[AusNOG] {Disarmed} Re: [AUSNog] : Re Data Centre Fire Suppression Safety

[Tony Wicks]

I remember working at an underground DC (Auckland) that had a spontaneous
Halon dump due to a maintenance guy getting too close to one of the sensors
and transmitting on his RT (25 years ago). Tiles went flying, luckily no one
died. So the 1 minute only applies if everything is working. Part of the
procedure if the  1 minute warning lights and sirens went off was for us
network guys to check the DC for anyone working inside. As you can imagine
we refused to ever carry out that step should the warning occur again. That
DC was physically very large and had no portable oxygen systems available,
criminal negligence really. 

 

  _____  

From: AusNOG <ausnog-bounces at lists.ausnog.net
<mailto:ausnog-bounces at lists.ausnog.net> > on behalf of Nicholas Hobbs
<Nicholas.Hobbs at epworth.org.au <mailto:Nicholas.Hobbs at epworth.org.au> >
Sent: Thursday, 13 December 2018 12:54 PM
To: ausnog at lists.ausnog.net <mailto:ausnog at lists.ausnog.net> 
Subject: Re: [AusNOG] [AUSNog] : Re Data Centre Fire Suppression Safety 

 

In our modest sized data centres we have a 1 minute delay timer from VESDA
detection of a fire (requiring multiple sensor confirmation) to firing of
the gas. Once confirmed, loud hailers and flashing LED signs go off, making
it impossible not to know to that you need to evacuate. 

Our one minute is based on the maximum time it would take someone to walk to
the exit from the furthest point in the room plus a 30 second buffer. 

Entry doors have large signs and instructions on what to do which are
covered on induction of the very limited number of people allowed in there.

All of our installs (oldest is 8 years, newest is 2 years) had to be signed
off and approved by building and fire inspectors and receive maintenance
inspections every 6 months and monthly fire panel operation tests. There was
a standard they had to be signed off to, but can't recall it at the moment. 

 

I'm not across how halon systems we loaded/dispersed, or what system was
used in Antarctica (but as it is a power generator room mentioned, assuming
they're still using Halon.), however our systems discharge via an explosive
firing pin at the bottle head which is a pretty binary state system and
means the pipes are empty. They also have a hardwired pressure sensor that
alarms (and sets off the evacuation alarm) if there is a pressure change in
the bottle. Best practice also has the bottles in a separate room to the
data room. 

Pure speculation, however Antarctica's temperature may have caused unknown
factors such as a fractured pipe or valve from extreme exposure over time,
or as it was a power generator building, there may have even been a
flashover event causing a 'valid' discharge. 

 

Nick  

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20181213/0cdb7d42/attachment.html>