Key takeaways
- An unsuitable or poorly sealed hood causes leaks that distort the measurement by fifteen per cent.
- The SIA 382/1 standard governs the permissible tolerances on supply and extract air flow rates.
- The annual calibration of the anemometer determines the legal and technical validity of the acceptance report.
- Repeatability is achieved by taking three consecutive readings, each integrated over a minimum of thirty seconds.
Choosing and sealing the measuring hood to channel turbulence
Directly measuring the flow rate at a diffuser or an extract grille with a vane anemometer without an accessory produces major systematic errors. Terminal air flows show uneven velocity profiles, jet effects, recirculation and pronounced swirl caused by the diffuser vanes. Holding a handheld anemometer against the centre of a supply terminal underestimates or overestimates the volume depending on the angle at which the air streams strike the impeller.
- Airflow measuring hood
- Airtight funnelling device mounted on a measuring probe (vane or hot wire), channelling the entire air stream of a diffuser in order to turn a turbulent flow profile into a uniform straight flow.
The choice of hood depends directly on the geometry of the terminal. For a circular grille with a diameter of 100 to 160 mm, a round hood of suitable dimensions prevents the jet from being crushed. For square or ceiling diffusers of 600 x 600 mm, a flow capture hood with a fabric skirt mounted on a frame with a honeycomb jet straightener is required. The fundamental handling rule is to completely enclose the outer frame of the diffuser. The peripheral foam seal must be compressed evenly against the slab or suspended ceiling to prevent any peripheral leakage caused by the overpressure created by the device.
Calculating uncertainty and repeating readings in accordance with good practice
Acceptance of a ventilation installation is based on compliance with the design flow rates specified by the master energy plan and the requirements of the SIA 382/1 standard. For a measurement to be technically and contractually defensible during the inspection of the works, the technician must take into account the inherent uncertainty of the measuring chain and eliminate handling deviations.
| Source of error | Average impact observed | Corrective measure on the construction site |
|---|---|---|
| Hood sealing defect | 5% to 15% | Visual check of the foam seal and firm mechanical hold |
| Inherent sensor uncertainty | 2% to 3% | Valid calibration certificate less than 12 months old |
| Flow swirl effect | 4% to 8% | Flow straightener fitted upstream of the vane |
| Fan instability | 1% to 3% | Rolling time-averaged reading over at least 30 seconds |
- System conditioningCheck that all doors and windows of the room are closed and that the air handling unit has been running at a stable nominal speed for at least fifteen minutes.
- Positioning and centringApply the hood perpendicular to the wall while compressing the seal, without obstructing the outlet section of the measuring nozzle.
- Time integrationStart the average recording on the anemometer for thirty to forty seconds in order to smooth out air pulsations.
- Repeating the seriesTake three separate measurements, removing and repositioning the instrument between each cycle, to validate the repeatability of the procedure.
If the difference between the maximum and minimum values of the three readings exceeds five per cent, the measurement is rejected. This fault indicates airflow instability, a poorly locked adjustment damper or poor manual sealing of the hood. The contractor meets their obligations by recording the arithmetic mean of the retained series, the instrument reference and the date of its last calibration in the measurement report handed to the site management.
Measurement protocol at the diffuser: neutralising the jet and hand position
A 20% variation between two successive readings at the same supply terminal does not come from a fault in the network, but from airflow instability at the capture point or from an angular offset by the operator. Ceiling diffusers with helical or slot jets throw strongly diverging air streams with tangential velocity components. If the technician applies a vane anemometer directly to the grille without a flow straightener, part of the fluid stream strikes the blades obliquely, producing a major underestimation or overestimation depending on the tilt of the probe.
- Choice of funnel and honeycomb straightenerFit onto the diffuser a suitable measuring hood whose skirt fully covers the frame of the terminal, with foam seals compressed against the ceiling or wall to prevent any peripheral leakage. The instrument must include a honeycomb stabilising grid upstream of the counting section to turn the swirling motion of the flow into regular axial streamlines.
- Maintaining mechanical contact pressurePress the hood on using a telescopic pole, applying a constant axial vertical thrust. Any tilting or gaping of the base of more than 5 mm distorts the entire air passage and changes the back-pressure exerted on the terminal.
- Stabilisation delay of the fluid streamAllow the thermo-anemometric probe or the micro-manometer flowmeter to stabilise for a minimum of 10 seconds after the hood is applied. Introducing the hood creates a slight local overpressure that temporarily dampens the flow of the section before it reaches its steady state.
| Diffuser type | Recommended measuring instrument | Main disturbing phenomenon | Average deviation observed without a suitable hood |
|---|---|---|---|
| Ceiling diffuser with helical jet (supply) | Pressure-compensated balometer or hood with honeycomb | Swirling vortex and radial velocity component | +15% to -25% |
| Linear jet terminal with multiple slots | Airtight rectangular hood with flow-straightening grid | Peripheral induction and uneven velocity gradients | -15% to -20% |
| Single-deflection wall extract grille | Rigid hood on 100 mm vane anemometer | Fluid stream contraction effect at the edges of the frame | +10% to +15% |
| Self-adjusting sanitary extract valve | Circular hood with flexible sealing ring | Sensitivity to the additional pressure drop of the funnel | -5% to -12% |
Repeatability, statistical average and validation according to the SIA 382/1 standard
The compliance of a ventilation installation is never based on a single raw value displayed by a digital anemometer. The overall measurement uncertainty results from the inherent error of the sensor (often given by the manufacturer as plus or minus 3% of the reading plus a constant), the thermal drift at the measuring point and the variability in the operator's application. The professional method requires a series of three separate readings, separated by a clear removal of the hood followed by a fresh sealing. This manoeuvre makes it possible to estimate the real repeatability of the on-site acquisition.
The SIA 382/1 standard sets the rules applicable to ventilation and air-conditioning installations in buildings. To ensure the evidential value of the commissioning report towards the site management or the consulting engineer, the ventilation contractor must record the serial number of the measuring instrument, the date of its last calibration certificate from an accredited laboratory (valid for a maximum of 12 months), the ambient temperature during the test and the air density correction factor applied. When flow rates are adjusted at the regulating devices, the contractor is responsible for the proper performance of the services in accordance with the general obligations of art. 364 CO, by delivering networks whose measured flow rates match the design flow table within the contractual tolerance margins.
In the event of a dispute during the inspection of the works within the meaning of art. 367 CO, the reproducibility of the measurement is the company's only technical safeguard. A cross-check carried out with two different instruments leads to the same conclusion only if the hood placement protocol, the integration time and the pressure-drop corrections specific to the instrument are rigorously harmonised between the parties.
Resolving deviations and adversarial protocol in the event of a dispute
When a technical acceptance is blocked by a flow deviation exceeding the tolerances permitted by the SIA 382/1 standard, the clash between the installer's figures and those of the inspection body almost always stems from a difference in instrumentation or a biased application protocol. Two technicians measuring the same terminal with different instruments or different holding positions commonly find a twenty per cent deviation. To break the deadlock without needlessly reworking the ventilation network, the company must immediately formalise an adversarial procedure on site.
- Comparison and checking of calibration certificatesEach party presents the current calibration certificate of its equipment. A vane anemometer or differential manometer that has not been verified for more than a year is automatically excluded from the finding.
- Visual check of the sealing at the contact surfaceThe adversarial test requires checking the foam seal of the measuring hood. Any air leak between the terminal trim and the base distorts the reading through the intake of ambient air or peripheral back-flow.
- Cross-check with double readings and swapped operatorsThe installer's technician and the site management's technician each take three consecutive measurements with the same instrument, then repeat the operation with the instruments swapped, in order to isolate human bias from the measurement uncertainty.
- Recalculating the average and comparing it with the tolerancesThe arithmetic mean of the stable series serves as the reference value. The deviation is recorded in the report with express mention of the overall uncertainty range accepted by the SIA 382/1 standard.
If the flow rate remains outside tolerance after the measurement biases have been neutralised, the investigation moves from the terminal back up to the upstream regulating device. Checking the position of the balancing damper, measuring the static pressure in the branch duct and inspecting the fouling of the filters in the air handling unit make it possible to characterise the fault: overall fan drift or a simple branch imbalance.
Frequently asked questions
How should we respond if the engineer disputes our hood-measured flow rates?
Propose an immediate adversarial reading. Require the calibration certificates of both instruments to be presented. Carry out a cross series of three measurements per terminal with the same hood fitted with its jet straightener, recording the arithmetic mean. If the deviation persists, check the influence of the pressure drop added by the instrument on the diffuser against the tolerances of the SIA 382/1 standard.
Why does an extract terminal show unstable values?
Instability often comes from air stream separation or turbulence immediately upstream of the grille, caused by an elbow too close to the plenum. To stabilise the reading, press the airtight skirt of the hood firmly against the wall without over-compressing the seal, let the flow establish itself for at least ten seconds and record a rolling time average over thirty seconds.
Can a helical-jet terminal be measured with a simple vane?
No, this is strongly discouraged. The swirling flow of a helical-jet diffuser completely distorts the rotation of an unguided vane, creating errors that frequently exceed twenty per cent. You must insert a honeycomb flow straightener into the measuring hood or use a compensating matrix anemometer that neutralises the radial velocity components so as to measure only the axial flow.
What flow deviation tolerance is permitted at acceptance?
The SIA 382/1 standard defines the acceptance criteria for ventilation installations. As a general rule, an overall tolerance of ten per cent on the total flow rate of the installation and of ten to fifteen per cent on each individual terminal is permitted during final balancing, provided that the sum of the flow rates per room meets the required hygienic air change rates.
Texts and standards cited
- norme SIA 382/1Ventilation and air-conditioning installations - General principles and required performance for the measurement and tolerance of flow rates
