Smoke detectors suitable for a very dusty environment

[kurnal]

Once again I would be grateful for any advice on the following problem.

A very large cement plant has about 20 major internal oil filled substations ranging from 66kv- 3.3kv and about 100 control panels throught a wide range of buildings. There are a number of internal and external cable chambers up to 4x4x3 metres and cable ducts up to 1m wide and 1m deep.  The life risk is very sparse but the potential for business disruption  enormous should a fire occur  and develop undetected.

So they wish to instal automatic detection to give early warning of overheating or incipient fires. This will augment prevention procedures, including regular  thermographic testing and oil testing.

I am trying to determine whether fire, infra red, flame, smoke or heat detection may be the best way to alert the control room giving an early warning of fire, but with the need to install a very extensive fire alarm system over about 20 acres of plant buildings, or whether to recommend temperature monitoring through a building management system, - cheaper, but a much less resilient networked system using existing BMS pvc cabling.

The client would prefer smoke detection but the the big issue is that the whole site is dusty and some areas extremely so.  Has anybody encountered a similar situation or experience of protecting underground cable ducts, substations and cabinets?

[Ashley Wood]

Kurnal, I believe  'Cerburus' (excuse my spelling if this is wrong) produce a flame sensor/beam detector that has an air blower. The air blower keeps the lens clear of dust. This may be what you are looking for?

[kurnal]

Thanks Ashley I will follow that up.

[Thebeardedyorkshireman]

Kurnal
I would be tempted to consider vesda detection with the holes at point detector spacings. This would give you good value, the filters will take care of the dust and are cheap to replace. Regular pipe 'blow down' will help. The new Gents dual optical with CO is a good head to prevent false alarms from dust.  'Fire wire' is good for cable tunnels if you run a length above each tray for rapid detection. A combination of all three would be horses for courses to protect a whole site.
Dave

[wee brian]

Cerberus got bought out a few years ago - cant remember by who

[Paul]

Kurnal

Try Protec.  Their latest breed of asperating smoke detection is imune to dust.  Operates much the same as the VESDA but lastest technology allows it to activate on the prodcuts of combsution only therefore reducing the liklihood of unwanted false fire alarms and in turn providing early detection.

Worth a look.  I have seen it run very well in industrial bakeries where the issues are much the same as in your customers.

I can give you a contact if you like.

http://www.protec.co.uk/cirrusafd.htm

Paul

[kurnal]

Thanks Paul I have mailed you.

Dave- When you suggest Gent are you thinking of the sQuad point detector?

In two of the substations the dust build up is about two inches deep on the floor and all surfaces. But some are reasonably clean.

Do you know if  there is anything available as an add on kit to make the blow down easy as part of a regular routine for site fitting staff?
I guess the aspirating detection system would give us fault alerts if the filters were blocked- low flow?

[afterburner]

kurnal, have experience of aspirating systems becoming blocked by 'normal' dust from small occupied room which has led to fairly high levels of maintenance in terms of powered cleaning of the tubes. This did indeed cause 'low flow' fault alarms which we have tried to engineer our way through by decreasing the flow and increasing the inlet guards. However, if the system can tolerate the dust they work well.

[Paul]

Hi Kurnal,

I have replied to your email.

Paul

[FireFly40]

I got invovled with commissioning an aspirating system in a tile factory where the dust was about 2" deep. OK for about 24 hours then repeated low flow and filter dirty. I would think anything with a filter will be a chocolate teapot. Also have tried the kidde laser thing (cant remember name) it monitors for particle size and behaved OK but was only in warehouse with fork lifts also networkable by themselves and software good. Detronics make a really good uv ir flame det that tells you when its dirty but well expensive and youll be cleaning lenses forever Personally I would go for fenwall stats in tranformer rooms and firewire along cable tunnels (the thermocouple type not the melty springy one) In 25 years I havent found any smoke detector that wont get blocked or fire with that amount of dust.

[Graeme]

http://www.systemsensoreurope.com/English/2_Products/2_3_1_Filtrex.html

[Allen Higginson]

Cerberus got bought out a few years ago - cant remember by who

That would be Siemens - I'd suggest the S-Line Neural head which can be put in dirty enviroments and ignore nuisence pollution and comes with a no false alarm guarantee.Any financial loss through down time is compensated (this only works with the Siemens protocol of course).
They also do a fibre optic linear heat detection system (Fibrolaser) that is installed like Alarmline but will pinpoint down to 3m (or thereabouts!) as to where the fire is via a graphical representation.It has been adapted for detecting trains that have broken down in the tunnels in Switzerland.

[kurnal]

Thanks Gentlemen- some brilliant help here and lots of potential solutions. I will let you know which way the client chooses to go.

I am moving to a position in which I am likely to recommend a mixture of your solutions- the linear heat detection for cable trays and ducts linked to a fire alarm system, thermostats in the equipment cabinets reporting via the BMS and for the cleaner substations probably optical or vesda.

Client doesnt want to have to carry out filter maintenance more frequently than 3 monthly so that will point the dustier substations towards heat detection.

I will be contacting all your links and asking them for quotes.

[Thebeardedyorkshireman]

Kurnal
Back in the ranch now. Yes, S Quad dual optical C.O. is the head in question.
I have done the reverse blow aspirating in cold stores to clear ice crystals. I also spent a great deal of time in Germany, talking with Werner Wagner ( he owns Wagner) about auto blow in Swedish sawmills. Give me a ring on the landline if you want to go through the theory. Stats are fine, used them for years ( used to be called Pye-Ether stats prior to Fenwall) but a bit slow to detect unless you get a flash.  The Kidde aspirator was called the uni-laser or the little one was the tele-laser. Great in its day but a bit long in the tooth now.
Dave
Dave