How UV, IR and Imaging Detectors Work

Fire detection is based on various properties of fire such as smoke emission, a rise in temperature and optical radiation emission.

In large spaces or outdoor environments, optical flame detection is one of the fastest and most reliable methods of detection. Such types of detection are based on radiation produced by the flames in visible (Vis), ultraviolet (UV) or infrared (IR).

What is a Flame?

The combustion of a hydrocarbon flame (oxidation reaction) mainly produces water (H2O) and carbon dioxide (CO2, in an excited state (*):
     2 C6H14+ 19 O2+ Energy ⇒ 12 CO2*+ 14 H2O*

The return to the original state of these molecules takes place through photon emission at specific wavelengths:
     CO2*⇒ CO2+ hν (2.8 µm - 4.3 µm)
     H2O* ⇒ H2O+ hν (2.9 µm)

In the emission spectrum of flames, there are some bands which are characteristic of their signature. These include:

A "bump" in the UV radiation around 200 nanometers; and a near infrared and visible part between 0.4 and 3 microns, which is more or less intense, based on the nature of the fuel.

Components of a burning flame

The second signature is of a temporal nature and is used to differentiate a flame from background radiation (environmental or industrial).

A flame will flicker randomly when it is burned naturally (with surrounding air), when compared to an artificially combusted flame that will not flicker as with the case of a Bunsen burner, for instance. The flame’s radiation is usually modulated from 1 to 20 Hz band which can allow efficient pre-filtering of the signal received by the detectors.

The Different Types of Detectors

One or more optical radiation sensors are used to design optical flame detectors, which can be configured on UV, IR, a combination of (UV/IR), or multiple IR bands (Multi IR).

False radiation from many environmental or industrial sources can impede the fire spectrum which can make it difficult to detect the flame. The device should not be affected by this environment and at the same time should have a good sensitivity to fire.

Ultraviolet (UV) Detectors

Ultraviolet detectors (UV) can only react to UV radiation that has a wavelength of less than 300 nm (a solar-blind region).

The sensor is a photo-tube that includes an anode and a cathode placed in a large potential difference (about 300 volts), and sealed in a quartz tube filled with inert gas. The photons are illuminated by UV radiation and strike the cathode release electrons, which, in turn, are drawn toward the anode. The gas molecules contained in the bulb are ionized by the electrons - energy carriers, producing a chain reaction. Then, an output signal is generated as a voltage pulses sequence by the detector.

Sometimes, this type of detector is employed for flammable liquid fires such as hydrocarbon fires but it is not only limited to this. The detector is one of the few devices to allow the detection of metal fires, ammonia, and hydrogen.

This type of detector can provide an extremely fast response time, hundreds of milliseconds or less, and reach a sensitivity up to pico watt/cm2. Yet, the detection capacity of this detector is attenuated in the presence of smoke. Therefore, in the event of a fuel fire in a closed environment, if the UV sensor fails to "see" the flames relatively early, then it may not detect the fire.

In a similar way, the UV radiation is absorbed by the oily films that may be deposited on the window of the detector, or by other organic compounds present in the environment. These factors will limit the detection sensitivity. The detector is also sensitive to a lightning storm, electric arcs, and X-rays, which can create false alarms.

Infrared (IR) Detectors

An IR detector is a pyroelectric sensor that is capable of detecting thermal radiation and is rather sensitive to variations of the received light signal. A tantalum/lithium crystal is associated with an OP amp or a field-effect transistor. An optical filter (a spectral gate) chooses a specific spectral band or wavelength (2.9 µm, 4.3 µm, 4 to 5 µm…).

The crystal perceives the random flickering emitted by the flame in the infrared band and creates a signal processed by a low-frequency bandpass filter (1 to 20 Hz) before it is interpreted by a microprocessor. Today’s components possess good signal to noise ratios which enable considerable amplification factors, and thus excellent sensitivity to flame radiation.

The IR detector is efficient in smoky environments and on bad burn rate fires (Gasoil for instance), but it is sensitive to aqueous environments (frost, fog, etc.) and many other interfering IR sources that are present in the environment. These sources may overlap the signals which had to be detected.

In order to protect against this strong sensitivity and prevent false alarms, several sensors are now being used in a single device (Multi IR). Generally more reliable and cheaper than a UV detector, this device has a longer mean time before failure (MTBF) due to less severe constraints on components.

UV / IR Combined Detectors

This type of detector combines the two technologies (gate AND between "UV channel" and "IR channel") and provides excellent rejection of false alarms and good detection distance. However, UV / IR combined detectors suffer from the combined limitations of UV and IR detectors.

UV / IR2 Combined Detectors

The combined UVIR detector using a performant UV sensor is limited in terms of detection range (a distance of detection), by the infrared part.

The IR sensor is curbed to keep a “reasonable” activation of the IR chain by the surrounding thermal background. By using extra information in the IR band, via a second sensor, this increases the gain of the amplification chains of the detector and thus increases the detection range, and at the same time maintains a good immunity to false alarms.

Multi-IR Detectors

The present market is obviously geared towards multi IR detectors. The aim is to remove the multiplying information and UV in the IR domain. A large number of devices depend on the CO2 IR band at 4.4 micron, but there are some devices that use the H2O IR band (2.9 microns) in order to detect ammonia or hydrogen fires.

Usually, multi IR detectors possess a longer detection range and maintain an extremely low rate of false alarms. In addition, these detectors are insensitive to attenuations related to oil vapors and fumes as well as other interference sources like a welding arc, lightning, etc. Again, the idea is to increase the gain of a detector’s amplification chain and select the wavelength and suitable signal treatments.

Three IR detectors are generally used in three different spectral bands, reaching detection ranges between 40 and 80 meters on standard heptane standardized fires, even in the presence of many interfering IR sources in the environment.

A multi IR detector that removes the multiplying information and UV in the IR domain.

Detection by Imaging

This technique is also employed for flame detection and is based on image processing issued from CCD matrices. Some models are limited to the visible range for a long time and currently operate infrared matrices with a spectral filter just like the one employed on conventional multi-IR detector to enhance the sensitivity to fire.

While the technique is more limited in terms of detection range, a major benefit is its excellent treatment of "friendly fire" which helps prevent flare reflection on metallic elements found in various industrial sites for example.

The technique’s limitations are intrinsic to the visible method (reduced efficiency in case of smoky fires or of extremely little emissive fires such as methanol, hydrogen, etc., or in the infrared range (range reduction owing to the presence of water-related compound, detection of hydrocarbon fires only, etc.).

A number of technological development areas are now possible for future devices:

  • Implementation of more advanced signal processing algorithms, based on multi IR technologies with three, four, or five sensors.
  • Multi-spectral infrared imaging - the cameras which can now reach the IR range in the appropriate spectral bands are still extremely expensive; however, these are likely to take over from the "classic" cameras limited to the visible.
  • Implementing semiconductor components in the UV range by replacing discharge tubes. Currently, no suitable component is available on the market to obtain this function in this application framework, but these types of components are being worked upon by many laboratories.

Metrological Characteristics

The market is primarily geared towards use in explosive atmosphere applications and hence, almost all the devices come with sufficient protection modes under regulatory requirements (IEC / ATEX / FM / CSA / UL) as well as classic information outputs signal (4-20 mA / ModBus / relay)

From a strictly metrological standpoint, optical flame detectors have some specific features. In particular, the concept of sensitivity is more difficult to understand than for a gas sensor and it is essential to clearly identify the factors affecting it.

  • First the fuel: all combustions do not produce the same amount of H2O or CO2 molecules. For instance, a fire of methanol (CH3OH + 2 O21 CO2* + 2 H2O*) is less emissive than a heptane fire (C7H16 + 11 O27 CO2* + 8 H2O*).
  • Secondly, the size of the fire: From a geometrical standpoint, the amount of signal received by the detector from the fire reduces as the square of the distance between them. In order to get a signal that is equivalent to the one achieved at a distance D of fire surface S, with double the distance (2xD), the fire surface should be multiplied by x4.
  • Finally, if the fuel is gaseous, then the diameter of the nozzle is not sufficient and this should be completed by the information associated with flow or flame height

Sensitive parameters of gas and flame detectors

Sensitive parameters.

There are other parameters that are more challenging to conquer while they can also interfere, such as the presence of interfering sources, the attenuation by rain, fog, smoke, wind speed on the fire, reabsorption of CO2 radiation in the optical path, etc.

The response time is related to a particular fire as it often decreases with the distance between the device and the fire. It differs considerably from technology to technology (UV detectors are usually faster than IR detectors) and can differ from less than 1 second to 30 seconds based on technologies, fires and Manufacturers.

The cone of vision is considered to be a fragile feature, while it is imperative in terms of positioning of devices.

Manufacturers provide values between 90° and 120° on the horizontal axis, but sometimes less on the vertical axis owing to the optical elements required for self-test.

Manufacturers provide values between 90° and 120° on the horizontal axis, but sometimes less on the vertical axis owing to the optical elements required for self-test.

It must be remembered that sensitivity to a fire is not constant over the entire extent of the cone of vision. The detection range will reduce when it moves away from the optical axis, along with a sudden fall on the cone edge. As a rule, vision angles are given for half distance sensitivity limit against the one acquired on the optical axis. Hence, it is appropriate to consider this value in the definition of the installation.

Manufacturers provide values between 90° and 120° on the horizontal axis, but sometimes less on the vertical axis owing to the optical elements required for self-test of flame detectors

The false alarm immunity is a very critical parameter since the financial outcomes can be huge in case of unintentional activation of the extinguishing system.

The devices supplied by key players possess excellent immunity with respect to false alarms and offer a range of evaluated sources, including the test parameters (distance, power, etc.), the type of source tested (welding arc, tungsten, halogen lamp, etc.), if the device activates an alarm in these conditions or not, and if the source is modulated or not.

During the time of installation, it must be remembered that the presence of a parasitic source (modulated infrared source such as sunlight through an open sky turbine or vapors) in the device field of view (multi-IR), even if cannot trigger an alarm, represents significant background radiation which can reduce the device’s sensitivity.

Choice of Technology

The following criteria help choose the most suitable technology for an application:

  • The fuel does not contain hydrocarbon molecules: The most suitable device would be a detector configured on UV or H2O emission-based infrared. The detection range is generally lower than for multi-IR versions for hydrocarbon fuels.
  • The risk source (potential fire) is at a very long distance and/or is small. One shall then chose multi-IR versions that achieve the best sensitivity.
  • The working area is a closed environment where smoke will rapidly accumulate or where there may be some vapors, such as hydrogen sulfide, ammonia, methyl methacrylate, ethyl or methyl acrylate, aromatic compounds such as toluene, styrene, hydroxybenzene, benzene, xylene, some nitrogenous compounds such as nitromethane, nitro propane, nitrobenzene, some chlorinated compounds such as tetrachloroethylene, chlorobenzene, dichlorobenzene, chloroprene, chloro-nitro-propane, vinyl chloride and lastly, some derivative from ethane or butane such as acetone or ethanol, acetaldehyde, butylamine, butadiene, and butanone. In that case, versions implementing UV should be avoided because this band is highly attenuated by these vapors or particles
  • The working area is outdoor. It is therefore risky to use a single UV technology because of its sensitivity to a number of relatively intense UV sources and the fact that UV can be reflected on all types of surfaces, especially metallic.
  • The coverage area includes hot equipment with opportunities for powerful air convection and the potential presence of CO2. It is recommended to use UV technology (UV, UVIR, UVIR2) in order to enhance resistance to false alarms.
  • The alarm triggered must be very short (<1 second). First, it is essential to cover the risk on a short distance, and subsequently encourage UV technologies that are probably faster.
  • “Friendly” fires such as a flare are seen in the device field of view, live, or by reflection, and the signal is strong enough to trigger a conventional detector. If the reduced sensitivity (see below) does not remove the activation, then this type of fire can be isolated using devices based on image analysis.

Installation and Setup

As mentioned before, the installation must consider the device cone of vision to cover the area that needs to be monitored, preferably with recoveries between detectors.

The installation of a flame detector must consider the device cone of vision to cover the area that needs to be monitored, preferably with recoveries between detectors.

In addition, devices are often installed in a high place and one must note the presence of a “shadowed” area at the base of the detector.

It would be better if the detectors are not installed using the IR bands on pipes which can shake in the presence of a strong wind, for instance, so that no background signal is generated which could otherwise interfere with the detection. Using the significant part of the optical flame detectors, users can adjust the sensitivity as well as the time delay before outputting an alarm, depending on the type of risk to monitor the environment.

Sensitivity alters the maximum distance at which a given fire produces an alarm. The temporization refers to the time (in seconds) during which the unit should receive a steady fire signal before triggering an alarm.

Normative Framework

Beyond the hazardous areas equipment standards, there is a normative framework for other aspects of the device:

EN 54-10 - Detection and Fire alarm systems - Flame detectors - Point detectors. This EN 54-10 standard has been attached to the Building Products Directive (BPD) through its amendment A1 (in March 2006). Among other things, the EN 54-10 standard provides the frame of the regulatory requirements with respect to performance, functionality, marking, shock resistance, vibration, EMC, light signaling, software design, manufacturing control of the equipment or documentation.

For instance, detectors are assessed based on their sensitivity to fire in the following classes: Class 1, Class 2, Class 3. Class 1 and class 3 being the highest level of sensitivity and the lowest level of sensitivity, respectively. On the American market, the repository is in accordance with the FM3260 standard. The method is slightly different from the EN 54-10 standard, but overall, the standard aims to check almost the same type of features.

Functional Safety

Although is it not compulsory, the evaluation devices in accordance with the IEC 61508 standard are increasingly being demanded by users. Certification, beyond defining reliability parameters (PFD, SFF, MTTR, MTBF, etc.), also covers the software, both in its development process and its structure.

The certification determines a Safety Integrity Level (SIL) number for the safety function afforded by the device. This level is increased from 1 to 4, with 4 being the highest. Several devices are SIL2 certified, but the best is to achieve SIL3 certification.

Conclusion

Therefore, selecting the right flame detector depends on the exact identification of the requirement.

Firstly, this involves questions like what type of fire needs to be detected (metal fire, flammable liquid fires, etc.), what is the type of working environment (outdoor, indoor, industrial, environmental interference sources, etc.), what is the response time anticipated and what are the distances that need to be covered.

Secondly, the installation should be properly configured: control panel voting function, device location, functional tests, and adjustment of sensitivity and time delay settings.

Finally, maintenance mostly involves cleaning the device window, either when the optical self-tests indicate a fault or as a preventive measure.

This information has been sourced, reviewed and adapted from materials provided by Teledyne Gas and Flame Detection.

For more information on this source, please visit Teledyne Gas and Flame Detection.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Teledyne Gas and Flame Detection. (2020, October 12). How UV, IR and Imaging Detectors Work. AZoSensors. Retrieved on November 23, 2024 from https://www.azosensors.com/article.aspx?ArticleID=815.

  • MLA

    Teledyne Gas and Flame Detection. "How UV, IR and Imaging Detectors Work". AZoSensors. 23 November 2024. <https://www.azosensors.com/article.aspx?ArticleID=815>.

  • Chicago

    Teledyne Gas and Flame Detection. "How UV, IR and Imaging Detectors Work". AZoSensors. https://www.azosensors.com/article.aspx?ArticleID=815. (accessed November 23, 2024).

  • Harvard

    Teledyne Gas and Flame Detection. 2020. How UV, IR and Imaging Detectors Work. AZoSensors, viewed 23 November 2024, https://www.azosensors.com/article.aspx?ArticleID=815.

Comments

  1. Jane Doe Jane Doe United States says:

    This report is phenomenal, extremely well written technical data.  Perfect clarity after just one or two readiings.  Meanwhile, a thorough knowledge is reflected, a reassuring expertise that keeps the reader reading, rather than tiring and going on to something else. If the author(s) of this report have written any textbooks, please let me know!

  2. Ripu Singh Ripu Singh United Kingdom says:

    best report read so far on internet.

  3. Jaap de Vries, Ph.D. Jaap de Vries, Ph.D. United States says:

    Great read, just an FYI, heptane does not radiate more than methanol due to the higher number of C atoms. Heptane burn much faster than methane and produces more soot, which causes the incandescent radiation. I don't think you will see significant increase burning methanol ethanol propanol etc. even though the number of carbon atoms keep going up.

The opinions expressed here are the views of the writer and do not necessarily reflect the views and opinions of AZoSensors.com.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.