top of page

Rectangular-Weir Head Readings: Check the Crest Datum Before Calculating Flow

Oct 27
4 min read
AI-generated illustrative scene: water passing over a rectangular laboratory weir. Not product or project evidence.

AI-generated illustrative scene, not a product screenshot or a completed-project photograph. The worked calculation below uses explicit example inputs.


A water-level reading is not automatically the head required by a weir equation. The calculation needs the vertical difference between the upstream water surface at the specified measuring station and the correct crest datum. A staff gauge can produce beautifully repeatable numbers while its zero remains displaced from that datum.


This worked example is a sensitivity exercise for water-measurement review. It is not a weir design, an overflow-capacity assessment or authorization to modify a channel. The applicable geometry, installation conditions and rating must be established by the responsible engineer before a measured head can be converted into defensible flow.


Define the rating and its units


The US Bureau of Reclamation's Water Measurement Manual describes the standard suppressed rectangular weir and the Francis relationship Q = 3.33 L H^(3/2), using Q in ft³/s and L and H in feet. The coefficient belongs to that unit system and rating. Do not enter metres while retaining 3.33.


For this original example, assume a 4.0 ft crest length and a true measuring head of 0.500 ft. Assume the installation satisfies the rating conditions; that is a calculation assumption, not a statement about an observed structure. The example deliberately retains the source equation's units to keep coefficient conversion out of the datum-error comparison.


Reference discharge = 3.33 × 4.0 × 0.500^(3/2) = 4.70933 ft³/s.


The crest length is unchanged in every comparison below. Only the head entered into the same equation changes. This isolates the numerical consequence of a datum error from other uncertainties.


A positive 0.020 foot head offset overstates calculated flow by 6.06 percent at 0.500 foot true head and 12.24 percent at 0.250 foot true head.

Original Francis-rating sensitivity example; not a weir design or full uncertainty budget.


Show the effect of a misplaced zero


Suppose the reading system overstates head by 0.020 ft, equivalent to 6.096 mm. It enters 0.520 ft rather than 0.500 ft. The head error is 4%, but the discharge error is not 4% because head is raised to the power 1.5.


High reading: Q = 3.33 × 4.0 × 0.520^(3/2) = 4.99470 ft³/s, approximately 6.06% above the reference.


Low reading: Q = 3.33 × 4.0 × 0.480^(3/2) = 4.42962 ft³/s, approximately 5.94% below the reference.


The positive and negative flow errors are slightly asymmetric. A compact check is Qindicated/Qreference = (Hindicated/Hreference)^(3/2). For a small head error, the fractional flow change is approximately 1.5 times the fractional head change; retain the exact power calculation when reporting the example.


Recheck the same offset at lower head


At a true head of 0.250 ft, the unchanged 0.020 ft offset is now 8% of the head. The corresponding high reading of 0.270 ft produces a flow ratio of (0.270/0.250)^(3/2), approximately 1.12237: a 12.24% overstatement. A fixed millimetre-scale datum problem can therefore be much more important at lower flow.


This comparison does not validate operation arbitrarily close to zero head. For the cited Francis rating, the manual limits head to at least 0.2 ft and no more than one-third of crest length. It requires crest height of at least three times maximum head and an upstream head station at least four times maximum head away, together with the other specified conditions. It recommends Kindsvater–Carter for general use. A spreadsheet returning a number outside the applicable conditions does not establish a valid measurement.


Separate datum checks from hydraulic checks


The manual identifies the crest as the head reference and specifies an upstream measuring location; its suppressed-weir guidance also addresses aeration and geometry. A correctly transferred datum does not compensate for downstream submergence, a clinging nappe, sediment or an inappropriate rating. Conversely, an apparently good flow pattern does not prove the gauge zero is right.


For an offline review, retain the crest-elevation reference, gauge-zero record, measurement location, observation time, water-level reading, rating identity and units as separate fields. Compare the survey and instrument records before blaming a change in calculated flow on the water system. Any physical verification or adjustment must follow the site's authorized access and safety procedures.


Report sensitivity without overstating accuracy


The percentages above quantify only the assumed head offset in a fixed equation. They do not include uncertainty in crest length, rating coefficient, approach flow, reading resolution or changing conditions. Do not label them total meter accuracy. If the datum has not been verified, show the uncorrected result and the unresolved datum issue rather than applying an unexplained correction.


A useful handover states what was measured, which reference produced the head, which equation converted it and which conditions remain unverified. That makes the result reproducible and prevents a small reference-level error from being mistaken for a real change in water flow.


Keep the next design task separate


Where this measurement forms part of a wider drainage-system review, do not substitute one observed weir flow for the design flow of connected pipework. For that separate piping task, the MEP Pipe Design and Sizing Handbook covers drainage and storm-water topics alongside plumbing sizing. It is a broader piping reference, not a weir-rating or instrument-calibration manual; the worked example above remains independent.



Continue the calculation review



Sources and calculation basis


Primary references checked on 27 October 2026. The worked examples and diagrams are original illustrative calculations, not manufacturer selections or project approvals. Apply the stated limitations and the requirements governing the actual project.





Comments

Rated 0 out of 5 stars.
No ratings yet

Add a rating
bottom of page