CORDIS - EU research results

Fire Fingerprint Sensor-On-A-Chip For Early Fire Detection

Final Report Summary - SMOKESENSE (Fire Fingerprint Sensor-On-A-Chip For Early Fire Detection)

Executive Summary:
The Smokesense project is addressing the need for improved smoke detection in order to prevent the outbreak of fire, or to detect and raise alarm at an early stage of the fire. At the same time the risk of false alarms shall be reduced or eliminated by adding the ability to recognize “smell patterns” to the sensing technology. In order to significantly increase sensitivity it was targeted to be able to detect and identify matter in the range of 1-10 PPM. A functional model was running towards the end of reporting period, and it has proven that sensitivity below 10 PPB is achievable.
The detection of various substances is not only very sensitive, but the output also provides a clear indication of the particular “smell”.
The aim was to find the fire-gas fingerprint, by analyzing various combustible materials chemically. The finding of this work was that the smoke or vapour output is very complex and varies with each material. This is not seen as a major obstacle, and will be solved by further studies of the actual detection output of the Smokesense unit in multiple scenarios. By the results from the operation of one test unit at the Fire Eater test room from month 8, where it is installed in parallel with other sensing technologies, it is possible to identify certain groups of smells, and combine it with a few known gases related to fire, and thereby achieve both early detection (high sensitivity) and ability to discriminate between different smells.
The technology will to a high degree depend on a controlled airflow over the sensor element. In order to reduce cost, and after analyzing the performance requirement for the sensor element it was decided to develop a metal diaphragm micro pump. Is has proven stable and with adequate flow capacity at low power consumption. The sensor element itself has during the project been undergoing a fundamental change by providing indirect activation of the sensor element fibres. The goal to reduce the power consumption of the complete unit to be less than 3 W, was fulfilled by running at less than 500 mW even without applying power management (software is available), and without compromising sensitivity, stability or discriminating ability. Patent has been applied for.
The project ended with a technological breakthrough and even an actual product that will be produced in a small series, and installed in a data Centre in Denmark, a Factory in USA, at DBI fire test laboratory, and at Fire Eater offices, warehouse, manufacturing and canteen areas. These pilot installations are expected to be in operation during April, and after a few months of operation the first small series production will be initiated, and be available for selected customers by end of 2016.

Project Context and Objectives:
The final product was delayed, and an extension of the project period was granted from 24 to 28 months. The main focus in the project was to secure that the new development of the sensor chip worked well, and this part of the project was agreed in the consortium to belong to background even though the patent application and development was done during the project. The SME's has secured the rights relevant for the exploitation.
At present there are not any patents or patent applications on foreground.

Project Results:
The developments within the project regarding the sensor technology was in the consortium agreement assigned as background, and the SME's (coordinator with JVI and IVV assigned rights to use and field of application exclusivity) assigned with exploitation rights. Therefore the following main S & T rightfully belongs to background but the development, and application into a physical product was done within the project:
- very high sensitivity of the sensor.
- Ability to discriminate between different smells even at very low concentration.
- very significant reduction of power consumption of the metal oxide sensor technology.
- A fully integrated unit with sensor, electronics, software, sample air dosing pump and communication module (wired and WiFi) and on board USB, RS232, and ethernet connectivity.

Potential Impact:
At the end of the project is going straight into exploitation. A business plan has been laid out which contains the 3 main stages:
-Production of the first small series of Smokesense units (February to march 2016).
- Installation of first production batch at 4 locations monitoring s data center, a fire test laboratory, water treatment plant (chlorine), forging machinery, compressor room, chemicals filling area, offices, kitchen/canteen, fire extinguishing system test room, manufacturing area, epoxy moulding and curing area, warehouse.
- harvesting experience from systems working live in targeted application areas.
- Scaling up "hand made" sensor chip production to 400-500 units, to enable production of similar number of Smokesense units, and available for sales/installations by Q3-Q4 2016
- 1st generation industrial production machinery available for sensor chip production by 2017, and a production capacity for that year of about 2,500 units.
- Continue production on 1st generation production machinery for 2018 with an expected output of 3,500 units.
- During 2017 and 2018 the 2nd generation production machinery shall be developed to enable cost efficient production volumes from 100-250 units per day.

These first 3 phases are very important for the project to:
a) gain experience from real life installations, before scaling up production. Any issues with stability or drifting can be adjusted.
b) roll out the product in high value areas where an unexpected service cost in the first years will have negligible or minor effect.
c) found the basis, and real life proof of concept, before the planned mass production and of further miniaturization

The impact for the SME's after the first 3 years is to a high degree depending on the initial phase (first 3 years) experiences. The business plan for the first 3 years will provide an additional net EBITDA for the 3 SME's of abt. 2,5 million €. If the SME's fail to develop high production capacity production machinery during the initial period, the project will still have a significant potential, and the 1st generation based production can be sustained and multiplied up to quite reasonable capacities, but the cost efficiency will not be suitable for the mass market potential of the product so therefore it is essential to reach this stage of development to harvest significant societal benefits of the technology. Otherwise it will be reserved for high-end applications and fail to reach the full potential and one of the goals of the project: to minimize general fire damages and death and injuries due to fire.

List of Websites:
Torbjørn Laursen
Fire Eater A/S
Vølundsvej 17
3400 Hillerød