Suitable type of detection?

[BCO]

Currently involved with a project that has a number of underground shafts and tunnels (serving underground plant areas) the tunnels are about 60m long and about 20m below ground. Access is via a number of vertical shafts (stairs and ladders). The tunnels contain a series of cable trays. Some plant is positioned at the base of the shafts. Access is for maintenance and repair only (therefore limited). The environment is very humid wet and subject to large climatic changes. There are no suppression systems proposed. The fire load is limited to the cables and plant. (which is not oil cooled)
The question relates to the most suitable type of detection to use in such an area. The primary objective is for asset protection. (Crucial)
My thoughts initially were an aspirating system would give the earliest warning of a fire (or smoldering) however I have experience of an aspirating system in a similar environment and it is plagued with problems (due to the environment). This leads me to considering marine type, point detectors. This method of detection is not likely to detect quite as early as the aspirating system, however this brought me on the thought, ‘if a fire was present and warning had been given, would the Fire Service commit crews to an underground area?’ I understand this would depend on the conditions within the enclosure, but in the majority of case I would imagine the answer would be no?

Given that a developed fire in the enclosure is unlikely to be tackled is there any merit in a more responsive (aspirating) detection device being specified?

[Redone]

Shooting in the dark here, for asset protection, perhaps FM 200 flood system with detection activated by 'double knock'...  problems with venting and recapturing after the event though.

I cannot see the fire service commiting crews, unless there is some serious functional compartmentation in place, after all you'd want the service engineer to be able to escape safely.

[kurnal]

Sounds an ideal application for wagner oxyreduct as a first choice. You wont have a fire. Alternatively are the spaces large enough for video detection to be used - such as Dtec? That way you can see what is going on without commiting staff into the danger area.
But it sounds like an application designed for oxyreduct!

[BCO]

Yes, oxyreduct or some form of flood suppression system would be the answer, as these will give the ultimate asset protection. However, as is always the case, less costly alternatives are initially being investigated. The cctv detection has crossed my mind, however, in the similar facility, (described in my question), beam detectors were installed only to constantly have steamed lenses making them in operable, my assumption is the same may happen with the cameras?
I suppose the key question presently is ‘will a series of durable point detectors (multi sensor, any suggestions??) provide the same level of asset protection as a correctly working aspirating system?   I.e. could they detect the fire as quickly and if not is the difference between the two systems likely to have a benefit in terms of asset protection?

[kurnal]

Oxyreduct will ensure you will never have a fire. Yes faults will arise and pyrolysis will take place but you will be alarted to it by the sampling detector in the oxyreduct system and subject to confined sapce permits to work - which you will have in place people can enter and work in the protected environment for up to 4 hours.
Now what is that worth in terms of business continuity?

Presumably the environment is regulary changing down there - particularly in terms of humidity and temperature and it is this that is affecting the beam system? The video system can use conventional high definition cameras- and these can be housed to work outside or in very hostile surroundings so this should not be a problem. fit it in a box with a wiper blade?

Fially aspirating systems have come a long way since the early days - what type of fault conditions were you getting? This should be simple to resolve.  For multi sensor heads I have only experience or working with the Gent SQuad range in which a dual optical chamber, forward and reverse scatter  work in conjunction with rate of rise in which as I understand it the sensitivity of the optical chambers increase if accompanied by a rise in temperature. The system can be programmed to 5 sensitivity states so false alarms can  be eliminated. Maintenance will be much more of a problem with point detectors.

The gut feeling is that it cannot be as quick as a properly configured aspirating system - but what are you going to do and how quickly can you respond to  an alarm? and how will you respond? I doubt the fire service will enter the space unless life is at risk and certainly will not have resources on board to deal with an incident underground in any case. ( apart from a lot of water).
The chances are that if an aspirating system costs you £x, a multisensor conventional system will also cost you £x, video detection will probably cost you £3x and  oxyreduct will cost you £4-5x. For business critical installations it may be worth the premium for the benefits?

[Fishy]

How on earth are you getting "large climatic changes" 20m underground?  Dampness and high humidity - sure - but the last thing you normally get at that depth is a large variation in climate.  Is there some process that's responsible?

[wee brian]

BCO's point is if the fire service wont go down there then it is a bit pointless getting early warning. Makes sense to me.

I think video systems need light don't they?

[kurnal]

Yes Wee B- artifiicial light or infra red will work with Dtec

[BCO]

There is a maintained emergency lighting system to 5266 so the light level for the cctv detection should be ok. The temperature changes are just normal external temperatures varying between near 0-20 degrees C.
So I suppose this is not that significant (probably a red herring on my part).
I think I have come to the conclusion that the speed of response (within reason) is not going to make a difference to the level of asset protection, however I do believe that if true protection is required then an oxyreduct system is definitely the answer. Whether it goes in is another question! That will be a decision for the money men!

[Ashley Wood]

BCO, Oxyreduct would do the job but have you considered high pressure water mist? I know of several large cable duct and cable flats projects across the world where these systems have been used; one is in the 'Woodhead tunnel'. This is the main power transfer tunnel for the National Grid across the Pennines and is 5km long! Some of these systems have approvals for cable tunnel applications as well. Have you considered fibber optic or digital/analogue linear detection cable? Again, thousands of kilometres of this stuff is in use in cable tunnels and works well.
Food for thought!

[wee brian]

or use cables that dont burn? (much)

[LouisK]

Part of the solution lies in your own text.  Detection without a response is a total waste of money.
Beam detectors will work well in tunnels and point IR/UV detectors such as the Apollo "Sharp Eye' mini UV detector would be suitable to cover the equipment and can handle extreme conditions. Linear detection has also been used sucsessfully for years. If the tunnels are sufficiently sealed then a reduced oxygen gas or CO2 would be a suitable response.  (Remembering that a purge system would have to form part of the CO2 system.) Ablative cable coatings such as the KSB range are also a viable risk reduction measure.  Since it is passive and only the malfunctioning cable would be affected it can reduce down and recovery time significantly.  It is also maintenance free.

[terry martin]

could you not use air sampling detectors, they use them in heritage buildings so there is no visible detector. it periodicaly draws in a sample of air and analyses it in a chamber. if i'm right they would not be affected by the climate.