Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Alarm Failure was Ionization Detectors



I don't have really solid references, but I get the impression
that ionization smoke detectors were designed against an
objective rate of a failure per thirty years, so that a house with
ten might see one fail every three years on average.

But I get the strong impression that
the battery energized type could exhaust its battery and
the warning alarm might pass quickly enough to be unnoticed
if a battery with a strong voltage knee at end of life were
employed.

Using rechargeables was also considered unwise, because these
self-discharge rather quickly. Apparently the building code
requirements for such things now often require a hard-wired
connection to a mains supply, with a battery as backup.

For free standing smoke alarms people recommend a ten year
battery, so that the whole device is ditched at that time.
I notice an undercurrent of preference for the IR side scatter
or forward attenuation sensing types - though I'm not sure
this is rational.
A photo sensing, ionization, mains wired hybrid with backup
is supposed to be the gold-standard. As to testing, fire marshals
want more than a press of the test button.

I suppose a snuffed taper would be ideal.

Brian Whatcott Altus OK


At 12:09 PM 9/29/2006, you wrote:
What's the problem? Just a bad battery, or an actual fault in the
system. How are they testing them?

Michael D. Edmiston, Ph.D.


Brian Whatcott Altus OK Eureka!