How do you answer that?!

[Tom W]

Q. But what if the sounders fail?

A. ?

The answer in my head is that they don't but I need to get more technical than that. If the theoretical question was put to you how would you answer it?

Im genuinely interested as to how you go about answering something like that! Sounder companies just seem to be saying "they don't fail" thats not enough for me! I need the rest of that statement!

Why?!

[Meerkat]

What?  All of them?  Is that even possible in a modernish_ system without some huge failre that takes out haklf the building?

[Big_Fella]

Nice question!

Assuming they mean 'the internal components' failing on all sounders on the fire alarm system within the building.  I would possibly approach it with the 'probability' of that happening.  If all circuits/cabling were to fail, the monitoring of the circuits would be shown as a fault at the control panel.  If the circuitry within the sounders failed, this may not get picked up until the sounders were activated during the weekly test.

So, whats the probability of all sounders developing this fault within one week? Virtually nil.

More chance of a plane parking in the MD's space I think  

[wee brian]

If they fail at the same time as there is a fire, then there is an increased risk that somebody might die.

But you are probably more likely to be struck by lightning so lighten up a bit.

[kurnal]

A lightning strike might just do it- multi seated fires, total meltdown of all building electrical systems.......

[SeaBass]

I agree that the chances of a catastrophic failure on a conventional sounder system are very low, but they are considerably better on a voice alarm system operating with a central amplifier, even when the amps and power supplies are duplicated. The faults usually revolve around software problems, but not always.

I've actually written system failure procedures in the past to ensure that buildings can remain occupied when such faults occur.     

[Tom W]

Cheers gang

So do your answers stand firm if we were to talk about a single sounder failure? 

[Graeme]

no-as sounders can and do fail individually.

[Tom W]

Thanks Graeme

Do you know of any case studies of them failing or why they fail individually?

[Wiz]

Anything can fail at any time for any number of reasons.

Electronic sounders are most likely to fail in normal operation due to a failure of a component part.

All manufacturers would normally select component parts to be used in their equipment based on the component part manufacturer's claimed failure rate per unit over time and against cost. i.e they would aim to use the most potentially reliable component they can afford.

In truth, very few electronic components fail during normal use. Most fail and are identified during post-production completed product testing and are thus never supplied to be installed anywhere.

However anything can fail at any time!

[John Webb]

There is also the fact that even small non-addressable panels often have several output circuits for sounders each individually fused or otherwise protected. So if properly used the failure of a complete sounder circuit would not completely silence the system within the building. You might not get the minimum sound levels in the affected area but something might be heard from adjacent areas if the designer has done their job properly.

[David Rooney]

In truth, very few electronic components fail during normal use. Most fail and are identified during post-production completed product testing and are thus never supplied to be installed anywhere.

Hmmm I need to put that to a couple of manufacturers we deal with..... and remind them about post production testing........

[Graeme]

Thanks Graeme

Do you know of any case studies of them failing or why they fail individually?

Hi Piglet

Not case studies but from experience.

[Big_Fella]

Even the odd sounder failing may not get picked up during the weekly test unless everyone stands by a sounder during the weekly test to ensure they operate.

Although there are a couple manufacturers on the market (and this is with addressable systems) who manufacture sounders with inbuilt microphones, this monitors the sound output when the sounder 'should' activate.  If no sound output is detected a fault is triggered at the main panel.

[Wiz]

There is also the fact that even small non-addressable panels often have several output circuits for sounders each individually fused or otherwise protected. So if properly used the failure of a complete sounder circuit would not completely silence the system within the building. You might not get the minimum sound levels in the affected area but something might be heard from adjacent areas if the designer has done their job properly.

John, interestingly, BS5839-1 doesn't specifically recommend the type of 'interleaving' of fire alarm sounders of different monitored circuits that might achieve the scenario you mention above, other than in areas designed to accommodate a large number of people.

Whilst 'interleaving' isn't specifically frowned upon, the actual drawings & text included in BS5839-1 relate to a system with a single sounder mounted close to the cie on one monitored circuit, and all other sounders in the building on the other monitored circuit.

Therefore a system designer could still comply fully with the recommendations of BS5839-1 in providing a system where the complete failure of one of the sounder circuits means that there is virtually no alarm sound level in large areas of a building during a fire condition.