whether your regulator has headroom or is about to lose control.

Q = Cd × (2/3) × √(2 × g) × L × H3/2

Variables

Symbol Meaning Unit (SI) Unit (Imperial)
Q Discharge over the weir crest m³/s ft³/s
Cd Discharge coefficient (typically 0.60-0.65 for sharp-crested flap) dimensionless dimensionless
g Gravitational acceleration 9.81 m/s² 32.2 ft/s²
L Effective crest length m ft
H Head over the crest (regulated by the flap) m ft

Worked Example: Self-acting Weir in a vineyard irrigation canal regulator

You are sizing a self-acting flap weir for an offtake regulator on a 1.8 m wide concrete-lined irrigation canal feeding a 60 ha vineyard outside Mendoza, Argentina. The canal serves a rotating water-rights schedule where supply flow swings between 0.08 m³/s overnight and 0.45 m³/s at peak afternoon demand. The downstream irrigation manifold needs the upstream pool held at 0.55 m ± 30 mm so the gravity-fed drip headers see consistent pressure. Crest length is 1.6 m effective, discharge coefficient Cd = 0.62.

Given

  • L = 1.6 m
  • Cd = 0.62 —
  • Qlow = 0.08 m³/s
  • Qnom = 0.25 m³/s
  • Qhigh = 0.45 m³/s

Solution

Step 1 — rearrange the weir equation to solve for the head H the flap must regulate to at nominal flow of 0.25 m³/s:

H = (Q / (Cd × (2/3) × √(2g) × L))2/3

Step 2 — plug nominal numbers. The constant (2/3) × √(2 × 9.81) = 2.953:

Hnom = (0.25 / (0.62 × 2.953 × 1.6))2/3 = (0.0854)2/3 ≈ 0.197 m

So at design flow the flap sits low enough that 197 mm of head spills over the crest. With the target pool at 0.55 m, the flap top must sit at 0.55 − 0.197 = 0.353 m above the canal invert. This is the sweet spot — flap rotated mid-travel, response linear, level rock-steady.

Step 3 — at the low end of the operating range, 0.08 m³/s overnight:

Hlow = (0.08 / (0.62 × 2.953 × 1.6))2/3 = (0.0273)2/3 ≈ 0.092 m

The flap rotates upward so only 92 mm of head spills. To hold the same 0.55 m upstream pool, the flap top now sits at 0.458 m. Real-world feel: in the cool of early morning the flap is nearly upright and only a thin sheet of water glides over — 30 mm thick maybe. Looks asleep. It's working perfectly.

Step 4 — at the high end, 0.45 m³/s peak afternoon demand:

Hhigh = (0.45 / (0.62 × 2.953 × 1.6))2/3 = (0.1537)2/3 ≈ 0.288 m

The flap rotates down so 288 mm of head spills. Flap top now at 0.55 − 0.288 = 0.262 m. You can hear it from 30 m away — a heavy curtain of water roaring over the crest. The flap is approaching the lower mechanical stop; if peak demand jumped another 30% the regulator would saturate and upstream level would start climbing above 0.55 m.

Result

At nominal 0. 25 m³/s the flap regulates to 0.197 m of crest head, holding the upstream pool at the design 0.55 m. The full operating range spans 92 mm of head at 0.08 m³/s through 288 mm at 0.45 m³/s — the flap rotates roughly 18-22° between these two extremes, and the level-control sweet spot sits firmly in the middle of that arc. If your measured upstream level drifts more than the design ±30 mm, the most likely causes are: (1) a partially blocked float-well damping orifice — silt builds up in 6-12 months in unfiltered canal water and makes the float lag inflow changes by 30+ seconds, (2) counterweight cable stretch on older units, where 5-10 mm of stretch shifts the calibrated set-point upward, or (3) a swollen rubber seat after rainy season, which adds friction and creates a stick-slip dead band of 15-25 mm before the flap breaks free.

Choosing the Self-acting Weir: Pros and Cons

The honest comparison is against the two structures a designer typically considers in the same slot — a fixed sharp-crested weir, and a motorised gate with a level sensor and PLC. Each wins on different dimensions.

Property Self-acting weir Fixed sharp-crested weir Motorised gate with PLC
Upstream level accuracy across 10:1 flow range ±20-30 mm ±150-300 mm (varies with Q) ±5-10 mm
Power requirement None — gravity only None Mains or solar + battery
Initial cost (1-2 m crest) USD 6,000-15,000 USD 1,500-4,000 USD 18,000-40,000
Maintenance interval Annual greasing, seal replacement at 8-10 yr Visual inspection only Quarterly — actuator, sensor, electronics
Service life 30-50 years (Canal de Provence units still running) 50+ years 15-25 years before electronics obsolescence
Response time to inflow change 5-30 seconds (float-damped) Instant but uncontrolled level rise 1-5 seconds programmable
Failure mode Drifts out of calibration, level creeps Floods or drys upstream pool Stops moving — full open or full closed
Best application fit Remote canals, no grid, stable demand profile Constant inflow systems Tight-tolerance industrial water systems

Frequently Asked Questions About Self-acting Weir

Hunting almost always traces to the float-well damping orifice being too large for the flap inertia, or to the counterweight being too heavy relative to the float buoyancy. The system is under-damped — every small disturbance kicks off a resonance the regulator cannot bleed off.

Quick diagnostic: throttle the orifice with a temporary plug or smaller bushing and watch what happens. If the oscillation period is roughly 8-15 seconds, you need more damping. If it's under 3 seconds, the issue is mechanical — usually loose linkage pins or a worn pivot shaft letting the flap chatter. Replace bushings before you re-tune the orifice.

Place the hinge so that at design head, the centre of pressure on the upstream face sits 5-15 mm above the hinge axis. For a vertical flap with water depth h on the upstream face, centre of pressure is at h/3 from the bottom — so the hinge typically lands at 0.30-0.40 of flap height below the crest at full pool.

Get this wrong by more than ±20 mm and you either get an unstable flap that wants to slam open (hinge too low, centre of pressure below hinge), or a sluggish flap with a wide dead band (hinge too high). It's the single most important geometric decision in the design.

Rule of thumb from canal practice: if you need upstream level held tighter than ±15 mm, or if the flow range exceeds 20:1, the self-acting design runs out of headroom and a motorised gate with PLC starts to win. Below 10:1 flow range and ±30 mm tolerance, the self-acting weir is cheaper over a 25-year life cycle by a factor of 3-5×.

The other deciding factor is grid access. A site needing solar plus battery plus actuator plus controller plus comms plus annual electronics service almost never beats a tuned mechanical regulator that just sits there for decades.

Two thermal effects stack up. First, the rubber seat swells slightly in warm water and adds 3-8 mm of static friction to the flap rotation, creating a small offset before the flap breaks free. Second, if the float is air-filled, internal air pressure rises with temperature — a sealed float at 35°C versus 5°C will sit roughly 1-2% lower in the water, shifting the calibration set-point upward by 5-10 mm on a typical arm.

Fix: use a vented float (small breather hole at the top) and accept the seasonal seal-friction offset, or recalibrate the counterweight twice a year if you need tight performance year-round.

Usually yes, but the upstream pool geometry decides whether it's worth doing. You need at least 1.5× flap height of clear water depth upstream so the flow approaches the crest without contraction effects skewing the discharge coefficient. You also need a stilling well location — typically a 200-300 mm diameter vertical pipe set into the upstream wall, connected to the float chamber via a damping orifice.

If the existing structure is masonry or unreinforced concrete, the hinge anchor loads (often 5-15 kN at peak head on a 2 m flap) frequently exceed what the original wall can carry without a stainless backing plate or chemical anchors. Check this before quoting the job.

Stuck closed (flap up, no spill) on a unit that previously worked is almost always debris jammed under the rubber seat or in the side seal channels — branches, plastic, weed mat. The flap can't rotate down to spill, so upstream level keeps climbing until water tops the embankment or finds an emergency overflow.

Stuck open (flap down, full spill) is usually pivot seizure from corrosion in the bushing, or a snapped counterweight cable. Check the cable termination first — galvanised cable in chlorinated or acidic water typically fails at the swaged fitting after 6-10 years. A stainless cable with bolted clamps lasts the life of the regulator.

References & Further Reading

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: