Duct smoke detectors sample airflow to trigger fire and smoke dampers; their placement follows NFPA 90A and the International Mechanical Code.
Where the Code Places Them
The detector samples air from the duct through a perforated tube or an integral probe, so it senses smoke carried in the airstream rather than in the room, giving earlier warning on a protected system before smoke spreads through the building.
A duct smoke detector senses smoke in the airstream rather than the room, so it can catch a fire in a concealed ceiling or riser before it spreads through the whole air-handling network. It is the detection half of a strategy that then shuts the fan or closes dampers on signal.
The detector uses a sampling tube with holes along the duct, drawing air to the sensing chamber so the device sits outside the flow and stays serviceable. The sampling points are placed where they represent the airstream, and the tube is kept clear of the fan's turbulent wake where it would read noise instead of smoke.
Air velocity sets the rules. Most duct detectors are listed for a band, commonly about 1-10 m/s or 1-20 m/s depending on the model, because too slow a flow fails to carry smoke to the tube and too fast a flow dilutes it below the threshold. The duct section is chosen where velocity sits inside that band at all operating points.
Code places the detector on the supply side downstream of the fan and any filters, or on the return where the standard requires, and it is wired to the building fire system so the response is automatic and logged. A detector fitted against the listed velocity or in the wrong branch gives a false sense of cover, so placement is verified at design.
The company's ventilation range is built under ISO 9001:2015 with CCC certification at its 80,000 m² production site, and its engineers respond to selection enquiries within 24 hours to advise detector location and duct velocity for compliant smoke control on a given air-handling layout.
Correct placement also avoids nuisance alarms: a detector in a high-velocity or turbulent stretch reads entrained dust as smoke, while one in a dead leg never sees a real event, so the section is picked from the airflow drawing and confirmed against the duct velocity at every operating speed. A visual smoke test or a calibrated challenge at commissioning validates the tube location before the system is signed off.
On a supply system, the sampling point is typically located downstream of the fan and upstream of the first branch, where smoke from a coil or unit would first appear in the distributed air that is about to enter occupied spaces.
On a return or exhaust system, the point is usually upstream of the fan and after the last branch, so smoke entering any return is caught before it reaches the unit or is recirculated back into the supply through a damaged or leaking coil section.
Codes also limit the air velocity at the sampling point, commonly 1.5-10 m/s, because too high a speed dilutes the sample and too low may not represent the flow, and both conditions can delay or prevent reliable detection.
Practical Rules
Maintain the required straight duct or tube insertion length so the sample is representative; a turbulent elbow right at the probe gives erratic readings and can cause nuisance alarms or, worse, a failure to detect during an actual event.
Keep the detector and its housing accessible for testing, since code compliance depends on periodic functional checks with a smoke source or test cap, and an inaccessible device cannot be proven to work when the authority asks for evidence.
Coordinate with the damper control so detection reliably closes the associated fire or smoke damper and signals the building system, rather than only alarming locally where nobody is present to act on the warning.
Do not place the sample where condensation or dust will false-trip it; a wet or dirty duct needs a suitable housing and maintenance plan, because repeated false alarms train occupants to ignore the system and delay real response.
| Item | Specification | Note |
|---|---|---|
| Supply point | after fan | Before first branch |
| Return point | before fan | After last branch |
| Air velocity | 1.5-10 m/s | Sample validity |
| Basis | NFPA 90A / IMC | Code reference |
| Action | closes damper | Signals BMS |
| Access | for testing | Required |
Selection Checklist
- Locate supply sampling after the fan and before the first branch.
- Locate return sampling before the fan and after the last branch.
- Confirm duct velocity at the probe is within 1.5-10 m/s.
- Provide access for periodic functional testing.
- Interlock detection with the correct fire or smoke damper.
- Protect the housing from condensation and dust false trips.
About the Manufacturer
Keruite Air Conditioning Group, headquartered in Dezhou, Shandong, has more than 30 years of HVAC manufacturing experience and operates an 80,000 m² production site with its own performance testing laboratory. Products are built under ISO 9001:2015 quality management and carry CCC certification, and are supplied to customers in over 50 countries.
Our engineers respond to selection enquiries within 24 hours and can provide project-specific sizing, performance curves, and documentation to support specification and procurement.