Theory explainer
Weir Flow Problems — FE/PE Civil Exam Worked Examples
A weir turns a head reading into a discharge, and nearly every exam question about it is one of five moves: run the rectangular equation, shorten the crest for end contractions, switch to the 5/2 exponent for a V-notch, invert the equation to find a head (the Cipolletti case), or correct for the velocity of approach. Learn which exponent belongs to which shape, and the topic is closed.
Last reviewed: 2026-10-31. Every worked number below was independently re-solved; steps are shown in full.
Rectangular weirs: the 3/2 exponent, and the coefficient that absorbs the constants
For a sharp-crested rectangular weir, discharge is Q = (2/3)·Cd·L·√(2g)·H3/2. With the standard Cd = 0.62 this collapses to Q ≈ 3.32·L·H3/2 in US units (cfs, feet, with g = 32.2 ft/s²) and Q ≈ 1.83·L·H3/2 in SI units (m³/s, metres) — the familiar Francis value of 1.84. A broad-crested weir uses the same form with a lower coefficient (≈ 1.70 in SI, ≈ 3.0 in US units); an ogee spillway crest is higher again. The coefficient is unit-system specific: 1.84 in SI is not 1.84 in US units, and grabbing the wrong system's constant is a planted distractor. The site's culverts and spillways page already carries the basic weir equation with mini-quiz questions; this page goes deeper, into contractions, V-notches, Cipolletti inversion, and the velocity of approach.
End contractions: shorten the crest first
When a rectangular weir does not span the full channel width, the flow contracts at each exposed end, and the effective crest length shrinks: Le = L − 0.1·N·H, where N counts the contractions (two for a weir contracted at both ends, one if only one end is free, zero for a suppressed weir). The correction is small — a tenth of a head per contraction — but it is exactly the kind of detail an exam stem plants: a stem that says "two end contractions" and a solver who uses the full length will land on a distractor answer a few percent high. Note the correction uses the head H, not the crest length, and it applies to the length only; the exponent stays 3/2.
V-notch weirs: the 5/2 exponent
A triangular (V-notch) weir discharges Q = (8/15)·Cd·√(2g)·tan(θ/2)·H5/2, with H measured above the notch vertex and θ the notch angle. For the standard 90° notch with Cd = 0.60 this becomes Q ≈ 2.57·H5/2 in US units and Q ≈ 1.42·H5/2 in SI units. Two consequences follow. First, a V-notch is the right instrument for small, low flows: the 5/2 exponent makes discharge change fast with head, so a small flow still gives a readable head. Second, it punishes head errors: differentiating gives dQ/Q = 2.5·(dH/H), against 1.5·(dH/H) for a rectangular weir, so a 2% head misreading is a 5% discharge error on a V-notch but only 3% on a rectangular weir. The classic exam trap is applying the rectangular 3/2 exponent to a triangular notch, or measuring H from the channel bottom instead of the vertex.
Cipolletti weirs and inverting the equation
A Cipolletti weir is a trapezoidal weir whose side slopes are 1 horizontal to 4 vertical. That slope is chosen deliberately: the extra triangular side area almost exactly cancels the contraction loss, so no contraction correction is applied at all, and the equation is simply Q ≈ 3.37·L·H3/2 in US units (1.86 in SI). Cipolletti problems often run backwards — a required discharge and crest length are given and the head must be found — which means raising to the 2/3 power: H = (Q/(C·L))2/3. The backward direction is where unit slips bite: a head in feet raised to 3/2 gives cfs only if the coefficient matches the unit system.
Worked problem 1 Suppressed rectangular weir (FE pace)
Given: a sharp-crested weir spans the full width of a channel, crest length L = 6 ft, head over the crest H = 1.00 ft, Cd = 0.62. Find the discharge.
Solution:
- Suppressed means no end contractions: N = 0, so Le = L = 6 ft.
- Coefficient in US units: (2/3) × 0.62 × √(2 × 32.2) = 0.6667 × 0.62 × 8.025 = 3.32.
- Q = 3.32 × 6 × 1.003/2 = 19.9 cfs.
Answer: Q ≈ 19.9 cfs. Trap note: using the SI coefficient 1.84 here gives 11.0 cfs — a unit-system error, not a rounding difference. And a head measured from the channel bottom (crest height plus head) instead of from the crest would inflate the answer far more.
Worked problem 2 Contracted rectangular weir (FE pace)
Given: a rectangular sharp-crested weir, L = 8 ft, contracted at both ends, head H = 0.75 ft, Cd = 0.62. Find the discharge in cfs.
Solution:
- Effective length: Le = L − 0.1·N·H = 8 − 0.1 × 2 × 0.75 = 8 − 0.15 = 7.85 ft.
- Head term: H3/2 = 0.75 × √0.75 = 0.75 × 0.866 = 0.650.
- Q = 3.32 × 7.85 × 0.650 = 16.9 cfs.
Answer: Q ≈ 16.9 cfs. Trap note: skipping the contraction correction (using L = 8 ft) gives 17.2 cfs — close enough to look right, which is why it is a planted distractor. The correction is only 2% here, but on a shorter weir or a larger head it grows.
Worked problem 3 90-degree V-notch weir (FE pace)
Given: a 90° triangular weir carries water under a head H = 1.20 ft above the vertex, Cd = 0.60. Find the discharge in cfs.
Solution:
- Coefficient in US units: (8/15) × 0.60 × √(2 × 32.2) × tan(45°) = 0.5333 × 0.60 × 8.025 × 1 = 2.57.
- Head term: H5/2 = 1.20² × √1.20 = 1.44 × 1.095 = 1.577.
- Q = 2.57 × 1.577 = 4.05 cfs.
Answer: Q ≈ 4.05 cfs. Trap note: the rectangular 3/2 exponent on this head gives 3.38 cfs, and measuring H from the channel bottom instead of the vertex inflates the head outright. If the stem gives a different notch angle, tan(θ/2) is the only change — for 60°, tan(30°) = 0.577 scales the discharge down by that factor.
Worked problem 4 Cipolletti weir — head from a required discharge (PE pace)
Given: a Cipolletti weir with crest length L = 5 ft must pass Q = 12 cfs. Find the required head over the crest.
Solution:
- Cipolletti equation in US units: Q = 3.37·L·H3/2, with no contraction correction (the 1:4 side slopes cancel it).
- Invert: H3/2 = Q / (3.37 × L) = 12 / (3.37 × 5) = 12 / 16.84 = 0.713.
- H = 0.7132/3 = 0.80 ft. Check: 3.37 × 5 × 0.803/2 = 16.84 × 0.713 = 12.0 cfs.
Answer: H ≈ 0.80 ft. Trap note: applying a contraction correction to a Cipolletti weir double-counts the loss the side slopes already cancel — the correction would shrink the effective length and push the computed head up. The other trap is inverting with the wrong exponent: H is the 2/3 power of Q/(C·L), not the square root.
Worked problem 5 Velocity-of-approach correction (PE pace)
Given: a suppressed rectangular weir, L = 6 ft, Cd = 0.62, sits in a 12 ft wide approach channel. The crest stands 2.00 ft above the channel bed and the measured head over the crest is H = 0.90 ft. Find the discharge allowing for the velocity of approach.
Solution:
- First estimate, ignoring approach velocity: Q0 = 3.32 × 6 × 0.903/2 = 19.90 × 0.854 = 17.0 cfs.
- Approach area: depth above the bed = 2.00 + 0.90 = 2.90 ft, so A = 12 × 2.90 = 34.8 ft². Approach velocity Va = 17.0 / 34.8 = 0.488 ft/s.
- Velocity head: Va²/(2g) = 0.488² / 64.4 = 0.00370 ft. Total head = 0.90 + 0.00370 = 0.9037 ft.
- Corrected discharge: Q = 3.32 × 6 × 0.90373/2 = 19.90 × 0.859 = 17.1 cfs. A second iteration changes the result by less than 0.01 cfs, so one pass is enough.
Answer: Q ≈ 17.1 cfs. The correction here is only about 0.6% — that is the lesson. On a wide approach channel the velocity-of-approach correction is small, and an answer far from the uncorrected 17.0 cfs signals an error elsewhere (usually the approach area, computed with the head alone instead of crest height plus head). The correction only becomes material when the approach channel is narrow or the crest is low relative to the head.
Frequently asked questions
What is the discharge equation for a rectangular sharp-crested weir?
Q = (2/3)·Cd·L·√(2g)·H3/2. With Cd = 0.62 this collapses to Q ≈ 3.32·L·H3/2 in US units (cfs, feet) and Q ≈ 1.83·L·H3/2 in SI units (m³/s, metres) — the familiar Francis value of 1.84. H is the head above the crest, not above the channel bottom.
How do end contractions change a rectangular weir's discharge?
Each end contraction shortens the effective crest length: Le = L − 0.1·N·H, where N is the number of contractions (two for a weir contracted at both ends, one if only one end is contracted, zero for a suppressed weir spanning the full channel width). Use Le in place of L in the discharge equation.
What is the discharge equation for a 90-degree V-notch weir?
Q = (8/15)·Cd·√(2g)·tan(θ/2)·H5/2, with H measured above the notch vertex. For a 90° notch with Cd = 0.60 this becomes Q ≈ 2.57·H5/2 in US units and Q ≈ 1.42·H5/2 in SI units. The 5/2 exponent makes a V-notch far more sensitive to head errors than a rectangular weir.
When does the velocity of approach matter on a weir problem?
When the approach channel is narrow or shallow, so the approach velocity Va is not negligible. Replace H with the total head H + Va²/(2g), where Va = Q divided by the approach cross-sectional area, and iterate once — the first correction is usually well under 1% for a wide approach channel, which is itself a useful check that the correction was small enough to ignore.
How sensitive is weir discharge to a head measurement error?
Differentiate the discharge equation: for a rectangular weir dQ/Q = 1.5·(dH/H), so a 2% head error gives a 3% discharge error; for a V-notch weir dQ/Q = 2.5·(dH/H), so the same 2% head error gives a 5% discharge error. Head must be measured upstream, away from the drawdown near the crest.
Free FE/PE flashcards + study tips, by email
Join the Water Resources Hub email list and get the free 40 water-resources flashcards (PDF + Anki) straight away, then study tips and practice problems by email. Email address only — no spam, unsubscribe any time.