Cracking the Codes
International testing of smoke, heat, CO and multi detectors at the heart of global standards
- By Bill Rossiter, Rick Heffernan
- Oct 01, 2006
A leading detector manufacturer recently recalled smoke/heat detectors that had been manufactured over a period of several months last year. The recall was necessary because an internal contact could have had insulating film, preventing the detector from reporting a heat alarm signal to the control panel.
This occurrence underlines strongly the need for proper functional testing -- both at installation, commissioning and acceptance and on an annual basis. Using examples from the national codes of the UK, United States, France, Germany and Canada, it is evident that other countries have similar requirements, but some can be clearer than they are now. The product recall underlines why such standards need to be adopted where they do not yet exist, clarified where they are misunderstood and upheld and enforced where they loosely followed.
Using examples from the national codes of the UK, United States, France, Germany and Canada, it is evident that other countries have similar requirements, but some can be clearer than they are now.
The Functional Check
A functional check involves a physical stimulus:
"In the case of detectors, tests must ensure that products of combustion are capable of passing unhindered from the protected area to the sensing chamber, for elements of the detector tests should focus on the ability of the detector to sample or verify the status of the atmosphere already in the sensing chamber," said the UK's BS 5839 1: 2002, a code of practice for system design, installation, commissioning and maintenance of fire alarm systems for buildings.
"The detectors shall be tested in place to ensure smoke entry into the sensing chamber and an alarm response," said NFPA 72.
But the UK's code also explains what is not acceptable. A functional check is not something that can be conducted by only checking analogue values, nor is it something that can be accomplished with a magnet. Use of a test button or a test magnet does not satisfy the recommendations states the UK's British Standard 5839 1: 2002 45.3.
In France, Regle APSAD R7 Detection automatique d'incendie, Edition 02.1997.4 (fevrier 2003) also highlights the importance of a proper functional check.
Its goal is to check that each detector has the capability of picking up that which it is designed to detect. As to how the functional checks can be performed, R7 explains that the person performing the work should be equipped with the necessary non destructive tools for the job that are compatible with the detectors and produce the appropriate stimuli to activate the detectors under test (smoke, aerosol, heat, IR or UV generators).
In a harmony rarely seen among these five nations, the Germans concur and DIN 14675:2003-11 states that the functional testing of the automatic fire detectors is, at the very least, to be carried out through the simulation of the relevant physical characteristics of fire outside of the detector (e.g. using test aerosols for smoke).
Extending this theme many national standards now refer not only to the need to check the device, but also the need to avoid danger or other damage to -- or from -- the environment in which the detector is installed. This is encapsulated by the UK's BS 5839.
"Every heat detector should be functionally tested by means of a suitable heat source. The heat source should not have the potential to ignite a fire, live flame should not be used and special equipment might be necessary in explosive atmospheres," said the UK's British Standard.
The French and British are of one mind and France's R7 states that devices producing live flames, such as lighters, are prohibited.
The theme of avoiding damage also is explicitly recognized for other detectors. Smoke detectors, for example, might react to various stimuli, but setting them off properly with a genuine physical functional check is only part of the art. Avoiding damage to them, for example, is another. Again, this is picked up by the British Standard.
"Point smoke detectors should be functionally tested by a method that confirms smoke can enter the detector chamber and produce a fire alarm signal (e.g. by use of apparatus that generates simulated smoke or suitable aerosols around the detector). It should be ensured that the material used does not cause damage to or affect the subsequent performance of the detector," said the British Standard.
"Each heat detector shall be tested to confirm operability, NOTE: An open flame shall not be used for testing the operation of heat detectors," said the British Standard.
When all these tests should be conducted? The answer is at commissioning and annually, though the annual tests are often split over the course of two or more occasions.
In the United Stats, the National Fire Alarm Code requires both initial, reacceptance and annual functional checks. Similarly, the British standard states that tests should be conducted at commissioning.
At commissioning, the Germans also require that 100 percent of all installed system components be tested and that, in the case of automatic fire detector tests, the alarm be triggered out by simulating the characteristics of the fire at the detector.
In France, Regle APSAD R7 Detection automatique d'incendie also highlights the importance of the commissioning tests. In Section 5.2, it says that commissioning tests should be cross-referenced to periodic annual checks. The checks can be split into two, six-per-month checks. Nonetheless, the checks must not be less frequent than six per month.
Six monthly visits by a competent person to every detector is, of course, a wise step for a number of reasons. Not only can a proper functional check on an ongoing basis highlight the inability of a detector to raise an alarm for reasons which may range from component failure, to wiring damage to dust covers, or other barriers to detection. It also provides an opportunity for an expert to assess building, usage and other changes that may impact the reliability or suitability of the detector relative to its installed environment.
So, with all this clarity, what confusion can remain?
One area relates to the ongoing sensitivity of a detector. This is something that cannot be assessed at all with most conventional detectors without a specialist test device. It also is one where even the faith in analog detectors,assessing as they do only the sensors as opposed to the ability of them to receive stimuli, is flawed.
NFPA 72 said that to ensure that each smoke detector or smoke alarm is within its listed and marked sensitivity range, it shall be tested using any of the following methods:
Calibrated test method.
Manufacturer's calibrated sensitivity test instrument.
Listed control equipment arranged for the purpose.
Smoke detector/control unit arrangement whereby the detector causes a signal at the control unit where its sensitivity is outside its listed sensitivity range.
Other calibrated sensitivity test methods approved by the authority having jurisdiction.
Detectors or smoke alarms found to have a sensitivity outside the listed and marked sensitivity range shall be cleaned and recalibrated, or be replaced.
The Canadian code is in concert with this, as well. CAN/ULC S536-04 said that each smoke detector sensitivity measurement and, if applicable, the cleaning date shall be recorded on the individual device record. Acceptable methods of determining the smoke detector sensitivity are:
Manufacturer's recommended test instrument, equipment or method.
Installed control units or transponders designed to test the sensitivity of individual smoke detectors.
Calibrated instruments that provide the operation as described in CAN/ULC-S529, Standard for Smoke Detectors for Fire Alarm Systems, for the purpose of testing smoke detector sensitivity.
Finally, the other area looking for clarification is that of multi-sensor detectors. Or multi-criteria detectors. The confusion has a single clarification for the purpose of field testing -- it does not matter whether one defines them as multi-criteria or multi-sensor. Neither does it matter how they are configured or when. Wherever possible, sometimes limited by the test modes available, each of their separate sensing abilities must be tested in the same real functional manner and at the same times.
This article originally appeared in the October 2006 issue of Security Products, pgs. 84-86.