On 25 May 2019, the Meridian Gate data centre in London was evacuated after a fire alarm activation — but critically, the fire was detected at the incipient stage before any visible smoke or flames had formed. The system responsible? A VESDA aspirating smoke detection network that identified combustion byproducts from an overheating UPS battery cabinet while the concentration was still measured in parts per billion.
This is the defining capability of aspirating smoke detection (ASD): detecting fire at the pre-combustion stage — before a flame exists, before visible smoke forms, and before conventional detectors could respond. For environments where the cost of a real fire is catastrophic — data centres, museums, pharmaceutical labs, semiconductor fabs — this capability is not a luxury. It is the engineered foundation of asset protection strategy.
How Aspirating Smoke Detection Works
Unlike conventional spot detectors that wait for smoke to reach them by convection and diffusion, ASD systems actively sample the protected environment by drawing air through a network of sampling pipes fitted with laser-drilled sampling holes. The sampled air is delivered to a central detection unit containing an ultra-sensitive laser detection chamber:
- Sampling Pipe Network: Perforated pipes (typically 25mm CPVC or ABS) are routed above or below the protected zone. Hole spacing and diameter are calculated using proprietary design software (VESDA Pipe Design Tool, aspirAPT) to achieve balanced flow and equal sensitivity across all sampling points.
- Aspirator (Fan/Pump): A variable-speed fan continuously draws air through the pipe network at a calibrated flow rate — typically 1–4 litres/minute per sampling hole.
- Filter Stage: Coarse particulate filters remove dust and debris that would otherwise contaminate the laser chamber, dramatically extending service intervals in dusty industrial environments.
- Laser Detection Chamber: A high-intensity laser beam passes through the sampled air. Any smoke particles present scatter the laser light, which is detected by a photodiode array. The signal intensity is directly proportional to smoke concentration — measured in %obs/m with a resolution as low as 0.0015%.
- Four-Stage Alarm Thresholds: VESDA systems provide four independently configurable alarm levels — Alert, Action, Fire 1, Fire 2 — enabling staged response protocols (inspect → investigate → evacuate → suppress) rather than a binary alarm/no-alarm output.
Critical Applications for ASD Systems
| Application | Why ASD is Required/Recommended | Relevant Standard |
|---|---|---|
| Data Centres | Detect overheating components before ignition; avoid suppression discharge from nuisance alarms | NFPA 75/76, EN 50600-2-4, TIA-942 |
| Museums & Archives | Protect irreplaceable cultural heritage; avoid water damage from sprinkler discharge | BS 5839-1, PAS 79, NFPA 909 |
| Pharmaceutical Cleanrooms | Detect contamination events; minimise maintenance access under ISO 14644 | ISO 14644, EU GMP Annex 1 |
| Semiconductor Fabrication | Detect chemical fire precursors; clean room integrity maintenance | NFPA 318, SEMI S2 |
| Cold Storage (-40°C) | Conventional detectors fail below -10°C; ASD systems operate to -20°C standard | BS 5839-1 Section 21 |
| Raised Floor Environments | Sample below-floor void where smoke stratification prevents ceiling detection | NFPA 72, EN 54-20 |
| Historic & Heritage Buildings | Non-intrusive pipe routing; no ceiling penetrations; compliant with planning constraints | BS 5839-1, Historic England guidance |
VESDA-E Product Family: Key Specifications
The market-leading VESDA-E product family from Xtralis (now part of Honeywell) offers four principal detector variants, each optimised for specific environments:
- VESDA-E VEP (Pipe Only): Standard aspirating detector; 4 pipe inlets; protects up to 2,000 m²; suitable for most commercial and industrial applications.
- VESDA-E VEA (Addressable): Integrates with Notifier and Honeywell addressable fire panels via built-in protocol interface; device-level reporting on the SLC loop.
- VESDA-E VEC (Clean Room): Optimised for pharmaceutical and semiconductor cleanrooms; ultra-filtered design; HEPA exhaust to prevent re-introduction of sampled particles.
- VESDA-E VES (Storage): High-sensitivity variant for deep-rack sampling in data centres and cold stores; enhanced flow management for large pipe networks.
Competing ASD platforms in the market include Securiton SEC Graph, Hochiki HSD-ASD, Wagner TITANUS, and Siemens SITRONIC ASD — each offering comparable laser detection sensitivity with proprietary protocol integrations.
ASD Pipe Design: Critical Engineering Considerations
The performance of an ASD system is entirely dependent on the quality of its pipe network design. Key engineering parameters include:
- Pipe Length Limit: Maximum 100m total pipe length per inlet in standard configurations; longer runs require airflow calculations to ensure transit time remains below alarm response time targets.
- Sampling Hole Spacing: Typically 4m in low-hazard environments; 1–2m in high-value or high-speed fire risk environments per EN 54-20 and NFPA 72 Chapter 17.
- Sampling Hole Diameter: 3–6mm, calculated to achieve balanced flow throughout the network. Holes are laser-drilled for precision — field drilling introduces significant performance variation.
- Transit Time: The time for air to travel from the furthest sampling point to the detection chamber must be calculated and managed. Long pipes require larger diameter or additional aspirators to maintain transit times below target response thresholds.
- Below-Floor/Above-Ceiling Sampling: ASD uniquely enables detection in void spaces inaccessible to point detectors — a critical advantage in raised-floor data centres where the majority of cabling and power infrastructure resides.
Molecular-Level Contaminant Differentiation
By 2030, next-generation ASD systems will move beyond laser obscuration measurement to spectroscopic analysis of sampled air — identifying specific molecular signatures of different combustion sources (PVC insulation, lithium battery electrolyte, wood cellulose, hydraulic fluid) at concentrations below 1 ppm. This will enable intelligent alarm routing: a lithium battery combustion signature in a server rack triggers a precision suppression response targeted to that rack, while simultaneously notifying the data centre operations team with the specific rack ID, battery module reference, and estimated time-to-thermal-runaway. The system becomes not just an alarm, but an intelligent incident management platform with sub-minute warning and automated response coordination.
Frequently Asked Questions
Related Articles: Addressable Fire Alarm Systems · Self-Testing Detectors · Cloud Remote Diagnostics · FDAS Design Consultancy