Fire detection, false alarm caused by liquid spray

[Chris Houston]

Ladies and Gents,

A client of mine does some engineering work.  They have a room that contains some expensive and important equipment that is protected by a gaseous fire suppression system.  The gas is dumped upon activation of the fire detection system.  The system was upgraded in the past year.  During the upgrade the smoke heads were changed from ionization to optical.  Since then the equipment in the room (as will happen sometimes, but rarely) caused and unexpected release of high pressure liquid.

The liquid went into the optical head and set of the system causing an unwanted gas dump.  Previously when this happened, ionization detectors got gunked up, but did not activate.

Question 1 - is the differing reaction of the differing types of smokes heads expected?

They have asked their servicing company (let's call the ABC for the purposes of this) to chagne the heads from optical to ionization, but have been told they are obsolete.  The heads are intrinsically safe (while there are no flammable liquids in the room, the a high pressure spray of the liquid is ignitable in certain circumstances).  The control panel is Minerva type.

Question 2 - Does anyone believe ABC when they say that intrinsically safe ionization heads are unavailable for a Minerva panel?

[GregC]

http://www.tycosafetyproducts-europe.com/english/products/fire/minervadetectorminervais.asp

Looks like they are still advertised as available, might be a protocol issue etc, there might also be a better type of detector available to suit the room and protect against false alarms.

edit: A thought, shouldnt the system be set to double knock ie 2 different detectors on different zones need to be activated to release the gas, you usually have ionisation and optical present, not sure how your clients set up is but seems a bit strange to dump a lot of very expensive gas on a single detector activating

[Chris Houston]

The gas is expensive (£4,000 per dump) but the equipment is absolutely business critical.  But yes, perhaps a coindance detection might still be prudent.

[Graeme]

.   a coindance  might still be prudent.

is that a disco for the rich?

[Chris Houston]

Oops, and I was trying to use the correct terminology as per BS 5839 ("Coincidence" rather than "double knock").

Note to self: use spell check in future.

[kurnal]

In acordance with the relevant BS (cant remember the number off the top of my head - the wrong side of a few cans now) all such systems should be configured to only discharge on a double knock. The detectors will be at half the normal spacing- ie twice as many per area covered.  When the equipment relased the liquid, did it directly affect just one detector or did it cause the release of a mist over a wider area affecting a number of detectors? Opticals will always be susceptible to mists.  In these days of environmental awareness it can only be a matter of time before ionisation detectors are no more. I understand that we are one the last countries in Europe to still use them?

[Chris Houston]

Are you regs you are thinking of not for gas systems in IT rooms (not that I'm sure either)?

Not sure how many heads were affected, but hard to imagine it spraying into 2 heads, but I'll check this and report back.

[Galeon]

Evening Chaps ,
If we assume that it is a general extinguishant set up , all that has been muted makes sense , however I have come across jobs that were analogue addressable , dumping gas , so it could be analogue system that has been configured wrongly.
The reason for my thought on this , is generally once you have a zener barrier fitted outside the area any conventional detector (and this is a general comment) should indeed work.
Unless Minerva had there own barrier to suit there panel and detectors the above would not come into play.
If it transpires it is only a conventional extinguishing arrangement , it would have had to be in 'auto' mode to dump, and the spray might have hit a void detector that has gone unnoticed.
Sorry to ramble on but a bit more digging on the arrangement might lead to shedding more light.

[Chris Houston]

What is a zener barrier?

What is a conventional extinguishing arrangement?

It was in auto mode.

[Galeon]

Hello Chris ,

Conventional arrangement would be minimum 2 radial wired circuits that cross with each other throughout the gas area , so the gas area would  be 1st detector zone 1 detector next to it zone 2 and so on in rotation.
Upon fire condition when the panel is in auto 2 fires on 2 zones would initiate timer circuit to put voltage out to the solenoid which would fire the agent , sometimes via another pilot cylinder to pnematically dump further cylinders.

Zener barrier , is basically used to ensure the supply voltage of the fire alarm does not produce a spark, once the circuit has entered the risk area .

mail me if you want a chat , on a night shift .

[Allen Higginson]

Ladies and Gents,

A client of mine does some engineering work.  They have a room that contains some expensive and important equipment that is protected by a gaseous fire suppression system.  The gas is dumped upon activation of the fire detection system.  The system was upgraded in the past year.  During the upgrade the smoke heads were changed from ionization to optical.  Since then the equipment in the room (as will happen sometimes, but rarely) caused and unexpected release of high pressure liquid.

The liquid went into the optical head and set of the system causing an unwanted gas dump.  Previously when this happened, ionization detectors got gunked up, but did not activate.

Question 1 - is the differing reaction of the differing types of smokes heads expected?

They have asked their servicing company (let's call the ABC for the purposes of this) to chagne the heads from optical to ionization, but have been told they are obsolete.  The heads are intrinsically safe (while there are no flammable liquids in the room, the a high pressure spray of the liquid is ignitable in certain circumstances).  The control panel is Minerva type.

Question 2 - Does anyone believe ABC when they say that intrinsically safe ionization heads are unavailable for a Minerva panel?

Hi Chris - I'm assuming it's a conventional extinguishing system.This being the case then you can still get IS ion's fron your service orgainization or Apollo to suit.

[Galeon]

Are you regs you are thinking of not for gas systems in IT rooms (not that I'm sure either)?

Not sure how many heads were affected, but hard to imagine it spraying into 2 heads, but I'll check this and report back.

Bs 6266

[Mr. P]

Wakey wakey you lot, You have a long day ahead! If something is is interfering with drinking, give that up and carry on drinking! Coffee anyone?

[Chris Houston]

BSI BS 6266 Code of practice for fire protection for electronic equipment installations

This is not an "electronic equipment installation" (i.e. it's not an IT room).

[Allen Higginson]

Chris
It is unusual to have an extinguishing system set up as single knock using "normal" technology detection methods due to the risk of false alarms.
Instances where we have had single knock are an adhesive mixing and coating production area which has Hawco Fenwall stats installed and a printing process area that uses solvent based processes which uses flame detection.
Is it an addressable Minerva controlling this (and is it an 80 or an MX?)?