Re-think the Spec?
Are installers and specifiers spending more money than is necessary by installing costly cabling in fire protection installations? The money invested in cabling would be far better spent on fitting more detectors to give a faster response to fires, says Don Scott of Siemens Building Technologies Fire Protection.
Keep the number of detectors down – cabling for them costs a fortune! That’s a thought that most specifiers will have in mind when planning a new fire protection system. But, as the far superior practice in Europe proves, it doesn’t have to be that way if we’re prepared to be realistic about real-world requirements and the interpretation of the relevant standard, BS5839.
So how many detectors should you include in a fire detection system and where do you site them? Those are not simple questions, of course, but hopefully the answer will be based on a thorough risk assessment of the people and areas to be protected, and a detailed consultation of the standard. But even then the answer is not usually clear-cut. The standard defines five types of life protection system, where the principal objective of an installation is to protect people from loss of life or injury.
With Type M systems, the alarm is raised manually, typically by using a manual call point. Type L5 systems are custom-engineered to meet specific requirements and, since these are so specialised, we won’t consider them further in this article. Type L4 systems have detectors to trigger the fire alarm automatically, but these are sited only on escape routes. With Type L3 systems, the detectors are sited not only on the escape routes, but also in adjoining rooms.
Type L2 installations take the use of detectors a stage further, by adding them in areas that, following a risk assessment, are deemed to be of a sufficient risk to warrant protection. The widest application of detectors is provided in Type L1 applications where they are installed, with minor exceptions, throughout the whole property providing protection for each area.
Now, suppose the risk assessment of a particular building shows that Type L4 protection will be acceptable, what possible reason could there be for fitting more than the minimum number of detectors to meet this requirement? A couple of interesting statistics from the Fire Service suggest the answer to this question.
The first is that in buildings where automatic detection is provided, 16% of fires failed to trigger an automatic alarm at any stage. It’s easy to see why this might be the case. Consider a building with an L4 system, with detectors only on the escape routes and, for the sake of this example, let’s consider that the escape routes are corridors. A room opening directly off the corridor may have within it another inner room.
If a fire starts within this inner room, it is going to take a very long time indeed for the products of combustion to reach the corridor and trigger the detectors mounted there, if indeed they ever do. Is this a rare case? Not according to the second interesting statistic, which is that over 60% of fires in buildings with automatic fire detection start in areas where no detectors are fitted.
It seems clear from these arguments that there is a very good case for fitting more detectors than are actually prescribed by the risk assessment and, in fact, in Continental Europe, this is standard practice. However, it is an approach that’s hardly ever adopted in the UK, and the reason is simple – cost. That’s not to suggest that European specifiers are prepared to spend more on fire safety, rather that an installation with a given number of detectors in the UK costs far more than it would in Europe. At first, this may seem puzzling, as the same detectors are usually on sale at much the same prices throughout the EU.
The solution to this conundrum is a key difference in practice between the UK and Europe. In the UK, detectors are almost invariably wired with fire-resistant cable, whereas in Europe, they are wired with ordinary Belden-type cable. The reason is that BS5839 requires that “all critical signal paths” within a fire protection installation should be wired in fire-resistant cable, and this invariably includes the wiring to the detectors.
Fire-resistant cable, however, is a very specialised product since, to meet the requirements of the standard, it must be capable of withstanding a temperature of 930ºC for up to two hours. The cost implications of using this type of cable are dramatic. Typically, detectors wired with fire-resistant cable – often mineral insulated cable that is more akin to copper pipe than ordinary electric cable – cost around £200 per detector, including cabling and installation. The equivalent cost for detectors wired with Belden-type cable is around £30 per detector.
It’s not hard to see that if Belden-type cable is used, then many more detectors can be fitted, providing much better coverage of the protected area, without increasing the cost of the installation.
But is the switch away from fire-resistant cable for detectors justifiable? First of all, note that we are talking here only about the cabling for detectors. There is no argument for switching to non-fire-resistant cable for other parts of the installation including, for example, cabling to sounders or other life safety systems.
Now let’s return to the detectors. Their function is to raise the alarm when a fire starts. After they’ve done this, their job is done. So why do they need to be wired with cable that will allow them to communicate reliably to the fire panel long after the fire has started and they’ve done their work? Also remember that after even a few minutes at temperatures far lower than 930ºC, most detectors would cease to function.
This would seem to show conclusively that there is no point in using fire-rated cable for detectors. One argument to the contrary, however, is that if the detector cabling is still functioning when the fire brigade arrives at a fire, the sequence in which the detectors are triggered will indicate the direction of spread of the fire, allowing the brigade to enter the building behind rather than ahead of it.
With older systems, this argument had some merit as the detector wiring had short-circuit isolators fitted only between zones. The result was that a cable that had burned through would disable the signals from all of the detectors in a zone, robbing the fire service of useful information. Modern detectors, however, such as those in the Siemens Sinteso range, have integral short circuit isolators, so even if a cable burns through, signals are only lost from that detector itself and those immediately adjacent to it on either side.
It seems then, that there is little reason, apart from the dictates of BS5839, to go on using fire-resistant cable for wiring detectors. Indeed, there is very strong argument that by using non-fire-resistant cable and installing more detectors, since fires will be detected sooner giving people more time to evacuate the affected area. Essentially, the installation will be made more efficient and therefore safer.
This is backed up by statistics for fire fatalities from the UK and various European countries. According to the World Fire Statistics for 2002, the number of fatalities from fire for the UK was 1.18 per 100,000 people. In Germany, the corresponding figure was 0.98, in Italy it was 0.82 and in Switzerland it was just 0.62. Clearly, the use of fire-resistant cabling for detectors has no discernable beneficial effect.
At this point, it’s worth noting that it is perfectly possible to meet the requirements of BS5839 without using fire-resistant cabling for detectors! Given the clear references in the standard to the use of fire-resistant cable, this may seem contradictory, but the standard itself allows variations from its prescribed terms “for good reasons”.
Such variations have to be agreed by all interested parties, such as the specifier, the contractor, the building owner, fire authority and the insurers, and are officially described as “agreed variations”. A variation covering the wiring of detectors is certainly possible, and it’s important to understand that an installation in line with BS5839, which includes one or more agreed variations, is still fully compliant with the standard.
Agreed variations are, without doubt, rather troublesome to implement and there can sometimes be problems in getting all of the interested parties to understand the arguments well enough to agree. In addition, it is hard for fire protection contractors to put forward an offer based on a possible agreed variation when their competitors are adopting a more conventional approach.
Nevertheless, it can be worth persevering with agreed variations partly because they lead to better, safer installations but also partly because the regular adoption of agreed variations of this type will lead to pressure for the standard to be changed to bring it more closely into line with European practice. This would ultimately be the best solution, but don’t necessarily expect it to happen soon – revisions to standards have to be very carefully evaluated, and that is no speedy process.
Hopefully, this article has demonstrated that the traditional UK practice of using fire-resistant cable for detectors in fire-protection installations delivers no benefits. On the contrary, the high cost of wiring detectors in this way leads specifiers to opt for minimum acceptable coverage, rather than the coverage that will give the best possible protection. In short, it’s high time for a change!
Contact Details and Archive...
Related Articles...